Monday, August 31, 2015

On the Poverty of Literary Cognitivism 5: It’s Time to Leave the Sandbox

Note, Sept 1, 2015: I've added three paragraphs about theory to a section, Characteristics of the Critical Sandbox.
* * * * * 
The child is father to the man.

– William Wordsworth

Despite the wide range of literary study that has taken place under the rubric of the newer psychologies – cognitive science, evolutionary psychology, and neuroscience – these initiatives remain impoverished by their instance on functioning as a form of hermeneutic literary study. Since World War II academic literary criticism has used literature as a vehicle for the investigation of the human condition as seen across half the departments of the modern university [1]. All the cognitivists and evolutionists are doing is inviting more disciplines to the party. But it really is the same old party, not the new ones the evolutionists, on the one hand, and the cognitivists, on the other, are proclaiming.

Text as Sandbox

After World War II all academic literary criticism, at least in the American academy, came to agree on one basic modus operandi: The literary text is a mental sandbox in which the critic plays with his or her favorite conceptual toys. The New Critics used the idea of form to isolate the text from external influences, authorial intention, reader affect, and history. Because their conceptual toys – the ideas of ambiguity and paradox coupled with humanism (secular or Christian) – were invisible as such, they could present this as an act of austere purification in which messy value judgments and textual entanglements could be left behind.

Once the New Critics had thus transformed the text into an intellectual testing ground, a sandbox, other critics brought in more obtrusive toys: psychoanalysis, phenomenology, Marxism, structuralism, deconstruction, feminism and the other isms, and now cognitive science and evolutionary psychology. The new historicists have been fighting a highly successful rear-guard action by bracketing all theory toys and privileging a somewhat different set of toys, non-literary texts roughly contemporaneous with the texts under examination. At various points along the way the new set of toys would be brought online through a rhetoric of revolution and renewal, but the basic procedure remained the same: declare the sandbox to be liberated, toss out the old toys and bring in the new. Same sandbox, new toys.

This modus operandi was underwritten by a tacit agreement that any reading of a text was legitimate as long as it was supported by a suitable rationale, of which there are now many kinds. This professional courtesy inevitably led not merely to a multiplicity of readings, but to divergent and contradictory readings. Some critics saw this as evidence of the richness of canonical literary texts while others saw it as evidence of epistemological and methodological inadequacy. The latter have argued their case from time to time, but have yet been able to inspire a discipline-wide austerity program that has narrowed the range of legitimate readings to one per text. They’ve had little choice but to agree to the same old unwritten tacit agreement: Let 10,000 flowers bloom.

Thus much of the practical criticism produced by these most recent revisionists reads like 1950s humanist criticism but with a different set of tropes, motifs, and themes. Thus, in a review of an anthology of Darwinist literary criticism [2] Steven Pinker offered this observation about Joseph Carroll’s analysis of Pride and Prejudice [pp. 166-167]:
Carroll dissects the novel with skill and verve, and will make many readers wish that they had had him as their college English prof. Nonetheless, one is left wondering how essential the evolutionary biology is to his insights. The mating criteria that obsess the Bennett women may reflect universal impulses, but the specifics of the novel depends on the way that these impulses were exaggerated and codified in their time and culture. Today, a depiction of a contemporary middle-class family that worried aloud about finding wealthy husbands for the daughters, and about their being disgraced by a daughter running off with the son of a steward, would elicit guffaws, not a flash of recognition. In Pride and Prejudice, to be sure, these worries are set in tension with other concerns, but a skeptic could say that the tension is between individual and cultural demands, not individual and evolutionary ones.
That certainly accords with my own impression of Carroll’s work, and that of others as well, such as Brian Boyd’s treat of Iliad and Horton Hears a Who in The Origin of Stories [3].

Later on Pinker remarks [p. 175]:
It’s conceivable that evolutionary thinking will raise, and eventually solve, the scientific question of why we enjoy fiction without offering anything to the field of literary criticism beyond our folk theories of human nature. (Under this scenario, evolution would explain why those folk theories are true, but would not add anything to the power of those theories to illuminate specific works of literature.) I don’t think it will come down to this, but advocates of Darwinian lit-crit should be prepared to spell out what they hope evolution will add to literary analysis beyond a rehabilitation of the relevance of a conception of human nature.
It’s that first sentence that caught my attention, for THAT is how Darwinian criticism reads. The descriptions it provides of characters’ actions, desires, and motives is much like that of a common sense understanding of human actions except that the categories of folk psychology are replaced by those of evolutionary psychology.

On the cognitive side, we have these remarks that Tony Jackson made about Mark Turner’s conceptual blending back in 2003 [4, p. 164]:
[…] it must also be said that the idea is equally indisputable without cognitive science to make the case. As far as I can tell, if we set out to unpack any metaphor or other linguistic figure and we consider the relevant elements of the (at a minimum) two items brought together to render a third meaning, we can use much the same method Turner does without knowing anything about cognitive science.
Later, in his discussion of Francis Steen in Aphra Behn’s Love Letters between a Nobleman and His Sister Jackson notes, “Too often it seems that the vocabulary of cognitive rhetoric is simply being plugged into the interpretation” [p. 173].

Since I’ve certainly not read all the specific interpretations produced in these newer schools, I won’t assert that they’re ALL like that, but enough of them are so as to be less than encouraging. Moreover, I find it telling that, after two decades or so of literary work employing these ideas and with thousands of MLA members interested in this work, that Richardson [5] can still remark (p. 367): “Cognitive literary approaches, however, stand to greatly widen their appeal if they can continue to demonstrate their potential for producing original readings, as well as addressing issues in theory and poetics […].” The emphasis has been on theoretical matters, not on practical criticism.

Discursive Thought Remains Central

One can see why that is the case. When, for example, literary critics went to psychoanalysis for ideas they found a discipline that dealt in dreams, fantasies, and stories, the stuff of literature. These newer psychologies deal with a lot of things, but not so much with dreams, fantasies, and stories – and where they do trade in stories, the stories are too simple to yield insight into literary texts. So literary critics have been busy gathering psychological ‘components’ into assemblages usable for thinking about literary phenomena.

Theory-building has thus far been the central activity, and accomplishment, of these literary investigators. Their theories, like most work in the humanities, are prose-centric or discursive in the sense that their central thinking takes place in the prose on the page. But the cognitive sciences have other ways of thinking, ways in which the central thinking takes place in some other medium and the prose exists to support that non-discursive thought. Some of this thinking takes the form of laboratory observations and the subsequent statistical analysis of data. Other thinking takes place in the proof structures and equations of mathematical modeling. And of course we have thinking that is embedded in computer simulations.

My early work in the cognitive sciences was of the latter kind. As I’ve already indicated in the second post in this series, What I Learned When I Walked off the Cliff of Cognitivism, I did that work by drawing diagrams, hundreds of them. You draw the diagrams and then you write the prose needed to explain them. (Some of my colleagues, of course, wrote computer programs realizing the structures in the diagrams they had drawn. Then they wrote the prose, and it explained the programs as well.)

And, unfortunately, if you’ve not had experience in actively working with such models, reading and even studying the proces, and the diagrams, doesn’t really get you into the “thought space” of the model. I’d seen and studied many such diagrams before I met David Hays. I knew that I had to learn how to do that. When I learned that Hays could teach me, that’s when I decided to apprentice myself to him. [6]

That’s how I got out of the interpretive sandbox. My impasse with “Kubla Khan”, which led me to a tree describing on the poem’s form, told me I had to get out of it. But it wasn’t until I’d learned how to think the mind in diagrams that I managed to crawl out.

The cognitive and evolutionary critics are still in the sandbox. They may read and study the work of investigators who think in experiments, statistical analysis, mathematical models, or computer simulations, but they do not, in a significant degree, follow them there.

Characteristics of the Critical Sandbox

This critical sandbox has a number of characteristics. Though each deserves extended discussion of its own, I’m simply going to list them here and provide short characterizations.

1) Reading: The distinction between ordinary reading, which everyone does, and interpretive reading, the province of literary critics, is elided. Critics “read” texts and so create “readings”. While the interpretive reading of religious texts is an ancient activity, the routine interpretive reading of literary texts is, as I’ve said before, relatively new. This usage has become so ingrained that critics will sometimes talk as though interpretive reading were the only real reading, leading you to wonder how literary culture ever managed to survive without interpretive critics. [7]

In the conceptual world I’ve been exploring since the 1970s, “reading” is restored to its ordinary meaning and is to be investigated through psychological experimentation and computer simulation. Those diagrams I produced for Shakespeare’s Sonnet 129 would be components in a computer simulation of just plain old reading the text. That the current art won’t support such a simulation is unfortunate, but those diagrams can serve to recognize and mark the principle, which has worked with much simpler texts.

2) The Text: The distinction between the text as physical object (marks on pages, pages bound into books) and whatever it means and whatever it represents is elided. While there are times and places when critics mean only physical marks or sounds, the signifiers, these tend to be in discussions of theory and/or method. In practical criticism the text is always something more than, in excess of, the marks on the page. The job of the critic, then, is to explicate that excess in by providing a reading. [8]

If you have an explicit computational model of language processing, that model can deal with the string of signifiers as a physical object. The model itself embodies (much of) the excess ordinarily in play when critics talk of “the text”. I note that the newly emerging practice of digital criticism works by finding patterns in collections of texts where the texts really are nothing but signifiers because that’s all the computers can deal with. But, of course, these digital humanists proceed under the assumption that those signifiers have been produced by human minds. The objective, of course, is to learn about those minds by examining the patterns discovered in a corpus of texts.

3) “Form” becomes either a synonym for genre – tragedies and sonnets are forms – or a philosophical declaration of textual autonomy. The purpose of that declaration is to enable a critical practice that focuses exclusively on “the text” as its object so the critic can then “read” it. Thus “close reading” does not involve sustained attention to a text’s form. Formal features may well be noticed as they contribute to the critic’s interpretive strategy, but form is otherwise a secondary matter. Thus, in a recent article on form, Sandra Macpherson notes that, while looking for form, critics still are unable to focus on the physical shape of the text [10].

The need for formal analysis is, of course, one of my main hobby horses, and I rode it in the third post in this series, The key to the treasure IS the (form of the) treasure [11]. My focus on form is, of course, a complement to my interest in explicit computational models. A computational model simulates the operations of the literary mind as it reads (1) the text (2). Literary form is the trace of that process; it is a clue to the working of the literary mind.

4) Theory: Over time the theory of literature morphed into critical theory which in turn became Theory, through the capitalization of the initial “t” is optional. During the 1950 and well into the 1960s theory of literature was just that, theory about literature. Theory of Literature (1948) by Rene Wellek and Austin Warren was the touchstone text. In the later 1960 and into the 1970s critical theory came to the for; it was about, not literature, but about literary criticism. Perhaps the central issue was whether texts had only one legitimate meaning or several.

In the 1980s critical theory collapsed into theory, and then Theory. Critical practice was no longer an object of inquiry. Rather theory/Theory had become a menu of interpretive schemes to be applied to texts – Marxist, psychoanalytic, deconstructive, feminist, and so forth. And criticism merges with literature as critical reading, that is explicit written interpretation, becomes the only true reading. The distinction between literature and criticism is all but gone.

In this respect cognitive and evolutionary criticism both restore the distinction between literature and talk about literature, that is, theory. And much of the work in these disciplines is literary theory, rather than offering full explications of individual texts. This gives some cause to believe that these thinkers may see the need to leave the sandbox.

* * * * *

There is one final property of the critical sandbox: It denies criticism, as it is ordinarily understood. Academic literary critics do not pass aesthetic judgments on literary texts nor do they explore the ethical implications of texts. At least that’s the more or less official ideology of literary criticism. In practice both things are done.

Following Wayne Booth [12], I propose that both ethics and aesthetics be explicitly acknowledged and that they be recognized as a different, but equally important, branch of literary investigation. The naturalist critic investigates how we have used texts to organize our minds, societies, and lives in the past while the ethical critic uses texts, of whatever era, to explore as yet unrealized possibilities for the future. The naturalist critic will show how, in the past, poets have been unacknowledged legislators of the world while the ethnical critic will explore what and how to implement poetic legislation in the future.


[1] J. Hillis Miller has made some interesting remarks about the profession as it existed in the 1950s and as interpretive criticism emerged. I have collected some of them, with references to the original documents, in a working paper, J. Hillis Miller on the Profession: “Literature is … made out of words”. Working Paper (2015) 13 pp. Download:

[2] Pinker, Steven. Toward a Consilient Study of Literature. Philosophy and Literature. 2007, 31:161-177.

[3] My review: Nature Culture Tweedledum Tweedledee. Blog post, New Savanna, May 23, 2011, URL:

You should also consult Michael Bérube’s review: The Play’s the Thing. American Scientist. January-February 2010. URL:

[4] Jackson, Tony E. Issues and Problems in the Blending of Cognitive Science, Evolutionary Psychology, and Literary Study. Poetics Today. 2002, 23,1: 161-179.

[5] Alan Richardson, Once upon a Mind: Literary and Narrative Studies in the Age of Cognitive Science, MFS Modern Fiction Studies, Volume 61, Number 2, Summer 2015, pp. 359-369, 2015.

[6] For what it is worth, I’d been thinking in diagrams since childhood. I designed layouts for my model trains and I spent a great deal of time ‘designing’ rockets and space craft and still have some of those designs.

[7] I’ve written a number of posts about this. You can find them collected under the tag “reading>reading”, URL:

[8] See, for example, Rita Copeland and Frances Ferguson. Introduction. ELH. Volume 81, Number 2, Summer 2014, pp. 417-422. This introduces a suite of articles: Essays from the English Institute 2012: Text.

See also my blog post, Texts, Traces, and Hyperobjects. March 19, 2011. URL:

[9] See my post, Signifiers, the Material Text, and the Digital Critic, August 30, 2015, URL:

[10] Sandra Macpherson, A Little Formalism, ELH, Volume 82, Number 2, pp. 385-405, Summer 2015.

[11] The key to the treasure IS the (form of the) treasure. August 26, 2015, URL:

[12] Booth, Wayne. The Company We Keep: An Ethics of Fiction. University of Chicago Press, 1988.


  1. Perhaps ridiculous to ask: do critics realize the experience of the poem? or is the investigation the critic's way of experiencing the poem? (Keeping in mind, for what it's worth, that rich poems will be experienced differently when read on the page as when heard performed aloud.)

  2. " critics realize the experience of the poem?" There's no way to tell from the criticism, but I expect that many do.