You Are as You Read: the role of texts in the production of subjectivity(1)
Jeff Evans (Middlesex University, UK)
Anna Tsatsaroni (University of Patras, Greece)


In this paper, we discuss an approach to the development of subjectivity that is social and based in social practices constituted by discourses (rather than on concepts of the individual such as "personality", "attitudes", "characteristics"). We shall discuss the way in which contexts and the material might be said to be "textualised", and hence the importance of intertextuality. We consider how these ideas (in relation to those of competing theoretical frameworks) can aid the understanding of the subject's "positioning", the basis of their readings of a mathematical problem, and their related thinking, affect and "performance".


Our theme is how texts produce subjectivity. We are interested in adults' numeracy and how their thinking about problems and their affect in particular situations is specific to the context. We argue that subjects' cognition, affect and the context are all based in or constituted by practices.

So we need to talk about texts first of all. The point that we want to develop is that the text is not something given: multiple readings of the text are possible; there isn't one privileged reading. There are several ways in which the text is fluid even though it looks like the words stay the same on the page. First, its meaning is not constituted once and for all, it's susceptible to change over historical time and it's susceptible to change over life of a subject. If we reread a text 10 years after, its meaning has changed; see for example, the menu in Fig.1. Second, at any given moment it's not unambiguous, it can be read in different ways.

The third thing is that the text is potentially open: there are going to be links outwards in all kinds of directions that are very important to understanding how someone understands the text and what they do with it (or what the text does to them) . Therefore, if multiple readings of the text are possible, if there isn't one privileged reading, then it seems reasonable to accept the post-structuralist position that the text is not the letters on the page, but it is that which is produced in a reading (Derrida, 1976).

What about context? People have been tempted to think about a context as something material, something outside of the text, as background. However, the problem is that this suggests that there is a "pre-discursive" context, that exists prior to language, or outside of it. There are a number of ways to deal with this problem.

Jean Lave (1988) presents a two-sided approach in studying adult shoppers. In one sense the context of their shopping, and of the sorts of numerate or mathematical things they do, is the supermarket as an arena with certain objective properties. At the same time she talks about it as a setting which is different for different shoppers. So in her work there is a tension between the supermarket as an arena, an objective site of shopping - which can therefore be studied as an object in political economy terms (cf. Lave, 1988, Ch.8) - and the supermarket as a setting which has different meanings for different shoppers.

Valerie Walkerdine (Walkerdine et al., 1989, ch.11) seems to play down the arena as objective, when she says "No easy materiality exists outside the practices through which it is read" (p.192). But this does not mean that she is seeing the setting as subjective, dependent on the individuals working or shopping or whatever. Rather, reality / materiality is read, interpreted, through practices, which are organised (regulated) by certain discourses and which therefore are themselves social: discursive practices. Thus we can perhaps offer a synthesis between the objective and the subjective, between the individual and the social.

We could say that the context itself has to be seen as, in some sense, textualised in the same way as we argued for text. In other words, for the context to be grasped, I need to be able to present it to myself. So there is no way that there is a pre-discursive context: to describe it and to engage with it is already to be engaged in certain kinds of language and discourse.

Although we often talk about contexts where texts are produced through readings, the idea of intertextuality points to the inseparability of text and context. As Derrida writes, "The phrase...there is nothing outside the text...means nothing else: there is nothing outside context.......the notion of text/context embraces and does not exhaust the world, reality , history. The text is not the book, it is not confined in a volume itself confined to the library. It does not suspend reference to history to the world, to reality, since these things always appear in a movement of interpretation which contextualizes them according to the network of differences and hence of referral to the other"(1988, 136-137).

Intertextuality is discussed by Fairclough (1992, Ch.4). Bakhtin shows the ways in which texts or utterances are shaped by prior texts and therefore can be seen as a response to them. So today's news reports on the Middle East relates to last night's and previous ones, and also - this is slightly harder to imagine - they relate to subsequent texts, in a sense they anticipate them, because those subsequent texts will be related back to today's. Thus, in David Lodge's Small World (1984), one of the main characters causes a stir in an international conference by claiming that T. S. Eliot's writing has affected Shakespeare's work. On the face of it, that is absurd - because how could Eliot writing in this century do that? But what he's saying is that when we go back and read Shakespeare, we re-read him through ideas of the 20th Century, which include those of Eliot.

A slightly more social, more material view of intertextuality comes from Kristeva (1986), a disciple and translator of Bakhtin. She talks about intertextuality being the insertion of history or society into a text and insertion of this text into history.

So what is the importance of all this, for learning mathematics? It is that the meaning of a mathematical operation or a mathematical signifier like a numeral or taking a percentage, is not universal. It is in principle ambiguous, and the way it gets its meaning is from that meaning being constituted by, being supported and shaped within, the discourses and practices in which it is inscribed or used. So if I am shopping, a unit price calculation has certain characteristics because it's come up while shopping. It would have different characteristics - perhaps in terms of the accuracy that would be required - if I were doing it as a problem in school maths. So, we argue, what appears to be "the same calculation" is not the same - because it is part of a different practice, which uses different terms which make different kinds of distinctions, and which represent different goals and values. When you're doing a calculation in shopping, you have different purposes and constraints than when you're doing it in the maths classroom. The calculations have to be more accurate in the classroom, because that is what is required, or what it takes to keep the teacher happy.

For example, in JE's interview (Evans and Tsatsaroni, 1994; Evans, 1999), there were two questions where a 10% calculation was required for both (for most respondents). The earlier question was Qu.2: What is 10% of 6.65?, given to them simply on a sheet of paper, whereas the later one, Qu.4, was put in a broader everyday context; see Fig. 1. I (i.e. JE - see NOTE 2) led into the task by asking them ... in a restaurant with a menu like this , how they would tip (if at all), and I asked them to choose a dish.... So here's another example of text - a calculation of 10% - that has quite different meanings. So for Qu.4, the same piece of text, the menu, generated at least four different rules and practices for tipping.

To put this in a more general way, the way we understand text and context must be reevaluated, "textualized", since "Every sign, linguistic or non-linguistic, spoken or written, in a small or large unit, can...break with every given context..engendering and inscribing itself, or being inscribed, in new contexts" (Derrida, 1988, p. 79). Thus, contextual transformation remains an always open possibility.

So let us begin to try to describe what that "context" was like.... First, it clearly had a lot to do with language, with discourse. Second, there were also different social relations in a restaurant, or in an interview about being in a restaurant, than in an interview with a maths teacher, if that's how the students saw it. You tend to have different "resources" - you may have a calculator with you, or some restaurants actually work out the 10% tip for you! But above all they are different calculations in the whole sense of what is going on. because they are inscribed in, or they form part of, different activities or practices where different things are at stake. So we look at the context as something based in, formed by, or constituted by, practices - indeed practices based in language so discursive practices. (eg everyday activity, a school maths activity, etc.) - each with its own set of social relations and resources.

That leads us on to the question how is subjectivity produced within these discursive practices.There are available a number of theoretical frameworks in maths education for investigating how subjectivity is produced, how a person's understanding of the meaning of a mathematical object is produced, how their thinking and feeling about maths has developed. First are two sorts of cognitive approaches, one relatively transitory, the other more durable: on the one hand, we could look at their cognitive processes of thinking and their misconceptions. Or on the other hand we could look at something a bit more durable such as cognitive style or a notion of an "ability" - perhaps in percentages, or in fractions. Third, we might look at affective characteristics, e.g. affective barriers to doing maths as a fairly permanent characteristic of a person, maths anxiety, beliefs in mathematics as a male domain. Or affective barriers in problem-solving processes (e.g. McLeod and Adams, 1989). Then there are social structural determinants for example social class, or gender itself. Finally, there is the notion of "experience": with adults; it is tempting to explain differences in performance as being due to experiences they have had as a waitress or experiences they've had as somebody who had to eat out a lot because of a job. Again, it can be tempting to think of this experience as based on a "reality" - which is somehow simple, pre-discursive, not depending on language in the way we argue the text and context are. However again we can make the same points we made in connection with context. Experience in order to have any effect must be perceived by the subject, it must be presented and represented to itself and again that is done through the medium of language. It cannot be done pre-discursively, on non-discursively. As with the term reality, "experience" appears in a movement of interpretation which contextualizes it.

What we argue, on the question of how is subjectivity produced, is that texts put subjects into positions, subject positions, so school maths texts, say, may put girls into the position of somebody who can't really do anything very mechanical, or somebody who's a helper in the classroom, or somebody who's really not concerned with the more intricate or powerful matters of business in the office but is relegated to a sort of subordinate role. However, the problem about positions, or subject positions, is that it may feel somewhat determinist: if texts position someone, there isn't much space left for their freedom or their agency. For example, the early work of Foucault focussed a lot on how institutions like the mental asylum, the clinic, the prison in 18th and 19th century France produced certain kinds of subjectivity in the people working in them, not just in the inhabitants but also the warders.

As an alternative view, we consider a way to acknowledge the subject's agency, their choice: "Man makes his own history but not in the conditions of his own choosing," propounded by Marx in his Theses on Feuerbach (1845 / 19xx). We read that as an attempt to balance the constraint and partial determinism in the historical conditions in which people find themselves, with the possibilities of freedom (for a class, at least) to make their own history. Therefore, in trying to avoid the over-determinism entailed in investigating subject-positions only, when we were looking at how JE's students approached any of these problems, we postulated a two-stage process. First of all we tried to analyse the situation in a way similar to the way Foucault did (or Walkerdine, 1984) although not in such detail, looking at what discourses or what practices are in play in this situation of the interviews. We thought that we could analyse the discourses that might possibly be positioning subjects in general here. In this general analysis, we thought for all problems that there was more than one practice at play in the situation.

Second, JE then read the transcripts, and tried to make a decision on which of those practices at play, or mix of practices, was actually positioning that particular person - and therefore was related to all the ways of producing subjectivity - making available certain ideas, giving an emotional charge to the problem, giving them possibilities for critical reflection, having certain aims and certain things at stake.

So a two stage process: the first stage involves setting out in general the possible practices at play, acknowledging the partial determinism; and the second stage is judging the particularity and variation across different subjects, looking at the possibility for partial freedom. And this two-stage process is analytical: we are not claiming this is the way the subject sees it.

Thus when I showed people the menu in Qu.4, when I came to the point where I asked them about a 10% tip, the practices in play at that moment could be either of two: school maths or eating out. That is the general stage of the analysis. Then I looked at each particular transcript (including their worksheets, because they had paper to work on), to see what they had actually said and done, so that is the particular stage of my analysis. For example, some of them got out a pencil and used the paper to calculate: for me that was an indicator that they had their positioning in school maths because you don't normally use those resources when you're in the restaurant. But mostly, the analysis of transcripts was done by looking for the use of key signifiers, terms (or symbols) that I thought were located, or had meaning, in one practice but not another. Part of making sense of any situation for a subject is deciding - or recognising - what activity, what practice this is and therefore what discourses they are going to be able to draw on, to make conversation or solve a problem or whatever.

The result of that two-stage process we called positioning, to distinguish it from a more determinist view of this whole process which would tend to talk about positions or subject-positions, to mark that this is an attempt (at problematizing the distinction between structure/agency, or determinism / freedom). This positioning in our view both supports and constraints subjectivity: it gives resources for thinking for example, but constrains the kind of critical reflection you can do - in the sense that it puts some limits on the play of signifiers, therefore on the production of meaning.

What else there might be that might have an effect is a very interesting question. Might there be something 'residual' in the subject, something that the person carries with him/herself, rather than its being provided by the discourse out there - that might affect what is called up, which signifiers are activated (i.e. are inscribed in other contexts)? We certainly feel that calling up is going to depend on memory. However, again, like experience and like context, memory needs to be understood as something subjects presents to themselves using language, discourse. Memory is also textualized.


Applying these Ideas

As maths educators, we are particularly interested in the practice of school maths although, as argued here, it relates to many others (see e.g. the work of Merttens, Vass, Brown et al. on the relations between home and school practices). What can we say about how subjectivity is constructed in a maths classroom?

We could begin with a structuralist analysis of maths in a maths classroom, as an activity which is organized by classification and framing (e.g. Bernstein, 1996) whose strength depends on the way it is recontextualsed as a subject of the curriculum. Therefore, first it has to be analysed as such, in terms of what type of domain of practice is constructed - maths as an esoteric, a public domain, etc.(e.g. Dowling, 1998). Then, we need to consider how the way school maths is constructed as a practice affects the way different subjects (seen structurally as middle-class and working-class, boys and girls, etc) perceive it differently, how they are differently positioned; there is a range of studies of maths and science done within this approach (e.g. Dowling, 1998; Cooper, Dunne and Rodgers, 1997; Morais et al., 1995).

However, the idea of "positioning" points to something beyond the structuralist analysis, using social categories. The argument about textualisation means that every signifier can be inscribed in new contexts; it cannot be absolutely controlled or confined within the context of its supposed original production, nor within the context to which the school activity attempts to confine it, and which structuralist approaches attempt to study.

Therefore, these issues can be studied better with interview material, because in such a setting, the elements of the context seem to be more fluid.

Now let us look at an interview transcript from one of JE's subjects. These were students at the end of their first year in the social science degree. They all had to take maths as about 1/6th of their study time. I had given all students (almost 1000) a questionnaire at the beginning of the year. In addition, at the end of the year, I did semi-structured interviews with a subsample of 25, using stratified sampling to try to ensure a certain representation of men and women, pre-21 and mature students, and middle-class versus working-class students. The interviews included both life-history questions, e.g. on what had been their experience with maths, and a number of practical problems (see e.g. Fig.1), many of which were designed by Bridget Sewell (1981), in her study of the use of maths by adults for the Cockcroft Report. What was different about JE's use of these questions was his use of what I call "contexting questions", asked both before showing them anything mathematical, and afterwards - to get an idea of what the text of the problem meant to them in terms of their everyday activities and if any memories were brought up. So that led to a lot of fruitful material as well as (normally) an answer of some kind to a mathematical question.

The contexting question that was asked before was of the form: "Does this remind you of anything you currently do?" And the one afterwards was: "Does this remind you of any earlier experiences?" They were commonsensically formed but they led to a lot of very interesting material.

When we analysed in general terms the practices in play in the interview one was the practice of school or college maths and in that the two positions that are normally available are teacher and pupil. The other practice at play was an interview practice where I was a researcher and they were an informant or respondent and that clearly involved different power positions. It involves a different relationship to knowledge. So those were the two basic practices at play in general in the interview.

We thought that the research interview practice would lead to the student calling up other practices and in the case of the menu (Qu.4) for example, the other practice we call eating out at restaurants. There are quite a number of related positions available when you eat out: you can take somebody out or you can be their guest - that's one kind of pairing which has different power and other implications; you can "go Dutch" where on the face of it you're going out as equals, certainly in terms of paying. And also in restaurants there is a customer -- waitress/ waiter relationship. So all of those depending on which of those you've been positioned in, in the past, might affect how you would respond to this question.

So basically, in our general analysis, we saw those two practices were positioning the people in the interview. And the research interview in particular opened up space for them to call up other practices such as eating out or shopping or baking cakes, depending on the kind of (often everyday) activity that was recontextualized in the material used in the interview problem.

What indicators do we have for positioning? First, we had my scripted talk: I tried to talk about research rather than a test, and I tried to talk about using numbers rather than maths and so on. I tried to use terms that were more consistent with the research interviewer practice than with the academic maths. Also very important is, second, the unstructured talk of the subject; for example, in Qu. 3 on the changing gold price, if they talked about 'graphs' and 'gradients', I saw that as an indication that they had their positioning in college maths because they were the terms we used. If they talked about 'charts' and 'trends', I took those as equivalent terms but indicating that s/he was thinking about it from within what we might call business maths or business practices. Besides, third, describing the interview setting, there were, fourth, what I call general reflexive accounts, for each of the two year groups I interviewed. Fifth, I wrote a particular reflective account of my relationship with each individual student because I had different relationships with each.

Let's have a look at "Donald" (not his real name). I started off the interview with a few questions about how he'd used numbers in the past and so on. He was a man in his 40s, he'd worked in the London money markets before he came to the college. In the first semester or two, he got very interested in philosophy and also maths, though he hadn't been very good in maths before, and he decided he was going to do a social science degree so he was doing town planning. This was almost the end of the third term, within a four year degree.

Let us consider Question 3; see Fig.2. I ask him my first contexting question: "Does that remind you of anything that you do these days or that you've done recently?" and he says "some of the work we did in Phase One [the first two terms] but, if you ask me straight out of my head, what it reminds me of I worked once with a credit company and we had charts on the wall trying to galvanise each of us to do better than the other and these sodding things were always there and we seemed to be slaves to the chart". He goes on a few more lines and he says: "That's what that reminds me of, a bad feeling in a way. I felt that a human being was being judged by that bit of paper." What I want to do here is to see if we can judge what's been called up. The first thing he says is that it reminds me of the maths course that we did. But then he pulls right away and he says that if you ask me, what it reminds me of, is working with this credit company and the charts, and the bad feeling.

The "graph" as signifier links with "chart" and pulls him in two directions but we think more towards business maths: he remembers both practices, but what is more strongly called up is the business maths. A minute later or so, I go back and ask: "Does it remind you of phase one?" and he says "yeah , well, we did some questions like this and it was the run over the rise and that kind of thing...." Then, after a hesitation, "Trends, I suppose if you were judging a trend it would remind me of that", and then a few lines later "I like the fact that I can do a chart now, but even to do a chart like that now, I couldn't sit down and do it straight away....". He talks a bit more about his feelings about maths. "With maths I have to go back to the basic things all the time." What's interesting here is again it's almost as if he's called up both practices. He's gone back to college maths, he talks about "the run over the rise" (NOTE 3).

Our conclusion is that he moves back and forth between the two practices; he is positioned in both and he is able to use terms - and ideas - from both. So let us see what happens when I ask him to try these problems: "May I ask you which part of the graph shows where the price was rising the fastest?" He says, first of all,"if I was to make an instant decision I'd say that one" and he points to the time before lunch, "but obviously I'd want to make it on a count of the line wouldn't I, I'd count a line"; he measures along the line and the vertical rise - and comes to the right conclusion that it's going up fastest in the first part of the slope. I ask, "What was the lowest price that day?" and he says this one here and points to the graph and says 580 and then, just as he finishes he says, "For some reason the price went higher at the close...." That is, his positioning in business maths leads him to seek an explanation for the situation depicted in the graph.

Thus he seems able to address this problem within either business maths as an "instant decision", or within school maths, by "counting a line", because that's what he's learned to do in college maths. (He was probably the only one of 25 interviewees who took the trouble to do the latter: most of the others made an instant decision but didn't mark it as such). He can think within both discourses about this problem and he's called up rather more strongly the business discourses than the college maths, although the latter is fluent, comfortable.

Is he just drifting in and out of the two discourses, or more deliberately "crossing the boundaries"? Or is he in his mind closing the gap, in some sense, between his experiences in the money markets and the college maths because these confrontations with the text, the graph, are allowing him to create "bridges" or "channels" that may be somehow more permanent? We might say that the graph signifies within both practices, and hence facilitates intertextuality, by opening a "channel" between the two practices. (NOTE 4)

A number of people including Donald were making distinctions that might be analysed via Bernstein's concept of classification (Bernstein, 1996). For Bernstein, classification is an expression of power relations which organises an activity and constitutes practices. It leads one to apply a structural analysis to define how practices constitute the subject, according to its social position (class, gender,etc.) Bernstein can be seen as making a "continued attempt to link the societal, institutional, interactional and interpsychic realms and to demontrate how the microprocesses of schooling relate to the complex institutional and societal forces" (Sadovnik, 1995, p.25). Yet as others have argued (Ladwig in Atkinson et al., 1997), he maintains some notion of a coherent, unified rational subject.

My interviewees were making distinctions which appeared to be based to a great extent on feelings and what they liked and disliked and what they didn't want to have anything to do with or did. We find this in about half of the 25 interviews. Donald's distinction is captured by his talking about 'figures' which is what you use in business, and maths or 'formulas and rules' on the other hand; this distinction is related to the two practices - work maths versus college maths - "the figures were a job" which certainly had strong positive components for him, whereas maths "was there to trip you up". Figures have a point, were meaningful, whereas maths was divorced from reality. He was confident with figures, he was still frightened by maths. These are distinctions he makes throughout the interview.

Other examples: "Jean" articulates her distinction between the two exam courses - CSE Maths and CSE Arithmetic on the basis of 'usefulness', and whether it should be optional or compulsory; she relates this to a similar division in topics in the 1st year course. "Peter" uses the dimensions numerical / not, useful / not, and pressure / relief, in a somewhat fluid way to distinguish mathematics and physics, from the sorts of subjects he took at A-level and now college.

So we can see a very interesting thing about affect here. At the social level, as Bernstein argues, it is experienced as values, perhaps sacredness or profanity, wonder, identity, commitment to rules - whereas at the individual level, it is experienced as feelings such as anxiety, emotion, liking. In between, perhaps spanning the two 'levels', we have attitudes and beliefs. These two levels also correspond roughly to the two fields of sociology and psychology. Perhaps social semiotics is what can link them together.

Concerning the question of an element of freedom or agency for the subject. Scott (1991), looking critically at the notion of experience, argues that 'experience' cannot be seen as separated from the discourses that are the basis for the subjectís positioning. She, like us, argues that subjects are positioned discursively but that doesnít mean theyíre completely determined. There are several reasons for this. First, certainly to the extent that subjects are positioned in more than one practice, thereís a possibility of conflict between the practices in terms of aims, in terms of the way you calculate, the values, the feelings attached to signifiers. Donald's interview shows very well that the first possibility for escaping determinism is conflicts among the multiple discourses one is positioned in - or spaces or gaps between them. Secondly there may be contradictions within a discourse. Valerie Walkerdine (Walkerdine et al., 1989) has written about double-binds that girls find themselves in as far as doing well in maths:.youíre supposed to do well in maths but itís quite dangerous, you may be seen as less feminine. Itís not good as a girl to excel at things that arenít girlsí subjects. Third, weíve seen that signifiers can have multiple meanings and indeed they can be connected in chains that go outside the immediate text. Here is where the idea that discourses are not closed, is important. So Donald sees the graph / chart, and thinks about his work practices in the city. Another subject, Fiona, thinks about her father who is a stockbroker and there are all kinds of meanings come up that get attached to what seems to be clearly a mathematical object but it isnít simply that - since words have to be used in order to describe it or even to present it to ourselves. The words clearly have a shared cultural meaning but may also, to some extent, mean different things to different subjects. For example, the graph problem is also about gold: that might have a particular meaning for some subjects.

These are three bases for the determinism to be broken up. As a result, we can say that subjects do have agency, they do have freedom, they can choose to some extent, but they are not indivisible unitary individuals that are exercising their free-will. Their agency is created within the situations and statuses or positions that are conferred on them.

So to come back to Marx and the earlier quote: "Man makes his own history but not in the conditions of his own choosing". What we can say in parallel is: "Subjects have their own agency but not through discourses and positioning of their own choosing." So within the frame of the discourses that position us that we donít choose, we have a certain element of agency and choice, but itís not complete in the same way that Marxís historical actors had some freedom of action but not in conditions of their own choosing.

Thus "how you position yourself" - as an interviewee, for example - is certainly important. It depends on how you read the elements of a situation, and you may negotiate a change in this rule, and/or you may learn to change that positioning. However, the analysis here suggests that we should more precisely say: "how you adopt a positioning within the activity that positions you". As is clear now your intentions, rather than being simply freely chosen, are shaped within the discourse(s).

An important issue which is central to the possibility of social change is: Are there variations of a particular text which might allow the reader to be repositioned in a different subject-position, hence producing a basis for a different subjectivity? For example, is it possible to change the texts in such a way that the double bind is lessened for girls?

A first response is itís not so easy: feminists have tried to produce changes in language, but apparently there is a lot of resistance to change in discourses producing gender inequality. This resistance comes from individuals, of course, but also from society in general - that is, all of us to the extent that we repeat and reproduce the discourses. We do it all the time unconsciously, especially in areas that are very important to us, that are effectively highly charged as sexuality and gender are. (Discourses of gender are related, for example, to feelings about mothers and fathers.) Thus it is very difficult to produce radically different texts as it is harder to change highly charged discourses than we wish. (NOTE 5)

This raises the further question of what is effective in the long term. The interview (like most research) just sampled briefly subjects' lives: all we could really show was a transitory characteristic. If Donald was angry or feeling bad, we canít be sure whether that was only that day - although of course, as the interview allows him to talk about different times of his life and these things come up, then we can have a sense that they are based on repetitions. Now, these are not repetitions of "behavior" which is rewarded, in any simple behaviourist sense - but rather, a repetition of engagement with certain strings of signifiers, i.e. certain texts. So that for us is an important consideratiion in understanding the establishment of long-standing characteristics. Teachers (and others) repeat the same text that positions the llearner in the same way; they see a certain idea of being used again, they have certain feelings about it. The subject judges him/herself and evaluates him/herself. Repetition will establish some chains of signification - and limit the construction of others - better than if it there is no repetition.



The approach discussed here is based on the view that context, memory and the material may be said to be textualised, and hence texts play a crucial role in the production of subjectivity. We think this view is an important supplement to studies of the mathematics learning, affect and classrooms, which use either a more structural approach or interaction in the classroom.



1. An earlier version of this paper was given by JE at a University of Exeter School of Education Seminar in Nov. 94. The presentation in this paper has been influenced by the dialogic structure of the seminar, and we especially acknowledge the contributions of Dennis Almeida and Paul Ernest.

2. Since the interviews and the original analysis were done by JE, the use of 'I', in a number of places in the paper, simply indicates this.

3. Though that is not quite right for the gradient: it should be "rise" over "run".

4. A nice example of intertextuality is provided by Donald's use of "chart" within business discourses, which may have anticipated the later use of "chart" instead of "graph" in computer graphic packages - where terms have been chosen, presumably, so as to appeal to businessmen.

5. However, powerful institutions like the Conservative Party in the 1980s had as a political strategy, "discursive shifting", for example working on words like 'enterprise' (Fairclough, 1992; Lerman 1990).



(All places of publ. London, unless mentioned.)

Atkinson P., Singh P., and Ladwig, J.G. (1997), "Review Symposium" of Bernstein (1996), British Journal of Sociol. of Educ., 18, 1, 115-128.

Bernstein, B. (1996). Pedagogy, Symbolic Control and Identity: Theory, Research, Critique. London: Taylor & Francis.

Cooper B., Dunne M. and Rodgers N. (1997), "Social Class, Gender, Item Type, and Performance in National Tests of Primary School Mathematics: Some Research Evidence from England"; paper presented at Annual Conference of the AERA, Chicago, March.

Derrida J. (1976), OF Grammatology.

Derrida J. (1988), Limited INC.

Dowling P. (1998), The Sociology of Mathematics Education. London: Falmer.

Evans J. (1998), "Building Bridges: Reflections on the problem of transfer of learning in mathematics" Educational Studies in Mathematics: Special Issue on Context (forthcoming).

Evans J. (1999), Adults, Practices and Numeracy: Mathematical thinking and emotions in context. London: Falmer Press (in preparation).

Evans, J. & Tsatsaroni, A. (1994). Language and subjectivity in the mathematics classroom. In S. Lerman (Ed.), The culture of the mathematics classroom. Dordrecht, NL: Kluwer.

Evans, J. & Tsatsaroni, A. (1996). Linking the cognitive and the affective in educational research: Cognitivist, psychoanalytic and poststructuralist models. British Educational Research Journal: Special Issue on Poststructuralism and Postmodernism, 21(3), 347-358.

Fairclough N. (1992), Discourse and Social Change. Cambridge: Polity Press.

Hollway, W. (1989). Subjectivity and method in psychology: Gender, meaning and science. London: Sage.

Kristeva J. (1986), Word, dialogue and novel; pp. 34-61 in T. Moi (ed.) The Kristeva Reader. Oxford: Basil Blackwell.

Lave, J. (1988). Cognition in practice: Mind, mathematics and culture in everyday life. Cambridge: Cambridge University Press.

Lerman S. (1990), in R. Noss et al. (eds.), Political Dimensions of Mathematics Education (PDME-1).

Lodge D. (1984), Small World.

McLeod D. and Adams V. (1989), Affect and Mathematical Problem Solving. New York: Springer.

Marx K. (1845 / 19xx), Theses on Feuerbach.

Morais A.M., Fontinhas F. and Neve I. (1993), "Recognition and Realisation Rules in acquiring school science", British Journal of Sociol. of Educ., 13, 2, 247-270.

Sadovnik A. (ed.) (1995) Educational Knowledge and Pedagogy. Norwood NJ: Ablex.

Scott Joan W. (1991), The Evidence of Experience, Critical Enquiry, Summer, 773-797.

Sewell B. (1981), Use of Mathematics by Adults in Everyday Life. Leicester:ACACE.

Walkerdine V. (1984), Ch.4 in Henriques J. et al., Changing the Subject. Methuen.

Walkerdine V. et al. (1989), Counting Girls Out. Virago.


Figure 1 Details of the 10% questions used in the interview

Qu. 2: "abstract" 10%

(C) Does this remind you of any of your current activities?

What is 10% of 6.65?

(R) Does this remind you of any earlier experiences?


Qu. 4: 10% tip on selected restaurant meal

(The problem was introduced by reading out several contexting questions - CA, CB and A)

[See "Menu" below]

(CA) Do you ever go to a restaurant with a menu anything like this?...

(CB) Would you please choose a dish from this menu?...

(A) Suppose the amount of "service" that you leave is up to the customer: what would you do? ...

(B) Could you tell me what a 10% service charge would be?...

(R) Does this remind you of any earlier experiences?

The "menu" for Qu.4 in the interview.

CHICKEN Served with sweet corn, banana fritter,

MARYLAND bacon, fresh tomato, whole French beans,

jacket baked potatoes with sour

cream and chives or French fried potatoes.

Roll and butter.

Ice cream, or a selection from our cheese

board, biscuits and butter.


SEA FOOD Served with tartare sauce, whole French

PLATTER beans, jacket baked potatoes with sour

cream and chives or French fried potatoes.

Roll and butter.

Ice cream, or a selection from our cheese

board, biscuits and butter.


GRILLED Served with tartare sauce, whole French

TROUT beans, Jacket baked potatoes with sour

10 OZ cream and chives or French fried potatoes.

Roll and butter.

Ice cream, or a selection from our cheese

board, biscuits and butter.


Coffee Special blend black or with cream 27p

Connoisseur        Served in large goblet glass with


Connoisseur coffees include sugar unless

otherwise requested 67p


Figure 2 Question 3 in the interview

The London Gold Price - January 23rd 1980 (in $ per fine ounce)

This graph shows how the price of gold varied in one day's trading in London. Which part of the graph shows where the price was rising fastest? What was the lowest price that day?





Source: Evans (1998), based on Sewell (1981)