A week ago I published long guest post at Language Log, Two Disciplines in Search of Love. The two disciplines, of course, are literary criticism and computational linguistics. Two days later I posted a conclusion as a long comment. I've now taken those two pieces, edited them together, added a post script, and created a PDF which I've uploaded to my SSRN page: Two Disciplines in Search of Love.
I've put that postscript into this post, followed by the abstract and table of contents of the PDF.
A Personal Postscript
I believe that the work currently being done in natural language processing (NLP) in literary studies will, in time, redeem the work that I did back in the 1970s and 1980s. That work is predicated on the idea of computation and the expression of that idea in the early cognitive sciences and in computational linguistics, in which I was instructed by David Hays.
Literary studies missed the boat on the so-called cognitive revolution and, despite the emergence of cognitive criticism in the middle and late 1990s, literary studies still hasn’t gotten the message. Briefly and crudely, the cognitive revolution is what happened when the idea of computation hit the human sciences. By the mid-1980s, however, the first wave of cognitivism had died down and newer work arose in which one could easily miss the importance of computation. That work – I’m thinking particularly of cognitive metaphor and conceptual blending, theory of mind, and mirror neurons – is what captured the attention of literary cognitivists and allowed them to go cognitive while not having to come to grips with computation.
That blissful ignorance may persist for another decade or two, but not much longer, for the implications of literary work based on the techniques of NLP will require critics to think about computation, not simply as a tool for crunching large piles of data, but as a process in the minds of readers and writers. For the moment work in NLP is largely empirical in character: We’ve got these cool tools, let’s put them through their paces. But also: Yikes! we’ve got to get all these texts in usable digital form! That’s fine; it’s early days.
But it won’t do over the long haul. Sooner or later the question, just what ARE we doing? will become pressing. The program I outlined above (in the complete essay), large-scale and long-term investigation into romantic love, cannot be brought to fruition without thinking about the computational view of the mind. For, so far as I know, that is the only conceptual approach we’ve got that can support richly structured models capable of orchestrating our basic biological proclivities into a dance of desire and fulfillment that transformed an adulterous liaison into a prerequisite for marriage.
When those days come, people will have no choice but to think more deeply about the strange conceptual objects coughed up by these high tech picks and shovels. Investigators will find themselves inching inevitably toward a computational understanding of texts themselves, and of the minds of which those texts are evidence and from which they are products. When those days emerge investigators will find a rich and various literature awaiting their attention.
Though computational linguistics (CL) dates back to the first efforts in machine translation in the mid 1950s, it is only in the last decade or so that it has had a substantial impact on literary studies through the statistical techniques of corpus linguistics and data mining (know as natural language processing, NLP). In this essay I briefly review the history of computational linguistics from its early days involving symbolic computing to current developments in NLP and set that in relationship to academic literary study. In particular, I discuss the deeply problematic struggle that literary study has had with the question of evaluation: What makes good literature? I argue that literary studies should own up to this tension and recognize a distinction between ethical criticism, which is explicitly concerned with values, and naturalist criticism, which sidesteps questions of value in favor of understanding how literature works in the mind and in culture. I then argue that the primary relationship between CL and NLP and literary studies should be through naturalist criticism. I conclude by discussing the relative roles of CL and NLP in a large-scale and long-term investigation of romantic love.
Let’s Set Critique Aside
Oh, Those Meddlesome Value Judgments
CL and NLP in Naturalist Criticism
NLP as a Descriptive Tool
Romantic Love: Cultural Invention or Biological Universal?
Ethical Criticism and Naturalist Criticism (Slight Return)
A Personal Postscript
Appendix: Further Reading