This article was downloaded by: [The University of Texas at El Paso]

On: 18 August 2014, At: 01:42

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Science

Education
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/tsed20

Multivoicedness and univocality in

classroom discourse: an example
from theory of matter
a
Eduardo Fleury Mortimer
a
Faculdade de Educação , Universidade Federal de Minas
Gerais , Belo Horizonte ‐ MG, Brazil
Published online: 24 Feb 2007.

To cite this article: Eduardo Fleury Mortimer (1998) Multivoicedness and univocality in
classroom discourse: an example from theory of matter, International Journal of Science
Education, 20:1, 67-82, DOI: 10.1080/0950069980200105

To link to this article: http://dx.doi.org/10.1080/0950069980200105

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information
(the “Content”) contained in the publications on our platform. However, Taylor
& Francis, our agents, and our licensors make no representations or warranties
whatsoever as to the accuracy, completeness, or suitability for any purpose
of the Content. Any opinions and views expressed in this publication are the
opinions and views of the authors, and are not the views of or endorsed by Taylor
& Francis. The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and Francis
shall not be liable for any losses, actions, claims, proceedings, demands, costs,
expenses, damages, and other liabilities whatsoever or howsoever caused arising
directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan, sub-
licensing, systematic supply, or distribution in any form to anyone is expressly
forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 INT. J. sci. EDUC., 1998, VOL. 20, NO. 1, 67-82

Multivoicedness and univocality in classroom

discourse: an example from theory of matter

Eduardo Fleury Mortimer, Faculdade de Educação, Universidade Federal de

Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

Minas Gerais, Belo Horizonte - MG, Brazil

The analyses of talk amongst students in three episodes, video recorded from a teaching sequence on the
particle model for Brazilian students (age 14-15) is presented. The analysis has two dimensions: the
discoursive dimension is an attempt to understand how meanings are constructed in the classroom as a
result of discoursive interactions; the conceptual dimension comprises the description of the patterns of
evolution of ideas of matter based on the notion of a conceptual profile of matter, in which conceptual
and historical features of the atom concept are profiled with pupils' and adolescents' ideas on atomism.
From the students talk we can hear the interaction among different 'voices', representing different zones
of a conceptual profile of matter. In this interaction, the students resort to both 'authoritative' and
'internally persuasive' discourses. The analysis emphasizes the role of these two types of discourse and
of epistemological and ontological obstacles in the construction of scientific meanings in the classroom.

Introduction
The investigation of classroom talk, mainly in the area of social psychology and
linguistics, has generated results which have implications for science education.
Some authors, drawing on Vygostky (1978), consider that discourse is at the heart
of the study of teaching and learning, as language is, at the same time, both a
psychological and a communicative tool (Mercer 1996). According to these authors
we use language not only to communicate and to share meaning but also to make
sense of our experience, to constitute our thoughts.
Empirical studies of classroom talk have shown the implicit rules working on
classroom talk, the particular features of this kind of discourse (as, for example, the
I - R - F sequence), and how teachers use the discourse to guide and to evaluate the
learning process (e.g. Edwards and Mercer 1987, Newman et al. 1989, Lemke
1990, Mercer 1995). These studies have indicated that the analysis of classroom
talk should take into account the content of the discourse and how the context of
schooling frames the content and shape of this discourse.
Several authors have attempted to study the discourse of science and of science
classrooms (Lemke 1990, Sutton 1992, Halliday and Martin 1993, Scott 1996,
1997, Ogborn et al. 1996, Mortimer 1997). Scientific language, according to
Halliday (1993), has its own grammar, in which the functions of verbs and
nouns are different from those of everyday language. The processes of science,
which would require an entire phrase to be expressed in everyday language, are
nominalized in scientific language. T h e verbs function as a causal link between
these nouns/processes in a phrase. Through this grammatical metaphor (Halliday
1993) processes are put into a relationship in the construction of a scientific argu-
0950-0693/98 $12 · 00 © 1998 Taylor & Francis Ltd.
 68 E. F. MORTIMER

ment. The language of science is, therefore, a way of talking about the world that is
different from everyday language, even in its grammar. Studies of science class-
rooms demonstrate that they are populated by abstract entities (electrons, mol-
ecules and so on) whose meanings are constructed as much as through talk as
through experiments or practical activities. Scott (1996, 1997), for instance, refers
to a 'teaching narrative', a sequence of talk, and activities mediated by talk, that
enact the transformation of knowledge in the intermental plane of the classroom.
These studies also suggest 'that the way in which the teacher "talks around" the
evidence or activity is at least as important as the evidence or activity itself (Scott
1997:127).
The study presented here aims at relating the discoursive analysis of science
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

classrooms with conceptual profiling - a personal and conceptual analysis of the

development of students' ideas through which the content of the classroom dis-
course can be highlighted. Conceptual profiling (Mortimer 1995) is a way to
describe the evolution of students' ideas of matter that is based on the conceptual
and historical features of the atom concept (Gregory 1931, van Melsen 1952,
Knight 1968) and in studies on children's ideas of atomism (e.g. Piaget and
Inhelder 1941, Novick and Nussbaum 1978, Driver 1985). This article examines
the ways through which classroom discoursive practices frame the development of
students' ideas of a particular concept - the concept of matter. How different
'voices', representing different zones of conceptual profile of matter, frame the
interactions among students in three episodes, and what the roles of different
types of discourse are in overcoming epistemological and ontological obstacles in
the construction of scientific meanings, are the two rnain issues addressed. The
consequence of this analysis for constructivism theory is briefly discussed.

Categories for a conceptual profile of matter

The categories to trace the evolution of students' atomistic ideas and to determine
the obstacles to the learning of scientific ideas were developed from an analysis of
the conceptual profile of the concept of matter (Mortimer 1995). This notion
shares with Bachelard's epistemological profile (Bachelard 1968) the idea that an
unique form of thought is not enough to explain a single concept. Nevertheless,
each zone of a conceptual profile, besides its own characteristic epistemology, has
also an ontological component. An important feature of the idea of a conceptual
profile is that its 'pre-scientific' divisions are constrained by the epistemological
and ontological commitments of individuals. As these individual characteristics are
strongly influenced by culture, it is possible to define a conceptual profile as a
'superindividual system of forms of thought' (Marton 1981) that can be assigned to
any individual within the same culture. Despite the differences between individual
profiles, the categories by which each conceptual profile is drawn are the same. An
individual conceptual profile is strongly rooted in the individual's distinctive back-
ground, and is content-dependent, since it refers to a particular concept. At the
same time its categories are context-independent, as within a culture we have the
same categories by which the zones of a specific profile are determined. In our
Western, industrial civilization, the scientific divisions of the profile of a scientific
concept can be defined by the history of scientific ideas. The pre-scientific zones
for many concepts can also be defined, as a consequence of the last two decades of
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 69

intensive research on students' alternative ideas, that has identified the same sort
of conceptions related to the same scientific concept in different parts of world.
In this article, the categories of the conceptual profile of matter (Mortimer
1995) will be used to analyse the content of the students' discourse. The first zone
of a conceptual profile of matter is called 'sensible', which means appealing to the
senses, and it is characterized by the absence of any discontinuous notion of mat-
ter. A student who only has this notion of matter represents it as continuous,
without any reference to particles. Its main obstacle is the negation of the possi-
bility of the existence of empty space between the particles. Related to this concept
of matter there is a notion of the physical states of matter being closely linked with
external appearances and physical features of materials, which states, for example,
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

that a solid is hard and a liquid is soft (Stavy and Stachel 1985, Stavy 1988, 1990).
The second zone of the profile is called 'substantialist atomism'. The notion of
substantialism leads to the conclusion that, despite using particles in their repre-
sentations, the students think of such particles as grains of matter than can dilate,
contract, change state, and so forth. Students, thus, make an analogy between the
behaviour of the particles that they draw and that of the substance. They are not
referring to the atom as a scientific concept, but to grains of matter that show
macroscopic properties. This analogy between the macroscopic and the micro-
scopic worlds is the main epistemological obstacle to further progress for students
whose concepts can be classified in this region. Moreover, the fact that they use
particles in their representations of matter is no guarantee that they believe in the
existence of the vacuum between them. This is particularly important in the sense
that someone in this zone has not necessarily overcome the obstacle of the previous
zone.
The kind of ideas that characterize these two first zones of the profile of
matter, i.e. matter as continuous and 'substantialist atomism', are described in
almost all articles about students' ideas related to the concept of matter (e.g.
Piaget and Inhelder 1941, Doran 1972, Novick and Nussbaum 1978, 1981,
Nussbaum 1985, Griffiths and Preston 1992, Garnett and Hackling 1995).
Although it is not possible to relate all the students' ideas about matter to these
two zones, the notion of a conceptual profile of matter provides a theoretical back-
ground that can highlight the main characteristics of such ideas and its main
obstacles to the construction of scientific ideas about matter in the classroom. In
the previous descriptions there was no reference, for example, to the idea that a
molecule is 'macro' in size (Griffiths and Preston 1992). However, if students
think about atoms and molecules as grains of matter that show macroscopic beha-
viour such as dilating, melting and so forth, they should think of these grains as
'macro' in size as well.
The third zone of the profile of matter corresponds to a classic notion of the
atom as the basic unit of matter, which is conserved during chemical transforma-
tions. The atom is a material particle and its behaviour is governed by mechanical
laws, like any other body. Substances are made up of molecules that result from
the combination of atoms. Atoms of the same type have the same atomic number.
This article does not deal with the atom as a system of electrical particles nor
with the quantum mechanical zones of the profile of matter. Nevertheless, it is
important to realize that classical atomism still has some 'substantialist' character-
istics as a legacy of its mechanistic origins. Despite the epistemological difference
between classical atomism and the other two areas of the profile, all these concep-
 70 E. F. MORTIMER

tions consider the atom as a kind of material thing, a basic block from which
substances are built. In this sense, all these 'atoms' belong to the same ontological
category. The main difference is that, in a classical view, we cannot attribute all
material behaviour to atoms, just because some forms of behaviour (such as melt-
ing, boiling, dilating) are a consequence of the motion of atoms, molecules or ions
in a vacuum and of the interaction between them, which can vary as the energy of
the system is modified. Consequently, an individual atom does not show properties
like boiling or melting points; these are interpreted as resulting from aggregating
the behaviour of a great number of them in macroscopic amounts. Nevertheless, a
classical atom shows some other material properties like mass, volume, radius, etc.;
it is a material thing, that belongs to the ontological category of substance. Matter
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

is only changed to another ontological category with quantum mechanics, where

particles of matter are seen as quantum objects described by mathematical equa-
tions. The 'reality' of those particles is the subject of a polemic that began with the
Einstein-Bohr debate (Bohr 1935, Einstein et al. 1935) and is still far from reach-
ing an agreement.

Teaching theory of matter for a conceptual profile change

There are some implications for establishing a strategy for teaching elementary
atomism that follow from the categories of the conceptual profile of matter. As
suggested in previous work (Mortimer 1995), the instructional approach and its
steps depend on the specific epistemological and ontological features of the zone of
the conceptual profile to be taught. In addition we should consider two distinct
parts in the learning process, as a consequence of using the conceptual profile
notion. The first corresponds to the acquisition of the concept in a specific profile
zone, and the second relates to the pupils achieving consciousness of their own
profile, allowing comparison between different zones of the profile, as well as an
evaluation of their relative power. In this process, students become conscious of
the limitations of their alternative conceptions without necessarily giving them up.
Another implication for teaching elementary atomism that follows from the
categories of the conceptual profile of matter is how to elicit the students' views of
phenomena in order to access their profile before teaching. To achieve this it is
convenient to choose phenomena in which matter is in the process of transforma-
tion, such as dilation, compression, change in physical state and so on (Appendix),
which might somehow be experienced by the students themselves, in the labora-
tory or in their everyday life. If students are asked to draw models for a system
before and after any sort of transformation they might use an intuitive atomism
(Piaget and Inhelder 1941, Bachelard 1975) which is an alternative to the scientific
atomism (Novick and Nussbaum 1978, Driver 1985) and has some obstacles to the
construction of the scientific concept that have to be addressed during the teaching
process.
Moreover, consideration of the conceptual profile of matter suggests that it
would be more sensitive to students' thinking to ask them to represent models
rather than draw what they imagine might be happening 'inside the material', or to
draw the material as if it were 'seen' through very powerful magnifying glasses.
These last questions might lead students to an answer based on external features of
the materials, since the approach suggested (seeing) would not be adequate for
dealing with constructed models.
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 71

The classroom as a social setting

The transfer of categories of the conceptual profile used to analyse evolution of
individual understanding to the social setting of the classroom presented several
problems. It is impossible to analyse classroom talk by looking for the ideas of
individuals and the way those ideas evolve. Evolution of ideas in the classroom
happens in a social space, as a result of the contribution of several individuals.
Newman et al. (1989) offer some insight into this sort of problem by suggesting
that the individual is not the most useful unit of analysis to deal with cognitive
change. Instead, they propose the 'analysis of social events involving negotiation
between participants with different understandings or analyses of the situation'
(P-62).
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

The analysis of the social and linguistic features of the classroom has incor-
porated some aspects of research on classroom talk. Studies of the development of
understanding in the classroom have generated several works in recent years and
some of them have a direct relationship with the sociocultural approach and the
semiotic analysis of Vygotsky.
Drawing on the work of Vygotsky, Bakhtin and Lotman, Wertsch (1991)
offers an insightful way to look at the generation of meaning in social settings
that can be used to analyse the construction of shared knowledge in the classroom
and that complements the approach based on conceptual profile outlined here. If
we look at the classroom as a space where at least two different languages - the
scientific and the everyday - are put in contact to generate new meaning, then
dialogicality and multivoicedness offer fundamental categories to analyse this
process. According to Voloshinov, any true understanding is dialogic in nature:
To understand another person's utterance means to orient oneself with respect to it, to
find the proper place for it in the corresponding context. For each word of the
utterance that we are in process of understanding, we, as it were, lay down a set of
our own answering words. The greater their number and weight, the deeper and more
substantial our understanding will be (Voloshinov 1973:102, cited by Wertsch
1991:54)
In this way, an utterance involves not only the voice producing it, but also the
voices to which it is addressed. This is a consequence of the multivoicedness that
characterizes the process of understanding.
Bakhtin's (1981) notion of voice refers to more than an auditory signal. It
involves the much more general phenomenon of the speaking subject's perspective
which is related to his/her world view. Related to the issues of dialogicality and
multivoicedness, Lotman's account of the functional dualism of texts in a cultural
system is very useful for analysing classroom talk. According to Lotman, the two
basic functions that texts fulfil are 'to convey meanings adequately, and to generate
new meanings' (Lotman 198: 34, cited by Wertsch 191:73). The first function -
called by Wertsch the univocal function of text — 'is fulfilled best when the codes of
the speaker and the listener most completely coincide and, consequently, when the
text has the maximum degree of univocality' (Lotman 1988:34). The second
function of a text — called by Wertsch the dialogic function — is to generate new
meanings.
In this respect a text ceases to be a passive link in conveying some constant informa-
tion between input (sender) and output (receiver). Whereas in the first case a differ-
ence between the message at the input and that at the output of an information circuit
 72 E. F. MORTIMER

can occur only as a result of a defect in the communication channel... in the second
case such a difference is the very essence of a text's function as a 'thinking device'.
What from the first standpoint is a defect, from the second is a norm, and vice versa
(Lotman 1988: 36-37, cited by Wertsch 1991:74).
Bakhtin's distinction between 'authoritative' and 'internally persuasive' dis-
course is clearly related to the univocal and dialogic functions of text. According to
Bakhtin (1981), in an authoritative discourse the utterances and their meanings are
presupposed to be fixed, not modifiable as they come into contact with new voices.
It is not a free appropriation and assimilation of the world itself that authoritative
discourse seeks to elicit from us; rather, it demands our unconditonal allegiance.
(Bakhtin 1981:343)
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

In contrast, the internally persuasive discourse seeks for the 'counter words', it is
'half-ours and half-someone else's', allowing dialogic interanimation.
The semantic structure of an internally persuasive discourse is not finite, it is open; in
each of the new contexts that dialogize it, this discourse is able to reveal ever new ways
to mean. (Bakhtin 1981: 346; emphasis in original).
According to Wertsch, for any text there is a tension between its univocal and
dialogic functions. Any discourse has authoritative and internally persuasive char-
acteristics. We can look to classroom talk searching for the context where one or
another feature predominates. T h e construction of a new meaning within a pre-
dominantly 'internally persuasive' discourse can be followed by recapitulations
where the 'authoritative' prevails. T h e use — or not — of the appropriate discourse
for each context can help us to explain the matching and mismatching between
teacher and student undertandings.

Analysing the classroom discourse

In this section, analyses of three episodes from a Brazilian Year 8 classroom (age
14—15), where the atomic model was taught in connection with the physical states
of matter, are presented. The data were collected by video recording both one of
the students' groups and the whole class. The episodes to be analysed belong to a
sequence of 12 lessons that lasted three weeks. The camera was introduced into the
classroom two months before this sequence began, so the students were able to
ignore its presence when the lessons began. All the classroom talk produced during
the 12 lessons was transcribed. From the whole transcription, 14 episodes of the
students' group discussion were selected for analysis such that they should repre-
sent all the students' ideas which arose in the teaching process. In this article the
analyses of three of these episodes are presented.
The lessons were based on group discussion alternated with whole class dis-
cussion. In the lessons the students had to discuss and then choose one of the
'models' they suggested in a pre-test, for the following phenomena: compression of
air in a plugged syringe; expansion of air heated in a test tube with a balloon over
its neck; the release of a vacuum in a flask connected to a large syringe; the diffu-
sion of gas odour in the kitchen as it escapes from its container; dilation, by means
of heating with the hand, of the alcohol (or mercury) column of a thermometer;
melting and vaporization of naphthalene heated in a test tube. The activities and
lessons that were planned for teaching atomism as a model to explain the physical
states of matter are summarized in the Appendix. This set of activities was planned
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 73

'">•

w
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

(b)

- So'/.cfc>

(c)
Figure 1. Examples of students' representations for matter undergoing
change (a) discontinuous representations, (b) mixed representations,
(c) sensible representations.

to fit a minimum of 10 lessons of 50 minutes each, and a maximum of 14 lessons,

depending on the individual classroom.
It is important to notice that the task of drawing models for matter undergoing
change might not be understood in the same sense by all of the students. The
examination of some of their drawings [figure l(c)] suggests that some of them
actually drew sensible representations of matter, which is an indication that they
do not understand the word 'model' in the same sense as the teacher does. This
distinction is important because when a student represents matter undergoing
change by means of drawings, even when his/her model does not coincide with
the scientific one, it can be considered a model if it goes beyond the sensible
aspects of the material, relating macroscopic aspects of the transformation to some-
thing that is imagined. Nevertheless, a second student whose representation does
 74 E. F. MORTIMER

not relate something imagined to the phenomenon but, instead, simply makes a
drawing of the situation [figure l(c)], cannot be considered as having the same
understanding of the word model as the first student.
Episodes 1 and 2 are based on classroom discourse that took place at the
beginning of the teaching sequence, during the discussion of the first two phenom-
ena (compression of air in a plugged syringe; expansion of air heated in a test tube
with a balloon over its neck). Two types of obstacles, belonging to the first two
zones of the conceptual profile of matter, were addressed in these episodes: the
negation of the empty space between particles and substantialism. Episode 3
belongs to the end of the teaching sequence and the analysis attempts to show
how the students use features of a classical view of matter — the third zone of the
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

profile - that they have acquired through teaching.

Episodes 1-3 show a pattern of evolution of the students' ideas as they dealt
with the obstacles to the construction of an atomistic model of matter. In Episode
I, the students discussed and chose a model for the compression of air, addressing
a question that the teacher had asked about the nature of space between particles.
This was the strategy used by the teacher to make these obstacles explicit in the
students' thinking.

Episode 1: What is this empty space?

1. Car: Which do you think is the best?

2. Edw: That one with the dots (Edw is referring to the dots into the drawing).
3. Car: I think that the most scientific is the one with the dots.
5. Raq: But, that one with the dots . . . What is that empty space there? In fact,
the air is all around, and not only in the dots. Then, we could colour all
around, this is what I think. Did you hear what he said? [referring to the
teacher] We should think about the space...
6. Car: But then it is not that with dots . . . It is that one [she points to another
drawing].
7. Raq: I think we could colour it, in this way: you make this first one very
light... then, when you compress it, it would be darker because it is
more concentrated. In this way there won't be space and there won't be
any doubt.
8. Car: But what happens is that air is not continuous like that... air is formed
by several particles.
9. Raq: [Ask Edw] What do you think?!
10. Edw: . . . The air... is as she said, air has particles.
I I . Raq: But, between the particles, is there a vacuum?
12. Car: No, b u t . . . here we don't have only air particles, we have other
particles . . . Nitrogen, pollution, dust, all sorts of things.
13. Raq: H u m . . .
There are at least two different conceptions of matter represented in the dis-
cussion which characterize two distinct zones of the conceptual profile of matter.
Car and Edw have chosen the atomistic model (utterances 2 and 3), as 'the most
scientific'. Raq was not convinced that air has particles, which is consistent with
the ideas that she had revealed in the pre-test — that matter is something contin-
uous. She made her view explicit in utterances 5 and 7. If the teacher did not refer
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 75

to the space between particles as something to be addressed, the students might not
take this as a problem. When Raq referred to the space between particles, at the
end of utterance 5, it was not her 'voice' that was speaking, but the teacher's, which
began to frame the student's negotiation of meaning. Raq would prefer to use a
continuous model because in this model 'there won't be space and there won't be
any doubt' (utterance 7). Although her view of matter was not accepted, her
question had to be addressed by her classmates.
The presence of the teacher's voice in the students' dialogue is an example of
the dialogicality that characterizes the generation of new meanings. In this text,
each utterance involves not only the voice producing it, but also the voices to
which it is addressed (Voloshinov 1973). The model that emerges at the end of
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

the episode (utterance 12) also reflects its predominately internally persuasive
nature. According to Bakhtin, 'the semantic structure of an internally persuasive
discourse is not finite, it is open; in each of the new contexts that dialogize it, this
discourse is able to reveal ever new ways to mean' (Bakhtin 1981: 346; emphasis
original). To counter Raq's objection ('But, between the particles, is there a
vacuum?') Car had to re-build her argument. Understanding the difficulty of
the idea of 'nothing' between the particles, she filled the empty space with particles
of things other than air ('we have other particles... nitrogen, pollution, dust, all
sorts of things').

Substantialism and atomism

Episode 2 offers an example of the presence of a substantialist view - the second
zone of the matter's profile - in the students' discourse. The students were dealing
with the second task of the teaching sequence (Appendix), drawing a model of air
before and after expanding as a consequence of being heated in a test tube with a
balloon over its neck.
Episode 2 took place after a long class discussion about the first phenomenon
(the compression of air in a plugged syringe), in which the teacher tried to deal
with the 'nature abhors a vacuum' obstacle. In doing so, he also argued that it
would be better to have a model which does not depend upon the nature of the
particle. In the teacher's words,
I can use that model [he is referring to the atomistic model] without having to talk
about the exact nature of the particle: I don't need to say whether the particle is a ball
or a square, whether it is elastic or non-elastic. It would be better to work with a
model which doesn't specify the nature of the particle, as I cannot observe it.
This is predominately authoritative utterance, and he used it when he was trying to
close the discussion. The mismatch between this authoritative teacher's discourse
and the understanding of the students in Episode 2 is evident, as they chose a
substantialist model, where the particles expanded themselves.

Episode 2: Expanding the particles

1. Car: What happened is that the particles got bigger.

2. Raq: Then the particles expanded . . .
3. Car: Expanded...
 76 E. F. MORTIMER

4. Raq: That is model three (she is referring to the number of the model that
they have on the sheet distributed by the teacher).
5. Car: What do you think, Gla?
6. Gla: Nothing... I don't know.
7. Car: Hey! We have to answer h e r e . . . We observed that the balloon filled up,
didn't we? But we have no answer: explain...
8. Raq: We have to explain: Air, when heated, expands.
9. Car: Expands. The air particles expand when heated because there is empty
space between the particles.
10. Edw: It is the air that expands.
11. Car: It isn't the air that expands, it is the particles that expand.
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

12. Car: Ah, now we have to draw h e r e . . . the model that we've chosen... (They
draw.)
13. Car: Here, see: we have to write the characteristics. What is the character-
istics of the first (tube)?
14. Edw: Normal.
15. Car: Normal. T h e particles have their normal size... now, in the second
(tube), they've got bigger, expanded, filling a bigger volume, haven't they?
16. Edw: Yes.
Episode 2 shows that Raq and Car had, at the outset of the episode, a consensus
about attributing the macroscopic property of expanding to the particles
(utterances 1 and 2). As Gla and Edw did not challenge their views, the text
became predominantly authoritative in nature. In contrast to Episode 1, where
'counter-words' had an important role in building a common understanding,
Episode 2 shows no space for voices other than the substantialist one. Car's
emphatic statement in utterance 11 in answering Edw's objection (utterance 10)
shows that she was not trying to negotiate her view any more, but only to find the
best way to express it, which was achieved in utterance 15.
Another characteristic of this episode is that the teacher's voice was not repre-
sented as in Episode 1. The students were not addressing a teacher's question nor did
they take any of his arguments against substantialism into account. The idea of empty
space between particles seems not to be a problem any more, and Car referred to it
in utterance 9 as an 'explanation' for the idea that particles could expand themselves.

The third zone of the conceptual profile of matter: the

classical atomism
Episode 3 occurred at the tenth lesson, near the end of the teaching sequence. It
shows how the students addressed a question which belongs to an activity aimed to
review and summarize the characteristics of the atomistic model (Appendix). The
episode shows the students using several characteristics of a classical atomistic
model which they have learned in these three weeks, since they did not use any
of these characteristics - motion, energy, arrangement - in their pre-tests nor in
the first two episodes.

Episode 3: Using the characteristics of an atomistic model

1. Raq: Then, how is energy related to the particles' motion?

 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 77

2. Ale: We only have to take into account... as a gas has more motion, it has
more energy. N o w . . . , the problem is how to explain gas.
3. Edw: It is this . . .
4. Raq: In the gaseous state the particles have more energy and their motion is
greater . . . wait a minute . . .
5. Ale: Please, you are driving us crazy . . .
6. Raq: Yes ... I am crazy too.
7. Car: If we were talking about solids, how much energy is associated with the
motion of these particles... Because it is solid and they vibrate, I
m e a n . . . t h e n . . . If we chose the liquid it means that they move them-
selves in groups, but if we chose the gaseous they move with more
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

freedom.
8. Raq: But it has to be logical.
9. Ale: Yes, you can't say it is because it is, it is because it i s . . .
10. Car: But it is because it is . . .
11. Gla: The gaseous..., when a gas is heated, its volume increases and its
motion increases as well.
12. Edw: But you can't say that a gas has motion only when it is heated.
13. Car: Good god! It is related to the particles . . .
14. Raq: Why is its motion greater than that of solids and liquids? Because the
particles move individually.
15. Edw: Because they can be considered as individual particles.
16. Car: Then!...
17. Raq: Then they can move faster.
18. Edw: Yes! Move faster.
19. Raq: Then we have to explain it this way. Then we can say that because they
move themselves . . . , no, they move themselves because of what? What
is this?
20. Edw: Because they are individual, the particles are individual... it is t h a t . . .
21. Ale: The problem is the quantity of...
22. Edw: Wait a minute . . .
[They stop and become thoughtful.]
23. Ale: I'd say that they are independent and move more quickly, then . . .
24. Raq: Yes, it is necessary to say that they move more quickly because . . .
25. Ale: How is energy related to motion? How is the amount of energy related
to motion? The greater the energy, the faster the motion. And that is all.
We notice here that the students did not have a full understanding of the question -
how is energy related with the particles' motion - when they began to deal with it.
At the outset Ale seemed to have understood the question, as he tried to relate
motion to energy in the gaseous state (utterance 2), but he was not convinced that
his statement could be the answer. The dialogue that follows shows how the
students try to construct this understanding by building an argument around
the notions of energy, motion and arrangement. In doing so they review several
characteristics of an atomistic model for solids, liquids and gases: which kind of
motion predominates in each state (utterance 7); how heat can affect the volume
and the particles' motion in a gas (utterance 11); that the particles have an intrinsic
motion, as 'y°u can't say that a gas has motion only when it is heated' (utterance
 78 E. F. MORTIMER

12); that the particles in a gas are far apart and therefore 'they can be considered as
individual particles' (utterances 15 and 20); and that being individual they move
more quickly (utterances 17, 18 and 23).
Some of these statements are not completely true. For example, the particles of
a gas do not necessarily move faster than the particles of a liquid, as kinetic energy -
and consequently motion - depends on the temperature and not on the particles'
condition of being individual or not. Nevertheless, the students in this episode are
framing a question with much more appropriate variables — motion, energy and
arrangement - than they framed the questions in Episodes 1 and 2. The teacher's
voice is present throughout the episode, but in a way that is different from Episode
1, where it was only referred to. In Episode 3 the teacher's voice becomes the
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

students' voices. Therefore, we have an appropriation of the teacher's voice by the

students and not only the teacher's voice speaking through the students. The
classical atomistic view of matter appeared initially on the intermental plane and
the students had several difficulties in using it, as shown in Episodes 1 and 2. As
this view of matter became internalized it moved from the intermental to the
intramental plane and the students became able to use it as their own view
(Vygotsky 1978, Wertsch 1991).
The discourse of Episode 3 has an internally persuasive nature as the students
are trying to build a suitable answer to the teacher's question. It also has an
authoritative nature, as the students' statements throughout the episode are not
in negotiation. The students used them as true statements to build a new argument -
how energy is related to the particles' motion. They use logical prompts to assure
the continuity of thinking through discourse (utterances 16 and 18) which is an
indication they they are building new thoughts in the intermental plane through
this discoursive interaction.
How did the students grasp this atomistic model? To answer this question we
would need to analyse other types of episodes, mainly teacher-students dialogue,
which is beyond the scope of this article. What could be suggested - although this
is based on data not analysed in this article - is that the alternation of internally
persuasive and authoritative discourse that characterized the students' talk was
also present in the teacher-students dialogue. In several of these dialogues the
teacher considered each student's argument and tried to build a counter argument.
Even when this strategy led to a model that was different from what he had in
mind he did not avoid the arguments or refuse the model suggested. In contrast to
this more persuasive discourse, the teacher resorted several times to assertive
statements and to univocal questions, that demand a unique and 'true' answer.
The alternation of these two types of discourse seems to be a very pervasive
characteristic of classrooms, one that contributes to the process of negotiation
and elaboration of meaning.

Conclusions
The analysis presented in this article articulates two dimensions of the science
classroom - the discoursive and the conceptual. In describing the evolution of
students' ideas in terms of a conceptual profile of matter, the obstacles to the
construction of a scientific idea and the difficulties students face when trying to
explain some phenomena with their own ideas could be highlighted. However, if
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 79

the analysis had had only this conceptual dimension, we could not say anything
about the way the new ideas emerged in the classroom talk.
The analyses of the episodes demonstrated that the students resorted to three
different zones of a conceptual profile of matter in different parts of a teaching
sequence, and showed the different kinds of discourse they used in building their
arguments. We might expect not to find an authoritative discourse when analysing
the students' talk. Nevertheless, the analysis of these episodes indicates that dis-
course in the classroom can be authoritative or internally persuasive independently
of being enunciated by the students or by the teacher. Both types of discourse were
present in the students' dialogues. An alternation between these two types of
discourse seems to be an important feature of the classroom talk. The phenomenon
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

of multivoicedness associated with this alternation enables an encounter between

two different voices in the classroom to take place: the scientific one, represented
by the teacher's discourse, and the everyday voice, represented by the students'
talk. When the students begin to share meaning with the teacher, we can observe a
shift in their discourse as they begin to use the teacher's voice as their own dis-
course.
The alternation between 'internally persuasive' and 'authoritative' discourses
seems to be part of the rhetorical design of the classroom talk. While the 'internally
persuasive' discourse allows alternative explanations and contradictory versions to
be considered through argumentation and justifications, the 'authoritative' dis-
course stresses the shared knowledge already constructed. In rhetorical psychology
(e.g. Billig 1987, Kuhn 1992), argumentation is not only a general feature of dis-
course, but of thinking as well. In this sense, the alternation of 'internally pre-
suasive' and 'authoritative' discourse in the intermental plane of the classroom is
important for developing conceptual thinking in the intramental plane.
The development of ideas in the classroom seems to depend at least as much
on social constraints as on empirical results, discrepant events of crucial experi-
ments. The same phenomena used to support the atomistic view can be interpreted
by students, using a substancialist or continuous view of matter, as shown in
Episodes 1 and 2. The construction of knowledge in classrooms is, therefore, a
guided process by which the teacher introduces students to the discourse of science
(Mercer 1995), helping them to build new meanings that could not be acquired
only through active participation in experimental activities and individual struggle
to make sense of these experiments.
The data presented in this article suggest that a movement from multivoiced-
ness to univocality characterizes the construction of meaning in science classrooms.
A contradictory consequence of the 'social pressure' towards the univocal voice of
scientific knowledge in the classroom is that this kind of pressure can hide the
epistemological and ontological obstacles that, despite being invisible, remain
alive. They are hidden by the circumstances, but they are still alive. The obstacles
that seemed to be overcome in classroom discussion, can reappear in group dis-
cussion, where the social pressure to conform is less than in the whole class, or in a
test, where the students are alone with their ideas.
The pressure towards univocality that characterizes the science classroom's
discourse suggests that knowledge is both constructed and transmitted in this
social setting, as the limits of what can be 'constructed' are clearly established in
the curriculum, which frames the functions and forms that the discourse can
assume. There seems to be no opposition or contradiction between transmission
 80 E. F. MORTIMER

and construction of meanings in the classroom, as some constructivist approaches

to science education have suggested. The teaching and learning process could be
referred to as a dialogue between scientific and everyday discourses where new
meanings are both constructed and transmitted.
The methodology of research presented in this article, articulating two dimen-
sions — conceptual and discoursive — of science classrooms, provides a useful tool
for understanding the construction of scientific concepts in the classroom. T h e
alternation between internally persuasive and authoritative discourse is only one
amongst several rhetorical aspects of science classrooms that can be revealed
through this type of research. The investigation of these aspects can provide
resources for professional development of teachers which combines the designing
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

of teaching strategies with rhetorical and reflexive tools.

Acknowledgements

This work was supported by grants from CNPq and CAPES (PADCT-SPEC).

References
BACHELARD, G. (1968) The Philosophy of No (New York: T h e Orion Press) (trans. La
philosophie du non (1940) by G. C. Waterston).
BACHELARD, G. (1975) Les intuitions atomistiques (essai de classification) (Paris: J. Vrin).
BAKHTIN, M. M. (1981) The Dialogic Imagination (edited by Michael Holqust, trans. Caryl
Emerson and Michael Holquist) (Austin: University of Texas Press).
BILLIG, M. (1996) Arguing and Thinking: A Rhetorical Approach to Social Psychology
(Cambridge: Cambridge University Press).
BOHR, N. (1935) Can quantum-mechanical descriptions of physical reality be considered
complete? Physical Review, 48, 696-702.
DORAN, R. (1972) Misconceptions of selected science concepts held by elementary school
children. Journal of Research in Science Teaching, 9, 127-137.
DRIVER, R. (1985) Beyond appearance: the conservation of matter under physical and
chemical transformations. In R. Driver, E. Guesne and A. Tiberghien (eds),
Children's Ideas in Science (Milton Keynes: Open University Press).
EDWARDS, D. and MERCER, N. (1987) Common Knowledge: The Development of
Understanding in Classrooms (London: Routledge).
EDWARDS, A. D. and WESTGATE, D. P. G. (1987) Investigating Classroom Talk (London:
Falmer).
EINSTEIN, A., PODOLSKY, B. and ROSEN, N. (1935) Can quantum-mechanical descriptions of
physical reality be considered complete? Physical Review, 47, 777-780.
GARNETT, P. J. and HACKLING, M. W. (1995) Students' alternative conceptions in chemistry:
a review of research and implications for teaching and learning. Studies in Science
Education, 25, 69-95.
GREGORY, J. G. (1931) A Short History of Atomism (London: A. and C. Black).
GRIFFITHS, A. K. and PRESTON, K. R. (1992) Grade 12 students' misconceptions relating to
fundamental characteristics of atoms and molecules. Journal of Research in Science
Teaching, 29, 611-628.
HALLIDAY, M. A. K. (1993) Some grammatical problems in scientific English. In M. A. K.
Halliday and J. R. Martin, Writing Science: Literacy and Discoursive Power
(Pittsburgh: University of Pittsburgh Press).
HALLIDAY, M. A. K. and MARTIN, J. R. (1993) Writing Science: Literacy and Discoursive
Power (Pittsburgh: University of Pittsburgh Press).
KNIGHT, D. M. (ed.) (1968) Classical Scientific Papers: Chemistry (London: Mills & Boon).
KUHN, D. (1992) Thinking as argument. Harvard Educational Review, 62, 155-178.
LEMKE, J. L. (1990) Talking Science: Language, Learning and Values (Norwood: Ablex).
 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM 81

LOTMAN, Y. M. (1988) The semiotics of culture and the context of a text. Soviet Psychology,
26, 32-51.
MARTON, F. (1981) Phenomenography: describing conceptions of the world around us.
Instructional Science, 10, 177-200.
MERCER, N. (1995) The Guided Construction of Knowledge: Talk Amongst Teachers and
Learners (Clevedon: Multilingual Matters).
MERCER, N. (1996) Las perspectivas socioculturales y el estudio del discurso en aula. In
C. Coll and D. Edwards (eds), Enseñanza, aprendizaje y discurso en el aula:
Aproximaciones al estudio del discurso educacional (Madrid: Infancia Y Aprendizaje).
MORTIMER, E. F. (1995) Conceptual change or conceptual profile change? Science and
Education, 4, 267-285.
MORTIMER, E. F. (1997) Múltiplos olhares sobre um episódio de ensino: 'Por que o gelo
flutua na água?'. In Anais do Encontro sobre Teoria e Pesquisa em Ensino de Ciências:
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

Linguagem, Cultura e Cognição: reflexões para o ensino de ciências. (Belo Horizonte:

Faculdade de Educação da UFMG.)
NEWMAN, D., GRIFFIN, P. and COLE, M. (1989) The Construction Zone: Working for Cognitive
Change in School (Cambridge: Cambridge University Press).
NOVICK, S. and NUSSBAUM, J. (1978) Junior high school pupils' understanding of the par-
ticulate nature of matter: an interview study. Science Education, 62, 273-281.
NOVICK, S. and NUSSBAUM, J. (1981) Pupils' understanding of the particulate nature of
matter: a cross-age study. Science Education, 65, 187-196.
NUSSBAUM, J. (1985) The particulate nature of matter in the gaseous phase. In R. Driver,
E. Guesne and A. Tiberghien (eds), Children's Ideas in Science (Milton Keynes: Open
University Press).
PIAGET, J. and INHELDER, B. (1941) Le développement des quantités chez l'enfant. Conservation
et atomisme (Neuchâtel: Delachaux & Niestlé).
POPPER, K. R. (1972) Objective Knowledge: An Evolutionary Approach (Oxford: Oxford
University Press).
OGBORN, J., KRESS, G., MARTINS, I. and MCGILLICUDDY, K. (1996) Explaining Science in the
Classroom (Buckingham: Open University Press).
SCOTT, P. (1996) Social interactions and personal meaning making in secondary science
classrooms. In G. Welford, J. Osborne and P. Scott (eds), Research in Science
Education in Europe: Current Issues and Themes (London: Falmer).
SCOTT, P. (1997) Teaching and learning science concepts in the classroom: talking a path
from spontaneous to scientific knowledge. In Anais do Encontro sobre Teoria e Pesquisa
em Ensino de Ciência: Linguagem, Cultura e Cognição: reflexoes para o ensino de ciencias
(Belo Horizonte: Faculdade de Educação da UFMG).
STAVY, R. (1990) Children's conceptions of changes in the state of matter: from liquid (or
solid) to gas. Journal of Research in Science Teaching, 27, 247-266.
STAVY, R. (1988) Children's conception of gas. International Journal of Science Education,
10, 530-560.
STAVY, R. and STACHEL, D. (1985) Children's ideas about 'solid' and 'liquid'. European
Journal of Science Education, 7, 407-421.
SUTTON, C. R. (1992) Words, Science and Learning (Buckingham: Open University Press).
VAN MELSEN, A. G. (1952) From Atomos to Atom: The History of the Concept Atom
(Pittsburgh: Duquesne University Press).
VOLOSHINOV, V. N. (1973) Marxism and the Philosophy of Language (trans. L. Matejka and I.
R. Titunik) (New York: Seminar Press).
VYGOTSKY, L. S. (1978) Mind in Society: The Development of Higher Psychological Processes.
M. Cole, V. John-Steiner, S. Scribner and E. Souberman (eds) (Cambridge, MA:
Harvard University Press).
WERTSCH, J. V. (1991) Voices of the Mind: A Sociocultural Approach to Mediated Action
(Cambridge, MA: Harvard University Press).
 82 MULTIVOICEDNESS AND UNIVOCALITY IN THE CLASSROOM

Appendix

Summary of the activities and lessons

(1) Classifying materials as solids, liquids and gases and identifying the criteria used in
the classification.
(2) Choosing, from the models presented in the pre-test and selected by the teacher,
the best model to explain:
2.1. Compression of air in a plugged syringe
2.2. Dilation of air submitted to heating in a test tube with a balloon over its neck
2.3. Vacuum in a flask connected to a large syringe
2.4. Gas odour scattering in the kitchen as it escapes from its container
Downloaded by [The University of Texas at El Paso] at 01:42 18 August 2014

For each of these four phenomena three steps were planned (after the second
phenomenon, step 1 was omitted as it seemed to be redundant):
(a) Identification of the features of the model drawn by the students (class as a
whole)
(b) Discussion in groups to choose the best model (groups of students)
(c) Discussion with the whole class to choose the best model
(3) Summarizing the characteristics of a model for gases:
3.1. How the model explains compression, dilation, vacuum and diffusion:
(a) Group discussion
(b) Class discussion
3.2. Generalizing the model to explain temperature and pressure of gases
(teaching exposition and class discussion).
(4) Generalizing the atomistic model to liquids and gases by choosing, from the mod-
els presented in the pre-test and selected by the teacher, the best one to explain:
4.1. Dilatation, by means of heating with the hand, the alcohol (or mercury)
column of a thermometer
4.2. Melting and vaporization of naphthalene heated in a test tube
For each of these two phenomena two steps were planned:
(a) Discussion in groups to choose the best model (groups of students)
(b) Discussion with the whole class to choose the best model
(5) Summarizing the atomistic model for solids, liquids and gases (group discussion
followed by class discussion).
(6) Generalizing the atomistic model to new situations:
6.1. Spontaneous dissolving of potassium manganate(VII) in water
6.2. Dissolving of sugar and salt in water
For each of these two phenomena two steps were planned:
(a) Discussion in groups to explain the phenomenon based on the atomistic
model
(b) Discussion with the whole class to choose the best explanation
(7) Different conceptions of solids, liquids and gases:
7.1. Analysing the criteria to classify solids, liquids and gases revealed in the first
activities, in the light of the new atomistic concept of solid, liquid and gas.
How does the new concept explain the properties of solids, liquids and gases
identified in the first activity?
7.2. Materials that 'resist' classification as solid, liquid or gas: glass, cloud, fog,
etc. The conception of colloidal solution.
For each of these two themes two steps were planned:
(a) Discussion in group to answer several questions related to the theme
(b) Discussion with the whole class to choose the best answer to each question