Cognitivism Unbound: A Response to Meir Sternberg
Alan Richardson
Boston College
Introduction | Gerald Cupchik | Andrew Elfenbein | Art Graesser | Uri Margolin | | Meir Sternberg
Let me begin with a disclaimer. I am a Romanticist, not a narratologist or narrative poetician, nor am I deeply read in narrative theory, cognitive or otherwise. Nevertheless, when invited by Els Andringa to respond to Meir Sternberg's massive two-part critique of cognitive work on narrative and narrativity, I felt compelled to do so out of a sense of responsibility. The first part of Sternberg's article appeared, after all, in a special issue of Poetics Today, "The Cognitive Turn? A Debate on Interdisciplinarity," that was provoked by and, in many cases, directly responded to a previous special issue, "Literature and the Cognitive Revolution," edited by Francis Steen and myself. Reading over the response issue, not least Sternberg's contribution to it, I found myself seized more than once by ominous feelings of remorse and self-doubt well-known to readers of Mary Shelley's Frankenstein. In attempting to bring more widespread attention to and even help define a new subfield in the making, cognitive literary studies, had I inadvertently given the alarm to a growing host of critics now aiming to do all they could to strangle the infant field in its cradle? Would the literary theater of the cognitive revolution soon prove overwhelmed by the reactionary forces so rapidly assembling?
Looking more recently, and in a somewhat calmer mood, at Sternberg's article and its follow-up piece, I saw that my initial fears were overblown. In fact, I found much to agree with in Sternberg's lengthy critique, a serious and even rather monumental undertaking (lengthier, at 220 pp., than many a recent monograph) that one might in fact consider in the light of a disinterested service to cognitive literary studies. Sternberg's extensive research and pointed criticisms confirmed my own amateurish sense that much of the cognitive-oriented work done on narrative in the 70s and 80s lacked nuance and had too quickly adapted overly simplified computational processing models. I also had felt, and now had Sternberg's authority to back up my feeling, that such work was ill-equipped to handle modernist literary traditions, while ignoring important post-modern developments in theory. I was happy to concur with Sternberg regarding the need for cognitive theorists and experimentalists to keep affect centrally in sight (as do, for example, Miall and Kuiken in their work on reader response) and appreciated his citation of Ellen Spolsky's cognitive literary criticism as a corrective and challenge to some work done by cognitive narratologists. I found certain leading terms of Sternberg's critique quite helpful, especially his charges of isolationism (ignoring work in relevant neighboring fields, such as literary narratology) and amnesia (forgetting about a long tradition in narrative theorizing going back at least to the Greeks -- Patrick Hogan would want us to take early Sanskrit and Arabic theorists into cognizance as well). Sternberg's diagnosis of an implicit neoclassicism informing much of the work he addressed also scored points with me, and might well become a watchword in future assessments by literary scholars of cross-disciplinary work in narratology.
So perhaps I should simply thank Sternberg and sit back down. But I do have a few points of disagreement, not least concerning the rhetorical stance that Sternberg takes in his two-part article, one that I worry leads to a bleaker outlook than need be, particularly in tandem with his emphasis on earlier, largely outmoded work. I would begin with Sternberg's use of the term "cognitivist," which to me is by no means an innocent descriptive adjective but a loaded and potentially misleading term of invective. "Cognitivist" and "cognitivism," as Steen and I pointed out in the response of Poetics Today in answer to an earlier set of critics, imply a unity of aim and approach that has no real existence either in the nascent field of cognitive literary studies or in the related one of cognitive narratology. "As an '-ism,'" we wrote, "'cognitivism' seems to imply a shared worldview or ideology rather than an overlapping set of varied research interests and theoretical reference points. (Compare the difference between 'social theory' and 'socialist theory.') While the 'cognitive' in 'cognitive literary criticism' does meaningfully relate to the same term as it occurs in disciplinary compounds like 'cognitive psychology,' 'cognitive anthropology,' 'cognitive linguistics,' and 'cognitive neuroscience,' these rapidly developing fields manifest far too much diversity -- in theory, method, and sheer range of subjects of inquiry -- to yield anything like the common essence suggested by 'cognitivism.'" Similarly, Sternberg's use throughout his two-part essay of "cognitivist" and its variants gives a spurious sense of unified purpose and method that occludes the diverse critical interests and far-flung disciplinary homes of the writers and researchers whose work he amasses together for the purposes of critique. There simply is no such entity as "cognitivism" that could serve as a common denominator for work done by (to evoke some of the disciplinary categories fused by Sternberg) Artificial Intelligence researchers, philosophers of mind, cognitive psychologists, discourse processing analysts, cognitive linguists, social psychologists, and cognitive narratologists. Many of the theorists and researchers cited by Sternberg simply never encounter one another's work, which is not suprising, given that they belong to different disciplines, reside in different departments, submit their work to different review systems, read different journals, and go to different conferences. Even this conference, the IGEL, which might provide a common biennial meeting place, would fail to attract many of the figures listed in Sternberg's copious bibliography, particularly those who would not consider themselves working with empirical approaches (even if subscribing generally to an empirical or naturalistic world view).
It should come as no surprise, then, that (as Sternberg remarks) "Surprisingly for an enclave… the cognitivist one does not even boast anything like the solidarity of a unified, continuous research program" (314). Because, in point of fact, this "enclave" exists nowhere outside of Sternberg's critique. An enclave, the OED tells us, is a "portion of territory entirely surrounded by foreign dominions," as, say, the ethnic Armenian Nagorno-Karabakh region, separated from Armenia proper by the Karabakh mountains, can be considered a minority Armenian enclave within Azerbaijan. The motley collection of psychologists and literary theorists, computer programmers and philosophers, film critics and poeticians gathered together within Sternberg's capacious if sometimes bruising embrace constitute the opposite of an enclave. No unified or homogenous group surrounded by foreign ground, they might be rather considered as a collection of disciplinary nomads or interdisciplinary squatters, passing through the same territory on occasion, to be sure, but passing one another, as it were, often in the night. If I want a spatial metaphor, I think not of an enclave but of something like Rick's café in Casablanca, a place where stray representatives of disparate populations may mingle or not, where alliances are quickly formed and readily abandoned, where acquaintances are made only to be dropped or forgotten, where the emergence of a long term association might be considered more the exception than the rule.
This is understandable. Cognitive psychologists have very good professional reasons for caring more what other cognitive psychologists think of their work than what Sternberg or I or most anyone else working out of a literature department might think. I, for my part, can try out a cognitive critical take on, say, the narrative representation of mind in an Austen novel without much concern with how it might appear to a discourse analyst, since such a reader is far less likely to encounter my essay than is either an Austen critic or a fellow cognitive literary critic with like interests such as Uri Margolin or Lisa Zunshine. There are no pressing professional reasons for me to read widely in discourse analysis, any more than for a cognitive psychologist to read widely in Jane Austen criticism, though surely Austen's best critics have had much to say of interest regarding representations of the mind.
Sternberg would and should reply, I think, first: that intellectual interests should not be subordinated to professional ones, and second: that anyone engaged seriously in work between disciplines has a duty to read in the relevant fields and to avoid isolationism. And I agree. But human nature being what it is, I doubt that intellectual and moral imperatives will do all that much in the absence of institutional aid and support. If cognitive-oriented work on narrative is to become more like the common pursuit of an enclave than the myriad tangential schemes taking place among a den of exiles, we could all do with more resources -- in print, online, in person at conferences such as this one -- for discovering, keeping track of, and mutually criticizing one another's work. Sternberg's critical overview of so much relevant work provides us with one inspiring example, one I hope to see followed soon by a more optimistic assessment, featuring the most exciting and promising work currently being done in this area across our disciplinary divides. Should anyone at present be working on such an overview, please send me a preprint as soon as it's ready to come out.
