HomeIntroducing CybercultureBook ReviewsCourses in CybercultureEvents and ConferencesFeatured LinksAbout RCCS

View All Books

Cognitive Fictions

Author: Joseph Tabbi
Publisher: Minneapolis, MN: University of Minnesota Press, 2002
Review Published: April 2004

 REVIEW 1: Kathleen Fitzpatrick
 REVIEW 2: Jen Webb

Cognition Against Interpretation
March 26, 2004

I was well enough aware of my book's flaws before reading the set of reviews published in the Resource Center for Cyberculture Studies. What I wasn't so sure about was the book's "potential to radicalize literary criticism" (Webb). That's more than I dared to claim, although at high points in the writing, I may have thought as much. But since I've been called out, I'll try now to follow through on the question of what changes, substantially, when critical writing is done within a cognitive framework.

Is there a cognitive approach to Literary Studies? Must the work of literature, submitted to such an approach, be regarded as a technology, the tradition itself a “collective system” (Morretti, “Graphs”)? If criticism is to avoid “[treating] metaphors as actualities,” as Jen Webb thinks I tend to do in Cognitive Fictions, we have to know where to draw the line. Yet such distinctions are far from settled in the cognitive sciences themselves, concerned not only with “the metaphors we live by” (George Lakoff and Mark Johnson, http://theliterarylink.com/metaphors.html) but also their material basis in the brain. That metaphors can be more than decorative, that they do not so much reflect as create the world we see and orient ourselves within, is by now generally accepted. What is still debated is whether the parts of a metaphor, source and target, are processed in the brain as coded or visual representations, or if the brain avoids symbolic activity altogether. So long as these questions are open, and openly discussed, in the sciences, we should not be too quick to assume, on the literary side, that we already know the difference between metaphor and actuality, representation and material inscription, a sign and the world that emerges in signification. A redrawing of the lines is arguably what a cognitive literary criticism needs to be about.

When I began writing Cognitive Fictions in the Winter of 1998, the Modern Language Association had just established its permanent session, “Cognitive Approaches to Literature.” I was drawn to these meetings because they promised a disciplinary alternative to a certain skepticism that had overtaken the literary field, producing a dreary conceptual uniformity despite an apparent opening of the profession and practice of literature to previously excluded voices and identities, a number of neglected texts and writers, and what felt like a still greater number of mutually excluding approaches. What unified all these approaches, however, and too many of the newly valued texts was, in the first place, skepticism over the constitution of any reality that is not text-based or “constructed” by language and human cultures. Secondly, at just this moment when new interpretations and a range of new critiques could be voiced, few newcomers were willing to question whether interpretation and critique were themselves the most important activities on the literary agenda. Among the cognitive researchers, by contrast, I found a willingness to question the universality claims of critique and constructivism, not by asserting a reality untroubled by the presence or absence of language, and not by denying the importance of culture, discourse, metaphors, and narratives in the development of texts and minds. A more general "approach" would at least keep track of the lost reality, starting with the mind's own desire for what its concepts, discourses, and narratives no longer even pretend to touch.

It would be six years before a summary of work in the field, and a number of exemplary essays, would be gathered together by Ellen Spolsky and Alan Richardson in the edited collection, The Work of Fiction (Palgrave, 2004). Here we find a number of named approaches, “Cognitive Rhetoric and Conceptual Blending Theory,” “Cognitive Poetics,” the much maligned “Evolutionary Literary Theory,” Cognitive “Narratology,” “Materialism,” and “Historicism.” Of necessity, my own approach was eclectic, and concerned less with literary convergences than with ideas that I found directly in the writings of scientists, mostly on the question posed forthrightly by Gerald M. Edelman and Gulio Tononi, of “How Matter Becomes Consciousness.” I knew that such work was controversial and that I, an outsider, would have to take the scientists' credibility on trust. (And I would have to trust, in turn, that the scientists wouldn't start playing pranks, Sokal style, if I took a speculative rather than a popularizing attitude toward the scientific work.) It always seemed to me that literature gains more from science in progress, in all the uncertainty of open exploration, than any writer ever gained from a dutiful study of results. I was not looking to science to tell me the nature of the world in certain terms; least of all was I concerned with the specifically “literary” results produced by cognitive researchers (cataloguing the mistakes made by a child learning to read, tracking a reader’s eye movements, and so forth). I decided early on not to be limited by any one “school” or position within the cognitive debates. What seemed most desirable in an emerging cognitive criticism was the very eclecticism and flexibility that researchers were discovering in the mind’s operations - its ability to bootstrap, to switch levels from substantive operations to conscious meanings, and to retrofit perceptual organs that had evolved for one function so as to perform altogether different functions in a changing environment.

The goals of criticism, I expect, will never coincide with scientific goals of repeatability; and writing about literary works does not actively produce knowledge the way a scientific discovery adds to or revises current knowledge. Critics for example rarely use a hypothesis or mathematical protocol to make literary predictions or to make sense of literary texts. Franco Morretti is a notable exception, one of his comparative studies actually maps the plots and character movements in nineteenth-century European novels, and so offers visual and statistical evidence of the ways that geography is not destiny, exactly, but rather an active informational space that makes certain stories possible and prevents others from taking place: the characters in Dickens's London, with its social stratifications evident in the city layout, carry a different set of longings and motivations than those of Balzac's Paris (with its Left and Right Banks) or Scott's frontiers (See Atlas). More recently and more ambitiously, Morretti has offered an evolutionary account of the spread of novelistic forms and conventions over the space of European and Asian empires, from colonial center to colonized periphery but only rarely back (and then, only after the genres and conventions have already stabilized). The novel's multiple "rises" over time, in terms of number of editions produced each year, are given as the ‘world-system’ itself develops. These observations are in fact suited to algorithmic and statistical description but only insofar as the project remains at the level of formalism or literary history, not at the level of meaningful sentences.

The avoidance of interpretive “meaning” in Morretti, though it may separate him from the majority of his critical contemporaries, puts him in the company of quite a few - although not all - cognitive and materialist critics who wish to contradict the universality claims of interpretation in the humanities, and who seek, from a philosophical standpoint, forms of critical practice that do not deny meaning so much as they consider “What Meaning Cannot Convey” (to cite the subtitle of a recent book by Moretti’s Stanford colleague, Hans Ulrich Gumbrecht). Like Gumbrecht and Morretti, I have grown weary of criticism that seeks only to make a little more sense, to locate an unnoticed source or context, or to add a still more brilliant twist to the interpretation of the same restricted set of texts. According to Morretti, critics of nineteenth-century British fiction write on "less than one per cent of the novels that were actually published" ["Graphs"]). So a bit more archival interest in works of literature as such could itself transform a discipline, and (today) it might not hurt if we had a few less dissertations on DeLillo and Morrison. Better still, we might make the uncovering of at least one under-appreciated novelist a requirement for advancement in the literary Humanities.

Like Moretti, Mark Turner in Reading Minds (note 1 below) has sought to revise the scholarly focus on what is exceptional and complex in literary texts and consider instead the unnoticed complexity of what they have in common, the schemas, conceptual pattenings, and unconscious apparatus that together make meaning possible. Rather than look to the sciences for just one more approach among competing approaches, Turner proposed a cognitive program that would be, in his words,

    a new common ground for the profession of English: the analysis of acts of language, including literature, as acts of a human brain in a human body in a human environment which that brain must make intelligible if it is to survive. (Reading Minds vii-viii)
The wording of this straightforward but carefully paced sentence is itself an indication of how the cognitive program might encompass, rather than further fragment, the literary field. Literature is to be considered as a process not an object of study, as "acts of language" not a specific set of texts to be interpreted. As Turner says elsewhere, concepts are "made intelligible" not by extracting meaning but rather by following the "activated set of links in a pattern" (Reading Minds 45). Also noticeable in the above formulation is a Darwinian inflection that would subordinate human history, and its short-term constructions, to the pre-cultural constructions of evolutionary history. What took billions of years to come into existence, now becomes understandable as a set of mental processes and that are modularized, protected from the environment that produced them, and largely closed to consciousness. Literature itself, in Turner’s view, “hardly belongs to consciousness” - rather the work of literature, like the movement of thought, is done using a large fund of unoriginal conceptual structures, "linked subprocesses that are themselves composed internally of the activation of links" (Reading Minds 45). And because these common structures cut across narratives from all human cultures, Western and non-Western, Turner does not elaborate differences so much as he looks to common embodied experiences for a shareable cultural knowledge and a generalized literary appreciation.

The terms of comparison, in such a research program, are no longer individual works or sets of works from different genres, periods, or national frameworks (English vs French literature, French vs German, American and Russian singularity in the face of European traditions, and so forth): in a cognitive approach, what is comparable, rather, are modes of cultural production and the organized production of natural systems, not least the mind. Instead of undertaking literary study on a case by case basis, "giving" and "arguing for" "readings," both Moretti and Turner seek a general knowledge of the whole -- in Morretti, the measurable movements in literary production and circulation; in Turner, "the linguistic or conceptual resourses" used in reading (Reading Minds 14). Indeed, the emphasis on such large-scale, unifying structures as maps (the familiar, geographical mapping in Morretti; the cognitive mappings below consciousness in Turner), systems, and patterns of activated links, brings both approaches into the same conceptual territory as contemporary work in hypertext and network architectures - structures that also originated in a desire (often expressed by hypertext pioneers Ted Nelson and Vannevar Bush) to grapple with an exponential expansion in the number of texts produced and published.

The commitment of all these critics to a greater openness and a fuller involvement in the wealth of literary production carries its own danger, however - that readers will forget the specific, uniquely literary presence, and the necessity through direct citation to bring this work into contact with other living texts and ideas. The number of books a reader reads, the reader's memory of plots and characters and unique poetic utterances - none of this matters nearly so much as the conceptual schemas and generic structures that get carried forward into the novels we are reading at the present moment, and which in the past few decades have migrated into films, video arts, aesthetic theories, and networked narratives. Here is where a cognitive approach has something to contribute to a uniquely literary program: through attention to the moment by moment workings of the mind, readers can better appreciate the moment by moment workings of thought and experience that is recoverable in the shape of sentences themselves, and knowable only in the present moment of reading (as a thought is knowable only during the two or three seconds when it is held in mind, before the thinker moves on to the next thought; and this is true despite the fact that no thought can be had without reference to past thoughts, schemas, and images whose selection is made quite easily, even if usually we cannot say why.)

Rather than stand in awe at the impossible-to-cognize production of novels and poems, a plenitude that only increases in today's multi-mediated multiculture (to the point where the word "awesome" expresses only indifference), a cognitive approach inverts the perspective and asks instead: of all the millions of possible choices, why at this moment do we chose this work for reading, and not the others? And once we have come to an understanding of one particular work, grounded in the half-forgotten knowledge of all the other works we happen to know, what new work, as yet unknown or unrealized, do we desire to read next? In a media environment where all works are linked to all other works - literally, materially linked - what contacts does this work, the one that preoccupies us at the moment, compel us to follow? A cognitive approach is likely to be radically of the present, primarily spatial, not temporal in its orientation: the critic looks outward toward opportunities for connection, not back in time for historical trajectories or forward in pursuit of an aesthetic project consciously directed and continually checked by critique. To be sure, a cognitive literary criticism must account for change from one moment to the next, in literary works and literary traditions no less than in psychic systems. But such a criticism tends to address change spatially, through selections out of a present environment that match up with a work's own developing themes, motifs, and meanings.

The operative metaphor of Cognitive Fictions is that literary creativity as such - and not only experimental writing - participates in the same reflexive, self-creative or autopoietic structurings that characterize mental processes. What makes the cognitive metaphor operative, not arbitrary, are parallel developments of a “materialist, biologically based cognitive science” (Spolsky, Gaps in Nature) and a criticism newly concerned with how texts, and the working imagination, are shaped and constrained by materialities of communication. It certainly does make a difference, to meaning or to possibilities for textual circulation, whether a work is written with “characters on wax, papyrus, or parchment” (Gumbrecht, 9). Still more, historical studies of media from Walter Benjamin in the 1930s to Marshall McLuhan in the 1960s to Friedrich Kittler in the present, have been demonstrating in ever more convincing detail how “intellectual movements” themselves are “triggered by technological innovations and by their application in the invention of new communication media” (Gumbrecht, 9). Hence another factor capable of enriching a cognitive approach has been the contemporary emergence of networked media, an environment where thought’s inscription can be observed in process, not only in its printed results, and where non-linear and multi-linear narratives can be readily experienced as the rule rather than the exception, by a mass readership with the capacity to become, at any moment, a multitude of authors.

Such a rich production of literary potential, the mutual observation among artists working in film, print, and digital regimes that John Johnston (1998) has termed "information multiplicity," is both the promise and the bane of contemporary literary studies. Trying to avoid extremes of high or low culture, populism or elitism, I worked hard in Cognitive Fictions to engage the media environment without losing the particularity, the literariness, of the literary work as written. That is why I kept coming back to the concept of autopoiesis, a self-production that preserves the unity of a system (at the level of organization) while remaining open to information multiplicity at the level of content. Whether the system is animate or inanimate, whether it is a text or a material texture like the human brain, the system first has to mark itself off from an environment - but this very act of boundary definition itself belongs to the system, not the environment (which is unmarked). There is no neutral boundary capable of distinguishing absolutely between meaning and noise, self and other, truth and untruth, and so forth: the boundary is always a self-referential product of systems operations. Drawing a boundary allows the system to observe itself and (some) elements in the environment (while at the same time preventing the observation of countless other elements). Every distinction between a literary work and the many operative systems in the expansive environment of media, materials, and social communications is essentially an act of self-reference, not because the work exists independently; rather, the drawing of distinctions, and the preservation of the work's internal organization is a way of selectively binding the work. The selections, however, are never wholly conscious or predictable, any more than our next thought can be consciously directed for very long and still produce fresh and interesting ideas. (note 2)

Without this key distinction between organizational closure and informational openness (which I may not have worked out fully in my book), I realize that the autopoietic approach is easily confused with old-fashioned literary autonomy, a key concept within the modernist literary tradition in which I was trained and whose signal works I continue to read and enjoy. Still, the readings I undertake in Cognitive Fictions are not modernist close readings, nor is the book's "media theory of the unconscious" identical with McLuhan's conception of media technologies as "extensions of man." Media are, rather, props and supports for consciousness, necessary not to the (usually harmful) extension but rather the creation and reproduction of selves: an infant who is raised in a crib and who moves about in a stroller will have a different world outlook than an infant in a hammock who learns to walk with the help of ropes and the supporting hands of children and adults. A text that is written with print publication in mind will come into the world differently than one whose audience is present all together at a conference, passing individually through a website, or posting to an email list. In both cases, a certain kind of presence is produced, which is certainly human, brought to us through hands and vocal cords and a material page or screen. But what is present is not only human, and what we imagine in reading is not what goes on in another person's psyche.

Jen Webb criticizes me for ignoring the "the importance of authors in crafting, organizing, and finally producing and disseminating texts"; she says I even forget the "fundamental principle" that communication "requires an actor (or two)." But why stop at two, when the contemporary media environment has so many human and non-human actors who exert their presence in any communicative circuit? Because I emphasize systems rather than individual psyches, Webb misses in my work an account of the "various literacies" used by readers and authors in "decoding texts." But texts, if they are literary at all, are not embedded meanings waiting to be decoded. A mixture of elements, including texts, psyches, whole societies, are imbricated in any communicative process. To limit communication to human actors, as Webb would have us do, and to limit the study of literature to what is consciously crafted, organized, produced, and disseminated by authors and readers - this is itself a reduction of diversity, producing in place of the literary an endless array of "literacies": subject positions, professional competencies, stylistic techniques, and so forth. Instead of stimulating individual creativity through a recognition of what is common to human creativity, texts confront readers with ever more specialized, ever less familiar languages whose differences demand "decoding" among knowing equals, explication by pedants, or (when the author is thought to be politically or socially more powerful): ideology critique. Communication itself, when it works, is endless and endlessly productive of differences that do make a difference. In conversation we are always interpreting and always being interpreted, always on guard against being appropriated, and not always averse to appropriating. We are only human - in conversation. But for this reason, human communication is the last place we should look to, when modeling an approach to works of literature. The value of a cognitive alternative, in my view, is precisely its willingness to recognize agency not exclusively in the production of cultural and linguistic differences, but in the material supports and systems that are needed to bring authors and readers together.

Kathleen Fitzpatrick raises similar concerns, when she fears that I run the risk of depoliticizing criticism: "is it a coincidence," Fitzpatrick asks, "that all of the authors of print cognitive fictions that Tabbi encounters are white men?" Of course not. Among the last generation of writers in the U.S. to work exclusively in print, white men have been goaded the most by the loss of world-reference; it is not "hegemony," but its transformations, that hurt the imagination into self-reflection. But neither is it coincidental that the one hypertext I encounter at length, by Stephanie Strickland, deals specifically with the problem of re-writing, although Strickland does not cast the problem exclusively as an act of political resistance, postmodern irony, or reversed perspective (the way that recent novelists have, for example, given the narrative over to Lolita or Ahab's wife). Changing material constraints on what can be said is no less important, in new media environments, than the loosening of cultural restrictions on who can speak. In Strickland's series of interlinked poems, the figure who emerges as "Re.Rainna," the woman re-writing, is in fact not only a woman, but a material presence at work re-drawing boundaries, changing levels from the literary to the mathematical to the cultural and back. Such a narrative self-consciousness, understood as a relocation of selfhood and agency in an environment where boundaries have shifted, is way more interesting, I believe, than the "writings about the process of writing" (Webb) that one finds under the term "meta-fiction."

The one thing I would wish to avoid above all, in a criticism informed by science, is for the specifically “cognitive” dimensions to be marked off and treated as one among many highly specialized areas of communication, best left not to ordinary readers, but to experts specially equipped to articulate and interpret certain refined techniques. But as the literary terms join up with the terms of an emerging science of mind and brain, what once seemed particular and special - even a faded fashion - becomes much more general and common. Within my home field of contemporary American fiction, at least one commentator has noticed how the cognitive framework helped to correct some "misunderstandings" and identify "a new style of metafiction" beyond the narrowly literary reflexings one finds in a certain strain of postmodern U.S. fiction (see Klinkowitz). A genealogy of contemporary American fiction needs to see William Vollmann, David Foster Wallace, and Lynne Tillman not as self-conscious successors to Thomas Pynchon or Robert Coover but as differently inclined writers within a developing media ecology. Similarly, in poetry, the great long poems of John Ashbery, A. R. Ammons, Gary Snyder, and Stephanie Strickland become contributing channels toward a mediated ecological mainstream, whose similarities and divergences from this or that "school" or formalism must be seen as incidental.


1. There seems to be a tradition of ambiguously titled books in the cognitive field, my own included.

2. Bruce Clarke has clarified the nature of the systems/environment boundary in a set of papers distributed at the Fall 2003 Science and Literature Conference in Austin, Texas. In an email follow-up to those sessions, Clarke and I have addressed the question of whether a literary work can be regarded as a system. Clarke thinks works of literature are structures, not systems -- they do not operate, but rather they provoke operations in psychic and social systems. From this perspective, self-reflection in literature might be understood as a special kind of operation, one that does not change anything materially but which produces a defamiliarizing sense of familiar, worldly elements in the literary work.

Works Cited

Gumbrecht, Hans Ulrich. Production of Presence: What Meaning Cannot Convey. Stanford, California: Stanford University Press, 2004.

Johnston, John. Information Multiplicity: American Fiction in the Age of Media Saturation. Baltimore: Johns Hopkins University Press, 1998.

Klinkowitz, Jerome. Remarks on Cognitive Fictions in American Literary Scholarship: 2002. Durham: Duke University Press, 2004.

Moretti, Franco. Atlas of the European Novel: 1800-1900. London and New York: Verso, 1998.

—. “Graphs, Maps, & Trees.” New Left Review (November-December 2003). http://www.newleftreview.net/Issue24.asp?Article=05.

Spolsky, Ellen. Gaps in Nature: Literary Interpretation and the Modular Mind. Albany: SUNY Press, 1993.

Spolsky, Ellen, and Alan Richardson. The Work of Literature: Cognition, Culture, and Complexity. London: Palgrave, 2004.

Turner, Mark. Reading Minds: The Study of English in the Age of Cognitive Science. Princeton: Princeton University Press, 1991.

Joseph Tabbi


©1996-2007 RCCS         ONLINE SINCE: 1996         SITE LAST UPDATED: 12.10.2009