To judge by recent trends in literary criticism that fall under the general rubric of material culture studies—practices as varied as history-of-the-book, thing theory, and various types of quantitative methods in the humanities—the matter of what counts as materialism in literature is far from settled. These trends tend to read literature's materiality literally, considering matter, objects, and things as the physical remainders of a pragmatic phenomenalism whose ultimate horizon of inquiry is the referential context in which the literary operates, rather than the operations of the literary itself as a historical referent. Much methodological stock is put into the work of analogy, so that, by treating the literary object as just another object in a world of objects, literature's historical impact is determined by tracing the specific modes of production, distribution, and consumption that make literature a cultural commodity in the global marketplace. And even when the literary object is understood as a symbolic repository of material history, an object that contains other objects, it is its formidable mimetic powers that are made to account almost entirely for its cultural work. The historicity of literature, in this view, rests on its ability to influence, coerce, or otherwise discipline its readers, who are often portrayed as the always-already interpellated subjects of an ideology literature can do no more than re-inscribe as a narrative of subject construction.

The materiality of literature, in any case, remains elusive: on the one hand, to study literature as an object is to bracket its defining characteristic as a literary work–the fact of its literariness; on the other, to limit the historical role of the novel to the effects it has on its readers is to subscribe uncritically to its referential function–the fact that it can only be read literally, as a medium to the material culture that is said to exist outside of, and in spite of, its qualities as literature. Accordingly, despite the claims they make about literature's historical impact, these accounts tend to occlude the materiality of that which makes literature matter: its language. This is a surprising occlusion, not least because there has been a longstanding awareness of the materiality of the literary that in many cases predates current critical trends within material culture studies and object-centered literary analyses that ostensibly seek to specify literature's historical role.

I don't wish to suggest that we should rehabilitate the notion of linguistic materiality uncritically, but we ought nevertheless to recall its speculative force in the arguments we make about literature's material history, if only as a first step toward providing a more robust account of the specificities of literary practice as well as its institutional determinations. A very partial critical genealogy—from Roland Barthes's formulation of the "le degré zéro de l'écriture," through Lacan's "agency of the letter," and Paul de Man's formulation of materialism as a "radical formalism," to more recent elaborations by Catherine Malabou of a material plasticity that would reject the literary as a quasi-metaphysical category—suggests, at any rate, that critical debates on the materiality of the literary are far from having been exhausted. The challenge is to find a new way of formulating literary empiricism that lends specificity to the literariness of the literary object while enhancing its claims to historical relevance.

In what follows I address in a preliminary and provisional way two questions that move in this direction: 1) Can a meaningful distinction be made between linguistic and literary materiality? And, if so, 2) is there a conceptual advantage in construing this singular essence of the literary as a form of immanence?

The most commonly cited model of linguistic materiality is the one formulated by Paul de Man in his reading of Immanuel Kant's third Critique (1981, but not published until 1996). In the essay “Kant's Materialism,” de Man refers to the radical formalism of the language of poets, who, according to Kant, look at the world as one sees it ("wie man ihn sieht"), as an absolute materialism of language that entertains no notion of reference or semiosis. To the extent that language at its limit does not act as a repository or conveyor of meaning, it can no longer be said to be the immaterial medium of communication that we assume to be operative when we speak of the inside of a text and of its outside referent. Yet, however distant from a materialist conception of culture or history this definition of materialism may seem at first glance, de Man insists that it is precisely the "a-referential, a-phenomenal, a-pathetic" nature of this materialism that allows Kant to gain access to the "moral world of practical reason, practical law, and rational politics" (128).

In de Man's formulation, materialism is a non-phenomenalizable attribute of language that nevertheless has effects in the world that are empirically verifiable, insofar as they belong to a pragmatic order of discourse. This form of materialism is therefore historical in a strong sense, even if it does not strictly conform, in this telling, to what we would recognize as "historical materialism." When, in The Political Unconscious (1981), Frederic Jameson makes a distinction between a mechanical form of materialism of the sort found in Lacan (the phallus as "material signifier") and the properly historical materialism elaborated by Karl Marx, he is trying to distinguish a conception of matter that corresponds to signifying systems, and thus to bourgeois ideology, from a notion of materialism determined by the mode of production. For Jameson, literary labor can at best be subsumed under the rubric of ideology critique; it cannot, at any rate, be assimilated to "real work on the assembly line and to the experience of the resistance of matter in genuine manual labor" (45). De Man's formulation of materialism is not dialectical in the way Jameson suggests a properly historical materialism ought to be, but it is nevertheless closer to this mechanical model of history than would at first glance appear to be the case.

Indeed, even among Marxist critics who focus on the "mode of production," there is no consensus regarding the best method of approaching the materiality of literature. In a 1991 essay aptly, though perhaps optimistically, titled “The Moment of Materialism,” David Simpson notes: "[T]he very popularization of materialist methods, with their accumulating adjectival qualifiers—cultural materialism, historical materialism, and (let us not forget) dialectical materialism—brings with it another anxiety of definition" (6). For Simpson, this anxiety has to do less with matter as such than with history. To make this point, he goes on to quote Engels, who wrote in 1890: "In general, the word 'materialist' serves many of the younger writers of Germany as a mere phrase with which anything and everything is labelled without further study, that is, they stick on this label and then consider the question disposed of. But our conception of history is above all a guide to study, not a lever for construction after the Hegelian manner. All history must be studied afresh" (qtd. in Simpson 6). Leaving aside the question of whether or not literary criticism uses the term "materialism" as a "lever for construction," it does seem to be the case that, to judge from the process of redefinition to which it is periodically submitted, the history of the term, if not history itself, is certainly "studied afresh."

Frances Ferguson makes a valuable distinction between what she calls "deconstructive materialism" and "social materialism" on the basis of the relation between language and the objects to which it refers. In the latter, objects are endowed with meaning to the extent that they can enter or be entered into a causal relation with individuals; the former, in contrast, emphasizes the "primacy of the material signifier in both embodying and interfering with any and all such casual accounts" (10). It follows that the view of language as a referential relay that transparently connects our perceptions of objects with the objects themselves comes under suspicion as the materiality of the letter becomes itself an object of knowledge whose very opacity seems to be derived from its apparent insubstantiality.

In de Man, the materiality of language is what remains when the text is evacuated of its historical and humanistic determinations, and can therefore only be accessed in language by means of its allegorizations—even—or especially, when these are of a performative order. It is tempting in this context to equate the materiality of language with the institutional history of performative speech acts so as to be able to claim that literature is itself an act of speech that has real effects in the world. But this would be to confuse, as de Man once put it, "the materiality of the signifier with the materiality of what it signifies." Andrzej Warminski coins the term "super-performative" to designate not a performative that "functions within an established juridico-political system, but rather one that itself is the inaugural act of positing that puts such a system into place in the first place" (26). When Derrida refers to what de Man, at the closural moment of his essay “Phenomenality and Materiality in Kant,” calls the "prosaic materiality of the letter" as a materiality that is not a thing or a body but that yet "operates" and "forces," a "materiality without matter" (350), he is attempting to preserve the radical unreappropriability of this force of language to language itself.

Formulated in these terms, the materiality of language seems to be an attribute of language that operates without recourse to any properly literary function. It is "prosaic" in that it would not even be determined by language itself. Yet, the "absolute, radical formalism" that marks for de Man, in his reading of Kant, the passage from reference to performativity should be read literally. It makes sense from this perspective to speak of a literary materialism as opposed to a linguistic materialism, provided we revise our commonsense notion of the literary as a referential medium that captures objects, things, matter. We must understand literature, in other words, as poets do; that is, as language itself.

But how can we access the historical, empirical effects of language-as-literature? What form of historical agency does it constitute? Gilles Deleuze's conceptualization of immanence (or the plane of immanence) can help us stage an empirical thinking of the literary that might provide an alternative model for historical agency. In “Immanence, A Life,” Deleuze maintains that the virtualities, events, and singularities that constitute life elude the transcendence of both subject and object. The transcendental is distinguished from the transcendent in that the empiricism of the transcendental plane is unencumbered from the appeals of Some Thing as a unity superior to all things, or of a Subject as an act that brings about a synthesis. Immanence is not immanent to anything other than itself. Can a description of literary immanence equally elude the insertion of subject and object as categories mediated by a consciousness that would reabsorb them as transcendents?

Consider how Deleuze illustrates his notion of immanence: "No one," writes Deleuze, "has described what a life is better than Charles Dickens, if we take the indefinite article as an index of the transcendental" (28). Deleuze goes on to describe the passage from Our Mutual Friend in which Rogue Riderhood, found nearly dead in the river, is revived before the expectant eyes of his neighbors, whose sympathy swiftly evaporates as soon as they realize Riderhood is once again restored as Riderhood. This is how Deleuze glosses it: "The life of the individual gives way to an impersonal yet singular life that releases a pure event freed from the accidents of internal and external life, that is, from the subjectivity and objectivity of what happens...a life of pure immanence, neutral, beyond good and evil, for it was only the subject that incarnated it in the midst of things that made it good or bad" (28-9).

Deleuze does not quote Dicken's text directly, but something similar can be said about the way Dickens himself renders the moment of Riderhood's death and his curious resurrection. At first, Riderhood's inert body inspires a modicum of sympathy from those whom Riderhood, in life, has otherwise alienated. ‘Doctor examines the dank carcase, and pronounces, not hopefully, that it is worth while trying to reanimate the same. All the best means are at once in action, and everybody present lends a hand, and a heart and soul. No one has the least regard for the man; with them all, he has been an object of avoidance, suspicion, and aversion; but the spark of life within him is curiously separable from himself now, and they have a deep interest in it, probably because it IS life, and they are living and must die. (Dickens 443)’ Notice in this passage the use of the present tense, a device common enough in Dickens when a sense of urgency gives narrative expediency to the event being described, but here explicitly bracketing Riderhood's reanimation as though the chapter as a whole in which it is effected had emerged—had come alive in the present or been restored—from an otherwise conventional narrative flow rendered in the preterite. The preterite tense, as Barthes notes in Le Degré zéro de l'écriture, is one of the Signs of Literature, together with third-person narration, that gives the Novel the "formal guarantee of the real" (33). The preterite is a mythological object that appeals to the ideology of the real and, at the same time, points to the artificial nature of novelistic narration. Dickens's deliberate use of the present tense in this extended episode has the effect of making it an object of literary curiosity while momentarily deadening the rest of the narrative.

Similarly, when the passage apostrophizes Riderhood directly, as though to call him back to life, the structure of address breaks with convention by creating a first-person plural subject position from which those attending to the "flabby lump of mortality," and possibly those "below," are all given voice. ‘If you are not gone for good, Mr. Riderhood, it would be something to know where you are hiding at present. This flabby lump of mortality that we work so hard at with such patient perseverance, yields no sign of you. If you are gone for good, Rogue, it is very solemn, and if you are coming back, it is hardly less so. Nay, in the suspense and mystery of the latter question, involving that of where you may be now, there is a solemnity even added to that of death, making us who are in attendance alike afraid to look on you and to look off you, and making those below start at the least sound of a creaking plank in the floor. (444)’ The distance separating the more formal "Mr. Riderhood" and the familiar name "Rogue" with which most of the characters in the novel refer to him is also marked as a grammatical distinction between a form of address that would refer to him as though absent in the third person and a positing of the name that would make him a familiar presence. Moreover, the phrase "no sign of you" both signals Riderhood's precarious hold on life, and also, if read literally, the absence of the "you" being apostrophized, as though the sign "you" had no purchase in the scene of address staged between life-and-death. The effect is a radical suspension of reference that yields no "you"—or, for that matter, no "I"—that could become a subject of enunciation, whether plural or singular, and thereby suspends the action of the novel insofar as the subject pronoun carries the narrative. Indeed, it is the "he," according to Barthes, that performs the service of both signifying the novelistic as artifice and naturalizing this artifice as the "real."

Riderhood's precarious narrative existence is accompanied (indeed accomplished) by a process of formal de-realization that Dickens began in this episode by reversing and displacing conventional novelistic forms of utterance such as grammatical tense and narrative voice. This process is radicalized by a virtual suspension of novelistic discourse itself. ‘Stay! Did that eyelid tremble? So the doctor, breathing low, and closely watching, asks himself.’ ‘No. ’ ‘Did that nostril twitch? ’ ‘No. ’ ‘This artificial respiration ceasing, do I feel any faint flutter under my hand upon the chest? ’ ‘No. ’ ‘Over and over again No. No. But try over and over again, nevertheless. ’ ‘See! A token of life! An indubitable token of life! The spark may smoulder and go out, or it may glow and expand, but see! (444)’ The almost theatrical disposition of this paratactic passage (almost because the "No"s could hardly be staged) suggests the rejection or the abandonment of novelistic convention, as thought the convention that would portray Riderhood coincided with his ability to remain alive. To be sure, the iteration of "No" in this passage marks the interruption of narrative coherence and representational verisimilitude, but the loss of referential logic also makes visible a linguistic struggle whose stakes are nothing short of "life" itself. The use of the "No" seems at first to negate the force of the exclamation "Stay!" (a locution with no stated subject of enunciation) and of the doctor's questions, asked in silence, to himself, "breathing low." Yet, the insistence of the evidentiary "See!" and the ambiguously referential "a token of life!" (ambiguous because while "life" has no adequate referent in reality, the use of "token" in the phrase suggests that it is figuratively accessible) constitute a catachrestic positing of the materiality of language itself. The performative force of the utterance "No" may well deflate the attempt to reanimate Rogue through the mere act of naming, but a rhetorical counterweight is found to take the place of a nominal accounting of "Mr. Riderhood" in the figure of the "spark."

The figure of the "spark" returns in another passage that marks Rogue's reinscription into the social coordinates of his past life, and thus into the novelistic elaboration of his identity. Indeed, the language of the passage hints at a return to conventional novelistic modality: ‘The spark of life was deeply interesting while it was in abeyance, but now that it has got established in Mr. Riderhood, there appears to be a general desire that circumstances had admitted of its being developed in anybody else, rather than that gentleman. (446-7)’ The radical formalism of the entire episode makes the rendering of a life a matter of great literary interest, as though the text itself were coming alive before our very eyes. The text's use of the "spark of life" to mark this passage suggests that, as a figure for life—for a life, as Deleuze reads it—the "spark of life" can also be read as a figure for what I have been calling literary immanence, a material existence that renders literature alive. This rendering-alive of literature might get us closer to a thinking of empiricism from which the historical, doubly empirical determinations so familiar to material culture studies could be used to think of agency as a material becoming of the literary.