SOURCE: Weber, Jean Jacques. “Three Models of Power in David Mamet's Oleanna.” In Exploring the Language of Drama: From Text to Context, Edited by Jonathan Culpepper, Mick Short, and Peter Verdonk, pp. 112-27. New York: Routledge, 1998.

[In the following essay, Weber explores the interaction of the “social context” and the “cognitive context” in Oleanna.]

INTRODUCTION

Oleanna is David Mamet's recent and highly controversial intervention in the political correctness debate. It stages a confrontation between a male professor, John, and his female student, Carol. John is about to be granted tenure and, on the strength of this promotion, has started negotiations to buy a new house for his family. Carol is a rather shy and confused student who is afraid that John will fail her, and John treats her with a mixture of surface concern and underlying condescension. He asserts that he likes her and promises her a grade ‘A’ if she comes back to his office to talk about the course. When Act II opens, it turns out that Carol has accused John of sexual harassment. Her newly gained self-confidence contrasts with John's gradual loss of confidence, as he fears that he may not be granted tenure, that he may not be able to buy the house and that he may even lose his job. In Act III, Carol offers to withdraw her complaint to the tenure committee on one condition: he must sign a list of banned books, including his own. John indignantly rejects what he sees as an attack against academic freedom and when, in the final climactic scene, he finds out that Carol has lodged criminal charges of battery and attempted rape against him, he physically assaults and verbally abuses her, thus committing the very acts of which she had previously accused him.

The play is divided into three acts, each of which consists of an office-hour interaction between the professor and his student. These literary representations of office-hour interactions have greater complexity than real-life office-hour interactions in terms of what is usually referred to as discourse layering:¹

level 1: author-reader/audience discourse
level 2: inter-character discourse

In other words, the inter-character discourse in the play between John and Carol is embedded within the discourse between the dramatist David Mamet and his readers or audience. And if we look at the play as performance, we also have to allow for the interpretative contributions of the production director and the actors.

In the analysis that follows, I will largely focus on the inter-character discourse layer and consider the dynamics of the office-hour interaction in an educational institution such as a university. I trust that many readers—to the extent that they themselves are either students or academics—have taken part in such encounters and will therefore have certain expectations about them. For example, we assume that the main roles are played by a lecturer and a student (or more than one student). The former is the more powerful participant, whose authority relies on, first, his expert knowledge and pedagogic expertise and, second, on the external, social legitimation of his role as a lecturer. The student, on the other hand, is the less powerful participant, because she is the novice possessing less knowledge of the subject area that she is studying. I refer to the professor as ‘he’ and the student as ‘she’ because this accords with the gender distribution in Mamet's play.

Another plausible expectation that we may have about the office-hour interaction is that the dominant discourse-type associated with it is academic discourse, although the impersonal academic discourse may at times shade into a more caring, more personal type of discourse. This is what happens in Act I, where John makes up for Carol's insufficient socialization into academic discourse by trying to establish a positive affect bond between them. In Acts II and III, on the other hand, the dominant status of academic discourse is threatened or even subverted by Carol's increasing use of legal discourse. A particular genre such as the office-hour interaction can thus be seen to draw upon a range of discourses, or even become a site of conflict between institutionally and ideologically diverse discourse-types.²

The main point that we are concerned with here is how the characters manage these discourse conflicts, and how the reader and the audience interpretatively (re)construct the discourse processes attributed to the characters by the author. A narrow and exclusive focus on the text will not take us very far here; I shall try to show that we need to bring in aspects of the social and cognitive context. It is to these contextual matters that we turn in the next two sections.

THE SOCIAL CONTEXT: POWER

Power can be defined as the ability of people and institutions to influence or control the behaviour and material lives of others.³ Though ‘control’ has more negative connotations, ‘influence’ suggests that power can be exerted both in a positive and a negative way. In the positive sense, power is something that can be given to marginalized groups (e.g. women in patriarchal society, or students in an educational context) so that they are empowered. But power can also be used to constrain, oppress, disempower, and marginalize certain people or groups of people. Note that there is a continuum between the two extremes, so that power often has good and bad aspects, in the sense that, for example, giving power to one person may imply not giving it to, or taking it away from, somebody else. Hence there is a first important distinction to be made between power as enablement and empowerment versus power as domination and oppression.

The other important distinction that I want to make is between institutionally or socially legitimated power and discursively constructed power. Perhaps a good way of approaching it is via West and Zimmerman's (1985: 116) concept of ‘participant identities’. They distinguish between three types of participant identities:

1 ‘master identities’, which crosscut all occasions of discourse: these are our more permanent identities such as age, sex, social class; on all these dimensions, John has power over Carol, who is young, female and lower class;

2 ‘situated identities’, which inhabit particular social settings: these are less permanent identities such as professor and student; on this dimension, too, John has power over Carol;

3 ‘discourse identities’, which constantly shift between discourse participants: these are ephemeral identities created by the verbal activities that we engage in; for example, in many social situations, apologizing is a self-threatening act, whereby we put ourselves in the position of the powerless discourse participant; uttering a command, on the other hand, is an act which threatens our interlocutor's face or self-image, and puts us in the position of the powerful discourse participant (though it all depends on whether our command is taken up or ignored or rejected).

It is our master and situated identities which invest us with social and institutional power, a power which, as we have seen, has not a single source but multiple sources: age, sex, class, job status, etc. But social power is not completely stable; it constantly has to be renegotiated. The renegotiation is done through our discourse identities which provide us with discursive power so that, to some extent at least, power has to be seen as an effect of discourse.⁴ This means that any power relation is inevitably dynamic: it is possible for socially powerless participants to temporarily gain discursive power over socially powerful participants in particular discursive encounters. The powerful participants may tolerate or even encourage this as long as they do not feel threatened in their own superior social power. In that case, the powerless participants may be lulled into a false sense of being taken seriously and having a say in things, or they may become aware of this paradoxical state of affairs and experience the seemingly emancipatory moves as oppressive ones.

This, in fact, is the way Carol reacts to John's moves in Oleanna: she feels that rather than trying to empower her, he is really only interested in upholding his own institutional power and hence in oppressing her. One of the questions we shall address in the analysis below is why this should be so. I shall argue that it is largely due to the nature of Carol's background assumptions about teaching and power. In the remainder of this chapter I will try to uncover this underlying conflict in what Carol's background assumptions are, what John's are, and to what extent they clash with each other. But first of all I have to explain more clearly the role played by background assumptions in interpretative processes.

THE COGNITIVE CONTEXT: SCHEMATA

According to Durant and Fabb (1990: Ch. 7), the creation of meaning is an inferential process, which combines new information with information already stored in memory. The information in the mind is stored not individually but in chunks, sets of beliefs, assumptions and expectations; and it is these sets that I refer to as ‘cognitive models’ or ‘schemata’.⁵ If the speaker's schemata are highly similar to the hearer's, with many shared assumptions, they will find it easy to communicate and understand each other. In this case, speaker and hearer can be said to belong to the same interpretative or discourse community, all the members of which use highly similar assumptions in their inferential processing of discourse. Such a set of shared presuppositions, assumptions, beliefs, values and cultural practices constitutes a world-view, a version of reality which comes to be accepted as ‘common sense’ within that particular community. If, on the other hand, speaker and hearer belong to different discourse communities and hold differing world-views, they will nevertheless understand each other to the extent that they share at least some interpretative assumptions. If their background schemata are widely divergent, then the result is likely to be misunderstanding or even a breakdown in communication. In this case, we can say that the discourse participants are locked in a situation of schematic or presuppositional conflict.

In Act I of Oleanna, Carol and John use very different interpretative schemata, and therefore find communication extremely difficult. The conflict can be seen as that between an elaborated code and a restricted code (Bernstein 1971), with the latter being associated with disempowerment, marginalization and inarticulacy. Carol is the restricted-code speaker, who has to acquire a new ‘language’—namely, academic discourse. Indeed, John frequently provides ordinary-language synonyms for the more specialized lexical items that he uses (e.g. pointer for index (24), liking for predilection (31), model for paradigm (45), bill of particulars for indictment (63), happen for transpire (66)). But using a new language also means taking on a new identity (as a member of the academic discourse community) and acquiring a new belief-system, a new world-view. It is this process of socialization, as we shall see, that Carol seems to resist.

The important theoretical point is that our belief-systems, and the cognitive schemata in which they are enshrined, are not fixed and static, but are highly dynamic structures which are constructed in and through discourse, and are constantly revised and updated. Schemata, as Cook (1994: 188) says, ‘are used in processing, but also changed by processing’. So there is constant flux and change. For example, our schemata for teacher and student are influenced by our self-image (if we ourselves happen to occupy either of these social roles) and by our real-life encounters with particular teachers and students. But they are also influenced by the representations of teachers and students in the verbal and visual texts that we come across in our daily lives, particularly if a large number of texts use similar ways of representing them. In other words, our cognitive schemata are socially, culturally, discursively and intertextually constituted.

But the very fact that certain representations tend to dominate in a particular culture also works against change. These dominant representations are naturalized and become part of the culture's ‘common sense’. We do not question them any longer, and the prejudiced beliefs or stereotyped images that they may include become almost invisible. At one stage in the play, John reminds Carol that the minimum common ground that is needed for communication to take place at all is to look upon the other person as human (53). But cognitive schemata—precisely because of their stereotypical simplifications and overgeneralizations—encourage a depersonalization of the Other. Therefore it is important to be aware of the way in which our schemata are constituted. A critical awareness is the precondition for positive cognitive change, for identifying and deconstructing the more pernicious aspects of schemata, and helping towards the construction of alternative, more positive ones.

In Oleanna, we witness a schema change in Carol, but it is a negative one, towards greater stereotypicality and depersonalization: she moves from an early view of John as an empowering tutor to a later view of him as oppressor, male predator and obstacle to her empowerment who has to be ruthlessly removed. Why and how she shifts from the early to the later model is a basic question to which, in the rest of this chapter, I will try to provide an answer. In order to do this, we shall attempt to (re)construct the putative cognitive processes of the characters, to uncover the cognitive models and assumptions underlying Carol's and John's verbal exchanges. The relevant assumptions that we shall look at in the next section are those concerning the nature of teaching, power and the roles into which the two protagonists cast each other.

DAVID MAMET'S OLEANNA

Mamet's play thus presents us with an inherently unequal power relation situated within the institutional context of higher education: Carol goes to see her professor in his office in order to talk about her coursework. There is a big difference for students between having a casual chat with a lecturer and being in a tutorial. In the latter case, there are a wide range of assumptions and expectations, which both lecturer and student rely on as mutual knowledge. They include, first of all, the assumption that the student will go to the lecturer's office (rather than the lecturer going to the student's room). Some lecturers might require students to make an appointment. For instance, in Oleanna John points out to Carol that ‘this was not a previously scheduled meeting’ (13). Next, we assume that the student goes to the lecturer's office because the former wants to talk to the latter about a specific topic, a topic which usually will be connected in some way with a particular course that the lecturer is teaching. In other words, the student does not go to the lecturer for a general chat; on the contrary, there is a constraint on topic relevance that the student will have to observe. Consequently their verbal interaction will not be unfocused conversation, but goal-orientated, instrumental discourse. In this way, we can gradually build up a general teaching schema for the office-hour interaction, which is widely taken for granted within the academic community and which contains the following assumptions amongst many others:

1 the lecturer possesses some piece of information, knowledge or advice that the student needs in order to get on with her work;

2 the lecturer is both willing and able to pass on that information to the student.

These assumptions presuppose the lecturer's competence and pedagogic expertise, neither of which would be expected in a general conversation, where the above assumptions would not be in force and there would be fewer constraints on topic selection. In Oleanna, for example, when John is talking on the phone to his wife, Carol overhears him using the expression ‘term of art’ and questions him about it. John, however, assumes that she has come to talk about her essay, and is so intent on introducing this delicate topic that he fails to see the point of her question:

CAROL:
[Pause] What is a ‘term of art’?
JOHN:
[Pause] I'm sorry … ?
CAROL:
[Pause] What is a ‘term of art’?
JOHN:
Is that what you want to talk about?
CAROL:
… to talk about … ?
JOHN:
Let's take the mysticism out of it, shall we?
Carol? [Pause] Don't you think? I'll tell you: when you have some ‘thing’. Which must be broached. [Pause] Don't you think … ? [Pause]

(Mamet 1993 [1992]: 2-3)

What is going on here is that Carol is still operating within a conversational schema, in which she can develop whatever topic comes up, whereas John already operates within a teaching schema, presupposing that there is something specific and relevant to the course that Carol ‘want[s] to talk about’, that there is some ‘“thing” [w]hich must be broached’.

However, John eventually realizes that he has been too brusque; he apologizes and proceeds to answer Carol's question:

JOHN:
It seems to mean a term, which has come, through its use, to mean something more specific than the words would, to someone not acquainted with them … indicate. That, I believe, is what a ‘term of art’, would mean. [Pause]
CAROL:
You don't know what it means … ?
JOHN:
I'm not sure that I know what it means. It's one of those things, perhaps you've had them, that, you look them up, or have someone explain them to you, and you say ‘aha’, and, you immediately forget what …
CAROL:
You don't do that.
JOHN:
I … ?
CAROL:
You don't do …
JOHN:
I don't, what … ?
CAROL:
for …
JOHN:
I don't for …
CAROL:
no …
JOHN:
forget things? Everybody does that.
CAROL:
No, they don't.

(Mamet 1993 [1992]: 3-4)

The fragmented nature of the two interlocutors' turns points to a major dysfunctionality in the interaction. Carol is shocked that John does not know for sure or has forgotten what ‘term of art’ means. She seems to make a confusion here between a conversational schema and a teaching schema, by applying the presuppositions of the latter to the former. She expects John to be able to supply an exact definition of the concept ‘term of art’, although this information is not relevant to the course that John is teaching and in which he alone is normally expected to have the necessary expertise.

So is Carol simply guilty of a conversational faux pas here? Or could the fact that Carol expects expertise from John in this area, too, be a first indication that she is operating in a teaching schema of her own, the assumptions of which differ significantly from the office-hour interaction teaching schema set out above? If so, we would be witnessing a schematic conflict, with Carol expecting a much more wide-ranging expertise from John than would usually be the case.

In the next extract, Carol again tries to raise a general conversational topic, but this time John does not follow up her topic and instead imposes his own:

CAROL:
Oh, oh. You're buying a new house!
JOHN:
No, let's get on with it.

(Mamet 1993 [1992]: 5)

From here on, John is in the dominant position and in control of the topic, whereas Carol is reduced to a submissive position. She is almost inarticulate, unable to talk about her work, and is finally reduced to a barren enumeration of the physical actions that she has undertaken in her fruitless attempts at socialization into the specialist academic discourse which is so alien to her:

JOHN:
I know how … believe me. I know how … potentially humiliating these … I have no desire to … I have no desire other than to help you. But: [He picks up some papers on his desk.] I won't even say ‘but’. I'll say that as I go back over the …
CAROL:
I'm just, I'm just trying to …
JOHN:
… no, it will not do.
CAROL:
… what? What will … ?
JOHN:
No, I see, I see what you, it … [He gestures to the papers.] but your work …
CAROL:
I'm just: I sit in class I … [She holds up her notebook.] I take notes …
JOHN:
[simultaneously with ‘notes’]: Yes, I understand. What I am trying to tell you is that some, some basic …
CAROL:
I …
JOHN:
… one moment: some basic missed communi …
CAROL:
I'm doing what I'm told. I bought your book, I read your …
JOHN:
No, I'm sure you …

(Mamet 1993 [1992]: 5-6)

Researchers such as Erickson et al. (1978) distinguish between a ‘powerful’ and a ‘powerless’ speech style. Whereas the former is marked by disaffiliating, non-supportive interruptions, the latter is marked by hesitations, repetitions and uncompleted turns. Could John be seen as using a powerful speech style and Carol a powerless one in the above passage? After all, John frequently interrupts Carol or his turns overlap with Carol's, which might indicate his power; and Carol uses many hesitations, repetitions and uncompleted turns, which might indicate her lack of power. But, if we look more closely at the dialogue, we find that many of Carol's turns also interrupt, or overlap with, John's; and John, too, has hesitations, repetitions and uncompleted turns. It thus becomes extremely problematic to identify certain linguistic features as reflexes of a powerful speech style and others as reflexes of a powerless one. A more promising route might be to look at the differing functions that one and the same linguistic form can have in context: whereas Carol's hesitations and repetitions indicate her lack of power, John's can be interpreted as instruments of power, aiming to lessen or soften the anticipated antithetical emotional reaction that Carol may have to him.

John's evaluations of Carol's work are quite negative here: he repeats ‘no’ three times (‘no, it will not do’; ‘No. I see, I see what you, it … but your work’; ‘No, I'm sure you’). But he mitigates these negative evaluations by using a variety of redressive strategies to minimize the threat to Carol's face or self-esteem. Buck and Austin (1995: 65), relying on Brown and Levinson's (1987) politeness model, describe three different types of redressive strategies:⁶

negative face strategies: choosing more polite forms;
positive face strategies: emphasizing one's high regard for the hearer;
off-record strategies: performing the face-threatening act in an indirect way.

Significantly, John uses all three types: in the dialogue that immediately precedes the quoted passage, his language is almost exaggeratedly polite (‘You paid me the compliment, or the “obeisance”—all right—of coming in here.’ (5)), he emphasizes his feelings of empathy for Carol (‘I know how … potentially humiliating these …’), his basic solidarity with her (‘I have no desire other than to help you.’) and his high regard for her (in the dialogue immediately following the quoted passage, he refers to her as ‘an incredibly bright girl’ (7)); and he employs linguistic indirectness in order to avoid a direct assignment of blame. Instead of telling her how bad her essay is, he prefers to talk about ‘some basic missed communi[cation]’ between them. When he eventually gets round to discussing Carol's essay, he uses a series of questions which on the surface are requests for information and only function as accusations in an indirect way:

JOHN:
[Picks up paper.] Here: Please: Sit down. [Pause] Sit down. [Reads from her paper.] ‘I think that the ideas contained in this work express the author's feelings in a way that he intended, based on his results.’ What can that mean? Do you see? What …

(Mamet 1993 [1992]: 8)

Carol's unexpectedly forceful reaction to this indirect attack allows us to glimpse hidden potentialities in her which will not be fully revealed until Acts II and III:

CAROL:
I did what you told me. I did, I did everything that, I read your book, you told me to buy your book and read it. Everything you say I … [She gestures to her notebook.] [The phone rings.] I do … Ev …
JOHN:
… look:
CAROL:
… everything I'm told …
JOHN:
Look. Look. I'm not your father. [Pause]
CAROL:
What?
JOHN:
I'm.
CAROL:
Did I say you were my father?
JOHN:
… no …
CAROL:
Why did you say that … ?
JOHN:
I …
CAROL:
… why … ?

(Mamet 1993 [1992]: 9-10)

John feels that Carol would like him to be more like a father than a lecturer; he feels that she moves out of the teaching schema into a family schema, where parents support their children. Carol's reaction, however, shows that there is a basic misunderstanding here: she was still operating within the assumptions of the teaching schema, not a family schema. It gradually dawns on us that all along Carol has been operating in a teaching schema which is significantly different from John's. In other words, the conflict between them is a presuppositional one, a clash of schemata, which explains their incapacity to reach some form of common ground.

For Carol, assumptions like ‘doing whatever you are told by the lecturer’ (9) belong to the teaching schema. According to her, the lecturer has to ‘teach’ (11) the students, in the sense of passing some information or facts on to the students, who then also ‘know something they didn't know’ (12). If the students have difficulty in understanding, the lecturer has to provide support for them: ‘You have to help me’ (10). Her teaching schema is a highly specific and also highly traditional extension of the office-hour interaction teaching schema that we encountered above:

CAROL'S TEACHING SCHEMA:

1 the lecturer possesses knowledge and expertise in all areas;

2 the student lacks knowledge and expertise in all areas;

3 the student does whatever she is told by the lecturer (e.g. to read a particular book);

4 if the student understands, then she has acquired knowledge;

5 if the student is unable to understand, then the lecturer has to ‘help’ her;

6 the lecturer knows how to help the student.

John, on the other hand, feels that Carol is treating him as if he were her father. His teaching schema is also an extension of the office-hour interaction teaching schema, but includes more innovative and critical assumptions such as the following:

JOHN'S TEACHING SCHEMA:

1 the lecturer tries to ‘awake’ the students' interest (26);

2 the lecturer teaches the students how to think for themselves: ‘What do you think? … What do you think, though?’ (29);

3 the lecturer tries to ‘provoke’ the students (32);

4 the students are expected to be independent and critical, to ‘question’ things (33);

5 the lecturer can tell the students what he thinks, and then the students ‘decide’ (53);

6 the lecturer does not try to ‘fix’ the students.

(54)

We are faced here with a clash between two opposed teaching schemata, which I shall refer to as (Carol's) ‘power of’ model and (John's) ‘power to’ model. Carol sees power as the acquisition and possession of knowledge, with the lecturer possessing power because he possesses knowledge, and the student acquiring power by acquiring knowledge. She has a naive view of empowerment as a passive process of transfer. John, on the other hand, sees empowerment as an active process in which the lecturer gives the students the power to achieve their own goals. And this is to be arrived at not so much by the students acquiring knowledge but through their developing critical skills which open up new horizons of understanding.

John's schematic or background beliefs also lead him to treat Carol sensitively as a person. He shifts repeatedly from an impersonal academic discourse towards a more caring, more personal type of discourse, in which he and Carol play the roles of mentor and protégée rather than lecturer and student.⁷ In his attempt to make Carol feel better, John now tells her that he himself used to be considered stupid. Carol, however, is shocked by John's revelations: ‘People said that you were stupid … ?’ (Mamet 1993 [1992]: 17). The point is that John unwittingly destroys some of Carol's assumptions about education: the lecturer possesses expert knowledge and passes it on to the novice, the powerless student, and then the student comes to possess expert knowledge and gains power, too. John also tells Carol that he has problems with his work (21-2); again, this does not fit into Carol's ‘power of’ model, which stipulates that:

• the student has problems with work;

• the student goes to see the lecturer;

• and the lecturer solves the student's problems.

Finally, John tells Carol that tests ‘were designed, in the most part, for idiots. By idiots’ (23). All that Carol can offer in reply to this is a shocked ‘no’ (23). After all, in her model, exams have a clearly defined function: they test whether the student has acquired the necessary expert knowledge.

In this way, John destroys the main assumptions of Carol's ‘power of’ model. Will she now, as a consequence of the collapse of her own model, embrace John's ‘power to’ model? We might have expected that she would have experienced the ‘power to’ model as a truly liberating one; but what she experiences instead is a contradiction between theory and practice inherent in the ‘power to’ model:

CAROL:
… that it is prejudice that we should go to school?
JOHN:
Exactly. [Pause]
CAROL:
How can you say that? How …
JOHN:
Good. Good. Good. That's right! Speak up! What is a prejudice? An unreasoned belief. We are all subject to it. None of us is not. When it is threatened, or opposed, we feel anger, and feel, do we not? As you do now. Do you not? Good.
CAROL:
… but how can you …
JOHN:
… let us examine. Good.
CAROL:
How …
JOHN:
Good. Good. When …
CAROL:
I'M SPEAKING … [Pause]

(Mamet 1993 [1992]: 30)

Again and again John interrupts Carol, inhibiting her discourse and imposing his own meanings. This abuse of power contradicts his stated conviction that students should be taught to think for themselves. John's interruptions are seemingly supportive and co-operative, but in fact Carol experiences them as devices for maintaining power.

Thus a deep-seated hypocrisy is revealed within John's ‘power to’ model. And a fundamental emptiness has been revealed within Carol's ‘power of’ model. The result is a moral vacuum which is filled by yet another teaching model, which I shall refer to as the ‘power over’ model. For the two models of empowerment, Carol now substitutes a model of disempowerment and oppression, in which education is seen as an arbitrary relation of dominance and submission, which can be reversed.

And this is exactly what we find in Acts II and III: the normal roles have been inverted. Here, the lecturer has asked the student to come to his office (46, 59). Moreover, it is the lecturer who has a problem, who needs something that the student can provide him with. In Act II, Carol asks: ‘What do you want of me?’ (45), and in Act III: ‘What is it you want?’ (61). John still tries to pretend that Carol has a problem which he can help her solve: ‘Look, I'm trying to save you …’ (57), but Carol has already turned the tables on him. She has publicly accused him of politically incorrect behaviour, and therefore she now has power over him. He needs her to drop her charges; she, on the other hand, does not need his help any longer: ‘I don't think I need anything you have’ (49). On the contrary, she is the one now who decides what words and actions mean, who has the power of imposing her own meanings:

CAROL:
My charges are not trivial. You see that in the haste, I think, with which they were accepted. A joke you have told, with a sexist tinge … To lay a hand on someone's shoulder.
JOHN:
It was devoid of sexual content.
CAROL:
I say it was not. I SAY IT WAS NOT. Don't you begin to see … ? Don't you begin to understand? IT'S NOT FOR YOU TO SAY.

(Mamet 1993 [1992]: 70)

And John is the one who does not understand any longer, who in his worries about losing his job is reduced to virtual inarticulacy, e.g. ‘Well, I … I … I … You know I, as I said. I … think I am not too old to learn, and I can learn, I …’ (Mamet 1993 [1992]: 71).

Even the basic teaching situation has been inverted here, with John ‘learning’ and Carol taking over the role of the teacher: ‘I came here to instruct you’ (67). What she reveals to him is that the lecturer-student relation is an accuser-accused relation, that the ‘power to’ ideal with its attendant beliefs in free thought and free speech is illusory and impossible to realize in the oppressive reality of a ‘power over’ institution:

CAROL:
Why do you hate me? Because you think me wrong? No. Because I have, you think, power over you. Listen to me. Listen to me, Professor. [Pause] It is the power that you hate. So deeply that, that any atmosphere of free discussion is impossible. It's not ‘unlikely’. It's impossible. Isn't it?

(Mamet 1993 [1992]: 68-9)

The roles have been inverted. The student has become the powerful accuser and the lecturer the powerless accused—and the dominant discourse-type has shifted from academic to legal discourse, which allows Carol to gain discursive power over John:

CAROL:
Do you hold yourself harmless from the charge of sexual exploitativeness … ? [Pause]
JOHN:
Well, I … I … I … You know I, as I said. I … think I am not too old to learn, and I can learn, I …
CAROL:
Do you hold yourself innocent of the charge of …
JOHN:
… wait, wait, wait …

(Mamet 1993 [1992]: 71)

Carol's accusations followed by John's hesitant replies enact the prototypical question and answer structure of courtroom discourse. But Carol goes further than this. By formally accusing John of rape, and thus moving her case out of the educational setting and into a legal one, she gains not only discursive but also (in the terminology of West and Zimmerman) ‘situated’ power over him. In all these ways, Carol appropriates power for herself. However—and this is what is so negative about it—she is not trying to change the system which she experiences as an oppressive one into an empowering and emancipatory one, but simply appropriates as much power as possible for herself within the existing system, which ultimately of course only serves to uphold and reinforce it.

CONCLUSION

What I hope to have shown in this chapter is that it is not possible to interpret a text produced by a particular speaker as mere text, but that interpreting involves (re)constructing relevant portions of that speaker's social and cognitive context. In other words, understanding Mamet's play is not just understanding the text but also understanding the social context (power relations) and the cognitive context (background schemata), and the extent to which the two are enmeshed. Paradoxically, however, such a (re)construction of the cognitive ‘worlds’ of the characters and, by implication, of the author, can only be achieved through the reader's own socio-cultural schemata. And since you as a reader will doubtlessly use different assumptions from mine in your cognitive processing of the text, you may well come up with a different interpretation.

In this way, we cannot but use our schemata to process discourse; but we must not forget that this is a two-way interaction and that our schemata are also changed in response to the discourses that we process. More specifically, particular textual representations either harden the stereotypical elements contained within our schemata, or they have the opposite effect of breaking up their stereotypicality, forcing us to look at the world in defamiliarizing ways.⁸ I should like to suggest that Mamet's play has both these effects on its audience and readers: an effect of gender polarization and one of cognitive defamiliarization. On the one hand, it seduces us into a brutal ‘power over’ model by tempting us to empathize with John's final actions. Indeed, several reviewers and critics have commented on theatre audiences being elated or roaring with approval as John begins to assault Carol.⁹ In our support of John's stance towards academic freedom, and our opposition to Carol's fanatic and repressive attitude of political correctness, we are seduced into an equally fanatic wholesale rejection of political correctness. In the words of Kureishi (1995: 111), we could ask: ‘So who's the fanatic now?’ Is it Carol, or John assaulting Carol, or Mamet tempting his audience into supporting John's assault, or the audience who allow themselves to be seduced into supporting John's assault?

The critical reader or spectator will resist this textual seduction and be more concerned with Oleanna's ambiguities and its effects of cognitive defamiliarization. This is what I have attempted to do in the above critical discussion, using schema theory in order to sharpen our awareness of the ways in which the play forces us to rethink our assumptions about teaching and power. Whatever model of power in education we may have, it is bound to be shaken up by this play in which the ‘power of’ model is deconstructed and ridiculed right from the beginning, the ‘power to’ model is upheld and yet also discredited in more subtle ways, the ‘power over’ model wins and yet is also revealed in its full brutality and inhumanity. By exposing the different types of power relations inherent in the process of education, Mamet leads us to realize that education is not either a ‘power of’ or a ‘power to’ or a ‘power over’ process but all of them at the same time; that there is a dynamic tension between social and discursive power, as well as between power as domination and power as enablement; and that this precarious balance can easily be abused, from both sides, by the more powerful as well as the less powerful participants.¹⁰

Notes

1. See, for example, Bruce (1981) or Leech and Short (1981: 257-72).

2. I follow Fairclough (1995: 14) here, who distinguishes between genre as ‘a socially ratified way of using language in connection with a particular type of social activity (e.g. interview, narrative, exposition)’ and discourse as ‘a way of signifying a particular domain of social practice from a particular perspective’.

3. This definition is based on Fowler (1985: 61).

4. For a similar argument, and an application to E. M. Forster's Howards End, see Buck and Austin (1995).

5. Readers interested in schema-theoretic approaches should refer to Barlett (1932), Rumelhart (1975, 1980), Tannen (1993a); and for applications of schema theory to literature, Cook (1990, 1994) and Semino (1995, 1997).

6. According to Buck and Austin (1995: 65), negative face refers to the desire to be able to act freely and to avoid having one's privacy violated, and positive face to one's aspiration to be respected and liked.

7. See Hubert-Leibler (1992: 75) on the nature of the mentor-protégé relationship, where ‘the exercise of power is mitigated by feelings of solicitude and love, and a real concern for the other's well-being’. The two nouns, ‘solicitude’ and ‘love’, bring out the dangerous potential ambiguity of this relationship, with John later claiming that he was motivated by pure solicitude, whereas Carol interprets his behaviour as a form of ‘rape’ (1992: 67).

8. For an argument associating literary texts with the ‘defamiliarizing’ effect and non-literary texts with the ‘hardening’ effect, see Cook (1994).

9. See Csicsila (1995: 6) and Piette (1995: 185). It is, I think, this seductive effect of the play that Showalter (1992: 17) has in mind when she accuses Mamet of playing a rigged game.

10. I should like to thank Clara Calvo, Marion Colas-Blaise and the editors of this volume for their most perceptive comments on an earlier version of this chapter. I have not always followed their advice, so all remaining errors are mine.

Bibliography

Bartlett, F. C. (1932) Remembering: A Study in Experimental and Social Psychology, Cambridge: Cambridge University Press.

Berstein, B. (1971) Class, Codes and Control 1, London: Routledge and Kegan Paul.

Brown, P. and Levinson, S. C. (1987) Politeness: Some Universals in Language Usage, Cambridge: Cambridge University Press.

Bruce, B. (1981) ‘A Social Interaction Model of Reading’, Discourse Processes 4: 273-311.

Buck, R. A. and Austin, T. R. (1995) ‘Dialogue and Power in E. M. Forster's Howards End’, in P. Verdonk and J. J. Weber (eds) Twentieth-Century Fiction: From Text to Context, London: Routledge, 63-77.

Cook, G. (1990) ‘Goals and Plans in Advertising and Literary Discourse’, Parlance 2: 48-71.

Cook, G. (1994) Discourse and Literature: The Interplay of Form and Mind, Oxford: Oxford University Press.

Csicsila, J. (1995) ‘Review of Oleanna’, The David Mamet Review 2: 5-6.

Durant, A. and Fabb, N. (1990) Literary Studies in Action, London: Routledge.

Erickson, B., Lind, A. E., Johnson, B. C. and O'Barr, W. M. (1978) ‘Speech Style and Impression Formation in a Court Setting: The Effects of “Powerful” and “Powerless” Speech’, Journal of Experimental Social Psychology 14: 266-79.

Fairclough, N. (1995) Critical Discourse Analysis, London: Longman.

Fowler, R. (1985) ‘Power’, in T. A. van Dijk (ed.) Handbook of Discourse Analysis 1, London: Academic Press, 61-82.

Hubert-Leibler, P. (1992) ‘Dominance and Anguish: The Teacher-Student Relationship in the Plays of David Mamet’, in L. Kane (ed.) David Mamet: A Casebook, New York: Garland, 69-85.

Kureishi, H. (1995) ‘My Son the Fanatic’, in A. S. Byatt and H. Hollinghurst (eds) New Writing 4, London: Vintage, in association with the British Council, 100-11.

Leech, G. N. and Short, M. H. (1981) Style in Fiction, London: Longman.

Mamet, D. (1993 [1992]) Oleanna, London: Methuen.

Piette, A. (1995) ‘The Devil's Advocate: David Mamet's Oleanna and Political Correctness’, in M. Maufort (ed.) Staging Difference: Cultural Pluralism in American Theatre and Drama, New York: Peter Lang, 173-87.

Rumelhart, D. E. (1975) ‘Notes on a Schema for Stories’, in D. G. Bobrow and A. M. Collins (eds) Representation and Understanding, New York: Academic Press, 211-36.

Rumelhart, D. E. (1980) ‘Schemata: The Building Blocks of Cognition’, in R. J. Spiro, B. C. Bruce and W. F. Brewer (eds) Theoretical Issues in Reading Comprehension, Hillsdale, N. J. Lawrence Erlbaum, 33-58.

Semino, E. (1977) Language and World Creation in Poems and Other Texts, London: Longman.

Semino, E. (1995) ‘Schema Theory and the Analysis of Text Worlds in Poetry’, Language and Literature 4: 79-108.

Showalter, E. (1992) ‘Acts of Violence: David Mamet and the Language of Men’, Times Literary Supplement 4675: 16-17.

Tannen, D. (1993a) Framing in Discourse, New York: Oxford University Press.

West, C. and Zimmerman, D. H. (1985) ‘Gender, Language and Discourse’, in T. A. van Dijk (ed.) Handbook of Discourse Analysis 4, London: Academic Press, 103-24.