Professional Documents
Culture Documents
F
G
H
I
Springer- Verlag
Berlin Heidelberg NewYork Barcelona
Budapest Hong Kong London Milan
Paris Santa Clara Singapore Tokyo
3. High Technology
4. Science and Technology Policy
5. Computer Networking
Kluwer Academic
Springer-Verlag /
Kluwer Academic
Kluwer Academic
Kluwer Academic
Kluwer Academic
Publishers
Publishers
Publishers
Publishers
Publishers
NATO-PCO Database
The electronic index to the NATO ASI Series provides full bibliographical references
(with keywords and/or abstracts) to about 50 000 contributions from international
scientists published in all sections of the NATO ASI Series. Access to the NATO-PCO
. Database compiled by the NATO Publicat ion Coordination Office is possible in two
ways:
- via online FILE 128 (NATO-PCO DATABASE) hosted by ESRIN, Via Galileo Galilei,
1-00044Frascati, Italy.
- via CD-ROM "NATO Science & Technology Disk" with user-friendly retrieval
software in English, French and German ( WTV GmbH and DATAWARE
Technologies Inc. 1992).
The CD-ROM can be ordered through any member of the Board of Publishers or
through NATO-PCO, B-3090Overijse, Belgium.
Discourse, Tools,
and Reasoning
Essays on Situated Cognition
Edited by
Lauren B. Resnick
Learning Research and Development Center
University of Pittsburgh, 3939 O'Hara Street
Pittsburgh, PA 15260, USA
Roger Saljo
Department of Communication Studies
Linkoping University, Sweden
Clotilde Pontecorvo
Dipartimento di Psicologia dei Processi di Sviluppo
e Socializzazione, Universita degli Studi di Roma
"La Sapienza", Italy
Barbara Burge
Learning Research and Development Center
University of Pittsburgh, USA
Springer
Published in cooperation with NATO Scientific Affairs Division
Preface
Not long ago, projections of how office technologies would revolutionize the
production of documents in a high-tech future carried many promises . The paperless office and the seamless and problem-free sharing of texts and other work
materials among co-workers were just around the corner, we were told. To anyone
who has been involved in putting together a volume of the present kind, such
forecasts will be met with considerable skepticism, if not outright distrust. The
diskette, the email, the fax, the net, and all the other forms of communication that
are now around are powerful assets, but they do not in any way reduce the flow of
paper or the complexity of coordinating activities involved in producing an artifact
such as a book. Instead, the reverse seems to be true. Obviously, the use of such
tools requires considerable skill at the center of coordination, to borrow an
expression from a chapter in this volume. As editors, we have been fortunate to
have Ms . Lotta Strand, Linkoping University, at the center of the distributed
activity that producing this volume has required over the last few years. With her
considerable skill and patience, Ms. Strand and her work provide a powerful
illustration of the main thrust of most of the chapters in this volume : Practice is
a coordination of thinking and action, and many things had to be kept in mind
during the production of this volume. We are all very grateful for her willingness
to engage in this project in the midst of all her other duties .
The volume is the result of a seminar entitled Discourse, Tools, andReasoning:
Situated Cognition and Technologically Supported Environments, which the
editors organized in Italy. The seminar was made possible with financial support
from the NATO Special Programme Panel on Advanced Educational Technology .
We are grateful to the Programme Director L. Veiga da Cunha for this
opportunity to explore important issues of technology and learning .
We thank Barbara Kester of the International Transfer of Science and
Technology office for her help in the early stages of planning the seminar and
locating a suitable site.
Finally, we also thank Bruno Giannasi, the overseas conference coordinator at II
Ciocco, for his cooperation and attention to details before and during the seminar.
Contents
XI
..
Part One
Distributed Cognition:
Discourse and Activity in Complex Work Environments
Chapter I Constructing Meaning from Space, Gesture, and Speech
Edwin Hutchinsand Leysia Palen.. ...... ..... ............. .............. ..... ...... .....
23
41
63
87
VIII
Contents
Part Two
Negotiating Identities:
The Construction of Sociocognitive Communities
Chapter 6 Reasonable Uncertainties: Parents ' Talk About
Caring for Children with Chronic Renal Failure
David Middleton ...... ...... .... ......... ............ ......... ................................. 143
Chapter 7 Syncretic Literacy in a Samoan American Family
AlessandroDurantiand Elinor Ochs................. ............................ ........ 169
Chapter 8 Other Voices, Other Minds:
The Use of Reported Speech in Group Therapy Talk
Alessandra Fasulo... ... .... ........... ...... ........... ...... .................... ..... ... ... .. 203
Chapter 9 Situational Effects in Computer-Based Problem Solving
Paul Light and Karen Littleton... .... ........... ................ ............... ........... 224
Part Three
Learning in Practice:
How People and Tools Shape One Another
Chapter 10 Discourse and Development: Notes from the Field
Joseph Glick... ...................... .. ...... .............. ......... ........... ... ... ... .... ... 243
Chapter 11 Interactional Perspectives on the Use of the Computer and on
the Technological Development of a New Tool: The Case of Word Processing
Michele Grossenand Luc-Olivier Pochon.. .......... .. ..... ..... ...... .... .... ........ 265
Chapter 12 What Organizes Our Problem-Solving Activities?
Terezinha Nunes............. .. ............. ........ ....... .................. .... ... .. ..... .. .. 288
Chapter 13 Understanding Symbols With Intermediate Abstraction s:
An Analysis of the Collaborative Construction of Mathematical Meaning
Baruch B. Schwarz.i....
312
Contents
Part Four
IX
361
Chapter 16 Seeing the Light: Discourse and Practice in the Optics Lab
Roger Siiljo and Kerstin Bergqvist
385
406
443
Author Index
459
Subject Index .... ........ ......... ........ .................. ......... ....... ... .. .... .. .. ..
469
List of Contributors
XII
List of Contributors
As we planned the conference, four major topics were the focus of our attention :
the role of interactive discourse (talk and, occasionally, written documents) in the
coordination of complex cognitive activities; reasoning and learning as socially
distributed activities; social and institutional constraints on cognition; and
technologies for learning. We sought to include research on cognitive processes in
a variety of social and institutional contexts--for example, families, the medical
and social service system, schools, laboratories, and factories--and asked
participants to discuss questions of methodology and fundamental research
assumptions throughout, thus heightening the prospect of future collaboration
across disciplines .
The themes with which we began our planning represented well the kind of
thinking that characterized the loosely defined field of situated cognition five years
ago. But it is striking that, although each of those themes can be found threading
through the chapters of this volume, they are no longer the most helpful
characterizations of what this book is about. Instead, the three terms in our tiUe-discourse, tools, and reasoning--came to represent the recurrent themes of the
chapters and of our discussion.
Discourse. Discourse and interaction are dominant themes throughout this
volume . These terms are in such widespread use in psychology, sociology,
linguistics, and other disciplines that our use of them here might lead readers to
overlook the theoretically radical view of the role of discourse in human cognitive
functioning that most of our authors hold. Although many scholars treat talk and
other forms of communication as a means of transmitting ideas between
individuals or perhaps of coordinating discrete cognitive actions, our authors treat
discourse primarily as constituting cognition . In this respect, they go well beyond
Vygotsky's (1934/1978) view of speech and linguistic interaction as the primary
means by which children are socialized into a community and its forms of
thought. In Vygotsky's theory, originally interpersonal exchanges become
internalized as intrapersonal modes of thought. The same concept is expressed in
George Herbert Mead's (1934) definition of thought as conversation with the
generalized other. Vygotsky and Mead both treated interaction as the source and
origin of thinking but viewed thinking itself as, eventually, activity carried out by
the individual . Our authors, by contrast, view thought and reasoning as inherently
(and throughout the life span) social activities in which talk and social interaction
are not just a means by which people learn to think, but also how they engage in
thinking . They might say that discourse is cognition is discourse. . . . One is
unimaginable without the other.
Because discourse is publicly embodied in speech, it is easy to interpret as
action . Thus the notion that discourse and cognition are coterminous helps us to
accountable within social communities and how people acquire the capabilities for
reasoning that their communities recognize .
10
When we enter human life, it is as if we walk on stage into a play whose enactment
is already in progress--a play whose somewhat open plot determines what part we
may play and toward what denouement we may be heading. Others on stage already
have a sense of what the play is about, enough of a sense to make negotiation with
a newcomer possible (p. 34) .
11
12
Littl eton, "The social context of a cognitive task embraces not only direct
interpersonal interactions but also the social norms, expectations, representations,
and comparisons that condition such interactions."
13
14
typewriter . But the changes created in the general world of office machines flowed
back into computer design to create what is today a historically unprecedented set
of mental processes for composing and using texts and an accompanying set of
institutional changes . From a sociocognitive point of view, texts are now viewed
as less permanent, and the use of word processing has enlarged the possibilities of
co-authoring. Word processing has also had an impact on the organization of
work, dramatically changing secretaries' jobs and distributing "keyboarding tasks"
throughout the organization, thereby changing their perceived meaning with
respect to status, knowledge, and power.
Word processing is but a single example of the ways in which tools participate
in the definition of culture and competence. Any environment-that is, any
activity system that must be learned by a newcomer-contains a set of tools
(technologies in the broadest sense of the term) and practices for using them that,
in effect, define the culture that the newcomer must enter. Nunes (chapter 12, this
volume ; see also Nunes, Schliemann, & Carraher, 1993) shows that these tools
need not be highly technological. Treating arithmetic procedures as cultural
artifacts, she shows how such procedures guide and constrain reasoning in ways
that are adaptive to the particular situations for which they have been designed.
Different procedures invite different kinds of errors, and Nunes argues that
procedures that have grown up in certain social activity systems protect against
errors that would have important social consequences. In this discussion, although
arithmetic procedures exist only as ideas and verbal rules, they are as constraining-and supporting--as the physical artifacts that are more commonly recognized as
technologies. In another example of the way in which symbols structure thinking
in relation to particular activities, Nunes shows that the use of tiles in learning to
measure area is more effective than the use of rulers, because tiles carry the
meaning of area. This is so even when there are not enough tiles available
physically to cover the space being measured. The tile as symbol for a unit of area
nevertheless continues to function to support reasoning.
Another example of how artifacts and systems of representation affect cognitive
functioning is offered by Schwarz (chapter 13, this volume), who shows how a
particular set of computer-based representations of quantity and functional
relations can affect children's conversation and reasoning about mathematics.
Schwarz's representational software is explicitly designed for pedagogic purposes.
It is an instance of an intermediate abstraction (cf. White, 1993) for learning
concepts, a representation that embodies formal mathematical relations yet is
familiar in form and thus cognitively accessible on the basis of everyday
experience. Such intermediate tools make possible the negotiation of shared
meaning among children working together to solve word problems. By talking
about the objects in the representation system, and by modifying the
representations in order to test hypotheses, children "talk math" even before they
have learned the formal language of the field (in this case, algebra). Children have
to attune to constraints and affordances by negotiating the situated environment
established by the symbolic representation system . In so doing, they develop
15
explanations of why the objects behave the way they do, in conversations that are
similar to the transactive ones described in a later chapter by Teasley (chapter 15,
this volume).
In another use of techn ology to support learning , readers effectively become codevelopers of the texts from which they learn . Hypertexts present the material to
be communicated in modularized sections; learners choose the order in which they
want to read the sections, presumably in accord with their preferred learning styles
and information needs. Schnotz (chapter 14, this volume) examines the ways in
which learners use hypertext resources . He considers the possibility that the
modularization of hypertexts interferes with the flow of thought that ordinary
texts help readers to develop and to which they are accustomed. Hypertexts may ,
like word processing, be a domain in which a new technology can eventually
change the learner' s usual way of functioni ng. At the sam~ time, hypertext
technology, if it is to be widely used, will need to be transformed and adapted to
the user' s logic. This process will be aided by careful study of how people use
hyper texts as a function of both the problem-solving task and the type of guidance
that is given to the subjects (Pontecorvo, Cesareni, & Romanelli, 1995) .
Schnotz's study of the processes students used in learning about time differences
in cities around the world is an example of the kind of research likely to be useful
in this process.
16
17
18
are often about who or what is counted as problematic. Frequently, the issue is
whether there is a problem, exactly what the problem is, or who is the focus of it.
In the family discussions analyzed by Pontecorvo and Fasulo, parents often treat
their children 's behavior and ideas as problematic. They then appeal to categorical
rules and values to educate the children in the categories that function to define
proper behavior. This socialization is managed through specific discursive forms:
reproaches, suggestions, imperatives, and examples . Demands for accountability
are often addressed to children, and children are questioned and challenged more
than parents . By giving explanations and successful arguments, children develop
situation ally appropriate reasoning skills . Particularly striking is the way in
which children use a strategic appeal to a superordinate normative rule (e.g., being
punctual , telling the truth, respecting the rights of children, saving food for
others) to justify an action that has been challenged. Both authorities (e.g., aunt,
grandmother, school teacher, books) and normative values are used, even by very
young children, to challenge others.
The passage from local topics and strategies of discourse to more general and
superordinate ones is shown in another context by Ramirez and Wertsch (chapter
18, this volume), who studied women in an adult literacy program. Analyzing
group discussions on topics relevant to the women (e.g., children's education,
unemployment, drug abuse, and work), the authors show transformations in the
language produced by these adults while acquiring literacy. The issues discussed
by the women shift from reference to their personal daily experiences to more
general social considerations. As they advance in their literacy training, the
women also shift from discussing specific cases to "arguing about categories"
(Billig , 1987). At the same time, they begin to engage complex forms of
expression that combine two discursive genres: the colloquial working class genre
and the rhetorical devices of public debates. This genre combination is aimed at
persuading the audience to identify with their point of view through a kind of
dramaturgical expression . We have here evidence that familiarity with written
texts and literacy practices invokes new forms of oral expression . There is, thus, a
cultural evolution at work, in which mastery of one set of tools (the alphabetic
code and texts embodied in that code) yields a new cultural tool (formal oral
rhetoric). This process of reconstitution of tools in use has been, like discourse
and reasoning, a persistent theme throughout our collection of essays on situated
cognition.
19
References
Bakhtin, M . M. (1991). The dialogic imagination: Four essays . (Ed. by M. Holquist,
Trans . by C. Emerson & M. Holquist) . Austin : University of Texas Press.
Billig, M. (1987) . Arguing and thinking: A rhetorical approach to social psychology .
Cambridge, England: Cambridge University Press .
Bruner, J. S. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.
Cole , M. (1985) The zone of proximal development: Where culture and cognition
create each other . In J. V. Wertsch (Ed.), Culture, communication, and cognition:
Vygotskian perspe ctives (pp. 146-161). Cambridge, England : Cambridge
University Press .
Goffman, E. (1974) . Frame analysis. New York: Harper and Row.
Goodman , N. (1976) . Languages of art. Indianapolis, IN: Hackett.
Greenfield, P. (1984) . A theory of the teacher in the learning activities of everyday
life. In B. Rogoff & J. Lave (Eds.), Everyday cognition (pp. 117-138). Cambridge,
MA: Harvard University Press.
Hill , J., & Irvine , J. (Eds.). (1993). Responsibility and evidence in oral discourse.
Cambridge, England: Cambridge University Press.
Jacoby, S., & Ochs, E. (1995) . Co-construction: An introduction. Research on
language and social interaction, 28(3) , 171-183 .
John-Steiner, V. (1995) . Cognitive pluralism: A sociocultural approach . Mind, Culture
and Activ ity, 2(1), 2-11.
Keller , C., & Keller, J. D. (1993) . Thinking and acting with iron. In S. Chaiklin and J .
Lave (Eds.), Understanding practice: Perspectives on activity and context (pp. 125143). Cambridge, England : Cambridge University Press .
Lave, J. (1991) . Situating learning in communities of practice. In L. B. Resnick, J. M .
Levine, & S. D. Teasley (Eds.), Perspectives on socially shared cogn ition (pp. 6382). Washington DC: American Psychological Association.
Lave, J., & Wenger, E. (1991) . Situated learning: Legitimate peripheral participation.
Cambridge, England: Cambridge University Press .
Mead , G. H. (1934) . Mind, self, and society. Chicago: Universi ty of Chicago Press .
Middleton, D., & Edwards, D. (1990). Collective remembering. London : Sage.
Nunes , T., Schliemann, A. D., & Carraher, D. W. (1993) . Street mathematics and
school mathematics . New York : Cambridge University Press .
Ochs, E. (1988) . Culture and language development : Language acquisition and
language socialization in a Samoan village. Cambridge , England : Cambridge
University Press .
Ochs, E., & Schieffelin, B. B. (1984) . Language acquisition and socialization: Three
developmental stories . In R. A. Shweder & R. A. LeVine (Eds.), Culture theory:
Essays on mind, self, and emotion (pp. 276 -320) . Cambridge, England : Cambridge
University Press .
Perriault, J. (1989) . La logique de l'usage . Essai sur les machines a communiquer [The
logic of use. An essay on communication machinery] . Paris: Flarnmarion.
Pontecorvo, C., Cesareni, D., & Romanelli, P. (1995). Apprendere esplorando
[Learning by exploring] . In M. A. Garito (a cura di), Multimedia and distance
learning for science and technology (pp. 217-237) . Roma: Garamond .
20
Part One
Chapter 1
Abstract
Face-to-face communication in the workplace is often conceived of as consisting
mainly of spoken language. Although spoken language is clearly a very important
medium for the creation of representations, in complex work settings, it is one of
several such media. Gestures and the space inhabited by speakers and listeners are
normally thought of as providing context for the interpretation of speech. In this
chapter we show how space , gesture, and speech are all combined in the
construction of complex multilayered representations in which no single layer is
complete or coherent by itself. We examine a brief explanation given by one
worker to two others. We show how the meaning of the explanation is carried in
the coordination among the spatial organization of specilized artifacts, the
positioning of gestures with respect to those artifacts, and the words that are
spoken .
24
chapter we show how space, gesture, and speech are all combined in the
construction of complex multilayered representations in which no single layer is
complete or coherent by itself. We examine a brief explanation given by one
worker to two others. We show how the meaning of the explanation is carried in
the coordination among the spatial organization of specialized artifacts, the
positioning of gestures with respect to those artifacts, and the words that are
spoken. Our inspiration for this analysis comes from the work of Charles
Goodwin on situated seeing and the cognitive uses of spatial organization
(Goodwin, 19943., 1994b; Goodwin & Goodwin, in press) and of Eleanor Ochs
and her colleagues on the layering of speech and gesture over graphic displays
(Ochs, Gonzales, & Jacoby, in press).
The computational properties of a cognitive system are in part determined by
the patterns of communication within the system. This is true whether the system
is contained in the mind of an individual or distributed across a number of
individuals (Hutchins, 1991, 1995). The representations that are created depend on
the resources available for their creation. What can be represented? How can it be
represented? When a team is engaged in joint reasoning activity, communicative
resources can be seen as media for creating the representations that move
information around inside the system. Communicative behaviors are the
representations by which a socially distributed cognitive system does its work.
Data Collection
The setting for our study is the cockpit of a commercial airliner. This is a
complex high-technology work setting in which the crew engages in event-driven,
high-stakes activities. The quality of the crew's performance depends on their
ability to coordinate their actions with one another and with the dynamic behavior
of the airplane (Hutchins & Klausen, in press; Hutchins, in press). The data were
obtained from a videotape of a simulated flight. The simulation was performed in
a Boeing 727-200 high-fidelity simulator in the Manned Vehicle Simulator
Research Facility (MVSRF) at the NASA-Ames Research Center in Mountain
View, California. Flight in a high-fidelity simulator is very close to the
experience of flying a real airplane. The simulator used in this study provided full
visual displays with dusk lighting conditions and six-degrees-of-freedom hydraulic
motion . A real airline crew composed of pilots employed by a major air carrier
flew a simulated flight approximately one hour in duration from Los Angeles to
Sacramento. The flight was designated NASA 900 in order to hide the identity of
company with which the pilots were employed. As part of the scenario, a
dangerous fuel leak occurred midway in the flight.
There are three cockpit crew members on a 727-200: a captain, a first officer
(FlO), and a second officer (S/O) who is sometimes called a flight engineer. Either
25
the captain or the FlO actually flies the plane, typically alternating with each
flight leg . This person is designated as the pilot flying (PF) . The pilot not flying
(PNF) handles radio communications. For the flight examined in this chapter, the
captain was the pilot flying , and the FlO was the pilot not flying . The SIO
monitors systems such as the fuel and the hydraulic systems on the SIO panel.
The SIO also troubl e-shoots and refers to the airplane operations manual for
procedures in non-normal situations when necessary.
A low-light infrared camera was positioned behind the crew facing forward. All
three crew members could be seen, as well as most of the main flight instrument
panel and some of the second officer's controls and instruments, including the fuel
panel.
The SIO in the NASA 900 flight discovered the fuel leak by monitoring the
fuel gauges located on his panel. The fuel panel is described at length in the
following section. The interactions that ensue between the crew members upon
the SIO's notification of the problem are the focus of this chapter.
26
one, tank two to engine two, and tank three to engine three. The aft auxiliary tank
also has a direct line to engine two.
Fuel from each tank can be fed to other engines as well. Cross-feed valves
control the flow of fuel between tanks and engines in the fuel lines. There are
three cross-feed valves, one for each main tank and engine combination. The fuel
from the tank feeds a manifold, and from there the fuel goes to the engine. If the
corresponding cross-feed valve is open, the fuel also feeds into another manifold,
the cross-feed manifold. This manifold can supply all the engines with fuel,
depending on the configuration of the other cross-feed valves (see Figure 1). When
all three cross-feed valves are open, the fuel is free to flow from all tanks to all of
the engines. Direction of flow is determined by the pressure in the fuel lines and
by check valves that permit fuel flow only in one direction. When the cross-feed
valves are closed (the default setting), the configuration is called tank to engine,
because each engine is fed by only its own tank. The crew can control how much
fuel is burned from which tanks by using the cross-feed valves to direct the fuel
flow.
= '.... .
.-
ax=-..
Fig. 1.1. A model of the fuel system as it appears in training and operations
manuals.
Boost pumps also control fuel flow by supplying the pressure necessary to move
fuel to the engines. Tanks one and three and the aft tank each have a pair of boost
pumps located in the tanks. Tank two has two pairs of boost pumps because the
27
tank is divided into sections, two of which are located over the root of each wing .
Boost pumps can be turned on or off. When the boost pumps are on, they extract
fuel from the tank and feed it into the fuel manifold and to the engine. When the
pumps are off, the fuel remains isolated in the tank.
Fuel Instrumentation
The fuel system gauges and switches are located on the lower left section of the
S/O panel on the 727-200 . The fuel panel in the simulated airplane displays four
fuel quantity gauges: one for each of the main tanks and one for the aft auxiliary
fuel tank. The corresponding boost pump toggle switches, the low pressure
indicator lights , the cross-feed valves, and the engine shutoff valves are also
displayed on the panel (see Figure 2).
Fig. 1.2. The fuel panel. The second officer constructed his gestures in coordination
with the spatial organization of this panel. The number three tank gauge and boost
pump control switches are on the right side of the panel. The fuel quantity test button
is at the far left.
28
A crew uses the cockpit instrumentation and cues from flight controls and other
aspects of the environment to monitor the status of the aircraft. The following
are some of the signs that may be available to a crew when there is a fuel leak in
tank three.
Flight Controls. A leak in tank three, which is located in the right wing,
should cause the plane to roll to the left as the right wing becomes lighter than
the left wing. This condition is called a lateral weight imbalance. In order to
maintain a wings-level attitude with such a lateral weight imbalance, the control
yoke would have to be tipped to the right. The need to adjust the control yoke is a
cue to the pilots that something may be amiss. If the aircraft is on autopilot,
however, the autoflight system would make the required control correction
without notifying the crew. Although the autopilot system still physically tips
the yoke, this visual cue is subtle, because the amount of displacement of the
yoke may be small.
Instruments. Another sign of a fuel problem would be a greater decrease in one
of the fuel tank quantity gauges relative to the other gauges as indicated by the
gauge needle levels. On the 727-200, this information is available most readily to
the S/O because the instrumentation is located on his panel.
When a fuel leak is suspected, a typical response for the S/O is to press the fuel
quantity test switch to confirm that the fuel quantity gauges themselves
are operating properly. This test confirms that an irregular gauge indication is a
result of the physical state of the fuel system and not a result of a malfunctioning
gauge. Pressing the fuel quantity test button moves each of the fuel tank gauge
needles simultaneously to different positions to test for responsiveness. When the
fuel quantity test button is released , the needles return to their original positions.
Once the S/O confirms that the gauges are working properly, the next step is to
locate the fuel leak. A leak could be in one of two places: in the tank itself or
somewhere in the fuel line.
To determine if the leak is in the tank, the fuel must be isolated in the tank by
turning off the boost pumps in that tank. If the gauge still indicates a decline, the
leak is in that tank. If there is no decline in fuel quantity when the boost pumps
are turned off, the leak may be somewhere in the fuel line. This is an even more
dangerous situation than a leak in the tank, because the fuel may be escaping into
the fuselage where it could ignite and destroy the airplane. An additional action
that can confirm a fuel leak is to perform a visual check outside a cabin window
29
to look for fuel escaping from the wing. In a simulator, a visual check is
simulated by asking an experimenter if fuel can be seen from the wing.
Another action must be taken before the diagnostic of turning off the tank three
boost pumps is performed. An alternate fuel supply to engine three must first be
established so that the engine will not stop working when the fuel it normally
receives from tank three is no longer available. To do this, the cross-feed valves in
the fuel lines between the new fuel source and engine three must be opened. Once
this step is taken, the tank three boost pumps can be safely turned off.
8 /0
FlO
FlO
12 .00 .56 8 /0
ohhhh
xxx
30
The S/O turned in his seat to face the front of the airplane while addressing the
captain and FlO . No gestures other than body orientati on accompanied this
announcement.
12 .00 .4 3 S/ O
The S/O's opening announcement was a call for the attention of the other crew
members . The language the S/O used was explicit but indicated some uncertainty.
A fuel leak is a potentially flight-threatening situation and requires the immediate
and coordinated attention of all the crew members. After the S/O's announcement ,
the crew members collectively knew what the S/O suspected (a fuel leak) and
where he thought the problem was located (in fuel tank three). With that
information, the crew members prepared to attend to the problem.
Fig. 1.3. A frame from the video tape showing the 727 cockpit. The captain,
first officer, and second officer are all attending to the fuel panel that is on the
instrument panel at the bottom right.
31
assume that the crew members' mental models of the fuel system were similar
because they all received similar company training for the 727-200. Additionally,
they all have access to the same manuals that describe the operation of the fuel
system and fuel system procedures. Still further, typical airline career trajectories
start in the SIO position and move to the FlO position and then finally to
captain . In most cases all the crew members have had SIO experience.
A salient part of a pilot's understanding of a fuel leak is that it is a situation
that must be dealt with quickly. In response to the SIO's announcement, the
captain and the FlO turned in their seats to face the SIO and the SIO's panel
(Figure 3). Each of them also produced a contentless verbalization with a rising
intonation.
12.00.50
Capt
hrnmn
12.00.51
FlO
FlO
12 .00 .56
S IO
ohhhh
The very act of the captain and the FlO turning around to face the SIO and the fuel
panel indicates that they heard the SIO's announcement and realized that their
attention was needed. Once the captain and the FlO were situated, the SIO began
his explanation of the problem without further prompting. As the SIO spoke , he
turned in his chair to face the fuel control panel.
SPEECH
GES'IURE IN SPACE
right now
With the fuel system, there is always a question of whether what is observed is
really the behavior of the fuel system or if it is simply a gauge malfunction. The
SIO began by gesturing to (placing his finger on, but not depressing) the fuel
quantity test switch while saying "right now."
There was nothing in the S/O's words about the fuel quantity test button .
Pressing it in the context of a suspected fuel leak would have been a meaningful
action. But the S/O did not press it. He only touched it. We believe that the other
crew members interpreted this as an indication that the S/O had already tested the
gauges (in fact, he had) .
The words "right now" gave a sense of immediacy to the situation. They place
something in the present time, but what it is not yet clear. The speech and the
32
GES'IURE IN SPACE
The S/O next made a motion over the number three tank boost pump switches
that mimicked the motion used to turn the pumps off. The switches were already
in the off position . The combination of the gesture and the state of the panel and
the knowledge that boost pumps are normally on in flight made this action
unambiguous. The words redundantly expressed that which the gestures had
described, but the words also included information about temporal relationships
that cannot easily be represented in gesture. Speech marked the gestures as a
reenactment of what the S/O had already done. The verbal statement did not
indicate which pumps had been turned off, but the fingers did. The location of the
gesture in the space of the fuel panel resolved an ambiguous reference in the
verbal stream . The verbal component provided temporal markings that were
lacking from the gesture, and the gesture provided aspects of indexical reference
that were ambiguous in the S/O's words.
If the pumps were off, one may wonder where the fuel for engine three was
corning from. The topology of the panel facilitates certain inferences about the
functional behavior of the fuel system, and the S/O next moved to demonstrate
these inferences to the other crew members.
SPEECH
GES'IURE IN SPACE
I tried
The S/O changed topics at this point and his gesture directed attention to the other
side of the fuel panel where subsequent events would be described. He was now
beginning to explain how he established an alternative fuel source for engine
number three. The use of the past tense placed the action referred to in the past
with respect to the present course of action.
GESWRE IN SPACE
to feed from
number one
33
Here the gesture and the speech were almost completely redundant. The gestures
indicated the states of the controls that feed fuel from tank number one as the
fingers moved along the lines painted on the panel that depict the pipes in the
system that move fuel from the number one tank, through the boost pu mps ,
and to the engine one fuel feed valve.
SPEECH
GESWRE I N SPACE
to both
The S/O's hand lingered a moment near the controls for tank number one.
SPEECH
GESWRE IN SPACE
engine one
The S/O pointed to the area of the engine number three cross-feed valve and main
fuel supply valve while saying "engine one."
SPEECH
GESWRE IN SPACE
and three
In the brief statement, "I tried to feed from number one to both engine one and
three," the S/O explained that he had remembered to feed fuel to engine three
before he turned the tank three boost pumps off. The gesture accompanying this
section was complex and quickly executed. The S/O pointed to the tank one
gauge, to the tank one pumps (which were on), then to the engine three cross-feed
valve controller, and to the engine one cross-feed controller. These gestures drew
attention to the controllers that indicate that the valves were open and supplying
34
fuel to engines one and three from tank one. Some of the motions of the hand
also followed the flow of fuel through the system .
SPEECH
GES'IURE IN SPACE
but
Having established the alternate source of fuel for engine three, the S/O pointed to
the engine three fuel gauge. This was the locus of the problem. The S/O marked
with gesture a return to the topic of the fuel level in tank three and, with speech, a
return to the present tense. But signals a logical disjunction . The elements that
stand in disjunction are not yet clear but will be made clear by what follows.
SPEECH
GESTURE IN SPACE
we're
The S/O flicked the gauge with his finger. This is a common technique among
pilots to free a gauge needle that is believed to be stuck. From a strictly
functional point of view, this is a useless action. The S/O detected the fuel leak
by observing the rapid movement of the fuel gauge needle. The fact that it was
possible to detect the fuel leak is evidence that the needle is not stuck.
This flick was not performed in the S/O's original diagnosis and was not a
report of a previous action. Rather, it was a new action performed while the other
crew members looked on. Because this action was not functional, we might ask
what other kind of role it might be playing here . For one thing, it returned the
narrative to the temporal present. It was a way of emphasizing that the fuel level
shown by the number three tank gauge is the salient problem. At a more abstract
level of description, flicking a gauge is a way to produce an expected reading when
an unexpected reading has been encountered. In that sense, this action could also
be read as a assertion by the S/O that he would have liked the behavior of the
gauge to be other than it was.
SPEECH
GES'IURE IN SPACE
still losing
35
The S/O then emphatically gestured to the tank three gauge, while he said, "still
losing ." This last gesture drew attention away from the function of the needle
(that which the prior gesture, the flick, demanded) to the actual fuel quantity level
that the needle was indicating.
SPEECH
GESTURE IN SPACE
Finally, the S/O returned his hands to his lap indicating that his turn was
completed.
The Panel
First, the spatial organization of the panel is a central element of the usefulness of
the panel as a communicative resource. The spatial layout of the panel is
topologically (but not metrically) identical to the spatial layout of the fuel system
that it depicts. Table 1 shows correspondences between components on the fuel
panel and components in the fuel system.
36
IN THE SYSTEM
ON THE PANEL
Fuel tank
Fuel line
Quantity gauge
Painted line
Purrp control switch
Pressure indicator
Valve control switch
Position of valve control with
respect to painted lines
Punp
Pressure sensor
Valve
Fuel flow established
by valve position
The topological relations among panel components (e.g., the quantity gauges,
painted lines, and pump control switches) are the same as the topological relations
among the system components (e.g., fuel tanks, fuel lines, and pumps). The
actual mapping of the space of the fuel system onto the space of the panel
is complex. Components that are higher on the panel generally correspond to fuel
system components that are forward in the airplane. Components that are to the
right on the panel generally correspond to fuel system components that are on the
right of the airplane. But there are exceptions. The gauge and pump switches for
the rarely used aft auxiliary fuel tank have been placed out of the way so that they
do not interfere with the depicted relations among the main tanks and the engines.
The panel is further simplified by omitting depictions of check valves that cannot
be controlled from the panel.
The topology of the painted lines and switch positions creates a representation
that permits the crew to do conceptual inferences with simple and robust
perceptual skills. For example, figuring out where fuel will flow can be
accomplished by visually following lines on the panel. The valve controller has a
line painted on its top surface. When the controller is in the cross-feed position,
this line appears to connect the painted lines that depict the fuel line arriving at
and departing from the valve. The rotational action of the cross-feed valve
controllers, combined with the shape of the controller knob, makes the open and
closed states of the valve "look like" flow through or blocked flow. These may
seen to be trivial design features, but they have important cognitive consequences.
Imagine valve switches of a different kind (e.g., toggle switches) and a readout
that lists the name of the valve and its state in text format. With such a
representation, it would be impossible to use simple perceptual skills to reason
about the behavior of the system.
The simplified topology of the panel as a representation of the fuel system
itself permits the pilots to reason about the state and behavior of the fuel system
by "seeing" the panel in a particular way (C. Goodwin, this volume; Goodwin &
Goodwin, in press). The fuel system itself as a collection of physical components
cannot actually be seen from any real vantage point, but the pilots can "see" the
fuel system by seeing through the fuel panel. In fact, only through seeing fuel
panels and diagrams such as Figure 1 do pilots have any experience of the
topology of the fuel system. As with any materially instantiated symbolic
37
Meaningful Gestures
The gestures superimposed on the space of the panel can be read as meaningful
actions and courses of action on the fuel panel itself, or they can be seen as events
in the fuel system . Seeing each of these things requires a different stance with
respect to the panel. To see the gestures as actions on the panel, one must see the
panel as a panel. To see the gestures as representations of events in the fuel
system, the panel must be seen as the system that it represents .
The first meaningful gesture in this sequence is the SIO placing his finger on
the fuel quantity test switch. The fuel quantity test switch differs from all other
elements of the panel. All the other elements are in some sense "about" the fuel
system, but the fuel quantity test switch is "about" a set of components , the
quantity gauges, on the panel. This gesture must be read as being about the panel
rather than about the fuel system. In order for the captain and FlO to interpret the
SIO's gesture to the fuel quantity test switch, more than a shared understanding of
its function was necessary. It was not enough that they all have a similar model
of the switch's function. They needed to know that the others had a similar model
of the function as well. This kind of intersubjectivity underlies all of the
meaningful actions on the panel.
The procedure for diagnosing the fuel leak involves two distinct courses of
action. The first course of action establishes an alternate fuel supply for engine
three. The second course of action is to turn off the pumps in tank three and to
monitor the gauge for continued fuel loss. These courses of action were executed
in this order by the SIO before he notified the crew of the potential problem. The
explanation he gave of his action, however, interwove the two courses of action,
placing all of the second course of action in between the elements of the first.
We find it interesting that, although the order in which the actions are reported
is not the same as the order in which they were executed, it is the same order that
would be encountered in a traditional problem-solving account. The goal of
turning off the boost pumps for tank three cannot be accomplished directly
because it will cause engine three to flame-out (quit running) . This leads to the
creation of the subgoal of establishing an alternate fuel supply for engine three.
Once this has been accomplished, the pumps can safely be turned off and the
gauge monitored for further fuel loss.
The gestures acquire their meaning by virtue of being superimposed on the
meaningful spatial layout of the fuel control panel. The same gestures produced in
38
the absence of the panel would, of course, be quite meaningless. Enacted over the
panel, though, these gestures take on meanings such as "turning off the pumps"
and "the newly established path along which fuel is flowing ." The functional
consequences of the actions re-enacted in gesture by the S/O are easily seen by the
other members of the crew.
Speech
The verbal layer of the representation does things that cannot be done in the other
layers . For example, it uses tense markers and other linguistic devices to indicate
temporal relationships among actions . Gesture by itself is always action in the
present. The verbalizations place the actions in a temporal framework. This is
what makes it possible for the S/O's actions to be seen as a re-enactment of
action already taken rather than as a proposal for action to be taken.
Speech is also used to indicate the S/O's relationship to the actions and to
belief states derived from the actions. The use of personal pronouns is interesting
here. The S/O speaks of (1) his own state of knowledge, "I don't know," (2) a
condition shared by them all, "we must be losing it very quickly," and (3) a
relation between the captain and F/O to the shared condition, "you see, right
now ." Responsibility for actions and even for the flight in general are often
implicitly expressed in the use of pronouns in such settings.
In this excerpt, speech is used to control conceptual and temporal relationships.
Consider the words, "but we're still." There is a conceptual disjunction in the
"but" and a temporal disjunction in the "still." The conceptual disjunction marked
by the use of "but" is between the expectation that the level in tank three will not
decrease if the boost pumps are turned off (in normal operation) and the fact that
the level is decreasing. The temporal disjunction is between the past action that
should have put an end to the decrease and the present fact of continued decrease.
These disjunctions, together with the S/O 's action of flicking the gauge with his
finger , move the discourse back into action in the present time. In addition to
expressing the S/O 's relation to the gauge reading and shifting attention
momentarily from the fuel system to the panel , the flick gesture brings the
account back into the present tense. It is the only action taken on the panel during
the explanation. All of the other gestures depict an idealized set of movements
that the S/O 's hands might have made in doing the diagnosis.
Discussion
Does gesture support speech? Clearly it does, but no more so than speech
supports gesture . This example demonstrates the creation of a complex
39
Acknowledgments
Funding for the research described here was provided by grant NCC-2-591 to
Edwin Hutchins from the Ames Research Center of the National Aeronautics and
Space Administration in the Aviation Safety/Automation Program. Everett
Palmer served as technical monitor.
40
References
Goodwin, C. (1994a) . Professional Vision. American Anthropologist 96(3) , 606-633 .
Goodwin, C. (1994b). Seeing in depth: Space. technology and interaction on a
scientific research vessel. Unpublished manuscript. Univers ity of South Carolina,
Columbia, South Carolina.
Goodwin , C., & Goodwin, M. (in press) . Formulating Planes : Seeing as a situated
activity. In Y. Engestrom & D. Middleton (Eds.), Cognition and communication at
work. New York : Cambridge University Press .
Hutchins, E. (1991) . The social organization of distributed cognition. In L. B.
Resnick, J. M. Levine, & S. D. Teasley (Eds.), Perspectives on Socially Shared
Cognition (pp . 283-307) . Washington, DC: American Psychological Association
Press .
Hutchins, E. (1995). Cognition in the Wild. Cambridge, MA: MIT Press .
Hutchins, E. (in press). How a cockpit remembers its speeds . Cognitive Science.
Hutchins, E., & Klausen, T. (in press) . Distributed cognition in an airline cockp it. In
Y. Engestrom & D. Middleton (Eds.), Cognition and communication at work . New
York: Cambridge University Press .
Ochs, E., Gonzales, P., & Jacoby, S. (in press) . When I come down, I'm in the domain
state . In D. Slobin, J. Gherhardt, A. Kyratzis , & J. Guao (Eds.), Social interaction ,
social context, and language: Essays in honor of Susan Ervin-Tripp.
Chapter 2
Centers of Coordination:
A Case and Some Themes
Lucy Suchman
Xerox PARC, Palo Alto, California, USA
Abstract
This chapter identifies a class of worksites characterizable in terms of
participants' ongoing orientation to problems of space and time, involving the
deployment of people and equipment across distances according either to a
timetable or to the emergent requirements of a time-critical situation. To meet
simultaneous requirements of mobility and control, centers of coordination must
function as centers to which participants distributed in space can orient, and which
at any given moment they know how to find. At the same time, to coordinate
activities distributed in space and time, personnel within the site must somehow
have access to the situation of co-workers in other locations. One job of
technologies in such settings is to meet these requirements through the
reconfiguration of relevant spatial and temporal relations. This general
characterization is explored through ethnographic materials from an investigation
of the work of airline ground operations at a metropolitan airport on the west
coast of the United States.
Introduction
Recent work within anthropology and sociology includes a growing body of
ethnographically based analyses of the place of technologies in human activity and
42
Lucy Suchman
Centers of Coordination
43
The Site
From 1989 through 1991, we carried out an extended study of the work of ground
operations at a metropolitan airport on the west coast of the United States.' We
began the project with a general interest in contributing to analyses of the
dynamic structuring of people's interactions with each other and with their
material environments, and in exploring the relevance of such analyses to
problems in design.
We were initially drawn to the airport as a study site by two considerations.
First, as a workplace, an airport includes a diverse collection of professions and
activities, which in some respects occupy clearly delineated, largely separate
territories . At the same time, the coordination of the work requires interaction
across territorial boundaries at certain critical junctures. Second, during the course
of our study period, a new terminal building was completed and opened. This
made it possible for us to learn from the changes involved in moving operations
from one work setting to another. We were interested to see how the relations of
work and technology established in one facility would be reproduced, or
transformed, as they were re-established in a new setting.
Our study focused on the two airlines scheduled to move into the new terminal
and, within those airlines, on the work of ground operations: that is, all of the
work involved in servicing arriving and departing airplanes." Within ground
operations, we took as a further focus a particular "backstage" area, called the
Operations room, charged with coordinating the work of the gates and the ramp .
The Operations room proved to be ideal as a setting for our interests. Within the
room, we were able to see the working practices of a small group of people copresent to each other and closely attuned to each other's actions. At the same
time, the division of labor was such that each person within the room was
assigned the task of maintaining communications with some other relevant
location outside the room, via various technologies. We therefore were able to see
the structuring of these distributed, technology-mediated interactions as well.
Finally, we were able to see how these two working orientations - to co-workers
within the room and to those outside - were organized in relation to each other.
3The research, under the heading The Workplace Project, was funded by Xerox and
Steelcase Corporations. Brun-Cottan et al. (1991) presents a video final report on the
study.
4The work of ground operations does not include air traffic control, but rather all of the
activities involved in servicing an airplane while it is at the gate, for example, loading
and unloading passengers and baggage, ensuring that connecting baggage from an
incoming plane is moved to the appropriate outgoing plane , replenishing the plane 's
store of food and fuel. For analyses of air traffic control as a center of coordination, see
Harper and Hughes (1993) and Harper et al. (in press) .
44
Lucy Suchman
Themes
Our project, as analysts of the work of Operations, was to see just what that work
comprises within this particular site. At the same time, the materials from the
site contributed to a developing understanding of the social and material
organization of skilled practice within complex, technology-intensive worksites
more generally. The analyses that resulted from the project explore a set of themes
that both organized our looking at the site and were developed and elaborated by it:
Technologies as material practice: The inseparability of technologies and
the activities of their use. This includes locating the functionality of technological
artifacts not in particular devices, but in densely structured courses of action
involving the assembly of heterogeneous devices into a working information
system.
Reading a scene: How competency in these settings involves learning how to
read a scene, through the juxtaposition and interpretation of verbal reports, visual
images, and various forms of text, in real time, into provisional assessments of
an emerging situation .
(Re)producing a normal order: How, through their management of everyday
contingencies , workers are able to maintain an accountable spatio/temporal order.
Structures of participation: How participants in a multi-activity setting
structure their focus of attention and engagement from moment-to-moment.
Constituting workspaces: How workspaces are dynamically configured
through interactions across visible and invisible boundaries .
Acquiring competency: How the identity of competent practitioner is
acquired through progressive rounds of increasingly demanding work, supported in
situ by experienced co-workers.
Authoritative knowledge: Relations between participants' access to
technologies and the distribution of knowledge taken to be consequential for the
work at hand.
Designing for change: Implications of the analysis for relations between
professional design and design-in-use.
Centers of Coordination
45
46
Lucy Suchman
Along one wall of the room is a row of video monitors that feed images from
cameras located at each of the airline 's gates into the ro om. By seeing the image
of an airplane in a particular mon itor, mapping the monitor to a gate, and
mapping the gate in turn to a flight number , workers can track the status of
arriving and departing flights.
The association of gates with flights is supported through additional
representational devices. Located just below the row of monitors is the complex
board. The complex board is a whiteboard on which is drawn a space/time grid,
each cell of which is filled with an arriving and departing flight number, along
with its point of origin or destination. As Latour has argued (1990), the creation
and use of a device like the complex board involves the alignment, in a twodimensional, manipulable array, of a number of different spatial relations,
temporal orders and heterogeneous objects distributed across great distances.
Through its matrix structure, the complex board provides a graphic surface on
which operations workers can juxtapose and relate gates, times , flight numbers,
and other sites in the airline network . The further association of a clock then
makes it possible to relate the order represented on the board to the observable
order that can be viewed in the video monitors. Finally, computer terminals
provide access to a network of databases shared among the distributed sites of the
airline 's operations , representing the scheduled order from a variety of views, and
continuou sly updated to reflect the actual course of unfolding events.
Centers of Coordination
47
Reading a Scene6
A central finding of our analyses concerns the extent to which the work of
Operations involves the assembly of knowledge about past, present, and future
events through the juxtaposition and relationship of a diverse range of
technologies and artifacts. Access to information and its timely communication to
relevant others involves interaction not with a single technology but rather with
multiple technologies (e.g., forms, computer screens, video monitors) held in
relation to each other and read off in ways specifically structured by the task at
hand.
The reading of these technologies does not proceed in a unidirectional sequence:
rather, they can be juxtaposed as needed in order to add to or elaborate on whatever
information already happens to be in hand (see M. Goodwin, 1995; Goodwin &
Goodwin, in press .) Workers address inquiries to these information resources not
from some neutral starting place, but always from a position within the midst of
an ongoing situation. So, for example, experienced Operations room workers,
knowing the schedule for a particular flight , can locate the associated plane on a
video monitor and read the activities around it for the flight's progress.
Alternatively, with a given plane in view on the video monitor, they can associate
that plane with a particular flight in order to make sense of the image they see.
Competency in the Operations room involves learning how to read a scene
through the juxtaposition and interpretation of verbal reports , visual images , and
various forms of text, in real time, into provisional assessments of an emerging
situation.
The example that follows is drawn from a chapter by Charles and Marjorie
Goodwin (in press) where it is extensively and elegantly analyzed. It concerns the
work of a position in the Operations room called the Flight Tracker. The Flight
Tracker is charged among other things with tracking arriving and departing
airplanes, maintaining communications with pilots on the ground via a radio, and
clearing their arrival at designated gates.
On the simplest account, the incident begins with a report from an incoming
pilot that another plane is already parked at the gate to which he has been
assigned . The report implicitly identifies a problem, and the question locates the
solution to that problem with the Flight Tracker":
6This phrase is due to Goodwin and Goodwin (in press). "Reading a scene" is closely
related to what Heath and Luff (1992) describe under the heading of "overseeing the
local environment" (p. 83).
7In transcript segments, colons U:" indicate prolongation of the immediately
preceding sound; italics mark stress . A dot in paretheses "(.)" indicates an untimed
pause ; numbers in parentheses indicate elapsed time in seconds . An equal sign "="
indicates "latching," that is, the beginning of one utterance following directly on the
end of the prior utterance with no gap.
48
Pilot :
Flight
Tracker:
Lucy Suchman
I understand gate fourteen is occupied?
Do you have any instructions for (it)?
Uh : . m,
(0 .3)
(0 .1)
Pilot :
In their analysis of this sequence, the Goodwins observe that, to understand the
skills that the Flight Tracker deploys in doing her work, we need to begin with
her actions as hearer during the course of the pilot's calL We can see this clearly
by considering a transcript of the same sequence, this time with an indication of
the Flight Tracker's orientation during the exchange with the pilot, as indicated by
her gaze:8
Pilot :
Flight
Tracker:
(0.3)
back
to
video
monitors]
Hopefully: ,
(1.0)
Pilot :
8Charles and Marjorie Goodwin among others have developed innovative means of
indicating non-vocal activity within transcripts (e.g ., see Marjorie Goodwin's
transcription of this sequence in Goodwin & Goodwin, in press, and also Heath & Luff,
1993, and Luff & Heath, 1993). Here I have adopted the simple strategy of inserting
textual descriptions of the Flight Tracker's orientation within brackets, in boldface,
into the transcript.
Centers of Coordination
49
With the pilot's mention of gate, the Flight Tracker is already orienting to the
video monitors . In her look to the monitors , we see her beginning a course of
action that involves the juxtaposition and relation of multiple, partial perspectives
onto the scene. Specifically, she finds the location referred to by the pilot as it
appears on a video monitor within her work site, maps the plane that she sees
there to an associated flight number by means of a paper record of scheduled
arrivals and departures, compares the observable plane and current time with the
represented status of that flight in the record and on a flight information display
screen in front of her, finds an unreconciled disparity between actual and
represented events, gazes again at the monitor, and there finds the actual situation
to have changed in such a way (the workers on the ramp having pulled the
passenger stairs away from the plane) that she can project a resolution (the plane's
imminent departure .) Her vantage point in operations does not provide her with a
single master perspective, in sum, but rather with a range of partial information
resources with which she can assemble a coherent view.
50
Lucy Suchman
they are engaged. So we see here that the Flight Tracker makes recourse to the
schedule and to an expected order of events at the gate, while at the same time she
implicitly acknowledges the kind of order that it is. Specifically, she notes in her
hopefully and in her use of the modal should that there is no necessary relation
between schedules and the timely movement of airplanes, or between stair-pulling
and momentary departure. Rather, the normal order is contingent on its faithful
reproduction through the artful practices of personnel dedicated to its accomplishment, on each next occasion and across the unique and unpredictable circumstances
that each occasion provides . It is on that artful accomplishment that the Flight
Tracker's expectation of a projected resolution to the problem relies .
Jordan (1992c) looks at the management of trouble with an interest in the
relation between routine and improvisation in the work of Operations. In
particular, she develops the notion of a typified action sequence as a way of
talking about the projectability that organizational routines provide . Her analysis
demonstrates that the reproduction of organizational routines is not provided for
from outside but rather is a continuous accomplishment from within the local
group . At the same time, within the group, members orient to a normal course of
events, as the framework into which unexpected contingencies can be absorbed and
through which they can be managed.
M . Goodwin (in press) explores in detail the workings of such action sequences
as a matter of resources used by Operations workers to design their talk for the
specific environment that the room affords, in particular, for the troubles that it
presents for hearers . Taking as a case in point the routine delivery of informings
and announcements, Goodwin's analysis shows how workers employ various
forms of stylized intonation reminiscent of ritual speech in order successfully to
convey these utterances through the "sonic soup" from which their intended
recipients must pick them out.
Suchman (1993) takes up the theme of the accountability of practical action and
relates it to the use of technologies within the work of Operations. The phrase
technologies of accountability is meant to take advantage of the term's ambiguities to suggest two senses in which we might pursue the investigation of
technologically mediated practice. In the first sense, the competent use of
technologies in Operations supports the production of an accountable course of
intelligible and effective action. In the second sense , the technologies and actions
of Operations incorporate regimes of control designed to administer a particular,
institutionalized spatio-temporal order.
Structures of Participation
We noted that the social organization of work within the Operations room
involves a division of labor that assigns to each position different responsibilities
for communication, via various technologies , with other relevant locations. One
consequence of this organization is a differentiated structure of attention among
Centers of Coordination
51
workers within the room. The most compelling evidence for this differentiation is
the ability of participants to disattend sights, sounds, and events that draw the
attention of their co-workers , and to keep on with their business at hand. At the
same time, members of the operations room are engaged in the highly integrated,
joint task of getting planes smoothly in and out of gates according to specified
schedules. This collective responsibility means that they are in some very
important sense "in it together," both in the sense of sharing responsibility for
the work's success or failure and sharing the facilities through which the work
gets done. The coordination required within the center means that Operations room
work is characterized by a strong mutual orientation among co-present workers to
each other and to developing situations .
The interactional order of the Operations room is characterized as well by a
continuous state of incipient talk. The peripheral monitoring made possible by
their co-presence in a shared auditory and visual space means that at any time
something overheard in the work of another may be assessed for its relevance and
taken up as the business of the hearer. So, for example, a question asked "of the
room" or even of a specific co-worker may be answered by anyone who has an
appropriate response (Jordan, 1992c). Similarly, people are oriented to the
possibility that they or their co-workers may initiate an interaction at any time
without any marked pre-announcement or inquiry into the others' availability for
that talk (Brun-Cottan, 1991). The multi-activity nature of the setting means that
the one who would initiate an interaction needs to do so in a way that is sensitive
to the engagement of others, for example, through the utterance's placement or, as
an alternative, through its intonation (M. Goodwin, in press).
Another sequence analyzed by Marjorie Goodwin (1995) involves an incident
similar to that of Goodwin and Goodwin (in press); that is, an arriving pilot has
been assigned to a gate that is already occupied. Again we see the Flight Tracker
(Fl') receive a call from the pilot and consult the video monitors and complex
board to establish, first, that there is an airplane at the assigned gate, and then its
flight number and scheduled departure time. In this case, however, the projected
resolution of the problem is provided not by what the Flight Tracker can read of
the activities around the plane but by a reading done for him by the Ramp Planner
(RP), a co-worker in the Operations room responsible for directing activities
around the plane . The Ramp Planner is retrospectively found to be attending to
the call-in from the pilot of Flight 1091 and to the Flight Tracker's resulting
problem:
Pilot :
RP :
52
FT :
RP:
FT:
Pilot :
FT :
RP :
FT :
Lucy Suchman
[Looks to monitors]
(To pilot vi a radio ) Roger, ten ninety-one .
Charlie- Alpha : seven:? uh : : :, (0 .2) Shoo : :
(0.9)
Hold on one second, ten ninety-one .
(Off radio, into the room) Alpha seven,
[Shifts gaze from monitors to complex board]
That plane should be pu shin.
[Re-orients back to radio] (To pilot) That aircraft
should be off the gate shortly .
St and by until seven clears, Ten ninety-one .
Roger . Could you tell them we're gonna need
ground power please .
That's affirm . [Gaze toward RP] Did you catch that,
Joe?
Yeah .
Okay .
In this case, it is the Flight Tracker who, in response to the radio call from the
pilot of Flight 1091, discovers, through a look to the video monitors, that
another plane is occupying the gate to which Flight 1091 has been assigned. The
Ramp Planner evidently sees this problem as well. Moreover, the division of
labor in this case provides the Ramp Planner with a different history and
consequently different ability to interpret the state of affairs at the gate; this
difference is a resource here because he, like the Flight Tracker in the previous
example, offers a projected solution based on his reading of the scene. The Flight
Tracker, in turn, evidently reasons that the Ramp Planner, having just displayed
his attention to the call, will have continued to monitor its course and will hear at
its close a request from the pilot for an electrical power source. He figures that,
but he is not sure. Given this uncertainty, the Flight Tracker fulfills his
responsibility to pass on the pilot's request to the Ramp Planner not by
delivering it as news, but by asking the Ramp Planner to confirm that it has, in
fact, been heard.
Workers in the Operations room are sensitive to the environment for
communication that the room provides and to changes in that environment from
one moment to the next. Much of the time, workers are able to track the activities
of their co-workers or to be enlisted into them at the same time that they respond
to the demands of their own position . By listening not only for events to which
they need to respond but for the responses of others as well, minor troubles in
communication are routinely identified and resolved? In some situations, the
supervisor effectively acts as an additional pair of eyes and ears, charged
specifically with watching for the possibility that events may be missed (see
Suchman, in press) .
9See Heath and Luffs discussion (1992) of surreptitious mon itoring and rendering
activities visible for a related analysis .
Centers of Coordination
53
Constituting Workspaces
54
Lucy Suchman
Acquiring Competency
Our interest throughout the project was to analyze interactions within Operations
not by recourse to a master plan or external order, but rather with reference to the
activity's ordering from within, through the sensitivities of the participants to
each other and to their joint situation . However, to say that social order is
produced from within, through the local interactions of participants, is not to say
that it is produced without any reference to previous activity . One way in which
we can see the presence of a historically and culturally constituted community
into which people come is in the process of learning the work (Lave & Wenger,
1991) . We can take as an example a routine task within ground operations: that
is, the work of establishing what are called an airplane's weights and balances.
Weights and balances involve a relation between the plane's total complement of
passengers, baggage, fuel, and other cargo and the settings required for the
stability of the aircraft (e.g., the wing flaps). On many newer aircraft, the final
weights and balance calculation is taken by the pilot directly from a computer in
the cockpit. For older aircraft, however , the numbers are called up on a computer
screen in the Operations room, then radioed out to the pilot in an exchange called
a radio close-out, a routine part of the Flight Tracker's work.
Charles Goodwin (1991) analyzed a case involving an apprentice in the work of
Flight Tracking doing her first radio close-out , supported by an experienced coworker. In his analysis, Goodwin draws attention to the exquisite coordination of
apprentice and teacher in the course of the call. Specifically, he shows how the
more experienced co-worker momentarily "redesigns" the computer screen for the
apprentice in response to her demonstrated difficulty in finding her way through,
by directing her reading of it with his finger (see Figure 2). Goodwin's analysis
10Por example, the constructs recipient design (M. Goodwin, 1980; C. Goodwin,
1981 ; Schegloff, 1972) and part icipation structures (Goodwin & Goodwin, 1989)
underwrite almost all of our analyses. (See also Atkinson & Heritage, 1984; Heath,
1986 .)
Centers of Coordination
55
shows how, through their precise attunement to each other's place in the course of
action and its possible problems, apprentice and teacher together achieve a
successful reading.
Authoritative Knowledge
Through a comparative analysis of an American obstetrics ward and the
Operations room , Jordan (1992c) explores the relation between access to and
control over technologies, and the forms of knowledge taken as authoritative: that
56
Lucy Suchman
is, as "grounds for legitimate inference and action" (p. 1) within a given situation .
In contrast to the highly specialized and power-differentiated order of the hospital
setting, she finds familiarity with relevant technologies to be distributed across
participants in Operations, with a corresponding distribution of knowledge
required to get the work done.
As an example, Jordan traces the work on a particular afternoon in Operations
required to effect a switch of airplanes between gates, in order to enable
mechanical repairs to one of the planes. This involves towing an airplane that has
arrived at one gate to a different gate for departure, as well as a complex reassignment of crew and transfer of passengers and baggage. Jordan points to the
relatively "horizontal" distribution of knowledge with respect to the work
required to accomplish this rearrangement, including the use of relevant
technologies. She argues that this distribution of knowledge is both produced by
and helps to explain the frequency of "out louds," or apparently undirected
comments on the situation, and questions asked "of the room," over the
maneuver's course (see also M. Goodwin, in press). That is, workers recognize
more or less implicitly that information about events may be relevant to anyone
or more of their coworkers at any given time, just as one or more of their
coworkers may have answers to a question, given their distributed access to
available information resources and social networks. Jordan concludes by
proposing an agenda of comparative analysis of the distribution of authoritative
knowledge across different settings, involving differently privileged or accessible
technologies.
Our findings on relations of work and technology imply that , in order to design
anyone aspect of a working order, one needs to understand that aspect's
relationship to the extended system of activities and technologies of which it is a
part. At the same time, products of professional design will always be based in
partial, specifically situated and historically constituted projections of the
circumstances of an artifact's use. As a consequence, professional design needs to
be understood not as an end point but as a starting place, or platform, for the
ongoing processes of design-in-use that are both inevitable and necessary for an
effective working environment.
The specific processes of change that we observed in the Operations rooms
illustrate more general characteristics of workplace design . Rather than developing
according to a single master plan, changes in facilities and work practices arise
from participants' emerging and, to some extent, idiosyncratic appreciation for the
problems and possibilities that a particular setting affords. First impressions of a
facility change as one goes to work in it. Requirements are discovered through the
contingencies of everyday use.
Centers of Coordination
57
Conclusion
The analyses reported here draw on a rich set of materials that make evident the
dynamic , moment-to-moment structuring of coordinated work activities in a
complex, distributed, technology-intensive workplace. The accumulation of
studies that locate regularities of everyday interaction within specific sites of work
practice provide the materials from which richer tapestries of research and
theorizing can be woven . Moreover, others engaged in the study of what I am
calling here centers of coordination have proposed generalizations that intersect
with those that I have presented. There is every indication that comparative
analysis across such sites would more than repay our efforts . For example, we
might compare and contrast workers' relative expectations regarding the identity
and circumstances of callers into the center in the cases of the Operations room
and a 9-1-1 public safety center, and the consequences for the problem of turning
calls into organizationally relevant objects (Whalen, 1992, 1993; Whalen &
58
Lucy Suchman
Acknowledgment
I am grateful to Charles Goodwin, Marjorie Goodwin, and Randy Trigg for their
reading and comments on earlier versions of this paper. Lauren Resnick and
Clotilde Pontecorvo provided further thoughtful review and suggestions .
IlThanks to Randy Trigg for suggesting the interest of this particular line of
comparison .
Centers of Coordination
59
References
Atkinson, M., & Heritage, J. (Eds .), (1984). Structures of social action. Cambridge,
England: Cambridge University Press.
Barley, S. (1986) . Technology as an occasion for structuring: Evidence from
observations of CT scanners and the social order of radiology departments .
Administrative Science Quarterly, 31, 78-108 .
Blomberg, J. (1987). Social interaction and office communication: Effects on user's
evaluation of new technologies. In R. Kraut (Ed.), Technology and the
transformation of white collar work (pp. 195-210). Hillsdale, NJ: Lawrence
Erlbaum Associates .
Bowers, J. (1994). The work to make a network work. In R. Furuta & C. Neuwirth
(Eds .), Proceedings of the Conference on Computer-Supported Cooperative Work
(pp. 287-298) . New York: ACM Press.
Brun-Cottan, F. (1991) . Talk in the workplace : Occupational relevance. Research on
Language in Social Interaction, 24, 277-295 .
Brun-Cottan, F., Forbes , K., Goodwin , C., Goodwin, M., Jordan, B., Suchman, L., &
Trigg, R. (1991) . The workplace project: Designing for diversity and change
(Videotape). Palo Alto, CA: Xerox Research Center.
Clark, A., & Fujimura, J. (Eds.). (1992). The right tools for the job: At work in
Twentieth-Century life sciences. Princeton, NJ: Princeton University Press.
Dugdale, A., & Fujimura, J. (Eds.). (In prep) . Making sex, fabricating bodies: Gender
and the construction of knowledge in the biomedical sciences.
Filippi, G., & Theureau, J. (1993) . Analyzing cooperative work in an urban traffic
control room for the design of a coordination support system. In G. Michelis, C .
Simone, & K. Schmidt (Eds.), Proceedings of the Third European Conference on
Computer-Supported Cooperative Work (pp. 171-186) . Dordrecht, The Netherlands :
Kluwer.
Forsythe, D. (1993). Engineering knowledge: The construction of knowledge in
artificial intelligence. Social Studies of Science, 23, 445-77 .
Garfinkel, H., & Sacks, H. (1970). On formal structures of practical actions. In J .
McKinney & E. Tiryakian (Eds.), Theoretical Sociology (pp. 337-366). New York:
Appleton-Century Crofts.
Goodwin, C. (1981) . Conversational organization: Interaction between speakers and
hearers. New York: Academic Press.
Goodwin, C. (1991) . Situated literacy in the workplace. Colloquium, University of
California, Los Angeles, Department of Applied Linguistics, February 21, 1991.
Goodwin , C. (1994) . Professional vision . American Anthropologist, 96(3), 606633 .
Goodwin , C., & Goodwin, M. (1989) . Context, activity and participation. In P. Auer
& A. diLuzo, (Eds.), The contextualization of language (pp. 77-99) . Amsterdam:
Benjamins.
Goodwin, C., & Goodwin, M. (in press) . Formulating planes : Seeing as a situated
activity . In Y. Engestrom & Middleton, D. (Eds.), Communication and cognition at
work. New York : Cambridge University Press.
60
Lucy Suchman
Centers of Coordination
61
Jordan, B. (1992b). New research methods for looking at productivity in knowledgeintensive organizations. In H. V. D. Parunak (Eds.), Productivity in knowledgeintensive organizations: Integrating the physical, social, and informational
environments (pp . 194-216). Ann Arbor, MI: Industrial Technology Institute.
Jordan, B. (1992c). Technology and social interaction: Notes on the achievement of
authoritative knowledge in complex settings (IRL Technical Report No. IRL920027) . Palo Alto, CA: Institute for Research on Learning .
Knorr-Cetina, K., & Mulkay, M. (1983) . Science observed. London: Sage.
Latour, B . (1987). Science in Action. Cambridge, MA: Harvard University Press.
Latour, B. (1990) . Drawing things together. In M. Lynch & S. Woolgar (Eds.),
Representation in scientific practice (pp. 19-68). Cambridge, MA: MIT Press.
Latour, B., & Woolgar, S. (1979). Laboratory life. Beverly Hills, CA: Sage.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation.
New York: Cambridge University Press .
Luff, P., & Heath , C. (1993) . System use and social organization: Observations 0 n
human-computer interaction in an architectural practice. In G. Button (Ed .),
Technology in working order: Studies in work, interaction and technology (pp .
184-210). London: Routledge.
Lynch, M., Livingston, E., & Garfinkel, H. (1983) . Temporal order in laboratory
work . In K. Knorr-Cetina & M. Mulkay (Eds.), Science observed (pp. 205-238).
London: Sage.
Lynch, M., & Woolgar, S. (Eds.). (1990). Representation in scientific practice.
Cambridge, MA: MIT Press .
Orr, J. (1990) . Talking about machines: An ethnography of a modem job. Doctoral
Dissertation, Cornell University. (PARC Technical Report SSL-91-07 [P9100132]). Palo Alto, CA: Xerox PARC o
Pickering, A. (Ed.) (1992) . Science as practice and culture. Chicago : University of
Chicago Press.
Schegloff, M. (1972) . Notes on a conversational practice: Formulating place . In D.
Sudnow (Ed.), Studies in social interaction (pp. 75-119) . New York: Free Press .
Sharrock, W., & Anderson, B. (1993) . Working towards agreement. In G. Button
(Ed.), Technology in working order: Studies in work, interaction and technology
(pp. 149-161). London : Routledge.
Star, S. L. (1989) . Regions of the mind: Brain research and the quest for scientific
certainty. Stanford, CA: Stanford University Press .
Star, S. L., & Ruhleder, K. (1994) . Steps towards an ecology of infrastructure. In R.
Furuta & C. Neuwirth (Eds.), Proceedings of the Conference on Computer-Supported
Cooperative Work (pp. 253-264). New York: ACM Press .
Suchman, L. (1983) . Office procedures as practical action: Models of work and system
design. ACM Transactions on Office Information Systems, 1(4), 320-328.
Suchman, L. (1987) . Plans and situated actions: The problem of human-machine
communication. Cambridge, England : Cambridge University Press .
Suchman, L. (1993) . Technologies of accountability. In G. Button (Ed.), Technology
in working order: Studies of work, interaction and technology (pp. 113-126).
London : Routledge.
Suchman, L. (in press) . Constituting shared workspaces. In Y. Engestrom & D.
Middleton (Eds.), Communication and cognition at work. New York : Cambridge
University Press .
62
Lucy Suchman
Suchman, L., & Trigg, R. (1991) . Understanding practice : Video as a medium for
reflection and design . In J. Greenbaum & M. Kyng (Eds.), Design at work :
Cooperative design of computer systems (pp. 65-89) . Hillsdale , NJ: Lawrence
Erlbaum Associate s.
Suchman, L., & Trigg, R. (1993) . Artificial intelligence as craftwork. In S. Chaiklin
& J. Lave (Eds.), Understanding Practice (pp. 144-178). New York: Cambridge
University Press.
Suchman, L., & Whalen, J. (1994, November) . Standardizing local events and
localizing standard forms. Paper presented at the annual meetings of the Society for
Social Studies of Science, New Orleans, LA.
Traweek, S. (1988) . Beamtimes and lifetimes : The world of high energy physicists .
Cambridge, MA: Harvard University Press .
Whalen, J. (1992) . Technology and the coordination of human activity : Computeraided dispatch in public safety communications. (Unpublished report) . Eugene, OR:
Univers ity of Oregon , Department of Sociology .
Whalen , J. (1993, August) . Accounting for "standard" task performance in the
execution of 9-1-1 operations. Paper presented at the annual meetings of the
Ameri can Sociological Association, Miami, FL.
Whalen , M., & Zimmerman, D. (1990) . Describing trouble: Practical epistemology in
citizen calls to the police . Language in Society, 19, 465-492 .
Yates , J. (1989) . Control through communication : The rise of system in American
management. Baltimore and London: Johns Hopkins University Press .
Chapter 3
Animated Texts:
Selective Renditions of News Stories
Christian Heath' and Gillian Nicholls'
'University of Nottingham and King's College London, UK
2Middlesex University, UK
Abstract
The following chapter is concerned with the ways in which journalists, working in
an international news agency, coordinate the production of news stories with each
other . In particular, it explores how journalists animate stories on which they are
working and thereby render their own activities visible to colleagues within the
newsroom. Although the analysis is princip ally concerned with the ways in which
textual embedded stories are voiced and shaped in and through talk and interaction,
it also considers , more generally, how tools and technologies feature in everyday
collaborative work .
64
Introduction
In recent years, we have witnessed the emergence of a growing corpus of
naturalistic research concerned with the organization of work and communication
in complex technological environments. Although, in part, such research has
emerged in the light of rapid technological developments and the transformation of
the workplace, it has also been driven by a sophisticated critique of more
traditional analytic orientations to the interaction between the individual and the
computer (Winograd & Flores, 1986; Suchman, 1987). In particular, the growing
dissatisfaction with conventional plan-based, goal-oriented models of human
conduct and its concomitant assumptions concerning the nature of human
cognition and action have led to an interest in exploring the in situ organization
of work and in explicating the tacit and indigenous resources through which
individuals accomplish and coordinate their tasks and activities.
Despite the contributions of such research and, in particular, their substantial
body of findings concerning work and communication in settings such as air
traffic control (Harper & Hughes, 1993), ship navigation (Hutchins, 1990),
airport operation centers (Suchman, 1993; Goodwin & Goodwin , 1996),
emergency dispatch centers (Whalen, 1995), urban transport control rooms (Heath
& Luff, 1992), and financial dealing rooms (Heath, Jirotka, Luff, & Hindmarsh ,
1995), we still have relatively little understanding of the ways in which
technologies "mediate" the interaction between organizational personnel. It is clear
from these studies that new technologies not only reconfigure tasks and the
division of labor within the workplace, but also introduce unanticipated
asymmetries to the communication between organizational personnel. Although it
has long been recognized that working environments are founded on and embody a
wide variety of forms of communicative asymmetry, especially between
organizational personnel and their clients (see, for example, Markova & Foppa,
1990; Drew & Heritage, 1992), the ways in which real time, co-present
interaction is configured to manage differential access to specialized information
and tasks (accessed and performed through new technologies) remain to be
explored.
Take , for example, a document such as a medical record, which is held in
electronic form. During the course of the consultations, it is not unusual to find
patients attempting to coordinate their actions, such as utterances, with the
doctor's use of the computer as he enters data into the record (cf. Heath, 1986;
Greatbatch, Luff, Heath, & Campion, 1993). The patient may be unable to to see
the screen and the text itself and may only have a passing understanding of the
sorts of data that the doctor may be entering, such as diagnostic and prognostic
information. All the same, we find patients attempting to infer the character and
progression of the activity from such "visible" elements of the activity as the
rhythm of the key stokes , the movement of the hands, and slight shifts in the
doctor's gaze . These forms of peripheral sensitivity to the activities of others and
Animated Texts
65
The Setting
A number of international news agencies provide real time, screen-based
information to the financial sector as well as to other customers, including
television companies and newspapers. These include Blomberg, Nightrider, and
Reuters. Reuters is by the far the largest concern and has the most customers. It
has offices in most major cities throughout the world and coordinating centers in
London, Tokyo, and NewYork. In London, the Financial News Section of Reuters
is divided into four desks, each with its own editor, journalists, and subeditor(s).
These desks are Money and Capital, Equities, Oil and Minerals, and
Commodities. The desks are positioned near each other in a large open plan office.
The desks receive stories from the various offices throughout the world. The
stories are coded for a particular desk and appear within a "basket" on the editorial
screen. Journalists take stories from the basket, check their address and topic
coding, and edit the contents of the headline (the Header) and the story. The
corrected story is then transmitted or, in the case of longer pieces (say a couple of
pages or more), passed to the subeditor for final checking.
Although individual journalists on the various desks have a fair amount of
discretion in coding, editing, and prioritizing stories, and largely work on
particular items individually, they are sensitive to the interests of colleagues both
on their own desk and those working in other areas. For example, a news story
addressed to Money and Capital might also be relevant to the customers served by
66
the journalists in Equities . In many cases, different desks will transmit different
versions of the same story, even though the original text was only received by
one desk. At the morning editorial meetings, which review the previous day's
work and discuss the more important stories of the upcoming day, it is not
unusual to hear critical remarks concerning the failure of certain desks to pick up
on, or inform, others of potentially relevant stories . A case in point was a story
about a press release from Kleeneze concerning innovations in battery technology.
It was felt that the story should have been picked up by the Oil and Minerals desk
because the news had potentially important implications for lead consumption and
sales . It is important, therefore, that journalists remain sensitive to the interests
of colleagues working on other desks. They have to put their mind to the concerns
and relevancies of others (cf. Sacks, 1992) and to inform colleagues of incoming
news that might be relevant to their particular customers . It is also important,
however, that, given the enormous traffic in news stories handled by the various
desks, journalists do not bombard their colleagues with the details of all stories
they receive that might have some implications for customers handled by other
desks. Journalists need to discriminate among news stories with respect not only
to their own readers but also to the readers served by their colleagues. They need
to inform colleagues of potentially relevant stories but not overload them with the
information that they are receiving.
Journalists need to remain sensitive to the relevance of their own stories for
colleagues on other desks and to listen for potentially interesting items being
handled elsewhere; but they also have to work closely with colleagues in their
own area. For example, in passing stories on to subeditors, they might need to
provide some instructions concerning the ways in which the story should be
handled, or, in editing a particular item, it might be important to inform
colleagues of how the news is potentially relevant to other stories in the same
basket. Or, for example , decisions to make major changes to a story, its priority,
or topic coding may often be done in consultation with colleagues . A case in
point is spiking, a term still used in the electronic office. Journalists often talk
through stories with colleagues on the same desk before deciding to veto an item.
There is a strong orientation to keep colleagues informed, however indirectly, of
the stories on which you are working and to .let others know of the more
consequential decisions that you might be taking. In this way, while dealing with
a substantial amount of stories during the day, colleagues keep tabs on the more
significant items that are being handled by colleagues both on their desk and
within the news room. Thus Reuters provides a news service that is timely,
coherent, and reliable. Given the substantial amount of news pouring into London
during the day and being handled by the four desks, this is no mean achievement.
On the one hand, journalists work individually, handling and editing stories
received by the desk. On the other hand, they have to remain sensitive to the
interests of colleagues both on their own desk and on other desks. They need to
provide their colleagues with a sense of the work in which they are engaged and in
particular any potentially relevant stories ; but this needs to be accomplished
Animated Texts
67
Reading Aloud
It seems to be a feature of many co-present working environments in which there
is a sharp division of labor, yet a strong interdependence of tasks, that personnel
take it upon themselves to openly voice bits and pieces of information that may
be relevant to various activities within the domain . Although calling out and
shouting, perhaps best exemplified in the trading rooms of financial institutions
(cf. Heath et aI., 1995), appear, at least initially, to be addressed to all those
within audible range of the speaker, it soon becomes clear that such objects are
recipient selective or at least designed to be heard by particular categories of
personnel within the ecological domain . All the same, such objects are rarely
addressed to particular personnel; they do not demand a response or even that the
object is acknowledged; rather the information is there if it is needed or relevant.
By openly voicing, calling out various pieces and bits of information, speakers
are able to respect the current commitments of personnel within the domain while
providing information that may have some significance. Indeed, in contrast to the
impression of visitors to such environments, often overwhelmed by the shouting
and noise, it is in part the very fact that information is voiced in this way, that
allows personnel to retain some semblance of concentration and distance from
each other.
At Reuters, especially when major news stories are breaking, we find the sorts
of open voicing and calling out that arise in related environments in other
organizations. For the most part, however, information needs to be made available
among journalists working in a particular area or those based at adjoining desks.
Much of this information arrives within the agency, not through the telephone, as
it does in dealing rooms, control centers, and dispatch services, but in particular
baskets that are accessed on screen through the Reuters editorial system . For the
journalists, therefore, it is largely textually embodied stories that may be of
relevance to the work of colleagues . So while journalists are editing particular
68
stories, it is not unusual to find bits and pieces of news information being made
available to colleagues within the local milieu.
The fragments that follow are simplified examples of the original transcripts.
Where necessary, features of the original orthography are included, using a system
developed by Gail Jefferson. Colons indicate that the preceding sound is stretched,
underlines that the word or part of the word is emphasized, and numbers in single
brackets represent pauses or silences in tenths of a second. A full stop with
brackets represents a mini-pause, roughly a tenth or two tenths of a second.
Details of the orthograghy can be found in Atkinson and Heritage (1984)
and numerous other monograghs and collections on conversation analysis.
In the first instance , Peter makes a joke of a story he is editing on screen:
Fragment 1, Transcript 1
Peter :
Peter :
Peter :
In making his comments on the interest rate fall, Peter adopts a pronounced
Jewish accent. The remarks are not addressed to colleagues on his own desk
(Money and Capital) or to those on the adjoining desk (Equities). While talking
aloud, Peter looks at his monitor and continues to edit the story.
Peter's talk animates the story. It voices the text, or at least elements of the
story, and gives it the character of a jo ke. Peter' s remarks are loud enough to be
audible to colleagues on the surrounding desks, yet they are not addressed to any
particular colleague. The story is rendered visible, but does not demand or even
implicate a response. The way in which it is articulated avoids generating
sequential import; that is, no particular person is placed under obligation to
respond to the utterance. Even the question, "Didn't it do this last week,"
elaborates the joke and is not treated as sequentially implicative. The gist of the
story is voiced, but it does not place anyone within its audible range under an
obligation to respond.
Twelve seconds or so later, Alex, who is sitting some six feet away at the
Equities Desk momentarily changes his orientation . Peter utters "ar :::::" and after
one second delivers the following utterances. In the illustrations, Peter is second
and Alex third from the right:
Animated Texts
69
(13 .0)
P:
P:
A:
P:
a .kay?
(3 .2)
(0 .6)
(0 . 6)
Peter's talk is now addressed to Alex. It is not simply a further gloss of the story
on which Peter is working, but rather a specific quote from the text itself. In its
delivery, Peter displays - through his orientation toward the screen, the pace of
the talk, and the self repair - that the information he is providing to Alex is an
extract from the news story he is currently editing.
The delivery sharply contrasts with the earlier version. It is not rendered as a
joke or as a precis but rather as part of the original, authentic story. The way in
which the talk is produced, coupled with the accompanying visual conduct,
provides colleagues with the resources to differentiate the status of the description
and in particular its relationship to the textual version of the story.
The delivery of the quote is occasioned by Alex's (re)orientation to Peter. The
(re)orientation is treated by Peter as inviting further information concerning the
story, as demanding an exposition of the story rather than continuing the joke.
Although the original joke is not specifically addressed, the informing is designed
for Alex and shaped to provide accurate and authentic information concerning the
recent shift in Israeli interest rates .
Even following Alex's acknowledgment of the story, Peter remains sensitive to
his colleagues' potential interest in and use of the item. In particular, some
seconds later, he reads aloud the sentence that describes the actual fall in interest
rates . By pausing in the delivery of the sentence, Peter momentarily renders the
description problematic and, on completing the sentence, goes on to make a
correction to the story. The correction is accomplished by the speaker realigning
70
his position to the text, from narrator to commentator. Peter differentiates his
version from the original text and publicizes , at least across the two desks, the
editorial correction.
Peter :
Peter :
Peter :
Alex :
Finally, Peter marks the finish of business, handling the Israeli Interest Rate
story, by checking with Alex concerning who should receive copies of the
corrected version.
What began as a joke, therefore, turns out to have some serious import for
news production . The Israeli Interest Rate story gets publicly corrected and
distributed to more than one desk and subsequently to the customers of both
Money and Capital and Equities . It also is featured in and referred to in other
stories that are handled by the two desks on that day. The story achieves its wider
circulation by virtue of Peter' s joke. The joke is delicately designed to establish
these possibilities. It does not demand that others abandon the activities in which
they are engaged or even take up the story. Rather Peter' s joke renders visible the
gist of the story that he is currently editing . It momentarily displays the activity
in which he is engaged. It provides colleagues with news concerning the Israeli
Interset Rates but does not demand a response. The talk is produced as if devoid of
sequential relevance and yet invites others to consider the import of the story with
respect to their own activities and responsibilities.
In gaining some indication that a colleague is interested in hearing more of the
story, Peter transforms the way in which he presents the text to the others. The
joke is abandoned, and the speaker provides an authentic rendition of the text,
(re)presenting the change in interest rates. The speaker therefore differentiates the
informing by virtue of the ways in which he presents the text, although in both
cases, it is as if he is simply reading aloud the story on the screen. In the final
part of the informing, the speaker again alters his standpoint vis-a-vis the text ,
visibly locating an error in the original copy. In rendering his activity visible, the
speaker exploits, through the ways in which he talks through the story,
differential standpoints with respect to the text itself. In this way, he ongoingly
tailors the sequential import and sense of the story for those within the local
milieu.
Animated Texts
71
A postrophic Readings
In the first example, we find the speaker reading aloud and then, in the light of an
almost minimal reaction by a colleague, delivering a detailed exposition of a
story. The exposition contrasts with the humor of the reading; the exaggerated
Jewish accent is abandoned as Peter provides his colleague with details of the
story and in particular the specifics that may be of particular relevance to the
Equities Desk. Peter therefore does not simply make visible, through reading and
explicating the story, the activity in which he is engaged but rather, through the
talk, differentiates elements of the text. The text, at least as far as the recipient is
concerned, is constituted through the talk, and the talk provides the resources
through which colleagues can differentiate the character and status of the story . As
the following fragment suggests, the ways in which the story is characterized,
even when the characterization is itself a quote or rendition from the text, may
transform, not only through successive utterances, but within the developing
course of a single turn at talk. Differentiating and transforming the text in this
way is embedded in and coordinated with the actions of the potential recipient(s)
and the ways in which he ongoingly participates in the text's rendition.
In the following instance, Peter returns to the desk after his lunch break. A
fellow journalist on the Capital and Money Desk begins to tell Peter what he is
working on:
Fragment 2, Transcript
Jan :
Jan:
Peter:
Jan :
Isa:
9h hah .
(.)
Peter:
A cigarette vendor?
Peter :
Jan :
Jan :
(0 .2)
72
Jan announces that he is working on the Nigerian Crisis and raises the possibility
that the story might be dropped. He continues by giving Peter a flavor of the
story, an utterance that serves to project further information. Peter, however, takes
up the possibility raised in the introduction (namely, spiking the story) and,
remaining oriented toward his screen, starts typing right on the completion of his
utterance.
Despite the response, Jan attempts to encourage Peter to listen to further details
concerning the story. He utters "erm," projects continuation, but withholds the
projected turn. Roughly one third of a second into the pause, Peter lifts his hands
from the keyboard and turns from his monitor toward Jan.
Fragment 2, Illustrations 1 and 2
v
v
=errn- - - - - - - -I .fear the way things
(0 .8)
In the light of Peter's commitment to listen to the story, Jan transforms the ways
in which he is voicing the text. Rather than describing what the story says, Jan
quotes a quote from the text. The exposition is hearable as a quote of a quote by
virtue of Jan's visual and vocal conduct. By preserving his visual orientation
toward the monitor, even after Pete has turned toward him, Jan displays that he is
reading the story to his colleague. Moreover, through the character of the text that
is read aloud and, in particular, the use of the first person pronoun, the present
tense, and an expression such as "I fear," Jan displays to the recipient that the
quote is itself a quote in the text. Jan seemingly produces an authentic version of
the text, which is itself a quote, a quote that is presented as a simple, unsoiled,
voicing of the voice quoted in the text itself.
The ascription of the quote to a cigarette vendor delightfully transforms the
utterance into a joke. It not only underscores retrospectively the absurdity of the
quote but also reconfigures the speaker's position vis-a-vis the talk he has
produced; it even renders questionable the authenticity of the quote, which it now
turns out was delivered dead pan. The joke, of course, turns on the idea that a
Animated Texts
73
quote from a cigarette vendor would feature in a Reuter's story . Through the
ascription, the speaker's voice is reconstituted retrospectively and prospectively
reconfigures the nature of the activity and its sequential relevance.
It is no longer relevant, at least in next turn(s), for the co-participant(s) to
address the import of the story with respect to its news worthiness, but rather to
recognize and respond to the joke. Isa giggles and Peter picks up on and escalates
the joke. Rather than deal with the organizational relevancies of the story, Peter
delightfully juxtaposes cigarette vendor with major analyst, with Jan in turn
producing keysources.
Although Jan's remarks may appear more concerned with telling a funny story
than dealing with the news, like Fragment 1, the exchange has some relevance for
the work in which the participants are engaged. It not only informs Peter what
story Jan is working on, but also allows Peter to know what story he should take
from the basket next. Jan's remarks also provide Peter with a characterization of a
story, a characterization that suggests that the story might be worth spiking, and
invites Peter's response. Although Jan may be more concerned than Peter as to
whether the story is worth salvaging, he establishes Peter's support in spiking the
story if he so wishes. By the completion of the episode, not only is the story
spiked, but also all those working on the desk know that it has been spiked and
why. In one sense, therefore, no matter how jocular or trivial Jan's remarks might
seem, they provide the foundation for a collaborative decision not to transmit a
news story concerning the ongoing crisis in Nigeria. This might seem
unimportant to us now, but for those in financial services and dealing rooms in
the City of London and elsewhere at that moment, the decision to spike the story
may not necessarily have been inconsequential. It is not surprising, therefore, that
you might seek the thoughts of your colleagues, however indirectly, before taking
such a decision.
The articulation of the story is accomplished progressively , in the light of both
the speaker's ability to establish particular forms of co-participation and the
recipient's willingness to cooperate as an interested listener. An interesting feature
of the story's articulation is the speaker's shifting alignment to the text displayed
within the talk. The different standpoints that the speaker adopts in relation to the
story and particular components within that story, including speakers and their
utterances, are contingent on and accomplished through the (co-)participation of a
colleague(s). Moreover, in the light of particular forms of co-participation, the
speaker not only can articulate particular voices and shift these voices
retrospectively but also, in so doing, can transform, within the articulation of a
single utterance, the activity in which he is engaged. In consequence, the
sequential implicativeness of a particular action and the trajectory that emerges
therein transforms as the speaker shifts the way he or she voices the text. The
active listener so richly described by Bakhtin (1976) not only responds to and
develops the utterances of his interlocutor but also, through the ways in which he
behaves both prior to and during the course of the talk, permeates the very things
that are said and done.
74
One element of this interactional organization is the way in which the ongoing
behavior of the co-participant informs, and is informed by, the varying
standpoints the speaker adopts with respect to the text in which the talk is
embedded.
Collaborative Viewings
In some cases, journalists encourage colleagues not only to listen to a story on
which they are working but also to actually look at the text with them. These
collaborative viewings seem to be a recurrent feature of working together on the
news desks, but they may be less concerned with reading than with establishing
an appropriate orientation for an extensive exposition . Consider the following
instance, drawn from the Equities Desk, in which Paul tells Nick about the
layoffs announced by IBM a few minutes earlier. ("*hhhh" represents an inbreath;
the brackets indicate that the utterances or parts of the utterances are produced in
overlap.)
Fragment 3, Transcript 1
Paul :
Nick :
Paul :
Nick :
Yeap
(0.3)
How many. Out of . how
I Fifty
Nick:
Viv :
Paul :
It
Paul :
Viv:
Animated Texts
75
The recipient's conduct provides a foundation for an exposition of the story and in
particular a lengthy quote from the text itself. It appears to be the case, at least for
a reading that will necessitate a number of phrases and/or clauses, that to have a
recipient orient toward the text with you provides a more suitable environment
than looking at the speaker (and in particular the speaker's face). Indeed, for
relatively lengthy readings, it appears that tellers go to some trouble to have the
recipient orient toward the text rather than the speaker.
The text is in some sense the source and the means of the activity, and
alignment toward the text can in consequence demonstrate participation and
involvement in the activity . It would also seem, however, that an orientation
toward the speaker rather than the text during the delivery of such readings can
prove problematic, a fragile state of affairs, almost in tension . Indeed, looking at
76
the speaker rather than the text can serve to display the temporariness of the
recipient's alignment and commitment to the exposition . Such tensions may be
related to the ways in which gaze toward another can serve to engender action, to
elicit, for instance, the gaze of the person at whom one is looking and to establish
mutual gaze (cf. Kendon, 1991, Goodwin, 1981, Heath, 1986). By looking at the
speaker during the delivery of a quote, you encourage a mutual, face-to-face
alignment, an alignment that itself can serve to bring the quote to a premature
end. In encouraging the recipient to look at the text with you, even though in one
sense it is futile because he is too far away to read the text for himself, you not
only secure commitment to the telling of the story but also avoid placing yourself
under undue pressure to address the remarks visually toward the co-participant. If
the speaker were to return the gaze of the recipient, it might well serve to
undermine the possibility of delivering a quote from the text. Mutual alignment
toward the text provides an appropriate interactional environment for one
participant to read aloud to another and thereby tell the story.
Toward the completion of the quote, the recipient turns back to his own
monitor. In response, he simply mumbles, "*Bloody Hell," and it is left to Viv
(the editor on the desk) to elicit further information about the story. Indeed, she
attempts to elicit just the information that Nick was asking for earlier: namely,
what proportion of ruM's work force is being sacked. As the discussion
continues, Viv later announces that they will write a short feature on the ruM
story .
So although the original telling may be making no more than a simple moral
point concerning the behavior of ruM and the stock market, the voicing of the
text brings the story to the attention of others on the desk and leads to a feature
that may not otherwise have been written. The moment of small talk turns out to
have some relevance for the organization of the work in which the participants
are engaged and in particular in collaboratively deciding what to do with a
particular story. Many of these tellings have this sort of character. In various
ways, they render some part of the participants' current activities visible to others
within the domain and in particular provide others with a sense of the stories on
which they are working. These tellings may simply lead to a collaborative
decision to spike a particular story, but they may also generate discussion
concerning the relevance of the story for customers who might not otherwise
receive the news. So although these tellings may in some way appear trivial and
irrelevant to the various responsibilities that journalists have within the news
room, they can have a profound impact on who receives what news and the
consequent behavior of the markets.
Animated Texts
77
Forestalling a Reading
Occasionally speakers fail to establish an appropriate orientation from a colleague
for the exposition of a story. Consider the following example, in which Barry
talks about a story he is working on, a severe typhoon that is gathering in
Argentina.
Fragment 4, Transcript 1
Barry :
Barry :
This
Barry :
Celia :
Carol :
Barry:
Urn yeh.
Barry :
(1.0)
( .)
Oh
~.
(.)
(9 .0)
((26 .00) )
Barry and Celia are alongside each other on the Commodities Desk. While Celia
is typing , Barry begins to talk. His initial utterance, "I' ve got a little more on the
typhoon," suggests more is to follow and is designed to secure some form of
commitment from Celia to listen to the story (cf. Sacks , 1992) . Although Barry
is unable to see exactly what Celia is working on, he positions his initiating
utterance to maximize the possibility of gaining some commitment. It is
juxtaposed with Celia momentarily removing her hands from the keyboard .
Unfortunately, as Barry begins to speak, Celia begins once again to type.
Barry does not abandon his attempt to tell Celia the news . Rather, he attempts
to encourage Celia to temporarily suspend the activity in which she is engaged
and orient toward the story on which he is working . He leans sideways toward
Celia while simultaneously exaggerating his orientation toward his own monitor.
In response, Celia reorients posturally towards Barry's monitor but steadfastly
continues with the activity in which she is engaged. Celia's divided bodily
orientation attempts to reconcile Barry's eagerness to tell the story with her own
interest in continuing the activity in which she is engaged.
Despite Celia's attempt to divide her orientation and encourage the speaker to
tell the story while she continues to work on some other text, Barry attempts to
secure more commitment. Prior to completing the preface to the story, Barry
stretches the word "the::," produces "er.m," and pauses.
78
B. reorientes
v
--I've 9Qt a bit more : on the : : er :m,----- - --this-the
C. ceases
and again
starts typing
C. reorientes
Neither the sound stretch nor the hesitation serves to encourage Celia to provide a
more wholehearted commitment to the story . She continues to type and remains
oriented toward her own monitor and text. The speaker pauses, withholding the
projected item and upcoming story. Roughly half a second into the pause, Barry
momentarily glances at Celia's screen. The glance may be more concerned with
encouraging her to orient toward him rather than with assessing the state of her
current activity. As Barry returns his gaze to his own monitor, Celia produces a
series of staggered taps on the return key that appear to project an upcoming
boundary in her current activity.
Barry exploits the opportunity and begins to speak precisely on the anticipated
break in Celia 's activity .
Rather than continue with the projected news, however, Barry once again
attempts to encourage Celia to upgrade her commitment to the story. He
temporarily delays telling the tale, thrusts his head toward his monitor, and utters
"this." The demonstrative pronoun, coupled with his visual actions, appears to be
designed to show simultaneously (the source of the) upcoming story, while
encouraging Celia to look at the story and thereby to temporarily suspend the
activity in which she is engaged.
If his actions are an attempt to draw Celia's attention to the text and telling and
to elicit some greater commitment from Celia, they fail. She once more begins to
type, and Barry abandons any further attempt to build the story in the light of a
mutual orientation toward the text.
He returns to the prior activity and completes the preceding utterance, replacing
"this" with "the typhoon ." He does not follow the initiating utterance with (the
Animated Texts
79
80
coordinate the production of the telling with whatever he can retrieve from the
visible elements of his colleague's conduct.
Although Barry fails to establish an appropriate audience for an exposition of
the story, the item he mentions does not pass unnoticed. A minute or so later,
Carol seeks further information about the item from Barry, clarifying how it fits
with some weather news within another story that has entered the basket. Some
fifteen minutes later, as Barry is leaving the desk to go home, Carol asks where
he put the typhoon story and to whom it was sent. So although these tellings in
themselves may not always have much immediate sequential significance, they
can provide resources for the production of news sometime later. In particular,
they not only display who is working with what story, but also provide a sense of
the gist of the story, which can turn out to be relevant for other sorts of items
with which colleagues are having to deal. Momentary exchanges or tellings, or
simply catching someone reading aloud parts of a story on which he is working,
can be retrospectively found to have some import for an activity in which you
are engaged some time later.
Animated Texts
81
establish recipient alignment toward the text for the delivery of a quote, whereas a
precis of the story might be delivered while the co-participant is looking at his
own monitor and editing a separate story . Moreover, the data suggest that the
teller, in failing to secure relevant co-participation for the accomplishment of a
particular type of action such as a reading, may transform the projected activity in
which he or she is engaged and deliver the news in a very different fashion than
was suggested in the preceding talk. Journalists not only render particular
activities in which they are engaged visible to others within the local milieu, but
also develop selective characterizations and portrayals of stories and differentiate
the various forms of co-participation they require for different actions within those
renditions. Quotes, precis, readings, and summaries are differentiated in the talk
itself (and the characterization that is being developed) and systematically
accomplished with respect to different forms of co-participation from the
recipient(s). The accomplishment of the tellings, the step-by-step production of (a
characterization of) a story, are produced with respect to the current conduct of the
co-participant and in particular his or her orientation to different components of
the characterization during its articulation . Mutatis mutandis, the characterization
itself and the ways in which the textual story is rendered visible, is contingent on
the co-participation of the recipient during the course of its production .
The relationship between the informings and the text is both curious and
complex. In the production of an informing, tellers differentiate the status of
different components of the characterization with respect to the original text. The
text itself, the existence of an authorized and written account, is exploited in the
telling for a variety of purposes. It can be used to produce and present the factual
version of some set of events to enable others to build or transmit stories that
will have a significant impact on the behavior of particular markets. It can be used
to make political comments or to ridicule the journalism of colleagues based in a
distant office. Within each fragment, we find the teller systematically displaying
(and exploiting) the relationship between the informing and the text and
demarcating his own standpoint and "voice" from both the original author and
even sources within the text itself. So, for example, we can observe the ways in
which the teller can display that he is rewriting that story within the course of its
telling and retrospectively recast the authority of the text. Or, for example, we can
see how tellers prospectively establish a quote of a quote and display their own
alignment toward the relevance of the story for news production . The text is
selectively rendered visible. It is revealed within the talk and through the ways in
which the teller animates or embodies the text. The developing rendition , the
ways in which the talk embodies the text, is dependent on and embedded in the
emergent interaction with others within the local milieu and, in particular , the
teller's ability to establish and sustain particular forms of co-participation during
the production of a telling.
The text, therefore, or at least the text displayed on the monitor, does not so
much mediate the interaction but rather is ongoingly constituted in and through
the interaction . The characterization that is developed, the animated versions of
82
particular elements, and the quotes and the summaries do not simply transmit or
present the text but rather constitute the text, the story, for the other(s). The coparticipant's access to the story, even when he looks at the screen with the
speaker, is constrained by both what he can see and his understanding of the
contents of the piece. His colleagues' talk constitutes the text and the ways it is
seen on this occasion, its intelligibility, and its sense here and now. In
constituting the text, personnel provide a sense of their current work and the
activities in which they are engaged . Access to the story is asymmetrical. It is not
simply that the co-participants may not necessarily be able to see the same story;
rather, what they know of the text, even during the course of a collaborative
reading, is permeated through and embedded in the ways in which the teller is
concurrently characterizing the text. The reading is interaction ally constituted, not
only in the traditional sense through the author's relationship with the reader (Iser,
1982), but also through the interaction between teller and recipient. The talk
elaborates the text, and text elaborates the talk within the developing production
of the interaction .
The observations discussed here provide further support of the critique of the
conduit metaphor of communication, a critique developed by Bakhtin (1976) and
others such as Wertsch (1991). Moreover, the observations underscore the dialogic
nature of talk, even where the talk itself is reporting reported speech. And, we
might, following Bakhtin (1976), consider the ways in which talk reproduces and
relies on a particular textual genre (news reporting) that may be theoretically
distinct from the current context and yet forms an integral part of retelling and
editing stories within the newsroom. But the character of the dialogicity and
textual rendition discussed here goes beyond the idea of genres characterized so
profoundly by Bakhtin and developed in diverse ways by Todorov (1990), Lodge
(1990), and Goffman (1974) in his frame analysis of talk. In particular, in the
materials at hand, we find that the delivery of a piece of news to a colleague , itself
a selective rendition of a textual report, is produced with respect to the shifting
alignment and participation of the (potential) recipient. How the text is
(re)produced is thoroughly contingent on the interaction as it emerges between the
participan ts. The relationship between the speaker and the author, the report and
the original text, and the distinction between de dicto and de re (cf. Coulmas,
1986) is thoroughly contingent on and embedded in the emergent interaction
between the participants, even during the shifting course of the utterance and
textual reading.
As in other organizational environments , personnel in the newsroom at Reuters
have asymmetrical access to each other's activities. Aspects of colleagues'
activity, even when they are discussing a particular story with a colleague and
appear to be in mutually focused interaction, may be unavailable to the coparticipant. Moreover, in the newsroom as in other working environments,
individuals may simultaneously participate in multiple overlapping and
interdependent activities that demand a complex and continually shifting range of
configurations between those within perceptual range of the event (cf. Goffman,
Animated Texts
83
84
Acknowledgments
We would like to express our gratitude to journalists at Reuters who so
generously cooperated with the research. We would very much like to thank
Lauren Resnick, Jon Hindmarsh, Charles Goodwin, and Paul Luff for comments
on an earlier version of this paper. We would also like to thank Bernard Conein ,
David Greatbatch, Suzanne Guenthner, Isaac Joseph, Hubert Knoblauch, and Per
Linell . We would also like to thank Lucy Suchman for her support and
encouragement for the project of which this study forms one part. The study is
supported by the Rank Xerox Research Laboratories Cambridge EuroPARC, the
European Commission RACE Programme Project MITS (Metaphors for
Telecommunication Services), and the ACTS project MEMO .
Animated Texts
85
References
Atkinson, J. M., & Heritage, J. (Eds.) (1984) . The structures of social action.
Cambridge, England: Cambridge University Press .
Bakhtin, M. M. (1976) . Speech genres and other late essays . (V. W. McGee, trans.)
Austin : University of Texas Press.
Brown, J. S., & Duguid, P. (1994) . Borderline issues : Social and material aspect s of
design . Human-Computer Interaction, 9(1), 3-36.
Coulmas, F. (1986) . Direct and indirect speech. New York: Mouton de Gruyter
Drew, P., & Heritage, J. (Eds.) (1992) . Talk at work: Interaction in institutional
settings. Cambridge, England: Cambridge University Press .
Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: PrenticeHall.
Goffman, E. (1974) . Frame analysis: an essay on the organisation of experience. New
York: Harper Row.
Goffman, E. (1981) . Forms of talk. Blackwell: Oxford .
Goodwin , c., (1981) . Conversational organisation: Interaction between a speaker and
hearer. London : Academic Press.
Goodwin , C., & Goodwin, M. H. (1996) . Formulating planes: Seeing as a situated
activity . In Y. Engestrom & D. Middleton (Eds.), Cognition and communication at
work: Distributed cognition in the workplace. Cambridge, England : Cambridge
University Press
Greatbatch, D., Luff, P. K., Heath, C. C., & Campion, P. (1993) . On the use of paper
and screen in the general practice consultation . Interacting with Computers, 20,
216-234 .
Harper, R., & Hughes, J. (1993) . What a f-ing system! Send 'em all to the same place
and then expect us to stop 'em hitting: Making technology work in air traffic
control. In G. Button (Ed.), Technology and the working order (pp. 127-144) .
London : Routledge.
Heath, C. C. (1986) . Body movement and speech in medical interactions. Cambridge,
England: Cambridge University Press. (Editions de la Maison des Science de
l'Homme, Paris).
Heath, C. C., Jirotka, M., Luff, P., & Hindmarsh, J. (1995) . Unpacking collaboration :
Interactional organisation in a city trading room . Journal of Computer Supported
Cooperative Work, 3(1), 147-175.
Heath, C. C., & Luff. P. (1992) . Collaboration and control: Crisis management and
multimedia technology in London underground line control rooms . Journal of
Computer Supported Cooperative Work, 1(1-2), 69-94.
Hutchins, E. L. (1990) . The technology of team navigation. In J. Galegher, R. E.
Kraut, & C. Egido (Eds.), Intellectual teamwork: The social and technological
foundations of cooperative work. Hillsdale , NJ: Lawrence Erlbaum Associates.
Iser, W. (1985) . The act of reading. London : Routledge.
Kendon, A. (1991) . Conducting interaction: Patterns of behaviour in focussed
encounters. Cambridge, England : Cambridge University Press.
Lodge, D. (1990) . After Bahktin. London : Routledge.
Markova, I., & Foppa, K. (1990) . The dynamics of dialogue. London : Harvester
Wheatsheaf.
86
Chapter 4
Abstract
88
(cf., for example, the works of Hastie & Penningt on, 1991; Hughes, 1990;
Resnick, 1991; Suchman, 1987).
However, how should this process be described with respect to its
conversational developments, as well as in terms of its logical accomplishment?
There have been relatively few studies carried out in this area, even though, in
principle, there is no reason to consider that the fundamental principle of
ethnomethodology, as stated by Garfinkel, should not be applicable to this sphere.
Every topic of logic, order, reason, meaning, and method is to be discovered and is
discoverable, and is respecified and respecifiable only as locally produced,
naturally accountable phenomena of order. These phen omena are immortal,
ordinary society's commonplace , vulgar, familiar , unavoid able, irremediable and
uninteresting "work of the streets." (1990, p. 77)
The preceding thesis reflects the concern for theoretical prudence and the
prevalence given to the interpretations effectively made by the speakers. But
89
90
91
and resolve when they have a common goal, whether this goal comes from
themselves or is externally imposed on them, as in the sequence examined here.
The Objectives of the Chapter
The three theses mentioned earlier - that rationality is accomplished and exercised
conversationally, that this emerging rationality is no different in nature from that
explaining logic systems, and that it is organically linked to the processes that
govern conversational organization - merit an epistemological, theoretical, and
methodological analysis beyond the scope of this article. This is why we tackle
them here in a practical manner, by analyzing a conversation.
We begin with an example of an interactive process of problem resolution .
After quickly describing the context, we provide the conversation that permits
resolution of the problem. Next, we attempt to show how this conversational
process progressively brings about the logical solution to the problem, which
will amount to highlighting the idea that rationality as rationality itself emerges
from the conversation .
92
paper pulp has been prepared and colored. This manufacturing work covers the
different operations carried out continuously on the sheet of paper: The formation
of the damp sheet on the fabrication table and its progressive drying out in the
damp part of the machine; drying out the sheet; and rolling the sheet of paper.
The situation was as follows . The work on the paper currently being
manufactured at this change of shift had begun four days earlier. This implies that
each of the two operators had already been involved in this manufacturing process .
The processing of the paper had already caused a number of difficulties, and the
process had therefore been subjected to adjustments by the operators since its start
up. Thus, a part of the manufactured product had to be discarded on two occasions:
the previous day, because of pollution coming from the machine circuits, and
during the night shift, because of pollution from the recycled paper pulp. The
latter pollution incident had resulted in the paper pulp circuit being stopped,
cleaning of the corresponding mixing vats, and then progressive restarting of the
recycled paper pulp. Several paper breaks had occurred at the first press since the
start up of this manufacturing.
~""
INPUT:
coloring
pulp composition
pulp concentration
type of pulp
characteristics of pulp
refined
recycled pulp rate
products added
continuously
CONTROllED PARAMETERS:
pulp flow
speed of jet of pulp
speed of manufacturing linen
speed of press
pressure of press
pulp level at the machine head
pulp gate opening
wringing
format of the sheet
recycled pulp ...
OU1PUT:
width of the sheet
thickness
weight of the sheet
tint of the sheet
porosity
hot pressing
ashes rate
acidity/basicity rate
(pH)
paper fibers
arranging ...
The operators of the machine, who constantly check the dampening component,
which is the most delicate section, diagnosed these malfunctions in the following
93
way. The paper currently undergoing manufacture was of relatively low gram
weight (70 gr.). Consequently, projections of paper pulp had settled at the edge
sprays, otherwise known as pissettes by the operators. The edge sprays are two
jets of water that pass across the sheet in the dampening part of the machine , thus
allowing its width to be demarcated so that it can be cut on either side. The
projections of paper pulp were caused by the low gram weight and a jet of water
that was not completely steady. The pulp had then accumulated at edge spray
level, thereby forming a pile that had come off and then fallen on the sheet being
formed: hence the appearance of paper breaks at the first press. Furthermore, two
paper breaks had occurred at the stack during the morning (this resulted from
another malfunction that is not related to the one discussed here).
The Extract
The following transcript concerns the malfunction caused by the paper pulp
projections. The purpose of the conversation is to inform the on-coming operator
about this. The information concentrates on the setting of the machine: in other
words, the operating modes. The mutual recuperation (and the cognitions) of the
operators is aimed at adapting the operation of the machine to the constraints
imposed on it by the product currently being processed. A is the out-going
operator (leaving the work station), and B is the on-coming operator (the
replacement).
[(...) indicates that some of the text has not been reproduced; ... indicates a
pause]
Al :
B1:
A2:
B2:
A3 :
B3 :
A4:
B4:
AS:
94
B5:
A6:
95
96
97
turns, we begin by showing this expression as a variable (x) until the fixing of its
semantic interpretation . From this point on, it will be represented by the constant
(p) .
This process is governed by the logical rule presented earlier, which states that the
non-fulfillment of a preparatory condition for an illocution leads to the nonfulfillment of this illocution . By stating an objection (B 1), which is expressed as
is usual during a conversation ( "And (.oo) Have you noticed?"), B denies a
preparatory condition that he attributes at the same time to the assertion from A I .
It would seem as though, in B's opinion, that A did not have good reasons for
stating AI . Let x be the proposition , "The pissettes. they seem to be working
better," and q the proposition, "it still lifts the sheet." Following x, the q
statement establishes two things. First, it establishes between the speakers, as a
product of their discursive cooperation, an interlocutory couple that introduces a
debate (Trognon & Larrue, 1994) and is a product of their discursive cooperation.
We express this in the formula (x -> not q), where x is statement Al and (not q)
is the preparatory condition, introduced by B, for this statement. Subsequently , q
expresses the position of B in the face of the previous formula; the effect of this
position is that the assertion from Al is rendered defective and thus false. This
exchange can be represented in the following way:
A1:
B1 :
A1 :
B1:
What B concludes
What B attributes to A
AI : x
Al B: (x --> not q)
BI: q
BI ' : not x
98
By linking up with A2, A now accepts that the rear edge spray is lifting up the
sheet. He accepts the q proposition. At least we can note this for the sake of
simplicity, because he does not simply accept that the rear edge spray lifts up the
sheet: He states that he noticed it (" haveyou noticed?/umm, yes"), which actually
amounts to accepting the willful content of this perception but in a possible
world ( "maybe H). Furthermore, by following up with "But I didn't get as much
pulp as yesterday I did nothaveanypulp afterhuh," A clarifies, on the one hand,
his initial formulation ("they seem to be working better" is instantiated by "not
having any pulp"), that is, (x =p), and, on the other hand, he attributes to B the
idea that pulp projections are the result of the rear edge spray lifting up the sheet,
that is, (q -> not p), thanks to the conjunction but. The assertion by A2 has the
retrospective effect of attributing the theory that (q -c-not p) to B. The proactive
effect of the same statement is the inference [not (q-c-not p)] to which A is
committed and which B is invited to do. Indeed, it could not be (not q), because A
has just accepted q. Hence, employing the same type of representation as
previously, we show their continuing exchange:
Al :
Bl :
A2 :
Al:
Bl :
A2 :
x
q
q but (not p)
99
Table 4.2. Representation of the logical process emerging from the three initial
illocutions.
The columns repre sent the same things as in Table I (A is the out-going operator and
B the on-coming operator). But we have added A's reply . In saying "yes, maybe , " A
agrees q. He says also that he "didn't have any pulp after, " that it is proposition p .
Thi s propositio n is a more preci se reformulation of x. A implicitl y attributes to B (q>not p) a proposition shown in the third column . Because BI A, q and p, A concludes
AI: x
Al B: (x --> not q)
BI : Q
BI ': not x
A2: Q
B I A: (q --> notp)
A2: p
A2' : not
(0
--> not
0)
In the model shown in Table 2, the first column shows the illocutions
successively expressed in the convers ation; the second column shows the
"conclusions" drawn from these illocutions; and the third indicates the allocations
that the interlocutors make during their illocutions. The third column is
particularly interesting. First, let us consider that Al B is the formula that B
attributes to A retrospectively at the second speaking turn. When A clarifies his
initial formulation at the third speaking turn (x=p), Al B becomes something of
the following type: (p-c-not q). This instantiated formula can now be compared
with the formula that A retrospectively attributes to B at the third speaking turn:
(q-c-not p). The latter is the contraposition of the former, which means that the
interlocutors allocate equivalen t formulae to each other. In some ways, therefore,
the interlocu tors share the same formula: namely, (q -c-not p). It defines the
proposition being debated : Both admit q (that the rear edge cutter is lifting up the
sheet ), but A insists that there is no projection of pu lp (q and p), whereas B
insists that there is (q and not p). This conforms exactly to that which is
accomplished by the participants in the (B2, A3) pair:
B2 :
A3 :
I got s ome .
You got s ome? Me, I didn 't get it , huh .
100
Table 4.3 . Representation of the debate betwee n A, the out-going opera tor, and B .
the on-coming operator.
Table 3 represe nts the debate between A and B and includes Table 2 and B2 and A3 .
which are the propos itions about which A and B contradict themselves . Table 3 differs
from Tables 1 and 2 because the first column represents the sequence of illocutions
successively uttered by the interloc utors and the first line represents the interlocutors.
Thus we see that A and B debate (q- >not p) and that B agrees with not p and A with p .
Their contradiction is situated between p and not p.
A
Illocutions
What
A
says
Al
What A
concludes
AandB
WhatB
attributes
toA
What is
common
toA
andB
B
What A
attributes
toB
WhatB
conclud es
What B
says
x--> not q
q
BI
not x
A2
A2
q -->
notp
not (q -->
not p)
q -->
not p
notp
B2
A3
101
A4 :
B4 :
AS :
I g ot some
and I cl osed it o f f a
b it
because I found it
was deflecting the
sheet a bi t and that
made it squirt out
B' s action
(not q)
B ' s hypothesis
(q --> not p)
and by invited inference
(not q --> p)
102
From this standpoint, the B2-B3 interventio n is decisive as B establishes the fact
that preventing the rear edge cutter from lifting (or letting it lift somewhat less)
does not mean the end of paper pulp projection. Indeed, "I got some. ... And [
closed it offa bit because [ found it was defl ecting the sheet a bit and that made it,
umm, squirt out" could be interpreted as "I got some" despite the fact that,
although "I closed it offa bit" because "[ f ound it was deflecting the sheet a bit"
which caused pulp projection. It can also be seen that, by accomplis hing B2-B3,
B, this time fully and clearly, accepts the (q-c-not p) proposition to which he was
committed in B1 (by contraposition), which A had attributed to him in A2 while
contradicting it. The reasoning made by B is a truly experimental form of
reasoning that could be presen ted as in Table 4.4 .
Intuitively:
- (If the rear edge spray separa tes the sheet, there will be pulp projection s);
- By invited inference (if the rear edge spray does not separate the sheet, there are
then no projections);
- Therefore, I reduce the rear edge cutter, and I expect the projections to stop;
- However, they do not;
- Consequently (implicitly), it is untrue that the separating of the sheets by the
rear edge spray is a necessary condition for the projection of pulp .
At this moment in the exchange, the conditions experienced by A and B, the
"common world" in front of them, are shown in Table 4.5:
Table 4.5. Representation of A's and B' s common world after B' s reasoning.
The last column represents the state of affairs (the world) discussed by A, the outgoing operator, and B, the on-coming operator. So the last column is a set of truth
values resulting from the composition of the truth values of two propositions : q ( "that
one behind, it still lifts the sheet") and not p ( "getting some p ulp "). For instance, in
saying B2, B says that it is true for him that not p is true and q is true.
B ' s observation
not p
1
A's observation
True for B
(cf. B2)
True for A
(cf. A2 &A3)
True for B
(cf. B3)
False for B
(c f, B3)
In short, A and B tested the configuration (q and not p) together (see Table 4.6).
103
Table 4 .6. The truth table emerging from A's and B's interaction after B's
reasoning .
This table is simply Table 5 interpreted as a truth table of propositional logic. This
truth table is the truth table of a logical disjunction. Thereby the state of affairs
discussed by A and B is the disjunction of q ("that one behind, it still lifts the sheet")
and not p ( "getting some pulp ") .
q
not p
(q or not p)
104
I , reopened it ,
the front one
A's action
(p)
(not r)
A's hypothesis
(r --> not p)
and by invited inference
(not r --> p)
We have just structured this reasoning in the same way as that of B. It is, in fact,
more complex, because it seems to incorporate double reasoning: The first is that
previously stated; the second can be paraphrased as "It is even more true that I
didn't have any pulp thanks to the setting I made myself, that I was driven to pull
the tail end three times for a completely different reason (?) but the band wasn't
cut." "To pull the tail end," is re-starti ng the manufacturing process with a narrow
strip of paper on which the justification is gradually increased until the desired
width is achieved. In this way, if the edge sprays had been blocked up with pulp,
the tail end would have automatically broken . The second form of reasoning
reinforces the first, whether it be independent from the first (A had already carried
out the setting, and, for a completely different reason, he had had to re-start the
process), or whether it be integrated into the first one (A had to re-start the process
because he had not previously set the edge spray). Moreover, A draws a single
conclusion from the two forms of reasoning.
105
we know!) that the rear edge spray was not the problem; and thanks to A, B
knows (and we know!) that the malfunction was due to the front edge spray. The
different types of behavior adopted by the interlocutors toward theses sustained by
B and A should be noted. The first is logically, experimentally, and exhaustively
assessed. The simple confirmation of the second is sufficient for the interlocutors.
On the other hand, B5, referring to the two edge sprays, now explains the initial
proposition made by A at the same time as B5 justifies the illocution that induced
it as the propositional content, to which he was opposed in B1. The pissettes
seem to be working better [= there is no more pulp projection p] because they
are properly set; you say p because they are properly set. As an explanation of
AI , B5 satisfies this assertion. B changes his attitude in terms of this illocution
by justifying this illocution , which he had initially considered defective. From
this point on, the disagreement between the interlocutors is resolved.
Thus, A himself can only confirm, which is what he does in A6 by asking a
question "implying" a positive answer and by passing the microphone to B:
A7:
B6:
A8:
Development of
Disagreement
Treatment of
Disagreement
B5, A6-A7
106
Al
Bl
of cause ]but
A2
B2
A3
B3
A4-AS
L-EJI------------
B5
A6-A7
Conclusion
The conversation analyzed here occurred on several levels simultaneously. The
conversational aspect with the sequencing that it implies is fundamental, because
it maintains the progressive organization of the logical structure of the debate.
The logical aspect that develops simultaneously on the illocution level and on
that of proposals, which they are aimed at, is no less essential, because its process
through the conjunction of the analyses is that of the problem resolution .
107
It would still be necessary to add a textual aspect where the connectors (e.g.,
but, and, because) and the verb tenses playa very important role. Indeed, the entire
enunciative structure contributes to the resolution of the problem . For example,
"it still lifts the sheet " is stated in the present tense and indicates that, for B, the
situation has not really evolved and hence reinforces the illocutionary function of
the objection. On the other hand, "1didn't get as much pulp as yesterday, 1 didn't
have any pulp after" indicates a finished event that belongs to the past. This
suggests that a transformation took place and, in a certain way, outlines the
analysis that will follow. Furthermore, after having given an account in the past
tense of the adjustments he carried out and their results, A moves suddenly to the
present as if he is suddenly repositioning himself in the interaction he is having
with his interlocutor , meaning by that that things have now changed.
The superimposition of these different levels of operation contributes both to
the conversational production of the theory of recuperation relative to the
malfunction of the machine elaborated by the participants, and to the resolution of
the technical problem facing them. On the whole, the in situ accomplishment of
rationality is a global interactive process. In addition, what we have logically
formalized is also more or less expressed in the textual content of the
conversation; we now have to state that the distributed intelligence in the
conversation is overdetermined.
Furthermore, the conversational process creates implied social relationships
through shift work on the various operational levels. The two laborers working
on a machine in 3 eight-hour shifts must exchange information and share their
respective actions. They are required to construct a common history regarding their
relationship with the machine, which is not too far removed from reality. The
intricacy of the textual and logical processes of this operative conversation is thus
also the overall accomplishment of their work, this time, however, on a social
and economic level. It is a global social process.
108
References
Brassac, c ., & Trognon, A. (1993, July). The logic of speech acts as a universal
grammar of intercomprehension . Paper presented at the 4th International Pragmatics
Conference, Kobe, Japan .
Davillerd, C., & Grusenmeyer , C. (1993). La releve de poste: Une periode specifique de
travail collectif [Shift changeover: A specific period of collective work] In F. Six, &
X. Vaxevanoglou (Eds.), Les aspects collectifs du travail [Collective aspects of
work] (pp. 81-88) . Toulouse , France: Octares .
Evans, J. S. B. T. (1984). Heuristic and analytic processes in reasoning. British
Journal of Psychology, 75, 451-468 .
Evans, J. S. B. T. (1989). Bias in human reasoning: Causes and consequences. Hove,
England: Lawrence Er1baum.
Garfinkel, H. (1990) . The curious seriousness of professional sociology. In B .
Conein, M . de Fornel, & L. Quere (Eds.), Les formes de la conversation [Forms of
conversation] (Vol I, pp . 69-78). Paris: CNET.
Ghiglione, R., & Trognon, A. (1993) . OU va la pragmatique? [Where is Pragmatics
going?] . Grenoble, France: Presses Universitaires de Grenoble.
Goffman, E. (1969) . Strategic Interaction. Philadelphia, PA: The University of
Pennsylvania Press .
Grusenmeyer, C. (1991) . La releve de poste: Une phase critique du travail en equipes
successives [Shift changeover: A critical phase in shiftwork] . Cahiers de Notes
Documentaires de l 'INRS, 144, 407-418 .
Grusenmeyer, C. (1992, September, 6-11). Interest of the notion of shared functional
representation in shift changeover phase . Paper presented at the 6th European
Conference of Cognitive Ergonomics , Budapest, Hungary.
Grusenmeyer, C. (1995a). De l 'analyse des communications a celle des representations
fon ctionnelles partagees. Une application a la releve de poste [From
communications to shared functional representation analysis. An application to
shift changeover] . These de doctorat , Universite Rene Descartes (Paris V).
Grusenmeyer, C. (1995b) . Shared functional representation in cooperative tasks . The
example of shift changeover. International Journal of Human Factors in
Manufacturing, 5(2), 163-176.
Hastie, R., & Pennington, N. (1991) . Cognitive and social processes in decision
making. In L. B. Resnick, 1. M. Levine, & S. D. Teasley (Eds.), Perspectives on
socially shared cognition (pp. 308-327). Washington, DC: American Psychological
Association.
Heritage, J. (1990) . Interactional accountability: A conversation analytic perspective.
In B. Conein, M. de Fornel, & L. Quere (Eds.), Les formes de la conversation [Forms
of conversation] (Vol. 1, pp. 23-49). Paris: CNET .
Hoc, J. M. (1987) . Psychologie cognitive de la planification [Cognitive psychology
of planning]. Grenoble , France: Presses Universitaires de Grenoble.
Hughes , S. (1990). Planning-in-action . Cognitive phenomenon or social
accomplishment? In B. Conein, M. de Fornel, & L. Quere (Eds.), Les formes de la
conversation [Forms of conversation] (Vol. 2, pp. 217-236) . Paris : CNET .
Johnson-Laird, P. N., & Byrne, R. M. J. (1991). Deduction. Hove, England: Lawrence
Erlbaum.
109
110
Chapter 5
Abstract
In what rem ains one of the central accomplishments of cognitive anthropology, Berlin
and Kay (1969) demonstrated that the diversity of human color systems was built on a
univer sal infras tructure, with black and white being the most basic colors in all
systems. The analytical focus of their work is a structural system divorced from the
messy tasks of actually using color terms to make relevant distinctions within specific
courses of action situated within the concrete settings that constitute the lifeworld of a
particular society. By way of contrast, Wittgenstein' s later philosophy argues that it is
precise ly such endoge nous activities that provide the necessary framework for the
analysis of human langua ge. Using as data video tape of chemists attempting to
determine when to stop a reaction by deciding when the material they are working
with is jet black , this chapter explores (1) the diverse practices they deploy to
establish what can count as black; (2) how such a distinction is embedded within a
local activity system lodged in turn within a relev ant community of practice; and (3)
the embodied apprenticeship required for new members to become competent in the
use of such a category. For the chemists, jet black (e.g., the most prototypical example
of black) is not a preformulated, context-free universal color category, but instead a
problematic j udgme nt to be artfully accomplis hed through the depl oyment of a
collection of systematic work practices. This analysis contributes to the development
of a practice-based theory of knowledge and action.
112
Charles Goodwin
Two of the central and enduring topics in the analysis of cognition are the study
of vision (in neurophysiology, vision is the cognitive system that is best
understood, and its architecture provides a point of departure for the analysis of
how the brain organizes other types of representations) and semantic categories
(which at times have formed the essential subject matter for whole fields such as
cognitive anthropology) or, more generally, processes of classification. One
crucial place where these two lines of research intersect is in the analysis of color
categories, terms provided by language that are used to codify and structure
perception of the visual field. Different languages classify the color spectrum in
different ways. This has been argued to provide evidence for the Sapir-Whorf
hypothesis that language structures perception of the world that a particular
society inhabits (Bruner, Oliver, & Greenfield, 1966; Greenfield & Bruner,
1966). In what remains one of the central accomplishments of cognitive
anthropology, however, Berlin and Kay (1969) demonstrated that the diversity of
human color systems was built on a universal infrastructure, one almost certainly
linked to structures in the brain.
The focus of analysis in such work is an abstract structural system, divorced
from the messy tasks of actually using color terms to make relevant distinctions
within specific courses of action situated within the concrete settings that
constitute the lifeworld of a particular society. In contrast, Wittgenstein' slater
philosophy argues that it is precisely such endogenous activities that provide the
necessary framework for the analysis of human language.! Analysis of category
use from such a perspective has been a major focus of research by
ethnomethodologists and conversational analysts (e.g., Cicourel, 1964; Garfinkel,
1967; Heritage, 1984; Jefferson, 1987 ; Lynch, 1991; Sacks, 1992; Schegloff,
1972; Suchman, 1987) . Recently, some scholars have begun to analyze seeing as
a social process lodged within endogenous communities of practices (Goodwin,
1994, in press; Goodwin & Goodwin, in press; Heath, in press, this volume;
Heath & Luff, in press; Saljo & Bergqvist, this volume).
Iproblems with Wittgenstein's initial treatment of color perception provided the Achilles '
heel that led to the dismantling of the Tractatus (Hacker, 1986, p. 109) and its replacement
with a theory of color terms that emphasized their situatedness within the grammars of
diverse natural actitivies.
113
moreover, that a universal pattern exists for adding basic color terms to the
language:
Black
White
~ Red ~ Green ~
Yellow
Blue
~ Brown ~ Purple
Pink
Orange
Gray
Urhe work of Berlin and Kay generated a substantial body of subsequent research, some of
which led to modifications of their original typology. Such modifications, however, are not
consequential for the analysis in this chapter.
3According to Saussure, langue "is a storehouse filled by the members of a given
community through their active use of speaking, a grammatical system that has a potential
existence in each brain, or, more specifically , in the brains of a group of individuals . For
language is not complete in any speaker; it exists perfectly only within a collectivity"
(1959, pp. 13-14).
114
Charles Goodwin
Despite the brilliance of Saussure's insights here and the very great payoff they
have had in the subsequent development of linguistic theory, the way in which he
formulated langue had the inevitable effect of divorcing cognition from practice.
All of the cognitive work involved in coordinating talk and meaning with the
actions of coparticipants and of using language to build a relevant social world in
actual settings is treated as epiphenomenal. Cognitive phenomena, including
categories for the organization of perception , are situated analytically within the
structural system of the language as a whole, and the process of constituting their
meaning lies beyond the grasp of speakers.
115
Moreover, work in a variety of different fields has called into question the
assumption that human cognition operates within such neatly bounded packages .
Thus , in different ways, both Heidegger (1962) and Wittgenstein (1958) argue
that human cognitive activity is inextricably lodged within the activities and
settings of the lived social world: that is, that knowledge is intrinsically situated.
Strong support for such a position has come from the investigation of how
scientists actually do their work (Latour, 1987; Lynch & Woolgar, 1988;
Pickering, 1992), studies of cognition in the workplace (Heath & Luff, in press;
Middleton & Engestrom, in press; Rogoff & Lave, 1984; Suchman , 1987),
anthropological investigations of systems of cognition encompassing multiple,
differentiated actors and tools (Hutchins, 1993), practice theory (Chaiklin &
Lave, 1993), and conversation analysis (Heritage, 1984; Sacks, 1992; Schegloff,
1972, 1992b). All of this work demonstrates that a diverse collection of
heterogeneous phenomena and processes are implicated in human cognitive
activity : for example, not only mental representations but also material tools,
historically shaped and socially distributed forms of knowledge, processes of
social interaction and the forms of social action they produce , and recognizable
patterns of activity in a specific setting.
Restricted taxonomies provide analytical coherence by restricting research to
the study of patterned variation in a single domain of possibilities: for example,
ways of naming colors. If one wants to move beyond single taxonomies ,
however, the order provided by a data set structured in terms of comm on
underlying features is lost. The question thus arises concerning how a somewhat
ad hoc collection of very different kinds of entities can be related to each other
within a common analytic framework; for example, how can objects as diverse as
language categories, physical tools such as the vats and sticks that will be
examined later in this chapter, and social distributions of knowledge and power in
a specific work setting be studied as interdependent components of an integrated
cognitive process? The solution chosen in this chapter is to focus on what I call a
situated activity system." that is, the range of phenomena implicated in the
systematic accomplishment of a specific activity within a relevant setting . An
example of a situated activity system is provided by a game such as hopscotch
(M. H. Goodwin, in press), which integrates into a common framework of action
4-rhe term situated activity system was introduced by Goffman (1961 , pp . 95-99) to
describe repetitive encounters in social establishments in which an individual is brought
"into face-to-face interaction with others for the performance of a single joint activity, a
somewhat closed , self-compensating, self-terminating circuit of interdependent actions"
(1961, pp. 95-96). Goffman 's interest in moving role theory in new directions is different
from my use of the concept to investigate how properties of an encompassing, situated
activity shape cognition within specific settings. Such differences do not, however, detract
from my complete agreement with Goffman 's central argument: "The point about looking
at situated activity systems is that the complexities of concrete conduct can be examined
instead of by-passed" (1961, p. 99). For an analysis of situated activity systems constituted
through talk in local settings, see M. H. Goodwin (1990).
116
Charles Goodwin
117
cognitive phenomena, such as color categories, are constituted through the social
deployment of a collection of diverse practices lodged within the lifeworld of a
relevant community of practice.
118
Charles Goodwin
The participants in this setting are a geochemist and his students who are making
a scientific instrument. I videotaped what they were doing (from a specific
location that gave me better access to some aspects of what they were doing than
it did to others) but did not help in the work. By depositing manganese oxide on
acrylic fibers, the participants create a tool that is capable of extracting virtually
all of the radium ions present in a sample of water. Different bodies of water
(e.g., different rivers) have distinctive radium signatures. By using the fiber, the
7See Goodwin (1994) for an analysis of the socially organized practices of seeing that must
be mastered by a young archaeologist in order to make one of the standard documents of
her discipline : a map of an excavated section of dirt. Although the young archaeologist
know s the linguistic meaning of rules telling her where to take measurements (e.g.,
"wherever there ' s a change in slope") , finding what counts as such an event in the
complicated perceptu al field provided by the landscape in front of her is a complicated,
contingent process. Mastery of such ability is something that all competent archaeologists
expect of each other, and it constitutes part of the embodied infrastructure required for
proper understanding of the writing practices that constitute archaeology as a profession .
119
120
Charles Goodwin
As van Gogh recognized, the existence of a term such as black within the
semantic space of a particular language in no way solves the problem of how
what counts as black in nature is to be determined. Practitioners, such as the
geochemists being investigated here, who wish to use the category to locate
something relevant to their work are not given a solution to that problem by the
term itself. Instead, they are faced with a task: that is, how to find a specification
for black that can distinguish tones within "the endless variety of grays" in a way
that is appropriate to the activities in which they are engaged.
Several general processes provide organization for the work involved in the
task of determining a relevant specification for jet black.
Determining what will and will not count as a proper referent for a category in a
specific setting is lodged within larger activity structures . Establishing when the
fiber is jet black is important to the geochemists, because that color is the
diagnostic sign that the chemical reaction has proceeded to the point where it
should be quickly stopped. If the fiber is put into the quenching bath before it
exhibits the proper shade of black, it will absorb less radium ions when used later
121
as a tool for measuring different bodies of water. The color change is the simplest
measurement that can be made to indicate the progress of the reaction.f The
larger activity of making the fiber thus provides a motivational framework that
leads those involved in the activity to make particular perceptual distinctions in
the first place (i.e., it establishes a texture of relevancies, a focus for perception).
The activity also establishes the parameters of what will count as a correct
solution to the task of identifying black in these specific circumstances (i.e., a
range of shades that will lead to usable fiber if the reaction is stopped when they
appear). Clearly, other tasks would set other parameters. Moreover, different
tasks will set relevant standards for accuracy and precision at different places?
For some, a very wide range of shades might count as acceptable solutions,
whereas, for the geochemists being investigated here, a much more limited,
precise sense of what can count as an acceptable black is necessary if they are to
succeed in terminating the reaction at the proper moment. This precision arises
not from the status of their work as science, but rather from the specific task at
issue. In other scientific tasks involving the fibers, measurement could encompass
a much wider range of variation. Thus, when the fibers were used to gather data,
the scientists simply loosely filled the collection tube with fiber. Because the
fibers were so efficient in extracting radium from the water, in essence getting all
of the radium in a given sample of water, it was not considered necessary to
measure precisely the amount of fiber being used. For more extensive discussion
of measurement as a situated phenomenon, see Cicourel (1964), Lynch (1991),
and Sacks (1989).
122
Charles Goodwin
professor) had already successfully made the fiber (he had invented it). His
memory (as the sedimented product of prior practical engagement in the process)
can be used to shape not only his own actions but also those of the newcomers he
is supervising. In the lab, this is formalized into an organized system of
apprenticeship (e.g., a professor guiding the work of his students). In the lab
process examined here, the professor let his students carry out the tasks involved
in making the fiber (and also helped himself), while monitoring what they were
doing and evaluating their decisions.
Let us examine several examples of their interaction. Talk is transcribed using
the system developed by Gail Jefferson (Sacks, Schegloff, & Jefferson, 1974).
Talk receiving some form of emphasis (e.g., talk that would be underlined in a
typewritten transcript using the Jefferson system) is marked with bold italics .
Punctuation is used to transcribe intonation: a period indicates falling pitch; a
question mark, rising pitch; and a comma, a falling rising contour such as would
be found after a nonterminal item in a list. Comments (e.g., descriptions of
relevant nonvocal behavior) are printed in italics. Numbers in parentheses mark
silences in seconds and tenths of a second. To make it easier for the reader to find
a place in the larger transcript that is being discussed in the current analysis, I
have sometimes highlighted that talk by drawing a box around it.
In Example 1, Billy notices that one of his students, Gina, is about to check the
color of the fiber in the batch she is working on. He shakes his head from side to
side (i.e., No) (line 2) and says, "It' s not-" (line 3). She interprets this as an
evaluation of the current state of the fiber and asks (line 6), "It' s not even clo:se?"
Before she has completed the word clo:se, he overlaps with a No, affirming that
the fiber is notyet ready.
Here judgments about what constitutes the proper shade of black are calibrated
within the work group. The participants treat applying the category to the fiber as
something to be artfully accomplished, a topic for discussion in its own right.
Note that, although the view of the professor prevails, his word is not blindly
accepted . In lines 13-14, Gina offers a mild challenge to his assessment of the
situation ("I don know. It's not looking that far... "). He counters this proposal by
noting that the fiber "still has that tinge" (line 17). Through this exchange, her
attention is drawn to finer perceptual discriminations that she should take into
account in judging the color of the fiber. The notion of what can count as black is
not static but rather something that is progressively shaped and modified as
participants inspect the changing materials they are working with, while
interacting with each other.
123
It's not-
Gina:
Billy:
8L...---'-----;::7.':"---:------,;--;--;-----,:;-------:----;------'
10
11
12
13
(1.0)
Billy:
(1.9)
Gina:
14
15
16
17
18
Billy:
Gina:
19
20
21
22
23
24
I'don'know
it's not lookin that far (
How is it.
(An::: d)
It still has that-rimae.
L(Yamean tal
Ye(h)a(h)h.
(1.7)
Gina:
Billy:
We:ll,
Billy:
(1.2)
25
26
(0.9)
124
Charles Goodwin
his lab. Fiber that had reached the desired color was referred to as gorilla fur,
whereas fiber that was not yet the right color was called orangutan hair. Rather
than focusing discrimination entirely on two very similar shades of color, this
new set of terms contrasted two distinct types of animals, each of which
incorporated within its gestalt one pole of a color distinction that was relevant to
the activity at hand. The perceptual distinction that counted for those making the
fiber was thus highlighted. The humor in the new contrast set not only facilitated
memory and heightened salience but also incorporated an affective stance into the
perceptual distinction . In addition to color, the gorilla fur/orangutan hair contrast
encodes another dimension of the material being manipulated: its fibrous
qualities. It thus provides a richer evocation of the sensory environment of the
task than a color term alone would.
By being more salient, specific, concrete, and humorous, the new contrast
provides a tool that is simultaneously more powerful and better adapted to the
specifics of the environment within which it will function than the more abstract
jet black was. However, much research on the language used by scientists has
taken as its point of departure precisely the opposite set of assumptions. Thus,
Bernstein (1972) distinguishes elaborated codes from restricted codes specifically
in terms of how context-bound they are. Restricted codes that "sensitize their
users to particularistic meanings" are inferior to elaborated codes, the language of
science, which "orient their users toward universalistic meanings" (p. 164).
Similarly, Parsons' view of science emphasizes "conceptual abstraction from the
concrete" (Heritage, 1984, p. 19). In these data we find scientists moving in
precisely the opposite direction, actively inventing particularistic, restricted codes
when they already have access to far more universal, less context-bound
categories. Moreover, they have very good reasons for doing this: Categories
attuned to the particulars of the work that they are doing make relevant features of
that work more vivid and salient and thus help them to perform that work.
Viewing such issues from a broader perspective, Schegloff (1972) has argued that
the really difficult and interesting issues posed in the analysis of cognition
concern not the development of abstractions but rather the analysis of systematic
procedures capable of building the particulars of local events in a way that is
sensitive in detail to the structure of relevant context in those events .
If, following Vygotsky, we think of language as a tool for mediating our
relationship with the world (e.g., a term such as black mediates our perception of
the materials being worked with), the gorilla fur/orangutan hair contrast provides
an example of a second level of mediation being bootstrapped on top of a first.
These terms mediate between black and the fibers whose color is being assessed
by tailoring the general color distinction encoded in black to the perceptual tasks
faced in a specific local situation.
Recent work in the sociology of science has called into question traditional
notions of authorship by focusing attention on the contributions of workers, such
as lab technicians, whose crucial, highly skilled practice was central to the
findings reported in publications but made invisible there (Shapin , 1989). Here
125
we find that similar processes occur with cognitive structures as well. The
humorous contrast between gorilla fur and orangutan hair provides organization
for the perceptual work involved in making the fiber but disappears in
publication, to be replaced by a less useful but more abstract and general
category.
126
Charles Goodwin
It would be impossible to make a fine color discrimination while the fibers were
sitting in this liquid. Before any color judgment can be made, they must be
positioned for perception. To do this, a stick was used to lift a sample of fiber
from the liquid and hold it over the vat. The purple solution clinging to the fiber
on the stick was then washed away by spraying water on it.
Only after the fiber had been extracted from the very complex background
within which it was otherwise enmeshed could its color be evaluated. The activity
of assessing the fiber is thus supported not only by work-relevant cognitive
structures (e.g., an encompassing activity, category systems of various types), and
frameworks for interaction, but also by sets of situated practices for appropriate
manipulation of artifacts in the setting and by a tool kit that makes those practices
possible (e.g., plastic spray bottles of deionized water, sticks, vats, buckets,
chemicals). Although operating in the physical world, the practices engaged in
here are analogous to the cognitive structures given such prominence by the
gestalt psychologists (e.g., extracting a figure from a ground) and more recently,
used by anthropological linguists to shed new light on the nature of reference and
indexicality (Hanks, 1990, 1992) . Moreover, such processes of enhancement are
central to the production of knowledge in science more generally, constituting
what Lynch (1988) has termed an externalized retina. They are very important in
other work situations as well. For example, some of the most pervasive activities
found at the airport studied by Xerox PARC's Workplace Project involved tools
and practices designed to highlight phenomena (e.g., the use of yellow highlighter
to make information relevant to the task at hand stand out on a document) and
amplify perception (e.g., video links that allowed visual access to distant
locations). A major topic that has been virtually overlooked in the analysis of the
organization of gaze and other body behavior in face-to -face interaction is the
range of movements and activities involved in positioning for perception . The
cognitive task of assessing the color of the fiber in the lab would be impossible
without practices designed to extract the fibers from an irrelevant background and
to massage them into phenomena that are available for work-relevant perception.
Seeing Activities
The systematic incorporation of these practices into the activity of manufacturing
the fiber produced a framework for intelligibility that enables one party to make
inferences about what another is doing. In Example 1, Billy's negative head
shake, which was interpreted by Gina as proposing that the fibers were not ready,
occurred just after Gina positioned her squirt bottle over the vat. By seeing this
action within the larger framework of the activity system, Billy was able not only
to infer what she was doing but also to make an assessment about her perceptual
competence. These practices thus form part of the texture of intelligibility that the
participants are deploying to infer intersubjectivity and to make sense out of the
activities that they are collaboratively performing.
127
Embodied Cognition
Overwhelmingly, theories of cognition have divorced the mind from the body,
treating the latter (with the exception of the brain) as irrelevant to analysis of how
human beings think. This prejudice extends to other aspects of human social and
cultural behavior. Thus , Scheper-Hughes (1994) has noted critically that "The
body in social anthropology emerges as a passive, inert, dead weight attached to a
lively , responsive, nomadic mind, the true agent of culture" (p. 231) . Although
science is frequently depicted as the prototypical exemplar of disembodied
abstract thought, these geochemists were consistently attuned to what the
experience of working with the fiber might reveal to their bodies in an ad hoc
fashion as the activity unfolded. In Example 2, after making several batches of
fiber, a process that has required her to squeeze and manipulate the fibers at
various stages within the production cycle, Gina comments to the professor, Billy,
that, when the fiber is done, it has not only a particular color but also a distinctive
texture:
Exarnple2
Gina :
En it gets a certain
""1
to it
(0.3)
Billy:
Gina :
[Yes.
Also,
When it getstght. (fingers "feeling"fibers gesture)
Billy:
Gina :
Billy:
Gina:
Billy:
It certainl}1[~oes.
It gets sorta
It certainly does.
Billy immediately agrees with Gina's observation (he starts nodding before she
has finished the word texture). Alertness to the sensations experienced by her
body as she manipulates the fiber while evaluating its current state of progress
has revealed to Gina the possibility of another diagnostic criterion, one that is
available through touch rather than sight. This possibility was made known to her
not by instruction from her professor , but rather through the embodied process of
physically working with the fiber. The presence of her mentor is not, however,
irrelevant. By talking to him about what she has experienced, she is able to
transform what might otherwise remain private sensations and hypotheses into
128
Charles Goodwin
public events that can be evaluated and confirmed (or denied) by a more
competent practitioner. Note that an informed evaluation of her observation is
possible only from another body that has also physically worked with the fibers .
The embodied nature of the phenomenon that Gina and Billy are constituting
together is aptly demonstrated by the way in which both specify the experience of
the texture with hand gestures rather than words. A main focus of recent work on
gesture has been on how gestures externalize internal mental representations
(McNeill, 1992). The gestures performed here reveal a way of knowing that flows
in the opposite direction, from the hand as a sensory actor alive to the ad hoc
sensations it encounters as it works with external materials, to theories about how
those sensations are relevant to the accomplishment of the activities in progress.
The gesture points not to some hidden image lodged within the speaker's brain,
but instead to the hand as an agent of experience in its own right, encountering
specific phenomena in the world within which it is working . It is true that Gina is
trying to make Billy aware of something she has experienced, (e.g., a mental
event), a sensation of texture. However, rather than constituting a private point of
origin for the gesture, that experience is embedded within and emerges from the
embodied activity that the gesture makes visible. Indeed her interlocutor's ability
to recognize and evaluate the sensation she is talking about requires coparticipation in that same activity. The frameworks that make possible mutual
understanding of this gesture and of the sensation it makes visible are not
constituted by preformulated representations, but through co-membership in a
relevant community of practice.
The way in which Gina learns from her ad hoc engagement with a relevant
environment (i.e., the fibers she is manipulating), within a situation where she is
able to talk about what she finds with a more competent practitioner, is
compatible with Vygotsky's notion of a zone of proximal development (Cole,
1985). The way in which the professor sometimes learns from his students and
changes aspects of the manufacturing process in light of what they discover is
also compatible with Engestrom's (1987) expansion of the zone of proximal
development beyond individuals to processes of change within organizations.
129
But that prefaces her turn in line 21) by drawing his attention to the current
texture of the fibers (lines 21-22):
Example la
16
Billy:
17
18
19
Gina:
20
21
22
23
24
25
26
Ye(h)a(h)h.
1.
But it's gettin a little- (points with stick)
Billy:
(0.9)
Billy:
Assessment of the current state of the fiber is not made by the simple application
of a single category, the meaning of which is known in advance, but instead
emerges within a situated matrix of action encompassing multiple perspectives. In
this single sequence, color , texture, and temperature are all used to contest a
diagno sis. Some of the criteria being used to make the assessment are
pr ogressively changing (e.g., Gina's sense of what can count as an acceptable
black ), whereas others have only just been discovered by some of the participants
(e.g., Gina's recognition of the importance of texture). Rather than being
explici tly taught, many of these criteria are acquired through embodied
participation in the activity. The use of multiple criteria in this fashion cannot be
accounted for within the analytic framework of Berlin and Kay , who, like many
other cognitive anthropologists, carefully isolated for study a well-bounded
taxonomy restricted to a single perceptual domain . It is, however, consistent with
Wittgenstein's (1958) proposal that many categories are organized, not via
underlying essences, but rather in terms oi family resemblances: "a complicated
network of similarities overlapping and criss-crossing" (66) . When faced with
the practical task of locating a category, participants artfully make use of and
creatively discover a range of different methods drawn from a variety of sources.
130
Charles Goodwin
(1) The task itself and its material infrastructure (i.e., will the fiber actually
work?). Participants are not free to ignore the fit between the decisions they make
and the usefulness of the tools that are thus produced .
(2) What others will hold one responsible for.
There is thus a reciprocal relationship between the development of new tools
within a discipline and the development of socially organized structures of
perception by practitioners of that discipline. By working in concert with others
on relevant tasks, the body of the geochemist is transformed into a tool of the
trade. When Gina, one of Billy's graduate students, reached the shore of the
Amazon, she wanted to know approximately how much sea water it contained. To
determine this, she scooped a handful of the river water into her mouth and used a
chemist's sense of taste to evaluate its salinity.
Making judgments that others can not only recognize as appropriate but also
rely on constitutes being a competent practitioner. That knowledge can be gained
only through embodied practice, by working with the relevant materials and
having the judgments made in such circumstances evaluated by other competent
practitioners. Thus, by the end of the day (approximately one hour after Example
1), Gina was in a position to justify her decision to give a batch of fiber more time
by noting tinges in it (i.e., precisely the criterion to which Billy had earlier drawn
her attention) and to have Billy agree with that assessment after inspecting the
fiber himself:
Example 3
1
Billy:
2
Gina:
3
4
Billy:
6
7
8
9
10
11
12
13
Gina:
Billy:
Billy:
Gina:
Billy:
Billy:
The crucial importance of such embodied practice for being able to know what
black means within this activity system is well illustrated by my own inability to
make such a decision competently. I was physically present throughout the entire
fibermaking process . Several hours of that time were spent less than a yard from
131
the vat where these judgments were being made . Because I was videotaping the
process, I spent a lot of time looking at both the fiber and those working with it. I
was, however, outside the structures of accountability that linked Billy and Gina
to the fiber. I was not required to make decisions about when to withdraw the
fiber, did not have my color judgments evaluated by others, and did not feel the
fiber repetitively as it was undergoing the reaction .
Although I did not share with these geochemists a workable notion of jet-black
fiber so that they could rely on my judgment, this does not mean that their use of
the term was idiosyncratic. Instead, as Billy's careful work with Gina amply
demonstrates, the ability to make such an assessment was very much a social fact,
something that competent practitioners could hold each other accountable for and
that they were careful to teach to newcomers. The problem lies in assuming that
the language as a whole, as a cognitive structure abstracted from the messy details
of practice, is the place to study how categories encoding perceptual judgments
are organized. Instead , as argued by M.H. Goodwin (1990), the proper locus for
the analysis of culture, including the categories and practices through which it is
constituted, is not the society or disembodied language but situated activity
systems. By defining langue in the way that he did, Saussure found it impossible
to include human agency in the constitution of meaning or structure. For the
geochemists being analyzed here, determining what could count as black, so that
they could successfully perform the tasks in which they were engaged, was a
contingent, ongoing achievement, something that they had to work at to
accomplish (i.e., a prime locus for the analysis of human agency as a socially
embodied process).
Conclusion
The analysis in this chapter has focused on one of the central topics in
anthropological studies of cognition: semantic categories used to encode
perception of color. How does the view of cognition developed here differ from
that in the classic treatment of the subject by Berlin and Kay (1969)? Rather than
contradicting their findings, it uses as a point of departure a notion of what counts
as human cognition, and where it is located, that leads to the investigation of a
range of phenomena that were systematically excluded from the domain of
scrutiny they so ingeniously probed. It is useful to note explicitly some of these
differences.
132
Charles Goodwin
For Berlin and Kay, the primary objects of study , the phenomena they are
attempting to uncover and describe, are universal structures divorced from the
messy contingencies of situated practice. These structures are located in two
related places : the human brain and the semantic systems of particular languages .
This theoretical agenda led to a specific methodology : First, the relevant units of
analysis are separate languages (e.g., English is compared with Japan ese and
Tzeltal). Second, within each language, analytical criteria (e.g., that the term be
monoleximic, that its application not be restricted to a narrow class of objects) are
used to locate a small set of basic color terms. One effect of this is to isolate color
distinctions as a self-contained semantic domain; criteria from other sensory
modalitiesl? and from task-relevant use of color vocabulary are eliminated . Third,
in line with accepted experimental procedure in psychology, a standard stimulus
was prepar ed: an array of 329 Munsell color chips mounted on stiff cardboard.
Fourth, basic color terms were elicited from speakers of different languages.
Finally, these native speakers were asked to locate on the chart both the best
example of a specific color term and the boundaries of that term (e.g., all chips
that the term could validly designate).
These procedures were used to systematically collect data from a wide sample
of languages from many different parts of the world. However, the use of color
terms in locally relevant , endogenous activities was never probed . All speakers
were performing exactly the same experimental task, and, with the exception of
the Tzeltal speakers, all of the speakers resided in the San Francisco Bay area.
The notion of a relevant community of competent practitioners was completely
irrelevant to Berlin and Kay's analysis; for many languages, only a single speaker
was used. Thus, in a very strong sense, the basic analytic unit in these studies was
a context-free component of langue, located in discrete languages that were
treated as relatively homogeneous rather than as a set of endogenous speech
communities .
In contrast to this, the basic unit of analysis for the study in this chapter was the
situated activity system. Investigation was focused on a group of geochemists
who had to determine when a fiber they were working with was jet black in order
to know when to stop a relevant chemical reaction . Black and white are the most
basic color names in Berlin and Kay's analysis. However, for the geochemists, jet
black (i.e., the most prototypical example of black) was not a context-free
10Although as Berlin and Kay (1969) acknowledge in the very fIrst paragraph of their
book, Conklin (1955) had drawn attention to the presence of non-colorimetric information
in Hanunoo color words.
133
134
Charles Goodwin
specific category system to the activity they are engaged in: When the fibers
reach jet black, the reaction being monitored has to be terminated. The
encompassing activity thus provides a motivational framework within which
color discrimination becomes a relevant and expected thing to do .
Simultaneously, the structure of that activity sets parameters for what will count
as an acceptable solution to the task set by the relevant use of a color term (e.g.,
those shades of black that will produce usable fiber) . In turn the successful
accomplishment of that task leads to the deployment of a range of other practices
and tools (e.g., the invention of new category systems that highlight subtle
differences among similar colors, tools for extracting the material being examined
from a confusing background and positioning it for perception). The use of these
tools within the framework of the activity provides the participants with a visible
texture of intelligibility, enabling them to make inferences about what each other
is doing . By virtue of the encompassing activity, a heterogeneous collection of
very diverse phenomena - color categories, spray bottles, descriptions of animal
fur, sticks - is integrated into the accomplishment of a common cognitive task.
Proper use of these tools (what counts as "proper" is defined by the encompassing
activity) requires the mastery of socially organized embodied competencies (e.g.,
the ability to see, feel, smell, and taste as a geochemist). Rather than being private
perceptual structures lodged within the individual brain, such professional vision
(C. Goodwin, 1994) is socially organized by the tasks set by activities, such as the
one investigated in this chapter, and is something that members of the
communities responsible for doing these activities hold each accountable for if
one is to be recognized as a competent practitioner. Using the situated activity
system as a basic framework for analysis thus opens up to systematic study an
expanded view of human cognitive activity.
In searching for cognitive universals, Berlin and Kay were reacting against
ethnographic particularism. It might be argued that focusing on the situated
activity system leads right back to particularism; for example, instead of
examining color distinctions common to all speakers of English or even to all
geochemists, this chapter has investigated a perceptual discrimination used in the
work practices of a small group of geochemists. However , other ethnographic
work has demonstrated that versions of many of the practices described here are
found generally. Consider, for example, highlighting: the way in which the
geochemists extracted the fiber they wanted to examine from a confusing
background so that its perceptual salience was emphasized, and it was positioned
for focused, intense scrutiny. Similar practices constitute part of the professional
craft of archaeologists who both lift objects from the dirt to examine them and
annotate the earth with lines drawn with such tools as trowels, brightly colored
flags, and bits of string in order to make dim features stand out from a confusing
135
background (for a detailed analy sis, see Goodwin, 1994; Lynch, 1988) . Such
highlighting is an instantiation in concrete practice of a most general cognitive
structure, the figure-ground relationship. As noted earlier, highlighting of
documents, so that information of relevance to a particular workgroup is made
salient, was one of the most general work practices found at the airport studied by
the Xerox PARe Workplace Project. By virtue of the way in which such
highlighting structures the perception of others, by reshaping a domain of scrutiny
so that some phenomena are made salient, whereas others fade into the
background, it can have strong rhetorical and political consequences. The lawyers
defending the policemen who severely beat an African American motorist,
Rodney King, highlighted the videotape of the beating through gesture, category
systems, and by drawing white lines around Mr. King's body in order to focus the
attention of the jury on "aggressive" body movements of Mr. King and away
from the actions of the policemen beating him (Goodwin, 1994) . In brief, human
cognitive activity characteristically occurs in environments that provide a very
complicated perceptual field. It is, therefore, not surprising that a general class of
cognitive practices consists of methods for structuring that perceptual field so that
phenomena relevant to the activity in which participants are engaged are made
salient, a process that simultaneously helps classify those phenomena (e.g., as an
archaeological feature rather than an irrelevant patch of color in the dirt, or as an
aggressive movement). However, such processes remain outside the domain of
what can be studied, if the notion of what counts as cognition is restricted to
structures hidden inside the brain . Moreover, a central component of this process
is the framework of relevance provided by the situated activity system within
which the act of highlighting is embedded. Practices such as highlighting
precisely link relevant features of the setting to the activity being performed in
that setting. When setting and activity are lost, these cognitive practices
disappear. The issue, therefore, is not particularism but rather access to a range of
basic cognitive processes that requ ire for their analysis detailed study of actual
work in endogenous settings (Scribner, 1984).
An excellent example of how scientists use highlighting to make complex
phenomena amenable to rigorous investigation can be found in the procedures
developed by Berlin and Kay to extract the domain of basic color terms from a
very confusing environment. To accomplish this, Berlin and Kay had to proceed
on several fronts, developing, on the one hand , semantic criteria that would apply
across languages to systematically separate a small set of basic terms from the
much larger color vocabulary found in each language and, on the other hand,
constructing a relevant perceptual target, the Munsell chart, that would enable
explicit comparison between languages. Assembling this package of procedures
so that basic color terms uncontaminated by extraneous phenomena could be
rigorously measured and compared was a major accomplishment that led to new
knowledge about how the brain and language work together to structure
perception. As such practices reveal, however, the context-free universals of
Berlin and Kay were themselves shaped into distinct kinds of entities and made
136
Charles Goodwin
Acknowledgments
I am deeply indebted to Dr. Willard S. Moore for making possible the research in
his lab and to Alessandro Duranti, Cathryn Houghton, Ed Hutchins, Candy
Goodwin, Elinor Ochs, Billy Moore, Clotilde Pontecorvo, and Curtis Renoe for
helpful and insightful comments on an earlier version of this analysis.
137
References
Baker, G. P., & Hacker, P. M. S. (1980). Wittgenste in: Understanding and meaning .
Chicago : The University of Chicago Press.
Berlin, B., & Kay, P. (1967). Universality and evolution of basic color terms. (Working
Paper #1). Berkeley, CA: Laboratory for Language Behavior Research.
Berlin, B., & Kay, P. (1969) . Basic color terms : Their universality and evolution.
Berkeley, CA: University of California Press.
Bernstein, B. (1972) . Social class, language and socialization. In P. P. Giglioli (Ed.),
Language and social context (pp. 157-179). New York: Penguin.
Brun-Cottan, F., Forbes, K., Goodwin, C., Goodwin, M. H., Jordan, B., Suchrnan, L., &
Trigg, R. (1991). The workplace project: Des igning for divers ity and change.
(Videotape) Palo Alto, CA: Xerox Research Center.
Bruner , J., Oliver, R. R., & Greenfield, P. M. (1966). Studies in cognitive growth. New
York: John Wiley and Sons.
Chaiklin, S., & Lave, J. (Eds.). (1993). Understanding practice : Perspectives on activity
Gild context. Cambridge, England: Cambridge University Press.
Cicourel, A. V. (1964). Method and measurement in sociology. New York: Free Press.
Cole, M. (1985). The zone of proximal development : Where culture and cognition create
each other. In J. Wertsch (Ed.), Culture, communication, and cognition: Vygotskian
perspectives (pp. 146-161). Cambridge, England: Cambridge University Press.
Collins, H. M. (1985) . Changing order: Replication and induction in scientific practice.
London: Sage.
Conklin, H. (1955). Hanunoo color categories. Southwestern Jounal of Anthropology, 11,
339-44.
Damasio, A. R., & Damasio, H. (1992 , September). Brain and language . Scientific
American, 89-95.
Enge strom, Y. (1987) . Learning by expanding : An activit y-theoretical approach to
developmental research. Helsinki: Orienta-Konsultit Oy.
Engestrom, Y., & Middleton, D. (Eds.). (in press). Cognition and communication at work.
Cambridge, England: Cambridge University Press.
Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall.
Goffman , E. (1961). Encounters: Two studies in the sociology of interaction. Indianapolis,
IN: Bobbs-Merrill .
Goodenough, W. H. (1956). Componential analysis and the study of meaning . Language,
32, 195-216.
Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 606-633.
Goodwin , C. (in press). Transparent vision. In E. Ochs, E. A. Schegloff, & S. Thompson
(Eds.), Interaction and grammar. Cambridge, England: Cambridge University Press.
Goodwin , c., & Goodwin , M. H. (in press). Formulating planes : Seeing as a situated
activity . In Y. Engestrom & D. Middleton (Eds.), Cognition and communication at
work. Cambridge, England: Cambridge University Press.
138
Charles Goodwin
Goodwin , M. (in press) . Co-construction in girls ' hop scotch. Research on Language and
Social Interaction.
Goodwin, M. H. (1990) . He-said-she-said: Talk as social organization among black
children. Bloomington, IN: Indiana University Press .
Greenfield, P. M., & Bruner, J. S. (1966). Culture and cognitive growth . International
Journal ofPsychology, 1, 89-107.
Hacker , P. M. S. (1986). Insight and illusion: Themes in the philosphy ofWittgenstein (rev.
ed.). Oxford, England: Oxford University Press.
Hanks, W. F. (1990). Referential practice: Languag e and lived space among the Maya.
Chicago: University of Chicago Press.
Hank s, W . F. (1992) . The indexical ground of deictic reference. In A. Duranti & C.
Goodwin (Eds.), Rethinking context: Language as an interactive phenomenon (pp. 4376) . Cambridge, England: Cambridge University Press.
Heath, C. (in press) . Virtual looking : Spatial transformation and communicative
asymmetries. In P. Pelligrino (Ed .), Proceedings of the Colloquiem on the Semiotics of
Space. Geneva, Switzerland: University of Geneva .
Heath, C., & Luff, P. (in press). Convergent activities: Line control and passenger
information on the London underground. In Y. Engestr6m & D. Middleton (Eds.),
Cognition and Communication at Work. Cambridge, England: Cambridge University
Press.
Heidegger, M. (1962). Being and time (1. Macquarrie & E. Robinson, Trans.) . New York :
Harper & Row.
Heritage, J. (1984). Garfinkel and ethnomethodology. Cambridge, England : Polity Press.
Hutchins, E. (1991). The social organization of distributed cognition . In L. Resnick, 1. M.
Levine , & S. D. Teasley (Eds.), Perspectives on socially shared cognition (pp. 283-307) .
Washington, DC: American Psychological Association .
Hutchins , E. (1993). Learning to navigate . In S. Chaiklin & 1. Lave (Eds.), Understanding
practice: Perspectives on activity and context (pp . 35-63) . Cambridge, England:
Cambridge University Press.
Jefferson, G. (1987). Exposed and embedded corrections. In G. Button & J. R. E. Lee
(Eds .), Talk and Social organisation (pp. 86-100) . Clevedon, England: Multilingual
Matters Ltd.
Knorr-Cetina, K. (1981). The manufacture ofknowledge. Oxford, England: Pergamon.
Latour, B. (1987). Science in action: How to follow scientists and engineers through
society. Cambridge, MA: Harvard University Press .
Lave, J., & Wenger, E. (1991). Situated learning : Legitimate peripheral participation.
Cambridge, England: Cambridge University Press.
Lynch, M. (1985).Art and artefact in laboratory science. London: Routledge and Kegan
Paul.
Lynch, M. (1988). The externalized retina : Selection and mathematization in the visual
documentation of objects in the life sciences . Human studies, 11, 201-234.
Lynch, M. (1991). Method: Measurement - ordinary and scientific measurement as
ethnomethodological phenomena. In G. Button (Ed.), Ethnomethodology and the human
sciences (pp. 77-108) . Cambridge, England: Cambridge University Press .
139
Lynch, M., & Woolgar, S. (Eds.). (1988). Representation in scientific practice. Cambridge,
MA : MIT Press .
McNeill, D. (1992) . Hand &: mind: What gestures reveal about thought. Chicago:
University of Chicago Press.
Moore, W. S. (1976) . Sampling Ra<228> in the deep ocean . Deep-Sea Research, 23, 647651.
Parsons, T., & Shills, E. (1951) . Toward a general theory of action . Cambridge, MA:
Harvard University Press .
Pickering, A. (Ed.) . (1992) . Science as practice and culture. Chicago: The University of
Chicago Press .
Polanyi , M. (1966) . The tacit dimension. Garden City, NY: Doubleday .
Rogoff, B. (1990) . Apprenticeship in thinking. New York : Oxford University Press .
Rogoff, B., & Lave , J. (Eds .). (1984) . Everyday cognition: Its development in social
context. Cambridge, MA: Harvard University Press.
Roskill , M. (Ed.). (1983) . The letters of Vincent van Gogh. London: Fontana Paperbacks.
Sacks, H. (1972) . An initial investigation of the usability of conversational materials for
doing sociology. In D. N. Sudnow (Ed.), Studies in Social Interact ion (pp. 31-74) . New
York : Free Press .
Sacks, H. (1989). On members ' measurement systems. (Edited by G. Jefferson from
unpublished lectures : 1966, 1967, 1970.) Research on Language and Social Interaction ,
22,45-60.
Sacks, H. (1992) . Lectures on conversation: Volume I. (Edited by G. Jefferson, with an
Introduction by E. A. Schegloff. ) Oxford , England : Basil Blackwell.
Sacks, H., Schegloff, E. A., & Jefferson, G. (1974) . A simplest systematics for the
organization of turn-taking for conversation. Language, 50, 696-735 .
Saussure, F. d. (1959) . Course in general linguistics . (Edited by Charles Bally & Albert
Sechehaye, in collaboration with Albert Riedlinger; W. Baskin, Trans .) New York:
Philosophical Library .
Schegloff, E. A. (1972). Notes on a conversational practice: Formulating place. In D.
Sudnow (Ed.), Studies in social interaction (pp. 75-119) . New York: Free Press.
Schegloff, E. A. (1992a) . On talk and its institutional occasions . In P. Drew & J. Heritage
(Eds .), Talk at work: Interaction in institutional setting (pp. 101-134). Cambridge,
England : Cambridge University Press .
Schegloff, E. A. (1992b) . Repair after next turn: The last structurally provided defense of
intersubjectivity in conversation. American Journal ofSociology, 97(5), 1295-1345.
Scheper-Hughes, N. (1994) . Embodied knowledge: Thinking with the body in critical
medical anthropology. In R. Borofsky (Ed.), Assessing cultural anthropology (pp. 229239). New York: McGraw -Hill.
Scribner, S. (1984). Studying working intelligence. In B. Rogoff & J. Lave (Eds.),
Everyday cognition: Its development in social context (pp. 9-40) . Cambridge, MA :
Harvard University Press .
Shapin , S. (1989) . The invisible technician . American Scientist, 77, 554-563.
Star, S. L. (1988 a). Introduction: The sociology of science and technology . Social
Problems, 35(3), 197-205 .
140
Charles Goodwin
Star, S. L. (1988b). The structure of ill-structured solutions: Heterogeneous problemsolving, boundary objects and distributed artificial intelligence. In M. Huhn s & L.
Gasser (Eds.), Distributed artificial intelligence 2 (pp. 37-54). Menlo Park, CA: Morgan
Kaufmann .
Suchman, L. (1992). Technologies of accountability: Of lizards and airplanes . In G. Button
(Ed.), Technology in working order: Studies of work, interaction and technology (pp.
113-126) . London : Routledge.
Suchman, L. A. (1987) . Plans and situated actions: The problem of human machine
communication. Cambridge, England : Cambridge University Press .
Wittgenstein, L. (1958). Philosophical investigations. (Edited by G. E. M. Anscombe & R.
Rhees ; G. E. M. Anscombe, Trans ., 2nd ed.). Oxford , England : Blackwell.
Part Two
Chapter 6
Reasonable Uncertainties:
Parents' Talk About Caring for Children
with Chronic Renal Failure
David Middleton
Department of Human Sciences, Loughborough University, LEI I 3TU, United Kingdom
Abstract
Examples of parents' talk about their care of children with chronic renal failure are
discussed. These are taken from a series of parent support group meetings run over a
32-month period at a Regional Paediatric Renal Unit within the U.K. National Health
Service. All the par ents are directly involved in dialysizing their children at home
rather than in hospital settings. The success of these home-b ased treatments is
dependent on paren ts achieving a sophisticated level of paramedical skills and dietar y
knowledge. The analysis examines the discursive functioning of these support groups
in establishing and maintaining what parents argue as ordinary within the extraordinary circumstances of their lives. Parents tell of routine dilemmas they face in the
care of their children. The analysis demonstrates the way parents use these dilemmas
as a resource for arguing the basis of what, for them, constitutes reasonable outcomes
and circumstances in the care of their children- reasonable both as a form of jointly
accomplis hed reasoning over the salience of their experiences as parents, and
reasonable as acceptable in terms of the circumstances of the lives they live.
144
David Middleton
Introduction
This chapter examines the talk of people who have been projected into a
technologically supported environment by force of circumstance. They are
parents of children who have suffered various forms of kidney failure. To
maintain the viability of their children 's lives, either prior to kidney
transplantation or after any episodes of rejection, these parents are offered the
option to be trained to engage in complex medical and dietary procedures. This
involves the use of technologically sophisticated medical equipment, associated
consumables , drugs, and the maintenance of a carefully monitored dietary regime .
Examples of parents ' talk about their involvement in this demanding and timeconsuming health care are the focus of analysis . My aim is to examine how
parents argue the basis of what, for them, constitutes reasonable outcomes and
circumstances in the care of their children-reasonable both as a form of jointly
accomplished reasoning over the salience of their experiences as parents, and
reasonable as acceptable in terms of the circumstances of the lives they live.
The examples of transcribed talk are taken from a series of parent support group
meetings run over a 32-month period up to the point of this analysis at a Regional
Paediatric Renal Unit within the U.K. National Health Service. All the parents are
directly involved in management of the dialysis of their children at home rather
than in hospital settings. The particular forms of peritoneal dialysis used allow for
the treatment of infants from birth onwards.
The success of these home-based treatments is dependent on parents
developing their competence in a range of paramedical skills and dietary skills .
To achieve this, systematic training in the necessary procedures is offered by
members of a Regional Paediatric Renal Team. The direct involvement of parents
in peritoneal dialysis , either as a precursor to transplantation or as a result of
kidney transplant rejection , locates medical care as part of domestic rather than
hospital-centered routines (see Collier & Watson, 1994).
One of the main benefits for the children is that their exposure to long periods
of hospitalization are reduced. However , the extra burden of care places heavy
demands 'on the parents , both in terms of the disruption to conventional routines
of domestic life, and in terms of the emotional stress of adapting to and living
with a child who has a life-threatening chronic illness. This increased burden of
care sets families apart from the normal run and organization of domestic and
community life and makes for a form of daily routine and experience that is by
definition out of the ordinary.
To help ameliorate this burden of care, particular attention is paid by the
Paediatric Renal Team to what is termed psychosocial support. Such support
Reasonable Uncertainties
145
Analytical Approach
The analytic approach adopted here to the parents' accounts of health care
experiences represents an inversion of the conventional methodological positions
in health-related psychological studies. The conventional approach in
psychosocial studies of health care is to take description as mapping in some
indirect way onto peoples' perceptions of the realities of health care experience
(cf. Abraham & Hampson, in press). In contrast, the focus of analytic inquiry here
is the to-be-established nature of peoples' experiences accomplished in their talk
about the care of their children. This research forms part of a body of work
concerned with developing a discursive psychology informed by conversational
analysis and discussions in ethnomethodology (see Atkinson & Heritage, 1984;
Billig, 1987; Billig et aI., 1988; Buttny, 1993; Edwards & Potter, 1992; Potter &
Wetherell , 1987; Sacks, 1992). Within such an approach, what and how people
communicate in talk and text about the world in which they live, and their
psychological states, are treated as categories and interpretative resources that
they themselves use to construct accounts of their lives and their accountabilities
within those lives.
This chapter examines the construction of health experiences through peoples'
descriptions and accounts of those experiences. Such accounts are not treated as
representing the facts of the matter but as functional constructions that handle
exigencies of the interactive situation of their involvement in health care. The
particular focus is on the way parents voice uncertainties about the circumstances
and outcomes of their children's care. Their care experiences are examined as
IThe Paediatric Renal Team incorporates within its core membership a permanent renal
social worker, a clinical psychologist, and a health psychologist, in addition to specialized
nursing staff, consultants, associated doctors, dietitians, and a hospital teacher.
146
David Middleton
socially intelligible through an analysis of the way they construct accounts that
are functional to the circumstances in which they take themselves to be in the
support groups (see also Middleton, 1996a).
Summary Aims
Overall, the analysis examines the discursive functioning of these groups in terms
of what parents argue as "ordinary" within the extraordinary circumstances of
their lives . The aim is to demonstrate how the narration of parental experience
affords a discursive context for exploring dilemmas and difficulties in the care of
their children , and how parents use these dilemmas as a resource for arguing the
basis of what, for them, constitutes reasonable circumstances in the shared care of
their children and health-related outcomes for their children.
The analysis takes up these issues in a number of ways. First, examples are
presented of how parents formulate their memberships of the support group .
These include the way they report the support group's value in terms of the way
participation enables sharing of common problems with other parents and the
development of common understandings concerning the processes and outcomes
in the type of treatment they are all involved in. Their working of group
membership is also examined in terms of the way they contrast their
circumstances and understandings with other people they come into contact with
(including medical experts, the general public, acquaintances, and family
relations). Second, the organization of the parents ' talk is examined for the way
collaborative completions accomplish and display affiliation between parents as
group members.
The analysis then moves to examine directly how parents' reasoning about the
circumstances and outcomes of their care of their children is organized in terms
of the uncertainties about what to expect. Uncertainties of circumstance are
illustrated with an example of talk about the practical dilemmas of mundane child
care practices in the unusual circumstances of child care that parents face as a
matter of routine experience . Uncertainities of outcome are illustrated in terms of
the sorts of expectations it is reasonable to assume might attend their children's
care. Finally, the analysis illustrates three recurring problems in parents' attempts
to grapple with uncertainties in presenting and re-presenting the circumstances
and outcomes in the care of their children: uncertainties in representing the
implications and consequences of transitions and changes in the regimes of care
children are subject to (e.g., transitions from home to hospital); uncertainties
about the circumstances families find themselves in now (e.g., are we doing it
right? how far can we deviate from prescribed care schedules?); and uncertainties
of trustworthiness (e.g., to whom can parents entrust the care of their children).
Reasonable Uncertainties
147
Anne
Mrs . T
3
4
Mrs . T
Mr. 'IN
Mrs. C
Mrs . C
Mrs . TH
Mr. I
148
David Middleton
Mr. I
10
Mrs. I
11
12
13
Mrs . ?
Mrs. I
Mr. 'IN
14
15
Mrs . I
Mrs . N
16
Mrs . I
17
Mrs. TH
18
19
20
Mrs . I
Mrs . TH
Mrs. IJ
21
22
Mrs. TH
Mrs. IJ
23
Mrs . IJ
24
Mrs. I
24
Mr. I
Reasonable Uncertainties
149
Problems in Common
The changing status of becoming a parent of a child with "kidney problems" (8:
Mr . I) is constructed as entering a condition of being the "odd one out" (8). His
contribution takes up what it is to be a parent in these circumstances and
expresses an argument concerning the potential benefits of group attendance.
Mrs. TH has already enumerated the benefit of the group as a conversational
resource in the organization of contact beyond the meeting . Potential attendance
at the meeting provides a topic people can orient to in their incidental contacts, in
addition to being an actual location where "new people especially can feel that
they belong to something" (7). Mr. I's contribution does more than declare his
particular feelings, or at least make a claim that his feelings can be construed as
such. His contribution is constructed in such a way that it provides an opportunity
for other parents to align themselves with his conclusion and in so doing provides
the basis for corroborating his claim as being generally legitimate for parents in
such circumstances.
Mr. I presents the changing status of becoming a parent of a child with "kidney
problems" as entering a condition of being the "the odd one out" (8). His reported
speech of the news "oh my child's got kidney problems" is the premise for a
claim in general with others that "you tend to feel as though you are the odd one
out" (9). This drew agreement from others present. Its form as a piece of reported
speech is also interesting. In moving from a general claim in the second personal
plural ("you tend to feel") to something that can be heard as him reporting his
own experience ("nobody else knows what I'm going through and I've got all
these problems") makes it available as something others might care to go along
with without taking as a given that everyone shares the same experiences. The
first person reporting ironises his experiences and in so doing makes them
150
David Middleton
The group then proceeds to identify further aspects of what it is to occupy such a
unique status as parents handling the routines of dialysis and medical care, a
status that marks them as different from others in ways that only insiders can
understand. A whole range of social contacts with others (be they medical
professionals, other folks in general, and even family members) are argued as
challenging the assumptions on which contact with such people is usually
conducted. Theirs is a different world for all sorts of reasons.
Reasonable Uncertainties
151
Even professionals with medical knowledge can fail to understand the nature and
normality for them of their expertise. Me. TN (13) presented a short account of
his contact with a chemi st (pharmacist). The specific details concerning the
reason he needed to visit the chemist are imprecise : "I had to pick up some new
medicine or something" (13). This serves to emphasize the ordinariness of his
task. As part of such a mundane commercia l transaction, the ordinariness of his
position in having access to the use of the gastrostomy button' was expressed as
"simply said to him quite naturally" (13). The claim that "he hadn 't a clue" (13)
about what he was referring to despite "working in a chemist shop" (13) is
presented as the evidence for the specialty of the world they as parents inhabit.
Even professionals associated with medical care do not know about the most
basic procedures they as parents have to deal with routinely. What is ordinary to
them is extraordinary and unknown to others, even those who have medically
related expertise. These parents inhabit a world whose normality is defined in
terms of uncommon and irregular circumstances .
Again , in representing the group as a place where you can exercise very basic
forms of social engagement, "yes so nice to be able to get together for a chat
really" (16: Mrs. I), while at the same time knowing that the person you are
talking to "can rel ate" (Mrs. TH : 17) to what you are talking about, the
functionality of group attendance is contrasted with a social contact with others
from the outside who "tend to shut off' (Mrs. 11: 20). (The organization of this
particular sequence will be examined in more detail shortly.)
David Middleton
152
they just just go away" (24). Again, dilemmas in the extraordinariness of their
circumstances provide a resource for identifying criteria for accomplishing
commonalities of experience in terms of what is reasonable to expect.
Apart from the argument concerning what the group affords and the joint
construction of warrants for such arguments in terms of insider/outsider
boundaries, there are some interesting features in the way the organization of
their talk accomplishes and displays affiliation among the parents in their group
participation. In Sequence 1 (Sections 15-21), the parents do not just "talk about
the same sorts of things" (Mrs. N: 15); they mutually complete and repair each
other's talk. Their interruptions do not have to do with establishing alternative
outomes but with contributing to build up the same project.
Mrs . N
16
Mrs . I
17
Mrs . TH
18
19
20
Mrs . I
Mrs. TH
Mrs. IJ
21
22
Mrs. TH
Mrs. IJ
The projected termination of Mrs. I's turn (concluding what she may, in fact
"know" about the consequences of group participation) is preempted by Mrs. TH:
"you're not boring anybody because they can relate to it" (17). This warrant for
chatting , knowing that you share like concerns, is ratified by Mrs. I's expression
of agreement. Mrs. TH's continued comparison with the "outside" is, in turn,
Reasonable Uncertainties
153
interrupted by another speaker, Mrs. Il, who offers a reason it is difficult to talk to
those outside: because they "tend to shut off' (20). This, in turn, is both repaired
and ratified by Mrs. TH: "shut up now yeah"(21). However, her repair is qualified
in Mrs. IJ's next turn, where she qualified the implication of "shut off' as not
listening because "they don't know what you're talking about so they don't
listen" (22). The conclusion to be drawn from no response is that people have no
common experience to share with those in the know.
Such close coupled joint reasoning accomplishes a mutuality of common-sense
understandings of the situation. These interruptions do not signal that speakers
have alternative communicative projects to pursue. They do not interrupt to
propose alternative conversational consequences. Their completions of what
someone else has begun display sharing the same description of the world rather
than an alternative one (Perakyla, 1994). Lerner (1993) has examined how such
collaborative completions are part of the way in which people make relevant to
their action their memberships of groups. The collaborative completions in this
sequence mutually display the parents' involvement in common circumstances.
This no longer has to be argued for. In their talk, they practically constitute a
community of experience, what it is that they hold in common.
Mr . I
154
David Middleton
3 Mrs. I
4 Mr. I
5
6
7
8
Reasonable Uncertainties
155
Mrs . TIS but they were very urn . (.) 0 : h serious when she
was in urn baby unit I expected her to be enn well
have loads of things round her she was actually in
an incubator for about 24 hours. but having it
been three weeks I expected her to be in there
three months you know it was a bit of a shock when
they said she could corne horne urn but they were
very ..ill:.:.ious I mean they were ( . .. )
2?
huh huh huh
3
Mrs . TIS yes so serious you thought you know is she going
to pop off tomorrow or the day a : :fter uh huh you
know you didn't know quite how how to feel urn but
I got used to how doctors talk so huh huh huh
4
Mrs . ST when we brought Jeffrey in we were prepared for
quite a long stay so surprised when he said we
could go horne you know
5
Mrs . TIS I suppose it's nice if you are prepared for the
worst but things aren't that bad then. (.) you
know
6
Mrs . ST Doctor Smith said to us how long do you think
you'll be in for I said oh about 'til Christmas he
says no you 'll be back before then before
Christmas and it was such a shock when he said you
can horne you know
7
Mrs. I
we found that when we carne ' caus e we carne on the
Monday afternoon at tea time and we'd been in er
we'd had 5 weeks? 6 weeks in {City 2} in and out
of hospital . staying for a fortnight at a time and
we carne here on the Monday and we thought <I mean
to say I brought clothes for a fortni : :ght huh hu
$
8
Mrs . TIS hah hah hah
156
9
10
11
12
13
14
15
16
17
David Middleton
Mrs. I
Reasonable Uncertainties
157
158
David Middleton
know urr:: you look at them like a fish thats dried out of water don't you? got
(.). you know (.) couldn't believe-" (7; 9)
This is no mere repetition of the previous speakers' reasons and experience. She
deploys evidence in terms of her preparations and her stunned reactions as
detailing the equivalence of her experience. But "even nicer than that, "Mr. I.
(12) caps the telling through the unexpected outcome of being able to go on
holiday:
"even nicer than that we'd actually booked to go over to Jersey on the
Saturday. {City 2} had said we'll give you all the letters for your insurance
and everything. you can forget your holiday and all the rest of it. transferred
us down here on the Monday er:m with us having lots of problems with
Jeffrey draining out and er within a few hours of us being down here they
whipped his tube out he was of- off dialysis all together and on a low protein
diet and er. within the week we'd got clearance to go to Jersey for us
holiday" (12)
Reasonable Uncertainties
159
their children that were identified in the seven sessions that are the subject of
analysis in this chapter. They involved uncertainties in representing the
implications and consequences of transitions and changes in the regimes of care
children are subject to; uncertainties about the circumstances families find
themselves in now; and uncertainties of trustworthiness (e.g., to whom can
parents entrust the care of their children).
Sequence 5 is taken from the initial part of Mr. and Mrs. TN's account of their
initial experiences in dealing with the first few weeks of their preparation for
160
David Middleton
Reasonable Uncertainties
161
Mrs. 'IN $ then she comes off at half past seven "half seven"
Mr. 'IN at least with the dialysis machine we find that we
can jygge when to put it on and when we want to get
it off but we've compared notes with other people
about using the overnight fe :ed $
6
?
tuh
7
Mr. 'IN $ and i t seems to have a mind of it's own $
8
?
[yeh
9?
[of course
10??
[huh huh huh
11 Mr. 'IN $ and it's very difficult
5
1
2
3
4
Mrs L
you were talking about gastrostomy ( . . . )$
Mr. 'IN ermn ermn
Mrs. L $ do you find it's not finished when you want to take
her off in the morning
Mrs. 'IN yeah but on the quiet occasion we want to get up we
just 0 ( . . ) 0 don 't tell Jean hu huh
?? ?
hah huh huh
Mrs. 'IN ( ... ) seven hundred or so she's on seven hundred a
night and it says seven hundred and twenty nine and
you're only putting about seven hundred in so it's
not as if ( .. . ) but she eats a bit of it
?
( ... )
Mrs .?
it takes a while to just get it right
9
Mr. 'IN nm nm
10?
it depends on what bag you use as well
8
162
David Middleton
Uncertainties of Trustworthiness
In addition to the working of extraordinary normalities, there are also discussions
about uncertainties of trustworthiness: whom from the outside can they trust, be
they medical, lay persons, acquaintances, or relatives:
Uncertainties of trustworthiness
Sequence 8 (Session 5)
1
Mr . 'IN
2
3
4
?
???
?
. .. )
Reasonable Uncertainties
5
6
7
8
9
10
11
12
13
14
15
16
17
18
163
Within any normal pattern of child-rearing, issues of whom to trust your children
with is a typical concern. For parents in these circumstances, the basis for giving
over the responsibility of the care of their child to others is very problematic. In
the preceding sequence, the parents are describing their first use of the babysitting service that utilizes people skilled in dialysis. This afforded both the
possibility of going out and the possibility of handing over the responsibility for
dialysis for one night. The description of taking advantage of such an opportunity
was met with considerable laughter by the group (8). They clearly recognized the
trade-off in responsibilities being reported. In addition the issue is one of
knowing that those with whom you leave your child are known to have expertise
in the special regime of care. However, it is not enough for you to know that they
might be competent. Others at the hospital who might be called on to give advice
in an emergency need to know that the person left in charge is competent: "it's
nice to be able to go out and think well if anything goes wrong they can pick the
phone up and. phone somebody and talk to somebody at the other end that- they
know that she knows" (Mrs . I: 12; 15). Notwithstanding this, in the particular
circumstance being reported by Mr. and Mrs. TN, their intimate knowledge of the
child 's recent reactions and difficulties still places them in a dilemma concerning
whether they should or should not have relinquished temporary responsibility: "I
was sitting in the theatre thinking ah she has got peritonitis in there a blocka::ge
[huh huh huh ] and you know thinking are they going to get in touch with us you
164
David Middleton
Conclusion
This chapter has aimed to demonstrate how joint ratifications of entitlements to
experience accomplish commonalities concerning the extraordinary
circumstances of home-based peritoneal dialysis of children with renal failure.
The groups do more than consolidate some parent support network. The
communicative action in the groups works consensus about the ordinariness of
the extraordinary circumstances within which parents live their lives. The
descriptions of experience become organizable in terms of a consensus
concerning the legitimacy of that experience as commonly understood. Common
ownerships are worked." The point is that what it is to be ordinary within these
extraordinary circumstances is something that is the accomplishment of the group
discussions: A normativeness is accomplished that has no basis beyond the
parents' accountabilities as carers in these extraordinary circumstances.
The talk that makes up those discussions is organized through accounts that
express dilemmas that handle uncertainties about what is reasonable to expect as
parents who find themselves coping with the rigors of this form of shared care .
This is not to claim that the uncertainties outlined and illustrated exhaust the
possible dilemmas that parents may express in the future or those they might have
expressed in the past. Indeed, that is precisely where the generative potential of
this type of communicative action resides. Uncertainty concerning the nature of
their care of their children is a recurring feature of their talk. Resolving such
uncertainty is always provisional; alternative versions and interpretations are
continually made available (see also Middleton, in 1996b; Middleton & Curnock,
1995). This gives such talk about care its contradictory and dilemmatic quality.
4 See Perakyla and Silverman (1991) for a conversation analytic examination of owning
experience in multiparty AIDS counseling sessions.
Reasonable Uncertainties
165
As a parent dealing with the idiosyncrasies of chronic illness, you are not part or
at least cannot assume to be part of some "default" consensus about what, how,
when, with whom, for what purposes you should proceed in the care of your
child. This uncertainty of how you should represent the circumstances you are in
is definitional of what it is to be the parent of a child with a chronic illness. Under
the more routine circumstances of parenthood, we can on the whole assume there
is a solution , a way, something acceptable, appropriate in the way we conduct
ourselves. Such default assumptions are precisely what is missing in caring for
children who have developmental difficulties or are diagnosed as having some
form of chronic illness. The accomplishment of the talk in the parents' support
groups discussed here is the ratification they afford of forms of "ordinarinesses"
in the extraordinary circumstances of the parents' child care and in terms of
outcomes and regimes of treatments. In accounting for their actions, and their
action in telling their experiences, parents articulate the basis for understandings
held in common, understandings that can be claimed as warrants in future
elaborations of the significance of their struggle to define new forms of what
constitutes the normal.
Sacks (1992) pointed out that what it is to be an ordinary person in the world
cannot be taken for granted. Being ordinary is something we work at in our
communicative relationships with others. Ordinariness is not an inevitable
product of being an ordinary person. Rather we expend much effort in
demonstrating in our communicative relationships with others that our lived
experience of life can be taken as being the experience of an ordinary person.
In the context of parents who care for children with chronic renal failure, what
is it to be ordinary? Their doing of being ordinary has to be done with the
resources that are the common currency of their unusual circumstances. It is not
that they are ordinary parents of children with an unusual condition. Rather they
have to establish, they have to do in their talk with others , what it is to be
ordinary as parents-of-children-with-renal-failure. We have seen that there are
recurrent uncertainties in doing this that are a key resource in doing being
ordinary. One of the practical consequences of the analysis presented is an insight
into why narrating health care experiences provides such an effective device in
group meetings.
This is psychosocial support as examined within the communicative actions of
parents. We see joint reasoning in settings of extreme uncertainties about just
what to expect. The parents' experiences of isolation are narrated in terms of
these uncertainties. Uncertainties over what is normal in their extraordinary
circumstances, about just how to represent what was, is, or might be, give their
talk its dilemmatic quality . Their normality is an extreme of uncertainty. The
group affords joint reasoning about what is reasonable both in terms of
circumstance and outcome: that is, in terms of their incorporation and
David Middleton
166
engagement in the process of care for their children and in terms of the sort of
outcomes they might reasonably expect.
Transcription Conventions
distinctly quieter than surrounding
talk?
continuing talk $
interru [ption
[or spoken in
[unison
emphasized
[additional information]
hh out breath
C..) indecipherable
Reasonable Uncertainties
167
Acknowledgments
Grateful thanks are extended to the parents and families who consented to have
their conversations recorded; to Dot MacKinlay and Judith Argles for their
collaboration in recording the parent groups; to Jacqueline Collier for the initial
transcription of the conversational material; and to members of the Nottingham
Paediatric Renal Unit for their support and enthusiasm. Other people have been
most generous in their critical comments and suggested amendments, including
members of the Department of Communication Studies (Tema K) and Child
Studies (Tema B) at Linkoping University, Sweden ; members of the
Loughborough Discourse and Rhetoric Group; delegates at the 1993 NATO
Workshop on "Discourse, Tools and Reasoning," and Anssi Perakyla, Clotilde
Pontecorvo, Roger Saljo, and Barbara Burge.
168
David Middleton
References
Abraham, C. S., & Hampson, S. E. (in press). A social cognition approach to health
psychology: Philosophical and methodological issues. Psychology and Health.
Argles, J., MacKinlay, D. R. E., Middleton, D. J., & Watson A R. (1994) . Parent support
groups : A local account. Maternal and Child Health : The Journal ofFamily Medicine.
19(5), 152-158.
Atkinson, J. M., & Heritage, J. (Eds .). (1984) . Structures of social action : Studies in
conversational analysis. Cambridge, England: Cambridge University Press .
Billig, M. (1987) . Arguing and thinking : A rhetorical approach to social psychology.
Cambridge , England : Cambridge University Press .
Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D., & Radley, A. (1988) .
Ideological dilemmas: A social psychology ofeveryday life. London : Sage Publications .
Buchanan, K., & Middleton, D. (1995). Voices of experience: Talk, identity and
membership in reminiscence groups. Ageing and Society, 15,457-491.
Buttny , R. (1993) . Social accountability in communication. London : Sage Publications .
Collier, J., & Watson, A R. (1994) . Renal failure in children : Specific considerations in
management. In H. McGee & C. Bradley (Eds.), Quality of life following renalfailure:
Psychosocial challenges accompanying high technology medicine (pp. 225-245). Chur,
SwitzerlandlPhiladelphia, PA: Harwood Academic Publications .
Edwards, D., & Potter J. (1992). Discursive psychology. London: Sage Publications.
Goffman, E. (1981). Forms oftalk. Oxford, England: Basil Blackwell.
Heritage, J. (1984) . Garfinkel and ethnomethodology. Cambridge, England: Cambridge
University Press.
Jefferson, G. (1984) . On the organisation of laughter in talk about troubles. In J. M.
Atkinson & 1. Heritage (Eds.), Structures of social action: Studies in conversational
analysis. Cambridge, England: Cambridge University Press.
Lerner, G. H . (1993). Collectivities in action : Establishing the relevance of conjoined
participation in conversation . Text, 13(2), 213-245.
Middleton, D. (1996a) . A discursive analysis of psychosocial issues : Talk in a "parent
group" for families who have children with chronic renal failure . Psychology and
Health, 11, 243-260.
Middleton, D. (1996b). Talking work: Argument, common knowledge and improvization
in multi-disciplinary child development teams . In Y. Engestrom & D. Middleton (Eds.),
Cognition and communication at work (pp. 233-256) . Cambridge, England: Cambridge
University Press.
Middleton, D., & Curnock, D. (1995). Talk of uncertainty : Doubt as an organisational
resource for co-ordinating multi-disciplinary activity in neonatal intensive care. In G.
Loomes (Ed.), Risk in organisational settings . York, England: Economic and Social
Research Council.
Perakyla, A, & Silverman D. (1991) . Owning experience: Describing the experience of
other persons . Text, 11(3),441-480.
Perakyla, A (1994). Personal communication.
Potter, 1., & Wetherell, M. (1987). Discourse and social psychology: Beyond attitudes and
behaviour. London: Sage.
Sacks, H . (1992) . Lectures on conversation. (Vols 1 & 2; G. Jefferson, Ed.), Oxford,
England : Basil Blackwell.
Chapter 7
Abstract
Multiculturalism
In this chapter, we examine the multicultural organization of literacy instruction
within the Samoan American community of urban Los Angeles . In particular, we
introduce the concept of syncretic literacy as a framework for analyzing how
170
171
another in the course of conducting their daily social lives. As noted earlier, in
this view, just as one switches between codes, so one switches between systems
of social values, beliefs, emotions, practices, identities, and institutions.
Although it may be useful to treat cultures as coherent and separate and although
in some cases members of multicultural communities do draw boundaries between
what they consider traditional and what they consider new, more typically cultural
threads from diverse sources are interwoven into a single interactional fabric. In
the Samoan American community, members tend not to abruptly culture-switch
as they move from one activity to the next or one setting to the next; rather, they
blend cultural orientations in the course of carrying out a single activity,
including the activities of reading and writing Samoan and English .
Our research in this area is consistent with recent studies of immigrants in the
United States that stress the importance of seeing these groups as operating
simultaneously in two communities (Chavez, 1994; Rosaldo, 1989; Zentella,
1990). Our work, however, differs from these studies in our emphasis on the need
to study the daily, moment-by-moment confluence of multiple cultural models
and language-mediated practices. We believe that multiculturalism as a sense of
belonging to more than one community is not only imagined as an ideology of
connections (Chavez, 1994), but it is also enacted in daily routines that need to be
unpacked, if we want to uncover the different cultural threads they both imply and
sustain. A substantial part of this chapter is dedicated to the discussion of some of
these routines and threads in a Samoan American family.
172
Syncretic Literacy
In counterpoint to these three misconceptions of multiculturalism, we propose
considering literacy activities in all communities as syncretic . By syncretic
literacy, we mean that an intermingling or merging of culturally diverse traditions
informs and organizes literacy activities. This use of syncretism is drawn from a
number of sources, including studies of the interaction of religions in contact
situations (Apter, 1991; Herskovits, 1937, 1952, 1966), Bakhtin's work on
heteroglossia and hybridization (Bakhtin , 1981, 1984; Voloshinov, 1973), and
contemporary studies of multilingual communities and postcolonial discourse
(Hanks , 1986, 1987; Hill & Hill, 1986; Kulick, 1992).
In the study of religion, especially in Africa and the New World, syncretism is
generally defmed as the combining, intermingling, or merging of different belief
systems (Droogers, 1989). Herskovits (1952) identified syncretism as a form of
reinterpretation, a cultural process that affects all aspects of cultural change. This
anthropological use of the concept of syncretism differs from the way in which
the term has been used in psychology, where it connotes a somewhat incoherent
and transitional developmental stage, as in the acquisition of word meaning
(Vygotsky, 1986).
Our work is also inspired by Bakhtin's view of how different dialects,
perspectives, and voices can coexist within the same national language or even
within the same utterance. Bakhtin (1981) and his collaborator/alter ego
Voloshinov (1973) expressed a harsh criticism of synchronic investigations that
took monolingual, monological speech genres as the normal or unmarked form of
communication. They believed that any approach that takes an individual system
or code as the ideal object of inquiry is problematic because it does not allow us
to uncover the richness of language used by speakers as well as novelists. As
pointed out by Hill and Hill (1986) , bilingual situations in which codes are
routinely mixed in creative and sometimes contested ways are ideal contexts for
testing out Bakhtin's theory and, in particular, his notion of hybridization: "a
mixture of two social languages within the limits of a single utterance, an
encounter, within the arena of an utterance, between two different linguistic
173
174
These studies, among others (Heath, 1983; Lave, 1988; Scribner & Cole, 1981),
emphasize the importance of shifting the almost exclusive focus on writing (or
print) as a generalized technology characteristic of earlier studies of literacy
.(Goody, 1977; Goody & Watt, 1968) to an appreciation of the importance of
specific activities organized around and through print. In this chapter, we continue
in this tradition by integrating the emphasis on activities with an
ethnographically informed attention to the physical environments in which
literacy activities take place and the tools and artifacts that are made available in
such environments. When we look at literacy practices outside the school
environment, we find that the material resources utilized for literacy tasks have a
crucial role in establishing the breadth as well as the limits of what literacy can do
or mean for the children engaging in it. To be able to participate successfully in
literacy activities, children must learn to utilize resources that were not originally
designed for literacy practices. Even in those cases in which the materials used
were originally conceived as literacy-tools, the context in which they are used
forces a new interpretation of their original meaning. Placed in a new context, old
tools not only bring in remnants of the past but also force participants to face
issues of tradition, change, and social identity.
175
Field Research
Starting in 1993, we have been video recording Samoan American children and
adults interacting in a variety of settings, including four households and a church
compound in the greater Los Angeles area. The present study of children's literacy
activities is part of a more comprehensive study of how Samoan American
families educate their children into problem-solving discourse activities. This
project is a component of a larger project funded by the U.S. Department of
Education and coordinated by the National Center for Research on Cultural
Diversity and Second Language Learning at the University of California, Santa
Cruz.
In carrying out this urban multicultural and multilingual study, we are building
on a number of projects previously conducted by the two principal investigators,
including (a) a study of language socialization and language use across a number
of social settings in a rural village in Western Samoa (Duranti & Ochs , 1986;
Ochs, 1982, 1988); (b) a study of problem-solving discourse in decision-making
councils in Western Samoa (Duranti , 1981, 1990, 1994); and (c) a study of
problem-solving discourse during dinner conversations among white families in
the Los Angeles area (Ochs, Smith, & Taylor, 1989; Ochs & Taylor, 1992a,
1992b; Ochs, Taylor, Rudolph, & Smith , 1992).
176
caregivers, the more senioror higher ranking one will expect the younger and
lower ranking one to be the active caregiver. This is realized either through other-
selection - the higher caregiver directs the lower caregiver to carry out a task on
behalf of the child - or through self-selection - the lower ranking caregiver carries
out the task without being so instructed. Thus, for example, the practical care of
even young infants - rocking the infant to sleep, picking her up upon awakening,
changing clothes and bringing her to be breastfed, or, in the case of older infants,
feeding the child whole foods as well - tends to be the responsibility of older
siblings of the child or perhaps a younger sibling of one of the parents, if these
persons are not at school or called for other chores. If no one younger is present,
the child's mother or other adult takes charge of the child's care. Even when the
child is cared for by her siblings, an adult is usually within relatively close
proximity and intermittently monitors the caregiving activity (see Figure 7.1).
In this sense, caregiving is not only hierarchically organized but also socially
distributed. For example, caregiving is distributed when a sibling carries a hungry
infant to her mother and the mother breast feeds the child . Similarly, when an
adult directs a sibling on how to carry the infant or to wash the infant, the task is
collectively accomplished.
177
Fig. 7 .1. Young boy oversees sleeping baby while mother sits nearby (Western
Samoa, 1979)
178
important quality a child must display is faaaloalo [respect]. The term includes
the term ala, meaning "(to) face (in the direction of)." From infancy, children are
held or placed facing outward, paralleling the orientation of the caregiver toward
some interaction at hand. This is shown in Figure 7.2, where a baby is fed facing
outward, and in Figure 7.3, where he is made to sit up to look at the rest of the
interaction in the house (where his mother and other young women are sewing
colored feathers to sleeping mats) .
Young children are expected to position themselves at the edge of dwellings to
observe and ultimately report on what is taking place outside. In addition, toddlers
are explicitly socialized to identify and call out names of family and community
members as they pass by (see Figure 7.4).
Fig. 7.4. Young boy observes at the edge of a dwelling (Western Samoa, 1979)
In every community, members interact with children and other novices in ways
that allow them to appropriate cultural knowledge, beliefs, stances, expectations,
preferences, and practices (Rogoff, 1990). Further, by and large, in every
community, members rely on a similar core of strategies for facilitating the
appropriation of these cultural orientations. However, communities differ in their
reliance on one or the other socializing strategy. That is, they differ in terms of
179
strategies they usually employ and in terms of the activity settings in which they
opt for one or another strategy. In traditional Samoan households, three
socializing strategies prevail: repeated demonstration of an activity, prompting,
and action imperatives.
Generally, Samoan children are allowed a prolonged period of observation of
repeated denwnstrations of actions and activities before being expected to assume a
central role in them. For example, young children are not pressured into talking
before they articulate words and show themselves to be interested in
communication. Children are given ample opportunities to observe a wide variety
of culturally relevant activities . They accompany older siblings, parents, and
grandparents in work and, to some extent, in formal school environments. As
siblings and adults weave mats, carve boats, braid string, shuck coconuts, cook,
do laundry, read the Bible, and deliberate political matters, toddlers are close at
hand. Often a sibling will be carrying out some other work activity as he watches
over a younger brother or sister. The toddler may stand close by, attending to the
activity at hand.
Children and adolescents spend a long period of time as overhearers of
ceremonial speeches. Only much later in life are they expected to display their
own oratorical skills. Young children are often brought to ceremonial events by
their grandparents to sit by them and occasionally fetch cigarettes and other items .
Adolescents are expected to be present and ready to act as messengers, food
bearers, and gift announcers in their traditional role of servant in political,
religious, and other formal activities. Similarly, children spend long periods of
time as audience to dance performers as they practice dance routines over and over
again before they are themselves public performers. Young children may sit on
the laps of older community members, occasionally imitating the movements of
the dancers, but the focus of audience and dancers is on the performance. In these
ways, by the time they are asked to participate more centrally in an activity,
children have usually had many occasions to witness its completion (see Figure
7.5).
Similar to the strategy of repeated demonstration is the socializing practice of
prompting. Whereas repeated demonstration involves a member reproducing a
cultural activity in the presence of a novice, prompting involves the novice
reproducing or attempting to reproduce an act or activity. Prompting is a
socializing interaction in which a (usually) more knowledgeable interlocutor
elicits the repetition of an action or activity by a (usually) less knowledgeable
interlocutor. Although prompting is universal, some societies, such as rural
Western Samoan communities, rely heavily on this practice as a form of
instruction (see also Demuth, 1986; Miller, 1986; Schieffelin, 1990; WatsonGegeo & Gegeo, 1986).
180
Fig. 7.5. Samoan children and caregivers watching a ceremonial exchange (Western
Samoa, 1979)
181
place. For example, a mother sweeping the household compound will give a
basket to her two-year-old and order her to pick up fallen leaves on the compound
or to rearrange mats in a dwelling. In our observations, generally children are not
ordered to carry out tasks that they cannot easily carry out by themselves.
ITo those familiar with Samoan orthography, our transcription will appear, in some
respects, unconventional . We follow the tradition in using the inverted apostrophe C)
to represent a glottal stop (+ ), but we replaced the macron found in traditional Samoan
orthography with the circumflex accent (e.g., a, e). See section on Transcription of
Interaction that follows and Footnote 2 for an explanation of our transcription of
colloquial Samoan .
182
Transcription of Interaction
' In our transcripts, wefollow a modified version of the conventions introduced by
Gail Jefferson for conversation analysis (Sacks, Schegloff, and Jefferson, 1974). A
detailed description of our conventions is found in the Appendix.
Here, as elsewhere in our publications on Samoan language and culture, we
have chosen not to modify speakers' colloquial expressions, and thus we have not
hidden any use of what Samoans call "bad speech" (tautala leanai, a phonological
register where the two alveolar sounds represented in traditional Samoan
orthography as t and n are pronounced as the velar sounds Ik/ and 111/,
respectively- (Duranti & Ochs, 1986; Shore, 1982). Whereas bad speech
pronunciation is itself common in formal and ceremonial events in Samoa
(Duranti, 1981, 1994), and for this reason alone cannot be simply labeled casual
or informal, the combination of this pronunciation with the colloquialisms and
morpho-syntactic reductions characteristic of many household interactions gives it
a rather private connotation that many Samoans do not like to see displayed
publicly, especially in a written form. Although recognizing the complexity of
Ufo avoid ambiguity, we replaced the letter g introduced by the missionaries to
represent the velar nasal with the phonetic symbol zn/ (in other Polynesian languages ,
the same sound is represented by ng) . It should be pointed out that the sound IT]I is also
found in good speech, but in bad speech , IT]I replaces all Inl found in good speech .
Thus , certain distinctions found in good speech (e.g., Ifanal "gun , shoot" vs. /fana/
"bay ") are neutralized in bad speech - where bothfana andfaTJa are pronounced [fana],
183
184
tv ..... /vldc09l1mc
bockyo rd
ki tchen
front door
STREET
Key
Fig. 7.6. Map of section of Siks's house with participants ' locations at the
beginning of first visit
On May 27, 1993, one of us (A. Duranti) went for the first time to visit Sike's
family, one of the four families in our study. The shoes and sandals left in front
of the entrance were the first of a long series of reminders that the external shape
of the building, a two story house with a small front yard and a garage on the
side, was perhaps the only feature in common with the other families on the
block .
185
When the researcher walked in, he found several members of the family in the
living room area, including Sike, his cousin Mata, Sike's grandparents, and
Sike's aunt (his grandparents' unmarried daughter). Despite Duranti's repeated
attempts not to upset the flow of events with his presence, he was immediately
offered some food and encouraged to sit down and eat it. Before sitting down,
however, he set up the tape recorder on the floor and the video camera on a tripod.
Figure 7.6 provides a map of where everyone is located at the moment when
Duranti sits down to eat, and Figure 7.7 shows a frame from the video taken
while he is consuming the meal and talking with the grandparents .
Fig. 7.7. (From left) Grandfather, grandmother, Sike and his cousin, Mata, sitting at
the TV screen, and Sike's aunt (resting on the couch ; only her head is visible).
As can be seen in Figure 7.7, the grandfather is sitting on the floor to the left; the
grandmother is sitting on the floor farther away from the camera. Sike (6 years
old) and Mata (12 years) are sitting facing the opposite corner, playing a video
game; Sike's aunt (who had prepared and delivered the plate with food for the
researcher) is lying on a couch watching the children play while also listening to
and occasionally participating in the conversation among the adults in the room;
the researcher (A.D.), who is not visible in the video frame, is shown in Figure
7.6 to be facing the grandfather and the grandmother.
Several aspects of this setting are worth reflecting on, especially from the point
of view of the syncretism of material resources and the distribution of activities
and roles.
186
Despite the presence of the two TV screens, the telephone in the center of the
room (next to the grandmother), the furniture , the central chandelier (off screen),
and the thick walls, the camera has captured a scene here that, in several respects ,
resembles what one might see in a very different type of house in a Samoan
village. In particular, the seating arrangement closely resembles the type of spatial
organization that we have documented in Western Samoa. If we take as a point of
view the entrance to the house and the road, participants appear seated according to
traditional categorical distinctions, with the guest and the grandparents located in
the front part of the room and the grandchildren and the unmarried daughter in the
back region (an area that in Samoa would be considered an extension of the basic
floor plan of the house). In the front part of the room, a further distinction can
also be drawn between the guest and the grandfather (who is also a chief) who are
seated facing one another, as appropriate to people of equal rank (Shore , 1982), in
the more prestigious region, and the grandmother in the relatively lower position
(the back) . As shown in Figure 7.7 , only the section of the room with the
grandparents and the guest is covered by sitting mats, 3 whereas the back region
with the children and the unmarried daughter shows the house floor rug.
further beck regi on
for food preperettcn.
young women end
children
RACK->
reg ion
for high
chi ef s
FRONT ->
reg ion
for high
ch iefs
rood
Fig. 7.8. Scheme of traditional spatial distinctions in a Samoan house when guests
and other people of high rank are present.
Figure 7.8 represents the hierarchical division of space in a more traditional house
in Western Samoa with an extension in the back (adapted from Duranti, 1992). In
Figure 7.9, the seating arrangement in the living room of the Southern California
3These are an imported brand of industrially produced mats that are not found i n
Western Samoa , where the sitting mats are hand-woven, like all other mats, by the
women of the household.
187
tv wlvideogame
Ch
iaCr
~Aunt
beck->
front -)
Ibi9l
ITV.-I
CAMERA
RORD
Fig. 7.9. The two maps are matched to show salience of traditional organization of
space.
Syncretism of Activities
A few minutes later, the activities as well as the positions of the same
participants have shifted considerably. Mata and the grandfather are the only two
people left in the living room area. Mata is doing her homework, and the
grandfather is watching television. The grandmother is at the sink in the kitchen;
Sike is sitting at the kitchen table reading his instruction sheets for his homework
assignment; and the aunt is outside in the backyard. While the researcher with
camera moves to the kitchen to follow Sike's actions, Sike gets up from the table
and heads outside where his aunt is . He asks for help . The aunt first gives him
directions on how to proceed; then, just as Sike is heading back toward the
kitchen table, she instructs him to get a box that is behind her, against the
external wall of the living room . After an attempt to get the wrong box (the
plastic crate in front of the sliding door), Sike identifies the box his aunt was
indicating and drags it to her (see Figure 7.10).
188
Aunt
~:-:-IU- -~-
Sil< e
bOH
The aunt adjusts the box on the side between her and Sike and indicates to him
that he ean write on it. The box has been transformed into a desk (see Figure
7.11).
189
As the camera moves closer and eventually outside of the house to follow Sike's
activity in the yard, we get a sense of the multiplicity of activities within the
same physical space and the roles played in them by people and material
resources. The visual record makes evident that homework is not the only activity
occurring in the backyard. Another competing activity is emerging, one that has
the box-desk as one of its essential elements . Next to Sike and his aunt, on a
couch, there are a number of traditional fine mats Cie toga) (see Figure 7.12).
These are precious goods that are going to be sent back to Western Samoa to be
exchanged in a ceremony, called saofa'i, during which the grandfather will be
given a new and prestigious chiefly title . The box transformed by Sike and his
aunt into a desk is one of the boxes to be filled with fine mats. This makes the
box a tool with multiple but by no means equal or neutral functions. Each use of
the box indexes not only different types of activities within different value
systems (e.g., doing homework vs. packing fine mats for a ceremony), but also
different sets of culturally mediated expectations about children's and adults' roles
and about the goals of socialization. These different expectations are exhibited in
the following two segments. In the first, the aunt is instructing Sike while the
grandmother watches them, apprehensive and somewhat irritated.
190
Segment
1
2
3
Aunt;
4
5
Aunt;
Sik@;
7
8
Gm;
Sik@ ;
"(three. four.)
okay.
starts to write , turns to look at camera, goes back to write
changes position , leaning harder on box with his right arm
raises eyebrows exhibiting disapprovaljj'i
(2 .0)
10
Gm;
11
Sik@;
In line 10, the grandmother expresses her concern about the box with a warning in
fast speech: e:! leana le ki(Tl)ipusa!, literally "the empty box (atinipusa) is
bad (leana)." Sike, however, continues to write on the box. A few minutes later,
the grandmother's fears come closer to reality as Sike leans over and pushes hard
enough with his elbow to cause a dent. She upgrades her warning (line 28),
and this time her words are briefly echoed by her daughter (Sike's aunt) (line 29):
191
Segment 2
27
Sike;
28
Gm;
29
Aunt ;
30
31
32
Sike;
Aunt ;
Gm;
le kinipusa i 5 leana!
'the box 's gonna get ruined! '
'aua le so 'oga'don't over-' (i.e. 'don't press too hard')
(l write 'Tom'?)
yeah. write your "(name). write your middle name,
'a'e!
'oh! (disapproving)'
33
Aunt ;
dent
Several aspects of these two interactions are worth examining from the point of
view of the syncretic nature of the literacy activity we find in it. First, we will
focus on the different ways in which the aunt and the grandmother interact with
the child. Whereas the aunt accommodates to the child by attending to the task
that he is proposing (doing homework) and letting him use the box that was
needed for another activity (packing fine mats), the grandmother is much less
accommodating and more concerned that the child does not jeopardize the adults'
forthcoming activity (packing) by ruining the empty box on which he is writing.
The aunt and the grandmother are thus following two different models: The
grandmother is situation-oriented (or accommodate-child-to-situation) and the aunt
is more child-oriented (accommodate-situation-to-child; see earlier section, Child
Expected to Accommodate to Situation . The point here is not that the grandmother is insensitive to the child's needs but that she is more concerned with the
5 This is a colloquial expression in which the preposition i before the predicate (leaga)
is used to introduce a likely and unwanted event. In careful good speech, the word
kinipusa would be pronounced atinipusa and in careful bad speech akiuipusa.
192
child learning how to accomplish his goals without interfering with those of adult
family members. At the same time, the aunt is not just assuming a western role
of accommodating caregiver. She is also acting within the logic of traditional
Samoan child-care by assuming the lower ranking role of active caregiver, hence
allowing the grandmother (the higher ranking caregiver) not to be directly engaged
in the task at hand (see earlier section, Caregiving: Hierachical and Distributed).
In the previous section, we have seen how two different caregivers, Sike's
grandmother and aunt, attending to two different activities in contact with one
another, use two different socialization strategies toward Sike to control his
behavior as he does his homework. In Segments 1 and 2, the codes the caregivers
speak match such diverse strategies, with English being used (by the aunt) to help
Sike in his homework (adapt situation to child) and Samoan being used to remind
him of his need to adapt to the situation.
In this section, we show that the same caregiver may produce a syncretic blend
of teaching strategies within the same activity. Sometimes the blending takes
place within the same code (English), other times in two separate codes (English
and Samoan) . In line 2 of Segment 1, for instance, 'the aunt asks a test question
(how many books ?), a strategy typical of American teachers but not of Samoan
caregivers in traditional communities. At other points (in line 1 in Segment 1 and
lines 13, 14, and 24 in Segment 3), she uses explicit action directives to Sike ,
thereby following patterns characteristic of Samoan caregivers in traditional
households (see earlier section, "Socialization Through Repeated Demonstration,
Prompting , and Action Imperatives") .
Segment 3
12
13
14
Aunt;
15
16
Sike;
17
18
19
20
21
22
23
24
Gm;
sns.
Sike;
Aunt;
193
Segment 4
35
36
37
38
39
40
Sike
Aunt;
41
42
43
44
45
46
Aunt;
47
48
49
50
51
Sike;
Aunt;
sne.
Aunt;
OCT/d.)
'Like that.'
The homework activity is thus syncretic in that certain acts that comprise it orient
toward western modes of instructions, whereas other acts orient toward traditional
modes of instruction.
6An alternative interpretation of this sequence is : "right above this one. right here ."
194
Segment 5
1
2
3
4
Aunt;
5
6
Mata;
Aunt;
8
9
10
11
12
Mata;
sue.
195
oh no this one you have to:(puts paper back on flap of box in front of Sik~))
you have to nllld.
moves behind Sike to be able to read))
well you havepoints to paper)) you go:(t) uhmm underline it.
When Sike seems to have a hard time explaining the assignment to his aunt, his
cousin , Mata, becomes more involved in the activity. In line 16, which follows,
she reads the relevant passage from the assignment sheet; in line 20, she indicates
where to find the relevant information (the picture of the cave); and in line 22, she
proposes the solution to the problem.
13
14
15
16
17
Aunt;
Sik~ ;
Mata;
18
Aunt
19
20
Mata;
21
22
Mata;
Sik~ ;
196
At this point , the aunt instructs Mata and Sike to get another box, implicitly
suggesting that they can go and work on it, creating a second desk, but the
grandmother this time explicitly sanctions the use of a new box by suggesting
that they use something different. The solution is a nearby bench, where Mata and
Sike are eventually sent by the aunt to complete the homework assignment (see
Figure 7.13).
Fig. 7.13. The two activities of packing mats and doing homework become
separate.
With this new spatial configuration in place, the two activities that earlier
coexisted in one location, sharing one tool (the box), are now divided. This is a
very traditional solution whereby the adults are free to continue with their work
(in this case, folding mats and placing them in boxes) while the older children in
the household take care of the younger ones. The syncretism is this time realized
through the blending of a western task (English homework) with a form of social
organization that is part of the traditional Samoan learning environments
(hierarchical division of labor in caregiving).
Conclusions
Our work in a Samoan-American community in Southern California has been
used here to make two general points about the relationship between
multilingualism and multiculturalism and, thus, to rectify what in our view
are common misconceptions about immigrant communities in the United States
and elsewhere. By concentrating on a homework task carried out in one of the
families in our study, we have shown that, in a multilingual , multicultural family
setting:
197
(a) language (in the sense of the specific code used at any particular time in an
interaction) is not always a good predictor of the cultural orientation or
interpretive frame that is being activated by the participants. Thus, in a
Samoan American family, English may be used fluently but in ways that
are consistent with the socialization practices typical of traditional learning
environments in the home country.
(b) Members of multicultural communities can be in more than one culture at
a time. This becomes apparent in the homework sequence analyzed here as the
same space and material resources are being used for two very different tasks
by different participants. Whereas the grandmother sees the space in the
backyard and the boxes placed in that space as needed for the accomplishment
of a task oriented toward a rather traditional aspect of Samoan culture (the
preparation and exchange of ceremonial Objects), her grandchild sees it as a
new location on which to accomplish his math and English homework
assignments. The child's aunt (and grandmother's daughter) accommodates to
both cultural orientations by mediating between them. She first transforms a
material object designed for one task (a box) into a tool for the other (a desk)
and then, when a younger potential caregiver arrives on the scene, redirects the
child to a different location so that she can accommodate to her mother and the
accomplishment of the traditional task.
An implication stemming from the concept of syncretism employed here is that
becoming an English speaker does not necessarily entail adopting strategies
characteristic of other groups who use English. Although many Samoan
American family members use the dominant language of the United States, they
may do so primarily at the level of the communicative code rather than at the
level of communicative conduct.
198
Appendix
Transcription Conventions
Aunt;
(1.0)
you have
to read
(mh)
"(oka-)
points
e leona
(??)
no=no=no
199
References
Anderson, A. (1984). Social and institutional influences on the development and
practice of literacy. In H. Goelman & A. Oberg (Eds.), Awakening to literac y (pp.
24-37) . London: Heinemann .
Apter, A. (1991). Herskovits 's heritage: Rethinking syncretism in the African
diaspora. Diaspora, 13, 235-60 .
Bakhtin, M . (1981) . The dialogic imagination: Four essays. (M. Holqui st, Ed.; C.
Emerson & M. Holquist, Trans .). Austin : University of Texas Press .
Bakhtin, M. (1984). Problems of Dostoevsky 's poetics (C. Emerson, Ed. & Trans.) .
Minneapolis: University of Minnesota Press .
Bauman, R., & Sherzer, J. (1975) . The ethnography of speaking . Annual Reviews, 4 ,
95 -119 .
Blom, J.-P ., & Gumperz, J. J. (1972). Social meaning in linguistic structures: Codeswitching in Norway. In J. J. Gurnperz & D. Hymes (Eds.), Directions in
Sociolinguitics: The Ethnography of Communication (pp . 407-434) . New York:
Holt.
Chavez, L. R. (1994). The power of the imagined community: The settlement of
undocumented Mexicans and Central Americans in the United States . American
Anthropologist, 96 , 52-73.
Cicourel, A. V . (1992). The interpenetration of communicative contexts: Examples
from medical encounters . In A. Duranti & C. Goodwin (Eds.), Rethinking context:
Language as an interactive phenomenon (pp. 291 -310). Cambridge, England :
Cambridge University Press.
Demuth , C. (1986). Prompting routines in the language socialization of Basotho
children. In B. B. Schieffelin & E. Ochs (Eds.), Language socialization across
cultures (pp, 51-79). Cambridge, England: Cambridge University Press.
Droogers, A. (1989). Syncretism: The problem of definition, the definition of the
problem. In J. Gort, H. Vroom, R. Fernhout, & A. Wessels (Eds .), Dialogue and
syncretism: An interdisciplinary approach (pp. 7-25) . Grand Rapids , MI: Eerdmans .
Duranti, A. (1981). The Samoan fono : A sociolinguistic study. Pacific Linguistics
Monographs, Series B, 80. Canberra: Australian National University, Department of
Linguistics .
Duranti, A. (1990) . Doing things with words : Conflict, understanding and change in a
Samoan fono . In K. Watson-Gegeo & G. White (Eds.) Disentangling: Conflict
discourse in Pacific societies (pp. 459-89). Stanford, CA: Stanford University Press .
Duranti, A. (1992). Language and bodies in social space: Samoan ceremonial
greetings. American Anthropologist, 94, 657-91.
Duranti, A. (1994). From grammar to politics: Linguistic anthropology in a Western
Samoan village. Berkeley and Los Angeles : University of California Press.
Duranti, A., & Ochs , E. (1986) . Literacy instruction in a Samoan village. In B. B .
Schieffelin & P. Gilmore (Eds .), Acquisition of literacy : Ethnographic perspectives
(pp. 213 -32) . Norwood, NJ: Ablex .
Duranti, A., & Ochs , E. (in press) Change and tradition in literacy instruction in a
Samoan American community. In B. McLaughlin, B . McLeod , & S. Dalton (Eds.),
Teaching for success : Reforming schools for children from culturally and
linguistically diverse backgrounds.
200
201
Kulick, D. (1992) . Language shift and cultural reproduction: Socialization, self, and
syncretism in a Papua New Guinean village. Cambridge , England: Cambridge
University Press.
Lave, J. (1988) . Cognition in practice. Cambridge , England: Cambridge University
Press .
Lave, J., & Wenger , E., (1991) . Situated learning: Legitimate peripheral participation.
Cambridge, England : Cambridge University Press.
Leichter, H. J. (1984). Families as environments for literacy. In H. Goe1man & A.
Oberg (Eds.), Awakening to literacy (pp. 38-50) . London : Heinemann.
Macpherson, C. (1991) . The changing contours of Samoan ethnicity. In P. Spoonley ,
D. Pearson, & C. Macpherson (Eds.), Nga take: Ethnic relations and racism in
AorearoalNew Zealand (pp. 67-84). Palmerson North, New Zealand: Dunmore Press .
Mannheim, B. (1991) . The language of the Inka since the European invasion. Austin:
University of Texas Press .
McDermott, R. P ., Goldman , S. V., & Varenne, H. (1984) . When school goes home:
Some problems in the organization of homework. Teachers College Record, 85(3) ,
391-400.
Miller, P. (1986) . Teasing as language socialization and verbal play in a White
working-class community. In B. Schieffelin & E. Ochs (Eds.), Language
socialization across cultures (pp. 199-212). Cambridge , England: Cambridge
University Press.
Morgan , M. (1994) . The African-American speech community: Reality and
sociolinguistics. In M. Morgan (Eds.), Language and the social construction of
identity (pp. 121-48). Los Angeles: UCLA, Center for Afro-American Studies.
Ochs, E. (1982) . Talking to children in Western Samoa. Language in Society , 11, 77 104 .
Ochs, E. (1988) . Culture and language development : Language acquisition and
language socialization in a Samoan village. Cambridge, England: Cambridge
University Press .
Ochs, E., & Schieffelin, B. B. (1984) . Language acquisition and socialization: Three
developmental stories. In R. A. Shweder & R. A. LeVine (Eds.), Culture theory:
Essays on mind, self, and emotion (pp. 276-320). Cambridge , England : Cambridge
University Press .
Ochs, E., Smith, R., & Taylor, C. (1989) . Dinner narratives as detective stories .
Cultural Dynamics, 2, 238-57 .
Ochs, E., & Taylor, C. (1992a) . Family narrative as political activity . Discourse and
Society, 3(3), 301-340.
Ochs, E., & Taylor, C. (1992b). Science at dinner . In C. Kramsch (Eds.), Text and
context: Cross-disciplinary perspectives on language study (pp. 29-45). Lexington,
MA: D.C. Heath.
Ochs, E., Taylor , C., Rudolph, D., & Smith, R. (1992) . Story-telling as a theorybuilding activity . Discourse Processes, 15(1), 37-72 .
Rogoff, B. (1990) . Apprenticeship in thinking. New York: Oxford University Press .
Romaine, S. (1994) . Language standardization and linguistic fragmentation in Tok
Pisin. In M. Morgan (Ed.), Language in Creole situations: The social construction of
identity (pp. 19-41). Los Angeles: Center for Afro-American Studies .
Rosaldo, R. (1989). Culture & truth: The remaking of social analysis. Boston: Beacon
Press .
202
Sacks, H., Schegloff, E. A., & Jefferson, G. (1974) . A simplest systematics for the
organization of turn-taking for conversation. Language, 50, 696-735.
Schieffelin, B. B. (1990) . The give and take of everyday life: Language socialization
of Kaluli children. Cambridge, England: Cambridge University Press .
Schieffelin, B. B., & Ochs, E. (1986). Language socialization across cultures.
Cambridge, England : Cambridge University Press .
Scribner, S., & Cole, M. (1981). Psychology of literacy. Cambridge, MA: Harvard
University Press .
Shore , B. (1982) . Sala'ilua: A Samoan mystery. New York: Columbia University
Press.
Vo1oshinov, V. N. (1973) . Marxism and the philosophy of language (L. Matejka &
I.R. Titunik, Trans .). New York: Seminar Press .
Vygotsky, L. (1986) . Thought and language (A. Kozulin, Ed.). Cambridge, MA: MIT
Press .
Watson -Gegeo , K., & Gegeo, D. (1986) . Calling out and repeating routines in the
language socialization of Basotho children . In B. Schieffelin & E. Ochs (Eds .),
Language socialization across cultures (pp. 17-50). Cambridge, England : Cambridge
University Press .
Zentella, A. C. (1990) . Returned migration, language, and identity: Puerto Rican
bilinguals in dos worlds/two mundos . International Journal of the Sociology of
Language, 84, 81-100 .
Chapter 8
Abstract
This study analyzes the different types and uses of direct reported speech in the
context of psychotherapeutic discussions . The data are drawn from four sessions of
group therapy . The participants were 6 men, doing a program of recovery from
drug addiction, and a psychologist. The entire sessions have been videotaped and
transcribed .
Direct reported speech occurrences have been classified with regard to author,
content, and time location of the quote. An overarching distinction concerns the
fictional versus realistic character of the quotes. Data show that the therapist ' s
talk is strongly characterized by fictional quotes, whereas the p atients make a
more narrative use of reported speech . Specific function s of reported speech are
also examined in relation to therapeutic work, prop osing a view on identity and
self-understanding as based on intertextuality.
204
Alessandra Fasulo
205
thinning out and socializing its central focus, rather than absolutely relocating it
at a distancefrom the animator" (p. 13).
It appears fruitful then to consider reported speech as a linguistic structural
resource co-occurring in the determination of meaning, particularly with regard to
ascribable intentionality anddegree of accountability of the spoken words. Rather
than assuming reported speech as detached from the reporter, we want to think of
it as a manipulation of the "otherness" of discourse, containing cues for the
interpretation of the speaker's stance toward the quote.
Self-quotation can also occur. One's thoughts and speech acts, when framed in
reported speech, makefor a weakened form of reflexivity, that is, discourse about
a self that does not entirely overlap with the present one, incorporating a certain
degree of otherness.
In autobiographical narratives, splitting the omniscient narrator of the present
from the figure actingin the past can create suspense in the telling (Hill & Irvine,
1993), which sustains listeners' involvement (Tannen, 1988). In fact, if, on the
dimension of accountability, reported speech increases distance, it also creates a
kind of co-presence of quoting and quoted discourse, transporting speakers and
listeners in a different here-and-now(Blanche-Benveniste, 1991).
206
Alessandra Fasulo
207
the next section, I list the different types that were identified, trying to keep a
record of the discourse context in which they took place and the different "owners"
of the quoted words.
Initial letters refer to the fictionalized names of the participants : the therapist,
Antonio (T); Mauro (M); Luca (L); Nino (N); Francesco (F); Andrea (A); Daniele
(D).
Quotations are indicated by asterisks. The numbers in square brackets stand for
the number of the therapeutic session and the page of the transcript. The turn
where the quote begins is indicated by an arrow. The translation does not mirror
dialect aspects and jargon; intonation marks are reproduced partially to permit
comprehensibility.
T-4
This form appears mainly to have a topic-tying function, but it can assume
evaluative nuances, such as, in this case, where the young age of Mauro is
repeated and followed by a significant pause.
(2) Self-quotation of therapy talk: A participant refers back to something he said
in the current or in a former session. Usually it is anticipated by introductory talk
that specifies the sense of its reappearance.
208
Alessandra Fasulo
Excerpt 2 [4 :21]
[The therapist turns to the other participants with a question asked him by
Nino]
T
N
Here Nino specifies the exact wording of his question, which had just been
challenged through a procedure of group judgment. Metapragmatic work of
explanation and redefinition is common in self-quotation occurrences. When this
is the case, they can be considered as articulated forms of self-repair (Schegloff,
Jefferson, & Sacks, 1977).
(3) Other-quotation of therapy talk: Words previously uttered by another
participant are again made to enter the circuit of discussion. They can be taken up
in the immediately following turn (as in the example) or reprised at a different
point of the discussion .
Excerpt 3 [3:12]
D
T
D
209
(4) "Doing the other": The speaker produces words on behalf of another. The
person to whom the words are meant to belong is the immediately preceding
speaker or the next one, after turn allocation.' The sequence positioning of this
kind of utterance makes it possible to read them as reformulations, anticipations,
or free interpretations of the other's thoughts or feelings. Often the quote is
produced in a parodic style and has a mocking spirit.
Excerpt 4 [4 :32]
[The therapist looks at Nino, who is sitting back on the sofa with his arms
folded]
After the conclusion of an interview with one particip ant, the next speaker is
selected by the therapist. The direction of his look is a first indicator of the
choice, and a change of posture by the called person is common . The immobility
of Nino was thus a visible absence that got voiced by the therapist.
(5) Candidate utterances for another: This is a more complex way of enacting the
role of another participant. It is the offer of one or more propositions of which the
grammatical first person corresponds to the addressee, whereas the content comes
from the point of view of the speaker. I analyze this kind of reported speech in
more detail in the next section.
2When not otherwise indicated , all references to turn-taking organization are based on
Sacks, Schegloff, & Jefferson , 1974.
210
Alessandra Fasulo
Excerpt 5 [2 :15]
T
it has no value.
Here the therapist offers Luca a question that, in his view, Luca should ask
himself. This excerpt also shows a systematic feature of candidate utterances,
namely, that they are a translation in reported speech of an immediately preceding
assertive statement by the same speaker (Fasulo, 1994b).
(6) Candidate utterances for himself: The quote belongs to the speaker himself,
but what he is presenting is an utterance that he is now constructing as a possible
one. This is made clear by the conditional mode of the introductory verb.
Excerpt 6 [4 :35]
T
w.ho pu t on masks .
The reported speech is framed here as the equivalent of putting on a mask. Because
Nino said he could do that too, Luca reacts, asking to whom he is referring. It
could be argued that Nino is reporting a sample speech of other members of the
group, but, because the reported words are accompanied by a negative evaluation,
he makes them up as a candidate utterance for himself.
211
Excerpt 7 [2 :26]
lYi
D
T
The therapist here is testing Daniele's availability for helping the others. He does
so by a short hypothetical narrative (the setting is the after-lunch pause in the
community activities) in which a member of the group advances a request. Given
the exemplar character of the situation, Daniele's response represents an instance
of his general attitude.
Direct reported speech in these cases performs in the present of hypothetical
event, yet maintaining the abstractness of the argument.
(8) Anim ation of psychic entities: The child, the parent, or the adult are internal
unconscious instances, according to the psychol ogical theory adopted by the
therapist. Here they are represented as internal voices.
Excerpt 8 [4 :26]
A turn is intended here not as the whole contribution of the current speaker, but as the
reported utterance of a single author or principal.
212
Alessandra Fasulo
(9) Narrative self-quotation and (10) Narrative other-quotation are the canonical
forms of quotation, where an utterance or a dialogue is presented by replaying
episodes that happened in the past. These types come within a specified temporal
frame but often are instantiations of prototypical episodes; therefore, it is not
fidelity with the original utterances that is central but the transmission of the
affect connected with the situation . As in the following excerpt, the verbal tense
is the Italian imperfetto (translated here with conjunctive), which is an iterative
form. Both types are illustrated in the following recalled dialogue.
Excerpt 9 [4:3]
L
no all'inizio : in comunita :
no in the beginning in communi ty- when I had
[appena entrato in comunita : me dicev'
just come in the conmunf ry they'd say
T
[eh in comunita si .
yes in the communi ty
N -7
*fai gygs:to fai quell'altro me la pijavo proprio:
*do this and do that* I was really bothered=
-7
dicevo *oh rna tu che vuoi da me rna tu :
I 'd say *oh what do you want but you :
perche aggi 'a fa' ~ta co :sa* capl*[~
why should I do this * y'see
[*che rni paghi *
T
*are you paying me? *
This is part of a typical narrative of the patients, which 1 have called the "I am
changed" narrative (Fasulo, 1994a). This implies the contraposition of two
moments that, when displayed in conjunction, show a change in the patient's
attitudes or feelings. Here Luca reports his initial discomfort (second arrow) with
the obligations of the community life, as represented by a generic older resident
giving him instructions (first arrow). The temporal reference is not just a neutral
landmark but also sensitive to the point of the story," in that it places the story
itself just after Luca's entrance in the community, which is also the beginning of
the therapeutic path.' The continuation of this narrative is presented in the next
example.
4Por a discussion on temporal and spatial references in conversation, see Sacks (1992)
and Schegloff (1972) .
5Luca had had four months of community life. Before moving into the home where I
found them, some of the members were staying in a different place run by the same
organization, although they had another therapist following them.
213
214
Alessandra Fasulo
F
N
Anto rna tu ti : hai trovato qualAnto but you- did you ever findcioe in percentuale no? gente che ha detto
I mean in percentage no ? people who said
[*no*no[una persona che vuole fa la comuru ta hehe
a person who wants to do the communi tj! hehe
aggio a anda vi :a- m'aggio fa due anni*
I must go I must do two years*
cioe gia dai primi quattro cinque mesi .
I mean from the very first four five months .
o generalmente: siamo tutti quanti titubanti
or generally: we are all doubtful about leaving
per uscire dice- siamo tutti quanti cosi vero?
one says- we are all like that isn 't that right?
Here the therapist is asked whether, in his experience, he has found some people
who were determined to complete the program from the very beginning of their
stay in the community. The operation is that of a normative confrontation (the
information requested is, in fact, a statistic) on the dimension of will as expressed
in public statements.
The cases that I have presented reveal that reported speech is used in the
psychotherapy situation in a highly situated fashion. Its pragmatic implicatures
and aspects of performance make it a useful tool for conveying affect, past and
present attitudes , and envisaged possibilities. Let us see how the different forms
are distributed in the talk of the patients and of the therapist. Asterisks mark the
types that are consideredjictional, that is, those that are not meant to refer back to
something that had already been uttered (or thought).
6"Doing the community" is a conventional saying that expresses well what is both an
ideology and an observation: Leaving the addiction behind is an act ive
accomplishment and not a simple change of residence.
215
* 1.
Therapist
Simple repetition
13,9
* 1.
Simple repetition
6,1
2.
Self-quotation within
therapy
2,4
2. Self-quotation within
therapy
3.
Other-quotation within
therapy
5,9
3. Other-quotation within
therapy
5,9
5.
1,1
6.
2,4
*4 .
33 ,7
* 8.
9.
Animation of psychic
entities
Narrative self-quotation
0,9
14,4
* 4.
2,2
14,9
8,8
38,7
12,7
*7 . Fictional dialogues
*8 . Animation of psychic
entities
4,9
9. Narrative self-quotation
0 ,5
10 . Narrative otherquotation
19,3
4 ,4
1 1. Reported thoughts
10,6
11 . Reported thoughts
0 ,5
1 2 . Generic member
6, 1
5,1
1 2 . Generic member
Realistic
quotations
Fictional
quotations
(*)
66,6
Realistic
quotations
33,4
Fictional
quotations
100
28,7
(*) 71,3
100
All the types of direct reported speech are available resources for participants in
both roles, showing that the two roles are performed not via thoroughly
differentiated verbal modes but via a difference in their rate of use. Patients and
therapist have a mirroring pattern in their use of direct reported speech, with regard
to fictional versus realistic types. As is predictable on the basis of the roledefining traits, the patients make much use of narrative quotations, including
reported thoughts, whereas the therapist shows a higher degree of offered candidate
utterances and other-quotation of therapy talk. The fictional type of reported
speech most often present in the participants' talk is that of fictional dialogues;
that is, they actualize possible problematic situations to be submitted under the
therapist's scrutiny. The simple repetition is also higher in the patients' talk .
Presumably, given that most of their dialogues are between one of them and the
therapist, it is the therapist's words that get embodied in their following
utterances and used as a point of departure of the following discussion.
216
Alessandra Fasulo
Exercises on Future
In his analysis of psychotherapy in radio call-in transmissions, Gaik (1992)
argues that the use of irrealis is an index of therapeutic activity, by which he
means a particular type of conversational move of the doctor. Through this mode,
a proposal is conveyed concerning the patient's internal states that the listenerpatient can decide to acknowledge or reject. In my view, the interesting feature of
therapists' utterances in the irrealis mode is that it occupies an intermediate zone
between the interactants: the joint outcome of the patient's telling of his story and
the therapist's interpretive tools. I think this is the domain of candidate utterances.
Among the therapist's candidate utterances are positive and negative ones, by
which we mean that they are presented to be accepted or disregarded, plus some
that take on a "Hamletic" form. The last ones are interrogative sentences in which
an alternative is posed, and they never appear in patients' talk.
Therapist's positive candidate utterance:
T
~
Here is a candidate utterance that represents an act of will, introduced by "you can
decide": Note that this therapeutic action is not addressed to outline a possible
state of the world, because he is not saying "you can have a normal family" but
"you can decide *1 want to have a normal family* ." The action is messed
primarily to communicate to the patient his psychological capability of making
decisions and pursuing his own will.
Therapist's "Hamletic" candidate utterances :
217
Excerpt 13 [4:39]
ognuno lsi conosce I!lQltQ ~ dopa La comunita
everyone knows himself very well after the corrrnuni ty
[vabbe questa si
T
M
[questo si
The same type of action in the previous excerpt is carried on here, but we see that
two options are presented, and the patient, Mauro, is provided with the illustration
of what it is like after the community. He is only in the third week of his stay, so
these words also have the meaning of a definition of the therapeutic work in terms
of what it does to the members. And again, what it does is not to rule out the
possibility that a person, once out, can feel like using drugs again . The focus here
is on the mental activity of pondering different options and acting according to a
decision. Reported speech again helps in describing internal processes as actions
to be performed at a given moment (when realizing one is "slipping" again and
just before doing it).
Patient's negative candidate utterances referred to others:
Excerpt 14 [3 :15]
F
no [s(h)e: :
no
A priest who is the founder of the community and to whom requests for help and
counseling are often addressed.
218
Alessandra Fasulo
Here one patient is describing to another what he deems a wrong pact with parents
and also a wrong approach to the community, namely, entering it under the
condition of some material reward.
Patient's negative candidate utterances for oneself: see Excerpt 6.
Patients
Se lf
Other
Posit ive
37,8
20,7
16,6
N eg ati ve
27,1
46,6
16,6
Hamletic
35,1
100
100
219
Trying on Identity
The continous word-sharing in the therapy situation also appears with regard to a
particular type of utterances that have the aspect of claims of identity. The
following is one of the most relevant, given what the group is doing :
(1) Therapist provides candidate utterance:
Excerpt 15 [4:43]
M
T
... )
Excerpt 16 [4 :41]
M
T
N
allora noi rna qua non siamo usciti dalla ro :ba Anto,
then now we here are not out of the junk Anto,
no .
Daniele e io ieri stavamo a parla proprio di questa
Daniele and I yesterday were just talking about it
qua . *dopo due me:si sono gx tossicodipendente?*
*after two months am I an ex drug-addict?*
no Mauro . diciamocelo chi aro e tondo Ma' .
no Mauro . let 's be frank about it Ma'
220
Alessandra Fasulo
In Excerpt 15, the therapist stresses the need for adapting behavior with the claim
of being an ex-drug-addict, once out of the institution. In Excerpt 16, a participant
wonders about how long it takes to become entitled to utter the claim, and, in
Excerpt 17, a problem of labeling is dealt with by suggesting that they counter
the label by saying "I was."
The hot topic of being or not being a drug-addict is dealt with in an interesting
fashion: It appears as if the movement were from the utterance to the actor. The
utterance is "there" and circulates, but the actor has to work on himself in order to
become entitled to pick it up and use it when characterizing himself.
Identity claims can be tried on in a relatively safe environment until they get
attached to the person who "wears" them and are felt to belong to that person .
Conclusion
The analysis of reported speech in a group-therapy context shows that others'
words enter speaker talk at various levels of complexity, from repetition of some
words of an immediately preceding utterance to the alleged quotation that serves as
a point of departure for further discussion (where the utterance can be changed in
the process) and to the temporary substitution of one speaker with another.
In the microculture of six persons living together, sharing a past of drug
addiction and interacting with their regular therapist, some utterances appear
consistently throughout the therapy sessions, with a changing actor behind the I
who speaks. The fact seems to account for the primacy of discourse on single
individuals , the contemporary presence of different voices within one's repertoire,
and a conceptualization of psychotherapeutic work as the handing over from
221
therapist to patient a set of different voices that, entering into a dialogue with the
preexisting ones, can help patients on their way to change.
Certainly reported speech must be considered a socialized mode of expression
and a resource for pulling back to earth abstract concepts, especially in
interactional contexts where a gap in education is present among participants .
Nevertheless, other forms of reported speech are used without any problems in the
context I have observed. Therefore, I would not be satisfied with an explanation
invoking simplicity and rapidity at the basis of quoted speech usage. This appears
to be another piece of evidence in support of a view of the social person as a
crossing of possible discourses, none of which is truer than the other; on the other
hand, some can be felt as beyond one's grasp, and some have a limited horizon .
Such a theory of self works against the idea of monolithic individuals as much as
against an image of the person as the expression of gross sociological categories,
but it does not claim that we can say what we want: Doing "being normal," as we
have seen, can be no trivial undertaking.
Acknowledgments
The work presented here has benefited from discussion with the participants at the
workshop and especially from suggestions of further readings by Clotilde
Pontecorvo and Alessandro Duranti. Roger Saljo's revisions were also of great
help on both formal and content aspects. The remaining faults are due only to the
author's stubbornness.
222
Alessandra Fasulo
References
Bakhtin, M. M. (1991). Discourse in the novel. In M. Holquist (Ed.), The Dialogic
Imagination (pp. 259-422). Austin : University of Texas Press. (Original work
published 1934)
Besnier, N. (1993). Reported speech and affect on Nukulaelae Atoll . In J. H. Hill & J.
T. Irvine (Eds.), Responsibility and evidence in oral discourse (pp. 161-181).
Cambridge, England: Cambridge University Press.
Blanche-Benveniste, C. (1991) . Le citazioni nell' orale e nello scritto [Quotations in
oral and written texts]. In M. Orsolini & C. Pontecorvo (Eds.), La costruzione del
testo scritto nei bambini [The construction of written text in children] (pp. 259273). Florence, Italy: La Nuova Italia.
Fasulo, A. (1994a) . L'autobiografia. Modelli narrativi, contesti evocati e interazione
sociale nel processo di costruzione della storia personale [Autobiography. Narrative
models, evoked contexts and social interaction in the process of construction of
life-stories] . Unpublished doctoral dissertation, University of Rome "La Sapienza."
Fasulo, A. (1994b). La psicoterapia come traduzione tra linguaggi dell'esperienza
[Psychotherapy as translation among languages of experiences] . Rassegna di
Psicologia, 3(9) , 123-142.
Foucault, M. (1992) . Tecnologie del se [Technologies of the self]. In L. H. Martin, H .
Gutman, & H. Hutton (Eds.), Un seminario con Michel Foucault. Tecnologie del se
[Technologies of the self: A seminar with Michel Foucault] (pp. 11-47). Turin , Italy :
Bollati Boringhieri.
Gaik, F. (1992). Radio talk-show therapy and the pragmatics of possible worlds. In A.
Duranti & C. Goodwin (Eds.), Rethinking context (pp. 271-289). Cambridge,
England : Cambridge University Press .
Goffman, E. (1981). Forms of Talk. OXford, England : Basil Blackwell.
Gumperz, J. J. (1982). Discourse strategies. Cambridge, England : Cambridge
University Press .
Hewitt, J. P. (1976). Self and society. A symbolic interactionist social psychology.
Newton , MA: Allyn and Bacon.
Hill , J. H., & Irvine, J. T. (1993) . Introduction. In J. H. Hill & J. T. Irvine (Eds .),
Responsibility and evidence in oral discourse (pp. 1-23). Cambridge, England :
Cambridge University Press.
Jefferson, G. (1985) . On the interactional unpackaging of a gloss. Language &
Society , 14, 435-466.
Lee, B. (1994). Metapragmatics and subjectivities. In J. A. Lucy (Ed.), Reflexive
Language. Reported speech and metapragmatics (pp. 365 -391) . Cambridge,
England : Cambridge University Press.
Leech, G. (1978) . Natural language as metalanguage. Transaction of the Philological
Society of London. 1976-7, D, pp.l-31.
Macauley, R. K. S. (1987) . Polyphonic monologues: Quoted direct speech in oral
narratives. IPRA Papers in Pragmatics, 1(2), 1-34
Sacks, H. (1992). Story organization; Tellability; coincidence, etc. In G. Jefferson
(Ed.), Lectures on Conversation (Vol. II, pp. 229-241). Oxford, England : Blackwell.
Sacks, H., Schegloff, E.A., & Jefferson, G. (1974) . A simplest systematics for the
organization of turn taking for conversation. Language, 50, 696-735.
223
Chapter 9
Situational Effects in
Computer-Based Problem Solving
Paul Light' and Karen Littleton'
'Department of Psychology, University of Southampton, UK
2Faculty of Social Sciences, Psychology , The Open University, UK
Abstract
Possible bases of peer facilitation of children's learning are explored , drawing
upon concepts from developmental psychology. Against this background,
studies of social aspects of children 's computer use in school are reviewed . A
series of experimental studies is described in which pairs of children are found to
perform better than individual children when faced with novel computer-based
problem-solving tasks . Peer benefits are associ ated with patt erns of verbal
interaction in some degree, but facilitation by the presence of one or more
partners even occurs in the absence of direct verbal inter action. Gender
differences in performance are shown to be highly software dependent and also
prove to be sensitive to the gender of the partner, even (in fact, especi ally) when
there are no opportunities for inter action . The contextual sensitivit y of
performance is discussd in terms of situational constraints on cogniti ve
performance. It is argued that the important dimensions of context extend
beyond direct interpersonal interactions to encompass wider social processes.
One of the most salient aspects of the content of most human activities is the
presence or absence of other people. In the case of children' s use of computers in
learning , the presence of other children affords opportunities for collaborative
modes of learning . Computers lend themselves well to collaborative use. But the
225
226
227
228
1992; Mevarech & Light, 1992). It is apparent from this research that working
with others at the computer can confer benefits, both in terms of group
performance and individual learning, although it is equally clear that it does not
always do so. In looking for explanations, researchers tend to concern themselves
with the factors that influence how effectively the participants are engaging with
one another and with the task. Interpretations are characteristically framed within
cognitive-developmental perspectives traceable to the influences of Piaget and
Vygotsky . Although Vygotskian concepts such as scaffolding have for the most
part been developed in relation to markedly asymmetrical (usually adult-child)
exchanges, they have also been applied in relation to peer learning interchanges
(e.g., Mercer, Phillips, & Somekh, 1991). The capacities of the computer for
reshaping the interaction among participants have also been discussed within a
Vygotskian perspective (e.g., Crook, 1994; Jones & Mercer, 1993).
Whether they are approached from a Piagetian or a Vygotskian standpoint,
studies in this area typically construe the benefits of the presence of a partner at
the computer in terms of overt processes of interaction, most frequently
involving discussion and negotiation. Evidence of a role for cognitive conflict
has been provided by, for example, Teasley (this volume), whereas Rubtsov
(1992) analyzes his findings in terms of the progressive coordination of joint
actions.
With various colleagues, we have been researching school children's computerbased problem solving for almost a decade. Our first studies of computer-based
problem solving by pairs (as against individuals) involved some rather standard
tasks, such as the Tower of Hanoi and the Balance Beam. Here, turn-taking was
the predominant pattern, and gains from having a partner present were only
apparent when rather artificial steps were taken to force the joint engagement of
the partners with each turn (Light & Foot, 1987; Light, Foot , Colbourn, &
McClelland, 1987).
Influenced by some of the adventure game software that was appearing in
primary school classrooms in the late 1980s, we designed some more elaborate
problem-solving tasks that arguably made better use of the distinctive potential
of the computer. Hypermedia authoring tools lend themselves particularly well
to the creation of rich nonlinear information bases in relation to which planning
problems can be posed. Such software also facilitates relatively straightforward
development of very user-friendly and highly motivating programs. Using
HyperCard on Macintosh computers, we produced a planning task couched
within an adventure-game format. In the first version (Blaye, Light, Joiner, &
Sheldon, 1991), the scenario involved a quest to rescue a king's crown. The
basic screen was a map showing various towns on a mainland, some sea (replete
with pirates), and various islands. Air, sea, and land routes were marked. All the
place names and other labels on the map were buttons. The only interface device
used was a single mouse. If this was used to move the cursor to one of the
buttons and clicked, additional screens of information were revealed. In this way,
children could discover the whereabouts of all the available characters and modes
229
230
That the verbal interaction measures did not account for much of the variance
in performance or learning may simply reflect the inadequacy of the measures
used. On the other hand, it could be seen in rather different terms. In these
studies, unfamiliar adults were taking children from their classrooms to a less
familiar part of the school and presenting them with a very novel computer
(many had never used a mouse before) and a novel task. Given only scant
introduction, they were supposed to find out what had to be done and to work out
how to do it. It could be that the pair advantage in this situation reflects not so
much that the pairs were doing particularly well as that the individuals were
doing particularly badly under these conditions.
To address this, we conducted a further study (Joiner, Messer , Light, &
Littleton, 1992) in which we simply brought children to a room in the school
for a single session on the task, working on their own at a computer. The only
contrast was between children who (a) came on their own from the classroom and
worked alone (apart from an experimenter/observer) in the room and (b) children
who came to the room in groups of four and each worked individually at one of
four machines. In neither case was there any verbal interaction, and the children
could not see one another's screens. Nevertheless, the children in the latter
condition, with peers present, still did significantly better.
In subsequent studies, we have directly contrasted this parallel working (or
coaction) condition with a paired (or interaction) condition. Under these
conditions, we have found only a marginal advantage for the latter, which
disappears at individual posttest (Littleton, Light , Joiner, Messer, & Barnes,
1992). Thus, it seems that at least some of the benefits that we attributed in the
first instance to cognitively productive forms of overt interaction between
children working at the computer together may not, in fact, depend on such
interaction . Rather, they may depend, at least in part, on more covert processes
arising from participants' perceptions of the social situations created within the
experimental situation.
Meanwhile, we also had to consider a worryingly large gender difference in
response. One obvious possibility in relation to this is that superficial
characteristics of the software (namely, the scenario within which the task was
set) might make a difference. We recast the software in another version,
isomorphic in all respects with the first, but with a very different storyline . Here
some bears had gone for a picnic but had forgotten the honey, so they had to go
back for it. There were honeymonsters in the river, however, which necessitated
a detour by hot air balloon. The substitution of this version for the King and
Crown version made a dramatic difference. From being markedly inferior on the
King and Crown, the girls showed themselves as good or better than the boys on
the Honeybears (Littleton et al., 1992; Littleton, Light , Barnes, Messer, &
Joiner, 1993). We have made direct comparisons of boys' and girls' performances
on the King and Crown and Honeybears with children (N=48) all working one to
a machine in a coaction condition. Figure 1 shows the resulting pattern of
performance, which generated a significant gender by software interaction . We
231
repeated the study with a new sample of children and with an even closer
isomorphism of the two versions (the new King and Crown here being called
Pirates).
6
5
4
Girls
Boys
Software Type
Honeybears
Fig. 9.1. Levels of performance for boys and girls using either the King and Crown
or the Honeybears version of the task.
e
"3
:;j
'c.5
Girls
Boys
c
3 2
Pirates
Honeybears
Software Type
Fig. 9.2. Levels of performance for boys and girls using either the Pirates or the
Honeybears version of the task.
232
The results, as can be seen in Figure 9.2, show a close replication of the
previous findings. A not dissimilar pattern of responses also emerges from a
recently completed study (Littleton & Light, 1994) in which we looked at the
performance of girl-girl and boy-boy pairs working for one half-hour session on
either the Pirates or the Honeybears software. Here again, the girl-girl pairs
showed markedly superior performance on the Honeybears, whereas the boy-boy
pairs responded similarly to the two software types.
The reason for the girls' greater responsiveness to the Honeybears is difficult
to articulate precisely . It was apparent that many of them identified with the
characters, and afterwards they spontaneously talked about which bear was their
favorite. More than one talked about taking particular bears on their journey to
get the honey because "they wouldn't want to be left behind." This kind of
identification with the characters was not apparent among the boys or with the
King and Crown . It seems that these software hooks served to engage the girls
with the Honeybears' software. In themselves, these kinds of identification and
projection will not help to solve the task. Indeed, they could potentially hinder
solution. It is apparent, however, that, given an initial motivation to engage
with the task, the girls were every bit as capable of handling the interface and
thinking their way through the problem as were the boys. Whatever the precise
nature of the mechanisms underpinning the lability of the girls' response, the
results as they stand serve as a caution to all who develop software for children,
whether for use in the home or the school. Gender differences in response can be
massive, but they are far from immutable. The imagery or metaphors used in the
presentation of the task can have an influence out of all proportion to their
significance to the designer. Here, as elsewhere (e.g., Donaldson, 1978; Light &
Butterworth, 1992), context effects exert a critical influence on cognitive
performance and can affect not just the absolute difficulty of a task, but also its
relative difficulty for different groups of children.
In addition to the question of gender differences in response to computers, there
is the question of patterns of interaction between boys and girls while working
with computers . The issue of whether girls are disadvantaged by working with
computers in the presence of boys is one that has attracted a good deal of
attention in educational circles recently, and there is a significant move in the
United Kingdom toward segregating the sexes for computer-related activities,
especially out-of-classtime computer activities (Culley, 1993). Existing
empirical studies, however, show rather conflicting results . For example,
although Underwood, McCaffrey, and Underwood (1990) found evidence that
girls were disadvantaged by working with boys, Hughes, Brackenridge, Bibby,
and Greenhough (1988) found the opposite, with the girls actually having an
advantage in one study and with no differences in several subsequent studies.
Using the Honeybears, we have, in collaboration with David Messer and
Richard Joiner, recently undertaken a large study (N=120 eleven-year-olds) in
which we had children working either in coactive or interactive pairs, in girl-girl ,
girl-boy, or boy-boy pairings. The children were matched on the basis of an
233
individual on-task pretest , both within the particular pairs and across the
conditions . Overall, while they were actually working together, the children in
the interaction condition did slightly better than the children in the coaction
condition. This advantage disappeared, however, when the children were
individually posttested a week later on a slight variant of the task.
6
::IE
..~
<:
:3
::IE
Single Gender
Mixed Gender
Boys
Girls
Fig. 9.3. Levels of performance for mixed and single gender pairs for the
interacting pairs
Single Gender
Mixed Gender
Boys
Girls
Gender
Fig. 9.4. Levels of performance for mixed and single gender pairs for the coacting
pairs
234
In the interaction condition, the boys and girls did about equally well, and it
made no apparent difference to their performance whether they were working in
same or mixed gender pairings. In the coaction condition, however, where the
children did not have any opportunity to interact with one another, the mixed
gender pairings produced a marked and statistically significant polarization of
performance, with the girls in mixed pairs doing worse at posttest than girls in
other conditions and the boys in mixed pairs doing much better than boys in
other conditions. This result is illustrated in Figure 9.3 and Figure 9.4.
The absence of gender polarization of performance in mixed pairs in the
interaction condition was consistent with our previous findings using this
software. However, the discovery of such polarization in the coaction condition
was very unexpected. As a check on these results, we ran a further study, with
Stuart Bales, the results of which were reported briefly by Light (1994) . Using
the Honeybears software again, we compared the performance of pairs of elevenyear-old children working on the task during a single session under coaction
conditions. One third of the pairings were boy-boy, one third boy-girl, and one
third girl-girl. Once again , exactly as in the previous study, we obtained a gender
x pair-type interaction, with boys performing markedly better than girls only in
the mixed gender pairings. The effect that a number of educationalists have been
concerned about (namely, the fostering of gender differences in mixed gender
groupings) occurs quite markedly but only when the children are working
alongside one another without interaction. When they are actually collaborating,
the same polarization does not seem to happen .
We had not predicted these results, and explaining them is not straightforward.
We know from attitude questionnaires and interviews that the children subscribe
to the view that, in general, boys are more interested in and more capable with
computers than girls. But we also know that, with this software, the girls are
actually at least as good as the boys, and, in the first study , we had paired the
children so that (although they did not know it) the partners always had the same
pretest scores. We might speculate, then, that the opportunity for interaction
provides some feedback to the children on how they are doing relative to their
partner and that this feedback will tend to undermine their gender stereotypes. In
the case of the coactive pairs , however, these stereotypes may be evoked by the
situation of working alongside a partner of the opposite sex and cannot be
ameliorated by feedback. In the first of these studies, the two children did not
come from the same school class, so their knowledge of one another's abilities
was limited. Although the situation was not explicitly competitive, one could
imagine that the children would be anxious about their own performance relative
to their partner's, and that the threat of losing face might be particularly acute for
the boys in the mixed gender pairs. The absence of any feedback during the
session could only serve to heighten this process of social comparison.
These results obviously need to be explored further. But, for now, the message
for teachers might be that, instead of segregating boys and girls when working
with computers, they should foster as much interaction between the genders as
235
236
some way to go to establish at what stages and in what ways the change in
software impacts on performance.
The most recent finding, that gender polarization in the mixed pairs is much
more marked in non-interactive than in interactive pairs, is also interesting, not
least in the light of current debates about the best way to redress gender
differences in the context of educational computing. The finding also serves to
highlight the centrality of processes of social comparison in these kinds of
learning situations. The contemporary work of Monteil and colleagues in
Clermont-Ferrand (e.g., Monteil, 1993; Huguet, Chambres , & Blaye, 1994),
which stems from social rather than from developmental psychology, illustrates
the extent to which children's expectations of their own success relative to their
classmates can impact on their learning. Huguet et al. (1994) have established
recently that an artificially raised expectation of success on our Honeybears task
can lead to enhancement of children's performance.
Developmental psychologists working in a number of fields are gradually
coming to appreciate the all-pervading nature of contextual effects on cognition
(e.g., Forman, Minick, & Stone, 1993; Light & Butterworth, 1992). One aspect
of this belated shift involves the recognition that the social context of a
cognitive task embraces not only direct interpersonal interactions but also the
social norms, expectations, representations, and comparisons that condition such
interactions. In the end, taking this wider sense of social into account in our
research may turn out to demand richer and more diverse research methods than
those used in the research reported here. Nonetheless, our rather tightly
circumscribed experimental approaches have perhaps justified themselves, at least
to the extent that they have served to highlight their own limitations!
Acknowledgments
The work reported in this chapter was supported in part by the Leverhulme Trust
and in part by the United Kingdom Economic and Social Research Council. We
would like to thank the Year 7 staff and pupils of Rickley County Middle
School, Bradwell Village Middle School, and Alex Campbell County Middle
School , all in Milton Keynes, England, for their support and cooperation.
237
References
Barbieri, M. S., & Light, P. H. (1992) . Interaction, gender and performance on a
computer-based problem solving task. Learning and Instruction, 2(3), 199-213 .
Barbieri, M. S., & Light, P. (1993) . Risolvere insieme un problema al computer:
Differenze di genere e del tipo di interazione [Problem solving together on the
computer: Differences according to gender and type of interaction]. In C .
Pontecorvo (Ed.), La condivisione della conoscenza [The sharing of knowledge]
(pp. 305-326) . Firenze : La Nuova Italia .
Blaye, A, Light, P., Joiner , R., & Sheldon, S. (1991). Collaboration as a facilitator
of planning and problem solving on a computer-based task. British Journal of
Developmental Psychology, 9, 471 -483.
Blaye, A., Light, P., & Rubtsov, V. (Eds.). (1992). Interactional learning situations
with computers. European Journal of Psychology of Education, 7(4), 257-372.
Crook, C. K. (1994). Computers and the collaborative experience of learning.
London : Routledge.
Culley, L. (1993) . Gender equity and computing in secondary schools: Issues and
strategies for teachers. In J. Beynon & M. Mackay (Eds.), Computers into
classrooms: More questions than answers. London : Falmer Press.
Doise, W. (1990). The development of individual competencies through social
interaction. In H. Foot, M. Morgan, & R. Shute (Eds.), Children helping children
(pp. 43-64) . Chichester, England : Wiley .
.
Doise, W., & Mugny, G. (1984) . The social development of the intellect. Oxford,
England : Pergamon Press .
Doise, W., Mugny, G., & Perret-Clermont, A-N. (1975) . Social interaction and the
development of cognitive operations. European Journal of Social Psychology, 5 ,
367-383 .
Donaldson, M. (1978). Children 's minds. Glasgow, Scotland : Collins .
Forman, E. A., Minick, N., & Stone, C. A (1993) . Contexts for learning:
Sociocultural dynamics in children's development. Oxford, England: Oxford
University Press .
Griffin, P., Belyaeva, A, & Soldatova, G. (1992) . Socio-historical concepts applied
to observations of computer use. European Journal of Psychology of Education, 7,
269-286 .
Howe, C., Tolmie, A , Anderson, A, & Mackenzie, M. (1992) . Conceptual
knowledge in physics: The role of group interaction in computer-supported
teaching . Learning and Instruction, 2, 161-183.
.
Hoyles, C., Healy, L., & Pozzi, S. (1992) . Interdependence and autonomy : Aspects of
group work with computers . Learning and Instruction, 2, 239-258.
Hughes , M., Brackenridge, A, Bibby, A, & Greenhough , P. (1988) . Girls, boys and
turtles. In C. Hoyles (Ed.), Girls and computers (pp. 31-39). London: University of
London : Institute of Education . (Bedford Way Papers No. 34).
Huguet, P., Chambres, P., & Blaye, A. (1994) . Interactive learning: Does social
presence explain the results? In H. C. Foot, C. 1. Howe, A. Anderson, A. K.
Tolmie , & D. A. Warden (Eds.), Group and interactive learning (pp. 445-451).
Boston : Computational Mechanics Publications
238
Joiner , R., Messer, R., Light, P., & Littleton, K. (1992 ). Peer presen ce and computerbased learning. Hatfield Polytechnic Research Report, PSY/92-10JOI. Hatfield,
England .
Jones , A., & Mercer, N. (1993) . Theories of learning and information technology. In
P. Scrimshaw (Ed.), Language, classrooms and computers (pp. 11-26) . London :
Routledge.
Light, P. (1986). Context, conservation and conversation. In M. Richards & P. Light
(Eds.) , Children of social worlds (pp. 170-190). Cambridge, England : Polity Press .
Ligh t, P. (1994, September). Social comparison and computer-based learning. Paper
presented at the International Conference on Group and Interactive Learning,
University of Strathclyde, Scotland
Light, P., & Blaye, A. (1990). Computer-based learning: The social dimensions. In
H. Foot, M. Morgan , & R. Shute (Eds.), Children helping children (pp . 135-137) .
Chichester, England: Wiley.
Light, P., Buckingham, N., & Robbins, H. (1979) . The conservation test considered
as an interactional setting. British Journal of Educational Psychology, 49 , 304310 .
Light, P., & Butterworth, G. (1992) . Context and cognition. London: Harvester
Press .
Light, P., & Foot, T. (1987). Peer interaction and micro-computer use. Rassegna di
Psicologia, 2/3, 93-104.
Light, P., Foot, T., Colbourn, C., & McClelland, I. (1987). Collaborative
interactions at the microcomputer keyboard. Educational Psychology, 7, 13-21.
Light, P., & Gilmour , A. (1983). Conservation or conversation? Journal of
Experimental Child Psychology , 54, 356-363 .
Light, P., Gorsuch, C., & Newman, J. (1987). Context and communication in the
conservation task. European Journal of Psychology of Education, 2, 73-82.
Light, P., & Perret-Clermont, A.-N. (1989) . Social context effects in learning and
testing. In A. Gellatly, D. Rogers , & J. Sloboda (Eds.), Cognition and soc ial
worlds (pp. 99-112). Oxford , England : Clarendon.
Littleton, K., & Light, P. (1994 , September). Gender and software effects i n
children's computer-based problem solving: Analy sing interaction . Paper
presented at the International Conference on Group and Interactive Learnin g,
University of Strathclyde, Scotland.
Littleton, K., Light, P., Barnes , P., Messer , D., & Joiner, R., (1993, March). Gender
and software effects on children's computer-based problem solving . Poster
presented at the annual meeting of the Society for Research in Child Development.
New Orleans, LA.
Littleton, K., Light, P., Joiner, R., Messer, D., & Barnes, P. (1992). Pairing and
gender effects on children 's computer-based learning . European Journal of
Psychology of Education , 7, 311-324.
Mercer, N., Phillips, T., & Somekh , B. (1991) . Spoken language and new
technology: The SLANT Project. Journal of Computer Assisted Learning, 7, 195202 .
Mevarech, Z ., & Light, P. (Eds.). (1992) . Cooperative learn ing with computers.
Learning and Instruction, 2 (3), 155-285.
Monteil, J. M. (1993) . Soi & le contexte . [The self and the context] . Paris : Armand
Colin .
239
Part Three
Chapter 10
Abstract
This chapter develops the view that workplace learning and, more generally,
development can be conceived of as comprising two different processes :
acquisition and coordination. Although many studies of learning and development
focus on the problem of acquisition, we develop the view that coordination is the
more central problem. Knowledge systems can be acquired, but, if they are not
coordinated with existing knowledge systems, there is little development - only
an accretion of more knowledge. Little is known about the conditions under which
such coordination can be optimized.
Empirically, this chapter examines some linguistic markers of conceptual
coordination, under conditions that we have identified as fostering coordination
of knowledge systems . In particular, we examine the discriminatory values of
several discourse-related measures , such as intrasentential code-switching. We
show that coordinated knowledge states show up in the way language is used and,
therefore, that linguistic markers not only are useful for producing cognitive
measures but also can be used to examine the means by which membership in
communities of change can be interactively produced .
244
Joseph Glick
245
stretching from the mid-1960s to the present, with special focus on a research
program begun by Mike Cole, John Gay, and me in the late 1960s, and continued
and extended by Sylvia Scribner in the 1970s and 1980s, and currently being
extended by Cole and his colleagues and by this author and several students (Lia
DiBello, Elena Zazanis, and Jessica Kindred) who had started work with Sylvia
and have continued the work with me since she died. The theme of this chapter
then is "how we got here," and how we got from a traditional starting point to
share in the common conception just laid out. I take this approach with the belief
that our laboratory(s) have taken a journey that, in many ways, is characteristic of
some of the shifts in topic and perspective that the larger field within which we
work is going through.
Our general project had its origins in an attempt to understand the workings of
culture in human cognition. In the 1960s and early 1970s, we became involved in
a project that formulated this effort in terms of comparative studies: looking at
and comparing, for example, the performances on cognitive tasks of people in
Liberia with performances on cognitive tasks of people in the United States or
Mexico. Our hope was that comparative studies would enable us to make
orthogonal certain classes of variation that are inevitably confounded in
intracultural studies of development. For example, in a book authored by Cole,
Gay, Glick, and Sharp (1971) called The Cultural Context of Learning and
Thinking, we made a great deal of being able to disentangle the influence of
schooling and age, by working in an area where schooling was not yet universal
and where people could opt in to schooling at almost any age, because the school
system (as understood in Western educational terms) was relatively new. Thus, we
could compare first graders, for example, who were 6 with those who were 40.
Although this approach was clever, it brought more problems than solutions.
Working cross-culturally allowed us to get a handle on some independent
variables (and even here we had to put on some pretty restrictive blinders), but we
quickly became disenchanted with our experimental strategy which , at its heart,
involved the assumption that one could compare different groups on common
dependent variables.
The problem was that the tasks we used seemed to have different meanings in
the different cultures. We found that slight modifications of the comparative task,
which were made to make the task more culturally relevant, often led to large
changes in measured performance . Recognizing this as a problem, we began to
systematically vary tasks, either in materials used or in format of the experiment,
and we found that performance was not stable across the variations; people could
perform well or poorly depending on how we arranged the task. The classic
experimental formulation of the issues, independent variables whose effects were
measured by dependent variables, suggesting some intervening variables, seemed
inadequate to the problems we were facing. If we could make some independent
variables nicely independent, we tripped over dependent variables that were not
stable enough across their possible embodiments. And, if we found stable
dependent variables, this stability was often due to the operation of some
246
Joseph Glick
presupposed and hence unseen independent variable that was producing the
stability, for example, tasks whose performance demands mirrored the expected
performances that might come from within a cultural group for whom schooling
and the discourse practices and activities of the schooled were common points of
reference. For reasons such as these, we found that schooling seemed to be a
major variable in determining task performance. This turned out not to be because
of the influence of school as some abstract independent variable, but because of
the influence of schooling on the interpretation of the task itself.
The most trenchant discovery from this period of our research effort was that
generalities about cultural differences or the influence of such variables as
schooling were too loosely specified. The world did not neatly arrange itself into
variables whose impact could be traced in a general form . Indeed, this is a mild
way of saying that we discovered that what we wanted to discover was
undiscoverable in the terms in which we set out to discover it.
It soon became clear that we had to seek new terms of engagement. This
direction took us to the point of realizing that, rather than looking for variables to
provide the bird's-eye view that would allow us to compare, we had to find ways
of getting a more actor's-eye view that would get us closer to the concrete. Our
focus was shifted from variables to more microscopically analyzed situations and
performances within situations. Although comparison and grand explanations in
terms of variables was a lofty goal, it was a goal that had to await more engaged
under standings of particularities. We had to get closer to have even the chance of
getting to an overview of the big picture.
A critical way station on this march was embodied in work on the impact of
literacy on cognitive functions (Scribner & Cole, 1981). Rather than finding
generalized cognitive effects of being literate, Scribner and Cole found highly
specific effects related directly to uses of literacy by the person studied. They were
able to define different kinds of literacy and different uses for the different kinds.
Rather than having a general cognitive impact, the measurable effect of being
literate was directly related to the uses to which literate practice was put. It
seemed that, rather than looking for a generalized cultural/developmental variable,
one had to look at much more specific cases: that is, at what people actually did
and what it took to do those things.
In other words, the work had shifted from looking for general developmental or
cognitive rules, structures, or variables to looking for very specific grounded
cognitions or, more properly, to looking at cognition that was situated in real
activities (Lave, 1988; and see McDermott, 1993, for an update of one of the
seminal documents of this period, a paper by Cole, Hood, and McDermott entitled
"Ecological niche picking.")
247
development
Situated
cognition
Knowledge is carried
with the person
Defined by flexibility
- Generalized skills
- Transferable
Defined by efficiency
- Specialized skills
- Maximally adapted to local conditions
248
Joseph Glick
249
The dairy studies led Scribner to expand her research effort to looking at other
work sites for similar embeddedpractices and cognitive principles. The research
strategy employed was one that began with ethnograph ic observations of selected
work episodes, which yielded candidate practices that seemed to embody some
cognitive rule. The observational data would then be followed by quasiexperimental tests that simulated the work practice but with certain experimental
controls to test the limits of the rule that was thought to have been found.
As the engagement with various work sites expanded, Scribner turned her
attention to an increasingly common phenomenon in the workplace: The
introduction of technologies that transform the nature of work itself. What would
happen to the local knowledge that skilled workers had when a new form of
knowledge was demanded within the workplace? What are the traceable
relationships that exist between expert knowledge (of the workers) and expert
systems (of the machines)?
When she died, Scribner had been working on this problem in two factories that
were undergoing such technological transformation: a primary site, here called
Kemps, and a comparison site, here called Intek. A cornucopia of topics was
under study, including the following: the manner in which practical knowledge of
stockroom practices is communicated by more expert workers to less expert
workers, according to what principles and involving what operations; did such
cognitive operations as thinking in terms of ratios , which was involved in the
common stockroom practice of "counting by weighing," change when new kinds
of scales got introduced in the workplace? ; and, generally , what was the specific
orchestration of skills that was necessary to the performance of particular jobs,
and how were these related to the structure of the physical artifacts and conditions
that applied in the workplace?
The general thrust of these studies was an extension of the earlier work on dairy
workers, which might best be described as compensatory. The studies of
cognition in practice represented an attempt to identify, in people who are often
given the social status of low-level workers in work settings, sets of important
cognitive abilities involved in doing the job , which, once identified, might
contribute to a rethinking of such fundamental social/occupational classifications
as the division of labor between mental and manual components and the
accompanying class/occupational structure that went along with it (e.g., the
division of employees into hourly and salaried and assumptions about what kinds
of prerequisite training employees should have for different job classifications).
These were all matters not only of great psychological interest but also with
enormous social and educational implications .
These issues assume even greater importance in periods of great social and
organizational change. And, if it was nothing else, the time period within which
the research was being conducted was a period of great social and technological
change . As the research unfolded , we found ourselves being drawn more and more
to the issue of change. During the mid-1980s, studies began to examine the
impact of changing local technology, such as the impact of the move from
250
Joseph Glick
251
Four principles underlie this description, all of which are critical to our analysis.
They are set forth in the form of the following assumptions:
(1) That the logic of management-guided production and the logic of production
systems as they exist in practice are different;
(2) That management systems had previously been unable to control production
systems , but that, with the increasing availability of computers integrated into
manufacturing, the control aspect of management practice is increasingly
possible ;
(3) That the embedded rationalities of production will have to meet, at some
point, with the ideological rationalities of management control;
(4) That progress in this environment of change will require either the total
supplanting of one system by the other, or that a point of integration must be
found.
Much would depend on the degree of fit between the system of production as it
existed and the new demands of the rationalized production system.
In essence, the logic of the MRP system is quite different from the logic of
production on a number of key points. Rather than thinking of manufacturing
production as a process that moves forward in time - from available materials,
upon which labor and machine processes are applied to yield a finished product MRP thinks of manufacturing as a process that moves backward in time from
customer orders. Thus, the system plans backwards. In order to do so, it needs
several key pieces of information. It needs to know the structure of manufactured
items so that it knows the constituent parts of finished items and hence can plan
manufacturing down to a relevant level of detail. It must also know the process of
part assembly so that it can plot what comes before what. This is represented in
an indented bill of materials that outlines the relations among the parts in terms
of a family structure, that is, of parent-child relations among raw parts and the
sub-assemblies into which they go. Once the system knows the item structure
and the parent-child relations, it can then introduce, into planning, the concept of
lead times for the acquisition of materials, so that the various parts come together
in an assembly process just as they become needed .
The logic of the MRP system introduces a new language for describing things.
Thus, a finished item, which we are likely to call "the thing that is made," is
redescribed as a "0 level item" that is composed of "levels" beneath the "0 level"
252
Joseph Glick
253
occur over longer historical cycles and thus allow for non-catastrophic
adjustments of work flow.
Typically, people engaged in production know the reality of their businesses.
They have real relations with real suppliers in a real industry in a real plant. What
they do not know is the constraint on the logic of the computer program and the
many ways in which it can be unreal. Thus, it is not surprising that such systems
work to a greater or lesser degree across implementations. In some cases the
computer "fouls things up" and becomes progressively less used, or, more
precisely, ignored, while in some cases it helps a manufacturing plant to become
more smoothly efficient in its operations.
Thus, the MRP system, although a powerful analytic tool, must be harnessed
and adapted to each work environment into which it is placed. According to
industry manuals (e.g., Wallace, 1990), the key to a good implementation of
MRP is a rather long period of implementation that operates under the following
major constraint:
Implementing MRP-II properly requires a great deal of time and effort on the part
of many people throughout the company : Data must be made more accurate ,
people must be educated and trained, new software must be acquired and installed,
new policies and procedures must be developed and made operational....
Successful implementations are done internally. In other words, virtually all of
the work involved must be done by the company's own people. The
responsibility can't be turned over to outsiders, such as consultants or software
suppliers. That's been tried repeatedly , and hasn 't worked well at all. (Wallace,
1990, p. 13)
254
Joseph Glick
255
Ethnographic observations at the two sites supported the hard business data of
differential success. At Intek, we found that a "culture of secondary artifacts" had
been created by employees to elaborate the read-outs of the computer screens.
Thus , computer terminals were surrounded by print-outs of Lotus spread sheets
and the like. Nothing like that was happening at Kemps. And Intek employees
were even able to use the output of the system to make savvy predictions about
production processes. On one tour of the factory floor at Intek, an employee
glanced at a computer screen and said, ''That machine is about to go down,"
pointing to a machine on the floor. When we asked, "How do you know?" , she
explained that the output of the machine was progressively deviating ("look there,
see those figures") from its expected output, and therefore it was failing. Knowing
this, the machine was pulled from production for fixing, and a secondary machine
was put on line. A costly breakdown in production was avoided. Although Kemps
had similar screens (but no secondary artifacts), the information on those screens
was not used the same way. Indeed, because the MRP system had not generated
materials savings at Kemps, it was not clear that the information on their screens
was being used at all.
256
Joseph Glick
switching in bilinguals. She analyzed (Zazanis & Glick, 1993) responses to probe
batteries in terms of whether a stretch of talk reflected the structure and semantics
of production language or of MRP language . Finding that these languages can be
analytically distinguished, she then examined their relation to one another in the
workers' talk. In particular, she focused on whether a worker's language
productions kept the two knowledge systems separate or blended them in
compound utterances.
There is some suggestion in the literature that the phenomenon of intrapropositional code switching can be used as an index of the degree of balance
between language systems in bilinguals. Intrapropositional switching, then,
might be an index of a greater degree of balance than switching that occurs across
propositional boundaries. When examined in this way, workers at Intek showed
greater evidence of intrapropositional code switching, as is shown in Tables 1 and
2.
Table 10.1. Intrapropositional and Interpropositional Code Switching Analysis
# OF PROPOSmONS
S1REAMS
1N1RA-
PROPOSmONAL
SWITCHES
Intek
Kemps
Production
MRP
125
156
42
58
Production
30
27
MRP
9
18
9
2
Discourseand Development
257
way on the first sort and another on a second sort) or not (i.e., sorting according
to one system on both sorts). The results of this analysis are indicated in Table 3.
Table 10.2. Mean Proportions of MRPlProduction Streams and Switches
at Kemps and Intek
Intek
Kemps
(N=12)
(N=14)
41.1
9 .1
39 .5
10.8
58.9
9.1
61.8
12 .2
66.1
18.0
57 .5
Proportion of
MRP Streams
Mean
SD
Proportion of
Production Streams
Mean
SD
Proportion of
Inter-Stream Switches
Mean
SD
21.9
Proportion of
Intrapropositional
Mean
SD
5 .7*
4.4
0 .9**
1.0
258
Joseph Glick
SORTTYPE
Flexible
Inflexible
BYPLANT
Intek
Kemps
9 (86%)
2(14%)
5 (36%)
9 (64%)
BY SWITCHING
Flexible
Inflexible
Intrapropositional
No intrapropositional
13 (76%)
1 (12%)
4 (24%)
7 (88%)
One of our research team, Jessie Kindred (Scribner, DiBello, Kindred, & Zazanis
1991) had been working most heavily on an ethnographic analysis of the work
sites. She noticed some differences in the way that jobs were performed at the two
plants. One way of specifying what a job entails is to look carefully at the kinds
of activities that are dominant in the performance of that job . We call this a
leading activity analysis. The leading activity of a job is not fully specified by a
job title. You have to look carefully at what people do and then try to make sense
of it.
Looking at the issue that way, we could distinguish between two types of
organizations for any job. One we call procedural; the other we call constructive.
A procedural activity is one in which there is a clear specification of both the goal
of an activity and the means by which the goal is to be accomplished. A
constructive activity may have the same goal as a procedural activity, but there is
a less clear specification of the means. For example, I might specify how to get
from point x to point y by providing a path statement: "First, you turn left; then
after 10 feet, you turn right , and after 15 feet, you turn left again, etc." Or, I
259
might provide you with a map and let you figure out how to get from here to
there. Jobs in manufacturing environments can be path-like or map-like.
Kindred's analysis looked at jobs in this way, carefully specifying activities and
carefully specifying the degree to which that activity was organized by a clear and
constraining specification of the means to accomplish it.
Several illuminating findings emerged from this analysis. Most obvious, it
became clear that job title did not completely specify the degree of constructive
latitude accorded to a particular worker. Job titles may be seen in this sense as a
statement of the goals implicit in a job. The procedural-constructive distinction
assumes roughly common goals but identifies different relationships between
goals and means . That having been said, the two plants differed overall in the
degree to which their workers were given such constructive latitude in achieving
job goals . Kemps used more specified procedures than did Intek. In fact, all of
those artifacts that festooned the area around MRP screens at Intek had been
constructed by the workers, sometimes for amusement, sometimes as a secondary
tool, and sometimes as a display of their own constructive latitude. In many
plants, such employee-created artifacts are considered undesirable, allowing for too
much individual variation in the workplace. Here we found that the constructive
engagements that such artifacts betokened were key elements of a success story.
Lia DiBello and I (DiBello & Glick, 1992) examined some of these data from
the point of view of plotting the relationship between the degree of constructive
activities on the job in each of the plants to one of our measures of conceptual
integration, namely, the card sorting measure. Without going into excessive
detail, a few things are clear. First, we found considerable variation within job
title within both plants. This suggests that there is some variation that relates to
individuals within organizations . Second, the overall level of constructive
opportunity is higher at Intek (mean of 70% constructive activities) than at
Kemps (59.8%). Third, the variability in constructive opportunity is higher at
Kemps. There is an interaction between variability attributable to individuals and
overall organizational ethos . Fourth, and finally, constructive opportunity and
conceptual integration covary. When one is high, the other is likely to be high,
too. This shows up in a Pearson r of .67 between amount of constructive
opportunity and our measure of cognitive integration. We plotted a similar
correlation for educational level and conceptual integration and found no
correlation at all.
The MRP system had come to these two plants in quite different ways. Intek is
a family-owned business, and the owner has taken a more or less hands-off
attitude, allowing his children and managers the freedom to invent new procedures
to make the business run more smoothly. Kemps is more traditionally organized.
This history seems to have some current consequences. At Intek, the system is
seen as a part of the factory as a whole, and all people are given some level of
training in its use (even the receptionist) . At Kemps, the system is part of a
management control system, and training in its use is reserved for the managerial
260
Joseph Glick
ranks . The system shows up in employees ' lives as a set of orders from the front
office rather than as a tool for all to use.
We might wonder whether this history has been responsible for the different
patterns of implementation success that we have observed, both in business terms
and in terms of cognitive and linguistic measures. This question brings us up
against one of the primary limitations of the project that has been described so
far. We started studying the integration of management control systems with
production after the systems had been in place for a number of years. Our late
entry did not allow us to look at the processes by means of which these patterns
had become established.
We were left with serious questions about how it came to be this way, which
we cannot answer with the current data set. We are now working in an area where
some of these questions can be answered.
261
SITUATED COGNITION
School/family as
- Prerequisite skills
agent of preparation
Guidance toward
progress
- Teaching, scaffolding
- Piggybacked demands
- Development as constraint
Focus on
Focus on legitimacy
as one who knows
- Displayable competen ce
knowledge
- Testable competence
- Younger to older
Notions of efficiency
alongside of progress
- Novice to expert
Focus on individuals
- Knowledge is at issue
Focus on organizations
- Legitimate participation at issue
Notions of progress
As can be seen from the contrast just shown, the cognitive development model
262
Joseph Glick
Our current research activities are exploring issues within the situated cognition
model. For our group, the issue has become one of identifying the means by
which people accomplish or fail to accomplish legitimacy of participation within
a culture of innovation . Clearly, some of the variance in this process will be
related to management philosophy and institutional ethos, but we also expect that
some of the variance will be explained by a closer understanding of how people
act within their work environments. Although there may be institutional agendas,
people nonetheless seem to be involved forever in carving out a niche within the
constraints of the agenda. Technological innovation can be a powerful force in
providing more latitude for niche finding, even though the underlying motivation
for the technological innovation (as is the case with MRP) is to institute more
top-down control.
We suspect that our finding that knowledge states show up in ways of
speaking, not just in terms of the content of propositions but also in the more
fine-grained structure of discourse, may be an important analytic tool for
investigating processes of niche construction. Discourse may not only reflect the
results of constructive engagement, but it may also be the means by which
constructive opportunity is earned. We are suggesting and already have some
evidence to support the notion that some people begin speaking knowledgeably
long before they are knowledgeable. It could be that display of linguistic
competence earns workers constructive latitude in their jobs .
If these suspicions bear out in findings, we may be in for some serious
rethinking of what the prerequisite skills are for functioning in a modern
manufacturing environment, where the kinds of changes that we have begun to
see with MRP will more likely be the rule than the exception. It could be that
conceptions of competence or required competence for various job classifications
will need to be shifted from the cognitive development inspired intrapsychic
notions, such as abilities or skills, to more situated cognition inspired
interpsychic notions, such as discourse. When it comes to cultural practice, what
improves the chances of a person's being a good practitioner more than being able
to enter into the interpsychic community of practice as a full member.
What I am trying to suggest here, perhaps too elliptically, is that the path that
I described earlier, from cross-cultural studies of general cognitive operations to
cultural practice notions of embedded and situated cognition, will be taken again
as we shift our focus from the skills embedded in work to issues of legitimated
participation in the workplace. It could be that if the opportunity is provided, the
rest will take care of itself. The problem then is to understand both how and why
opportunity is provided and how it can be earned.
I do not think that the issue will turn out to be simply an issue of expertise.
To account for differences in participation in terms of expertise is circular,
because what is often at issue is the opportunity to become expert. Rather I
suspect that there are some highly specific factors blended with some very general
factors involved. To be sure, knowledge and training in the intraindividual sense
263
are involved, we must but equally consider the general issues of interpsychic
selves and organizational philosophies.
This formulation of the issue would clearly recognize that modem workplaces
can and do support a number of different coexisting levels of competence. Under
stable conditions, these differences do not make much of a difference. However,
the situation changes once the workplace undergoes change. It is perhaps the
generalized competencies of impression management (such as language use in
discourse) that may be selecting for change survivors. The issue of earning
opportunity is a serious issue and one that has major relations to the current
manufacturing environment where re-skilling (as jobs are lost) and de-skilling
(where jobs are proceduralized) seem to go apace.
Although people leave schools, and the design of the institution of school is
predicated on that leaving, the design of the institution of the workplace has
become much less clear in recent times. On the one hand, the workplace seems to
be institutionally organized in terms of people staying within it. Thus,
knowledge in the workplace can be distributed between individual cognitions and
artifact design. However, as employment prospects in a global economy become
less certain, the stability assumption no longer holds for the workplace. People
may have to leave the workplace as well. And if they stay, they may have to stay
within environments that place more of the cognitive structure in the design of
job procedures and intelligent artifacts, raising the issue for the worker of whether
it is even worth staying in an environment where the worker counts less and less.
Although there is danger here, there is also opportunity. Many of the
technological innovations only seem to place the cognition in the artifact. Deeper
analysis generally shows that what is really involved is the construction of an
opportunity for there to be new conditions of engagement between workers and
workplaces. To the extent that the new conditions of engagement afford
opportunities for constructive engagements within the structure of the job, it can
be argued that the workplace is beginning to take on some of the characteristics of
the school. There are pockets of opportunity where quite general skills are
264
Joseph Glick
References
Beach, 1(., Zazanis, E., & Glick, J. (in preparation). Ratio in the scale or in the hand?
Does black box technology disconnect the mind?
Cole, M., Gay, J., Glick, J., & Sharp, D. (1971). The cultural context of learning and
thinking. New York: Basic Books.
DiBello, L., & Glick, J. (1992, October). The relative roles of on-the-job training and
classroom education for workers learning MRP-Il effectively. Presented at APICS
Society meetings, special academic session. Montreal.
Lave, J. (1988). Cognition in practice. New York: Cambridge University Press.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation.
New York: Cambridge University Press
Martin, L. M. W., & Beach, I(. (1992). Technical and symbolic knowledge in CNC
machining: A study of technical workers of different backgrounds. (Technical
report) University of California at Berkeley, National Center for Research in
Vocational Education.
McDermott, R. P. (1993) . The acquisition of a child by a learning disability. In S.
Chaiklin & J. Lave (Eds.), Understanding practice: Perspectives on activity and
context. New York: Cambridge University Press.
Sabel, C. (1991). Moebius-strip organizations and open labor markets: Some
consequences of the reintegration of conception and execution in a volatile
economy. In P. Bourdieu & J. S. Coleman (Eds.), Social theory for a changing
society (pp. 23-61). Boulder, CO: Westview Press & Russell Sage Foundation.
Scribner, S. (1984). Studying working intelligence. In B. Rogoff & J. Lave (Eds.),
Everyday cognition : Its development in social context. Cambridge, MA: Harvard
University Press.
Scribner, S. (1986). Thinking in action: Some characteristics of practical thought. In
R. J. Sternberg & R. I(. Wagner (Eds.), Practical Intelligence (pp. 13-30).
Cambridge University Press.
Scribner, S. (1990, May). The character of knowledge systems in the workplace from
the perspective of activity theory. Paper presented at The Second International
Congress for Research on Activity Theory, Lahti , Finland .
Scribner, S., & Cole, M. (1981) . The psychology of literacy. Cambridge, MA: Harvard
University Press .
Scribner, S., DiBello, L., Kindred, J., & Zazanis, E. (1991). Coordinating two
knowledge systems: A case study. Unpublished manuscript, City University of New
York, Graduate School, Laboratory for Cognitive Studies of Activity.
Wallace, T. E. (1990). MRP II:_Making it happen: The implementers ' guide to success
with manufacturing resource planning (2nd ed.). Essex Junction, VT: Oliver Wight
Limited Publications.
Zazanis E., & Glick, J. (1993). Discursive measures of cognitive structure.
Unpublished manuscript, City University of New York, Graduate School , Laboratory
for Cognitive Studies of Activity.
Chapter 11
Abstract
This chapter focuses on the user's actual behavior when he or she uses a computer .
With reference to concepts developed in the field of ethnotechnology and the
social psychology of cognitive development, a series of observations on humanmachine interaction are reported. The analysis of these observ ations shows that
human-machine interactions result in a specific inter actional space that can be
reduced neither to the user's abilities and technological knowledge nor to the
technological qualities of the machine. This result is further illustrated by a short
case study concerning word processing . First, some historical elements
pertaining to the development of this new tool are reported ; then, some examples
show the specific effect the use of word processors has in social and institutional
contexts . The conclusion is that using a computer consists of an indirect dialogue
between the users and designers . As a dialogue, the use of a computer includes
theoretical and methodological problems that are typical of human-machine
communication in general .
266
Introduction
With the emergence of new information technologies (from the television to the
computer), the number of studies on human-machine interaction has increased.
These studies have turned more and more to multidisciplinary approaches
pertaining to social sciences, computing, and technology. Such approaches have
already been adopted in some fields of psychology (e.g., industrial psychology)
and have been promoted in the teaching of technical disciplines (Podak, 1993).
Next, to give a broad outline of these studies, we mention some technological
fields that have extended their research to include a more multidisciplinary
approach:
- Computer-assisted design and production. In this field, users are more and
more dependent on the models implemented by software. This dependence raises
an important issue regarding the transfer of knowledge and competence from one
field to another: How can users make use of their previous knowledge and
competence in this new context? The answer to this question is very complex,
because, as several studies have shown (see, e.g., Waern, 1993), an analysis of
the characteristics of the tasks is not sufficent to predict transfer. To account for
this complexity and to improve conceptualization, it has proved necessary to draw
on other disciplines that have already tackled the same problems. But, of course,
the issue is then to examine to what extent concepts and results obtained in these
disciplines are applicable to the field of human-machine interaction.
267
Thus, in different fields, the use of a new technological tool has revealed that
social, ideological, symbolic, and affective dimensions, which go far beyond
technology itself, must be taken into consideration.
Within this debate, our own attention has been focused on the use of computer
science in school. Following the development of this educational innovation, two
striking facts have come to our attention: The first was the confrontation between
positive and negative debates, based mainly on ideological arguments, in which
little attention was paid to the actual use of the computer in context. The
assumption was that the users (at least the "skilled" ones!) would inevitably use
the computer in the "correct" way: that is, in accordance with the designers'
expectations . The second striking fact was that most of these ideological
arguments lacked reference to previous studies and knowledge acquired in the field
of educational psychology, giving the impression that the introduction of
computers into schools had rendered irrelevant previous knowledge about learning
and teaching.
In an attempt to pool our resources and benefit from our respective backgrounds
(in psychology for the first author and in mathematics and computer science for
the second), our general aim has been (l) to clarify some of the assumptions
underlying the debate surrounding the introduction of computers into everyday life
and into the school environment; (2) to collect a series of examples from existing
scientific literature and from our own observations of users' actual behaviors; and
(3) to examine to what extent results obtained in the field of developmental
cognitive psychology (in the post-Piagetian and post-Vygotskian streams in
particular) might help construct a theoretical framework that would enable us to
account for different studies and observations. We now present this framework and
show how it can explain different observations in the field of human-machine
interaction.
This chapter is divided into four sections. In the first, some basic theoretical
concepts pertaining to ethnotechnology and the psychology of cognitive
development are presented briefly. With reference to concepts such as semiotic
tools, sociocognitive conflict, and intersubjectivity, four issues regarding the
application of these concepts in the study of human-machine interaction are
raised. In the second section, these issues are illustrated with some of our own
observations and some examples discussed in the scientific literature. Our first
conclusion is that human-machine interactions result in a specific interactional
space that can be reduced neither to the user's abilities and technological
knowledge nor to the technological qualities of the machine. In the third section,
we present a short case study of a particular use of the computer: word-processing.
Some historical elements pertaining to the development of this new tool are
reported. Then, some examples are given to show the specific effect the use of
word processors has in social and institutional contexts. In conclusion, we argue
that using a computer is a socially shared activity mediated by a technological
tool and consisting of an indirect dialogue between users and designers. As a
268
269
interactions. Thus, a central subject matter for study is the analysis of the social
processes of creation and transmission of these semiotic tools.
The study of the role of social interaction in cognitive development and
learning is also a basic theme ofpost-Piagetian research, which we refer to here as
the socialpsychology of cognitive development. In a first generation of research
(for a more detailed account, see Perret-Clermont, Perret, & Bell, 1991), studies
relying on a three-step design (pretest, test, posttest) focused on the social
conditions enabling a peer interaction session to be efficient for subsequent
individual learning (see, e.g., Doise & Mugny, 1981/1984; Doise, Mugny, &
Perret-Clermont, 1975; Perret-Clermont, 1980). One of the main results of these
studies is that situations in which the interacting partners give diverging
solutions to a problem are likely, under certain conditions, to provoke a
sociocognitive conflict, which has the effect of leading each individual to
restructure his or her reasoning. In a second generation of research (e.g., see
Perret-Clermont & Nicolet, 1988), the emphasis was placed on the meaning of
the learning or test situationsin which children are requested to solve a problem.
Similarly, the focus shifted from observing the individual effect of a peer
interaction session to observing interactional dynamics of the partners. Under the
influence of scholars such as Ragnar Rommetveit (1979, 1992), the main purpose
of these studieshas been to examinehow the childreninvolved in a learning or in
a test situation (Grossen, 1988; Grossen & Perret-Clermont, 1994) give meaning
to the situation and understand the implicit assumptions on which the situation is
based. Attention has thus been paid to certain communication processes such as
iniersubjectivity, showing that the development of a new competence or the
learning of a given body of knowledge is intrinsically linked to the sharing of
certain states of intersubjectivity. As a consequence, the alternation of states of
sociocognitive conflict and states of intersubjectivity appears to be a condition
that enables individuals to coordinate their points of view and come to an
agreementwhile at the same time being able to benefit from their differences.
Other studies have explored the role of implicit rules, values, norms, and
expectationsin the learning process and in the construction of an answer in a test
situation, that is, analysis of what can be called an experimental or didactic
contract (Schubauer-Leoni & Grossen, 1993; Schubauer-Leoni, Perret-Clermont,
& Grossen, 1992). These studies constitute only a small proportion of a large
number of studies relying on the assumption that learning situations are basically
communication situations that are culturally rooted and whose meanings have to
be constructed intersubjectively during the interaction (e.g., see Edwards &
Mercer, 1987; Elbers, Maier, Hoekstra, & Hoogsteder, 1992; Light &
Butterworth, 1992; Nunes, Light, & Mason, 1993; Resnick, Levine, & Teasley,
1991; S1Hjo, 1992; and the work of those contributing to the present volume).
To sum up the main features of these two psychological approaches in
conjunction with ethnotechnology is to consider the individual's interpretative
activity as a fundamental part of their actions and as a determinant of the kind of
cognitive processes they use in concretesituations.
270
MicheleGrossenand Luc-OlivierPochon
The sociocultural approach referred to earlier leads us to study more closely the
creation of new semiotic instruments and their possible impact on human thought
processes at any given time in cultural history. This issue has often been raised
by authors from diversedisciplines and backgrounds (see, among others, Belyaeva
& Cole, 1989; Papert, 1980, 1993; Pea, 1985). Moreover, each medium
promotes certain thought processes: Books, radio, and television modify
intellectual processes, each in its own particular way (e.g., memory, mental
representations of situations, how connections are made; Greenfield, 1984). Two
examples illustrate this point: The first example shows how the computer
modifies the thought processes and research practices of a scientific community.
In 1976, a mathematical theoremcalled the four colour theorem (Appel & Haken,
1977) was demonstrated through the use of a computer. Many mathematicians
were disconcerted by the result: To verify the accuracy of the demonstration, both
the method and the demonstration program had to be validated. Was this still
within the realms of mathematics? Hence, with the birth of experimental
mathematics (Cohen & Norden, 1989), the method goes beyond a mere
broadening of single cases. In mathematics, the introduction of the computer has
provided, and still provides, the means to use specific cognitive abilities:
recursive, cybernetic, systemicreason. Recursive reasoning is a good example of
this evolution (see Vitale, 1989; Pochon, 1991). In fact, having first the structure
of a mathematical proof, recursiveness has become a relatively widespread
techniquefor the construction of computerprograms (control structure).
271
The second example shows that the computer could be seen as a sort of
272
a. The designers. who design the program according to the technical potential of
the machine and their own representations of the body of knowledge involved;
b. The teachers, who use the program in the classroom and who, depending on
their own assumptions about the body of knowledge involved and the use of
computer in the school context, give their own interpretation of, for example, the
program, its purposes, its "correct" use;
c. The pupils, who use the program according to their own interpretation of
how to use the computer, the discipline being taught, and the definition of the
situation (e.g., should they use the program to play or to learn).
Each of these partners has specific representations of the machine, of its use in
the school context, and of the body of knowledge contained within the program.
Thus, it is highly possible that the representations of one partner do not coincide
with those of another, a situation that could give rise to misunderstandings. Thus,
when someone uses a computer, he or she is confronted not just with a piece of
technology but also with the assumptions of other individuals.
This example also shows that using a computer is far more than a simple
interaction between two partners: a machine and a user. In fact, the computer can
be considered a screen hiding the designers, the programmers, and other
individuals (the teacher, for example) who put the computer at the disposal of the
end users.
Having its own logic and sometimes taking on the role of "adversary," the
computer is not a malleable object that will bend to the will of its user. It resists
and imposes its own way of doing things on the user. Thus , it can give rise to
certain sociocognitive conflicts that oblige the users to reflect on their own
cognitive activity and to solve some of the problems they encounter (see
Mevarech & Light, 1992; Light & Littleton, this volume) .
Furthermore, programmers also have certain assumptions and expectations with
regard to possible errors and the user's reaction to them. They have conceived a
model that is supposed to match the user's expectations. Vacherand-Revel and
Bessiere (1990) describe three ways in which programmers could approach the
user : They could defme the user type; they could take into consideration the
diversity of users as a function of their needs; or, through either direct or indirect
interaction, they could consider the reciprocal adaptation of the model as
originally conceived and the users' model in practice. The latter and most recent
approach fits our hypothesis regarding the construction of intersubjectivity,
namely , the creation of a common system of reference. In fact, our hypothesis is
that the computer does not come out of the interaction altogether unscathed but
undergoes changes that turn it into an object different from that originally
conceived of either by its creator or by its users . We expand on this point later.
273
When entering the "logic" of the machine, the user tries to find out how it works
and to interpret how it can be used. This can lead him or her either to modify the
original design of the computer by increasing its functions and performance or to
change its use by adding elements that the designers did not originally anticipate.
What is the nature of these modifications? In what ways do they change the
functioning of the computer?
The Modification of the Original Design of the Computer and
Software. The performance of computers is constantly improving.
Programming languages are evolving at increasing speed. At each stage, N, a
language facilitates the creation of tools, which in turn enables language of N+1
orders to be translated into language N. This evolution takes place at every level:
for example, system or partner. With this in mind, we have observed how, since
the appearence of the first Dbase 2, people use the system not only to manage
data but also to manage programs.
274
275
machine combined intelligence. Neither man nor machine is passive, and their
respective activities create an original interactive space; just as water is not found
in either hydrogen or oxygen alone, the global result of the man-machine system
is not determined by the sum of the parts.
By the same token, these observations enable us to put forward the hypothesis
that, depending on the context in which it is being used, the computer is not just
a set of determined technological characteristics . On the contrary, its good and bad
features may be deemed to result from the specific nature of the user's behavior.
In the next section, we apply this theoretical framework to a single example of
computer use: word processing. As we demonstrate, word processing offers
numerous examples of the specific impact that human-machine interaction has on
technological development.
276
277
Thread Typewriter
'Thread Microcomputer
Emergence of concepts
ter [
Processcontrol (1951)
Human machine
First interactive
device(screenand
communication (1955)
Interactive computer
(1960)
Eeclric t)'pewntet
1962
Word processing
(1963)
ypeWJieeler (J 970)
"Theprinter is a
file"(l972)
Apple D(1978)
Wordstar(1978)
mMpc (1981)
LISA 1983)
(1984)
MacIntosh
Visi04 (l984)
From then on, word processors developed rapidly under pressure from typewriter
manufacturers who played a mediating role between computer technologists and
laymen . Subsequently, and as a result of the interactions between typewriter
278
2More recently , a way to produce a help system for WINDOWS has been devised that
uses the word processor WORD .
279
280
impact on the social identities of those using them, because they introduce new
social distinctions between those who use them and those who do not.
At an institutional level, we also observed that some managers see this new
tool as a means of exercising their authority. For example, a director of a school
decided to impose the use of one particular type of word processor on the
institution. Because there was no economic reason for this decision and the
availability of compatible programs made this demand unnecessary, we
hypothesized that this decision had a symbolic function comparable to the choice
of a letter head or logo. The choice of a certain word processor represents an
institutional symbol , a sort of emblem, and becomes part of the institution's
identity.
Turning now to the impact of word processing on interindividual relationships,
five different, but not mutually exclusive, kinds of effects can be identified:
(1) Effects on the individual's self-definition and group membership identity.
Some word processors, such as WORD or WORDPERFECT, have generated
competing fan clubs who support the respective qualities of "their" program. As
in the preceding example, the program gains a symbolic value and becomes a
positive sign of the individual's identity and of his or her group membership.
(2) Effects on the type of interindividual relationships. For example, in the
school context, word processors are often used to publish newspapers written by
pupils and distributed to parents and other schools . Pupils of different schools
exchange their newspapers and keep in touch through this medium.
(3) Effects on the author's relationship withhis or herwritten production. Word
processing enables two or more individuals to easily write a text together : for
example, by attaching a file to an E-Mail message or by using a "travelling
diskette," which goes from one writer to the other. Because it is rather difficult , or
at least tedious, to compare the changes occurring from one version of the text to
the next, the writers' relationship with their own written product changes: The
text no longer seems to be the "work" of a given individual but becomes common
property .
(4) Effects on the written production itself. With the use of word processing ,
the text acquires a flexibility it did not have before. Small corrections, as wen as
deep changes in the textual organization, can be carried out quickly. Texts are
regarded as being more temporary, as drafts liable to change. French writer Michel
Butor provides some testimony on this point: Questioned about his last book
(Butor, 1992) , he states that the use of a word processor has changed his
conception of a text and led him to adopt a hypertext writing influenced by the
technological possibilities of the computer (Audetat, 1993; see also Dubuisson &
Weiss, 1989).
(5) Effects on the mode of transmission of expertise. Sometimes when they
work in a group (e.g., students working in computer rooms), users do not consult
the user's guide (assuming that they have one) but prefer to ask their neighbor for
assistance . Or, if they work alone, they might ask for information in informal
discussions with friends. In fact, users frequently become involved in discussions
281
about the problems that are presented by a certain program and how they solve
them. Such discussions sometimes result in the writing of a personalized user
guide that is distributed within certain social groups (e.g., students).
282
283
decided on (see the world of Dynabook to refer to Alan Kay's visions; Kay, 1992,
p. 199) . It would include multimedia capabilities and communication
possibilities. Thus, future word processors could, like the first ones, be machines
dedicated to that purpose only.
Conclusion
Using an interactional approach, our principal aim in this chapter has been to
show that human-machine interaction creates specific interactional spaces and that
an understanding of these requires that they be observed in concrete contexts.
At the current stage of our study, three main conclusions may be drawn:
- The first conclusion is that human-machine interaction is an interindividual
relationship mediated by a technological tool and consisting of an indirect
dialogue between users and designers.
- The second conclusion is that this indirect dialogue has many of the same
characteristics as any other communication between individuals, and it requires
interpretation activity on the part of the user in order to understand the logic of
the machine. Hence, ambiguity, negotiation of meanings, and conflicts resulting
from the confrontation of differing perspectives and assumptions are fundamental
parts of this indirect dialogue . Thus, the transparency- of a system raises a more
complex issue than is generally believed . An example is provided by the fact that
the implementation of database structures or the definition of fields (type and
length) is no longer made necessary by the designer. Users nevertheless come up
against these implicit definitions when they carry out unusual operations (e.g. ,
when going over the number of permitted characters or when changing the data
type) .
- The third conclusion is that this indirect dialogue results in a specific
interactional space that arises out of the intersection between the technological
qualities of the machine and the user 's own activities and skills . From this point
of view, technological innovations might also be considered to be a modification
of use resulting from the resistance of the machine to obey the user 's will and the
user 's interpretation of the use to which the machine was originally dedicated .
This chapter represents only an initial attempt to present a comprehensive
theoretical framework capable of accounting for heterogeneous observations. Such
an attempt involves some risks, one of which is the presentation of too generic a
3Computer scientists consider a procedure to be transparent if the different operations
necessary for it to be carried out effectively by the machine are imperceptible to the
user. This particular use of the term transparent (which hides by rendering "invisible")
reflects, on the part of those who conceived of it, a certain repre sentation of the user
and a strategy to make his use as natural as possible.
284
Acknowledgment
The authors wish to thank Anne-Marie Rifai for her translation of this chapter
from French.
Correspondance concerning this chapter should be addressed to M. Grossen,
Faculte des Sciences Sociales et Politiques, University of Lausanne, BFSH 2,
CH-1015 LAUSANNE, Switzerland.
285
References
Alter, N. (1985). La bureautique dans I'entreprise [Bureaucracy in busines s]. Paris: Les
Edition s Ouvrieres .
Appel , K. , & Haken, W. (1977). La solution du probleme des quatre couleurs [The
solution to the four-color problem]. Pour la Science, 2, 56-70.
Audetat, M. (1993) . Le voyageur Butor en zone de transit [The Butor voyager in the
transit zone]. L 'Hebdo, 4, 61.
Bangert-Drowns, R. L. (1993) . The word processor as an instructional tool: Ametaanalysis of word processing in writing instruction . Review of Educational Research,
63(1), 69-93 .
Belyaeva, AV., & Cole, M. (1989). Computer-mediated joint activity in the service of
human development: An overview . The Quarterly Newsletter of the Laboratory of
Comparative Human Cognitive, 11(3), 45-66.
Butor , M. (1992). Transit. Paris: Gallimard .
Cohen, H., & Nordon, C. (1989) . L'arithmetique assistee par la geometric et
l'ordinateur [Arithmetic assisted by geometry and the computer]. La Recherche,
20(208), 352 -358 .
286
MicheleGrossenand Luc-OlivierPochon
287
Chapter 12
Abstract
Structuralist models of cognition emphasize the role of reasoning principles,
invariants, or rules in the organization of thinking and pay little attention to the
culturally developed systems of signs and their impact on thinking. However,
complex psychological functions , as pointed out by Vygotsky (1962) and Luria
(1979), are always carried out with the mediation of historically developed and
culturally transmitted systems of signs . Within this perspective, systems of signs
have a major role in thinking because they are central to reasoning processes .
This chapter presents research that supports the hypothesis that systems of signs
playa structuring role in problem-solving activities as they are used to support
skilled action .
Two sorts of empirical findings are reviewed . The first line of work analyzes
how cultural practices that rely on diverse systems of signs affect their users '
performance as they solve mathematical problems. The second line of work is
experimental and examines how the use of different measurement systems affects
children's problem-solving performance. The results reported here suggest that
schools need to consider carefully the systems of signs that they select for
transmission in school, in terms of both the difficulties they create for learners
and their power as tools for thought.
289
290
Terezinha Nunes
291
The separation between the structure and content of thinking is helpful in many
ways. It is a powerful heuristic for the analysis of cognition in many
circumstances . It is particularly useful when subjects cannot articulate the
reasoning principles they are implicitly using (e.g., in the case of young children)
or when the very articulation of the principles may alter our observation (by
making the principles of reasoning into the object of thinking). It is also a
powerful way of demonstrating the similarity of competencies underlying
markedly different performances. Elsewhere (Carraher, 1985) I have argued, for
example, that Brazilian working-class and middle-class children in first and second
grade, who show markedly different error rates in spelling , appear to rely on the
same phonological principle to generate the orthography of words. Although they
use the same phonological principle to generate spelling, when they apply this
principle to their knowledge of the language (here, the content to which the
principle is applied), they come up with very different results . Middle-class and
working-class Brazilian children speak different varieties of the language, and the
middle-class children speak a dialect that is closer to the idealized and prestigious
form of Portuguese used as the source for orthographic representation. Thus,
middle-class children using the conception of phonological representation of words
generate fewer spelling errors than working-class children using the same
conception.
In his vast work, Piaget suggested that the logic of actions was the source of
these general structures of reasoning, whereas abstractions regarding objects
offered the knowledge of contents or empirical knowledge. The logic of actions is
not a question of particular events but of general representations of actions such as
grasping, adding, and separating. This hypothesis is perfectly consistent with the
notion of situated cognition. Its basic assumption is that subjects develop their
knowledge of situations and objects, on one hand, and their reasoning structures,
on the other, from their own organization of actions (or, more generally speaking,
their participation in situations). This hypothesis is also consistent with the
notion that knowledge may be both general and specific. When actions are
represented in their generalized form, as schemata, the specifics of the situation do
not constrain the knowledge of principles .
However, this hypothesis does not offer a complete description of knowledge. It
is necessary to examine the way in which knowledge is mediated by systems of
signs, because we cannot assume a direct interphase between the subject and the
objects of thought. The mediating systems of signs cannot be treated as byproducts of the subject's reasoning principles, because many of the mediating
systems of signs are not invented by the subject but are culturally transmitted.
Piaget (1962) argued that the child's ability to use collective systems of signs
still depends on the development of the child's thought, because children need to
construct meanings for the signs. Although this may be largely true, he failed to
acknowledge that, once the system is acquired, it also comes to structure
reasoning . Thus, a more complete description of the development of reasoning
would encompass a description of the uses of culturally transmitted systems of
292
Terezinha Nunes
signs . This is the core of the hypothesis that I explore here. More specifically, I
suggest that reasoning and problem solving are organized by two factors: the
generalizable principles related to the logic of situations and the organization of
the systems of signs used in the representation of particular situations. The latter
have sociocultural significance and not only make knowledge more situated but
also color it with connotations of affect and valorization.
In the next two sections, I review studies indicating that systems of signs play
a major role in reasoning. The first section reviews evidence from participation in
different types of activities that are characteristically associated with the use of
different systems of signs . The second section explores the effect of a single
experience with different ways of representing area and its consequence for
subsequent performance in solving problems about areas.
293
in an order are obtained in addition, regrouping needs to be done, and one unit
(meaning one ten, one hundred, etc.) will be carried over to the next column .
Typically, teachers refer to the rules learned for writing numbers as they
demonstrate the procedure on the chalkboard.
Fig. 12.1. The Place Value Frame (Quadro Valor de Lugar) for practicing written
numbers . The frame is composed of pockets labeled C (Centenas, meaning hundreds), D
(Dezenas, meaning tens), and U (Unidades, meaning ones) into which singles, bundles
of ten, or bundles of one hundred sticks are placed . When practicing how to write
numbers, children write the number of units in each pocket on a sheet designed in the
same way.
Although the addition facts need to be mentally produced by the child, I call these
algorithms written arithmetic because the written symbols become a medium onto
which the intermediary results of different steps are off loaded (see Hatano, in
press) . After instruction on the algorithm, children carry out large numbers of
computation exercises and word problem solving , the aim of which is to offer
further practice in the procedure. Multidigit subtraction is similarly taught with
reference to the previously learned principles of place value notation. The
borrowing procedure starts with the notion that "you can't take a larger number
from a smaller one," and thus you borrow one (one ten, one hundred) from your
neighbor. Once again, subtraction facts have to be produced by the children, and
the sequential results are off-loaded onto the written representation. As with
addition, extensive amounts of practice with the algorithm and in solving
multidigit subtraction problems follow instruction. Figure 2 exemplifies the
subtraction procedure and typical instruction provided.
Three significant characteristics of these algorithms are of interest here. First,
computation is carried out from right to left (that is, from smaller to larger). This
294
Terezinha Nunes
295
oral or written methods, thereby controlling for ability through randomization (for
a review, see Nunes , Schliemann, & Carraher , 1993).
A.
216
129
1.
Children are taught to align the digits from right to left (as shown
above) , often initially using notation sheets such as those presented in
Figure 1. Subtraction is also carried out from right to left.
2.
3.
For each column, the same procedure is repeated . Each digit is referred
to during the application of the algorithm as units; that is, the 3 in 236
is spoken of as three and not thirty, and the 2 is not referred to as two
hundred . This allows the children to use the same knowledge of
subtraction facts regardless of whether the operation is on the column
for units, tens, or hundreds .
B .
The problem presented to the child involved taking 57 away from 252 .
The child (Fabio, 3rd grader) first answered "185" after speaking very
low. When the examiner asked the child to explain it aloud, the child
said: "Two hundred and fifty two and fifty seven. I take fifty away, that
is two hundred . And two and seven. Five to take away, then, one
hundred and ninety five."
Note the direction of calculation (first, fifty is subtracted and then
seven) and the maintenance of the magnitude in focus during
calculation (the child does not say "five" but "fifty").
Fig. 12.2. An example of subtraction with borrowing and typically provided
instruction (top) and an example of oral subtraction.
296
Terezinha Nunes
Two within-subject studies were carried out in which the same children solved
problems in different situations that were likely to elicit different arithmetic
practices. In the first study, the children were interviewed first in street markets as
they carried out their sales and later on in a school-like situation, answering
questions about computation exercises and word problems (Carraher, Carraher, &
Schliemann , 1985). In the second study, the children were interviewed in schools
but were placed in a pretend-shop situation and in a school-like situation
(Carraher, Carraher, & Schliemann, 1987). Both studies revealed that children
performed significantly better in the vending situations (real or pretend) than in
the computation exercises.
Qualitative analyses of both oral and written procedures were also carried out in
the second study (Carraher, Carraher, & Schliemann , 1987). Children's
verbalizations during the interviews were tape-recorded, and, together with notes
taken by the interviewers, allowed for a detailed description of the steps used by
the children during calculation.
Written procedures, whether correct or incorrect, were attempts to reproduce the
algorithms taught in school. Errors observed in these procedures were of the same
nature described in the literature (see, for example, Brown & Burton, 1978;
Lawler, 1990). In subtraction, for example, some children took the rule "You
can't subtract the larger from the smaller" to imply "Subtract the smaller from the
larger." Thus, in calculating 252 - 57, some children took 2 away from 7 and
obtained the answer 205. In multiplication, for example, when calculating 15 x 5,
some children calculated 5 x 5 = 25, correctly wrote down 5 and carried the 2 but
then added 2 onto 1 before multiplying, calculating 3 x 5 = 15, and obtained the
result 155. Thus, even incorrectly used algorithms demonstrated children's
attempts to follow the rules that they had been taught but clearly had not
mastered.
As mentioned earlier, oral arithmetic is not taught systematically in school.
Thus, the qualitative analysis of these procedures involved the search for
underlying principles rather than a comparison between what had been taught and
what the children produced. In spite of great variations in the surface of the
problem-solving procedures, addition and subtraction problems seemed to be
described mostly by one method, referred to in the literature (e.g., Plunket, 1979)
as decomposition. Examples of decomposition in addition and subtraction are
presented in Table 1.
A careful consideration of these examples shows that decomposition implicitly
uses the property of associativity of addition and subtraction as a principle. In
other words, when children use decomposition, they seem to understand that it is
possible to divide into parts a number to be added or subtracted and to carry out
the addition/subtraction in steps, by subtracting each part sequentially. Errors in
oral addition and subtraction also revealed the same underlying principle, because
they were mostly memory errors. When a number was divided into more than two
parts, sometimes one of the parts was eventually forgotten, and the final response
was incorrect. For example, when subtracting 57 from 252, one child divided 252
297
into 250 and 2 and 57 into 50 and 7. He then subtracted 50 from 250, obtaining
200 and, subsequently, 7 was subtracted from 200. His final answer was 193
because he seemed to have forgotten the 2 that he had set aside.
Edilene (solving 243 - 75) : "You give me 200, I will give you 25. With the
43 that you have, 143, you will have 160, 168."
Carlos (solving 500 - 80):
298
Terezinha Nunes
BD
.oral calculation
D'M"ltten ca lculaIlon
III
CIl
0>
60
.sc:
...
CIl 4ll
U
CIl
11.
20
COrrect
Response type
BO
Oral calculation
D'M"'tten calculallon
60
III
CIl
0>
.sc
...u
CIl
4Q
CIl
11.
20
correct
Response type
Fig . 12.3 . Percentage of correct responses and of errors in the different error bands
for oral and written addition (top) and subtraction (bottom).
4):
299
300
Terezinha Nunes
301
E:
Carlos :
E:
Carlos :
2.
Finally, I consider an experimental study that examined the use of written versus
oral symbols in the solution of problems with negative numbers (Nunes, 1993).
The introduction of written directed numbers in Brazil involves a break with
previous written arithmetic practices taught in school. In the first four years of
school, the signs + and - (plus and minus) indicate the operation to be carried out,
and two operations with different signs cannot be carried out simultaneously. In
the representation of directed numbers from the 5th grade on, more than one sign
can be written in the same expression, and signs no longer represent the operation
to be carried out but the direction of the number. For example, -20 and -30 are
added, not subtracted, despite the fact that they are preceded by the minus sign.
These ambiguities in the use of written numerical representations seem to affect
students' ability to solve problems with directed numbers in the written mode.
302
Terezinba Nunes
However, the logic of the situation that suggests that two negatives should be
added and yield a larger negative value is understood by the subjects . When
students were randomly assigned to either a written or an oral solution group,
significant differences were observed between the groups. The students in the oral
solution group were able to use their understanding of the situation and performed
significantly better than the students in the written solution group . The ease of
solution in the oral mode contrasted with the difficulties in the written system in
a particularly clear way when subjects assigned to the written condition of testing
reexamined their answers through oral calculation (see Figure 4 for an example) .
Problem 5:
Antonio :
Problem 9:
Antonio :
Experimenter:
Antonio :
Note that, in Problem 9, the numbers were written down as the experimenter
had requested, but Antonio's solution was not worked out from the written
representation. In Problem 9, unlike Problem 5, the student did not try to use
the school form of representing negative numbers and succeeded.
Fig. 12.4. A written and an oral solution to similar directed numbers problems .
303
The attempts to follow the rules for manipulating written symbols while setting
meaning aside resulted in errors that would appear to indicate lack of
understanding. However, this apparent lack of understanding contrasts strongly
with the reasoning displayed by the same subjects when the problem was solved
in the oral mode.
Taken together, the results of these studies indicate that the subjects' knowledge
of reasoning principles and understanding of the situation is not sufficient to
account for their performance during problem solving. The systems of signs used
to solve problems have their own characteristics, which also affect the processes
involved in solving problems. The characteristics of manipulations of the written
system distance it from meaning, thereby making it difficult for subjects to
monitor results through meaning. Thus, subjects are more likely to make errors,
and the errors are of larger magnitudes. In contrast, oral arithmetic allows subjects
to monitor the magnitude of intermediary results during calculation, resulting in
higher rates of correct responses and errors of smaller magnitude.
It could be argued that all the subjects participating in the previous studies were
novices in their use of written arithmetic. Expert performance should show greater
flexibility and fewer errors. Intuitively, it seems that expert performance.is not
prone to error but shows little flexibility. After many years of schooling, we
make fewer errors in arithmetic but still use the same procedures when relying on
written symbols. This is only an intuitive suggestion with respect to written
arithmetic. However, evidence from studies with experts in the use of the abacus
indicates that flexibility does not come with large amounts of practice in the use
of abacus arithmetic. Hatano (in press) put forth this argument in a recent review
of studies on the performance of novices and experts in abacus arithmetic. The
influence of the abacus as a mediating system of signs on expert performance is
illustrated both by remarkable feats and by amazing rigidity in problem solving.
For example, experts in abacus arithmetic can accomplish results in number recall
and manipulation that are clearly beyond what novices can do. They can memorize
numbers composed of 15 or 16 digits, can say the digits forwards and backwards
with the same ease, can carry out sequences of operations more quickly than they
can write out the answers, and can calculate while answering simple questions.
These characteristics of their skill suggest that experts' numerical activities are
mediated by a visually represented mental abacus that is used following specific
procedural rules. Consistent with this hypothesis, visual stimuli were found to
interfere with experts' performance more than verbal stimuli did. However,
monitoring of the operation is suppressed in the development of this skill, just as
it is in written arithmetic. Thus, experts fail to notice that they have just added a
list of numbers when minimal changes are made on the list (such as having the
first number in one list be the last in a subsequent list) and add it all up again
when presented with the second list. They also do not rely on simplified related
number calculations when they solve large number problems (such as solving 99
x 38 by solving 100 x 38 and then subtracting 38).
304
Terezinha Nunes
The studies just described indicate how different sorts of culturally developed
signs affect calculation processes in particular ways. These studies, however , do
not show whether the use of different symbols to represent the same situation can
result in the development of different reasoning structures or conceptions of the
situation . To examine this second question, I turn to studies on the measurement
of area.
305
subjects in these studies had been taught in school the formula for the area of the
rectangle . Bricks and grid-lined paper are often used in introducing area work,
when children cover figures with bricks and count them in order to obtain a
measure for the area. The goal of instruction is to move the children from the use
of bricks to the use of rulers plus formulae in the measurement of area. However,
children typically show great difficulty in taking this step, even after carefully
designed instruction (see Dickson, 1989). Our goal in this series of studies was to
analyze the reasons for this difficulty in the change of measurement tools.
A parallel can be found between the change of tools in oral and written
arithmetic and the change of tools in measuring area with bricks and with a ruler.
The measurement with the support of a ruler seems very simple from the
viewpoint of the algorithm used. The obtained measures simply need to be
combined through a formula. However, the steps in the procedure are distanced
from meaning. Rulers provide linear measures, and their connection to the
measurement of area is by no means obvious. Bricks, on the other hand, can be
viewed as units of area. Thus, their relationship to area problems is much clearer.
The question for a subject measuring an area with bricks is "How many of these
bricks are needed to cover the figure?" No parallel question can be posed when the
child starts from a linear measure.
In the studies described next, we used both a choice paradigm (children had more
than one tool available, and we verified which one was chosen and how the
structure of solution was affected by the choice) and a preassignment paradigm
(children were randomly assigned to experimental conditions in which a different
tool was available) .
The task used in all three studies was the same. The children, always working
in pairs, were told that two friends had done some painting work for a neighbor
and each one painted a wall. When they received their money, the friends wanted
to figure out whether they had done the same amount of work but found it
difficult to decide because one wall was longer than the other but the second one
was taller. The children were asked to use the pictures of the walls to figure out
whether the two friends had done the same amount of work. A series of four
problems was used involving two simpler comparisons and two more difficult
comparisons. Feedback was provided between trials by covering one area with
previously cut pieces of colored paper, which were then moved and assembled over
the other figure. Figure 5 presents schematically the series of problems and
illustrates the feedback procedure.
306
Terezinha Nunes
_rn
Item I in the task involved comparing two rectangles (A) with the same area
but different dimensions. The feedback procedure involved covering each in
sequence with the same bits of pre-cut colored paper as shown in (B) .
/
Each figure in C was used for comparison with a rectangle in the more difficult
items .
Fig. 12.5. Two rectangular areas used in the easier comparison of area
problems and a schema of the feedback procedure (top); two other figures used in
the difficult problems, where a parallelogram and a U-shaped figure were each
compared with a rectangle (bottom) .
A choice paradigm was used in the first study (for further detail, see Nunes, Light,
& Mason, 1993; the description presented here is simplified). Thus, children had
access both to rulers and to bricks during problem solving and could resort to
whichever they preferred. Bricks were numerically insufftcient to cover any of the
areas to be measured in order to avoid simple counting procedures to solve the
problem. We anticipated that the analysis of problem-solving procedures would
307
indicate whether the same underlying conception of the problem was reflected in
the children's solutions, regardless of the tool used or whether different conceptions would emerge .
An initial quantitative analysis showed that children who chose the ruler
performed significantly poorer than those who chose the bricks. Approximately
half of the children who initially chose the ruler as the measuring tool, after
measuring the sides and calculating the perimeter, realized that they had not
obtained a good evaluation of the work involved in painting the walls . They then
moved to the use of bricks. Those children who started with rulers and did not
change to bricks calculated either the perimeter or half of the perimeter instead of
calculating the area. This type of error, which is well documented in the literature
(see, for example, Dickson, 1989; Douady & Perrin-Glorian, 1989), is of
ambiguous interpretation. It has been suggested that this error reflects the
children's failure to differentiate the concept of perimeter from that of area
(Douady & Perrin-Glorian, 1989), but it can also be argued that the multiplication
of centimeters by centimeters makes no sense to children, whereas addition is
meaningful, and thus children choose to do what makes sense to them.
Brick users, in contrast, were much more likely to solve the problem correctly.
They tended to discover that the number of bricks in a row in a rectangle is fixed
and then to come up with the formula "number of bricks in a row times number
of rows." This formula corresponds to a model of multiplication that Vergnaud
(1983) has termed isomorphism of measures and can be described as involving a
one-to-many correspondence scheme (that is, if one row corresponds to 8 bricks, x
rows correspond to x times 8 bricks). In contrast, the ruler-plus formula approach
to area corresponds to Vergnaud's product of measures model, where one measure
multiplied by another yields a third measure (i.e., centimeters multiplied by
centimeters yield square centimeters, a new measure).
These results indicate that the tool provided to children for solving area
problems plays a structuring role in the conception of area that they develop in
the course of problem solving. Bricks, which are area units, are more clearly
connected to the problem than linear measures obtained in centimeters. Thus , the
use of bricks as tools for measuring area allows children to use the one-to-many
correspondence scheme that they often use in multiplication problems, whereas
the use of rulers requires a different conception of multiplication.
In a second study, a pre-assignment paradigm was used to avoid possible
confounding of the results by self-selection. The same task was used, but this
time the pairs of children were assigned to either a ruler or a brick availability
condition (Nunes, Light, Mason, & Allerton, 1994). In an attempt to discourage
the most frequent errors observed in the previous study, extra feedback was
provided during problem solving to children who calculated either the perimeter or
half of the perimeter. The children were told that they had correctly calculated
"how much it was all around the walls," and then they were asked whether they
knew how much work it was to paint the whole wall.
308
Terezinha Nunes
Children in the bricks condition were, as in the previous study , more successful
in solving the problems than those in the ruler condition. In spite of our attempts
to discourage addition of the measures obtained for the sides, children in the ruler
condition still preferred this solution. Children in the bricks condition, as in the
previous study, tended to discover the regularity of number of bricks in a row in
the rectangle and to come up with the formula "number of bricks in a row times
number of rows."
All children were seen again individually about one month later and were asked
to calculate the area of some figures . In this second session, both bricks and rulers
were available. Approximately two thirds of the children chose to work with the
same tool with which they had worked in the previous session. The children who
had worked with bricks in the first session still showed significantly better
performance than those who had worked with rulers .
In the third study, we first offered the children an equal amount of instruction in
the calculation of the area of a rectangle in either a brick or a ruler condition.
Immediately after instruction, they solved another problem that involved the
comparison of rectangular areas. In this immediate posttest, no difference was
observed between the two groups. However, in the next trials, when the children
were asked to solve problems that involved comparing a rectangular area with the
area of new figures (a parallelogram and a U-shaped figure that could be
decomposed into rectangles), the children in the bricks condition were much better
at adapting the procedure that they had learned. The formula that they had come up
with, number of bricks in a row times number of rows, allowed them to maintain
the meaning of area in focus, and they were, consequently, in a better position to
adapt their knowledge to new figures than those who had worked with rulers. Area
units symbolize area in a much clearer fashion than linear units do.
This series of studies indicates that the system of signs that the children were
offered in the learning phase influenced the reasoning principles that they used in
organizing the problem solution. The structuring of the children's action was not
independent from the tool they had at their disposal in the problem-solving
situation. Consequently, the very reasoning principles developed in the situation
were structured by the system of representation used.
Conclusion
The studies reviewed here have both theoretical and practical implications. There
is clearly a need to reexamine the concept of ability, which was shown to be
determined not solely from within (neither by talent nor by developmental level)
but also to be significantly affected by the systems of symbols used by the
subject. Different systems of symbols pose different types of difficulties for
children. We have seen how oral and written arithmetic differ and how they
309
influence children's reasoning processes as they are used as tools for thought. In
this context, it is also necessary to consider what kind of power the different tools
for thought offer to users . Within the range of numbers used in the studies
described here, oral arithmetic was powerful enough, but memory errors did crop
up . In the realm of larger numbers, oral arithmetic would probably have been a
poor choice of tool. Similarly, calculating area with the formula "number of
bricks in a row times number of rows" is a good initial approach to area
problems. The use of area units allows the subjects to keep the meaning in focus
and to perform sensible calculations. However, area units are not practical in
everyday life, where it is much simpler to measure figures linearly and to calculate
the area using a product of measures approach. Thus representational systems pose
different types of difficulties for their learners and also afford different power as
tools.
The educational implications of this analysis are also significant. Children
invest a substantial amount of effort in the acquisition of culturally developed
systems of signs such as oral and written numeration systems, measurement
systems, mathematical notations, and today, forms of inputting commands into
computers. These systems are used both for communication with others and for
personal purposes during problem solving . If symbolic systems are an integral
part of the intellectual process, schools , as privileged institutions for the
transmission of many cultural practices, need to reconsider their approach to
teaching . Greater attention must be given to how systems of signs structure
thinking and what systems are to be chosen as basic for everyone .
310
TerezinhaNunes
References
Brown, J. S., & Burton , R. R. (1978). Diagnostic models for procedural bugs in basic
mathematical skills . Cognit ive Science , 2, 155-192.
Carraher, T. N. (1985). Exploracoes sobre 0 desenvo1vimento da competencia
ortografica em Portugue s. [Exploring the development of spelling in Portuguese]
Psicologia: Teoria e Pesquisa, I , 269-285.
Carraher, T. N. (1986). From drawings to building s: Working with mathematical
scales. International Journal of Behavioral Development, 9, 527-544 .
Carraher, T. N., Carraher, D. W., & Schliemann, A D. (1985). Mathematics in the
streets and in school s. British Journal of Developmental Psychology, 3, 21-29 .
Carraher, T. N., Carraher, D. W., & Schliemann, A. D. (1987). Written and oral
mathematics . Journal for Research in Mathematics Education, 18, 83-97.
Dickson, L. (1989) . Area of a rectangle . In D. C. Johnson (Ed.), Children 's
mathematical frameworks 8-13: A study of classroom teaching (pp. 89-125).
Windsor, England : NFER-Nel son Publishing Company.
Douady, R., & Perrin-Glorian, M. J. (1989 ). Un processus d' apprenti ssage du concept
d'a ire de surface plaine [A teaching experience on the concept of the area of a plane
surface]. Educational Studies in Mathematics, 20, 387-424.
Gardner, H. (1993 ). Multiple intelligences: The theory in practice. New York: B asic
Bo ok s.
Gay, J., & Cole, M. (1967). The new mathematics and an old culture: A study of
learning among the Kpelle of Liberia. New York: Holt, Rinehart & Winston.
Grando, N. (1988): A matematica no meio rural: Pequenos agricultores, estudantes e
professores [Mathematics in the rural area: Farmers, students, and teachers] . M.A.
Thesis, Mestrado em Psicologia, Universidade Federal de Pernambuco.
Hatano, G. (in press). Learnin g arithmetic with an abacus. In T. Nunes & P. E. Br yant
(Eds.), Teaching and learning mathematics: International perspectives. Falmer,
England: Lawrence Erlbaum .
Hughe s, M. (1983). What is difficult about learning arithmetic ? In M. Donaldson, R .
Grieve, & C. Pratt (Eds.), Early childhood development and education (pp. 204-221).
Oxford, England: Blackwell.
Lave, J. (1988). Cognition in practice. Mind, mathematics and culture in everyday life.
Cambri dge, England: Cambridg e Univer sity Press .
Lawler, R. W. (1990) . Constructing knowledge from interactions. In L. P. Steffe & T.
Wood (Eds .), Transforming children's mathemati cs education. International
perspe ctives (pp. 47-51). Hillsdale, NJ: Lawrence Erlbaum .
Light, P. & Perret-Clermont, A-N. (1989) Social context effects in learning and
testin g. In A Gellatly, D. Rogers, and J. Slobod a (Eds.), Cognition and social
worlds (pp 99-112) . OXford, England: Clarendon Press .
Luria, A (1979). Curso de Psicologia Geral [Course of General Psychol ogy]. Rio de
Janeiro: Civilizacao Brasileira.
Nunes, T. (1993). Learning mathematics: Perspectives from everyday life. In R. B .
Davis & C. A. Maher (Eds.), Schools, mathematics, and the world of reality (pp .
61-78) . Needham Heights, MA: Allyn and Bacon.
311
Chapter 13
In this chapter, I investigate how children understand concepts and the meaning of
symbolic representations when working with conceptual models. The particular
focus of this study is the nature of children's conversation when collaborati ng to
solve tasks embodied in what we call an intermediate abstraction. The analysis is
based on theoretical considerations developed by Greeno and his colleagues
(Greeno, 1989; Greeno, Engle, Kerr, & Moore, 1993) and on a line of research
carried out by Resnick and Schwarz (Schwarz, Kohn, & Resnick, 1993; Schwarz
& Nathan, 1993; Schwarz, Nathan , & Resnick, 1996; Schwarz, in press) with
several computerized systems . My approach focuses theoretical attention on the
nature of conversational turn-taking between children working with an
intermediate abstraction of mathematical operators. This study shows that
conversational processes of acceptance stemmed from collaborative learning , and
that those processes enabled the construction of shared meaning resulting in the
development of the concept of a mathematical operator. Three such processes were
found. The first one is a process of creation of a shared meaning through reference
to objects of the system . This process is initiated by one participant describing a
property of the intermediate abstraction. The second participant then accepts the
meaning uttered by the first participant and refines it. The second one is a process
of convergence of a different nature that is initiated by an impasse within one
particular representation and is characterized by a change of representation. The
third turn-taki ng occurrence is a process of confirmation through a different
representatio n. At the end of this chapter, I show how individuals who interacted
313
314
Baruch B. Schwarz
from the literal features of the world, in relation to (b) the interplay of physical
metaphors, through the constructive use of (c) iterative cycles of conversational
turn-taking, constrained by (d) the application of higher standards of evidence for
convergence. Therefore, according to Roschelle (1992), intermediate abstractions
trigger conceptual change because they create favorable ground for developing
metaphors, and because students collaborating together can gradually construct
shared meaning (what Roschelle calls convergence).
The approach adopted by Greeno et al, (1993) is slightly different. It recognizes
both interactional and cognitive processes as contributing to conceptual change in
collaborative work. Greeno and his colleagues focused on the "winch system" to
acquire symbolic representations. In that system, the labeled size of each spool is
equal to its circumference in inches, which equals the distance that a block moves
each time the handle is turned with the spool. This means that saying or writing a
numeral can be interpreted as referring to the size of a spool, a distance per turn,
or both . Further, if a phrase is used that specifies one of these (e.g., the four
spool), its referent can be extended to include the distance per turn of a block with
that spool if the speaker and the listener are attuned to the constraint that these
two quantities are equal. Greeno et al. (1993) generalize that learning symbols is
an activity that involves attunement to constraints and affordances about (a) the
domain that is referred to, (b) the domain of notational expressions, and (c) the
meaning relations between notations and their referents. In the referent domain,
attunement to constraints makes it possible to refer to co-constrained properties of
a situation. Constraints of the notations can be considered the syntax of the
representational system. For example, in equations, there are constraints on the
sequence of numerals, letters, and operations signs that the student learns , and
there is a simple grammar that determines a phrase structure for any algebraic
expression. In graphs, there are constraints, such as the continuity of the line that
is considered as an object, and affordances, such as considering any segment of a
line as an object.
Greeno's framework about understanding as attunement to constraints and
affordances is adopted in this study. I examine the kinds of relations between
symbolic and nonsymbolic representations developed in collaborative work with
an intermediate abstraction. The two components considered are to which degree:
(a) different symbols, concepts, or interpretations were differentiated from each
other and (b) the relations between them were specified systematically. I also
adopt the theoretical ideas given by Clark and Wilkes-Gibbs (1986) and Clark and
Schaeffer (1989) about conversational processes. In particular, referring is
considered a collaborative process, with the speakers trying to establish, by the
initiation of each new contribution, the mutual belief that the listeners have
understood what the speaker meant in the last utterance (the principle of mutual
responsibility).
315
The Merry Meadows Farm sells 12 crates of peaches to the local grocery store. To
deliver the peaches , the store has to rent a truck for $36 . Altogether, it costs the
store $210 to get the peaches and bring them to the store. How much does the
Merry Meadows Farm ask for each crate?
316
Baruch B. Schwarz
The equation is 12*x + 36 = 210. For the equation, the symbolic expression is
the formula itself. For the word problem, it is the written text (i.e ., the words,
nouns, verbs, prepositions, and adjectives included in the formulation). The text
here refers to its objects (e.g., the Merry Meadows Farm, the peaches, the truck)
and to all the events (e.g., renting a truck, bringing the peaches to the store)
described in it. The equation 12*x + 36 = 210 can refer to the quantities in the
word problem. It can also relate to abstract entities: for example, the function f(x)
= 12*x + 36. In general, symbolic expressions include written or spoken
instructions, and they refer to quantities in a concrete situation or to abstract
entities. The two symbolic systems can undergo transformations. For example,
the sentence, "The Merry Meadows Farm sells 12 crates of peaches to the local
grocery store," can be transformed into "The local grocery store has been sold 12
crates of peaches from the Merry Meadows Farm." Similarly, the equation can be
transformed into 12*x = 120 - 36 or into 12*(x + 3) = 120. Therefore, when
solving algebraic word problems, algebraic symbols refer both to mathematical
entities and to objects and events of the word problem. For example, 12*x + 36
can refer either to the function f(x) = 12*x + 36 or to the price the store pays for
having the 12 crates. In general, just a part of the operations that we perform on
symbolic expressions preserves the references of their terms in the domain in
which they appear. For example, the transformation of the equation 12*x + 36 =
120 into 12*x = 120 - 36 is consistent with the transformation of the situation
into "paying for the 12 crates without the shipment costs $120 - $36." The
transformation of the equation into 12*(x + 3) = 120 is possible because 12*(x +
3) and 12*x + 36 refer to the same function. In contrast, this transformation has
no easy mapping with a transformation of the word problem situation (something
such as "Because the overall shipment costs $36, the shipment for each crate costs
$3"). Therefore, it is not surprising that students have difficulties reasoning
algebraically (i.e., to be able to use algebraic entities) to solve word problems.
The constraints of the two symbolic systems are not considered as a coherent
whole. Subjects able to conduct formal computations correctly were often totally
unable to apply their skills in real world situations; rather, they chose ad hoc
methods triggered by the situations. Such a phenomenon stresses what Greeno
(1989) calls a "perverse semantic interpretation that maps the symbolic structures
back to the notations used to express them, rather than to a domain of real objects
and situations and abstract mathematical entities" (p. 294). This interpretation
implies that problems of translation between the two symbolic representational
systems are very difficult. For example, Clement (1982) showed that students in
various grades generally failed to solve the following problem, referred to as The
Students and the Professors:
Write an equation using the variables Sand P to represent the following statement:
There are six times as many students as professors at this university. Use S for the
number of students and P for the number of professors .
317
Clement showed that students often write the relation 6S = P (instead of the
correct relation S = 6P), because letters are not perceived as representing the
number of students, and the equal sign is not used to express an equivalence but
rather a comparison or an association. In this example, the correspondence is
syntactic , not semantic. A similar phenomenon was found by Schoenfeld (1987)
in geometry where students' knowledge was linked to "drawings" and not to
spatial relations and constraints. In contrast, it was found that intermediate
abstractions allowed students to reason about mathematical and scientific
concepts. For example, students collaborating to solve problems on the Planner
system could model and solve difficult arithmetic word problems (Schwarz, Kohn,
& Resnick, 1993; Schwarz, Nathan, & Resnick, 1996) and could reason about
negatives with different referents (e.g., temperatures, chronology, elevators). This
is because children could develop links between the actions they operated on the
objects of the system and the symbols, links that were based on affordances and
constraints of each of the representations. The symbolic-descriptive expressions
were branched to demonstrative representations, that is, as objects with properties
and relations that correspond to properties and relations of objects in the situation
or domain that the expression represents. The study described in this chapter
defines the learning situation created by the use of the system OPERA, an
intermediate abstraction for the acquisition of the concept of mathematical
operator. In particular , it traces the contribution of collaborative learning and the
processes that characterize it.
Baruch B. Schwarz
318
(a)
(b)
(e)
(d)
There are also what we have called teams of machines, in which operators consist
of more than one element. Machines operate on trains, the trains having lengths
that are indicated by a "flag." A machine operates on trains by putting them into
the entry portal of the machine and "clicking" on it. Figure 2 shows how the team
[+2] [*3] runs on a +2 train. The "out-train" is a +12 train. Clicking on a train
when it is at the exit portal will run it "backwards," producing a +2 train at the
entry portal. This representation shows operators as numerical processors, that is,
the number/train is processed by operation(s)/machine(s). The trains and the
machines do not display literal features of real trains but are metaphors that
express an essential common function of trains and operators: the fact that there is
a "vehicle" that undergoes a series of transformations .
......~~~~
. . . .~ Q . .<J.. ~
........
319
......
Fig. 13.3. Flipping the [+3] machine and the [+2][*3] team
..
0.1
1
2
I
4
4
7
10
II
320
BaruchB. Schwarz
JI:=='-
2
~
Ji:=:==l._-JL
Sa
J.!:=::==.L
. __
....J:::=~_
5b
J.!::==!JL
5c
~.
.=::=JL
5d
Fig. 13.5. The stack representations for the operators [+2] (5a), [*3] (5b), [+2][*3]
[5c], and [*3][+2] (5d)
The four representations alluded to thus far are symbolic. Intermediate abstractions
always contain components from which intuitions can stem. The stack
representation plays this role. It is not symbolic, showing the changes that
quantities undergo with operators in a dynamic visual way. Figure Sa shows a
stack representation of the operator [+2] as a process. It begins (as always) with
the creation of a generic block; then, a block labeled +2 is chained on its right
321
side. The operator [*3J is shown in Figure 5b as the gradual construction of a 3high stack. Figure 5c shows the creation of the compound operator [+2J [*3].
After the creation of the stack for the [+2] operator, it is expanded as a whole to a
stack with three generic blocks and three blocks labeled +2. Switching the [*3]
and [+2] totally changes the representation, as shown also in Figure 5d.
These examples show that the stack representation makes explicit the structure
of algebraic expressions as the formation of structures. The critical aspects of
those structures are unambiguous (in contrast with the traditional algebraic
syntax). Multiplication is an expansion in the vertical direction; addition and
subtraction are translated as putting blocks on the right side (in the horizontal
direction). The importance of order for operations is made explicit by the very
different stacks that result. Of course, this representation provides the child with a
set of constraints (such as the horizontal and vertical expansions for addition and
multiplication). These constraints are often arbitrary, although easily explainable.
In conclusion, OPERA provides a system of representations that express
different aspects of operators. The tables convey the idea of mapping; the
machines express operators as numerical processors; the stacks are an embodiment
of algebraic structures. All the representations can be displayed and run in parallel.
They provide co-constrained properties in a symbolic and nonsymbolic way.
Baruch B. Schwarz
322
the team of machines leaves them invariant. For example, for a 2 in-train, the
out-train is also a 2 train. To construct a team matching the black box, the
student can first disconnect it from the team [*2] [-1] and then flip this team. But,
of course, many other strategies based on the various representations available in
OPERA can also be applied.
Inlnln1
Inu.n . ]
(6 ' 2) - 1)-7
.,
In
Oul
.'
3
323
324
Baruch B. Schwarz
posttest, students were given clues displaying the three representations in OPERA
(see Figure 8).
CuIouboi
and nu.ben
..
I i
---j---
EJ
c. Slack Folders
The machines were given as two cut-outs: one in the form of a machine, the
other in the form of a */ machine. The students had to choose the right machine
or team and to instantiate it by means of number tiles. The table was a simple
double array that the students had to fill in. The stacks were sorted in six folders:
For example, the second folder contained the piles of 2, 3, 4 . . . generic blocks;
the third folder contained the stacks consisting of piles of 2, 3, 4 . . . generic
blocks, to which one block was added to the right side; the fourth one contained
two piles of equal height (see Figure 9). After reading a story (unsuccessfully
solved), the student was invited to try to solve the problem by using one of the
three sets of clues.
325
Slact.. fo'det .3
1 Exp :
2 Exp :
3
4 Leah :
5 Exp :
6 Leah :
7
8
9 Exp .
10 Leah :
11
12 Exp :
13 Leah:
14
15 Exp :
16 Leah:
17
18 Exp :
19 Leah :
326
Baruch B. Schwarz
In this short example , a very common phenomenon is illustrated: The students are
able to recognize the story of a problem through a stack or a team of machines;
and then, with the representation they choose, they are able to solve the problem.
I
I
.,
,
I
I
I
327
It is striking that, although Ron and Iris were not familiar with the system, they
were quickly able to develop metaphoric explanations based on their intuitive
understanding of how the objects of OPERA behave . First, Ron identified the first
stage of the stack representation as the addition of 2 (line 1). Iris identifies the
expansion of the two blocks as adding "two more of the stuff' (line 3). Then Ron
modifies her interpretation: "it gets them to the third floor" (lines 7 and 8).
Although Iris accepts Ron's interpretation, she refines it by using the term
stretches (line 9). All these terms constitute a coherent set of explanations based
on metaphors (such as "stretches," "two more of the stuff," or "it gets them to the
third floor"). Because OPERA was designed to represent all the aspects of the
concept of mathematical operator, or what Roschelle (1992) has called the deep
structure of a concept, developing a coherent set of metaphorical explanations of
the behavior of the objects of OPERA implies developing an intuitive
understanding of the concept.
Another very salient process in this excerpt occurs when the participants take
turns in their conversation. That is the process of convergence to a shared
meaning by referring to objects of the system. For example, in line 2, Iris adds
"to it" to the explanation "it adds the 2" given by Ron . Her remark refines the
meaning of "It adds 2" in a way that is essential: In a mathematical language, an
operator such as +2 needs to be applied on numbers. This convergence of meaning
occurred here by referring to an object of the OPERA system. Similarly, in line
9, Iris uses the term stretches to qualify the meaning of "it gets them to the third
floor" uttered by Ron. At last, Ron specifies "how much" the stack stretches.
Again, in this turn-taking sequence, the second participant helps to make the
shared meaning converge. In summary, convergence is a turn-taking process in
which (a) one participant describes a property of the intermediate abstraction, (b)
the second participant accepts the meaning uttered by the first participant, and (c)
the second participant refines this meaning. Convergence often occurs when the
objects to which the participants refer are visible.
Inverse 3 was given much later than In-train 7. Two different solutions of
Inverse 3 given by two pairs are presented here. These solutions uncover different
processes of collaborative learning, processes that both stem from convergence.
The first pair, Amos and Leah, uses the representations of OPERA extensively .
328
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Baruch B. Schwarz
Amos : If I put a 2 train in t he entry [Amos creates a 2
train, puts it at the portal entry of t he t eam [*2 ] [1]
[bla::k l:ax] , am cliCks en this tE:am] , i t give:; .
two. And if I put a 3 train , it gives a thre e .
So, these machines do not change. .
[Leah takes the mouse and creates another t rain, a 5 train,
and runs it through the team]
Leah : Yeah! It does not change the trains (pause). The two
is multiplied by two, it's four , minus one, i t 's
three . So the three in the black box gives, hmm.
Amos : I do a table for all the mach i ne s together and also
for the black box separately. This one is easy , yo u
know . [Amos enters the x-values 1, 2, 3, and 4 and the
s ame y-values ]. The b l a ck box . . .
[Leah points out on t he t eam [ *2] [-1 ] [b l a ck box] and on the
first en try of t he table o f t he bla ck box (s e e Figu r e 12)]
Leah: (whispers) One by two minus one, it's still one . . .
(pause)
Amos : Yeah! It does not change . .
Leah : But for three, uh , for two . . . If two enters here
[Leah points on the t eam of machines ] it gets to four
minus one , it's three , and through the black box, it.
Amos : must get out two .
Leah: hmm!
[Leah enters 3 i n t he x -entry and 2 in the y-entry for the
table of the black box]
(Leah and Amos continue to fill in the black box table; they obtain 1,3, 5, and 7
for the x-values and 1,2,3, and 4 for the y-values).
19
20
21
22
23
24
25
26
27
28
29
30
31
32
329
Inuene 3
..
In
2
3
4
Fig. 13.11. Leah tries to construct a table for the black box
330
Baruch B. Schwarz
to one block." The referent here is, of course, the stack representation . Then Leah
confirms Amos's assertion by saying, "So, I divide it by two." It is not clear to
which representation Leah refers when using the term divide (the machine "divide
by two" or the algebraic representation). What is certain, however, is that it is
symbolic in contrast to Amos's reference to the stacks. Another confirmation
done through a second representation occurs in line 28. It is Amos's reaction to
Leah in which he tries to confirm the "divide by two" by going to the table.
During most of the sessions, Leah's and Amos's discourse was governed mainly
by convergence turn-taking, with the language used by the pair often being
metaphorical . In contrast, the other pair, Ron and Iris, moved from convergence
turn-taking to confirmations through a different representation. For example, the
interaction between Ron and Iris while solving "Inverse 3" is shown in the
following excerpt:
1 Ron: I try some trains. One gives . . . one . And a two gives
two
[Iris clicks on the 2 train backward; It yields a 2 train]
2 Iris: It's the same in the two directions. If I rotate the whole
3
stuff, it will be the same .
[Iris rotates the team of three machines and checks that the 1
and 2 trains yield the same trains . She rotates the team again
to its initial position .]
4 Ron: Yeah the two do the reverse, they annihilate each other .
5
You know, I'll cut the black box from the two others . The
6
table for the black box is . . .
[Ron enters the 0, 1, 2, and 3 x-values through the black box
and obtains the y-values 1/2, 1, 1 1/2, and 2.]
7 Ron : Let's see. If I put half in the table of the two machines,
8
I get zero . Yeah . And three gives two . It's the reverse,
9
and the black box is the rotation of the minus one and / 2 .
10 Iris: I'll try it. (Iris flips the [*2] [-1] team of machines to
11
obtain [+1] [/2]) . If I attach this to a [*2] [-1], it will
12
behave like the black box .
[Iris creates the [*2] [-1] team and attaches it to [+1] [/2] .
She then enters a 2 train and a 3 train ; they go out
unchanged. ]
13 Ron : Yeah, they do exactly the same, but in the opposite
14
direction . They kill each other . . . . I f I see that with the
15
stacks, it will return to the same thing.
[Ron selects the [*2] [-1] [+1] [/2] team and runs its stack
representation. While the running occurs, Iris gives the
following comments:]
16 Iris: OK, we get the *2, and minus one . Now the plus one eats the
17
minus one. And the /2 reduces it again to the first brick .
331
332
Baruch B. Schwarz
Exp :
Ron:
Ron first goes wrong in choosing the multiplication machine but corrects himself
immediately by flipping it and saying, "It's the opposite; it will be a division by
6 machine." Ron uses the same object but changes its position to modify his
answer. In fact, Ron knows that the formal relations "multiply by six" and "divide
by six" are essentiality the same, depending on how the relation between the
professors and the students is put. The machines are not a way to facilitate
matching between a symbolic and a nonsymbolic system, but they are used in a
flexible way to express the relation .
Discussion
I have analyzed the construction and interpretation of symbolic representations in
terms of attunements to constraints and affordances. Learning a representational
system involves more than acquiring referential correspondences between symbols
and their referents . It requires knowledge about how to reason about the
represented system, how to construct and use symbolic expressions, and how to
create correspondences between symbols and their referents. However, the
achievement of coordination among all the representations is difficult: Multi-
333
334
Baruch B. Schwarz
References
Clark, H. H., & Schaefer, E. F. (1989) . Contributing to discourse . Cognitive Science,
13, 259-294.
Clark, H. H., & Wilkes-Gibbs, D. (1986) . Referring as a collaborative process .
Cognition , 22, 1-39.
Clement, J. (1982) . Algebra word problem solutions: Thought processes underlying a
common misconception. Journal for Research in Mathematical Education, 13, 1630 .
Greeno, J. G. (1989) . Situation models, mental models, and generative knowledge. In
D. Klahr & K. Kotovsky (Eds.), Complex information processing : The impact of
Herbert Simon (pp. 285-318). Hillsdale, NJ: Erlbaum.
Greeno, J. G., Engle, R. A., Kerr, L. K., & Moore, J. L. (1993 , July). Understanding
symbols: Situativity-theory analysis of constructing mathematical meaning. In
Proceedings of the Fifteenth Annual Meeting of the Cognitive Science Society (pp .
504-509). Boulder, CO.
Kirshner, D. (1989) . Critical issues in current representation system theory . In S .
Wagner & C. Kieran (Eds.), Research issues in the learning and teaching of algebra
(pp. 195-198) . Hillsdale, NJ: Lawrence Erlbaum Associates.
Larkin , J. L. (1989) . Robust performance in algebra : The role of the problem
representation. In S. Wagner & C. Kieran (Eds.), Research issues in the learning and
teaching of algebra (pp. 120-134). Hillsdale, NJ: Lawrence Erlbaum Associates.
Matz, M. (1980). Towards a computational theory of algebraic competence. Journal of
Mathematical Behavior, 3, 93-166.
Roschelle, J. (1992) . Learning by collaboration: Convergent conceptual change . The
Journal of the Learning Sciences, 3, 235-276 .
Schoenfeld, A. H. (1987) . When good teaching leads to bad results : The disaster of
"well taught" mathematics courses . Educational Psychologist, 23, 145-166 .
Schwarz, B. B. (in press) . Why can intermediate abstractions help acquiring robust
representations? Interactive Learning Environments.
Schwarz, B. B., Kohn, A. S., & Resnick, L. B. (1993). Positives about negatives . The
Journal of the Learning Sciences, 3(1), 37-92 .
Schwarz, B. B., & Nathan, M. J. (1993, July) . Assessing conceptual understanding of
arithmetic structure and language. In Proceedings of the Fifteenth Annual Meeting of
the Cognitive Science Society (pp. 912-917) . Boulder, CO.
Schwarz, B. B., Nathan, M. J., & Resnick, L. B. (1996). Acquisition of meaning for
arithmetical structures with the Planner. In S. Vosniadou, E. De Corte, R. Glaser, &
H, Mandl (Eds.), International perspectives on the construction of technology-based
learning environments (pp. 61-80). Mahwah, NJ: Erlbaum.
Smith, C., Snir, J., & Grosslight, L. (1992). Using conceptual models to facilitate
conceptual change : The case of weight-density differentiation . Cognition and
Instruction, 9, 221-283 .
White, B. Y. (1993) . Thinkertools: Causal models, conceptual change, and science
education. Cognition and Instruction, 10, 1-100.
335
Chapter 14
Abstract
A central issue in learning from hypertext is the nonlinearity of presenting
information. Nonlinearity, however, does not have a facilitative effect on
learning per se. The essential point is rather how learners make use of this
nonlinearity, that is, how they interact with hypertexts, which strategies they
use, and what kind of knowledge they construct. This chapter is concerned with
differences between successful and unsuccessful strategies in learning about a
complex subject matter from hypertext. In an empirical study, subjects had to
answer a series of questions with the help of a hypertext. Cognitive processing
was analyzed through the think-aloud method . A group of successful learners and a
group of unsuccessful learners were identified by their scores in a comprehension
test presented after learning . The results indicated that the two groups used
different strategies, which were reflected by different patterns of information
access. The successful learners tried to construct a mental model of the subject
domain, which was then successively elaborated . They combined model
construction, model analysis, and symbol processing in a flexible way, and tried
to solve inconsistencies by elaborating the respective model structure . In
contrast, unsuccessful learners tried to get by with a symbol processing strategy
and tackled the construction of a mental model only when the former strategy
could no longer be applied. Inconsistencies were solved by ad hoc repairs of
procedures rather than by systematic elaboration of the model structure. The
respective strategies can be related to a global and a local orientation of
knowledge acquisition. Instructional implications for the selecting and
337
Introduction
Knowledge acquisition is generally perceived today as an active process of
constructing complex mental representations. Learners who have become experts
in a specific field have acquired highly integrated knowledge structures of the
respective domain (Chi, Glaser, & Rees, 1982). Educationalists, however, usually
complain about the fact that learners only accumulate isolated pieces of knowledge
that are not integrated into a coherent whole (cf. Carey, 1986; diSessa, 1983;
Vosniadou & Brewer, 1990). Various reasons for this lack of mental integration
include the method of presenting information and the media used for this purpose .
The assumption is that traditional learning materials such as written texts do not
adequately take into account the active and constructive nature of human
knowledge acquisition, because learners are only in a passive role of receiving new
information. Furthermore, the linear format of such texts is said to be inadequate
for presenting complex subject matters, because the respective network-like
structure has to be transferred into a linear structure during writing by the author
and subsequently must be retransferred into a network during comprehension by
the learner. Due to these processes of linearization and delinearization, learners
frequently would not recognize the complexity of the content (Spiro, Feltovich,
Jacobson, & Coulson, 1991).
Today, the development of new technologies provides new possibilities for
presenting information in a way that is assumed to be better adapted to the active
and constructive nature of human learning and that allows the explicit display of
the network-like structure of knowledge. One kind of technology receiving special
attention in this context is the so-called hypermedia. Hypermedia are electronic
documents with a nonlinear, network-like structure consisting of nodes and links.
Nodes have the function of information containers, wherein texts, static or
animated pictures and graphics, sound, and simulations can be stored. If they
contain only text, the document is refered to as a hypertext. Links have the
function of relating the information of a node to the information of other nodes
(Gloor & Streitz, 1990; Jonassen & Mandl, 1990; McAleese, 1989; McAleese &
Green, 1990; Rizk, Streitz, & Andre, 1990).
On the one hand, hypertext systems provide new ways of individualizing
learning processes due to the possibility of a flexible access to large and
interrelated sets of information . The learner can inspect the node contents by
selecting his or her individual path to follow in a nonlinear structure in order to
fill in knowledge gaps (Yankelovich, Landow, & Cody, 1987). On the other hand,
however, such systems impose greater demands concerning the self-regulation of
338
Wolfgang Schnotz
learning . The individual needs to have his or her own criteria for information
access. Existing knowledge gaps have to be recognized as such, which is
frequently rather difficult (cf. Baker, 1985; Glenberg, Wilkinson, & Epstein,
1982). The information needed then has to be searched for in a large body of
accessible information without losing sight of the searching aim (Carmel,
Crawford, & Chen, 1992; Guthrie, Britten, & Barker, 1991; Guthrie &
Mosenthal, 1987). The more possibilities are available to access different kinds of
information, the higher the probability will be that learners become disoriented or
get lost in hyperspace (Conklin, 1987; Jonassen & Wang, 1993). Therefore,
learners need specific navigational aids, such as maps, indices, dynamic glossaries,
conditioned paths, or guided tours. However, such help also imposes additional
cognitive requirements (cf. De Young, 1990; Kobasigawa, 1993; Lucarella, 1990;
Saxer & Gloor, 1990).
Furthermore, some characteristics of human comprehension and learning can
hardly be appropriately taken into account with hypertexts . During comprehension
of a coherent text or discourse, a flow of thought ensues (Chafe, 1979): The
individual is constructing a topic-specific mental representation that is
successively elaborated from sentence to sentence as long as the respective topic is
maintained. Accordingly, it is possible to guide this flow of thought by means of
a well-designed thematic sequencing so that the learner can successively elaborate
his or her knowledge. Designing hypertexts, however, requires a modularization of
the respective knowledge so that the material displayed remains coherent and
understandable in any type of sequencing. It is very difficult in this case to
maintain a continuous flow of consciousness by thematic continuity (cf. Brown,
1990). Therefore, some authors consider hypertexts to be primarily data bases,
which serve experts in up-dating their knowledge rather than the acquisition of
new knowledge by novices (cf. Jonassen & Grabinger, 1990). Empirical findings,
however, are not clear in this respect. Samarapungavan and Beishuizen (1992), for
example, found that novices demonstrated better understanding of a complex
domain after learning from a hypertext, whereas other investigators found learners
had better understanding after learning from a traditional text (cf. McGrath, 1992).
Earlier discussion has frequently suffered from gross oversimplification of the
differences between hypertext and traditional text and from the erroneous
assumption that the new medium has inherent properties that have per se a
facilitative effect on learning. On the one hand, for example, it has been
downplayed that learning from traditional text is also an active process in which
learners construct network-like mental structures supported by linguistic devices
such as anaphora (cf. Smith & Weiss, 1988; van Dijk & Kintsch, 1983). On the
other hand, it has been overlooked that processes of linearization and
delinearization are also necessary in learning from hypertexts. The learner can
admittedly select his or her own navigation path through the material and decide
which information is to be accessed next. But this information ultimately also has
to be processed sequentially, the only difference being that the linearization is
made by the learner instead of the author. In other words, nonlinearity per se has
339
Theoretical Framework
In investigating processes of knowledge acquisition, a central issue is the structure
and use of this knowledge. The question that should be asked concerns the nature
of the respective mental representations and the processes that operate on them .
Another central issue is the dynamic of knowledge acquisition . We need to ask
how learners realize that they have knowledge deficits and how they are induced to
successively elaborate and modify their knowledge. These questions will be dealt
with here. Because the acquisition of knowledge is usually embedded in and
stimulated by the use of knowledge, acquisition and use of knowledge will not be
considered separately but rather as two sides of an integrated entity (cf. Brown,
Collins, & Duguid, 1989).
340
Wolfgang Schnotz
IF
TIffiN
could be applied to the earlier proposition (EAST (is: Rio de Janeiro, of: Mexico
and thus create the new proposition (LATER (is: time at Rio, than: time at
Mexico. To find out that the time is 5:00 pm in Rio, when it is 2:00 pm in
Mexico, one could apply the following procedure
IF
TIffiN
Such symbol processing requires no reference to the subject domain. The rules can
be used even when an individual knows neither where the respective places are nor
what the terms ~ or later mean. It is sufficient to handle the symbols in a
purely syntactical way.
Mexico
Rio
GreenWich
14
15
16
17
13
19
20
341
342
Wolfgang Schnotz
that a task-relevant procedure cannot be applied. If, for example, the question
concerning the time and date in Anchorage when it is 3:00 pm Thursday in Tokyo
is to be answered in a model-based way and the model contains no time zones,
then determining time differences is not possible. Another kind of a task-defined
structure deficit is that the procedures applied lead to inconsistent results. If, for
example, the model contained time zones but no dateline, learners would conclude
that the time is 8:00 pm Wednesday in Anchorage if they go to the west, and that
it is 8:00 pm Thursday in Anchorage if they go to the east. In both cases, the
model structure would have to be elaborated appropriately to avoid these
problems .
The assumption that knowledge deficits become obvious when the respective
knowledge is used implies that the existence of knowledge deficits can remain
unnoticed due to a limited, too-restricted use of this knowledge. According to this
view, the opportunity that learners have to recognize deficits in their knowledge
depends to a great extent on the demands with which they are confronted. In the
following study, learners were given the task to answer a certain sequence of
questions. The questions were designed and sequenced in such a way that the
learners were required to successively elaborate their mental representation. To
perform these elaborations, the subjects had access to a hypertext about the
respective topic. The analysis concerned the learners' interaction with the
hypertext and focused especially on the question concerning whether successful
and unsuccessful strategies could be distinguished.
Method
Subjects and Learning Task
Seventeen German university students participated in the study. They were asked
to answer seven questions step by step. Because these questions were expected to
stimulate learning activities, they are referred to as learning questions. The
questions (presented in German) were:
Ql)
Q2)
Q3)
Q7)
343
344
Wolfgang Schnotz
12 h
dayi
6h
18 h
.~"":""""""""f---
date line
m idnight line
Oh
day i+1
Learning Material
To answer lhe learning questions, the subjects had a hypertext about time and date
at their disposal . The text consisted of 67 paragraphs with a total length of 65] 6
words (including headings) . Because previous studies have shown that hierarchical
structures cause less orientation problems than associative network structures, the
hypertext system in this study was given an easily comprehensible hierarchical
structure (Gray, 1990; Mohageg, 1992). The paragraphs were subsumed under four
main topics: (1) Geographical relations on the earth's surface, (2) local time and
zone time, (3) time differences on the earth, and (4) date differences on the earth.
Nine paragraphs contained information about algorithms with numerical examples
of how to compute time differences between different locations on the earth.
Another 58 paragraphs contained information about spatial and temporal relations
on the earth's surface. The subjects had free access to the different text paragraphs
through a hierarchically organized menu. Furthermore, the subjects could navigate
sequentially by means of forward and backward buttons as well as by keyword
buttons within the hypertext. The text sections that were already accessed in a
learning session were signaled graphically in the menu but remained available .
345
Procedure
The study was performed in single subject sessions that included the follow ing
phases: diagnosis of individual learning prerequisites, training to use the hypertext
system, knowledge acquisition, comprehension test, and interview.
Diagnosis of learning prerequisites. The subjects were interviewed with regard
to their knowledge about time differences on the earth: They were asked what time
zones are and what relationship exists with the earth's rotation. In addition, they
were asked whether the time in the east is earlier or later, what the dateline is, and
how to proceed in crossing it. Furthermore, they were asked in which country or
on which continent the previously mentioned cities are located. Finally, the
subjects received a German intelligence test combining verbal, nonverbal, and
spatial measures (Amthauer-Test).
Training to use the hypertext system. After diagnosing their learning
prerequisites, the subjects were given explanations on how to use the hypertext
system with other learning content. Furthermore, they were stimulated to think
aloud, that is, to verbalize what came into their minds during learning and
thinking.
Knowledge acquisition. Subjects were then presented questions Q 1 through Q7
step by step on a computer screen. In addition, the screen contained the
hierarchical hypertext menu, which allowed the subjects to access freely the
various sources of information. The topical question remained visible all the time.
After answering a question , the subjects were presented with the next question and
still could freely access the learning material in the hypertext. They received no
feedback concerning the correctness of their answers. The subjects were asked to
continually verbalize what was going on in their minds . Their verbal statements
were tape-recorded.
Comprehension test and interview. After the learning phase, the subjects had to
answer 25 relatively complex questions to enable us to check how well they had
understood the facts. For example, they were asked what day it would be in Alaska
if one started by plane on a Thursday at 9:00 pm in Japan and landed in Alaska
after an 8-hour flight. Another question asked which days exist simultaneously on
earth and how broad are the respective date sections if it is Monday 8:00 pm in
Greenwich. The items could be answered without time limit. In a final interview,
the subjects were asked about their strategies for answering the learning questions
and their information-access strategies.
Scoring
Besides assessing the subjects' intelligence scores, prior knowledge about time
and date differences on earth was evaluated with regard to the number of domainspecific concepts mentioned in answering the prior knowledge questions. The
think-aloud protocols from the learning phase were analyzed qualitatively with
346
WolfgangSchnotz
Results
The students at each end of the learning spectrum were selected in order to
compare the strategies of successful and unsuccessful learners. Each of these two
groups, one with the highest and one with the lowest comprehension test scores,
represented one third of the original group. The six subjects with the highest
scores are called here the successful learners. Their average comprehension score
was 18.8 (SD = 1.8). The six subjects with the lowest scores are called the
unsuccessful learners. Their average score was 13.8 (SD = 2.3). With regard to
prior knowledge or intelligence scores, there was hardly any difference between the
two groups. The averageprior knowledgescore was 7.4 for the successful learners
(SD = 1.6) and 7.1 for the unsuccessful learners (SD = 1.4). All subjects knew
that it is later in the East due to the earth's rotation. They also knew about time
zones, and nearly all knew about the dateline and its approximate position.
However, all subjects wereuncertain about how the date had to be changed there,
depending on the direction in which one would cross this line. They were all able
to localize the cities correctly, except for Anchorage. Only one subject knew that
this city is in Alaska. The successfullearners had an average intelligence score of
110 (SD = 3.7); the unsuccessful learnershad an averageintelligence score of 109
(SD = 6.0).
Question Answering
Concerning the subjects' answers to the learning questions, there were clear
differences between the two groups. Figure 3 shows the relative proportion of
correct answers within both groups across the seven learning questions. Although
most of the successful learners answered all of the questions correctly, the
unsuccessful learners showed remarkable fluctuations. The number of correct
answers was clearly reduced in Q3 (where the date was questioned for the first
time). Q4 was again answered correctly by most subjects, but in Q5 (where
crossing of the datelinewas to be considered for the first time), another clear drop
was observed. Finally, in Q7 (where both the dateline and the midnightline were
to be accounted for), yet another decrease in the proportion of correct answers
ensued with the unsuccessful learners. The test for statistical significance of the
347
respective differences for Q3, Q5, and Q7 between both groups resulted in p<.05 ,
p-c.Ol, and p = .13 (Mann-Whitney).
100 %
\
\
5 0%
\
\
/
\
0% +------,J----+-
Q1
Q2
IQ..
Fig. 14.3 . Relative frequency of correct answers within successful and unsuccessful
learners across learning question s Ql through Q7.
In some cases, the subjects only copied the surface structure of the respective
examples of computing time differences from the text, as expressed in the
following comment:
348
Wolfgang Schnotz
I have to add the time of day of the city mentioned after
the time delay of the first-mentioned city . Then I have to
subtract the time delay of the second-mentioned city .
Obviously, whether the respective city is west or east of the other one and what
time differences should therefore be expected play no role here. Accordingly, the
unsuccessful learners often expressed syntactical problems, such as where to fill in
numerical values in a formula and how to take into account a minus sign. The
following extract provides some typical remarks:
What do I have to insert where now?
Now I have two negative numbers; which one do I have to
subtract from the other one?
Do I have to add or subtract now?
What shall I do with the negative sign?
These kinds of remarks appeared 1.5 times (SD = 1.0), on average, with the
unsuccessful learners, but only 0.3 times (SD = 0.5) with the successful learners
(p<.05, Mann-Whitney) . In contrast, the successful learners did not restrict
themselves to using only a symbol-processing strategy but rather employed a
model-based strategy from the beginning. They obviously tried to localize the
respective cities mentally, through their geographical prior knowledge, and to
integrate the respective entities into a coherent image. The protocols of those
subjects frequently included comments such as the following:
It is disturbing to know just the time zone value of the
city, although this is, in principle, sufficient to
calculate the result . I want to be able to imagine the
facts as well.
Now, Rio is west of Cairo, Mexico is even further west,
Tokyo is east of Cairo. That means . . . The time in Tokyo
is the most advanced of all those cities .
I would like to know Where Anchorage is. I will assume now
that it is somewhere in the West of the U.S .A .
349
The subject first started with a model construction by means of building a spatial
configuration: Mexico-Rio-Greenwich. Then a step of model analysis ensued, by
distinguishing between the whole and its parts within the configuration by
applying a part-whole schema. The result led to the decision that a symbolprocessing procedure was to be applied: In this case, the required difference
(Mexico-Rio) was a part of a whole (Mexico-Greenwich). Therefore, a subtraction
was necessary: The smaller amount (Greenwich-Rio) had to be subtracted from the
bigger amount (Mexico-Greenwich). If the required difference had been the whole
(such as in Rio-Cairo), however, the respective differences (Rio-Greenwich and
Greenwich-Cairo) would have been parts of a whole that, therefore, had to be added
in order to determine the required time difference. Finally, another model analysis
was performed testing whether Rio is west or east of Mexico. This analysis again
resulted in a decision about the next symbol-processing operation : Because Rio is
east of Mexico and it is later there, the difference determined before (3 hours ) had
to be added to the time in Mexico (10:00 pm Tuesday), which then (after taking
into account the change of date at midnight) resulted in 1:00 am Wednesday.
350
Wolfgang Schnotz
means of mental simulations (cf. Brown, Burton, & de Kleer, 1982). The
following extracts from subjects' protocols indicate the performance of such
simulations, which finally resulted in a discovery of the midnightline:
In the protocols of the unsuccessful learners, only two indications of such mental
simulations were found. Instead, those subjects tended to answer the question
(why could one not reach the past or the future by frequently crossing the dateline)
by stating that, in certain cases, the correction of the dateline had to be
suppressed. Related remarks made included the following:
If one returns to the same time zone from which one
started, one should not correct the date at the dateline .
The correction of the date at the dateline should probably
be performed only once in 24 hours .
Instead of elaborating the structure of the respective mental model, these subjects
performed, as it were, ad hoc repairs of their procedures by reducing the
applicability of those procedures . Indications of such procedure repairs were found
with three of the unsuccessful learners but with none of the successful learners.
351
Whereas the groups hardly differed from each other between Q2 and Q7, the
successful learners used nearly twice as much time before answering the rather
easy question Ql (475 sees (SD=207 as did the unsuccessful learners (245 sees
(SD = 109 (p<.05, Mann-Whitney ).
500
400
300
200
100
0+----f---1---f-----f---+---+----i
01
02
03
04
05
o6
07
Fig. 14.4. Pattern of the average learning time invested by successful and
unsuccessful learners across questions QI through Q7.
352
Wolfgang Schnotz
10
8
6
4
/
<,
''0--- --d'
O+---+---t------1f----+--+---+---i
02
-0 3
01
04
05
o 6 07
Fig. 14.5. Pattern of the average number of times that information was
accessed by successful and unsuccessful learners across questions Ql through Q7
Interpretation
The results just presented suggest that the two groups used different strategies to
answer the learning questions and, accordingly, interacted in different ways with
the hypertext system. The successful learners obviously used a model-based
strategy from the beginning. Because they first had to perform the initial
construction of a mental model, they retrieved much information and spent much
time at this point. Later, they concentrated their information access at those
points where an elaboration of the model structure was required to answer the
questions, that is, at Q3 (where the date was questioned the first time), at Q5
(where crossing the dateline was to be considered the first time), and at Q7 (where
both the dateline and the midnightline were to be accounted for). The successful
learners combined model construction, model analysis , and symbol processing in
a flexible way. Inconsistencies within the subjects ' knowledge were usually
solved by elaborating the respective model structure.
In contrast, at the beginning, the unsuccessful learners tried to get along with a
pure symbol-processing strategy and tackled the model construction only in the
last questions . The respective model structures were obviously not well
elaborated. Inconsistencies within the subjects' knowledge were solved by means
of ad hoc repairs of procedures rather than by a systematic elaboration of the
model structure . Because answering of questions with mere symbol processing
requires relatively little information, especially if learners mainly follow the
available examples of how to calculate time differences, those subjects showed a
lower and rather flat pattern of information access. Yet, due to the cognitively
demanding initial construction of a mental model, one would expect increased
353
information access as soon as the learner begins to construct a mental model. This
was not the case, however, as can be seen from Figure 5: The frequency of
information access was not remarkably higher in the last two questions than it had
been before. This pattern was possibly the result of difficulties in specifying the
aim of information search, because, as mentioned earlier, the unsuccessful learners
had no clear idea what information they should search for with these questions.
Discussion
Learners with more prior knowledge are usually better able to find and process
information that is relevant for further learning than are learners with less prior
knowledge (O'Donnell, 1993). In the present study, however, no such effect could
be observed; there was hardly any difference in prior knowledge between successful
and unsuccessful learners. Most subjects remarked that they had not yet thought in
a systematic way about this learning topic, so they could be considered novices .
Clear differences were found, however, concerning the strategies used in answering
the learning questions: The successful learners employed a model-based strategy
from the outset, whereas the unsuccessful learners tried first to get by with a
symbol-processing strategy and tackled the construction of a mental model only
during the last question . These different strategies were also reflected by different
patterns of information access among successful and unsuccessful learners.
These strategy differences can be related to a distinction between a local and a
global orientation of knowledge acquisition (cf. Hacker, 1986). With a local
orientation, the individual reacts ad hoc on the topical demands, whereas with a
global orientation , the attempt is made to systematically construct and elaborate
the respective knowledge structure in order to be useful also for other known or
anticipated tasks. Accordingly, the unsuccessful learners in the preceding study had
a local orientation in that they restricted their information access to the topically
relevant numerical-algorithmic information. The tendency to solve inconsistencies
by means of ad hoc repairs of procedures, instead of by elaborating the model
structure, can also be interpreted as an indication of a local orientation in solving
learning tasks. In contrast, from the outset, the successful learners in this study
were rather globally oriented in the strategies used to solve the learning tasks and
in terms of their information access strategies. At the beginning, they retrieved
much information for the initial construction of a mental model, which was then
elaborated step by step according to the respective task demands.
As mentioned earlier, the unsuccessful learners frequently reported that they did
not know exactly for which information they should search, especially in the last
questions. This corresponds to the finding that search for information is usually
rather difficult with complex search goals (Armbruster & Armstrong, 1993).
Because those subjects also reported having difficulties in imagining the facts, and
354
Wolfgang Schnotz
because their learning behavior indicated that they delayed tackling the
construction of a mental model, one can speculate that mental models constructed
by learners also function to specify the information required for further learning
(cf. Mayer, 1989). Accordingly, mental models would be not only the result of a
retrospective integration of previous information with topical new information
but also the basis for a prospective search for further information to continue the
knowledge acquisition process . Further research will be necessary to clarify this
point.
Instructional Implications
What practical consequences can be drawn from these considerations? Because the
acquisition of knowledge is usually embedded in and triggered by the use of
knowledge, the respective learning processes can be fostered not only by an
adequate design of hypertext systems, but also, in order to stimulate learners to be
active builders of their own knowledge, by selecting adequate learning tasks and
by bringing these tasks into an appropriate sequence.
Selection of learning tasks. If one follows the considerations mentioned earlier,
learning tasks are to be selected in such a way that learners are prompted to
construct mental models on the basis of propositional representations and to
systematically relate both forms of representation to each other. For this purpose,
questions can be asked that serve to localize entities correctly within the model
and that afford a determination of relations between these entities in the model
structure. With regard to the learning content of this study, such questions could
concern, for example, where the dateline is, whether it is earlier or later on its
western side than on its eastern side, where on earth (measured according to the
sun 's position) the time is 6:00 am and where it is 6:00 pm. Learners should be
forced to operate with their models in such a way that cognitive conflicts would
emerge if the achieved model structure is not yet adequate. Furthermore, they
should be prevented from utilizing ad hoc repairs in case of cognitive conflicts
without considering the consistency of their knowledge structure. Learners should,
for example, be encouraged to compare the topical solution with the solutions of
previous tasks and to check them for compatibility.
Sequencing of learning tasks. The selected learning tasks have to be brought
into a certain sequence, which prompts learners to successively elaborate their
knowledge and to use this knowledge in different ways. For this purpose, one can
follow the concept of a successive model progression (cf. White & Frederiksen,
1990). The first step is to determine for each task the model structure that is
required to solve the task with a model-based strategy. The second step is to place
those task-specific model structures in a sequence of increasing complexity. The
result is not only a sequence of model structures but also a sequence of learning
355
tasks, which should motivate the learner to move from one model structure to the
next, more complex structure.
Display of information. The concept of a successive model progression also
provides some indication about when learners need which kind of information to
proceed from one model structure to the next. In view of a possible cognitive
overload of the learner, in the case of unrestricted access to a very broad set of
information, it would be sensible to restrict the accessible information for
learning with hypertext in a task-specific way. According to the preceding
considerations, however, the information offer also has to be large enough that it
is possible for learners not only to gather information from the local view of the
momentary learning task, but also to have the opportunity to perform a more
globally oriented search for information, having all the information available that
has been necessary so far to construct a model on the task-specific elaboration
level. In order both to restrict the available information to a reasonable extent and
simultaneously to make this information offer broad enough, it would be possible
to use specific display techniques as, for example, so-calledfisheye views (Furnas,
1986). A fisheye view is a kind of presentation that displays (analogous to a lens
with a very wide angle or fisheye) nearby objects in more detail and the periphery
in less detail. In this way, the topical part of the information base could be
foregrounded, whereas other parts remain in the background although they are still
accessible to the learner. To date, very little is known about the cognitive effects
of such hypertext -specific types of staging (cf. Meyer, 1975).
Knowledge acquisition from hypertext is a complex research area that has
enlarged the traditional scope of text processing research by a number of additional
problems. It would be too simplistic to look only for positive or negative effects
of this new medium on learning outcomes. Instead, process-oriented studies are
required that analyze how different learners interact with hypertexts, which
strategies they use, and which mental representations they construct. Besides
cognitive investigations, affective and motivational aspects should also be
analyzed in order to achieve a comprehensive view of knowledge acquisition from
hypertext. A better understanding of these processes will gradually enable us to
design the respective learning environments in a way that takes into account the
psychological principles of knowledge acquisition as far as possible.
Acknowledgments
The author gratefully acknowledges the support of Emmanuel Picard in
developing the hypertext sytem used in this study. He also wants to thank Roger
Saljo and Barbara Burge for their comments on a former version of this article.
356
Wolfgang Schnotz
References
Armbruster, B. B., & Armstrong, J. O. (1993) . Locating information in text: A focus
on children in the elementary grades. Contemporary Educational Psychology, 18(2),
139-161.
Baddeley, A (1992). Working memory : An introduction. International Journal of
Psychology, 27(3/4), 96 .
Baker, L. (1985). How do we know when we don't understand? Standards for ev aluating
text comprehension. In D. L. Forrest -Pressley , G. E. MacKinnon, & T. G. Waller
(Eds.), Metacogn ition, cognition, and human performance. Vol. I: Theoreti cal
perspectives (pp. 155-205). Orlando, FL: Academic Press .
Brown , J. S., Burton, R. R., & de Kleer, J. (1982). Knowledge engineering and pedagogical techniques in SOPHIE I, II, and ill . In D. Sleeman & J. S. Brown (Eds.) ,
Intelligent tutoring systems (pp. 227-282) . London : Academic Press .
Brown, J. S., Collins, A, & Duguid, P. (1989) . Situated cognition and the culture of
learning. Educational Researcher, 18(1), 32-42 .
Brown, P. J. (1990). Assessing the quality of hypertext documents . In A Rizk , N.
Streitz, & J. Andre (Eds.), Hypertext: Concepts, systems , and applications (pp . 112). Cambridge, England : Cambridge University Press.
Carey , S. (1986) . Cognitive science and science education . American Psychologist,
41, 1123-1130.
Carmel, E., Crawford, S. F., & Chen, H. (1992) . Browsing in hypertext: A cognitive
study . IEEE-Transactions on Systems, Man, and Cybernetics, 22(5), 865-884 .
Chafe, W. L. (1979). The flow of thought and the flow of language . In T. Givan (Ed.),
Syntax and semantics, Vol. XII: Discourse and syntax (pp. 159-181). New York:
Academic Press .
Chi, M. T. H., Glaser, R., & Rees, E. (1982) . Expertise in problem solving . In R. J .
Sternberg (Ed.), Advances in the psychology of human intelligence (Vol. I, pp. 775). Hillsdale, NJ: Erlbaum .
Conklin , J. (1987) . Hypertext: An introduction and survey. IEEE Computer, 20(9), 1741.
De Young, L. (1990) . Linking considered harmful. In A Rizk, N. Streitz, & J. Andre
(Eds.), Hypertext: Concepts, systems, and appli cations (pp. 238-249). Cambridge,
England : Cambridge University Press .
diSessa, A (1983) . Phenomenology and the evolution of intuition . In D. Gentner & A.
L. Stevens (Eds .), Mental models (pp. 15-33). Hillsdale, NJ: Erlbaum.
Furnas, G. W. (1986, April). Generalized fisheye views . CHI Proceedings, 16-23.
Garner , R., & Reis, R. (1981) . Monitoring and resolving comprehension obstacles :
An investigation of spontaneous text lookbacks among upper-grade good and poor
comprehenders. Reading Research Quarterly, 26, 569-582 .
Garrod, S. C. (1985) . Incremental pragm atic interpretation versus occasional
inferencing during fluent reading. In G. Rickheit & H. Strohner (Eds.), Inferences in
text processing (pp. 161-181) . Amsterdam: North-Holland.
Glenberg, A M., Wilkinson, A. C., & Epstein, W. (1982) . The illusion of knowing :
Failure in the self-assessment of comprehension. Memory & Cognition, 10, 597602 .
357
8.
Kobasigawa, A. (1993) . Children 's retrieval skills for school learning. The Alberta
Journal of Educational Research, 29, 259-271.
Kosslyn, S. M. (1981) . The medium and the message in mental imagery : A theory .
Psychologi cal Review , 88, 46-66.
Lucarella, D. (1990). A model for hypertext-based information retrieval. In A. Rizk, N.
Streitz, & J. Andre (Eds.), Hypertext: Concepts, systems , and appli cations (pp. 8194). Cambridge, England : Cambridge University Press.
Marton , F., & Saljo, R. (1976). On qualitative differences in learning : II. Outcome as a
function of the learner's conception of the task. British Journal of Educational
Psychology, 46, 115-127.
Mayer, R. E. (1989) . Models for understanding. Review of Educational Research,
59(1), 43-64 .
McAleese, R. (Ed.). (1989) . Hypertext: Theory into practice. Worcester, England:
Billing & Sons .
McAleese, R., & Green, C. (Eds.). (1990). Hypertext: State of the art. Oxford, England :
Intellect.
McGrath, D. (1992) . Hypertext, CAl, paper or program control: Do learners benefit
from choices? Journal of Research on Computing in Education, 24(4), 513-532 .
Melton, R. F. (1978) . Resolution of conflicting claims concerning the effect of behavioral objectives on student learning . Review of Educational Research, 48, 291302 .
358
Wolfgang Schnotz
Part Four
Chapter 15
Abstract
Using data from two studies of scientific reasoning , this chapter explores whether
transactive discussion is the basis of productive peer collaborations and questions
what role the partner plays in the apparent effectiveness of this type of discussion. In
the first study, dyads who engaged in transactive discussion showed more
improvement than dyads who did not have transactive discussions . In the second
study, both dyads and children working alone showed improvement related to talk in
general. However, dyads produced more transactive types of talk and showed a more
complex understanding of the problem that they generated more quickly . Having a
partner was not a necessary or sufficient condition for producing transactive talk but
increased likelihood that it would occur. The data from these studies suggest that the
value of peer collaborations may be that the presence of a partner provides a natural
context for elaborating one's own reasoning.
The past ten years of research on children's collaborations have demonstrated that
children who work with peers can show improved reasoning and problem solving .
Although not all collaborations are successful, a growing number of studies have
pointed to the conditions under which children who work with peers learn more
than children who work alone. The earliest studies demonstrated that the degree
of mutuality is an important factor in determining the success of the
362
Stephanie D. Teasley
Transactive Discussion
363
confirmed that children produced more transactive reasoning when working with
peers than they did with adults.
Table 15.1. Types of Transactive Statements
TYPE
Feedback Request
Paraphrase
Justifi cation Request
Juxtaposition
Completion
Clarification
Refinement
Extension
Critique
Integration
IRANSACTIYE MOVE
Do you understand or agree with my position?
I can understand and paraphrase your position or reasoning.
Why do you say that?
Your position is X, and my position is Y.
I can complete or continue your unfinished reasoning.
No, what I am trying to say is the following .
I can elaborate or qualify my position to defend against your
critique .
Here is a further thought or elaboration.
Your reasoning misses an important distinction, or involves a
questionable assumption .
We can combine our positions into a common view.
364
Stephanie D. Teasley
those who do not. This evidence comes from studies of children collaborating on
tasks in various domains . In the domain of moral reasoning , several studies have
found that children show higher levels of moral judgment when they engage in
discussions that act on their own and their partner's reasoning (Kruger &
Tomasello, 1986; Kruger, 1992). Damon and Phelps (1989) have provided
evidence that dyads learn more from discovery and reflection tasks (e.g.,
proportional reasoning, spatial perspectives) when they engage in mutually
responsive dialogue in which ideas about the logic of the task are shared . Tudge
(1992) has shown that children benefit from collaboration when they shared their
reasoning about solutions for a mathematical balance beam task. Thus , these
studies on moral, conceptual, and mathematical reasoning suggest that peer
interactions can support learning, especially when children use trans active forms
of discussion to communicate ideas.
365
Example I
Note that, although the dispute ended, there was apparently no agreement
between the dyad partners on the causal status of the green olives. In this case,
even though the partners were acting on each other's reasoning, consensus was
not the result, and Child 2 maintained her own position. Although these children
experienced a conflict between positions, at least one member of the dyad did not
end up with a higher level of understanding. However, it is clear that both
partners understood each other's position and ended the discussion, even though
they did not agree.
Although the Azmitia and Montgomery (1993) study generally supports the
claim that trans active discussion is related to successful collaborative outcomes,
there were two differences in the results of their study compared with previous
studies of transactive reasoning. First, not all types of transactive discussion (e.g.,
transactive questions and responses) were correlated with better task
performance. The authors note that the overall trans active properties of the
discussions generated by their scientific reasoning task were best captured in the
366
Stephanie D. Teasley
367
represented in the microworld by how far the cars could travel in a standard
period of time . Children's predictions about speed were indicated by their
selection of one of several distance markers located along the racetrack.
In a study of children working alone, Schauble (1990 ) found that the students
who came to a more complete understanding of the micro world were those who
evaluated both the patterns of evidence they constructed and their own changing
theories. By using Schauble's racecar task in a collaborative context, we were
able to specifically examine whether children engaged each other in trans active
discussions about theories and evidence and the extent to which this type of talk
was related to theory change.
Children's understanding of the causal status of car features was measured by
their ability to predict accurately the speed of the cars . Because the children were
asked to say how fast each car would go before the cars ran along the racetrack,
we could determine how well each dyad unders tood the combined effects of any
particular set of car features. We looked for improvement in predictions over the
course of the 20 minute session to see which dyads were using their accumu lating
evidence to revise their theories about the feat ures. Of 24 dyads, 15 dyads
improved in their ability to predict the speed of any individual car . Correlational
analyses showed that improved prediction accuracy was not associated with the
number of cars constructed or the number of roadtests run. In addition , the dyads
who showed improvement in prediction accuracy did not construct more cars or
run more roadtests than did the dyads who did not improve.
Since the activity with the microworld itself did not determine which children
would better understand the racecar features , what did? Analyses of the children's
protocols showed that the verbal activity of the improvers and nonimprovers was
characteristically different. As will be seen, the dyads who engaged in transactive
discussion during more of their experiments were those who improved in
prediction accuracy.
Following the method of Berkowitz and Gibbs (1983) , I coded the presence or
absence of transactive discussion (not the number of turns codes as transactive).
Unlike Berkowitz and Gibbs, however, I considered whether or not transactive
discussion occurred in two phases of experimentation: evidence generation and
evidence evaluation . Specifically, discussions were coded as transactive if they
indicated reasoning about how to generate evidence (e.g., indicating why a
particular feature was chosen, how to test the effect of any feature) or evaluate
evidence (e.g., explaining the racecar's speed, stating a hypothesis about causal
effects of racecar features) . Discussions that were simply descriptive of the
child's activity during evidence generation (e.g ., how to operate the computer,
what commands were being entered) or of the racecar's behavior during evidence
evaluation (e.g ., merely describing the roadtest outcome without assessment or
speculation about why the car was fast or slow) were not coded as transactive.
Then, for each phase of experimentation considered separately, I calculated the
frequency of transactive reasoning on a roadtest-by-roadtest basis (as proportion
of roadtests with discussion coded as transactive). In doing so, I assessed whether
368
Stephanie D. Teasley
the frequency and timing - not total volume - of transactive reasoning are
important.
Example 2
Cl : Put on fins.
C2: It seems like when we put no fins on it, it
Cl:
Yeah
goes farther .
C2:
Cl:
than
when it does have a fin .
C2: Okay .
Cl: So let's try it with those same features that we had on
that one [previous car], except let 's put a fin on it
next time .
C2: Okay.
Cl : See how it, if it goes the same thing.
This discussion was coded as trans active because the partners were discussing
their reasoning about how to construct the roadtest. In fact, these partners were
talking about why they were selecting particular car features (the tailfin), as well
as their overall plan for the roadtest (to see if adding the tailfin affects the
outcome compared with a previous car configuration). Contrast the discussion
just presented with that in Example 3:
Example 3
369
In this discussion, neither child explained why the part icular features were
selected for each car; there were no explanations about why the car is expected to
travel a certain speed (the prediction flag chosen); and there was no rationale for
choosing to run these two cars in the same roadtest. Because the partners did not
display any reasoning, the evidence generation phase of this experiment was not
coded as transactive.
Analyses of the conversation of the dyads who improved in their ability to
predict speed showed that they engaged in trans active discussion during the
evidence generation phase of 36% of all roadtests. In contrast , the dyads who
showed no improvement in prediction accuracy engaged in transactive discussion
during the evidence generation phase of only 19% of their roadtests. This
difference was significant, t (22) =1.72, P < .05.
Engaging in transactive discussion during evidence generation was also related
to prediction accuracy, because these discussions usually involved a plan or
strategy for how to construct the roadtest in a meaningful way. These discussions
could lead to more systematic experimentation and therefore more intelligible
outcomes. Consider the discussion in Example 4:
Example 4
Cl:
Let's try to think of, like, for slow ones that color
didn ' t matter and both [cars] go to that [prediction
marker for slowest cars].
C2: Yeah .
Cl : If we leave this [the second car constructed] the same,
they will probably go to there.
C2: Yeah.
Cl: It will prove that color doesn't matter .
C2: Now we can prove that col or doesn' t matter ?
Cl : Yeah , because they'll both go the same .
C2: Oh. ...
In this discussion, Child 1 suggests how they might construct the roadtest with
two objectives: (a) making the cars go slowly and (b) testing to see if color
matters. Although Child 2 seems to agree on the proposed plan for the roadtest,
she does not understand the concept of a controlled test of variables (where all
features are held constant except one). Without Child l' s statement that leaving
the cars the same except for color will prove that color does not matter, Child 2
would not have been likely to understand the significance of this test or possibly
any subsequent controlled test of features. Thus transactive discussion served to
externalize each partner's logic and to make it available to the other partner for
agreement , dispute, or, in this example, clarification . That this dyad went on to
produce further controlled tests, sometimes suggested by Child 2, supplies further
Stephanie D. Teasley
370
evidence that both partners understood the nature of controlled tests and could
reason based on evidence from this type of roadtest.
Transactive Discussion and Evidence Evaluation
371
dyads, t (22) = 3.27, p < .001. The outcome discussions of the non-improvers
were more likely to be simple descriptive statements about what the cars were
doing rather than reasoning about why the car had a particular speed.
Summary
The results of this study show that short-term learning and theory change in a
causal reasoning task vary as a function of the difference in the kind of verbal
activity children engaged in while they were collaborating. Dyads who were able
to integrate their evolving theories about specific racecar features into global
assessments of car speed showed improvement in prediction accuracy over the
course of the session . These dyads engaged in transactive discussion as they
generated and evaluated experimental evidence . These transactive discussions
helped children to follow what Schauble (1990) calls normative scientific
reasoning. However, simply having a partner was not enough. Not all of the
children in this study were able to coordinate individual theories into correct
predictions about the combined effects of those features. In addition, the dyads in
this study engaged in many of the problematic behaviors that Schauble (1990)
and others (e.g., Klahr et al., 1993; Kuhn et al., 1988) have found to be typical for
children working individually on reasoning tasks, such as generating
uninformative evidence and misinterpreting experimental outcomes.
In considering the possible ways that transactive discourse might support
learning in peer collaborations, it is important to notice that we are analyzing
discourse in dyads without acknowledging that it may be the talk itself (of any
kind) and not the partner that facilitates learning . In any study of collaboration,
the collaborative condition necessarily confounds two factors: having a partner
and talking aloud. We cannot conclude that a particular type of talk, such as
transactive discussion, is the important factor in collaboration until we investigate
whether producing talk while working alone also facilitates reasoning.
372
Stephanie D. Teasley
was most interested in the performance of the two talk groups. Given the
literature on children's private speech (see Diaz & Berk, 1992), it seemed likely
that asking these children to produce verbal protocols as they worked alone could
improve performance. However, would the children who talked as they worked
alone engage in transactive reasoning (e.g., produce self-oriented transacts)?
Further, what are the comparative effects of transactive utterances with and
without a partner (a test of self-oriented versus other-oriented transacts)?
The spaceship micro world was very similar to the racecar microworld as
children reasoned about the causal status of unknown variables and ran
experiments to test their theories about these variables. In the spaceship
micro world , children were asked to determine the causal effect of only one
variable: the mystery key. Children were shown how five LOGO-like commands
affected the spaceship (e.g., forward, backward, right turn, left turn, fire) and then
asked to determine how the new mystery key worked. Unknown to the children,
the mystery key acted as a repeat key, making all commands entered before it
repeat. For example, FORWARD 2 FIRE 3 MYSTERY KEY 2 made the
spaceship repeat the FORWARD 2 and FIRE 3 commands in sequence two
additional times, for a total of three Forward-and-Fire sequences. Commands
entered in a program after the mystery key were executed after the repeated
sequence was complete. For example, FORWARD 2 MYSTERY KEY 2 FIRE 3
would make the spaceship go forward six times and then fire three times. Success
on the spaceship task involved identifying the correct function of the mystery
key. Children's understanding of the mystery key was assessed after they worked
with the microworld for one 20-minute session. Their understanding could be
very general (e.g., knowing that it made the program repeat) or more complex
(e.g., knowing how many times the program would repeat and that commands
after the mystery key would not repeat).
Analyses of children 's hypotheses about the mystery key showed that talking
affected performance but having a partner did not. Children who talked aloud as
they worked, either alone or with a partner, came to a better understanding of the
mystery key than did children in either silent group. Furthermore, the amount of
speech was related to performance because the total number of utterances! was
significantly correlated with children's understanding of the mystery key, r (23) =
.49, P < .05. These data suggest that the production of talk of any kind was
associated with better hypotheses. Previous collaboration studies finding superior
IBecau se I was coding the talk produced by children working alone as well as by dyads , I
coded at the level of utterance rather than the conversational turn. The unit of utterance,
although roughly corresponding to a sentence, was not considered here in the strict
grammatical sense but as it commonly occurs in spoken English (see Schegloff, 1991).
373
performance in dyads (who talk) as compared with children working alone (who
do not talk) may have been measuring the benefits of talking and not the benefits
of having a partner. Given that talk improved performance, were children in both
the alone and dyad conditions producing transactive reasoning ? Was there an
added advantage for having a partner?
As in the racecar study, the dyads' discussions were coded separately for the
evidenc e generation phase and the evidence evaluation phase of experimentation
to capture the timing of transactive reasoning . Because the previous analysis
suggested that the volume of talk was important, transacts were coded on an
utterance-by-utterance basis. Utterances were coded as transactive if they
indicated reasoning about how to generate evidence (e.g., indicating why a
program was designed, how to test the effects of the mystery key) and evaluate
evidence (e.g., assessing why the spaceship moved in a certain way, stating a
hypothesis about how the mystery key works) . Utterances that were simply
descriptive of the child's activity or of the spaceship's behavior were not coded as
transactive .
I asked whether children who worked alone produced any transactive reasoning
and, if so, how it compared with the transactive reasoning of the children in
dyads . It was not surprising that most of the children who talked aloud as they
worked alone did not produce as much talk overall (transactive or otherwise) as
did the children with partners . Children working alone showed longer pauses and
greater variability in the amount they talked (see Diaz, 1992, for discussion of the
difficulty in eliciting verbalization from children working alone) . Although the
children working alone talked less than the dyads did, they produced utterances
coded as transactive.
Because alones talked less than dyads, I followed the standard convention of
analyzing the transactive utterances as a proportion of total talk (e.g., Kruger &
Tomasello, 1986). Given the total amount of talk produced by any individual or
dyad, this analysis calculated how many of those utterances could be considered
transactive. The analysis showed a general pattern in which dyads produced
significantly more transactive utterances than alones, t (21) = 3.22, P < .01,
whereas alones produced more descriptive utterances than dyads, t (21) = 4.29, P
< .001 (see Figure 1). Transactive utterances were positively correlated with
performance, r (23) = .57, P < .01, and descriptive utterances were negatively
correlated, r (23) = -.51, P < .05. These results suggest that having a partner may
make trans active statements more likely and that performance is likely to be
enhanced, regardless of whether the transacts are self-oriented or other-oriented.
374
Stephanie D. Teasley
100
Alones
. Oyads
90
80
8
~
70
60
::> SO
"50
....
40
30
I0
20
,~
10
,
,
,
,
'4l;
1..0'"
,,",
"
Interpretive
Descriptive
Fig. 15.1. Mean Proportion of Transactive and Descriptive Talk for Children Working
Alone and Children Working in Dyads,
C2:
Cl:
C2:
[pause]
Cl :
375
C2: Well, which one you want to do is- we want to put one ,
like turn or something .
Example 8
Cl :
In both transcripts, children show much talk about their reasoning. For this dyad
(Example 7), the discussion concerned the intelligibility of a program with fire
commands entered before and after the mystery key. Both children wanted to see
if all commands are affected by the Mystery Key, but they held different ideas
about how to test that hypothes is. In the child working alone (Example 8), the
transcript shows that she was reasoning about the effects of the Mystery Key by
calculating in advance how many commands will be executed.
In contrast to the two transcripts just presented, consider the next two excerpts
from children who showed no transactive utterances while they generated
evidence. The first example is from a dyad; the second is from a child working
alone.
Example 9
C2:
Cl :
C2:
Cl :
376
Stephanie D. Teasley
C2:
Example 10
Cl :
In both transcripts, there was no talk about why commands were being entered or
how the spaceship might move. In the dyad (Example 9), Child 2 did not describe
all the commands he entered, so his partner was not sure when or if the mystery
key had even been selected. In the alone condition (Example 10), the child was
only narrating the sequence of commands being entered . Both examples contrast
sharply with the previous transcripts (Examples 7 and 8) in the amount of
reasoning displayed in the talk.
Example 11
[Program runs]
C2:
Cl :
C2:
Cl :
C2:
Cl:
C2:
377
Example 12
Cl :
378
Stephanie D. Teasley
itself'; ".. . maybe it repeats itself as many times as you pushed it in"). Child 1
then produces evidence that he not only agrees but also applies that hypothesis to
explain the behavior of a previous experiment. The child working alone produces
fewer transactive statements but, nonetheless , makes a conclusion based on the
observed behavior of the spaceship . Compare the preceding transcripts with the
following two protocols in Examples 13 and 14:
Example 13
[Program runs]
Cl :
C2 :
Cl:
What?
[pause]
C2:
Huh??
Example 14
[Program runs]
Cl : [hmming]
Okay, one, two, one, two .
Okay, it's done .
These two examples typify the talk when children were producing more
descriptive utterances than transacts. In the dyad (Example 13), although Child 1
asks a transactive question, Child 2 does not respond with a transact, and no
further transactive discussion ensues. Although the absence of trans active
utterances was more typical of the children working alone, as shown above in
Example 14, there were children in both conditions who produced few transactive
utterances .
Summary
Dyads were producing more transactive utterances in both the program generation
and evidence evaluation phases of their experiments than were the children who
worked alone. Having a partner was not a necessary or sufficient condition for
trans active reasoning, however, because some of the children in the alone
condition produced transactive talk, and some of the children in dyads produced
very little . The increased likelihood of transactive reasoning by dyads shown in
these data suggests that the value of the peer in collaborations may be that the
379
Hypothesis Type
Incorrect
Condition
Talk Alones (n = 12) 3
Correct
Correct
General
Complex
11
These data suggest that having a partner to talk with helps children to refine and
elaborate their ideas. In addition to a deeper understanding of the mystery key,
children who talked with partners also showed a second advantage: They arrived
at a correct general hypothesis much more quickly than did children who talked
alone. After only four experiments, 75% of the children who talked with partners
gave a correct general hypothesis about the mystery key . In contrast, it took 11
experiments before 75% of the children who talked alone gave a correct general
hypothesis. It seems, therefore, that talking with a partner leads to more rapid
generation of correct general hypotheses that then get further developed as
children continue to work together.
Several factors may account for slower and less complex hypotheses generated
by children in the talk alone condition . First, various cognitive studies in which
individuals have been asked to produce think-aloud protocols have found that the
production of concurrent verbalization can slow problem-solving activity (e.g.,
Ericsson & Simon, 1980) . In addition, studies of children 's private speech have
shown that the beneficial effects of verbalization may not appear immediately
(e.g., Bivens & Berk, 1990). Because all children in this study were limited to a
380
Stephanie D. Teasley
twenty-minute session, it is possible that the talk alones may have needed more
time for the full benefits of their verbalization to be realized. If the session had
lasted longer or delayed posttest performance measures had been administered ,
the talk alones may have shown more positive effects from verbalization.
Conclusion
The data from both the racecar study and the spaceship study provide evidence to
support the hypothesis that transactive talk facilitates learning in scientific
reasoning tasks . Children in both studies who came to understand the causal
relations of variables in the two microworlds were those who talked about their
reasoning as they worked. Data from the spaceship study suggest that
collaborations lead to learning not because of the partner but rather because
having a partner increased the likelihood that children would produce transactive
statements. Transacts are particularly crucial for success on the reasoning tasks
used here, because previous research has documented how easily children are
distracted by surface features of experimental outcomes and how often they fail to
coordinate theories with evidence (Klahr et al., 1993; Schauble, 1990).
The answer to the question, "How important is the peer in peer collaboration? ",
may fundamentally rest on the interdependency of cognition and social relations
in communication. Built into communication is an implicit obligation to make
sense to one's conversational partner (Grice, 1989). When children are talking to
partners as they work on scientific reasoning tasks, the situation requires
coordinating conversational coherence to maintain and make sense of the ongoing
task activity (e.g., Teasley & Roschelle, 1993) . To simultaneously manage
coherence and activity, children working with partners produce transactive
utterances that make explicit their intentions, current interpretations, and
expectations. The production of these types of utterances engages children in the
epistemic operations that are crucial to scientific reasoning (see also Pontecorvo ,
1987) . However, because some of the children working alone in the spaceship
study also produced transactive statements, the necessity of having a partner is
questionable. As children talk aloud to themselve s, the social implications of
communication are maintained (Wertsch, 1985), although the conditions are not
as stringent, because the failure to be coherent has no immediate ramifications .
Several studies provide evidence that the mere anticipation of having a partner
can lead to cognitive gains. In two studies of cognitive tuning (Cohen, 1961;
Zajonc, 1960), subjects who expected to transmit information to others were
found to have more organized cognitive structures than did subjects who expected
to receive information. Similarly, studies investigating accountability have shown
that subjects who expected to defend and justify their opinions engaged in more
careful deliberation and showed more cognitive complexity than did subjects who
381
did not anticipate justifying their opinions (e.g., Davis, Stasser, Spitzer, & Holt,
1976; Tetlock, 1983). Finally, a developmental study of children's problem
solving also found that the anticipation of having a partner produced conservation
answers that were as highly rated as in a condition in which children actually had
to work out a solution (talk) with a partner (Doise, Rijsman, Van Meel, Bressers,
& Pinxten, 1981, cited in Light & Perret-Clermont, 1989).
These studies point to the conclusion that the anticipation of having a partner
can lead to the same benefits as actually having a partner. This finding is
consistent with data presented here and provides a possible explanation of why
some of the talk alone children in the spaceship study produced proportionally the
same amount of transactive talk as some of the talk dyads. These alones may have
been more aware that their talk would ultimately be heard by the experimenter
and, therefore, maintained the social obligation to be coherent and explicit about
their reasoning. That is, the alones with a high proportion of interpretive talk may
have perceived the situation as being more social than did the alones who
produced less interpretive talk, and adjusted their talk accordingly . In contrast,
talk alones who verbalized few transactive utterances may have perceived the
situation as being less social and, therefore, produced speech that resembles inner
dialog more than external social speech (Wertsch & Stone, 1985). If this
assumption is correct, any procedure that reproduces the social obligation for
coherence that is implicit in conversation should engage interpretive thinking
with or without actual vocalization.
An interesting question for future work is whether language is ever divorced
from the implicit social contract regarding coherence. To test such a hypothesis, it
would be necessary to observe a child engaging in conversation with a partner for
whom coherence is not necessary . Interacting with a computer partner (e.g.,
computer tutoring) is one of the few situations in which natural language is
produced without the presence of a human partner. Some empirical evidence
suggests that conversational norms are changed when language is mediated by
computers (e.g., Sproull & Keisler, 1991). If the computer is the conversational
partner rather than just a medium for communication, would these norms
disappear completely? For example, one can imagine that students may not feel
obligated to explain their thinking or to be explicit about their ideas in response to
computer-based prompts . In this case, the computer prompts would not provoke
the same amount of transactive reasoning as if the prompts were delivered by
human partners. Such evidence would more directly address the issues of whether
the demand for coherence is fundamental to producing transactive reasoning and
if this coherence is an integral part of communication regardless of the actual
presence of a collaborator.
382
Stephanie D. Teasley
Author Note
The studies reported in this chapter were conducted while I was at the Learning
Research and Development Center at the University of Pittsburgh. The work was
supported by funding from the Office of Educational Research and Improvement
(OERI) of the United States Department of Education and the Andrew W. Mellon
Foundation. I am grateful for the cooperation of the schools involved: St.
Bernadette in Monroeville, Pennsylvania, and Lake Hills in Ferrysburg,
Michigan . I am also grateful for the guidance and support of Lauren Resnick over
the course of these studies. Margarita Azmitia and Clotilde Pontecorvo
generously provided helpful comments on earlier drafts of this paper.
Correspondence concerning this article should be addressed to Stephanie D.
Teasley, Colaboratory for Research on Electronic Work (CREW), , University of
Michigan, Ann Arbor , MI, USA. Electronic mail may be sent via Internet to
steasley@umich.edu
383
References
Azmitia, M., & Montgomery, R. (1993) . Friendship, tran sactive dialogues, and the
development of scientific reasoning . Social Development, 2, 202-221.
Berkowitz, M. W. (Ed.). (1985). New directions for child development: Peer conflict and
psychological growth San Francisco: Jossey-Bass ,
Berkowitz, M. W., & Gibbs, J. C. (1983). Measuring the development of features of moral
discussion . Merrill -Palmer Quarterly, 29, 399-410.
Bivens, J. A, & Berk, L. E. (1990). A longitudinal study of the development of elementary
school children's private speech. Merrill-Palmer Quarterly, 36, 443-463.
Cohen, A R. (1961). Cognitive tuning a factor affecting impression formation.Journal of
Personality, 29,235-245.
Damon, W., & Killen, M. (1982). Peer interaction and the process of change in children's
moral reasoning . Merrill-Palmer Quarterly, 28,347-367.
Damon, W., & Phelps, E. (1989) . Critical distinctions among three approaches to peer
education. International Journal ofEducation Research, 13,9-19.
Davis, J. H., Stasser, G., Spitzer, C. E., & Holt, R. W. (1976). Changes in group members'
preferences during discussion : An illustration with mock juries . Journal of Personality
and Social Psychology, 34, 1177-1187.
Diaz, R. M., (1992). Methodological concerns in the study of private speech. In R. M. Diaz
& L. E. Berk (Eds.), Private speech: From social interaction to self-regulation (pp. 5581). Hillsdale, NJ: Erlbaum.
Diaz, R. M., & Berk, L. E. (Eds.). (1992). Private speech: From social interaction to selfregulation . Hillsdale, NJ: Erlbaum.
Dunbar, K., & Klahr, D. (1989). Developmental differences in scientific discovery
strategies. In D. Klahr & K. Kotovsky (Eds.), Complex information processing: The
impact ofHerbert A. Simon (pp. 109-144). Hillsdale, NJ: Erlbaum.
Ellis, S., & Gauvain, M. (1992). Social and cultural influences on children 's collaborative
interactions . In L. T. Winegar & J. Valsiner (Eds.), Children's development within social
contexts (pp. 155-180). Hillsdale, NJ: Erlbaum.
Ericsson, K. A., & Simon, H. A (1980). Verbal reports as data. Psychological Review, 87,
215-251.
Gauvain, M., & Rogoff, B. (1989). Collaborative problem solving and children's problem
solving skills. Developmental Psychology, 25, 139-151.
Giles, H., & St. Clair, R. N. (Eds.). (1985). Recent advances in language , communication,
and social psychology. New York: Erlbaum.
Grice, H. P. (1989) . Studies in the way of words. Cambridge, MA: Harvard University
Press.
Klahr, D., Dunbar , K., & Fay, A L. (1990). Designing good experiments to test "bad"
hypotheses . In J. Schrager & P. Langley (Eds.), Computational models of discovery and
theory information. San Mateo, CA: Morgan-Kaufman.
Klahr , D., Fay , A. , & Dunbar , K. (1993) . Heuristics for scientific experimentation: A
developmental study. Cognitive Psychology, 25,111-146.
Kruger, A C. (1992). The effect of peer and adult-child transactive discussions on moral
reasoning . Merrill-Palmer Quarterly, 38,191-211.
Kruger, A C., & Tomasello, M. (1986). Transactive discussions with peers and adults.
Developmental Psychology, 22, 681-685.
384
StephanieD. Teasley
Kuhn, D., & Brannock, J. (1977) . Development of the isolation of variables scheme in
experimental and "natural experiment" contexts. Developmental Psychology, 13, 9-13.
Kuhn, D., Amsel, E., & O'Laughlin, M. (1988). The development of scientific thinking
skills. New York: Academic Press.
Light, P., & Perret-Clermont, A-N. (1989). Social context effects in learning and testing.
In A Gellantly, D. Rogers, & J. A Sloboda (Eds.), Cognition and social worlds (pp. 99112). Oxford, England: Claredon Press.
Piaget, J. (1965). The moral judgment of the child. Glencoe, IL: Free Press. (Original work
published, 1932).
Pontecorvo, C. (1987). Discussing for reasoning: The role of argument in knowledge
construction. In E. DeCorte, H. Lodewijks , R. Parmentier, & P. Span (Eds.), Learning
and instruction: A publication of the European Association for Research on Learning
and Instruction . (Vol. I, pp. 71-82). Oxford, England: Leuven University Press.
Pontecorvo, C., & Girardet, H. (1993). Arguing and reasoning in understanding historical
topics. Cognition and Instruction, 11,365-395.
Schauble , L. (1990). Belief revision in children: The role of prior knowledge and strategies
for generating evidence. Journal ofExperimental Child Psychology, 49, 31-57.
Schegloff, E. A (1991) . Conversation analysis and socially shared cognition. In L. B.
Resnick, 1. M. Levine, & S. D. Teasley (Eds.), Perspectives on socially shared cognition
(pp. 150-171). Washington, DC: APA Press.
Sproull, L., & Keisler , S. (1991) . Connections: New ways of working in networked
organizations . Cambridge, MA: MIT Press.
Teasley, S. D. (1995) . The role of talk in children's peer collaborations. Developmental
Psychology, 3, 207-220.
Teasley, S. D., & Resnick, L. B. (1993) . Social influences and cognitive change: Peer
collaboration in a causal reasoning task. Unpublished manuscript, University of
Pittsburgh, Learning Research and Development Center, Pittsburgh, PA
Teasley, S. D., & Roschelle, J. (1993) The construction of shared knowledge in
collaborative problem solving . In S. Lajoie & S. Derry (Eds.), Computers as cognitive
tools (pp. 229-258). Hillsdale, NJ : Erlbaum.
Tetlock, P . E. (1983). Accountability and complexity of thought. Journal of Personality
and Social Psychology, 45,74-83.
Tudge, J. (1992) . Processes and consequences of peer collaboration: A Vygotskian
analysis. Child Development, 63, 1364-1379.
Wertsch, J. V. (1985) . Culture, communication, and cognition: Vygotskian perspectives.
Cambridge, England: Cambridge University Press.
Wertsch, J. V., & Stone, C. A. (1985) . The concept of internalization in Vygotsky 's
account of the genesis of higher mental functions . In J. V. Wertsch (Ed.), Culture,
communication, and cognition: Vygotskian perspectives (pp. 162-179) . Cambridge,
England: Cambridge University Press.
Zajonc, R. B. (1960). The process of cognitive tuning in communication. Journal of
Abnormal and Social Psychology, 61, 159-167.
Chapter 16
Abstract
The general issue addres sed in this chapter concerns the relation ship between
perception and discourse. The empirical analyses reported focus on how students and
teachers, working in the context of a physics laboratory in school, communicate about
the properties of light. It is shown that, in order to understand the behavior of light in
the experiments that are set up, one has to have access to elements of a theory of light
that make the phenomena produced appear as significant according to a particular
perspe ctive. Thus, seeing in the sense of identifying something that is culturally and
contextually significant is a sociocultural process that relies on discursive resources.
Introduction
The theoretical background of this chapter reflects an interest in the mediated
nature of human mental and practical action. The study is based on an empirical
field study of student/teacher and student/student cooperation in the context of
the physics laboratory. The specific problem analysed concerns work done in the
area of optics on such topics as light sources, reflection and refraction of light,
and similar phenomena. The teaching and learning activities take place in a toolsaturated environment in which equipment such as the optical bench, prisms of
various kinds, mirrors, lenses, diaphragms , lamps, and a range of other devices
386
are used . Our interest here concerns the extent to which the participants share
perspectives on what is being talked about, and the sense in which they have
access to discursive tools that allow for the instruments to serve as a means of
providing shared - and new - definitions of the nature of light.
In the rhetoric surrounding teaching and learn ing within the natural sciences,
experimentation and hands-on manipulation of physical objects are often claimed
to serve as important vehicles in attempts to ground physical concepts and
explanations in the life-world of learners. Experimentation is seen as a significant
means for the creation of practical and concrete experiences in the teaching of the
abstractions that characterize theoretical concepts and mathematical models of
the physical world in science . The laboratory-like setting in the school offers an
arena in which systematic observations of phenomena can be made in a direct and
supposedly unmediated fashion, and where what is to be learned can be supported
by almost bodily sensations. The students when "using their hands, putting
together and taking apart things ... get much more out of it [the teaching]," as
one of the teachers participating in our study put it (cf. Bergqvist & Saljo, 1994,
p.153).
However, rather than discussing this general issue of the role of child activity
and experimentation in education, we focus on the more limited question of what
is presupposed by individuals in terms of their ability to perceive events in order
to make sense of certain kinds of physical phenomena illustrated in the context of
optical experiments . When "looking" at light, what do we "see"? And, how is this
"seeing" related to sociocultural experience?
The assumption that we perceive directly and by means of a simple registration
of visual stimuli is a prominent component of Western lay and scientific
conceptions of the process of seeing. The physical world is construed as objective
and out there, and, in most accounts, seeing is described as a democratic
phenomenon, as it were; when we look at the world from a similar perspective,
we have the same sensations and retinal images and, thus, "see" the same things .
In order to understand the difficulties that the young people whom we have
observed have in grasping the basic principles of optics when working with the
equipment designed to illustrate some of these principles , it becomes evident that,
ina sociocultural perspective (Wertsch, 1991), perception has to be accounted for
in a radically different way. Retinal images and sensations are part of the picture ,
but they offer very few clues to the situated meaning of what was illustrated in
the particular experiments that we observed. Yet, it was obvious that the teachers
were convinced that phenomena such as refraction of light, reflection angles, and
other similar basics of optics were there to be unproblematically seen by
everyone when looking at the optical bench and the other devices used in the
laboratory. However, from a sociocultural perspective, seeing must be construed
as intimately related to collective and individual experiences that provide a
person with perspectives on what to see in what is out there (Saljo, 1992). In the
optics lab, there is, we argue , a subtle interplay between technological and
discursive tools in the constitution of phenomena to be perceived, and the
387
artifacts very clearly embody distinctions that are discursive in nature. Categories
form part of situated practices as Goodwin (this volume) puts it.
388
cognitive agency outside the stimulus domain" (Rock, 1985, p. 40). At this level,
a broad set of factors that have to do with our experiences, memories , prejudices,
taken-for-granteds, and fantasies can begin to playa role for perception . Thus, "in
moving from biological and sensory processes (. . .) to perception, we have
reached a point in our discussion of psychology where the mind now steps in and
makes contact with the information" (Seamon & Kenrick, 1992, p. 120). The
statement that the "mind ... steps in" is a significant and revealing expression in
this context. The interaction between the individual and the outside world is
generally conceived in terms of the mind as an inner eye or homunculus
analyzing and observing the sense impressions entering the perceptual system . In
most accounts of perception, there is very little room for sociocultural experience
to play any decisive role for perception. The higher cognitive agency alluded to
by Rock is mostly given a very narrow interpretation and is more of a gatekeeping function; "Our senses are our only means of knowing about the world,
but they are not windows that let the world in. They are extensions of the brain
that respond to different forms of physical energy" (Seamon & Kenrick, 1992, p.
119). In the transition from external stimuli, via retinal projections, to neural
processing of the signals, there is generally very little room for shared discursive
tools, what Winograd and Flores (1986) referred to as resources emerging from
"the consensual domain," to enter the picture as mediating devices in perceptual
orientation , which we return to shortly .
Our intention is not to give a full account of theories of perception, but it
should be pointed out that there are radically different interpretations of the
nature of perception that imply different positions with respect to the relationship
between exper ience and perception . One notable example is a phenomenological
perspective of the kind offered by Merleau-Ponty (1962/1974) , which obviously
does not always qualify as a theory of human perception, because it is generally
not given serious attention in main-stream textbooks . In this case, the distinction
between sensation and percept is not accepted as a point of departure. In fact,
Merleau-Ponty argued that "once introduced, the notion of sensation distorts any
analysis of perception" (p. 13). This "empiricist construction . .. hides from us
'the cultural world' or 'human world' in which . . . almost our whole life is led."
(p. 23). In phenomenological interpretations, the topic of inquiry is instead
human consciousness and the intentionality of experience. The problem of what
is objectively out there and what is subjectively perceived, referred to in the
distinction between sensation and percept, disappears as the basic point of
departure for analysis. Instead, the assumption of the primacy of the life-world in
experiential terms forms the background for the understanding of perception (and
other cognitive phenomena).
389
390
391
Segall, Campbell, and Herskovits (1966) and Luria (1976), for instance, showed
how susceptibility to visual illusion cannot be explained by reference only to
physiological phenomena but has to be related to culture. In his studies of the
introduction of literacy among peoples in the southern republics of what was then
the Soviet Union , Luria (1976) concluded that "illusions are not universal" (p.
43) and that the higher the level of formal education, the more likely the
participants in his studies were to perceive illusions.
A different type of work, but documenting what can be interpreted as similar
results, was carried out by de Groot (1965). In his well-known studies of chess
players, de Groot found that there were very marked differences in which expert
and less expert players were able to reproduce the positions of the chess pieces
when being allowed to see the board for brief intervals only. In particular, if the
positions of the pieces could be seen to represent situations that are likely to arise
in a game played by people with considerable skills, players with a high level of
expertise - grand masters and masters - were very accurate in their ability to
describe the positions of the pieces. de Groot argued, "It is evident that
experience is the foundation of the superior achievements of the masters" (p. 329)
in such situations (although this is not necessarily the case with unusual or atypical games [cf. Djakow, Petrowski, & Rudik, 1927]), because highly skilled
chess players are able to pick up "the essential relations between thepieces, their
mobility and capturing possibilities, their co-operation or opposition" (p. 331).
Thus, the "entire perception occurs in fundamental relations and possibilities and
is in a sense dynamic" (p. 331), and the "M[aster] sees more than [the] E[xpert]"
(p. 334, all italics in original). The examples could be multiplied, but the
argument could very well be made that, when considering perception and its role
in social practices, what Rock referred to as "conceptually based, consciously
apprehended information" must playa critical role. We perceive reality by means
of the sociocultural resources that we find applicable to concrete practices
(Wertsch, 1991). The origin of these resources must be sought not in the
biological substrate of human thinking and perception, but rather in discursive
practices that have developed in specific contexts and that contain distinctions
that mediate these practices in rich detail and according to certain patterns.
392
0 U
D u
D
1
I~
13
14
.,
, ,
I ,
C::J Q]6
15
~
9
i..
1\
10
11
c:J - -
"
IS
19
Fi g. 16.1. Geometrical figures used by Luria (1976, p. 33) when studying perception and
categorization.
393
The Setting
In optics laboratory in the school where we have been studying student/teacher
and student/student interaction, the obvious purpose of the activities is to develop
new conceptions of light among the students. Students are to acquire, by means
of what is referred to as concrete experimentation, modes of understanding the
properties and behavior of light that are not necessarily part of their previous
experience. The data used in the following discussion derive from a participant
observation study of work in the physics laboratory among secondary school
students (aged 13-14). Students in two classes were followed during one year
through participant observation. One focus of this research was to study group
work among the students. The particular project we used as a vehicle for
illustrating some of the dilemmas of how light is construed in different
sociocultural practices dealt with some of the basic concepts of optics. During a
period of seven lessons spread out over three weeks, the students were to learn
about phenomena such 'as refraction and reflection, light sources and shadows,
and mirrors and lenses. Important concepts that were introduced and illustrated in
the experiments included inflection- and reflection-angle, and focus and focal
distance. A significant proportion of the time was devoted to hands-on
experimentation using the so-called optical bench, which is illustrated in Figure
2.
lamp
diaphragm
screen
394
In the experiment illustrated here students learn about the refraction of light as it
passes through different substances (cf. Figures 4 and 5 where students'
difficulties in interpreting this experiment are discussed in some detail. For an
illustration of a different experiment using this device, see Figure 3). During the
lessons, two teachers (Anders and Randi) shared the responsibility for instruction .
(Details about the design and the data collected can be found in Bergqvist, 1990).
395
time there is a prism in the middle, and on the screen the lights of different
wavelengths appear.
Anita
Inga
Anita
Eva, Anita
Anita
Inga
Anita
Yees!
Wow ... Wow
Discolights
Yes
The blue one is the
nicest
But what are we supposed
to use this one for?
You probably give an
angle to the light in
some way
Check this, what a nice
light that this!
Nothing is happening it
it seems to me. What is
this then?
Yes, that's right,
that's what we're
supposed to do. You
should point the light
onto here . And then you
would see how the light
is reflected.
This was really boring
No, I don't get it
Me neither . But that's
nothing to worry about
What are you supposed
to do with that thing
then?
What a cool light! He
Eva
Anita
Anita
Eva
Anita
Inga
Anita
Eva
Inga
Inga
Anita
Eva
ANDERS
Eva
... )
(Points to a mirror)
396
rnga
ANDERS
Eva
rnga
Eva
RANDI
397
RANDI
Armi
RANDI
Fig. 16.3. Drawing of optical bench illustrating that light cannot go through solid objects.
398
In commenting on the illustration, she pointed out that the pen, being a solid
object and placed in front of the light beam, would produce a shadow on the
paper. As an explanation of this phenomenon, she added that "light can only go
straight forward and it cannot go through a body" (Randi), and, thus, there will be
light on the side of the pen. This is what the students were to discover themselves
by experimenting with the equipment. When students were told this, they seemed
anything but enlightened. Again we would argue that this is an interesting point
at which the discoordination in perspectives between teachers and students
looking at the same physical event must be understood by reference to
sociocultural resources. For the students, the observation that light cannot go
through solid objects seemed normal and not in need of any explanation. They
were probably also aware that light can only go straight forward. Therefore, well
before they had ever been exposed to this problem in the physics class, they
would be able to predict what would happen in a situation of this kind . The
shadow on the paper would require no special explanation.
For Randi, however, who has been socialized into the discursive communities
of physics and physics teaching , this is a fruitful way of recontextualizing a
fundamental principle in a concrete fashion . She construes this situation as a
context in which basic principles of the properties of light are displayed and there
is a problem to be accounted for. The interesting observation from our point of
view is, however, that this socialization into specialized communities of practice
in a particular field to some extent seems to make it difficult for her to see what
nonmembers can possibly see when facing the experiment. What needs to be
explained in one discursive practice can remain tacit in the other one and vice
versa.
The technology in the laboratory thus already embodies a theory of optics. The
optical bench has been designed on the basis of insights into optics that have been
made during literally thousands of years. In this sense, the equipment is heavily
theory-laden, and it contains forms of knowledge built into a physical device .
399
phenomenon that students were familiar with from other contexts ; and the very
purpose of this inductive teaching strategy is to use the students' personal
experiences as a source for learning (cf. Wistedt, 1994). During the work on this
topic, the students were given the following tasks:
(I) Study and make a set of drawings showing the beams of light going through
different plastic objects, a rectangle , a triangle, a half circle , and check if the beam
rebounds. Make at least two positions for each object. Do the same thing with a lens,
which will make five beams. Make drawings on sheets of paper that you attach to the
screen.
(2) Discuss what happened with the light and why, on the basis of one drawing with a
fairly strong refraction .
Prior to working with these tasks , the teacher had introduced the topic of
refraction of light, among other things, by means of illustrations showing how
light "slows down and turns off' when it passes from a thin substance to a more
dense substance, for example, from air to plastic prism. The illustration made on
the board is reproduced in Figure 4.
Fig. 16.4. Drawing illustrating principles of refraction of light through different substances
with differing optical densities.
400
sand
sno w
Fig. 16.5. Drawing illustrating the principle of refraction by means of analogy with a
sledge moving across different surfaces.
The idea communicated is that the sledge slows down when friction increases as
it passes over the sand and that it gains in speed when back on the snow again.
This concrete reference to a bodily experience would then serve as an analogue to
how light behaves when moving through objects with different densities .
Even though the students had an initial understanding of the fact that light
refracts, their work on the task given is not very successful. Although the students
might in some instances perceive that the light refracts, they have constant
difficulties in constituting this as something that calls for any explanation . In
addition, they find it difficult to move between the two presentations: what they,
at best , would see when they work with the prisms, and the example with the
sledge .
ANDERS
Edvard
ANDERS
Edvard
Ola
Ola
Edvard
Ola
Edvard
Ola
401
A major reason for the difficulties students have is, thus, the explanation offered
in Figure 5 and present on the board during the lesson. The analogy between the
sledge and refraction of light is illuminating only if one already has an
understanding of the concept of optical density. In that case, one can subsume the
slowing down of the light as it passes through a solid object (with higher density
than air) and the movement of the sledge on different surfaces under the same
general principle. However, this connection has to be made on an explicit
conceptual level that codifies the slowing down as a shared feature of the two
examples. What the students at best have at their disposal is an intuitive, yet not
verbalized, understanding tha t light will turn off when it hits objects or
substances (e.g ., water), but they do not understand how this turning off is
represented in the picture through the angle on the line symbolizing the light
beam, and they are not able to grasp what the turning off (which is clearly visible
on the drawing) has to do with the light beam slowing down. Indeed, it is not
clear whether the idea that light can slow down is at all connected to the idea of
refraction for the students. Light simply seems to be there in a straightforward
manner, and there are no signs that the students, while working, think of the light
beam as moving from one point to the other. What is necessary, and what is
available within the community of practice within which the teachers operate, is a
distinct codification of the pri nciples of reflection tha t allows for the two
instances to count as similar . It is the familiarity with what Lemke (1990) refers
to as the thematic patterns of scientific reasoning that connects the two
illustrations as illustrations of something similar. In the students' work, the
movements of the sledge become the major topic of discussion, as is shown in the
preceding example when the discussion centers around how the sledge would
move if the teacher would be the person sitting on it. The connection to reflection
of light is clearly beyond reach at a thematic level.
402
Concluding Comment
These observations illustrate the very intricate way in which sociocultural
experience enters into our perceptions of events . The technology present in the
optics lab embodies perspectives on the nature of light; it is by no means theory
neutral . At the same time, it is reconstructed by teachers as instruments through
which students can experiment and make experiences more or less without
interpretation. We have shown that it is doubtful whether this is the case.
However, the main interest in this chapter has been to illustrate how
communities of practice offer a fixation of perspectives (Rommetveit, 1992) by
means of discursive distinctions . We thus learn to see phenomen a by reference to
conceptual resources that originate in specialized communities in complex
societies. Learning in such contexts implies appropriation of accounts of the
world that are neither out there in the objects themselves nor in our brains.
Rather, they are cultivated in institutional settings for particular and sometimes
highly specialized purposes. This implies that learning should not be reduced to a
matter of acquisition of information that is available. Rather, we have to consider
the decisive role of the social distribut ion of different forms of knowledge and
how knowledge is made available to groups and individuals in modern society.
The difficulties that our students have in understanding the principles of light,
and precisely how these are illustrated by means of the tools utilized in the
experiments, have less to do with problems of acquiring knowledge in a narrow
sense or with deficits in conceptual understanding and more to do with their lack
of access to communities of practice in which light is perceived and construed in
the way characteristic of physics. Thus, we might guess that most of the students
we have followed are not participants, not even peripherally (Lave & Wenger ,
1991), in cultural practices in which it makes sense to talk about light using
concepts from optics. On the contrary, in most contexts they would probably run
into problems of making themselves understood, should they try to do so.
A corollary of these observations is the profound extent to which institutions
socialize members into living in only partially shared social realities
(Rommetveit, 1988, 1992). Psychological and cognitive development in complex
societies thus is fundamentally dependent on the opportunities to participate in
institutionalized communicative practices in which specialized accounts and
ways of construing phenomena are cultivated. Thus, to the teachers in our studies,
the basic phenomena of optics illustrated in the laboratory settings were seen in a
very real sense; they were assumed to be there to be perceived by any observer
looking at the equipment. The decontextualization practices (Gustavsson, Linell,
& Saljo, 1993) offered by school subjects and scientific accounts provide filters
that become part of sociocultural resources when perceiving. Socialization of
members into communities of practice is often so effective that it becomes very
difficult to realize in what sense one's perception of an event is dependent on
conceptu al resources and distinctions that are not available to outsiders. We learn
403
The research reported here has been financed by the Swedish Council for
Research in the Humanities and Social Sciences. This chapter was written while
the first author was a Fellow at the Swedish Collegium for Advanced Study in the
Social Sciences (SCASSS) in Uppsala.
404
References
Atkin son , R. L. , Atkinson, R. C ., Smith, E. E., & Bern , D. J. (1990). Introduction to
psychology (10th ed.). New York: Harcourt Brace Jovanovich.
Bergqvist, K. (1990) . Doing schoolwork. Task premises and j oint activity in the
comp rehensive classroom. Link oping : Linkoping Studies in Arts and Science , 55.
Bergq vist, K., & Salj o, R. (1994). Conceptually blindfolded in the optics laboratory.
European Journal of Psychology ofEducation, 9(2), 149-158 .
Cole , M., Gay , J., Glick, A., & Sharp, D. W. (197 1). The cultural context of learning and
thinking. New York: Basic Books.
Djakow, Petrowski, & Rudik (1927). Psychologie des Schachspiels [Psych ology of ches s
playing] . Berlin: De Gruyter.
Edwards, D ., & Mercer, N. (1987). Common knowledge. The development of
understanding in the classroom. London: Methuen.
Gr aaf, W . de , & Maier, R. (Eds .). (1995 ). Sociogenesis reexamined. New York: Springer.
Groot , A. de . (1965 ). Thought and choice in chess. The Hague : Mouton.
Gu stavsson , L. , Lineil, P., & Salj o, R. (1993 ). Discourse on language and discourse in
language . International Journal of Educational Research, 19(3), 265-276.
Harr e, R., & Gillett, G . (1994). The discursive mind. London: Sage .
JOnsson , L., Linell, P., & Saljo, R. (1991 ). Formulating the past. Rem emb ering in the
police interrogation. Activity Theory, 9/10, 5-11
Koffka, K. (1935 ). P rincipl es of Gestalt psychology. New York: Harc ourt Br ace
Jovan ovich.
Lave , J. (198 8). Cognition in practice. Cambridge, England: Cambridge Univer sity Press.
Lave, J. , & Wenger, E. (199 1). Situated learning: Legitimate peripheral participation.
Cambridge, England: Cambridge Universi ty Press .
Lemke, J. (1990). Talking science: Language, learning, and values. Norwood, NJ : Ablex.
Luria, A . R . (1976) . Cognitive development . Its cultural and social fo undations. [M .
Lopez-M orillas and L. Solotar off, Trans.] Cambridge, MA : Harvard Uni versity Press.
Me rleau-Ponty, M. (1974 ). Phenomenology of perception. Lond on : Routledge & Kegan
Paul. (Original work published 1962)
Middleton, D., & Edwards , D. (Eds.). (1990 ). Collective remembering. Lond on: Sage .
Nei sser, U. (1967) . Cognitive psychology. New York : Appleton-Century-Crofts .
Neisser, U. (1976 ). Cognition and reality. San Francisco: Freeman.
Rivers, W . H. (1901). Primitive color vision . Popular Science Monthly, 59, 44-58 .
Rock, I. (1985). The logic of perception. Cambridge, MA : MIT Press.
Rommetveit, R . (1988) . On literacy and the myth of literal meaning . In R. Sii1jo (Ed .), The
written world (pp. 13-40). New YorklBerlin: Springer.
Rommetveit, R. (1992). Outlines of a dial ogi cally bas ed social-cog nitive approach to
human cog nition and com municatio n . In A. Heen-Wold (Ed.), The dialogi cal
alternative - Towards a theory of language and mind (pp. 19-44 ). Oslo : Scandinavian
Uni versity Press.
Siilj o, R. (1992) . Hum an growth and the complex society. Cultural Dynamics, 5( 1), 43-56.
Seamon , J. G., & Kenrick , D. T. (1992). Psychology. Englew ood Cliffs, NJ: Prenti ce Hall.
Segall, M . H., Campbell, D. T ., & Her skovits, M. 1. (1966 ). The influence of culture on
visual perception. Indi anapolis, IN: Bobb s-Merrill.
Shotter, J. (1990). Knowledge ofthe third kind. Utre ch t: ISOR.
405
Chapter 17
Abstract
The aim of the study is to identify both modes of children's participation in family
disputes and types of argumentative moves adopted, particularly in the act of
opposing (problematizing) others or defending oneself. The corpus consists of
twenty-seven dinner conversations of ten middle-class families living in Rome
and Naples, each with one child between three and six years and at least one older
sibling .
Data are shown concerning relative distribution of six different discourse genres
(according to temporal focus and presence of problematic events : Ochs & Taylor,
1993) and family members ' role in problematization. A qualitative analysis
illustrates ways in which children are involved and act in family disputes. The
quantitative results indicate that the problematizing activity occupies about 1/3 of
family talk, allowing children peripheral participation in conflict talk; 1/2 of the
problematizations are directed to the children . When discourse concerns past
events, children show a lower rate of problematizing activity (31.3% vs. 40.9% of
the whole of conflict talk) but - when challenged on their past behavior - they
appear to have already learned at 4 or 5 years how to justify themselves and to
provide rhetorically designed answers, using appropriate temporal markers,
authority references, and visual recall devices. Children 's orientation to social
and/or family norms and values is also discussed.
407
Theoretical Background
Exploring learning processes through family talk requires laying out some
theoretical premises, although increasing numbers of scholars have recently
focused on natural interactive contexts for understanding the acquisition of
competences . In this view, cognition and knowledge acquisition are considered the
products of collaborative, repeated, and regulated activities within social domains.
Cognition is seen as a property both of persons and of the relationships between
them and also of the local rules and affordances of the setting: in a word, as a
property of situations in a Goffmanian sense (Goffrnan, 1964; 1974).
School, family, and academic institutions are places where particular
sociocognitive practices are active, part of which is represented by the specialized
types of discourse that circulate within them. By looking at the discursive activity
taking place in a specific setting, we have a key to understanding both the kind of
social situation that we are observing and the nature of the competence that it
demands of its members.
Goodwin's (1994) studies of apprenticeship in the field of chemical and
archeological research have shown how the role of the expert is that of monitoring
and correcting the procedures used to deal with the specific reality with which the
apprentices are presented. The accomplishment of activities within these reality
domains is the joint product of information and actions brought forth by different
persons and applied to the concrete objects they are manipulating.
From a developmental point of view, the acquisition of children's cognitive
capabilities is supported by the scaffolding of adults (Bruner, 1986), so that adult
and child constitute a cognitive system in which one's developed skills are at the
service of the other's purposes (Kaye, 1982). This view can be located in a
Vygotskian framework that considers higher psychic functions as internalizations
of social actions through the mediation of language (Vygotsky , 1934/1990) . In a
sense, the very definition of what psychic functions are (e.g., problem solving,
memory, reasoning) emerges from the discourse that sustains activity (Resnick,
1991). Discourse analyses in educational settings such as school and home have
shown that what gets transmitted is not abstract information but a set of rules on
the right methods for achieving proper knowledge (Edwards & Mercer, 1987;
Orsolini & Pontecorvo, 1992); for displaying memory functioning (Edwards &
Middleton, 1988); and for building up sound narratives (Ochs, Taylor, Rudolph,
& Smith, 1992; Pontecorvo, Amendola, & Fasulo, 1994).
Children's mastery of conversational practices is a basic requirement for their
participation in social life, and it implies far more than learning word meanings
(Ochs, 1988). Talk-in-interaction is a finely organized practice with properties of
its own that risk being neglected by conceptualizing it as "the serial
externalization of some joint arena of batches of talk, hatched in private (or even
socialized) intentions, and filled out with the docile artefacts of 'language'"
(Schegloff, 1989, p. 140). The conversational machinery provides opportunities
408
for delivery of contributions that the others can acknowledge, ignore, or treat in
different ways, whereas the structure of participation is regulated by the turntaking organization so that every turn at talk must be seen as the joint product of
speaker and listener(s). The management of turn-taking, as well as the
organization of sequences and preferences (Sacks, 1987; Pomerantz, 1978; 1984;
Schegloff, 1990), is an ability that children acquire by various degrees of
participating in interactive talk.
Basically, talk is a means of doing things. Learning it, therefore, makes it
possible to carry out negotiation and to pursue others' approval by building
arguments requiring conversational , rhetorical, and cognitive skills (Duranti &
Goodwin, 1992). Arguing , then, is clearly a demanding task in terms of
perspective taking and reasoning; nevertheless, it occurs very frequently among
children (Corsaro & Rizzo, 1990; Pontecorvo, 1993), with peers (Garvey, 1993),
parents (Caron, 1990), and teachers (Pontecorvo, 1987). Arguing requires walking
on the edge between cooperation and competition (Schiffrin, 1990), defending
one's position by proposing evidence that can be deemed acceptable by the other
conversants. In her analysis of everyday arguments, Schiffrin (1985) remarks on
how the defense of one's stance, even when relying on personal experiences, needs
to replicate to some extent the culturally salient meaning of the debated issue.
That is to say, there is a background of conventional knowledge about how beliefs
can be related to facts in such a way that the connection can be acknowledged and
accepted.
A good argumentative strategy is offered by categorization, that is, the
definition of an event/object as a special case of a given class or as a new/different
case, according to the developing rationale of the speaker. Billig (1987) points out
that this process lies at the core of the argumentative activity. In such a view,
categorization is not a once-for-all, internal assignment of objects to classes, as
conceptualized in the cognitive theories; it is a social endeavor, based on the
speaker 's stance and adapted to his or her discursive aims. It is precisely when the
assignment to a category is ambiguous and problematic that arguing and thinking
start to operate, and divergence in assignment can be the issue at stake between
two speakers.
Categorization and definition are frequently used in children's disputes about
scientific and historical topics (Pontecorvo & Girardet, 1993), as an instrument
for stating a position in an ostensibly neutral and descriptive way. It has been
observed that defining and categorizing are ways of persuading listeners by the
artifactual construction of evidence. Naming is the basic form of categorization,
and there is a sense in which the use of a word or an expression for reference
always singles out a specific dimension from the area of possible meaning covered
by that word or expressi on. Learning to argue is, in an important sense , learning
to categorize according to the context-specific presuppositions and the
interactional goals one is pursuing (Edwards, 1991).
Interactive mechanisms of conversation, such as sequential functions, are likely
to offer some of the central learning mechanisms operating in discursive settings
409
in which adults and children are involved. The sequential organization of talk
(Levinson, 1983), with the set of expectations that it projects, provides a
powerful scaffolding for understanding social meanings and for the transmission of
knowledge.
In a context of disagreement and challenge, accounts or justifications are
expected, and these can be accomplished by different types of discourse products,
such as narratives, descriptions, definitions, and causal explanations, according to
the specific topic at hand or to considerations concerning the audience.
A crucial factor is the disputability of the topic or of the social context that
permits disagreement, opposition, and conflict talk (Grimshaw, 1990) . The
pragmatic function of justifying opposition forces the speaker to reconsider his or
her own previous claim and to turn it into an articulate discursive position. In
classroom discussions (Orsolini & Pontecorvo, 1992), children can be asked to
make their positions explicit, and then they offer adequate explanations .
But how do young children learn to argue? How are they socialized to the
relevant tools for answering and initiating arguments? How do they develop the
reasoning skills? How do they practice arguing in interactional settings in which
adults aim to transmit rules, knowledge , and ways of thinking and behaving?
Family discourse provides interactional settings that precede and accompany
school experiences and where sociocognitive exchanges are carried out by
participants who share a high degree of reciprocal knowledge and of customary
practices. In their heuristic research work, Elinor Ochs and her colleagues (Ochs &
Taylor, 1992a) have analyzed family dinner conversations, showing how versions
of past events are subject to continuous revision and new display of affect by the
participants. In a study on Italian family dinners (Pontecorvo, Amendola, &
Fasulo, 1994), we also found that narrative activities were accomplished with the
contribution of more than one member , with members other than the main
narrator being entitled to do the telling even when they had not been present on
the scene of the narrated facts. Such observations have drawn us to engage in a
deeper analysis of family arguments revolving around the past.
Methods
Aims of the Study and Research Questions
410
Data were collected in the natural context of family dinners. Ten two-parent
families participated. Our selection criteria were the presence of one child between
three and six years of age and at least one older sibling. The families were middle
class and were living in Rome and Naples.
Each family was videotaped from three to four times during family dinner
within a 20-day period. The first videotaped dinner was not used for research
purposes. In most cases, the researcher left the camera with the family and was
not present during recording.
The corpus consisted of 27 dinner conversations of the ten families. The
conversations were fully transcribed using the CHAT transcription method of the
CHILDES system.1 Participants' activities were observed and fully transcribed,
considering (a) the verbal utterances (including repairs, interruptions, overlaps);
(b) the paralinguistic aspects (such as intonation, stresses, pauses, laughs); (c) the
nonverbal behaviors (such as movements, gazes, mimicry).
Transcriptions and coding have been subject to cross-control by at least three
researchers.
411
412
needed. This is the operational criterion according to which both the problem and
the relative roles are identified.
The crossing of the temporal dimension with the presence or absence of a
problem gives rise to six types of discursive genres. Table 1 shows the
proportional distribution of the topics within each genre.
Table 17.1. Distribution of the different genres according to temporal focus and
problematization (%)
TEMPORAL
FOCUS
PROBLEMCENTERED
NON-PROBLEMCENTERED
PAST
SIDRY
REPORT
7 .2
11 .8
TROUBLESHOOT
BROAOCAST
21.4
47.3
PLAN
AGENDA
PRESENT
FUTURE
TOTAL
TOTAL
19 .0
68 .7
5 .5
6.8
12 .3
34 .1
65 .9
100 .0
The predominant temporal focus in the dinner conversations we studied was the
present (68.7%), followed by the past (19%), and then thejUture (12.3%) . This
overall result shows that the bulk of family conversation is devoted to the ongoing activity of the dinner (although the topic is not always food). Most
conversational topics are non-problem-centered (65.9%). However, the 34.1 % of
problem-centered narratives are particularly relevant because of the conflict talk
that the problematic rendering of actions, conditions, thoughts, and feelings can
induce.
Because the aim of this study is to describe how Italian children practice arguing
and reasoning competencies within the family learning environment, we gave
special attention to the ways in which children challenge other members or build
their defense when questioned.
413
Quantitative Results
Discourse Roles
Following the analysis proposed by Ochs, Taylor, Rudolph, and Smith (1992)
and by Ochs and Taylor (1992b), we identified, within each topic, the discursive
roles of Protagonist (the leading or principal character whose actions, conditions,
thoughts, and feelings are put "on the table"); Introducer (the participant who
makes the first move to open a topic); and Addressee (the participant to whom the
Introducer predominantly addresses discourse) (Pontecorvo, Amendola, & Fasulo,
1994).
In this study, we focused mainly on the ways in which the two discursive roles
of the Problematizer and Problematizee are implemented both by adults and by
children. The Problematizer is the participant who renders problematic an action,
condition, thought, or feeling of a protagonist or narrator. The Problematizee is
the participant whose action, condition, thought, or feeling is rendered
problematic. Actions, thoughts, or feelings may be problematized on several
grounds. They may be treated as untrue or doubtful, or they can be interpreted as
signs of incompetence or as being inappropriate on many different grounds.
We report one excerpt from a family discussion concerning a problematic topic
raised in the present to show to what these roles correspond. Topics that are
problematic in the present concern things that children or other members are
required to do hie et nunc: As in this case, food and nutrition are often in the
foreground.
Excerpt 1
Gravina family : String beans2
Dad: Paolo; Mum: Antonella; Son: Riccardo (Ricky), 7 years old; Daughters: Tiziana,
21 years old ; Silvana, 19 years old
[Arrows after the name indicate problematizing turns; arrows under the name indicate
consequent moves of Problematizee]
nun che buoni ' s t i fagiolini, vero
these string beans are good aren't they?
Ricky-> solo parcha me l i vuoi far mangiare
only because you want me to eat them
Tiziana : [aho! I
heyl
Mum:
2In all the excerpts, the words in italics, which are reported after the invented name of
the family, refer to the name of the specific topic and indicate the shared object of the
piece of talk .
414
Mum:
->
n o , ' cause
they are
really good
Dad:
Ricky
of them mysel f
eh!i
Silvana : sai che t u sei il campione, ti devi nutri!re
Mum :
champ ion,
you have
to
The problematization here comes from the younger child, who attributes to
mother 's positive evaluation of the vegetables the hidden intention of inviting
him to eat them. The problematic character of Riccardo's claim is evident from
the reactions elicited: Mother and the two sisters simultaneously start talking, the
younger sister with just a reproaching interjection presumably addressed to the
lack of respect shown by the child. The older sister admits the rightness of her
brother's attribution ("indeed") but repeats the mother's evaluation, whereas
mother rejects the child's accusation by stating that her comment was only due to
the goodness of the food.
This density of speaking is typical after problematizing turns, usually
displaying different alignments of the family members with regard to the "hot"
topic. The initial alignment takes place at the turns immediately after the child's
problematization (Tiziana: "hey!"; Mum: "no"; Silvana: "indeed"). Silvana later
adds an element to show her brother that his attack was unjustified (i.e., that
mother was talking to father and not to him) . And father readily agrees,
announcing his intentions of eating all the vegetables. In response, the child
reveals that he understands that he is the intended recipient of father's little show,
by remarking that he does not care. Again the child is warned, this time by
mother, to be more respectful. Silvana eventually cheers him up with reference to
his swimming activity, at the same time confirming the family intention of
making Riccardo eat and providing a reason for that.
We can observe here the complexity of the cognitive task posed to the child: He
receives both denial and confirmation of the content of his claim, plus repeated
reproaches to the way he addresses his parents. In these few lines, the social
organization of challenging authority is visible, as is the peculiar fact that the
offended authority does not defend himself or herself, but this is done by another
member of the family, who is lower in the hierarchy (the sister for the mother,
the mother for the father).
415
We now focus on the distribution of the family members in the discursive roles
of Problematizer and Problematizee (see Table 2).
PROBLEMATIZEE
FATHER
26 .1
18.2
MOTHER
29 .8
24.2
CHILDREN
40.9
48.6
As expected, children tend to be more the Problematizee (48.6%) than are parents
(42.4%), but the difference is not very large. Furthermore, children (at least two in
each family) are the Problematizer significantly less often (X2= 4.93, p<.03) than
are parents (40.9% vs. 55.9 % of the parents); however, they assume this role
more often than we had expected on the basis of similar American data (Ochs &
Taylor, 1992b) . In other words, Italian children problematize parents more than
American children do.
PROBLEMATIZEE
FATHER
25
16.1
MOTHER
37 .5
14.3
CHILDREN
31.3
48 .2
In this discourse context, children problematize much less (31.3%) compared with
the general distribution presented in Table 23 and with a ratio of 1 to 2 compared
with parents (62.5%) (X2 = 39.82, p< .OOOl). The increase in the parents' role of
3The proportion of children 's problematization slows down from present to past, with
future in the middle ; that is, Incidents = 44.5; Plans = 39.5; Stories = 31.3 .
416
problematizers is totally due to the increase in the mothers' share. Mothers , who
already were higher than fathers in problematizing (29.8% vs. 26.1%), increase
about 8% here. Overall, more than one third of the problematization in the story
genre is done by the mothers.
These results differ from those obtained in similar American research (Ochs &
Taylor, 1992b) where problematizing seemed to be done mainly by the father. In
the Italian families, we observed that there is more problematizing activity carried
out by children and by mothers, frequently in oppositional talk reciprocally
addressed. If we consider "who problematizes whom" (see Table 4), there emerges
a possible correspondence and reciprocity between the problematizing talk of
mothers and children.
TABLE 8.4. WHO PROBLEMATIZES WHOM IN THE PAST? (%)
PROBLEMATIZER
PROBLEMATIZEE
FAlHER
CIffi.DREN
OTlUERS
MOTlUER
SElJP
13 .4
6 .2
4 .5
CIffi.DREN
FAlHER
OlHERS
SELF
17 .9
10 .7
8.9
SmUNGS
12.5
9.8
4.5
3.6
(25)
M01HER
(37 .5)
CIffi.DREN
(31.3)
MOlHER
FATlUER
OlHERS
SELVES
0.9
0.0
0.9
[OlHERS
(6.2)]
417
418
the script andlor the categorization of the recounted facts. The discourse context
laid down by the challenge, although constraining the possibilities of
argumentative response, also offers elements than can be used for building the
defense.
The following is a qualitative analysis of a set of stories extracted from a corpus
of 75. The selection is meant to reproduce instances of different patterns of
participation and argumentation.
(a)
Peripheral participation
(b)
Challenge
1) being drawn in by others
2) building on another's introduction
3) introducing and challenging
(c)
Defense
1) redefining and recategorizing events
2) counter-opposing the Problematizer
Learning to Challenge
Peripheral Participation
Children are frequently the silent protagonists of parents' conflict talk. They are
referred to in the third person, although present, and their role is that of peripheral
participants (Lave & Wenger, 1991). We also find this kind of talk in the
aftermath of parent-child discussions, when parents continue the discussion
between themselves, often debating educational-organizational decisions (see
following excerpts : Violin and Coke).
In Excerpt 2, the child is used as prosecution evidence of guilt for which both
parents attribute responsibility and carelessness to one another.
Excerpt 2
Soldano family: Hands
Dad: Vittorio , 35 years old; Mum: Raffaella, 35 years old; Grandmather; Grandfather ;
Sons: Gianluca, 4 years old; Stefano, 8 years old; Daughter: Marina, 3 months old
Mum : - >
Dad:
si rna (
yes
419
but ...
Dad:
->
well
you
said it here at
the table
In the preceding short exchange between the two parents, about the dirty hands of
the child, they each discharge their own responsibility for washing the child's
dirty hands , and they each justify themselves by attributing the fault to the other.
The argumentative strategy used by the two parents is similar and is based on
displacing in a former past (as shown by the repetitive use of the declarative and
perceptual in past forms of verbs: see, say, hear) the negligent behavior of the
other and the origin of the present problem.
Mum: - >
save the
small pizza
eaten the
cake.
save
things
for
him
for daddy
420
Dad:
Mum:
save- which
(h)
Grandpa:
Dad :
[hahahahah
huhuhuhuh
[ah ve:di
Uh you see,
she ate my
small pizza
my-
Mum:
our
[e salata (
that
at
the pastry
are too
turning
salty
to
Granpa
Dad :
->
but
small
pizza.
small pizza.
There was
only
ate the
cake.
Grandma :
but
Dad
Dad :->
for
breakfast
didn't
you
leave
him anything
dinner?
[a chi?
to whom to Grandma
Grandma points toward Dad with her head))
vedi che gentili m'hanno lasciato la crostata
look how kind they were:
Mum:
for
Mum:
there
they left me
the
leave
the
dinner
for Vittorio?
cake
421
Excerpt 4
Selci family : Gigi
Dad: Francesco; Mum: Monica; Aunt: Stefania ; Daughters : Manuela, 10 years old ;
Federica, 7 years old; Francesca, 4 years old; Son: Alessio , 1 year old
Mum:
Aunt :
Gigi dov'era?
where was Gigi?
a casa .
at home.
Mum :
immancabile eh
Aunt :
as always uh?
mhm .
Mum finishes
up
Federica) )
Federica : [grazie mamma
thanks
Mum:
mum
[col cane?
with his
Aunt:
dog?
eh .
yeah
Federica : Manuela !
Mum:
e col gatto?
and with the cat?
Aunt :
Aunt's
[vanno d'accordo]
they get
along well
cat
422
Manuela:
->
Mum :
Stefania do
you
like
Gigi?
che dici?
what
are
Aunt:
( (shakes
Manuela:
a me no
Aunt:
no?
you
saying?
head
as
in
"so
and
SOH
don't
no?
not
Aunt:
to
say that
->
he
Gigi
is your father
is not my father
Federica: strano
Aunt:
Manuela :
strange
mh
Aunt :
Manuela:
he
mamma
Rori se vuole
mamma Rori
Aunt :
Mum:
didn't
no .
wants
sposa z
to
get married?
no .
[e fidanzata
she's
engaged
Manuela:
[Gigi?
Federica : invece si e sposato con
he
mamma
Rori
In this story, the two older sisters use a joint strategy for problematizing the
young aunt. Their topic is the live-in partner (Gigi) of the aunt's mother (mamma
Rori), and they like talking about this delicate issue.
The sisters use the apparently neutral format of asking informative questions,
offering personal evaluations, and stating matters of "fact." It becomes evident
that they know very well the answers to their questions and the falseness of their
asserted facts (the aunt's mother and her partner are not married, and the partner is
not the aunt's father) . The sisters' goal of challenging their aunt's known but
unadmitted view about the situation is highlighted by the mother's first attempt
to stop Manuela (what are you saying?) and then giving a socially acceptable
definition of the relationship (she's engaged).
A similar situation in which the two sisters cooperate in problematizing adults
occurs in the following excerpt drawn from conversations of the same family. The
youngest daughter introduces a topic about her baptism and seeks information
about her godmother; then the oldest daughter (who is ten) problematizes the
mother's choice of her godmother by questioning the choice of the person (who is
not highly valued by the mother herself) and complaining about being held by an
423
unknown person . Federica uses the opportunity to criticize the mother who,
according to the girl, was very badly dressed at her baptism.
Excerpt 5
Selci family : Baptism
Dad: Francesco ; Mum: Monica; Aunt: Stefania; Daughters: Manuela, 10 years old;
Federica, 7 years old; Francesca, 4 years old; Son: Alessio: 1 year old
Federica:
Aunt:
it
godmother of what?
(4 .0)
Federica :
Mum:
of
Aunt:
baptism
di battesimo?
of
baptism?
Mum:
Aunt :
Mum:
[te e Ugo] .
you
Aunt:
[ah
e vero] si
oh yes
Manuela:
dear ...
di battesimo .
that's
right
(2 .0)
Mum :
Manuela : - >
better to
lose than
friend
for me
Mum :
->
because
Manuela :ah.
->
Aunt :
Manuela:
ah ,
mamma!
[Marilu ?]
mummy
Mum:
->
424
Mum:
Prancesca
clean your
hands
with the
napkin
Mum:
Federica:
for my
bapt ism
As in the preceding example, when children challenge their parents, they often
advance complaints about the ways in which they have been treated. It is a kind of
appeal to children' s righ ts, which have not been properly attended to by the
parents.
This is also the case in the next story, a very short one in which the seven-yearold daughter questions why her father has not taken her to a nice small woods
where she would have liked to go with him.
Excerpt 6
Selci family : Nice small woods
Dad: Francesco; Mum: Monica; Aunt: Stefania; Daughters: Manuela, 10 years old ;
Federica, 7 years old; Francesca, 4 years old; Son: Alessio, 1 year old
Dad:
we n t
to get
so il
Dad:
the
small
woods
g rown?
eh qua vicino
just
in
the
surroundings
you
t el l
me;
wou ld
have
come
never went
Aunt:
there
in all my life
Dad:
stasera, andiamo,
( .) un'altra volta
Aunt :
this evening we
mhm
Dad:
->
when I
go
(1. 5)
go
for t h e s e
things
feel
like a
t hief
425
In the last two turns, the father justifies himself both by prormsing and by
furnishing a kind of justification by evaluating his uneasy state of mind in going
to the wood and taking some soil.
The stories of this particular family (Selci) show a high proportion of child
problematization. The children, three plus a baby, are problematizers four times in
ten stories and are often successful in constraining the adults to justify
themselves.
Excerpt 7
Fanaro family: Brood Hen
Dad: Silvano , 34 years old; Mum: Teresa, 34 years old; Daughter: Stefania, 5 years old;
Son: Sergio : 7 years old
Sergio :
went
Mum
ha f ame?
gets hungry
Sergio :
->
Sergio:
swimming and
well you
you were
swimming.
~natra.
their
on one of those
426
Marmna. :
Sergio:
Dad:
chair he
looked
like
duck
chi io ?
who me
Sergio:
eh . stavi cosi
ah you were going like this stretching his
neck and turning his head left and right
Da d :
->
of
course
had
had to
do you
Dad :
stand ing ,
never
in
the water
you had a
better time
than
in Rome
alone
Dad:
Mum:
really don't
you
Da d :
think
really don't
so
thi nk
so
don't
Mum :
Dad:->
would have
is to be at
home alone
alone
Here, the child teases his father for his ridiculous position at the beach, describing
the scene and even imitating him to the amusement of the whole family (in his
second turn, he addresses his mother ). The father reformulates the metaphor,
changing it from duck to brood hen, a smart move that has the effect of
427
transforming an image of voluntary and lazy sitting into one of forced and busy
watching. At the same time, it introduces a reason for the complaint about the
mother's absence.
In what follows, the ironical dispute is transferred smoothly to the two parents,
whose topic is who "suffered" more during summer : the mother, who was at
home alone and who worked, or the father, who looked after the children at the
beach. In the last part of the story, "other voices" (Fasulo, this volume) of the
people are reported by both, as a backing for the counter-opposed interpretation of
the situation they are proposing. In this last part, the children are exposed to
alternative and conflicting ways of reconstructing and evaluating the same
situation in the past. Through peripheral participation, they are exposed to a nice
dramatization of parents' sacrificing for children's good.
Learning to Defend
When children are the direct object of challenge, they can decide to defend
themselves. We have identified the two main ways of doing this: offering an
alternative interpretation and thus denying the content of the challenge, or counteropposing the Problematizer. Both strategies can involve reconstruction of past
events .
Excerpt 8
Fanaro family : Quarrel
Dad: Silvano, 34 years old; Mum: Teresa, 34 years old ; Daughter: Stefania, 5 years old;
Son: Sergio : 7 years old
MLun :
da Daniele e Fausto ?
with
Sergio:
Daniele
and
Fausto?
bene !
no we had a
mhrn
MLun :
mhm
t u e Sergio?
you
and
Sergio?
428
Stefania :
Dad:
ell
eh.
[corne rnai?
how was
Mum:
it?
[e perche?
and why?
Dad:
Stefania:
speak
->
louder
because: ::
Stefa
because
he
was
joking and
the
two
Stefania:
eh
Dad:
Stefania:
to
you?
as
one-year-old
children
then?
mess with
Sergio :
->
yes it looks as
she was doing
though she
is
telling what
guitar the
Mum:
->
well
you
have not
today
figured
being
tovagliolo [(
the
->
the
have told
truth
soprattutto Fausto
in
Mum :
)]
napkin
Stefania :
Sergio:
429
particular
Fausto .
not
understood
this
Here Stefania, the younger child, expresses opposition to her brother's laconic
account of the morning's events by initiating her turn with a no, followed by a
radically different account. The parents, speaking in complete overlap, ask for
clarifications, and Stefania answers by recalling some "facts" that clearly put the
responsibility on the brother's and cousin's side. In support of her thesis, she
offers the evaluation of another adult (their aunt) who got angry with them. She
builds a report clearly designed to obtain the consensus of her parents (i.e.,
recipient-designed: Schegloff, 1989). Because her mother asks a question and her
father makes a comment, both seemingly accepting her version (Mum: "why?
what were they doing?" and Dad: "just like one-year-old children"), the brother
reacts by challenging his sister's version as biased and introducing a new piece of
evidence about her questionable behavior. The parents, summing up the reports in
a reproachful tone addressed to both children, provoke a change in Stefania's
strategy: leaving aside the past facts, she casts a positive light on herself by
defining her present behavior as praiseworthy, using an appeal to a more general
set of rules ("at least we told the truth"). Although she includes her brother in this
clever escape, she was the one who decided to tell the parents about the quarreling,
despite Sergio's reassuring answer .
From the children's talk, we can infer that they are completely aware that
telling is not an innocent activity or a straightforward mirroring of reality. The
children know that what they say can meet the sanction of parental judgement, and
this knowledge is detectable as a built-in feature of their accounts. Stefania's last
move shows a deep understanding not only of the moral values but also of their
hierarchical order: Telling the truth is a morally superior value to behaving
properly, something that is also implicit in the Catholic culture, where the
practice of confession warrants forgiveness for one's sins.
430
In the next excerpt, a new and different definition of the problematic situation is
offered by a four-year-old girl who is problematized by her older sister. The
mother supports the claim of the sister, although apparently keeping a more
neutral role. The justification strategy of the younger child is to redefine the
situation by partially keeping the same objective tone used by the mother. In this
very short story, the redefinition of the situation is an epistemic operation carried
out by the four members of the family, who each look at the same event from a
different point of view.
Excerpt 9
Bianco family: Police
Dad: Giovanni, 42 years old; Mum: Sara, 33 years old; Daughters: Silvia, 4 years old;
Fabiana , 9 years old
tell
us
precisely what
you were
doing
Fabiana
came up
Dad:
()
Silvia : stavamo giocando a polizia e poi uno c'ha
->
we were playing police then someone
disturbato ch'eri tu .
came to disturb us who was you
about
(4.0)
Dad:
left
nap
Fabiana: no : :
Mt.un:
a dis fare i letti
to undo
the
beds.
431
The child responds to her mother's challenge with a report on the activity she
was engaged in with her friend Clara: "we were playing police ." This is an act of
categorization that contains a justification. Fiction games have features of their
own. They have to be consistent with the reality domain they are assumed to
imitate: If you play police, you cannot sit down and talk quietly. The temporal
construction is also significant: The play is rendered with an iterative form, and
the precipitating event is represented by Fabiana's coming in and disturbing ,
reported with a punctual verb tense. By this temporal construction, the
problematic focus is transferred from the activity to the interruption and from one
daughter to the other. Fabiana, called into question, substantially changes the
picture, defining her entering as the providential avoidance of an impending
dangerous event: namely, her young sister's breaking her head.
As in the preceding example, we have here a contest about different versions of
a past event in which the contending parties were both present, with the
problematized party objecting to the format of the report, as shown here by the
ironical reported speech that follows the redefinition ("I disturbed"). A parallel
exchange takes place in the following turns between father and mother: He reports
the account he had been given before leaving the children alone in the room. again
an innocent one (they wanted to take a nap), and she translates it at the action
level (to undo the beds).
Confronting this episode with the preceding one, we observe that a challenge
about a similar troublesome event (children playing in a far too energetic way)
elicits different justification strategies: Sergio, the child of Excerpt 8, when
challenged, goes down as a first step to the factual level, providing factual
information from which the desired inference can be drawn by the parents .
Conversely, in Excerpt 9, the questioned conduct is accounted for with a formula,
the name of an activity that involves a certain kind of action but does not require
describing them. Some data from social cognition studies suggest that a situation
of challenge provokes a retreat to the descriptive level (Semin & Fiedler, 1988).
This certainly happens, in that rooting a claim in a detailed narrative is an
effective device to achieve an appearance of truth. Nonetheless, there can be
discourse situations in which the opposite strategy can turn out more convenient,
as in our example. We are closer, therefore, to Billig's position (1987), inasmuch
as a differential categorization is always a choice and responds to the rhetorical
demands of the discursive occasion.
In accounting and justifying, children can also use the frame offered by the
Problematizer, as in Excerpt 10 :
Excerpt 10
Aloisio family : Coke
Dad: Marco ; Mum: Laura; Daughter: Gaia, 6 years old; Son: Adriano, 11 years old
Dad:
432
Gaia:
Adriano : - > Gaia 1 'ha finita tutta . oggi porneriggio 1 ' ha aperta
Gaia has finished
opened it
it
up.
This
afte rnoon
she
and
she
finished
it
al l.
Dad :
->
Gaia
look ,
it
is a
l ot that
that muc h of
uh
Mum :
[e ti fa male
->
and
Gaia :
no guarda (
no
Adriano :
it's
l o ok
couldn ' t
drink
anything
afterwords
Gaia :
->
daddy daddy
Mum:
[capi :to
you
Dad:
began
this morning
see :
Adriano :
look
even worse
eh infatti
yeah exac t ly
( ( Ga i a laughs
Adri ano :
e ride lei
and
she
laughs
Mum :
->
when I t o l d you n o t
it ' s too :
to buy it
) e troppo :
' cause
( .. . )
Dad:
->
no
Mum:
' caus e
well
Dad :
nobody looks
no because
after her
even i f
c l ear l y
addirittura?
really?
Mum:
Adr iano :
eh : !
a f are che ?
to do what?
Mum:
, cause
( ( laughing) )
Mum:
Adriano :
Gaia
is
somewhat
fe' vero ]
Gaia :
that 's
tru e
he he
he he ci ride pure capito
she laughs too , y'see
i mitating Ga ia
( ( t h ey
both
laugh
433
The father opens the problem by addressing a question preceded by a rna [but],
which in this case (and in many others we found in our corpus) is neither a
contrastive move to previous talk nor a way of blocking a possible inference.
Rather, it is a way of initiating a problematic issue, probably based on a
previously unexpressed assessment of a state of affairs. The older child, Adriano,
picks it up by accusing Gaia of having drunk all the Coke and touches on the
temporal dimension by specifically indicating when his sister opened the bottle of
Coke. After the father and mother have expressed their disapproval, Gaia selects
the temporal dimension to make up a defense, correcting her brother and claiming
she began drinking the Coke beforehand, in the morning. Her justification
strategy involves the reconstruction of the event, aimed at getting rid of the
attributed behavior of "drinking so much Coke in so little time." Here the case is
especially interesting, because her justification bumps into another blameworthy
conduct: "drinking Coke in the morning." Her false step is commented on in a
critical chorus, and the girl ends up laughing at herself.
In the second part of the story, the focus shifts to "who buys it" (the mother
reproaches the father) and "who should control the children" (the father reproaches
the mother). The shift from the level of facts (there is no more Coke) to the level
of rules and values involves all the family roles, and the children find a free space
here between the parental authorities. They learn to use it in arguing, as can be
seen in other stories (e.g., "Violin ," Excerpt 12).
The Problematizee, particularly a child, can use the same frame of discourse
proposed by the Problematizer in order to counter argue, as is the case in the
following excerpt.
Excerp t 11
Soldano family : Ghosts
Dad: Vittorio, 35 years old; Mum: RaffaeIla , 35 years old; Grandmother; Grandfather;
Sons: Gianluca, 4 years old ; Stefano, 8 years old; Daughter: Marina, 3 months old
[Stefano refuses to go to bed without his brother because he is afraid of ghosts . He says
tha t, in his friend's "Handbook of Ghosts," it is written that Rome is the town most
populated by ghosts 1
Dad : - >
scusa Roma
excuse me
is Rome
so
full
of ghosts,
is it?
you met
male i fantasmi ?
hurt
by t he
g hosts?
Stefano :
->
once
stumbled on t he
nothing
have you
434
Dad:
that was
nothing not
the
ghosts,
[canmina dai
come on
go,
Mum :
[dai Stefano su
Stefano:
come
no I
Dad
on
stefano
am afraid,
eh Stefano su!
come
on
Stefano
Dad:
has
finished,
he must
come
camera on you
and
someone'll
hear that
-ridiculous-
he
is
afraid of
Mum:
Dad:
->
because
Marco
and wants
he
to wait
for
listens
to
Gianluca
the
foolish
things
that
seen them in a
book made up
tell
stupid things
to
children
so
Stefano :
->
Dad:
lot of money
there
is
even a
photo
ghost?
seen
Stefano:
that
can be
seen
be a llora si vedono.
well
they can be
seen then
In this story, the father, adopting a rational stance, challenges Stefano, the eightyear-old son, about his fear of ghosts. In his defense, Stefano keeps within the
same line of discourse and provides evidence ("Once I stumbled on the nothing") .
435
The father, although rejecting the explanation, actually reifies a nonentity not too
different from ghosts ("and that was the nothing") . Then the mother depicts
Stefano's fear as being a shame for a child of his age, and the father builds on it,
stressing Stefano's easiness to believe in every foolish thing his friends say or
that is written in stupid books . Stefano then argues against the core argument
(i.e., that the book lies) by presenting a new piece of evidence, the photos. The
father's rebuttal is again weak: Although meant to present the visibility of a
ghost as an absurdity, it can be taken literally to strengthen Stefano's position,
which, in fact, he does.
The level of discourse established by the father is accepted and used here by the
child to dismiss the content of the challenge and to discharge the negative
evaluation attached to his conduct.
Excerpt 12
Naceri family : Violin
Dad: Giancarlo , 45 years old; Mum: Gaia, 44 years old; Daughters: Ludovica, 14 years
old; Irma, 10 years old ; Antoni a, 3 years old
I nna :
the
violin
teacher
Dad:
d idn't
understand
anything
she
started off
at
six thirty-five
say s
ten t o
seven
said ten to
seven
he
always
c ome s
about
half
an hour
l ate r
436
Dad:
up.
doesn't
even
realize
she
arrived
six thirty I
am
out at
call
for
the
Dad:
next
lesson
she went
out
six thirty-five.how is
it
possible
is
absurd.
bad
done
Mum:
NO : :
Dad:
no .
neanche lei 10 vuole fare pili
not
Mum:
it anymore
no: :no rna anche se 10 volesse fare non glielo farei fare
no no but even if she wished to do it I
wouldn't allow her
it's a
commitment
for me,
not
for her.
she
has
to
practice
that
she
doesn't
This story shows how the conflict between daughter and father is mediated by the
mother, who is addressed by both. The daughter partially succeeds in counter-
437
opposing the father by using the same argument ("he is always late") he was
using against her. But when the father addresses his counter-defense to the mother
by saying that he will be the one who sets the time for the violin lesson, she
answers laconically and coldly that "the time is this one." So the mother is not
only the one who does more problematizing on the past activity than any other
single participant (as it shown in Table 3) but she is also the privileged recipient
or addressee of many disputes. And, in this case, she is recognized as the deciding
judge. She has the last word in the dispute between father and daughter because
she is committed to having the daughter practicing the violin. Her conclusion has
the form of an articulated reproach; it embeds a counterfactual argument, is
expressed in the third person, and is also addressed to the father in the sense that
he is not as involved as she is in getting their daughter to study the violin.
Conclusions
Return ing to the questions posed at the beginning, we can confirm that conflict
talk is not dispreferred (Levinson, 1983) but is rather usual in Italian families'
shared discourse. The quantitative results indicate that the problematizing activity
occupies a large part of family discourse, given that about one third of family talk
is problem centered. This allows children peripheral participation in conflict talk,
both when they are and when they are not the object of the discussion. In such
cases, children are exposed to various strategies for challenging and defense.
However, half of the problematizations are directed at the children, and when the
discours e concerns past events, it is more difficult for them to have an active role
of problematizer. We found that the mother controls and sustains this family
genre.
We focused on the different kinds of problematic situations in which children
can participate, when past events are problematized. We found that it is easier for
them to build a challenge within a topic already introduced by someone else. This
also appears to be a successful strategy of turn-taking, in that their problematizing
moves get taken up more easily. Specifically, they formulate topic-tied questions
that strongly demand a second pair part (Sacks, 1992), that is, of some kind of
relevant response.
Compared with discourse interaction in school, family discourse has an obvious
difference. In teacher-pupils interaction, an opposition is referred to another's
claim and is always a second pair part because it requires that someone has taken a
position or advanced a claim (Orsolini & Pontecorv o, 1992); in family dinner
talk, the Problematizer can raise a question and problematize someone else from
the very beginning of the sequence because the referent is probably already shared
by the participants. On a fine-grained linguistic level , this can explain the use of
rna [but] at the very beginning of the problematic topic. Sometimes at the
438
beginning, there can be a description or a short report of the facts, but the story
can also begin with a question . Although older children seem to be able to
introduce a topic on which they problematize someone else, for younger children
it is easier to link their argument to a topic previously introduced by other, more
expert participants. However, children become aware rather early that questions apparently very innocent ones, as we have shown in the previous excerpts - act as
one of the more appropriate communicative avenues by which to enter into the
family discourse arena.
When some children's past behavior is challenged, even children as young as
Stefania (Excerpt 8: Quarrel), are able to justify themselves and to articulate
answers that are rhetorically designed and finely tuned to their particular audience,
by offering a seemingly accurate reconstruction of the quarrel with her brother and
by using temporal markers, authority reference, and visual recall. Moreover, while
they are exercising their conversational skills in justifying and arguing, children
are also coping with the evaluative, normative, and encyclopedic connotations of
various family discourse topics. That is the case of the superior value of truth to
which Stefania appealed as a compelling argument.
Almost every example that we have reported here could be analyzed for
unraveling the implicit normative assumptions underlying the debate, both in the
challenging and in the defending parts. For instance, in Excerpt 7, when Sergio is
teasing about his father's laziness, the father defends himself by appealing to his
superior duty to look out for the children and conjures up, in competitive
cooperation with the mother, a little scene of parental sacrifice for the children's
benefit. A shared Italian value of the centrality of children's needs is thus
dramatized by the two parents for their children's sake.
In a general way, all the problems described here are not strictly cognitive . They
are far from classical (and mostly artificial) problem-solving situations, because
the focus is on problem definition (as we said at the beginning), but they are also
more value laden than the ones in problem-centered topics oriented to the future in
which practical solutions are sought. The reconstruction of a past that is yet
problematic appears substantially structured within social-relational problems.
Once again, cognitive activity related to communication concerns social
relationships in the first place, as shown in Bruner (1990).
We have tried to show in our analysis that an underlying orientation to social
and/or family norms and values is detectable in all our short stories and enacted in
the collective discourse. More or less active participation in these types of
discursive situations, which frequently lead to arguing activities, enables children
to be exposed to and to explore different strategies. So children need to learn the
use of complex discursive tools that are embedded with both arguing and
reasoning dimensions. These include the pragmatic ways of responding to requests
and problematizations proposed by parents and siblings through argumentative
moves accomplished by particular linguistic forms. They also include the use of a
range of different justification strategies, such as recategorizing reported events,
439
transferring the problem focus from self to others, and gaining the right of
challenging and of problematizing others.
Acknowledgment
The study presented in this chapter is part of a research project on interactive
learning in different social contexts that was supported with a 1993-1994 grant
from the University of Rome "La Sapienza" to the first author.
440
References
Bakhtin, M. M. (1986) . Speech genres and other late essays. Austin : University of
Texas Press .
Billig , M. (1987). Arguing and thinking: A rhetorical approach to social psychology.
Cambridge, England : Cambridge University Press .
Bruner, J. (1986) . Actual minds, possible worlds. Cambridge, MA: Harvard University
Press.
Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press .
Caron, J. (1990) . Le developpement de l'argumentation chez I 'enfant [The
development of argumentation in children] . Paris : Presses Universitaires de France .
Corsaro, W. A., & Rizzo, T. A (1990) . Disputes in the peer culture of American and
Italian nursery school children. In A D. Grimshaw (Ed.), Conflict Talk (pp. 21-66) .
Cambridge, England : Cambridge University Press .
Duranti, A, & Goodwin, C. (Eds.). (1992). Rethinking context: Language as an
interactive phenomenon. Cambridge, England : Cambridge University Press .
Edwards, D. (1991). Categories are for talking. On the cognitive and discursive bases
of categorization. Theory &: Psychology, 1(4), 515-542.
Edwards, D., & Mercer, N. M. (1987) . Common knowledge: The development of
understanding in the classroom. London : Routledge.
Edwards, D., & Middleton, D. J. (1988). Conversational remembering and family
relationships: How children learn to remember. Journal of Social and Personal
Relationships, 5, 3-25.
Edwards, D., & Potter, J. (1992). Discursive psychology. Newbury Park, CA: Sage .
Garvey, C. (1993) . Diversity in the conversational repertoire: The case of conflicts in
pretending. In C. Pontecorvo (Ed.), Forms of discourse and shared thinking .
Cognition and Instruction, 11(3 & 4), 251-264 .
Goffman, E. (1964) . The neglected situation. American Anthropologist, 66(6), 133136.
Goffman, E. (1974) . Frame analysis: An essay on the organization of experience. New
York: Harper and Row.
Goodwin, C. (1994) . Professional vision . American Anthropologist, 96(3), 606-633 .
Grimshaw, A D. (Ed.). (1990) . Conflict talk: Sociolinguistic investigations of
arguments in conversations. Cambridge, England : Cambridge University Press.
Kaye, K. (1982). The mental and social life of babies. Chicago: University of Chicago
Press.
Labov , W. (1972). Language in the inner city: Studies in the Black English vernacular.
Philadelphia: University of Pennsylvania Press .
Labov, W., & Fanshel, D. (1977). Therapeutic discourse: Psychotherapy as
conversation. New York: Academic Press.
Lave, J, & Wenger, E. (1991). Situated Learning: Legitimate peripheral participation.
Cambridge, England : Cambridge University Press .
Levinson, S. C. (1983) . Pragmatics. Cambridge, England: Cambridge University
Press.
Ochs, E. (1988) . Culture and language development: Language acquisition and
language socialization in a Samoan village. Cambridge, England: Cambridge
University Press .
441
Ochs, E. (1994) . Stories that step into the future. In D. Biber & E. Flanagan (Eds.)
Sociolinguistic perspective on register (pp .106-135). Oxford, England: Oxford
University Press .
Ochs, E., Pontecorvo, C., & Fasulo, A. (1996) . Socializing taste . Ethnos , 61(1-2). 746 .
Ochs , E., & Schieffelin, B. (1989) . Language has a heart. Text, 9(1), 7-25 .
Ochs, E., & Taylor, C. (1992a). Family narrative as political activity . Discourse and
Society, 3, 43-57 .
Ochs, E., & Taylor, C., (1992b) . Mothers' role in the everyday reconstruction of
"Father knows best. " In K. Hall (Ed.), Locating power: Proceedings of the 1992
Women and language conference (pp. 447 -462) . Berkeley, CA: University of
California at Berkeley.
Ochs, E., Taylor , C., Rudolph, D., & Smith, N. (1992). Storytelling as a theorybuilding activity . Discourse Processes, 15, 37-72.
Orsolini, M. (1993) . "Because" in children's discourse. Applied Psy cho linguistics,
14, 89-120.
Orsolini , M., & Pontecorvo, C. (1989) . La genesi della spiegazione nella discussione
in classe [Genesis of explanation in class discussion] . In M. S. Barbieri (Ed.), La
spiegazione nell 'interazione sociale (pp. 161-190) . Torino, Italy : Loescher.
Orsolini, M., & Pontecorvo, C. (1992) . Children's talk in classroom discussion .
Cognition and Instruction, 9, 113-136 .
Pomerantz, A. (1978) . Compliment responses: Notes on the co-operation of multiple
constraints . In J. Schenkein (Ed.), Studies in the organization of conversational
interaction (pp. 79-112) . New York : Academic Press .
Pomerantz, A. (1984). Agreeing and disagreeing with asses sments : Some feature s of
preferred-dispreferred discourse. In 1. M. Atkinson & 1. Heritage (Eds.), Structures of
social action (pp. 57-101). Cambridge, England : Cambridge University Press .
Pontecorvo, C. (1987) . Discussing for reasoning : The role of argument in knowledge
construction. In E. De Corte, J. G. Lodewijks, R. Parmentier, & P. Span (Eds .),
Learning and Instru ction. European research in international context, 1, pp. 71-82 .
OxfordlLeuven: PergamonlLeuven University Press .
Pontecorvo, C. (1993) . Forms of discourse and shared thinking . Cognition and
Instruction, 11, 293-310 .
Pontecorvo, C., Amendola, S., & Fasulo, A. (1994) . Storie in famiglia. La narrazione
come prodotto collettivo [Stories in the family . Narration as a collective product] .
Eta Evolutiva, 46, 18-34.
Pontecorvo, C., & Girardet, H. (1993) . Arguing and reasoning in understanding
historical topics. In C. Pontecorvo (Ed.), Forms of discourse and shared thinking .
Cognition and Instruction, 11, 365-395 .
Pontecorvo , C., & Orsolini, M. (1993). Discussing a story in a pre -school setting . In
M. A. Pinto & M. Danesi (Eds.), Humanism in language studies: Essays in honor of
Renzo Titone (pp . 78-94). Milan : IMFE .
Resnick, L. B. (1991) . Shared cognition: Thinking as social practice. In L. B.
Resnick, J. M. Levine, & S. D. Teasley (Eds.), Perspectives on socially shared
cognition (pp. 1-20) . Washington, DC: American Psychological Association .
Sacks, H. (1987) . On the preferences for agreement and contiguity in sequences in
conversation. In G. Button & J. R. E. Lee (Eds.), Talk and Social Organization (pp .
54-69). Clevedon, England: Multilingual Matters.
442
Sacks , H. (1992) . Lectures on Conversation: Volume 1. (G. Jeffer son, Ed.) Oxford ,
England : Basil Blackwell.
Schegloff, E. A. (1989). Reflections on language , development, and the interaction al
character of talk-in-interaction. In M. H. Bornstein & J. S. Bruner (Eds.), Interaction
in human development (pp. 139-153). Hillsdale, NJ: Erlbaum.
Schegloff, E. A. (1990). On the organization of sequences as a source of "coherence" in
talk-in-interaction . In R. Freedle (Ed.), Advances in discourse pro cesses:
Conversational organization and its development (Vol. 38, pp . 51-77). Norwood ,
NJ: Ablex.
Schiffrin, D. (1985) . Everyday argument: The organization of diversity in talk . In T .
van Dijk (Ed.), Handbook of discourse analysis, Vol. 3. Discourse and dialogue (pp .
35-46) . London : Academic Press .
Schiffrin, D. (1987) . Discourse markers. Cambridge , England: Cambridge University
Press.
Schiffrin, D. (1990). The management of a co-operative self during argument. The role
of opinions and stories. In A. D. Grimshaw (Ed.), Conflict Talk (pp. 241-259) .
Cambridge, England : Cambridge University Press .
Semin, G., & Fiedler, K. (1988) . The cognitive functions of linguistic categories in
describing persons: Social cognition and language . Journal of Personality and
Social Psychology, 54(4), 558-568.
Vygotsky, L.S. (1990) . Pensiero e linguaggio [Thought and language] . Bari, Italy:
Laterza . (New Italian critical edition by L. Mecacci, based on the first Russian
edition; originally published in 1934) .
Chapter 18
Abstract
The aim of this chapter is to present observations on the role of literacy in adult
education, with special emphasis on the cognitive and discursive transformations
that represent the appropriation of a new communicative technology such as
writing (Goody, 1977; Ong, 1982). Starting with the contributions of rhetorics of
this century (Burke, 1969; Billig, 1989), we have studied an activity frequently
promoted by teachers inspired in the main ideas of Paulo Freire: cl assroom
debates . Through the study of this kind of adult classroom activity, we have
analyzed the following aspects involved in them: types of semiotic means (e.g. ,
contextualized vs. decontextualized signs, speech genres, social languages;
Bakhtin, 1986a; Wertsch, 1985, 1991) used in debates and similar forms of
conversation by students (all of them women of several adult centers situated in
southern Spain) ; modes of argumentation and ways for persuading the audience
displayed by participants in debates ; motives that emerge in the course of this
form of dialogue .
444
Introduction
Research into literacy has been carried out in many of the social sciences. Many
of the findings from research in psychology, communication ethnography, and
literary theory have contributed to a better understanding of educational processes
in general and of the learning of writing in particular. However, these findings
have not yet filtered through to one of the educational areas that is presently
undergoing rapid development: literacy and adult education. This study aims to
probe into the complex world of adult education . We will do so by observing and
analyzing the transformations that literacy produces in orality in people attending
courses at different educational levels at the adult centers that we have been
studying in the Andalusian Adult Education Program, southern Spain, Cala,
1991; Laboratory of Human Activity , 1993).
We have been studying a series of group debates on subjects that, more often
than not, concerned and motivated the participating students, who were practically
all women . These discussions, on topics such as children's education,
unemployment, and drug abuse in youngsters , were often stimulated by the
teachers and provide an ideal setting for obtaining generative meaning with which
to develop texts that are understandable and stimulating for the student (Freire,
1970, 1973; Freire & Macedo, 1987). From our standpoint, it seems that they
also function as an interesting setting for testing the changes that a
communication technology such as writing can promote within the framework of
oral communication. However, we believe that orality and literacy each have an
influence on the other, and, if it is true that the former is affected by the second,
then it is also necessary to practice some types of dialogue, such as a debate, to
increase mastery of oral forms of discourse to begin with, and then later on to
develop the necessary interpretative skills for the comprehension of written texts.
In future research, we hope to clarify the mutual transformations between both
forms of communication in the framework of the research being undertaken in the
Laboratory of Human Activity . The underlying hypothesis in this chapter relates
to this two-way relationship. We believe that the technologizing of the word by
writing has changed our orality significantly, giving rise to rhetorical forms of
arguing that came up in the debates we observed. But we should also a<tl
something that is rarely taken into consideration by researchers into literacy,
while at the same time being widely acknowledged by adult educators: Promoting
this type of discourse is necessary to stimulate the understanding of our audience's
discourse in the use of words. In the case we discuss here, this is tantamount to
saying the discourse of others participating in the debate. Enlarging the capacity
for understanding oral discourse is fundamental for developing oral skills that will
facilitate the interpretation and comprehension of written discourse.
445
446
447
this was a person characterized by compassion for others and respect for other
people's basic rights (see Burke, 1969).
Two aspects of rhetoric should be highlighted from an ethical perspective: the
role of persuasion of the audience and the personal transformation of the speaker
in the use of words. Understanding the rhetoric act as a communicative event,
motivated exclusively by the desire to persuade the audience, represents just the
ftrst of the two aspects referred to; the second is understanding this as an act that
affects both the speaker and the other interlocutors . For this double affect to
happen , we need to accept a dynamic vision of language that will allow us to
understand the rhetoric act in terms of the real action: a particular form of action
that affects the audience and simultaneously transforms the speaker who carries it
out. Taking into account the two-way direction of rhetoric, the concept of
persuasion, normally understood as the fundamental motive, gives way to another
motive of great importance, the achievement of social cohesion, so necessary for
living in groups. As Duncan, one of the most illustrious proponents of the new
rhetoric, has stated, the attaining of such an important objective is conditioned by
the fact that men act "rhetorically on themselves and on others" (Duncan, 1989,
p. 154).
The role of rhetoric can be found in relation to the commitment that emerges
with the use of the word. Far from the neutrality that formalist approaches have
tried to implant on meaning, the use of the word always involves addressivity,
otherness, and, in short, commitment and responsibility of the speaker toward
those listening to him or her (Burke, 1969, 1984; Clark & Holquist, 1984;
Voloshinov, 1986). In Burke's opinion, speech is not a neutral event. In fact, it is
far from lacking judgments and moral assumptions; it is dominated by them. All
of this is intensely moral. Names for objects contain emotional suggestions that
give us clues concerning how we are to act in relation to them (Burke, 1984). A
word such as automobile will normally involve a concealed selection (it
designates not merely an object but also a desired object). Speech is not
exclusively guided by the mechanical act of referring; instead, it acts as a system
of attitudes. Burke continues his reasoning, saying that calling a man friend or
enemy is the same as suggesting a program of action with which to consider him .
An important ingredient in the meaning of such a word is the set of attitudes and,
consequently, the availability of the action that this involves . And the author
adds: "regardless of whether we should call the implicit program of action adeqJ.ate
.. . or inadequate .. . these emotional or moral weights inherent in spontaneous
speech tend to reinforce the act itself, hence making the communicative and active
aspects of speech identical" (Burke, 1984, p. 177).
The perspective adopted by Burke is important to us for one fundamental
reason: By eliminating any supposition of neutrality, the word is re-infused with
the morality that it had lost, along with its capacity for creating attitudes for
action. If we also combine the two ideas we have already mentioned (that the
rhetoric act influences both the speaker and his/her audience, and that the use of
the word is moral and generates attitudes for action), we come to a conclusion that
448
coincides with the standpoints defended by Bakhtin and his circle: namely, that
human experience is created through discourse, and, as a result, it lacks real
existence outside the field of expression. From the very moment of expression , in
the act of producing an utterance, we are creating experience or recreating our
internal experience. But, at the same time, internal experience is created by the
same semiotic material that gives form and expression to its external
objectivation . From this, we can conclude that no experience exists that is not
linked to signs (Bakhtin, 1986a; Mead, 1934; Voloshinov, 1986, 1987).
If the use of the word is not a neutral event but the creation of attitudes, and if,
as Bakhtin points out, it is the way through which the speaker can externally
recreate internal experience, in spite of the word having, like all signs, an external
origin, we can conclude by saying that the rhetoric act affects both the audience
and whoever carries it out. But, in addition to this, expression-experience, the
utterance once it is produced, may turn itself into a motive for reflection for its
transmitter and enable him to discover new senses that were latent in him.
449
The analysis of this concept does not stop here. To gain a better insight into it,
we need to analyze the means/tools with which the agent acts, because in addition
to human agency being shared, it is also mediated agency, given the fundamental
role that tools play in it (Wertsch, Tulviste, & Hagstrom, 1993). Mediated agency
is, in Bateson's terms, that which extends beyond the skin (Bateson, 1972;
Wertsch, 1991; Wertsch, Tulviste, & Hagstrom, 1993). Apart from depending on
the behavior of others at the time of carrying out one's own action, one also needs
the tools for acting . So human agency is mediated in the sense that it is carried
out with the use of cultural instruments, of mediating procedures, such as the
tools of work and language. Whatever their structure and the type of activity that
they facilitate, tools enable us to extend our mind beyond the skin, beyond the
narrow limits of our bodies.
Apart from the way in which we have tackled the analysis of the agency
responsible for rhetoric action, we can see the latter as a subset of what Wertsch
and colleagues have labeled mediated action (Wertsch, Tulviste, & Hagstrom,
1993). Given the important role played by the settings and tools used for carrying
out the action, we must assume that variations in the settings will produce
important changes in the course of and in the results of the actions in which they
are involved. If we consider that rhetorical actions involve a peculiar use of signs,
of word, and, in general, of semiotic settings, we will have to accept that the level
of decontextualization of these will play an important role when they are carried
out.
Several social and interactive processes are directly responsible for the
deconiextualizaiian of signs and indirectly responsible for the development of
higher mental functions (Vygotsky, 1978). By adopting a cultural-historical
perspective, Vygotsky took into consideration all the social processes as a whole
to explain the genesis of higher mental functions. In practice, however, he
actually tackled only the study of the sign in the framework of basic social
processes, such as interaction between dyads and relationships within the
framework of small groups (Vygotsky, 193411987; Vygotsky, 1978). Subsequent
experience has shown that the development of these instruments and their
progressive decontextualization do not depend exclusively on interactions within
the framework of small groups; they also depend on large scale institutions
developed for this reason . In what is undoubtedly his most representative book,
Thinking and Speech (193411987), Vygotsky was aware of the importance of
formal education in the process of decontextualization of the word's meaning. It
was clear to him that, without the framework of formal education, the child could
not reach a level that would be sufficient for developing scientific concepts .
In the Western world, the school has evolved as a setting in which society
provides the instruments and sufficient knowledge for people to be able to live
within it. The general set of practices that are carried out there provide the student
with new meaning that would be unlikely to be transmitted without it. It is rare
for these meanings, which Vygotsky labeled scientific concepts, to have arisen as
the result of individual work . Instead, they represent the efforts of working groups
450
451
Types of Utterances
The utterances we present next are representative examples of the discourse uttered
by the women, all of them students at different educational levels of the Adult
Education Program, taking part in debates. In all the debates observed in the adult
classes , we can confirm the existence of utterances that reflect a wide variety of
personal experiences (utterance-type), in terms of both their content and the
interdependence of the dialogue with preceding utterances. During one of these
adult classes, on the subject of household chores , the women taking part produced
utterances similar to the following:
There is one common thread running through the possible meanings and senses
reflected by these utterances. They all depend on the speaking voice concentrating
on itself and on its daily experiences, and, given the conditions created by the
dialogue, the speaker decides to exteriorize dramaturgically a part of herself
(Habermas, 1984) through the presentation of fact from her daily existence .
452
However, these are not the only utterances expressed by those taking part in the
debate. Others of a more general content may transcend the limits of that existence
and place the problem described in a more complex social context. The following
examples reflect more general facts:
When someone produces an utterance like this, it does not refer to just one
particular person; it refers to many. The woman who says something like this is
producing an utterance whose meaning could be assumed by many other women.
Most likely it is applicable to such a wide variety of couples that it goes beyond
the boundaries of its group of reference. It is fairly common to see expressions
like 3 and 4 in propaganda sent out by political parties at election time in an
attempt to win over the female vote. Example 3 could also come from a feminist
text, or it might appear in the more progressive press reflecting present-day issues
affecting women.
Utterance 1 (My husband doesn 't help at home) reflects the use of
contextualized signs. It takes the self of the speaker as a reference point or, to put
it differently , as a deictic center (Levinson, 1983). In addition to this, the referent
is one particular person, and, as a result , the sign that refers to him cannot be
anything other than a contextualized sign (my husband). The same thing happens
with the predicate. In the latter, a very specific type of help is mentioned: the
chores that the speaker carries out in her own home. So, all the signs are highly
contextualized, and, as a result of that, we get a particular and specific image of
the situation .
Utterances 3 and 4 transmit a different meaning (the man should help with the
housework, and menshouldhelpwith the household chores). The speaker here has
used a different type of tool. The signs used in these two cases are
decontextualized signs (the man, men, the housework, household chores), and so
the effect achieved is that of a general meaning, that has no particular individual as
a referent. Instead, it takes a general individual or individuals that do not even have
to be men from the same reference community (the men from this town, from
this neighborhood, and even from this country) as referent, and it may refer to the
male collective in its totality.
If formal education is the most privileged setting in our culture for producing
the decontextualization of meaning, the question to be asked in relation to the
utterance-types presented here has to do with their varied use by the adult students
taking part in the debates. Of all the utterances produced by the students at the
first levels of training (total or partial illiterates) , those similar to type 2 (My
husband doesn 't evenhelp me to clear away the plates) were the most frequently
found in their verbal productions. However, the utterances of types 3 and 4 tended
453
454
455
456
arguing. The particular utterances will be under the control of the first motive
mentioned.
We should stress, however, that both hypotheses could mutually complement
each other. The recently acquired habit of expressing oneself in the educational
framework, faced with a public made up of the teacher and the rest of the women
in the classroom, makes the motive of expressing oneself dramaturgically take
precedence while at the same time highlighting the lack of tools and skills for
carrying out more complex rhetorical actions. But, at the same time, we should
also point out that only continued discursive practices (e. g., semi-structured
dialogues or debates) will allow the appearence of new motives and the
development of more sophisticated rhetorical instruments subordinated to them.
Whatever the different perspectives adopted in the study of debate, teachers,
experts, and researchers of literacy and adult education share one thing in common:
the great interest represented by the application of a rich variety of discursive
practices previously studied by rhetoric but largely forgotten by pedagogy and
other educational sciences.
It is worth considering the transforming role of literacy when this is acquired in
an appropriate educational framework. In the adult centers in which the researchers
from the Laboratory of Human Activity have carried out their observations, the
literacy practices are combined with various forms of dialogue. These dialogues
are sometimes informal conversations through which women students have an
opportunity to speak spontaneously about a wide variety of subjects. On other
occasions, the dialogues are organized in debates coordinated by the teacher. We
believe that only the appropriate combination of a new communication
technology, such as literacy, with the oral dialogue in all its variations can
promote the adequate decontextualization of the sign and the generation of new
forms of discourse (social languages and genres). Through these new semiotic
settings in the service of communication, the adult starting at the beginning of
the educational process or coming back to it after many years will be able to
widen his or her cultural horizon and find ways of adapting to the demands of a
changing world.
457
References
Bakhtin, M. M. (1986a). Speech genres & Other late essays. Austin, TX: University of
Texas Press .
Bakhtin, M. M. (1986b). The dialogic imagination. Austin, TX: University of Texas
Press.
Bateson, G. (1972). Steps to an ecology of mind. A revolutionary approach to man 's
understanding of himself. New York: Ballantine .
Bauman, R., Irvine, T., & Philips, S. (1987). Performance, speech communication and
genre: A critical review of concepts in the ethnography of speaking . Working
Papers of the Center for Psychosocial Studies (No. 11). Chicago: Center for
Psychosocial Studies.
Billig, M. (1989) . Arguing and thinking. Cambridge , England: Cambridge University
Press.
Brigg, C. L. (1993) . I'm not just talking to the victims of oppression tonight - I'm
talking to everybody: Rhetorical authority and narrative authenticity in an AfricanAmerican poetics of political engagement. Journal of Narrative and Life History, 3
(1), 33-78.
Burke, K. (1969) . A rhetoric of motive. Berkeley , CA: University of California Press .
Burke, K. (1984). Permanence and change. An anatomy of purpose . Berkeley , CA:
University of California Press .
Cala, M. J. (1991). El programa de educaci6n de adultos y el papel social de la mujer: un
estudio sobre valores , actitudes y modos de pensamiento [The Program of Adult
Education and the social role of the woman. A study about values, attitudes and
modes for thinking] . Memoria de investigaci6n . Sevilla, Spain : Instituto Andaluz de
la Mujer.
Clark, K., & Holquist, M. (1984). Mikhail Bakhtin . Cambridge, MA: Harvard
University Press.
De Certeau, M. (1984) . The practice of everyday life. Berkeley, CA: University of
California Press .
Duncan, H. D. (1989). Communication and social order. New Brunswick, NJ:
Transaction Publishers.
Freire, P. (1970) . Pedagogy of the oppressed. New York: Seabury Press.
Freire, P. (1973). Education for critical consciousness. New York: Seabury Press .
Freire, P., & Macedo, D. (1987) . Literacy: Reading the word and the world. Hadley,
MA: Bergin and Garvey.
Giddens, A. (1989) . Constitution of society. Cambridge, England : Polity Press.
Goody, J. (1977) . The domestication of the savage mind . Cambridge, England:
Cambridge University Press .
Habermas , J. (1984). The theory of communicative action (Vol. I). Boston, MA:
Beacon Press .
Holquist, M., & Emerson , C. (1986) . Glossary for The dialogic imagination (pp. 423 434). Austin: The University of Texas Press.
Laboratory of Human Activity. (1993). La educaci6n de adultos y el cambio de valores,
actitudes y modos de pensamiento en la mujer: Un estudio etnognifico del discurso
[Adult education and change of values, attitudes, and ways for thinking: A discourse
ethnography study]. Memoria de Investigaci6n. Madrid, Spain: CIDE.
458
Author Index
205,222,411 ,440,443,448,450,
457
Balmelli, A, 271, 286
Bangert-Drowns, R. L., 275, 285
Barbieri, M . S., 229, 237
Barker, K. G., 338, 357,374
Barley, S., 42, 59
Barnes , P., 230, 238
Bateson , G., 448, 457
Bauman , R., 182, 199,451 ,457
Beach, K., 250, 264
Beishuizen, J., 338, 358
Bell, N., 269, 286
Belyaeva, A , 227, 237, 270, 285
Bern, D. J., 387, 404
Bergqvist, K., 12, 112, 386, 394, 404
Berk, L. E ., 372, 379, 383
Berkowitz, M. W ., 362, 363, 367,383
Berlin, B., 111-114 , 120,129,131,
132, 134-136, 137
Berstein , B., 124, 137
Besnier, N., 205, 222
Bessiere, c, 272, 287
Bibby, A , 232, 237
Billig , M., 8, 18,19, 145,168,408,
440,443,446,453 ,457
Bivens, J. A , 379, 383
Blanche-Benveniste, C., 205, 222
Blaye , A , 227-228, 229 , 236, 237,
238
460
Blom, J.-P., 170, 199
Blomberg, J., 42, 59
Bower, G. R., 339, 358
Bowers, 1., 42, 59
Brackenridge, A., 232, 237
Brannock, J., 364, 384
Brassae, C., 89, 94, 95, 108, 110
Brewer, W. F., 337, 358
Brigg, C. L. , 450, 457
Britten, T., 338, 357
Brown, J. S., 65, 85 , 296, 310, 339,
350,356
Brown, P. J., 338,356
Brun-Cottan, F., 42, 43, 45, 51, 57,
59, 123, 137
Bruner, J., 112,137,407,438,440
Bruner, J. S., 9, 19, 115,137,138
Buchanan, K., 158, 168
Buckingham, N ., 226, 238
Burke, K., 443, 446, 447, 457
Burton, R. R., 296, 310, 350, 356
Butor, M ., 280, 285
Butterworth, G., 232, 236, 238, 269,
286
Buttny, R., 145,168
Byrne, R. M. J., 90, 108
Cala, M. J., 457
Campbell, D. T., 391, 404
Campion, P., 64, 85
Carey, S., 337, 356
Carmel, E ., 338, 356
Caron, J., 408, 440
Carraher, D. W., 14, 19,292,295,
296,299,300,310,311
Carraher, T. N ., 291, 292, 294, 296,
299, 310
Cesareni, D., 15, 19
Chafe, W. L. 338, 356
Chaiklin, S., 115, 137
Chalfont, B., 266, 286
Chambres, P ., 236, 237
Chen, R., 338, 356
Chavez, L. R., 171, 199
Author Index
Chi, M. T. H., 337, 356
Cicourel, A. V., 112, 121, 137, 173,
199
Clark, A., 59
Clark, R. R., 314, 334
Clark, K., 447, 457
Clement, J., 316, 334
Cody, D., 337, 358
Cohen, A. R., 380, 383
Cohen , R ., 270 , 285
Colbourn, C., 228 , 238
Cole, M., 3,19,117,128,137,174,
202, 245, 246, 264, 270, 285, 304,
310, 389, 404
Collier, J., 144, 151, 168
Collins, A., 356
Collins, H. M., 117, 137,339,356
Condor, S., 145, 168
Conklin, R., 114, 132, 137
Conklin, J., 338, 356
Corsaro, W. A., 408, 440
Coulmas, F., 82, 85
Coulson, R. L., 337, 358
Crawford, S. F., 338, 356
Crook, C. K., 228, 237
Cuisset, P., 273, 285
Culley, L., 232, 237
Curnock, D., 164, 168
Damasio, A. R., 136, 137
Damasio, R., 136, 137
Damon, W. , 363, 364, 383
Davillard, C., 91, 108
Davis, J. R., 381, 383
De Certeau, M., 454, 457
de Kleer, 1., 350, 356
Demuth, C., 179, 199
De Young, L., 338, 356
Diaz, R. M., 372 , 373, 383
DiBello, L., 245, 258 , 259, 264
Dickson, L., 305, 307, 310
diSessa, A., 337, 356
Djakow, 391, 404
Doise, W., 225, 226 , 237 , 269, 285
Author Index
Donaldson, M ., 226 , 232, 237
Douady, R , 307 , 310
Drew, P., 64, 85
Droogers, A., 172, 199
Dubuisson, C., 280, 285
Duchastel, D. D ., 356
Dugdale, A., 42, 59
Duguid, P., 65, 85, 339, 356
Dunbar, K., 364, 371, 380, 383
Duncan, H. D., 447, 456
Duranti, A., 11, 16, 170, 172, 175,
181-183, 186, 199,200,408,440
Edwards, D., 10, 19, 145, 168,269,
285,389,394,404,407,408,417,
440
Eibl-Eibesfeldt, 1., 190, 199
Elbers, E., 269, 285
Ellis, S., 362 , 383
Emerson, C., 450 , 457
Engestrom, Y., 115 , 128, 137
Engle, R A., 312, 314, 331, 334
Epstein, W., 338, 357
Ericsson, K. A., 379, 383
Evans, J. S. B. T ., 90, 108
Faggin, E, 274 , 285
Falzon, P., 89, 109
Fanshel, D., 417 , 440
Fasulo, A., 11, 16, 17,206,210,212,
222,407,409,411 ,413,441
Fay, A. L. , 364 , 371, 380 , 383
Feltovich, R L., 337 , 358
Fiedler, K., 431 , 442
Filippi, G., 42, 59
Fish, R., 266, 286
Flores, E, 64, 86, 388, 405
Foot, T ., 228, 238
Foppa, K., 64, 85
Forbes, K., 59, 137
Foucault, M ., 218, 222
Forman, E. A., 236, 237
Forsythe, D., 42, 59
Frederiksen, J. R., 354, 358
461
Freire, P., 443, 444, 457
Fujimura, J., 42, 59
Furnas, G. W., 355, 356
Gaik, E, 216, 222
Gal, S., 170, 200
Galegher, J., 266 , 286
Gane, M ., 145, 168
Gardner, H., 289, 310
Garfinkel, H., 42, 49, 59 , 61, 63, 85 ,
88, 108, 112, 137
Garner, R , 341, 356
Garrod, S. C., 341, 356
Garvey, C., 408, 440
Gaudin , T., 268, 285
Gauvain, M., 362, 383
Gay, J., 245, 264, 304, 310, 389, 404
Gegeo, D., 179,202
Ghiglione, R., 90, 94, 95, 106, 108
Gibbs , J. C., 362, 363, 367, 383
Giddens, A., 448, 457
Giles, H., 363, 383
Gillet, B., 273 , 285
Gillett, G., 389, 404
Gilmour, A., 226 , 238
Girardet, H., 364, 384, 408, 410 , 441
Glaser, R., 357,356
Glenberg, A. M ., 338, 357
Glick, A., 389, 404
Glick, J., 13, 245, 250, 256, 259, 264
Gloor, P. A., 337, 338, 357, 358
Goffman, E ., 9, 19,82,85, 95, 105,
108,115,137,160,168,205,222,
407,440
Goldman, S. V., 174,201
Gonzales, P., 24, 40
Goodenough, W.H., 114, 137
Goodman, N ., 3, 19
Goodwin, C., 4, 7, 16,24, 36,40,42,
45,47,48,51,54,59,64,76,85,
112 ,118,134,135,137,170,173,
200,387,392,407,408,440
462
Author Index
Author Index
John-Steiner, V., 3, 19
Joiner, R., 228-230 , 238
Jones, A, 228, 238
Jonassen, D. H., 337, 338, 341, 357
Jonsson, L., 389, 404
Jordan, B., 42, 50, 51, 55, 56, 59, 60 ,
61, 137
Karsenty, L., 89, 109
Kay, A, 283, 286
Kay, P., 111-114, 120, 129, 131, 132,
134-136, 137
Kaye, K., 407 , 440
Keating, E ., 173, 200
Keisler, S., 381, 384
Keller, C., 4, 19
Keller, J. D., 4, 19
Kendon,A, 76, 85,173,200
Kenrick, D. T ., 388, 404
Kerr, L. K., 312 , 314,331 ,334
Kiesler, S., 273, 287
Killen, M., 363, 383
Kindred, J., 245, 258, 264
Kintsch, W., 338, 339, 358
Kirshner, D., 315, 334
Klahr, D., 364, 371, 383
Klausen, T., 24, 40
Knorr-Cetina, K., 42,61,117,138
Kobasigawa, A, 338, 357
Koffka, K., 389,404
Kohn, AS., 312, 313 , 317, 333, 334
Kosslyn, S. M., 340 , 357
Kraut, R., 266, 286
Kroskrity, P. V., 170,200
Kruger, A C., 362-364, 373, 383
Kuhn , D., 364, 371, 384
Kulick, D., 172, 200
Labov, W., 411 , 417, 440
Lacoste, M., 89, 109
Landow, G., 337, 358
Larkin, 1. L., 315, 334
Larrue, J., 90, 94, 95, 96, 97, 106,
109, 110
463
Latour, B., 42, 46, 61, 115, 117, 138
Lave, J., 4,9, 13, 19,54,61, 115,
133, 137, 138, 139 , 174, 180, 201,
246,261 ,264,289,294,310,311,
389,402,404,4]8,440
Lawler, R. W., 296, 3]0
Lee, B., 204, 222
Leech, G., 204, 222
Leichter, H. J., 174, 201
Lemke, J., 401 , 404
Lerner, G. H., 153, 168
Levine, J. M., 269, 287
Levinson, S., 88, 89, 109, 409, 437,
440,452,458
Levy, P., 273, 286
Light, P., 11, 226-230, 232, 234,
236, 237, 238, 269, 272, 286, 289,
304,306,307,310,311,381,384
Linell, P ., 389, 402, 404
Littleton, K., 11, 230, 232, 238, 272
Livingston, E., 42, 61
Lodge, D., 82, 85
Lucarella, D., 338, 357
Luff, P., 42, 47 , 48, 52, 58, 60, 61,
64, 67, 85, 112, ] 15, 116, 138
Luria, A R., 288 , 310 , 391, 392, 404
Lynch, M., 42, 61, 112, 115,117,
121, 126, 135, 138, 139
Macauley, R. K. S., 205, 222
Macedo, D ., 444 , 456
Mackenzie, M ., 227, 237
MacKinlay, D. R. E ., 145, 168
Macpherson,
170, 201
Maier, R., 269, 285, 391,404
Mandl, H., 337, 357
Mannheim, B., 171,201
Markova, I., 64, 85
Martin, L. M. W., 250, 264
Marton, F., 341, 357
Mason, J., 269 , 286, 304, 306, 307,
311
Matz, M., 315, 334
Mayer, R. E ., 354, 357
c.
464
McAleese, R., 337, 357
McCaffrey, M., 232, 239
McDermott, R. P., 174, 200, 246 ,
264
McClelland, I., 228, 238
McGrath, D., 338, 357
McNeill, D., 128, 139
Mead, G. H., 2, 19, 448, 458
Melton, R. F., 341, 358
Mercer, N., 228, 238 , 269, 285 , 394,
404, 407, 440
Merleau-Ponty, M. , 388, 404
Messer, R., 230, 238
Mevarech, Z. R., 228, 238, 272, 286
Meyer, B. J. F., 355, 358
Middleton, D., 10, 19, 115, 137, 145,
146, 158, 164, 168,389,404,407,
440
Miller, P., 179, 201
Minick, N ., 236, 237, 446, 457
Moeschler, J., 95, 96, 105, 106, 109
Mohageg, M. F ., 344, 358
Monteil, J. M ., 236, 238
Montgomery, R., 364, 365, 283
Moore, J. L., 312, 314, 331, 334
Moore, W. S., 119, 139
Morgan, M., 183, 201
Moricot, C., 274, 285
Morrow, D . G., 339, 358
Mosenthal, P., 338, 357
Mugny, G., 225, 226, 237, 269, 285
Mulkay, M., 42, 61
Nathan, M. J., 312, 313, 317, 333,
334
Neisser, D., 389, 404
Newman, J., 226 , 238
Nicholls, G., 7
Nicolet, M ., 269, 286
Nisbett, R. E. , 90, 109
Nordon, C., 270, 285
Nunes, T ., 14, 19,269, 286, 289,
292, 295 , 300, 301, 304, 306, 307,
310, 311
Author Index
Ochs, E ., 11, 16, 19,24,40, 170,
172,175-177,181-183,199,201,
202,407,409,410,411 ,413,415,
416,440,441
O'Donnell, A , 353, 358
O'Laughlin, M ., 364, 371, 384
Oliver, R. R., 112, 137
Ong, W. J., 443, 445 , 446 , 457
Orr, J., 42, 61
Orsolini, M., 407, 409 , 410, 437, 441
Palen, L. , 4, 6
Palmer, S. E., 340, 358
Papert, S., 270, 286
Parsons, T., 117, 119, 139
Pea, R. D., 270, 271, 286
Pennington, N ., 88, 108
Perakyla, A, 153, 164, 168
Perret, J.-F., 269 , 286
Perret-Clermont, A-N., 225, 226,
237-239,269,285-287,289,310,
381, 384
Perriault, 1., 13, 19, 268 , 286
Perrin-Glorian, M. J., 307, 310
Phelps, E. , 363, 364, 383
Philips, S., 451, 457, 458
Phillips, T., 228, 238
Piaget, J., 89, 90, 109, 225, 239, 291,
304 ,311,362,384
Pichert, J. W., 341, 358
Pickering, A , 42,61,115,117,139
Pinker, S., 341, 358
Plunkett, S., 296, 311
Pochon, L.-O ., 7, 13, 270, 274, 286
Podak, C., 266, 286
Poirot-Delpech, S. L., 274, 285
Polanyi, M., 118, 139
Pomerantz, A , 408, 441
Pontecorvo,
15-17, 19,364,380,
384 ,407-411,413,417,437,441
Potter, J., 145, 168,417,440
Pozzi, S., 228, 237
Radley, A , 145, 168
Ramirez, J. D., 18
c,
465
Author Index
Randall, D., 42, 43, 60
Reed , H. J., 294, 311
Rees, E ., 337, 356
Reis, R., 341, 356
Resnick, L. B., 5, 16, 19, 88, 109,
269,287,312,313,317,333,334,
366, 384, 407, 441
Retornaz, A, 90, 110
Rheingold, H., 275, 276, 287
Rivers , W. H., 389, 404
Rizk, A., 337, 358
Rizzo, T. A., 408, 440
Robbins , H., 226, 238
Rock, I., 387,388,389,404
Rogard , V., 89, 109
Rogoff, B., 115, 133, 139, 178,201,
268, 287, 362, 383, 446, 458
Romaine, S., 171, 201
Romanelli, P., 15, 19
Rommetveit, R., 269, 287, 402, 404
Rosaldo , R., 171,201
Roschelle, J., 313,314,327,334,
366, 380, 384
Roskill, M., 120, 139
Ross , L., 90, 109
Roulet, E., 95, 96, 105, 106, 109
Rubattel, C., 95, 96, 105, 106, 109
Rubtsov, V., 227, 228, 237, 239
Rudik, 391,404
Rudolf, D., 175, 201
Rudolph, D., 407, 413, 441
Ruhleder, K., 42, 61
Sabel , C., 250, 264
Sacks, H., 8, 10, 19,49,59, 66, 77,
86 , 112, 115, 121, 122, 133, 139 ,
145,158,165,168,182,202,208,
209,212,222,223,408,437,441,
442
suje, R, 17, 112,269,287,341,
357,386,389,394,402,404
139
Saxer, K. H., 338, 358
Scardigli , V., 274, 285
Schaefer , E. E , 314,334
Schauble, L., 366, 367, 371, 380, 384
Schegloff, E. A, 96, 109, 112, 115,
122, 124, 133,139, 182,202,208,
209,212,222,223,372,384,407,
408,417,429,442
Schegloff, M., 54, 61
Schelling, M., 95, 96, 105, 106, 109
Scheper-Hughes, N., 127, 139
Schieffelin , B. B., 16,19,170,177,
179, 201, 202
466
Smith, N., 407, 413, 441
Smith, R, 175, 201
Snir, J., 313, 334
Soldatova, G., 228, 237
Somekh, B., 228, 238
Spiro, R. J., 337, 358
Spitzer, C. E., 381, 383
Sproull, L., 273, 287, 381, 384
Star, S. L. , 42,61,117,123,139, 140
Stasser, G., 381, 383
St. Clair, R. N., 363, 383
Stone, C. A , 236, 237, 381, 384
Streitz, N. A , 337, 357, 358
Suchman, L. , 4, 6, 42, 45, 50, 52, 54,
58,59, 61, 62, 64, 86, 88, 109,
112, 115, 116, 123,137, 140
Szeminska, A, 304, 311
Tannen, D ., 205, 223
Taylor, C., 175,201,407,409,410,
413,415,416,441
Teasley, S., 15, 16, 228, 269, 287,
366, 371, 380, 384
Tetlock, P. E., 381, 384
Theureau, J., 42,59
Thurstone, L. L., 289, 311
Todorov, T., 82, 86
Tolmie, A, 228, 237
Tomasello, M., 362, 364, 373, 383
Traweek, S., 42, 61
Trigg, R., 42, 59, 62, 137
Trognon, A, 4, 7, 89, 90, 94-97, 106,
108,109,110
Tudge, J., 364, 384
Tulviste, P., 449, 458
Turner, V., 206, 223
Underwood, G., 232, 239
Underwood, J., 232, 239
Vacherand-Revel, J., 272, 287
Vanderveken, D., 94, 95, 109, 110
Van Dijk, T. A , 338, 339, 358
Varenne, H., 174,201
Author Index
Vergnaud, G., 307, 311
Vitale , B., 270, 287
Voloshinov, V. N., 172,202,204,
223,447,458
Vosniadou, S., 337, 358
Vygotsky, L. S., 2, 3, 16, 19,86,
172,202,268,287,288,311 ,407,
442,449,458
Waern, Y., 266, 271, 287
Walkerdine, V., 394, 405
Wallace, T. E., 253, 264
Wang, S., 338, 357
Wason, P. C., 90, 110
Wathen, S. H., 16, 19
Watson, A R, 144, 145, 151, 168
Watson-Gegeo, K., 179, 202
Watt, 1., 174, 200
Weiss , J., 280, 285
Weiss, S., 338, 358
Wenger, E., 4,9, 13, 19, 54, 61, 133,
138, 180, 201, 261 , 264, 402, 404,
418 ,440
Wertheimer, M., 389, 405
Wertsch, J. V., 4, 16, 18, 19,82, 86,
268, 287, 380, 381, 384, 386, 391 ,
405 ,443,446,449,451,458
Wetherell, M., 145, 168
Whalen, J., 42, 57, 58, 61, 62, 64, 86
White, B. Y., 14, 19,313,334,354,
358
Wilkes -Gibbs, D., 314, 334
Wilkinson, A C., 338, 357
Winograd, T., 64, 86, 388, 405
Wiser, M., 313, 335
Wistedt, 1., 399, 405
Wittgenstein, L., 111, 112, 115, 129,
133, 136, 140
Wooigar, S., 42,61 , 115, 139
Yankelovich, N., 337, 358
Yates , J., 45, 62
Zajonc , R. B., 380, 384
Author Index
Zazanis, E., 245, 250, 255, 256, 258,
264
Zeitz, C. M., 16, 19
Zentella, A. C., 171,202
Zimmerman, D., 42, 58, 62
467
Subject Index
Caregiving, 169,176-179,192
hierarchical organization of, 176180, 192, 194, 196
socially distributed, 176-181, 191,
192, 194
Centers of coordination, 41-43, 45,51 ,
57,116
Chemistry, 445
Classification
processes of, 112-115, 126, 129,
135, 136
Classroom interaction, 393-398
Coaction, 230, 232-234
Co-construction, 363, 379
of knowledge, 229
Code-switching, 170, 171, 105, 243,
255-258
Cognition, 2, 64, 112-116, 124, 127,
133-136,236,246,249,250, 389,
390,402
developmental, 243, 245-248, 261263
distributed, 5-8, 12, 15-17, 24, 90,
106, 115, 133
embodied, 118, 127-131, 134, 136
situated, 1,3,4,9, 12, 15-17,24,
30,31 ,39,115,117,128,246-250,
254, 261, 262, 289-292, 303, 304,
387, 389, 396, 398, 407
470
situated action in, 112, 115, 116,
126, 128, 134
Cognitive
conflict, 362, 363, 365
development, 265, 267-269, 274,
279
Collaboration (see also Learning,
Problem Solving), 16, 63, 66, 74,
82, 90, 181, 224, 225, 227
peer, 361-364, 367, 369,371-373,
377-381
Color categories, 111,112, 114, 115,
121, 124, 131, 132-135
Communities
of innovation, 254, 262, 263
of practice, 111-113, 117, 128, 132134, 250, 252-254, 398
Competent practitioner, 44, 47, 116
Computer
use of, 265-267, 270-276, 278, 281
representations, 272, 274
Conceptual
change, 394,403
integration, 255, 259
knowledge, 391, 396,402
learning, 396, 401
models, 312, 313
Context, 224, 236
Contextualization, 451, 452
decontextualization, 448-450, 452,
456
Contingency, 44, 56-58
Control,49
Control room , 42, 43, 45-47, 50-54,
64, 67
Conversation (see also Dialogue,
Discourse , Talk), 90
analysis, 68, 88, 90, 91, 94-106,
112, 115, 133, 145, 182
family, 407, 409-411, 413, 414,
418-438
structure of, 90, 106
transactive, 15
Subject Index
Conversational processes, 314
of confirmation, 312, 329-331, 333
of convergence, 312, 327, 329-331,
333
of creation, 312, 329
turn-taking, 282, 312, 314, 327,
329, 331, 333
Debates, 443, 444, 446, 450-452 , 454,
455
Decomposition , 296, 297
Design, 43, 44, 57, 58
Design-in-use, 44, 56, 57
Dialogue, 443, 445, 449, 450, 453456
Didactic contract, 269
Dilemmas of uncertainty , 143, 146,
157, 158, 163
circumstance, 144, 146, 149, 153,
155, 160, 165
ordinariness/extraordinariness of
family life, 143, 144, 146, 151,
15~ 15~ 158, 16~ 16~ 165
outcome, 144, 146, 153, 155, 157,
158, 165
reasonableness , 144, 158, 164, 165
trustworthiness, 146, 149, 159, 162,
164
using technologies of care, 144,
151, 161
Discourse (see also Conversation,
Dialogue, Talk), 2, 3, 5, 17, 29, 38,
47-49,175,204,205,207,221,
243, 246, 256, 262, 263, 385, 398,
443,444,447,453
analysis, 8, 9, 206, 407, 452
classroom, 392, 395-397,400,401
communities,S, 9, 10, 17
genres, 406, 411, 412, 443, 449,
450, 454-456
narrative, 31, 32,34, 38
oral, 444, 445, 448
patterns of, 4
471
Subject Index
roles, 409, 411-413, 415
written , 444, 445, 448
Distributed activities, 43-45, 46, 50,
51,53,56
Division of labor, 42, 43, 53
Environment
home
cultural organization of, 169, 183,
189,197
spatial organization of, 183, 184,
186, 196
work, 45, 51-53, 55, 56, 67, 83,
254
technology-intensive, 42, 44, 57,
91,92
Ethnography, 255, 258, 444
of science, 117, 119, 124, 128, 134,
182
Ethnomethodology, 49, 87-89, 94,
112, 121, 145
Ethnotechnology, 13,41-42,265,
267-269, 274
Experimental contract , 269
Expertise
transmission of, 280
Learners
Subject Index
472
402
to reason, 15
tasks, 339, 354, 355
Linearity/nonlinearity, 336, 338
Linguistic markers, 243
Literacy, 173, 174, 246, 444-446,
452,455
activity, 170-175
church-based, 175, 182
home-based, 175, 182
syncretic, 169, 170, 172, 173, 175,
182, 183, 191
Material practice, 44, 45
Materials Requirement Planning
(MRP), 250, 251-257, 259, 260,
262
Mathematical operators, 248, 249,
312,315,317,319,321,327,331,
333
Measures
isomorphism of, 307-309
product of, 307-309
Mediated agency, 448, 449
Mental models, 340, 341, 344, 350,
354
analysis of, 336, 341, 342, 348,
349, 352
construction of, 336, 339, 341-343,
348-350,352-354
model-based strategy, 337-339, 341343, 348, 352, 353
Mexicano speakers, 173
Microworlds (see also OPERA,
Planner system), 366, 367, 371,
372, 380
Modification of use, 273, 274, 276279
Multiculturalism, 11, 169-173, 175,
176, 183, 196, 197
Multilingualism, 170, 172, 173, 175,
196
Mutual intelligibility, 49
Mutual orientation, 53
Nahuatl , 173
News , 65-67, 70, 80
Normal order, 44, 49
OPERA, 313, 317-323, 327, 331-333
Opposition, 409, 410, 416, 417, 429
counter opposition, 410, 417, 418,
427,433,435-437
Organizational routines, 50
Organizationally relevant objects, 57,
58
Partial perspectives, 49, 52
Past events reconstruction, 406, 410,
411,416,417,419,421 ,427,433,
437,438
Perception , 17, 112, 114, 116, 119,
121, 123-126, 131, 135, 145,385393,396,402
visual, 386, 387, 393
Perceptual development, 391, 403
Peripheral monitoring, 51
Peripheral participation, 418, 427, 437
Philosophy , 117
Piagetian theory, 225, 228, 267-269,
362, 363
Planner system, 313, 317
Pragmatics, 89
Problem-centered topics, 411, 412,
437,438
Problem resolution, 87, 90, 91, 94106
Problem solving, 361, 363, 365, 379,
381
collaborative , 312-314,317,323,
326, 327
computer based, 224, 227, 228, 230,
234, 235, 269, 272
group, 24, 30, 31, 37, 39,47-49, 51
mathematical, 288-290, 292-294,
296, 300, 303, 306-309
Problematization, 406, 410, 412, 413416,419,421 ,422,425,437,438
Subject Index
Problematizer/problematizee, 410,
413,415-419,427,431,433,435,
437
Professional design, 56, 57
Prompting, 178, 179
Psychology, 132, 206, 224, 236, 266,
387, 444
Psychosocial support, 144, 145, 149,
151, 165
common knowledge, 146, 162
formulating memberships, 146, 147,
149, 150
paediatric renal care, 144-146, 149151, 153, 154, 159, 160, 164, 165
problematic of normality, 149, 151,
154, 163, 164, 221
Rationalilty, 90, 91
Reading a scene, 44,47
Reasoning, 3-4, 15
as action, 4
practical, 52
principles , 228-292, 296, 297, 300,
303, 304, 308
scientific , 364, 371, 380
Repeated groupings, 300
'Representational devices, 46
Rhetoric, 443, 446-448, 450, 455
Samoan American community, 169,
171
Samoan
culture, 182
language, 182
Scaffolrung, 228,407,409
Schooling, 245, 246, 261, 263
Semiotic tools, 267-270, 279, 282,
283,443,446,447,452,454,455,
458
Sensemaking, 49
Shared meaning, 312, 314, 327
Shift changeover, 91
Signed numbers, 301-303
Situated identity, 206
473
Situativity theory, 4, 5, 12
Social comparison, 225, 235, 236
Social language, 443, 450, 452, 454456
Social marking, 226
Sociocognitive conflict, 225, 226,
228, 254, 267, 269-272, 279, 283
Sociocultural, 4, 12, 170, 268, 270
experience, 388-394,396,398,402,
403
theory, 385, 386
Sociology, 41,89, 117, 124,266
Spanish, 173
Speech
direct, 204-207, 211, 213, 214
private, 372, 379
reported, 203-205, 209, 210, 213,
214, 220, 221
Stories, 65-71, 72, 73,77,78
Structures of participation, 44, 45, 50,
53, 58
Study, 445, 446
Symbol processing, 336, 340-342,
348, 349, 352, 353
Symbolic representation , 14, 36-37,
46,94-106,117,175,279,294,
312, 314-322, 326, 327, 332, 333
non-symbolic representation, 320,
321, 332
Syncretism, 11, 172, 173, 176, 183,
192-194,196, 197
Synesthesia, 124, 128, 129, 134, 136
Systems of signs, 288, 289, 291, 292,
297, 303, 304, 308, 309
Talk (see also Conversation, Dialogue,
Discourse), 7, 8, 63, 66, 68-72, 79,
82,369,371-373,
accountable, 5-6, 8, 15
coherence in, 380,381
conflict talk, 409, 412, 418, 431,
437
dinner, 406, 409, 410, 412, 437
effects of, 371-373, 381
474
mediated by computers, 372, 381
therapy, 203, 207, 208, 215-218 ,
219, 220
Task completion , 181
Technology, 64, 83, 249, 250
of accountability, 50, 58
communication, 45, 47,53,58,
443, 444
information , 45
innovation, 275, 276, 283
Temporal order, 42, 44-46, 50
Territories, 43
Text, 68-72, 75, 76, 81, 82
Tools, 340
in measurement, 290, 304-309
Tower of Hanoi, 228
Transactive discussion, 361, 372, 374,
376, 378-381
categories, 362, 365
types of, 362, 363, 365
Transactive reasoning, 362-368, 372,
373, 375, 376, 378-380 .
Transformations
cognitive, 443
discursive, 443
Turn-taking (see also Conversational
Processes), 408, 431, 437
Typified action sequence, 49, 50
Uncertainty
discursive analysis of, 159-164
discursive reasoning, 145, 146, 150,
153, 154, 156, 158, 160, 162, 163,
165
Utterances (see also Verbalization),
451-454
Verbalization (see also Utterances),
367, 373, 379, 380
Vision , 112, 118
Visual behavior, 69, 72, 74-79 ,83
Vygotskian theory 2, 4, 16, 124, 128,
225,228,267,268,363,407,449
Subject Index
Word processing, 265, 267, 275, 280282
history of, 265, 275-279
Work practices, 43-45,51-53,57,6567,111,1 26,133,134,248-250,
252-254, 262
Workplace
design, 43, 53, 56, 57,65, 254, 263
learning, 243, 248, 249, 253, 261263
acquisition, 243, 248, 249
coordination of knowledge
systems, 243, 248-252, 255, 256,
258, 260
organization , 258, 260, 262, 263
constructive, 258-260, 262, 263
procedural, 258-260, 263
Workplace Project, The, 43, 126, 135
Workspaces,44, 53, 54
Written algorithms, 292-297, 300
67:
76:
78 :
80 :
81:
84 :
85 :
86:
87 :
89:
90:
9 1:
92:
93 :
95 :
96:
97 :
104:
105:
107:
109:
111:
112:
113:
115:
116:
117:
119:
121 :
122:
125:
128:
129:
132:
133:
135 :
137:
140 :
1986. (ROB)
Vol. 26 : Modelling and Analysis in Arms Control. Edited by R. Avenhaus, R. K. Huber and J. D. Kettelle .
VIII, 488 pages . 1986 . (out of print)
Vol. 27: Computer Aided Opt imal Design: Structural and Mechanical Systems. Edited by C. A. Mota
Soares. XIII, 1029 page s. 1987.
Vol. 28: Distributed Operating Systems. Theory und Practice. Edited by Y. Paker, J.-P. Banatre and
M. Bozyigit. X, 379 pages. 1987 .
Vol. 29 : Language s for Sensor-Based Control in Robotics. Edited by U. Rembold and K. Hormann .
IX, 625 page s. 1987. (ROB)
Vol. 30 : Pattern Recognit ion Theory and Applications. Edited by P. A. Devijver and J. Kittler. XI, 543
pages . 1987 .
Vol. 31: Decision Support Systems: Theory and Application. Edited by C. W. Holsapple and A. B.
Whin ston . X, 500 pages. 1987 .
Vol. 32 : Information System s: Failure Analysis. Edited by J. A. Wise and A. Debons . XV, 338 page s.
1987 .
Vol. 33 : Machine Intelligence and Knowledge Engineering for Robotic Applications. Edited by A. K.
C. Wong and A. Pugh. XIV, 486 pages . 1987. (ROB)
Vol. 34 : Modelling , Robustness and Sensitivity Reduction in Control Systems. Edited by R.F. Curtain.
IX, 492 pages . 1987.
Vol. 35 : Expert Judgment and Expert Systems. Edited by J. L. Mumpower, L. D. Phillips, O. Renn and
V. R. R. Uppuluri. VIII, 36 1 pages . 1987 .
Vol. 36 : Logic of Programming and Calculi of Discrete Design. Edited by M. Broy . VII, 415 pages.
1987 .
Vol. 37: Dynamics of Infinite Dimensional Systems. Edited by S.-N . Chow and J. K. Hale. IX. 514
pag es. 1987.
Vol. 38: Flow Control of Cong ested Networks. Edited by A. R. Odoni , L. Bianco and G. Szeqo . XII,
355 page s. 1987.
Vol. 39 : Mathematics and Computer Science in Medical Imaging. Edited by M. A. Viergever and A.
Todd-Pokropek. VIII, 546 pages. 1988 .
Vol. 40 : Theoretical Foundation s of Computer Graphic s and CAD. Edited by R. A. Earnshaw. XX,
1246 pag es. 1988 . (out of print)
Vol. 41: Neural Computers. Edited by R. Eckmiller and Ch . v. d. Malsburg . XIII , 566 pages . 1988.
Reprinted as Springer Study Edition 1989, 1990.
Vol. 42 : Real-Time Object Measurement and Classification . Edited by A. K. Jain. VIII, 407 pages .
1988. (ROB)
Vol. 43 : Sensors and Sensory System s for Advanced Robots. Edited by P. Dario. XI, 597 pages .
1988. (ROB)
Vol. 44 : Signal Processing and Pattern Recognition in Nondestructive Evaluation of Materials. Edited
by C. H. Chen . VIII,344 pages . 1988. (ROB)
Vol. 45 : Syntactic and Structural Pattern Recognition. Edited by G. Ferrate,T. Pavlidis, A. Sanfeliu and
H. Bunke . XVI, 467 pages . 1988. (ROB)
Vol. 64: Sensory Robotics for the Handlingof Limp Materials. Edited by P. M. Taylor. IX,343 pages.
1990 . (ROB)
Vol. 65: Mapping and Spatial Modelling for Navigation. Edited by L. F. Pau. VIII, 357 pages. 1990.
(ROB)
Vol. 79: Logic, Algebra, and Computation. Edited by F. L. Bauer. VII, 485 pages. 1991 .
Vol. 80 : Intelligent Tutoring Systems for Foreign Language Learning. Edited by M. L. Swartz and
M. Yazdani. IX, 347 pages . 1992. (AET)
Vol. 81 : Cognitive Tools for Learning. Edited by P. A. M. Komrners , D. H. Jonassen, and J. T. Mayes .
X, 278 pages. 1992 . (AET)
Vol. 82 : Combinatorial Optirnization . New Frontiers in Theory and Practice . Edited by M. AkgQI, H. W .
Hamacher , and S. TUfekg. XI, 334 pages. 1992.
Vol. 83 : Active Perception and Robot Vision. Edited by A. K. Sood and H. Wechsler. IX, 756 pages .
1992 .
Vol. 84 : Computer-Based Learning Environments and Problern Solving. Edited by E. De Corte, M. C.
Linn, H. Mandl, and L. Verschaffel. XVI, 488 pages. 1992. (AET)
Vol. 85 : Adaptive Learning Environments. Foundations and Frontiers. Edited by M. Jones and P. H.
Winne. VIII, 408 pages. 1992 . (AET)
Vol. 86 : Intelligent Learning Environrnents and Knowledge Acquisition in Physics. Edited by
Vol. 87: Cognitive Modell ing and Interactive Environrnents. With demo diskettes (Apple and IBM
compatible). Edited by F. L. Engel, D. G. Bouwhuis, T. Basser, and G. d'Ydewalie. IX, 311 pages .
1992. (AET)
Vol. 88: Prograrnming and Mathematical Method . Edited by M. Broy. VIII, 428 pages . 1992.
Vol. 89 : Mathernatical Problern Solving and New Information Technologies. Edited by J. P. Ponte ,
J. F. Matos , J. M. Matos, and D. Fernandes. XV, 346 pages. 1992. (AET)
Vol. 90 : Collabor ative Learning Through CornputerConferencing. Edited by A. R. Kaye. X, 260 pages .
1992. (AET)
Vol. 91 : New Direction s for Intelligent Tutoring Systerns. Edited by E. Costa . X, 296 pages. 1992 .
(AET)
Vol. 145: Advanced Educational Technology : Research Issues and Future Potent ial. Edited by T. T.
Liao. VIII, 219 pages . 1996 . (AET)
Vol. 146: Computers and Exploratory Learning. Edited by A. A. diSessa, C. Hoyles and R. Noss .
VIII, 482 pages. 1995. (AET)
Vol. 147: Speech Recognition and Coding. New Advances and Trend s. Edited by A. J. Rubio Ayuso
and J. M. Lopez Soler. XI, 505 pages . 1995.
Vol. 148: Knowledge Acquisition, Organization, and Use in Biology. Edited by K. M. Fisher and M. R.
Kibby. X, 246 pages. 1996. (AET)
Vol. 149 : Emergent Computing Methods in Engineering Design. Applications of Genetic Algorithms
and Neural Networks. Edited by D.E. Grierson and P. Hajela. VIII, 350 pages. 1996 .
Vol. 150 : Speechreading by Humans and Machines. Edited by D. G. Stork and M. E. Hennecke. XV,
686 pages . 1996.
Vol. 152 : Deduct ive Program Design. Edited by M. Broy . IX, 467 pages . 1996 .
Vol. 153: Identification, Adaptation, Learning. Edited by S. Bittanti and G. Plcci. XIV, 553 pages. 1996 .
Vol. 154 : Reliability and Maintenance of Complex Systems . Edited by S. Ozekici, XI, 589 pages. 1996 .
Vol. 155 : Cooperation: Game-Theoretic Approaches. Edited by S. Hart and A. Mas-Colell. VIII, 328
pages . 1997 .
Vol. 156 : Microcomputer-Based Labs: Educational Research and Standards. Edited by R.F. Tinker .
XIV, 398 pages . 1996. (AET)
Vol. 157 : Logic of Computation. Edited by H. Schwichtenberg . VII, 396 pages . 1997 .
Vol. 158 : Mathematical Methods in Program Development. Edited by M. Broy and B. Schieder . VIII,
528 pages . 1997 .
Vol. 160 : Discourse, Tools, and Reasoning: Essays on Situated Cognition . Edited by L.B . Resnick,
R. Salj6, C. Pontecorvo and B. Bunge. XII, 474 pages . 1997 (AET)
Springer
and the
environ ent
At ringer we firmly believe that an
intern ational science publisher has a
special obligation to the environme nt,
and our corp orate policies consistently
reflect this conviction.
We also expect our business partners paper mills, printers, packaging
manufacturers, etc. - to commit
themselves to using materials and
production processes that do not harm
the environment. The paper in this
book is made from low- or no-chlorine
pulp and is acid free, in conformance
with international standards for paper
permanency.
Springer