The fortunes of critique have lately been bound up with those of context. Critique is reductive, or so the post-critiquers say, because it reduces the complexity of a cultural product to a symptom of its social context or to a function of the determining forces of history; it reduces the “thickness and richness of our aesthetic attachments” by referring them to an explanatory framework.1 To critique is to contextualize, the argument goes, and to contextualize is to impoverish.

What do we mean, though, these days when we talk about context? Is context a surface formation or a depth discovery? Is it essential to the meaning of a work or merely incidental to it? Given the renewed importance of concepts of context to recent debates about how we read or ought to be reading, it seems to me that it might be worthwhile to return to a strong theoretical position on context, a deconstructive one, to see what it can contribute to our present ways of thinking. Simultaneously, we can observe what happens to familiar deconstructive arguments when our present critical contexts are brought to bear on them.

In what follows, I will take up Derrida’s “Signature Event Context” with both of these motivations in mind. The text hardly needs an introduction here; it’s known for being Derrida’s most extended engagement with analytic philosophy, in particular with J.L. Austin’s theorization of performative language, and it gained a certain notoriety because of the debate with John Searle that ensued, in part reprinted in Limited, Inc and which I will largely be bracketing here. Rather, I’ll focus on the earlier text in order to consider what Derrida understood there by context and by the act of contextualizing.

“Signature Event Context” (SEC, Signature événement contexte) puts questions of context front and center: “this particular communication,” Derrida writes, “will be concerned with the problem of context and with the question of determining exactly how writing relates to context in general.”2 SEC also works through a notion of writing that is developed at greater length in Of Grammatology and other pieces of Derrida’s from this period, systematically showing how a generalized conception of writing, as Derrida articulates it, disturbs, transforms, and inhabits from within the three terms in the essay’s title. “Writing,” over the course of the essay, is shown to transform what we ordinarily understand by signatures, events, and—most importantly for the purpose of my argument—contexts.

The essay was first given as a talk in 1971 at a French-language conference on the theme of “Communication.” Derrida begins the essay by recursively gesturing to the context of his talk (or in French, communication, the standard term for a talk or conference paper), asking about the meaning of the conference’s keyword: is it certain that the meaning of the word “communication” is something “unique, “univocal,” “controllable,” or “transmittable”; is it “determinate,” “identifiable,” or “describable?” He draws our attention to at least three possible meanings: “communication” can refer to a transmission of meaning, in the sense that I communicate my thoughts by means of this essay; it can refer to a physical proximity, in the sense of communicating rooms; or it can describe a transfer of force, in the sense that one ball communicates its motion to another in a game of billiards. In pointing this out, Derrida signals that he will bring to bear on his discussion not only considerations of meaning, but also considerations of force. What we usually assume can limit the equivocity of a word’s meaning, he goes on, is the context in which it is used (SEC, 2/368). Context is thus introduced in connection to a notion of limit and to the task of limiting. Context renders the meaning of a word or utterance controllable and identifiable; it determines, in the sense of its etymology from the Classical Latin determinare, to bound or to set limits.

Context determines, or rather, as we will see, context raises the problem of determinability. “But are the requirements [les réquisits] of a context ever absolutely determinable?” Derrida asks. “This is, fundamentally, the most general question that I shall endeavor to elaborate. Is there a rigorous and scientific concept of context?” (SEC, 2–3/369). As he proceeds, it becomes clearer that there are at least two sorts of determination at stake. Derrida will be concerned with whether context is determining the meaning of a given utterance, and he will be concerned with whether context itself is determinable.

He unfolds both questions as part of his broader consideration of writing. Working through a reading of Condillac, Derrida undertakes the task in the first part of his essay of defining writing, or rather, as he says, of determining it: “What are in effect the essential predicates in a minimal determination of the classical concept of writing?” (SEC, 9/377). Derrida uses the term “determination” in a traditional way here, equating it with predication, the kind of predication that ascribes properties to a subject in order to define it. And he draws on the framework of the genus-species relation when he notes that Condillac treats writing as a “species” of communication, a species being that which “admit[s] a relative specificity within a genre,” a type belonging to a general kind and drawing part of its identity from its difference from other types (SEC, 6/374).

Seeking a determination of writing, then, Derrida looks to identify its specific difference from other types of communication. This task becomes a strange one, however, when Derrida names “absence” as writing’s “essential predicate.” How can absence serve as a predicate of any sort? Moreover, how is this absence itself determined? When Derrida first refers to the idea of “absence” as a predicate, he notes that it is yet “to be determined.” So he writes: “let us attempt… to characterize the absence”; “how do we qualify this absence?” (SEC, 7/374).3

The move Derrida makes, extending Condillac’s argument and causing some confusion for Searle and others, is from an empirical possibility to a structural necessity. Whereas Condillac describes writing as a form of communication that can reach “persons who are absent,” Derrida argues that such an absence is not merely possible or incidental but pertains to the nature of writing as such: the possibility of absence constitutes the structure of writing in its specific difference from other forms of communication (SEC, 7/374). What is more, we see that Derrida extends Condillac’s statement about “persons who are absent” to at least three different determinations, or to three different figures. While Condillac refers only to absent receivers, Derrida adds to this the notions of an absent sender and an absent context. He reasons by analogy: “what holds for the receiver holds also … for the sender or the producer” (SEC, 8/376). Writing is a mark that can subsist in the absence of a sender, a receiver, or a context; writing is what subsists as a mark of difference from this absence.

Derrida enumerates three characteristics of writing over and against what is absent (he literally enumerates them, in one of many sections in this essay where he feels the need to number his points): writing is characterized by a “subsistence” [une marque qui reste], by a “spacing,” and by a “force of rupture” (SEC, 9/376). For reasons that will become clearer later, I think that these three notions—subsistence, spacing, and rupture—are best grasped as describing differential relations, though Derrida identifies them with “absence” as such. Regardless, in his definition of subsistence, spacing, and rupture, a multi-faceted image of context appears. “Spacing,” for example, entails the separation of a written sign “from other elements of the internal contextual chain.” By “context,” here, we are given to understand that Derrida at least sometimes has in mind those marks accompanying the mark in question and which together are internal to a signifying chain, text, or work. This is a notion of context as internal to a text. Spacing also separates the mark, Derrida further remarks, “from all forms of … reference,” so that we can also file “all forms of reference” under the heading of what counts as context. Finally, when Derrida writes that the written sign implicates a “force that breaks with its context,” he goes on to gloss context quite broadly as “an allegedly real context [that] includes a certain ‘present’ of the inscription, the presence of the writer to what he has written, the entire environment and the horizon of his experience, and above all the intention, the wanting-to-say-what-he-means [le vouloir dire], which animates his inscription at a given moment” (SEC, 9/377). In this passage, we bear witness to a capacious notion of context, one that absorbs what Derrida earlier referred to as sender as well as receiver. What is called context now allegedly includes: physical presence, temporal presence, physical environment, experience, consciousness, intention, will, and the real.

What would it mean to subtract context from writing, in each and every sense of “context” that Derrida proposes here? And would this be a remotely helpful exercise for thinking about the relation between text and context in literary studies? It’s hard to see how it could be. Even the strongest statements I can think of urging critics to turn away from context—take Rita Felski’s essay “Context Stinks!,” the last chapter of her Limits of Critique, or Joseph North’s arguments against the historicist/contextualist paradigm in Literary Criticism: A Concise Political History—even these apparently hardline positions in no way advocate for a total absenting of every form of contextual consideration. Rather, each seeks to limit appeals to certain contexts in order to boost the visibility of others. North, for example, seeks to challenge the priority placed on historical scholarship in order to shed light on present contexts of aesthetic education. For him, historical scholarship has become overly specialized and no longer realizes its vocation “to intervene in the ‘culture as a whole.’”4 Felski, for her part, wants to tamp down what she calls “fealty to the clarifying power of historical context” because she sees such fealty as engaging in a sort of one-upmanship of insight.5 Against the ascendency of historicist claims, she argues that we would do better to stay open to “the messy, mundane realities of how and why we read,” which includes the contexts of texts’ reception as well as their potential “future affinities.”6 For both Felski and North, in short, context as such is not the problem. Rather each seeks to shift from one version of context to another, more ample one.

Returning to Derrida, and to the role of absence in Derrida’s characterization of context, let us consider Derrida’s reference to the possibility that a producer or receiver of a written text may be absent. The first thing I would observe is that the absence to which Derrida repeatedly refers has the contours of a determinate absence. When Derrida argues that writing subsists “in the absence of such and such a person, and hence ultimately of every empirically determined ‘subject,’” or when he argues that writing remains readable “in the absence… of the empirically determined subject who… has emitted or produced it,” he describes the absent receiver and sender as empirically determined or determinable figures (SEC, 7/375, 9/377). In each passage, then, the absence Derrida cites is obtained through the negation of an empirical figure. True, he speaks of an absence of empirically determined figures “in general,” but this generality, though it is not itself empirical, is conceived on the basis of the empirically determinate; “in general” here means in every case. Derrida also refers to an “absolute” absence, but what is “absolute” is the independence of each determinate absence from any recourse to presence, whether distant or deferred. Indeed, Derrida’s focus in these passages is on limiting the claims of presence, not on absence per se.7

Now for Derrida’s argument about absence to be convincing, it matters how Derrida figures these particular receivers, senders, or producers, and I think the debates over meaning and intention that have taken place in the wake of this essay—with Searle’s disagreement being only one of them, and Steven Knapp and Walter Benn Michaels’s 1982  essay “Against Theory” being another, one which reiterates Searle’s position to a certain extent and which opens a line of argument that has found a home in the present journal, among other places—debates over intention very much rest on how intention is figured in each case. For Knapp and Michaels, who draw on analytic philosophy of language, the absence of intention is figured as the absence of intentionality as such, which the authors equate with an absence of meaning. Hence, they paint the picture of a Wordsworth poem absurdly washing up on a beach, as if it were a product of nature; if we subtract any possible structure of human intending, the words become mere squiggles in the sand. For Derrida, however, who draws on continental phenomenology, the absence of intention is repeatedly figured as the absence of a particular determinate intention or set of intentions. Derrida further characterizes intention in terms of consciousness, will, or desire, as for example when he refers to “what [the] alleged author-scriptor consciously intended to say at the moment he wrote it” or to “my intention of saying something meaningful … my wish to communicate” (mon vouloir-dire, mon intention-de-signification, mon vouloir-communiquer-ceci) (SEC, 9/377, 8/376). Figured in this way, as a conscious desire bound to the moment of inscription, intention can be subtracted without a total collapse of the structure of intentionality as such.8 Indeed, as Derrida points out subsequently in “Limited Inc a b c…,”  “[A]t no time does SEC [“Signature Event Context”] invoke the absence, pure and simple, of intentionality.”9 Derrida never argues for the latter but envisions instead a “differential typology” in which “the category of intention will not disappear; it will have its place, but from that place it will no longer be able to govern the entire scene and system of utterance” (SEC, 18/389).

I think literary scholars employ such “differential typologies” of intention all the time, where the delimited subtraction of one set of intentions makes possible an investigation into others. Excluding “what an author had in mind” as a standard of interpretation, for example, as Wimsatt and Beardsley argued, need not entail—and almost never does entail—a rejection of intentionality as such. It yields a shift in what we look for when we read or research, a shift away from forms of agency figured in terms of individuality, subjectivity, or consciousness, toward other agential forms; toward intentions that can, for example, be multiple or divided against themselves; towards forces of cultural production or institutional norms understood as intentional actors in a broader sense; towards forms of agency belonging to the unconscious or to a social milieu. To emphasize any one of these forms is to construct its difference from another. In none of these examples is intentionality dismissed tout court. In this sense, what Derrida describes as an “absence” of intention would be another name for the finitude of any given intention, and in turn the finitude of its explanatory power. What it does not describe is the exclusion of the total field of every possible intention. Poems are not natural objects; they are intended—I agree with Knapp and Michaels here—but every alleged intention or set of intentions is finite. The absence “in general” that Derrida describes would be another name for the generalization of this structure of finitude that inhabits writing.

As we return to Derrida’s remarks on context, it’s helpful to consider them in light of this reading of intention. While it’s true that Derrida seems to reject context in fairly strong terms, writing, for example, that “the sign possesses the characteristic of being readable even if the moment of its production is irrevocably lost …. No context can entirely enclose it” (SEC, 9/377)—we should avoid the patent absurdity of taking this to mean that Derrida conceives of a writing that would break with every possible context at every moment. Joshua Kates argues that what first appears in Derrida’s argument as the absence of context is later affirmed as “an irreducible multiplicity of contexts,” and I think this phrasing of an “irreducible multiplicity” is particularly helpful.10 As Derrida writes, “The possibility of disengagement … which belongs to the structure of every mark, spoken or written, does not imply that the mark is valid outside of a context, but on the contrary that there are only contexts without any center or absolute anchoring” (SEC, 12/381).11 If we consider context along the same lines as intention, we arrive at the idea that something like the finitude of any given context is what gives rise to an “irreducible multiplicity” of contexts in the plural, and in turn to what Derrida calls writing’s iterability. That is, the separability of writing from a particular determinate context is what makes possible writing’s iteration in another, just as the separability of writing from an intentional sender or receiver makes possible writing’s re-attachment to other forms of intention, to other moments of production or reception. Separability and iterability go hand in hand.

What does it mean, more precisely, to talk about the finitude of context? Derrida does not use this expression; and in fact, it may sound like he is describing something like the opposite when he says that the goal of his essay is to “demonstrate why a context is never absolutely determinable” (SEC, 3/369). This part of his argument develops through his engagement with Austin’s How to Do Things with Words, where Derrida seeks to show how the predicates he has just ascribed to writing are generalizable not only to speech but also to that kind of speech that is closest to action, to what Austin at least initially calls performative language.

Examining how it is that language can successfully do something in the act of its being spoken—Austin’s examples famously include wedding vows, which accomplish the action of marrying in their being uttered—Austin emphasizes the role played by convention. Speech acts succeed where conventions exist, he argues, and where they can be appropriately mapped onto the people and circumstances involved. Conventions and circumstances, rather than the intentions of the speaker, turn out to be the determining factors in whether or not a speech act is achieved.12 In this way, explaining the force of a speech act would seem to entail, as Jonathan Culler puts it, “set[ting] forth the conventions that make it possible.”13

But where exactly do conditions, conventions, or context begin and end? How do we determine which contextual factors count as the determining ones? Logically, we cannot simply defer to convention to answer the question of what counts as convention, unless we want to end up with an infinite regress, an infinity of nested boxes of conventions. In his reading of Austin, Derrida works through Austin’s exclusion of so-called non-serious performatives such as “joking,” “acting a part,” or “writing poetry,” ostensibly with the aim of showing how assumptions about intention actually underwrite Austin’s argument, despite Austin’s claims to the contrary.14 For distinctions between serious and non-serious speech would seem to imply some recourse to speakers’ intentions. By extension, I would argue, the same must be true for distinctions between relevant and non-relevant elements of context: what counts as relevant would have to be decided with recourse to a given speaker’s intentions, or at least to the intentions that listeners impute to her. This is another way of saying that a speaker must be able understand and make use of context, convention, and circumstance in order for her speech act to succeed, and her audience must be able do the same in order to understand it. As Knapp and Michaels put it in “Against Theory 2,” where they argue that intention, and not convention, determines the meaning of an act, “following conventions is only one way of doing what is essential, namely, giving clues to your intention.”15 Derrida does not elaborate his point in this way, but he seems to see something of this logic at work in Austin’s text when he claims that Austin surreptitiously identifies context with the intentions of the speaker, with “a free consciousness present to the totality of the operation” of the speech act (SEC, 15/384). Though Austin claims to treat communication as a species of force and action, Derrida shows him to be drawing on a traditional understanding of communication as a transmission of consciously intended meaning (SEC, 14/383). In this framework, context is determinable because intentions determine it.

I will come back to this point about the determining role of intention in the second part of my essay. For now, let me simply point out the pivotal place of intention in Derrida’s argument. Despite readings of Derrida that would claim the contrary, intentions persist in Derrida’s account of speech and writing; but to grasp the particular nature of these intentions, we have to grasp the fact that they are inhabited by what Derrida calls “writing.” Derrida uses this term in both an empirical and a transcendental sense. In its transcendental sense, “writing” names the structure of language as such, a structure characterized by self-difference. The iterability of any empirical utterance or speech act—its constitutive repeatability, the fact that it is subject to citation and redeployment—divides any given iteration from itself, to the extent that this very self-division defines the condition of possibility of language. “Writing” is another name for this condition, and it accounts not only for empirical speech and writing but also, we note, for the intentions that permeate them. Indeed, Derrida goes so far as to assert that the structure characterizing writing is valid “not only for all orders of ‘signs’ and for all languages in general but moreover … for the entire field of what philosophy would call experience” (SEC, 9/377). Because this structure entails iterability, it introduces a fissure into any given experience, intention, or iteration, as part of the meaning of the iteration lies in past or future occurrences. Thus, in the case of intention, Derrida writes, “the intention animating the utterance will never be through and through present to itself and to its content. The iteration structuring it a priori introduces into it a dehiscence and a cleft [brisure] which are essential” (SEC, 18/389). In turn, we reason, the same is true for context. Because it is identified with intention as its “determining center,” it must likewise be divided from within by its own iterability. “This essential absence of intention to the actuality of the utterance,” Derrida writes, “this structural unconsciousness, if you like, prohibits any saturation of the context” (SEC, 18/389).

In Derrida’s account, iterability yields the non-coincidence of context with itself, an internal lack or difference that makes it unavailable to a totalizing determination. For Derrida, context is indeterminable, then, not because it is too large empirically speaking, but because it is too problematic epistemologically speaking. It lacks a determining center in the form of an intentional consciousness that could determine it neatly and entirely. Derrida draws out this distinction between empirical excess and structural indetermination in a later essay, “Living On,” when he asserts the following about context: “This is my starting point: no meaning can be determined out of context, but no context permits saturation. What I am referring to here is not richness of substance, semantic fertility, but rather structure: the structure of the remnant or of iteration.”16 What is worth highlighting here is the particular reason that context is indeterminable—and indeterminable not in practice, but in principle. The reason lies not with a given context’s being too rich or consisting in too many empirical details to be captured adequately in a description. It lies with the structural indeterminacy of context and with the fact that context, like intention, is the kind of thing to which iterability applies.

Derrida’s comments in “Living On” recall his reading of Levi-Strauss in “Structure, Sign, and Play in the Discourse of the Human Sciences,” where he describes two ways of conceiving the limits of totalization in the project of a structuralist anthropology. In one, Derrida points out, totalization is impossible for empirical reasons because “there is too much, more than one can say.” In the other, totalization is impossible for reasons of language, “not because the infiniteness of a field cannot be covered by a finite glance or a finite discourse, but because the nature of the field—that is, language and a finite language—excludes totalization … instead of being too large, there is something missing from it: a center which arrests and grounds the play of substitutions.”17 In our case, again, this center would be an intentional consciousness that could organize elements of a context into a meaningful totality, one that could be brought to bear on a literary work without excess, contradiction, or remainder.

Is all this to say that we should give up on context because we can have no purchase on it? I think it is rather a way of pointing to the constitutively incomplete nature of what is still the necessary task of making sense of texts in context. For every context that fails to determine its text completely because it, itself, is structurally indeterminate, new contexts come to impose their force and their relevance. The flip side of Derrida’s claim that every text ruptures with its context is his assertion that texts are citable and reinscribable in other contexts, for belonging to the structure of every mark is both “the possibility of disengagement and citational graft” (SEC, 12/381).18 Here, the limitations of a given context would not refer to the meaninglessness or uselessness of the concept of context, just as the finitude of intention does not refer to the meaningless or uselessness of the concept of intentionality. It would refer to the limited and provisional nature of the intentional act that determines any given context, for a particular purpose, in support of particular interpretive claims.

What do literary scholars mean by “context” when we talk about context? I think we usually do not mean an indifferent heap of empirical facts. We usually do mean a meaningful assemblage of circumstances and events, norms and conventions, that we believe are relevant to the interpretation of a text at hand or of some other object of knowledge. In this sense, contextual claims are a form of evidentiary procedure. And if in using the expression “we mean,” I bring consciousness or intention back into the discussion, I do so because I want to underscore the role that intentions play in the making of what we call context, and of what we expect context to do. Contexts are not given; they are not natural kinds; they are made, and they are objects of our intending. And if we follow a Derridean line of thinking, they are made of something called writing.

***

Writing is the determination of context: this would seem to be the insight of deconstruction. In the preceding section, I arrived at two entwined claims, the second a consequence of the first. First, I noted the way Derrida associates context with intention in his reading of Austin. As Derrida sees it, only when context is fully present to an intentional consciousness is context in turn able to determine the meaning of a speech act, according to the logic of Austin’s account. Second, I noted the limitations of context. Derrida shows context to be indeterminable, or at least as having a determination that is “never entirely certain or saturated,” because, like intention, context is divided from itself, riven because it is written (SEC, 3/369). Intention determines context, then, but intention also fails to determine context completely or certainly because both are constituted by writing, understood here as a structure of difference and deferral.

Let me first unpack the consequences of describing context as a fundamentally written phenomenon. We might feel like this position is essentially skeptical, one that somehow makes context less “real.” We might recall that overly cited and poorly understood line from Of Grammatology, “il n’y a pas de hors-texte,” translated controversially by Spivak as “there is nothing outside the text,” and which has been held up, out of context, as evidence of Derrida’s flat-footed denial of the existence of history or the real outside of its representation.19 But if we retain the Derridean sense of writing as a transcendental structure, then we are obliged to jettison any pre-theoretical identification of writing with representation; writing would rather be the mark that makes possible the difference between representation and represented, signifier and signified, or text and context, and Derrida’s statement would refer to the impossibility of any position independent of this structure.

“Writing,” in fact, would name the necessity of context—the necessity of there being something other than the text—and the necessary limits of any particular text in its confrontation with, and relation to, something other than itself. The structure of writing, which is a differential structure, requires that there be context, that speech acts and written texts take place in relation to other texts as well as to the material world, where they are subject to norms, power dynamics, and material events, including the event of inscription.

While Derrida tends to characterize writing in terms of absence, spacing, and rupture, I noted earlier that it might be useful to describe writing as a differential relation. To emphasize its relational character is also to emphasize its productive nature: the active, putting-in-to relation that writing necessarily effects. In this sense, to claim that context is written is to claim that it is posited—or, I would argue, performed—in relation to a text and that the determination of both text and context depends on this performance. That is, the act of determining a context can be described as a performative speech act, an act of language drawing on conventional systems of meaning and behavior and producing concrete effects in the world.20

When I describe the act of determining a context as performative, I do not mean to advocate for a voluntaristic model of criticism, one where each critic would be free to pursue the construction of context however she wishes. Performativity appears this way only if we take the conscious, self-possessed individual to be the sole possible figure of intentionality. But if we retain Derrida’s insistence on intention’s non-coincidence with itself, and hence on the category of intention’s being subject to a differential typology, one that I suggested would include agents of the unconscious as well as institutional norms and social groups, then we arrive at a more comprehensive picture of performative utterances. Such utterances can be both intended in the ordinary sense and bound by multiple constraints. Moreover, we open up discussions of the persuasive nature of performative speech, of the way such speech serves to convince or transform others or comes to include or exclude other forms of intention.

Highlighting the persuasive aspect of claims about context gives us a way to understand why context is subject to so many different descriptions and figurations. Is context a box or a network? Does it reduce or amplify a text? Is it displayed on the surface or hidden below it? These spatializing figures, so prevalent in literary-critical debates, come into view more clearly as figures when we begin with a notion of context itself as a performative utterance. I mentioned above Felski’s writings on context. Her work is a particularly interesting case to consider because of the way it transports Bruno Latour’s language from sociology to literary studies. In Reassembling the Social, Latour rails against appeals to “social context” that would treat as a determining framework for various phenomena what should rather be an object of inquiry for the social sciences, namely “society.”21 For Latour, unlike most literary critics who make arguments for or against context, what is at issue is not “historical context” but an explanatory framework, or as we might call it, “theory.”22 Hence “context stinks,” Latour can claim in the words of Rem Koolhaas, because it is “simply a way of stopping the description when you are tired or too lazy to go on.”23 Felski makes a parallel move when she faults literary historicism for its efforts to control the narrative of what a text means, “to box it into a moment of origin, to lock it up in a temporal container.”24 Yet, as we observed earlier, what she advocates is not at all a jettisoning of context in favor of a narrower focus on the text “itself,” but rather an expansion of our considerations to include “the fact of association, the coming together of phenomena to create assemblages, affinities, and networks.”25 Felski seeks to replace one image of context—context as box—with another—context as network, and she clearly states her preference for the latter. What she prefers, more precisely, in the connection between text and context, are relations of contiguity and proximity over relations of determination or identification; for her, the text is “imprisoned” by historicist readings because it is identified with its historical context without remainder. At the level of rhetorical figures, we observe a preference for what we might call metonymic associations over metaphoric substitutions.26

The same preference appears in a number of literary critics’ objections to contextualizing claims. The relation of context to text is often portrayed as overly deterministic, with Marxist criticism serving as a familiar target. Raymond Williams, for example, in his 1977 book Marxism and Literature, describes a caricature of Marxist literary criticism in which it is seen as “a necessarily reductive and determinist kind of theory: no cultural activity is allowed to be real and significant in itself, but is always reduced to a direct or indirect expression of some preceding and controlling economic content.”27 More recently, Christopher Lane has referred dismissively to the “deterministic net” of contextual analysis,28 while Sharon Marcus and Stephen Best have valorized “surface reading” as a form of “sustained proximity to the text” against Fredric Jameson’s injunction to critics to “rewrite” the text “in the stronger language of a more fundamental interpretive code.”29

Again, it is attentiveness to the nature of the text-context relation as a performative relation that enables us to see these figures of context as figures, revealing the value judgments that trail in their wake, and allowing us to imagine the possibility of their being written or performed otherwise. What do critics stand to gain, we might ask, from their choice of metonymic figures over metaphorical ones? The language of metonymy is the language of chance encounter and fortuitous association—of “communication” in the particular sense of communicating a force. Paraphrasing Derrida, we might say that this associative context is “said to be pertinent [relevant] when it touches: the object to which it seems to refer, but also … its addressee, upon whom it produces certain effects.”30 This is a spatializing language that emphasizes the bald contiguity of things, whereas the language of metaphor, in contrast, refers us to the synthetic, mental activity of identifying and understanding—to “communication” in the sense of communicating semantic content. Metonymy is a language of power, then, more suited to descriptions of the natural world, while metaphor is a language of meaning, more suited to the world of consciousness and intention. There is a certain caché, I think, in denying our interpretive role as critics, positioning texts as “agents” (Felski) and ourselves as passive and receptive (Best and Marcus). It lends a veneer of natural necessity to our practice. But even those critics who hold up contextual claims under the sign of sheer proximity, who point to the relevance of historical details because of their contiguity in space or time, do so under the assumption that this contiguity will be a meaningful one. I see here a sort of anticipatory grammar of context, a formalism of the contextual act, in which the text-context relation is expected to be legible.

Derrida, at the end of his essay, likewise trades on the value of a language of power when he describes deconstruction as a “labor,” one that can “intervene” in a “field of nondiscursive forces” and “displace a conceptual order” (SEC, 21/392). Yet in his reading of Austin, we see a different path open up, one where power gives way to meaning. For while Austin turns to speech act theory to free language from evaluations of truth and falsity and open it to an analysis of force—an analysis of how language can “perform” or effect a change in circumstances—Derrida shows that Austin still reserves a central role for intention. On Derrida’s account, the total situation of the speech act in context is still underwritten by the conscious, intentional presence of a speaker. Extrapolating this insight to a literary-critical register, I draw the lesson not only that our literary contexts are constructed rather than discovered, but that they are constructed intentionally, in the service of particular interpretive claims or scholarly interests. What the analysis of context as performative utterance brings out is the way that both relations of force and relations of meaning are at work in acts of contextualization. These acts produce effects in the material world, but they also function as vehicles for intentional meaning. Contextualizing claims made by critics are performative because they are intentional acts, buoyed or constrained by convention, and subject to the same structural indeterminacies as other kinds of intentional acts. This is not to say that context does not matter but rather to suggest that the way it matters, the force that it has, and for whom, remains to be determined in our acts of reading.