The Agonistics of Reading: Playing, Gambling, Committing

From Michel Picard’s La lecture comme jeu to Umberto Eco’s model of the “game of chess”, reading has often been compared to a kind of game. Games serve as a useful template of interaction, highlighting both the exterior set of rules governing an activity, and the agency that these rules leave to the individual. Yet games are also a constitutive human activity, with myriads of variants, from the free play of children to the gambler’s thrill to the highly ritualized sets of actions and reactions seen in martial arts or chess. This article proposes to review classical and contemporary theories of reading based on their specific use of the metaphor of reading as a game. It first presents the structuralist and phenomenological approaches, which tend to define reading as a performance based on pre-established rules, like a game of chess. It then delves into theories that instead choose to highlight the incalculable aspect of every new reading, the possibility for the reader to go off the beaten path. These tend to see reading more as a game of chance than a game of chess. This does not mean, however, that they construe reading as licentious: gambles involve stakes. Theories like Jean-Jacques Lecercle’s Marxist pragmatics are most specific in explaining what the reader is actually committing to. The stance we take, our interaction with texts as “interpellations”, are part and parcel of our lives as social and political beings. They are the products of a certain context, but they may in turn influence or call into question the very structures that make them possible. This is why this article suggests reading be examined through the notion of agonistics. Taking up the ancient Greek word “agon”, which implies that games are forms of trial made to reveal something of the player’s nature, the agonistics of reading posits that reading must not be seen as an isolated phenomenon. On the contrary, the challenge that texts pose, to confront them or to accept them, is fundamental in the construction of our identity as readers and as human beings. This dialectics of self-revelation and self-construction, through the interaction with texts, is the often unspoken yet decisive game that every reader plays.


Introduction 1
In La lecture comme jeu (1986), Michel Picard remarks upon the increasingly ludic nature of texts, both in contemporary literature and in popular genres like thrillers. The involvement of the reader, figuring out clues, making indirect connections and treating the narrative voice with playful scepticism, is undoubtedly a widespread phenomenon, while genres like the "choose your own adventure books" deliberately blur the line between narrative and game. However, Picard adds that playfulness has always been part of the reading experience, although perhaps in less obvious ways: many forms that are now seen as "serious" literature were formerly considered as games (Picard 1986: 195). Rather than thinking of this hybridization as a process, invented by someone at a certain juncture to mesh the two distinct activities of reading and playing, it may be more fruitful to wonder if these in-between forms do not, in fact, tap into something inherent in all reading. Reader-response theories have long used the metaphor of games and playing -which are both inherent in Picard's notion of "jeu" -to define and describe the activity of reading. Louise-Michelle Rosenblatt presents an apt explanation as to why this metaphor is so prevalent: to understand reading, theoreticians must go beyond the traditional vision of the reader as a passive audience. "The reader's creation of a poem out of a text must be an active, self-ordering, self-correcting process" (Rosenblatt 1978: 11). The process of playing games provides a perfect template to articulate these three points: not only are games actively played, but the players "order" their own moves by conforming to some code of rules, and they make mistakes, from which they can learn in order to perform better. Nearly all reader-response theories rely on this basic parallel, highlighting the active, ordered and non-linear process that reading entails.
However, a game is much more than an interactive, iterative process based on rules. It is a human activity, with myriads of variations involving different processes, goals and faculties. From children pretending to inhabit a fantasy world with ever-changing rules, to a gambler betting it all on the roll of a dice, to martial artists or chess champions performing a highly ritualised set of attacks and counters, games can be widely different. One may wonder to what extent the complex and still ongoing debates between competing conceptions of reading can be understood by asking what kind of game the reader is actually playing. Going into the precise details of the metaphor of reading as a game leads us to question not only the process itself, its structures and the rules that govern it, but also what is at stake, the personal and social values attached to reading. Just as games are a mirror of those who invent and play them, visions of reading imply a certain stance towards interpersonal and social interactions, and a certain idea of their end-goals -that is to say, in the Aristotelian sense of the term, a vision of ethics. 4 This article presents an overview of several reader-response theories, focusing on how they metaphorically or concretely link the activity of reading to different forms of play. I will first present the structuralist and phenomenological approaches, which define reading as taking up a set of established rules and enacting a ritualised performance, with a view to reaching a pre-determined set of end-goals which all the participants can agree on. Just like a game of chess ends on indisputable victories, draws or defeats, and a dramatic or musical performance can be deemed good or bad depending on its faithfulness to the original score, reading according to these theories follows scripts and can be considered and judged accordingly. However, such conceptions are confronted with two major problems. The first is that some texts, especially in modernist and post-modernist literature, deliberately blur the rules that govern reading. By creating unsolvable ambiguities or giving contradictory information, they jeopardize the very ritual that is supposed to make them readable. The second is the fact that some readings can be artistically valuable and meaningful while eschewing or distorting the rules. While it is always possible to dismiss such a use of the text, by considering it as perverse or eccentric, other theories embrace this form of playfulness, and construe a definition of reading which makes room for it.

5
Theories of that kind tend to insist on the notion of "free play", in some form or other. They emphasise the rights of the reader and the inherent singularity of each new act of reading as a unique event. Be it from a psychoanalytical perspective, as in the works of Michel Picard, or from the deconstructionist standpoint of Jacques Derrida, reading is defined by its contextual and creative aspect. Like the play of children, it takes up preexisting elements, but never combines them twice in the same way, so that each new reading involves an irreducible fragment of chance. Like Stéphane Mallarmé's roll of the dice, each reading constitutes a new gamble, which never abolishes chance, never creates a fixed new protocol. Nevertheless, this inherent creativity must not be confused with pure arbitrariness, nor does the contextual nature of the reading event make it totally free. On the contrary, theories that highlight free play also claim that there is a specific form of seriousness to reading as a gamble, that its openness involves another set of values.
text. The rules that govern the reading of texts do not represent abstract ideals: they are embedded within their rhetorical structures and rely on broader social and cultural forces. This is why I think that the pragmatic model of reading proposed by Jean-Jacques Lecercle bridges the gap between proponents of reading as a structured game and those that stress the possibility of free play. It puts obedience (the realisation of the ritual) and subversion (the breaking of rules in order to effect another type of performance) on the same plane as possible forms of responses to an interpellation. A text does not offer one meaning to strive for, but a number of positions from which to interact with its contents, some of which its rhetorical apparatus favours and some of which it sidelines. Those postures even include closing the book or looking out the window, for instance -both actions which readers can take in response to stimuli from certain books. Most importantly, each reader's reaction also comes to bear some meaning. Even closing a book amounts to taking a stance, and this revelatory aspect of all possible responses may be seen as the basis for a form of dialectic evaluation, bearing both on the text and on readers themselves.This article posits that the notion of agôn is fundamental in understanding what game the reader is playing. The word agôn, in its original Greek context, was used mainly in the context of competitive sport. It implies that the confrontation with another is inherently revelatory. In the case of reading, our reaction to the otherness of the text, whatever it may be, cannot but have implications on our very identity. Our choice to take on certain roles, to obey or disobey the rhetorical guidance of the text, is directly linked to our way of responding to all forms of cultural and social interpellations. It is precisely because none of these messages have a fixed authority over their interpretation, or transcendent claims to our obedience, that choosing to obey or refuse them is so crucial. As Paul B. Armstrong puts it, "parity between the worlds of text and reader [...] would mean that the authority of the conventions governing both are at play and at risk" (Armstrong 11). In the end, the challenge that texts pose, to confront them or to accept them, to reduce them to predetermined protocols or open them up to incalculable gambles, is fundamental in the construction of our identity as readers and as human beings. This is the often unspoken yet decisive game that every reader plays.
Playing Chess and Playing Music: Reception theories and the traditional metaphors of reading 7 In Lector in Fabula, Umberto Eco compares reading to a game of chess (Eco 1986: 117). This simile underscores a certain vision of the interaction with texts, which stands at the core of his theory. Eco understands texts as frameworks, sets of rules and possibilities, which remain incomplete, and require a form of actualisation. Using the common knowledge of a certain culture, what Eco calls its "encyclopedia", the text sets up scenarios, from which the reader is supposed to expect a number of consequences. This enables the reader to enact complementary reading strategies, looking for hints and reconstructing the narrated events in order to fit certain "interpretive hypotheses", that is to say to provisionally give meaning to what he or she is reading. This interaction coalesces in two abstract figures: the Model Reader, as the hypothesis of a reader who would enact the perfect series of interpretive movements to attain a total understanding of the text, and the Model Author, the representation of all the interpretive virtualities of the text as intentional discursive strategies ("The Author The Agonistics of Reading: Playing, Gambling, Committing Angles, 11 | 2020 wants us to understand this" would be the symbolic equivalent of "the text's structure leads to such an interpretation") (Eco 1986: 71;81). Like in a game of chess, the web of narrative and stylistic devices within a text defines a set of possible moves, some more fruitful than others, and the "model" reader and author are abstract figures always playing the best possible move in a given situation to create the most appropriate interaction within the parameters. The other possibilities include imperfect moves, for instance a more superficial interpretation, but not breaking the rules. In a game of chess, swiping the pieces off with one's arm in anger or departing from the table are not moves, and do not need to be recorded as such. Only what constitutes "textual cooperation" in some form or another is the subject of Eco's research (Eco 1986: 236).

9
This structural representation of reading makes for effective and cogent interpretations, especially when dealing with texts that play on hints and require a certain form of reconstruction from readers, such as mysteries and thrillers. However, its vision of the "encyclopedia" as a fixed, structural container for all the meanings and connotations of a given word or phrase, raises questions. It seems to externalise something as intimate as the relation that each of us has with language.
10 Conversely, phenomenological approaches have tried to integrate the original perspective and relation to the world of the individual reader in their reflection. Instead of seeing the act of reading as a logically determined game, on the model of chess, they tend to rely instead on metaphors of theatrical or musical interpretation. In The Act of Reading (1976), Wolfgang Iser likens reading to taking up a role. This position keeps the reader in charge of how exactly the role is played, but nevertheless "commits him to a point of view", which is embedded within "a certain textual structure" (Iser 1976: 71). The "selective" interpretation is inscribed within a "horizon of meaning" (75), which is to say that the text acts upon its reader, giving hints as to how it should be construed. Borrowing from J. L. Austin's pragmatics, Iser calls this the "illocutory force" of fiction: it is meant to produce effects on us as readers. Through its rhetorical structure, it asks us to adhere to its perspective, and "construct the context for its reception" (114). As in Eco's vision of reading as a game of chess, Iser sees the text as delineating a reader-figure, the "implied reader", which actual readers are supposed to follow, or at least to integrate in a dialectics with their own preconceptions (Iser 1978: 293). The goal of interpretation is hence a form of harmonisation between the reading public and the "implied reader" which the text calls for. 11 The notion of a dialectics leading to harmony is a fundamental tenet of phenomenological approaches. It underlies what Hans Robert Jauss calls the "conciliation of horizons of expectations", which is not a structural effect of the text, but is defined by Jauss as a duty of the reader (Jauss 1988: 434). Phenomenological approaches tend to imply a form of value judgement, whereby the reader is morally encouraged to enact such a conciliation. This ties in with a broader philosophical movement, rising from a certain reading of Hegel and of the concept of "recognition", which posits that human understanding is a goal in itself. Jürgen Habermas is perhaps the clearest advocate of "communicational understanding" as the end-goal of rational human interactions, and the desire for "intersubjective recognition" as the basis of communications (Habermas 1987: 27;32).
12 This underlying assumption leads phenomenological reception theories to relegate tensions and conflict to a secondary role. Iser explains that the model for tensions within works of fiction is the beau désordre of Marivaux: it is inherently provisional, and The Agonistics of Reading: Playing, Gambling, Committing Angles, 11 | 2020 its own dialectical movement leads to its resolution. In concrete terms, this means that, when confronted with complexities and tensions, the reader is supposed to "participate in the solution" by finding a new interpretation, and reintegrating the anomaly within a broader frame (Iser 1976: 91). In Iser's theory, this breaking and remoulding of assumptions constitutes the hallmark of a literary style, as opposed to texts that propose only to reinforce existing stereotypes (Iser 1978: 284) However, it ensues that in certain cases, the reader must relinquish freedom in order to follow a certain path: "the communication between the text and the reader can only produce a felicitous outcome if it is under control" (Iser 1976: 297). Consequently, the reader is supposed to have agency only when the text leaves an opening in "places of indetermination"; even then, the potential for disjunction is limited (297). In the end, just like a chess player has a fixed number of moves to play and a performer only a limited amount of freedom within the parameters of a playtext or musical score, conceptions of reading that rely on these metaphors tend to highlight the "limits of interpretation", to paraphrase the title of one of Eco's essays. They "aim [...] at producing a possible reader whose profile is designed by and within the text" (Eco 1990: 52).
13 Although this line of thinking produces undeniably elegant analyses, delineating forms of "perfect reading" called for by the very structure of the text, it is of limited use when applied to more chaotic or confused situations. This of course includes complex texts. The works of James Joyce, in particular, have often been presented within readerresponse theories as a sort of margin within the literary corpus, which both fascinates and baffles theoreticians. As early as his seminal work Opera Aperta (1962), Eco saw Joyce's Finnegans Wake (1939) as a "terrifying document" on the possibilities of semantic ambiguity and formal instability, and the symbol of a temptation to go beyond all the accepted codes that make texts understandable (Eco 1962: 257;288). Eco's attempts to affirm the "internal logic" of a totalizing Joycean "encyclopedia" in The Limits of Interpretation (1990) have to come with a caveat on the irreducible "play" with these notions within the text. Likewise, Iser's musings on Joyce's Ulysses (1922), which lead him to deduce that the text abandons all possibilities of synthesis in order to create an "experience", reveal the inability of Iser's theoretical frame to explain the process of reading Joyce in terms other than negative (Iser 1989: 23-4). Rather than being fully explained by these lines of thinking, some works seem to remain at their limit, and call into question the very possibility of a comprehensive model of reception theories. These theories also tend to ignore that readers can, and often do, break the rules, cheat, or otherwise subvert the rituals that govern the "right" way of reading. Picard remarks that ideas such as that of pacts or duties of the reader exorcise the "tacit, obscure, uncontrolled" aspects of reading, which in reality is a "dangerous game" (Picard 1986: 165). In doing so, we may wonder if they do not also impose a certain standard set of values, which restores the basic structures of authority and intentionality by asking of the reader to accept a form of transcendent guidanceeven if they replace the traditional Author figure with the rhetorical structure of the text. As Rosenblatt argues, the notion of a predictive text often hides a willingness to uphold "public control" and "check the relevance of reading" according to institutional or social values (Rosenblatt 1978: 69). Emphasising the rights of the reader to more freedom, to more creative forms of play, may then be a way of calling into question such external standards, of making the act of reading more central and autonomous. Interestingly, a significant number of theories that emphasise the autonomy of reading have been written in French. Perhaps not coincidentally, the word "jeu" in French is open to a much broader range of interpretations than "game" or "play", enabling thinkers to highlight the inherent dialectics between the logical rules of games, the larger notion of playing, and even more specific meanings. For instance, Roland Barthes, in "Le plaisir du texte" proposed a conception of reading which would include the "logical contradictions" within the text and the unpredictability of reading as "jouissance", outside of the boundaries set by the rhetoric -in his own words: "que les jeux ne soient pas faits, qu'il y ait un jeu" (Barthes 1973: 9-11. My emphasis). In this sentence, he reuses the set phrase "les jeux sont faits", or "the bets are off" which the croupier or bookmaker shouts when new bets are no longer allowed, thereby introducing another fundamental term in the semantic field of games and reader theory: that of the gamble. The play of reading only intervenes if "bets are on", if there are still moves to make and chance involved. This defines another set of conceptions of reading, which oppose the predetermined parameters of structuralist or phenomenological reception-theories, in the same way that games of chance stand in contrast to games of chess.
15 This approach, which focuses on individual readings, their idiosyncrasies and their relation to chance, underlies Picard's notion of reading as a form of play. Picard stresses the similarities between the psychological relation towards games and towards art. Both call for a form of psychological "adaptive function" which involves forms of interpretation. The confrontation with artworks and with situations within games conjures up a half-way space between reality and fiction, with its own set of rules and possibilities. Taking up Freudian concepts, Picard identifies texts in particular as "transitional objects" between the mind's freedom and the resistance of the outside world (Picard 1986: 102). They enable the reader to experiment with interactions, dissociating them from a specific context to manipulate them, through a double mechanism that is both defensive and constructive (27-30). It is not a question of committing to a role, as in Iser's theory, but on the contrary of trying on many situations and postures for size. According to Picard, this Protean and creative path through the text is the very essence of the game the reader plays (Picard 1986: 93-4). 16 However, involving chance decisions and the freedom to hop from one position to another does not make the reading process haphazard. Because the reader is producing his own reading, "il joue gros jeu", as Picard puts it, that is to say, risks are involved. And Picard emphasises the two complementary psychological dangers that haunt every reader: getting caught up in the game, losing perspective to blindly follow a reductive reading path, or -to borrow a word coined by Donald Winnicott -"fantasying", that is following individual musings to the point of solipsism (Picard 1986: 120). Iser remarked that good fiction always oscillates between involving readers further and distancing them from the action (Iser 1978: 286). Picard sees this rhetorical balancing-act as a set of psychological safeguards which reveal that every interaction with literary texts walks the line between over-involvement and indifference. Far from being marginal phenomena, these unwarranted effects are part and parcel of the reading process and can be explained within Picard's theory as the direct consequences of the reader's freedom and risk-taking.
17 Thus, the conception of reading as free play replaces the opposition between objective "good" and "bad" reading with questions on what it means subjectively to choose a certain path: to adhere to a simplistic view of a complex text or to forget its otherness, which Picard likens to the "resistance" of reality, in order to let one's own ideas have free reign. All reactions are possible, all are within the framework of these theories, but they have different implications, and call upon what Hillis Miller calls an "ethics of reading", in the sense that each reading implicates the reader who "must take responsibility for it" (Miller 1987: 59). 18 The relation between play and responsibility in the context of reading has been particularly important in the deconstructionist movement, and has been theorised at length by Derrida. From the start of his philosophical career, Derrida sought to liberate what he called the pluri-dimensional "problem of reading" from unilateral visions of meaning (Derrida 1967a: 31). In De la grammatologie (1967), he claims that writing and reading existed before their subjection to a theory of meaning, which is a more recent, historical phenomenon (130). He also construes a new theory of "play" ("jeu") which precisely articulates the notions of free play and structured games. Taking up a very specific meaning of "jeu" in French as the 'looseness' within a mechanism (like a lock) which gives a small margin to work with, Derrida affirms that a form of play derives from the very fact that perfect, totalising structures, although they may serve as a regulatory ideal, have no actual existence. They are thought of as centres, but in fact constitute only "vanishing points" caught within the structure of "différence", in the differential between many objects and events, none of which can be the true standard for the others (Derrida 1967b: 423). This applies readily to theories of reading analysed earlier: it is only in how they differ from the ideal of a Model or Implied Reader that individual readings acquire their singular existence. Standard readings are not the most pertinent interpretation, but merely representations of all the protocols that are at play to regulate individual readings and make them go through sufficiently similar motions for them to be compared. In the end, these "standard readings" have nothing to say about what actually happens when we read, that is to say the uncontrolled mechanics of differentiation and dissemination by which my reading differs from yours, and even differs from my reading of the same text at a different moment. Michel Lisse, in his works on Derrida's conception of the reading process, calls this the "logics of ' maybe'". The question is not to reject any regulatory interpretative model based on etymological of philological facts, but, in Derrida's words, to oppose all "philological fundamentalism" which would erase the possibility of an eye-opening "invention" on the reader's part (Lisse 2001: 49). Speaking about Mallarmé, Derrida likens this to the notion of the "roll of the dice" in the poet's works: Le hasard ou le coup de dés qui "ouvrent" un tel texte ne contredisent pas la nécessité rigoureuse de son agencement formel. Le jeu est ici l'unité du hasard et de la règle, du programme et de son reste ou de son surplus. (Derrida 1972: 62) 19 Every new reading is a composite of probability and chance: it is probable that the reader will take a cue within the text, understand a reference, follow a beaten path, since that is where the text's structure leads. But at every turn, there is an opening, a "surplus". The play stems from the looseness of the mechanism, in its inability to actually be all-encompassing and at one with itself, which opens up the space for a broader form of play.
20 This conception also enables Derrida to assert the seriousness of what he calls "jeu". In "Plato's Pharmacy", he argues that it is only possible to relegate play to the notion of a fun diversion, of games, when there exists a "serious" underlying structure to which it can be compared. The "controlled and contained" reflection on "fun," alternate readings only works when the distinction with the serious, true meaning of a work is supposed to be founded on solid grounds. If the authoritative notion of the faithful "cooperative" reading is only an abstract ideal, then all takes on a text are serious on some level. Derrida theorises this possibility through the notion of "coup", playing on the large range of meanings that this word can take: nuances like that of striking (as in a political coup, but also a stroke of luck), of dramatic momentum (the coup de théâtre) and, of course, that of the gamble, since a roll of the dice is "un coup de dé" (Derrida 1972: 180;192). At each juncture, for each word of the text, there exists the inherent possibility for the reader to enact one of these "coups", be it a stroke of geniusfinding a hitherto hidden chain of connotations that hints at a character's hidden motive, or a subdued undercurrent that runs counter to the main narrative -or a personal, whimsical reversal of meaning. This is why every reading is an event, even if the text is the same -like any game of roulette or blackjack is a new, incalculable gamble, even if the parameters haven't changed since the last game which follows the same rules. Someone may always have a stroke of luck, and someone can always decide to hold up the casino. 21 One of the common points of these different meanings of "coup", however, is that they all involve some form of contextual value. A political coup, even a coup de théâtre, imply that there was a dramatic situation -real or fictional -, in which the action intervened. Like in a casino, there is always something at stake in play understood as a "coup". This is why Derrida rejects the idea that play equals "licence", and affirms the "severity" of his perspective. Just like Picard, he replaces objective notions of "good" and "bad" with subjective postures, "licence" and "severity". Once again, the set of values which he posits is based on the relation to an Other, to exteriority. The "text of the other", in its singularity, calls for a "responsible response", which puts the respondent's singularity "at stake" (in Dutoit and Romanski 2009: 199). This complex and compressed statement undoubtedly requires analysis, but it points towards the two questions that must be answered in order to gain a full understanding of the act of reading: how it relates to the otherness of the text, and what we concretely put at stake when we open a book, the exact nature and parameters of our commitment.
Understanding the Stakes: Interpellation and its counter-play 22 Theories of reading rely on the basic idea that the text in some way calls towards its readers, creating a tension between our ordinary way of reacting to people and events and a certain role that it assigns to us within its structure. Iser considers this call, and the fact that the author "sketches an image of himself and his reader" in certain positions, as the source of a tension which is consubstantial with the very notion of an "implicit reader" (Iser 1976: 73). But the concrete explanation for this tension, which implies a force acting upon the individual, differs considerably depending on the thinker. As Paul B. Armstrong puts it, calling upon the Sartrian reflections on literature, the possibility of mutual recognition between author and reader cannot completely hide the fact that "power is also at work in reading" (Armstrong 2005: 5). Furthermore, American theoreticians of reader response like Stanley Fish have stressed the social and cultural aspect of this relation of power. According to Fish, taking up a certain way of reading is a politically motivated action, which places the reader within a certain "interpretive community" (Fish 1980: 16). Rejecting the idea that texts can bear illocutory power -an idea which runs directly counter to Iser, as seen above -Fish argues that they in fact borrow it from their context, that certain readers, within certain communities "accept the underlying setting" which enables the textual rhetoric to function (Fish 1980: 230).
23 Thus, the concrete relation between a text's structure, its effect on the reader, and the social forces at play in the cultural world needs to be understood in order to define how these three instances -text, reader and social world -interact. In this task, the models developed by Marxist pragmatics, and particularly in the works of Jean-Jacques Lecercle on what he calls the ALTER structure, provide a number of cogent responses. Lecercle understands the "call" of the text through the Althusserian notion of "interpellation" as a process which uses historically determined institutional relations of power to make its readers -as well as authors -adopt socially accepted patterns of response and behaviour (Lecercle 1999: 75). Instead of imagining that the written word has its own power to make readers follow certain paths and understand certain references, Lecercle argues that texts (the T in ALTER) borrow both their linguistic structures (the L in ALTER) and web of connotations (the "encyclopedia" or E) from the external state of a certain language and a certain culture, and that these in turn determine socially coded positions ("actantial places") for the author and reader (A and R). In other words, the rhetoric of a text relies on the fact that readers are human beings in a certain society, and that they will respond to certain recognisable forms of discourses as they do in the rest of their lives, respecting the literary author like they respect other forms of authority -or, in certain social contexts, respecting them as little, and acting overtly against certain frames and clichés.
24 This is how Lecercle overcomes the limitations that he sees in Iser's "aesthetics of negativity", as wells as other theories which acknowledge the fact that rules can be broken but cannot fully grasp the "duality of activity and constraint" that defines the real agency of the reader (Lecercle 1999: 90-1). Lecercle argues that reading is entirely "le jeu d'une structure", that is a structure both "at work" and "at play". On the one hand, its constraints, like all social structures, live on and are reinforced through "sedimentation": if many people read trying to emulate the structures of Model Readers, or to follow the will that they attribute to certain Author-figures, and if reading is taught that way, it becomes an increasingly strong norm. But structures are also historically determined, and every new act of reading, like the Derridean "coup", can "break the crust" and be part of a broader "insurrection" within culture (Lecercle 1999: 166). It then has to face the defensive reactions of cultural institutions. University professors and cultural journalists are bound to rail against "eccentric" or "subversive" readings, which any Marxist will take as proof that there indeed was a spark in the struggle for the control of the cultural field. This explains how free forms of reading are always possible but how, in practice, standard readings are statistically normal: The Agonistics of Reading: Playing, Gambling, Committing Angles, 11 | 2020 opposing the rules amounts to taking a risky social stance, struggling with powerful, long-established institutions.
25 However, Lecercle asserts that resistance to interpellations -what he calls "imposture" -may use the very framework of the ALTER structure, albeit in a different way. Real subversion does not consist in explicitly opposing all social forces, since there is no standpoint, no language or tradition, that is completely outside these forces. They must be challenged from the inside. This is the meaning of "counter-interpellation": if every text borrows from the structures of a cultural field, it means that these structures are at play within it, liable to be subverted by a new composition or a new interpretation. Giving new meanings to words, changing the connotation and implications of certain phrases (which Lecercle likens to the Butlerian notion of "reworking", Lecercle 1999: 164-5), and even giving a completely revisionist interpretation of certain narratives, can have effects on the entire cultural framework of a society. Such a perspective also ties in with Michel de Certeau's reflections on culture and reading. Certeau envisions reading as an "operative process" of "reappropriation" of cultural codes which constantly updates old norms, and stops them from unilaterally exerting oppressive power (Certeau 1980: 11). In this everrenewed struggle, which he calls the "fundamental polemology" of language, he distinguishes between effects of strategy, that is institutionally imposed norms -how school programs and university curriculums teach people how to read -and effects of tactics, or individual resistances and "coups" ("playing on the occasion") that readers undertake to express their own, personal take on a text (Certeau 1980: 21).
The Challenge and the Revelation: Reading as agôn 27 Hence, from the perspective of Marxist pragmatics, the interaction between reader and text bears not only on their abstract status in a hypothetical Republic of readers and writers divorced from the world of everyday interactions. Who we are when we shop or eat, work for a salary or vote to elect our political leaders, and who we are when we interpret texts, are all interrelated. Hence, the forces at work to define the balance of political power or the redistribution of riches are directly linked with those that preside over our interpretations and our imagination. That is the very definition of "ideology" as "the imaginary relation to the real conditions of living" (Althusser 1976: The Agonistics of Reading: Playing, Gambling, Committing Angles, 11 | 2020 101). Culture, far from being immaterial, informs and changes our relation to the most concrete aspects of our everyday life.
28 With this in mind, it becomes easier to understand why most phenomenological approaches, and even some advocates of free play, fail to grasp the full stakes of reading. For as we have seen, notions like that of the illocutory force of literature, its effect upon the reader, and even that of commitment, are generally in line with phenomenological theories. Jauss, for instance, shows how texts can "interrogate" their interpreter and make readers call into question their own preconceptions, going so far as to use the idea of "interpellation". He also defends the reader's right to question the text in return, to "counter-question" (Jauss 1889: 57-9;94). However, he asserts that literary hermeneutics must function in such a way that "the quarrel of interpretations does not end with a struggle to the death, like in politics". He argues for a respectful and pacific vision of interpretation, where only a better mutual understanding can justify new questions (439)(440).
29 First of all, this stems from what I would argue is a flawed reading of Hegel. In the Phenomenology of Spirit (1807), the struggle to the death is a pre-political phase which institutes the power relations between humans through the "master-slave dialectics" (Hegel 1971: 158-61). This does not in any way mean that later phases of dialectics in Hegelian thought do not include deadly conflict, which is a fundamental drive in his vision of human affairs, the "work of the negative". Unlike later phenomenological thought, and the communicational theories of Habermas, Hegel's phenomenology gives conflict and struggles a decisive role to play in the movement of History. Most importantly, the dichotomy that Jauss proposes between literature and politics deliberately ignores the fact that what texts reveal about their readers, and what readers reveal about the texts that they encounter, have very much to do with the struggles of politics. As Lecercle argues in De l'interpellation (2019), there is no clear-cut division between the "communicational" and "strategic" value of interpellation, no way to distinguish between its rational appeal and underlying rhetorical effects. The two are intertwined, and separating them is in itself a politically consequential act which underlies the bias of a certain phenomenological standpoint (Lecercle 2019: 12).
competitors. The confrontation with others, and with the limits of one's body and mind, was a form of trial which brought to light one's true mettle. Similarly, in the world of ideas, the confrontation with texts and with the readings and interpretations of others, is a way to put one's own preconceptions and the limits of one's faculties of imagination on trial. As Gilles Deleuze and Félix Guattari put it, the agôn of ideas ruled the Greek "society of friends" within the city and, by extension, within the philosophical community: it was the standard by which each thinker's "pretensions" to truth or to faithfulness to an idea could be judged (Deleuze & Guattari 1991: 11). Instead of trying to judge reading through some external, neutral standard, a theory of agonistics thus argues that the act of reading itself can set and change values. It is a source, rather than an object, of ethical judgement. When confronted with a text, readers find themselves at the intersection of different forces. Firstly, the originality of the text itself, with its contextually determined set of rhetorical devices and discursive strategies. Secondly, the multiple cultural and institutional prescriptions of a certain social context, which depend on a myriad of factors -when and where the reader lives, how educated they are, what Bourdieu would call their "cultural capital." And, finally, the idiosyncrasies of their own personal lives, their intimate relation to books, stories, and to language. All these forces have pretensions to inform the construction of meaning, and none of them has any transcendent right to its claim. By deciding to read one way or another, to negotiate and arbitrate between all of these forces, the reader intervenes in the power-dynamics in a way which is perhaps not 'free' in the sense of 'free will', but always constitutes a form of "coup", a combination that cannot be determined in advance. This choice in turn reveals a lot, both about the forces at work, which may be challenged in the process and appear under a new light, and about the reader himself or herself. For if there is no "right" way of reading, every decision is meaningful, and has something to say about the one who made it. What is more, if the forces at work in the decision are inextricably intimate, cultural and social, then the posture of the reader that is revealed through their interpretation bears on all of these aspects of their identity. In the end, there is no good or bad reading, just a myriad of motivated interpretations, and choosing one over the other says a great deal about who we are.
32 This overview calls for two complementary explanations regarding conflict and cooperation to adequately articulate it with traditional reception-theories. Firstly, agonistics does not imply that conflict is the fundamental model for the reading process. It simply posits that it is not inherently marginal: cooperation and conflict are two kinds of response to the challenge of a text and a context of reading, and both should be studied on the same plane. Secondly, this does not make cooperation any less ethical. I would argue that, on the contrary, there is something particularly abstract and unconvincing about the way in which some theories of reading construe cooperation as a sort of transcendent, categorical imperative. Roger Sell, for instance, claims that "A reader responding to a literary text is humanly obliged to give its writer a hearing", while "misinterpret[ing] the writer by imposing their own world view" is a sort of amoral "temptation" (Sell 2011: 6). Such moralizing discourse, conjuring up a vague notion of "human obligation", does not do justice to the complexity and the dilemmas that readers face when they actually try to do justice to the otherness of the text. It is precisely because they have no obligation to cooperate that choosing that path is so interesting and valuable.  (2012), the fact that texts and reading protocols do not have any inherent power and are dependent on how readers actually respond to them entails more, not less, responsibility. This requires a specific understanding of the field of ethics, and Attridge shows how Derrida, particularly in his dialogue with Emmanuel Levinas, came to assert his own vision of that particular notion. Most importantly, Derrida differs from Levinas because he directly links the violence of confrontation to ethical evaluation, "in the pure and immediate ethics of the face to face" (Attridge 2012: 110), without resorting to a third, overarching term. As Hillis Miller puts it in his Ethics of Reading, "each reading is strictly speaking ethical in the sense that it has to take place, by an implacable necessity, as the response to a categorical demand, and in the sense that the reader must take responsibility for it." (Miller 1976: 59). What is ethical is the fact that there is a confrontation, forces and claims, which the reader cannot simply withdraw from by hiding behind a concept or a protocol: we can decide to read how we want, but we can't say that we didn't decide, or that the decision has nothing to do with who we are. Whatever choice we make, we must accept that it expresses something about us.
34 To respond to the primary question, that is to say what game the reader is playing, I believe we can draw from Derrida's notion of signature and counter-signature which he defines as "a duel of writing and reading" (Dutoit and Romanski 2009: 287). The question of what rules this duel follows, whether it takes the form of a chess match, a choreography, or a gamble, are secondary when compared to its stakes. In the choice of the protocol, in the oscillations between following and betraying the text, we must commit and open ourselves to be pierced and revealed, just like we may pierce and reveal some meaning that no other reading had brought to light. "Promettre de se compromettre". These are Derrida's exact words: to promise to commit, to promise to compromise and question your identity coming in, and to promise to respond to the challenge that the text confronts you with, whichever way you choose to do so.

Conclusion 35
In the end, the agonistics of reading adheres to the metaphor of reading as a form of play or game, but rejects traditional connotations of these words. It asserts that play is in fact a very serious activity, which we use to reveal, challenge and inform our identity as individuals, both psychologically and socially. That is the meaning of agôn as I have used it here. 36 In consequence, theoreticians of reading must continue to elaborate models of socially normative readings, with a contextual and historical framework in mind, seeing them not as overarching structures set in stone, but as institutions linked to broader political and social trends. Furthermore, they must endeavour to evaluate individual interpretations, whether it be those confirmed by documentary evidence at a certain time and place, or their own readings, as so many "coups", meaningful and revelatory responses to the call of the text. They must analyse the way in which such readings reinforce or call into question certain standards -academic standards of a "good" reading, social standards of "common sense", political standards of "biased" or "ideological" interpretations, and so on -while always trying to bring to the fore the relative singularity of each interpretation, and the way in which even faithfulness requires adapting the meaning to a new, original context.