Another world

« previous post | next post »

I'm nearly sure that the people who submit comments on this and other weblogs are all intelligent, knowledgeable and thoughtful. Often, this is plain in what they write. But sometimes, in order to maintain my commitment to this belief, I'm driven to conclude that a commenter has recently tunneled through the barrier separating us from some parallel — and fairly distant — universe. Yesterday, one such transdimensional pilgrim contributed this observation:

Transformational grammar hasn't been accepted as plausible for decades, except, for some reason, in English departments.

Now, I know several of the inhabitants of several English departments in American colleges and universities. And if these organizations are full of partisans of transformational grammar, they're keeping a very low profile. In fact, in my (admittedly limited) experience, English departments are among the last places on campus where you're likely to find any indication of interest in any form of linguistic analysis whatever.

I'd be very happy to find that I'm wrong about this.

But just to show that what I'm saying is consistent with the sorts of evidence available to me, at least in this particular space-time bubble, I'll supplement my earlier examination of the English curriculum at Stanford with a similar inspection of English at Yale. The department's front page offer this account of their interdisciplinary connections:

Our work is enhanced by connections to many other disciplines, including history, art history, political philosophy and theory, and literature in other languages, as well as the contemporary interdisciplines such as gender studies and cultural studies.

Linguistics, you'll notice, is not mentioned. And turning more specifically to the hypothesis that the Yale English Department might be a hotbed of die-hard transformational grammarians, I'll note only that the word "transformational" does not occur within the domain, and the word "chomsky" is likewise absent.

This does not mean that the Yale English department is atypically anti-Chomsky, but rather that it is (typically) uninterested in linguistic analysis of any type at all — also among the missing are phoneme, vowel, consonant, Lakoff, Whorf, "noun phrase", transitive, adverb, iamb, trochee, dactyl, pentameter, hexameter, and metrical.

(To make it clear that Google has actually indexed some text there, post-colonial occurs 22 times.)

As I expected, no form of linguistic analysis is mentioned in the requirements for the undergraduate major, nor in the graduate program requirements.

The word "grammar" does occur five times. One is in the CV of a writing instructor, who (between 2000 and 2004) taught a course called “Expository Writing and Grammar” in the English as a Second Language program. The second is in a guide to the form of essays, which tells us that

By American convention, periods and commas required by the grammar of your sentence precede the close-quotation marks, even when they are not found in the original.

The third instance is in the description of a Spring 2008 course called "Writing and the Environment", which mentions that

In this writing course, the written essays include first and final versions that receive teacher and peer evaluations with particular attention to ideas, argument, grammar, and style.

The fourth instance is another course title in an instructor's CV, this one in a different department: "Greek 111 Review of Grammar and Selected Readings (Plato, Xenophon)".

And the fifth mention of grammar is in a document called "Survival Kit for English Majors", which no longer exists on the departmental website. But according to Google's cached "view as html" version, this warned majors that "literature courses in other language departments" count towards the English department's course requirement, but

This category does not include grammar, composition, or conversation courses.

If we have any other readers from one of the universes in which English departments are full of transformational grammarians, I'd love to hear some stories about what it's like there. And also, please tell the rest of us how you do that tunneling trick, so we can pay a visit for ourselves.


  1. mike said,

    June 17, 2008 @ 1:52 pm

    Could you perhaps address a slightly different aspect of the question and explain to those of us in, um, not-English departments the "not considered plausible" nature of TG? Has TG gone the way of the geocentric universe? Are there still TG-ians, and are they the flat-earthers of linguitics? And finally, of course, what IS considered plausible these days?

  2. JakeT said,

    June 17, 2008 @ 2:18 pm

    But at least we CAN comment now.

  3. Adrian Bailey said,

    June 17, 2008 @ 2:19 pm

    After a hard day, this post gave me a wide smile. :) Thanks.

  4. Melanie said,

    June 17, 2008 @ 2:20 pm

    At least in some schools in Japan, Chomskyan transformational grammar is only accepted as plausible in English departments, where it's considered to only apply to the English language and not to linguistics more generally.

  5. Coppe said,

    June 17, 2008 @ 2:22 pm

    Transformational Grammar assumes that a sentence is represented at two levels, Deep Structure and Sentence Structure. At Deep Structure, a sort of basic word order is maintained, which is the same across sentence types (whether they end up active or passive or questions, etc.). Different transformations on the Deep Structure order yield different Sentence Structures. So, moving an object to the front of the sentence yields a passive (i.e. "pizza was eaten by me"), where leaving it in-situ yields an active sentence (i.e. "I eat pizza"). There's a little more to it than that, but that's the basic idea.

    The whole idea of Deep Structure and Sentence Structure is now outdated, but other parts of Transformational Grammar aren't. The ideas that language is innate and modular and the concept of x' bar theory (language structure can be represented as binary trees), for instance, are still core elements of Chomskian linguistics.

    What is considered plausible now is the Minimalist Program, a research program kickstarted by Chomsky in 1995 which encompasses a wide range of Chomskian theories that share a number of very interesting assumptions (which are too complex to go into here).

  6. Nassira Nicola said,

    June 17, 2008 @ 3:12 pm

    To be fair, some universities without linguistics departments nonetheless offer linguistics instruction in their English departments. I'm thinking, for example, of Northern Illinois University (, which employs Real Live Linguists Betty Birner and Gülşat Aygen. Gülşat, in particular, is a genuine syntactician in good standing, and though I'm not entirely sure she'd describe herself as a Transformational Grammarian per se (i.e. she doesn't write transformations), she certainly does teach out of Radford's red TG book. (See, for example, the course description for ENGL 518.1, "Morphology and Syntax":

    So while I grant you that to those of us in universes that overlap with Yale and Stanford – and U Chicago, and U Penn, and all those other places with standalone departments for the study of Language-with-a-big-L – the idea of linguists in an English department might seem bizarre, there is indeed a whole universe of linguists out there for whom that's just the way of it. And the tunneling process is as simple as clicking.

  7. Nassira Nicola said,

    June 17, 2008 @ 3:18 pm

    (Well, that's very odd. One URL got hotlinked, the other didn't. I thought I knew enough about the new platform not to put in an HTML link, but I suppose it's outwitted me again.)

  8. James Lyle said,

    June 17, 2008 @ 4:38 pm

    Indeed, both my undergraduate institutions were places where the only linguistics available was in English or other language departments. It seemed perfectly natural that for those benighted parts of the world (well, the U.S. anyway) where we didn't have linguistics departments, the English department was its natural home. I recall that when I was job-hunting in the late 90s, a lot of openings for linguists were in English departments, and it seemed that chomskyan linguists were generally as welcome there as anybody.

  9. pc said,

    June 17, 2008 @ 4:42 pm

    I second Nassira's point, though I'm kind of surprised the point even needs to be made, given that many (most?) smaller universities and certainly most colleges don't have standalone linguistics departments (yet). While it's probably a fair point that there aren't a lot of people doing "transformational grammar" in non-linguistics departments in general and English depts in particular, it seems unfair to claim that people in English depts in particular show no "interest in any form of linguistic analysis whatever." It makes me wonder if by "linguistic analysis of any type at all" you have actually a fairly narrow view of what that might comprise. Just in English departments in the state of Michigan, UMich has Anne Curzan and Dick Bailey; MSU has Dennis Preston and Geneva Smitherman; Wayne State has Ellen Barton, Walter Edwards, Geoff Nathan, Martha Ratliff, and Ljiljana Progovac; EMU has Anthony Aristar, Helen Aristar-Dry, and Dan Seely. While most of these scholars aren't doing syntax/"transformational grammar" (though there are several exceptions, notably Progovac and Seely), they are *certainly* doing linguistic analysis. Though I suppose there are people who could disagree with me about that.

    But this is really a discussion about how scholars are organized within the academy, not whether English professors are (or ought to be) interested in linguistics; people in English departments who are keenly interested in linguistics are probably going to be linguists by identification, who happen to teach in English departments. At any rate, depending on someone's experience with universities, there might be good reasons for them to believe that lots of linguists/syntacticians/TGians/Chomskyans are hanging out in non-linguistics departments (with English being, btw, the most immediately sensible place for them probably to be, at least in the US).

  10. john riemann soong said,

    June 17, 2008 @ 4:54 pm

    On the subject linguistics and teachers with low profiles, just this year I had a high school AP English teacher who *gasp* had familiarity with linguistics and was at least well-read enough on transformational grammar to hold a conversation, and had in fact, received a fair amount of linguistic training.

    Why didn't he talk about it before? Well it wasn't part of the College Board's AP syllabus, and as far as the marginal benefit of teaching linguistics (per marginal cost in time) was concerned, it was lower than many other things he could teach.

    So my high school's English Department does have a transformational grammarian, or at least a partial one. Nevertheless, linguistic terms refuse to show up in his materials, except where rhyme and meter is concerned.

  11. kay said,

    June 17, 2008 @ 5:04 pm

    Geoffrey Russom of the English Department at Brown fits the bill exactly – a Chomskyite whose classes cross over into linguistics. I'm not sure where he stands on transformations, however. The course description of his upcoming "Narrative Poetics" is as follows: "A study of literary language from the writer's perspective that draws on Chomsky's universalist linguistics to refashion structuralist poetics and narratology. Small written experiments with literary word choice and literary sentence structure prepare for a final experiment with narrative structure."

  12. Adrian Ryan said,

    June 17, 2008 @ 5:42 pm

    Being a Linguistics major at Yale University, I can very well tell you that the only departments other than my own at which one might find mention of any transformational grammar are the Psychology and Cognitive Science departments. I had always assumed that it was the Psych departments at smaller schools (and hey, Yale only has 5500 undergrads, not exactly a large school) that took care of the scientific approach to studying languages, and left the English and Lit departments to, well, literature.

    Hm, one last almost-related interesting comment: while studying in France, both my grammar and phonetics instructors were linguists, who taught practical French from a theoretical standpoint. I'm not sure it helped so much from my grammar teacher, as he was supposed to be instructing us on the finer points on writing at the French university level, and not explaining garden path sentences and transformational grammar in French, but the phonetics helped a lot in being able to better speak the language.

  13. lynneguist said,

    June 17, 2008 @ 5:47 pm

    I interpreted the quotation about Transformational Grammar/English departments not to refer to people actually doing TG in English departments, but that literary folk frequently have some odd and outdated ideas about linguistics. I've met folks who know all they know about linguistics from Neal Stephenson's novel _Snow Crash_ and its ideas about Deep Structure. I also have met literary people who seem to believe that Linguistics said all it had to say when Saussure's students published his lectures.

  14. Mark Liberman said,

    June 17, 2008 @ 6:14 pm

    Nassira Nicola: … some universities without linguistics departments nonetheless offer linguistics instruction in their English departments.

    Being serious for a moment, I'll confess that I know of quite a few linguists with positions in English departments, including some former students. And I'm sure that there are many more that I don't know about. But my point was a rather different one.

    The U.S. Department of Education

    …[makes] available a searchable database of postsecondary institutions and programs accredited by accrediting agencies or state approval agencies recognized by the U.S. Secretary of Education. This database includes approximately 6,900 postsecondary educational institutions and programs.

    I'm not sure how many of the 6,900 accredited postsecondary educational institutions and programs have English departments, but I imagine that half or more of them do. The LSA's Directory of Linguistics Programs in the U.S. and Canada lists 15 linguistics programs that are housed in departments with "English" in their name. What percentage of the remaining several thousand English departments have one or more faculty members with a significant background in any sort of linguistic analysis? I don't know, but I'd be pleasantly surprised if it's much more than one or two percent. Certainly in this universe, these thousands of English departments are not hotbeds of transformational grammar, or any other sort of grammatical analysis.

    A more serious and meaningful question would be how many of these English departments require their students to acquire the basic concepts and skills of linguistic analysis — say at the level of IPA transcription, simple syntactic description, scansion of metered verse, and basic concepts of semantics and pragmatics. The really interesting parallel universe would be one in a requirement of this kind was the norm.

    The undergraduate major at NIU, which is about as linguistics-friendly as English departments get, does require students to take two out of the following seven options:

    # ENGL 300A, ENGL 300B, or ENGL 300C, Advanced Essay Composition (3)
    # ENGL 318, Language and Linguistics (3)
    # ENGL 320, History of the English Language (3)
    # ENGL 321, Structure of Modern English (3)
    # ENGL 432, Topics in General Linguistics (3)
    # ENGL 433, Discourse Analysis (3)
    # ENGL 434X, Women, Men, and Language (3)

    But this is unusual, I believe. The English programs that I pointed to at Stanford and Yale, which are entirely empty of any required instruction in the analysis of the English language, seem to be much more typical, and not only at elite schools.

  15. John Lawler said,

    June 17, 2008 @ 6:24 pm

    In the universe of origin, "Transformational grammar" appears to mean something, whereas in ours, it's merely an indication that the speaker is not au courant — kind of like mentioning CP/M or Cobol, or (in a slightly more intellectual vein) Logical Positivism or Skinner boxes. Unless it's being used as shorthand for "Chomsky, and all his works, and all his pomps".

    In the US, there are schools with linguistics departments and schools without (by far the majority). In those without, linguists may have wound up in some other departments (English, languages, and anthropology are the most common) by the vagaries of history, and they may or may not have learned their trade when parts of it were called "Transformational Grammar", though that doesn't guarantee they ever found it, or Chomsky, plausible.

    Those in linguistics departments may or may not find Minimalism or GB or any other MIT-flavored syntactic variety plausible, either. They probably won't call any of them "Transformational", though.

  16. Jean-Sébastien Girard said,

    June 17, 2008 @ 6:45 pm

    Université laval is more explicit in grouping English and linguistics together. In the "Études anglaises" programmes (which is roughly equivalent to an English BA, AFAICT), there are two explicit linguistics class and two Grammar class (Can be characterized as advanced ESL classes, but elements of linguistics discussion in the one I took), and one of the concentrations (No idea what that is in more usual systems) is straight out linguistics.

    This is in addition to an explicit Language Science Degree (which was not available not so long ago; no idea why).

  17. Andrew Goldstone said,

    June 17, 2008 @ 7:00 pm

    Well, as a member of Yale's English department (grad student) who's had a lifelong interest in linguistics (and a multiyear interest in this blog!), I'd say I have to respond here. Prof. Liberman, your general impression is not totally wrong, but it is a distortion. First of all, the subfield of medieval literature has not lost touch with its philological roots; all of the medievalists I know have some linguistic training, especially in historical linguistics (esp. Indo-European, of course). You searched in vain for metrical terms on the department website, but metrics, of the old-fashioned kind, is frequently taught here in courses on poetry. (I am very grateful to you, by the way, for your posts of long ago on metrics; I used them in preparing to teach meter to my discussion-section students.) There is normally an undergraduate "History of the English Language" course, typically taught by a medievalist, and graduate students have the option of taking part of their general exams on the history of the language (as I did).

    But outside medieval studies, and in the field of English at large, the lack of interest in and the level of ignorance of linguistics is completely stunning. This post's chuckling over Chomsky is particularly apposite, since syntax is probably the weakest area of all; even in a department famed for teaching and practicing "close reading," nearly everyone, faculty and students alike, continues to use a grammatical terminology cobbled out of elementary-school memories and bits transposed from French, Spanish, or Latin. Though my colleagues have perhaps heard of Chomsky, they certainly couldn't analyze a sentence in any modern framework, X-bar, Minimalist, CGEL, or otherwise. (I brandish Huddleston and Pullum whenever I can; it hasn't made me any friends yet, but I'm still hoping. Forwarding "Prescriptivist Poppycock" posts to exponents of said poppycock has also not met with the acclaim I'd hoped for.)

    At the root of the problem, I suspect, is the near-universal antipathy of people in my field to any suggestion that they need or ought to develop competence in anything scientific. Far too many literary scholars think of the work they do as in conflict with, and superior to, all forms of science. Humanistic judgment and wisdom is held to supersede any need for technical competence, even when it comes to the very material out of which everything we study is made–language. Needless to say I would much rather we all became as competent as we could in all relevant fields.

    But of course the dialogue has to go in both directions, too. Even this wonderful blog has occasionally been guilty of taking gratuitous potshots at the fuzziness of humanists. Typically this takes the form of supposing that we in the humanities are all wholly in the grip of French post-structuralism. That this is also the kind of charge regularly levelled at us by The New York Times might give pause to the writers of this blog, who do such a terrific job of exposing that paper's linguistic incompetence. If anything popular culture has an even worse understanding of what contemporary research in English literature looks like than it does of linguistics. In fact the idea of "theory"-dominated English departments is at least twenty years out of date. If certain ways of writing and thinking distilled from Derrida and friends are still occasionally persuasive in my field, it's because they actually are compelling, in their distilled versions, to reasonable people. My field has done a pretty poor job of explaining why this is so to general audiences (with the occasional exception of…Stanley FIsh's New York Times blog?), but it's really a minor point. The dominant methods in our field now owe are "historicist," and most research aims at contributing to cultural history in a way that ought to be intelligible to anyone who thinks historical research is worthwhile. Slowly but surely, an empirical turn in our methods has been taking place which can only lead to a rapprochement with the social sciences, linguistics included.

    The point is this: linguists, don't give up on us here in the Yale English department or in the field of literature at large. The time is ripe for a real scholarly dialogue between these fields to begin again.

    P.S. Thank you for all you've taught me–and the rest of your readers–through this blog! You are a force for good.

  18. john riemann soong said,

    June 17, 2008 @ 7:00 pm

    Lawler: I'm a bit confused, because I thought minimalism, X-bar theory, etc. fell under transformative grammar.

    For example, when I access the Wikipedia article about the Minimalist Program, it is categorised under the category "Syntactic transformation," of which the main article is "transformational grammar".

  19. James A. Crippen said,

    June 17, 2008 @ 7:08 pm

    I think John Lawler is right here. The majority of English profs working in the US probably were required to take some sort of linguistics course during their graduate studies, and the course they took probably described transformational grammar, which was not terribly out of date at the time. Today they equate all of linguistics with “transformational grammar”, by which they mean everything that Chomsky and his followers have done since the 1950s. I doubt if most of them have any idea that what they call transformational grammar might at the very least be better called “generative grammar”, and that to linguists mentioning TG is like talking about ALGOL 60 (COBOL was never serious), continental drift, Lamarckism, or the Rutherford model of atoms.

    Indeed, my own school’s English department offers undergraduates a “Modern English Grammar” course which teaches a sort of rudimentary, debased transformational grammar for syntax analysis. This is largely because members of the Linguistics department have been unwilling to teach the class for many years (“they should just take our intro course instead”). At least they don’t use the crufty old Reed-Kellogg system of sentence diagramming.

  20. Shigekuni said,

    June 17, 2008 @ 8:08 pm

    Very strange

    "transitive, adverb, iamb, trochee, dactyl, pentameter, hexameter, and metrical."

    …these SHOULD get mentioned
    that they are not is a sad, sad comment on the state of English departments.

    a student of Literature and Linguistics at a German university

  21. Nathan Myers said,

    June 17, 2008 @ 8:09 pm

    Wait, what happened to continental drift? And, seriously, what is it that is wrong with sentence diagramming?

  22. Mark Liberman said,

    June 17, 2008 @ 9:02 pm

    Nathan Myers: …seriously, what is it that is wrong with sentence diagramming?

    Nothing, really.

    Or rather, it's a notation along with a set of cultural practices for using it. The notation was never formalized, as far as I know, but I expect that it could be. As a way of encoding basic syntactic descriptions, such diagrams are probably as good as dependency graphs or immediate-constituent trees, though I don't know enough about the Reed-Kellogg system to be sure of this.

  23. John Lawler said,

    June 18, 2008 @ 12:43 am

    @john riemann soong:

    Oh, my, no. Transformations are passé. There's been a new! improved! version every half-decade or so, you see, always with new names, new abbreviations, and new (and incompatible) rules and assumptions. Sort of like Windows.

    And for gods' sake, don't believe anything in Wikipedia about linguistics. Bits of it might be OK, but most of it is just hopeless.

  24. john riemann soong said,

    June 18, 2008 @ 1:49 am

    How updated are linguists not involved in investigating syntax on the goings on of each new technique?

    Most of what I know about linguistics has been self-taught, with a degree of error as you might imagine. While syntax was interesting for a while, hardcore syntax hasn't appealed to me. I suppose transformational grammar is the quintessential "hardcore syntax" topic, never mind if it's been outdated for decades. So my impression of it is quite similar to the public perception.

    Yet would a linguist — or a linguistics student, even — more interested in phonetics, sound and language change or the sociolinguistics side of linguistics necessarily be up to date about most major developments in syntax? Do they just come by in conversation?

  25. john riemann soong said,

    June 18, 2008 @ 1:52 am

    "And for gods' sake, don't believe anything in Wikipedia about linguistics. Bits of it might be OK, but most of it is just hopeless."

    Hmm — what is the problem you currently see? Are there currently too few linguists actively editing that the "with sufficient eyeballs, all errors are shallow" principle doesn't apply? I suppose now I'm worried because a great deal of why I became interested in linguistics in the first place was what I came across through Wikipedia.

  26. Nathan Myers said,

    June 18, 2008 @ 3:03 am

    On a tangential note, appropriately, I insist, when one discusses sentence diagramming, the name "Brainerd Kellogg" is pure unadulterated wonderfulness. I am tempted to study the history of linguistics just to learn the fate of his arch-rival, Wonknoodle Post.

    john r.: the problem with Wikipedia is that there is little correlation between who knows anything and whose edits survive. Often when there is any disagreement, the person who has less else to do wins, which bodes ill.

  27. dauthie said,

    June 18, 2008 @ 4:20 am

    Shigekuni, of course the below are taught in literature departments here in the US: "transitive, adverb, iamb, trochee, dactyl, pentameter, hexameter, and metrical." Don't let Mark fool you.

    And, Andrew, great post. I'm glad you hint that linguists like Mark can also be accused of arrogance, especially when he says: "it is (typically) uninterested in linguistic analysis of any type at all — also among the missing are phoneme, vowel, consonant, Lakoff, Whorf, 'noun phrase', transitive, adverb, iamb, trochee, dactyl, pentameter, hexameter, and metrical."

    We are "typically" uninterested in pentameters? The heck?!

    I must disagree with this, though: "Humanistic judgment and wisdom is held to supersede any need for technical competence, even when it comes to the very material out of which everything we study is made–language."

    The study of literature is not a technical science, and as for technical competence, it really only has to do with the metrics of poetry. Ours is a hermeneutical enterprise. So, yes, to a great degree, I feel that what we do is above the sciences and math.

  28. Cath said,

    June 18, 2008 @ 6:38 am

    They're not all so great with "phoneme" though.

    If you'll excuse the shameless self-publicisation, it's been noted before that Eng Lit professors don't always make the best use of basic linguistics terminology.

  29. Patrick Dennis said,

    June 18, 2008 @ 9:24 am

    On Wikipedia: Practitioners and professors of sciences other than linguistics make periodic attempts to provide interested outsiders with updates on current thinking and controversy in their respective fields. Are there any volumes on linguistics to be found on the popular science shelf? (No, not Lakoff on political discourse.) If there is a better source than Wikipedia for the non-linguist to find accessible information on latter-day "Aspects of the Theory of Syntax," let us know.

  30. john riemann soong said,

    June 18, 2008 @ 9:31 am

    There was something linked here about scientific methods in literature a while back, or indirectly linked by following a previous link by another LL reader. In any case, the article suggested that applying the scientific method to literature was the only way to save the flagging literature field. Or something like that.

    "john r.: the problem with Wikipedia is that there is little correlation between who knows anything and whose edits survive. Often when there is any disagreement, the person who has less else to do wins, which bodes ill."

    Other than vandals, I didn't know partisans intervened enough in linguistics articles to cause edit wars. ;-)

    What I observe that happens frequently in these areas is that you get the same topic expressed in different articles from different points of view (what we call a POV fork). Where phonetic and acoustic topics intersect with opera, the opera point of view dominates, etc. A visitor flags the topics and then we poor sysops have to reintegrate the articles back together, rarely without a fair bit of research on our part. But the barrier to a combined interdisciplinary approach doesn't seem like a flaw of the project itself, when you observe how biology and linguistics have yet to really embrace each other.

    Besides, POV-pushers rarely write in a manner that's coherent with the rest of the project, so the different register they write in becomes very obvious. Those who do not know anything, I observe, have a hard time coming up with meritable sources and inline citations, if they come up with sources at all.

  31. Nick said,

    June 18, 2008 @ 10:42 am

    In response to Soong's earlier post, there are many linguists (especially in more peripheral fields like Second Language Acquisition) who don't know about the many syntactic frameworks (not just Chomskyite ones, but optimality theory, categorial grammar, lexical-functional grammar, etc). There's a lot of work out there, so many linguists hold onto what they first learned in school and assume it to be true.

  32. Bill Benzon said,

    June 18, 2008 @ 12:28 pm

    About two years ago I posted a brief chronology at The Valve in which I ran up parallel developments in cognitive science and literary theory:

    Back in the late 60s and on into the 70s there was a fair amount of interest in Chomskyian linguistics among students of literature. Thus you'll find Jonathan Culler talking about literary competence (on the model of Chomsky's linguistic competence) and deep structure in his Structuralist Poetics (1975, though I believe it has recently been reissued). Students of stylistics also took a serious run on generative grammar, though Stanley Fish put the kabosh on that (see early essays in Is There a Text in This Class? 1980). There was some interest in generative story grammars among students of narrative, as well as text grammar. But none of this really stuck and made it to the mainstream, though it persists and pools and eddies here and there.

    Why not? Let me offer some observations from my intellectual history. I did my undergraduate work at Johns Hopkins from 65 to 69 and stuck around for an MA in humanities. Though I was a philosophy major, literature became my main interest. I discovered Chomsky in a psycholinguistics course taught by James Deese and went on to read generative linguistics on my own time. On the one hand, literature is constituted by language, so this linguistics stuff ought somehow to be relevant, no? On the other hand, I liked the mechanistic style of the subject; it seemed to me that there was something deep there (of course, there is). But however interesting all this syntax was, it didn't really get at meaning, and that's what the study of literature was about, then (and now). What do these wonderful objects MEAN?

    So I found my way to generative semantics, but that looked pretty much like syntax to me; it still didn't get at meaning. I drifted to Syd Lamb's stratificational grammar, which I like for its visual notation, and for the fact the he used the same abstract structures for phonology, grammar, and semology (to use the terms from the TOC of Outline of Stratificational Grammar). While he dipped into semantics, he didn't dip too deeply. Then Karl Pribram (neuroscientist) had a reference to some work by Ross Quillian. I dug that out of the library and that seemed like THE STUFF. So I looked for more stuff like that and found my way into the early work on semantic networks (Schank, Norman, Wilkes, etc.). In the mid-70s I went to graduate school in English at SUNY Buffalo, but spent a great deal of time studying with the late David Hays in linguistics and eventually wrote a dissertation on Cognitive Science and Literary Theory. That was about knowledge representation with literary examples.

    The point of that story is simply that the nature of my interest in literature drove me to something that looked like it really began to get at meaning. I don't think there was (or) is anything atypical about that interest, it was oriented toward the meaning of texts. That's what literary critics study, the meaning of texts. Linguistics, for better or worse, doesn't quite yet bring the study of meaning within range of literary concerns.

    That's beginning to change. It seems that literary scholars have at long last begun to discover (or is it rediscover) cognitive science. So you see papers, conference sessions, at whole conference (at U Conn, Storrs, last Spring), and books on cognitive science and literature. Cognitive linguistics is big, cognitive metaphor theory and conceptual blending in particular. I should note that critics of cognitive linguistics in literature point out that that stuff looks like old wine in new bottles. I'm sympathetic to that criticism. But if this interest in cognitive linguistics leads to a more substantial interest in the cognitive sciences, and linguistics . . . one can only hope.

    My own view is that students of literature need to come to terms with the idea of computation, both as a real physical process in some suitable substrate (brains, digital computers) and as an abstract mathematical idea. Whatever one may think of the specific proposals that have come out of the generative tradition, for example, it seems to me one can appreciate how certain fundamental abstract ideas allowed Chomsky to argue against Skinner. When students of literature are comfortable with that, then we can roll up our sleeves and get to work.

  33. Lester Piggot said,

    June 18, 2008 @ 12:37 pm

    Andrew Goldstone said about Humanities vs science, Needless to say I would much rather we all became as competent as we could in all relevant fields.

    Good point, linguists can become competent in English. I could be wrong, but it seems to me that linguists have chosen to see their subject as a science and only a science, and whatever the advantage of that strategy, the cost is that they have excluded people who can't (or don't want to) deal with maths and so on. Economics, on the other hand, is an example of a social science thatcan be approached through mathematics, but it doesn't have to be. Philosophy can be approached scientifically, as can architecture, I'm sure there are other subjects that swing both ways. And, of course, there are people who seem equally happy reading a paper about statistics and watching a production of Hamlet, but I suspect that most of them are the bloggers at Language Log. For the rest of us, either (multisubject condition and covariate analysis within SPM2), or reading The Wreck of the Deutschland are going to make us nod off. Lastly, one way linguists could get more attention from non scientists: omit needless jargon. Omit needless jargon.

  34. pc said,

    June 18, 2008 @ 3:41 pm

    @Lester: Linguistics definitely swings both ways (ALL ways, really). Lots of linguists don't know a thing about math, and I'd venture that most linguists who do know things about math-ish subjects (other than computational-type linguists) mostly know things about statistics, and then mostly about statistics as learned through a stats software package, not the theoretical underpinnings (I might well be wrong about this, but it's my impression, just knowing what linguistics grad programs tend to look like). Linguistics is still considered part of the "humanities" in many places, for whatever the division between "humanities" and "social sciences" is worth, and many linguists use methods that align them more plainly with that side of the universe of frameworks. (And some of us are even happy that this is the case!)

    Also, just because I happened to see this, Penn's grad syntax course description reads: "The approach taken is that of contemporary generative-transformational grammar." So the term IS still in non-ironic usage, apparently.

  35. Lester Piggot said,

    June 18, 2008 @ 4:07 pm

    PC said, Linguistics definitely swings both ways (ALL ways, really). Lots of linguists don't know a thing about math, and I'd venture…

    Ah. You're making me feel a lot better about linguistics.

  36. Andrew Goldstone said,

    June 18, 2008 @ 6:33 pm

    dauthie, Thank you for the compliment. It's great to know other people out there are thinking about these issues! But I have to disagree both with the idea that scholars of literature are engaged in a uniquely "hermeneutical enterprise" and with the idea that this is what enables them to ignore work in scientific fields. Instead, I see interpretation as part of every knowledge-producing endeavor; and, conversely, I think that literary criticism, insofar as it aims to produce viable knowledge, ought to be pushed increasingly in certain more "scientific" and technical directions, and to stop using hermeneutics as a shield.

    Even if it were the case that literary critics wanted only to produce interpretations of individual literary works, they would still need the best possible account of all the facts about those works. Meter is not the only such fact that has a linguistic foundation. Understanding patterns in syntax, diction, etc., requires having the right terms to analyze syntax, diction, etc.; and those terms ought to come from linguistics. More significantly, linguistics and literary studies ought to be in a continuous dialogue about such problems as the relation of language to thought. Now as Bill Benzon's comment makes clear, linguistics is not the only relevant scientific field, partly because it does not have a monopoly on the study of verbal meaning; cognitive science and sociology are, for example, of equal interest and importance to work on literature, which in the course of its interpretations makes constant use of assumptions and generalizations about psychology and social behavior. Relying on the folk versions of these fields is a dangerous business for even the most science-phobic hermeneut.

    And Cath, I can't tell you how many times I've seen and heard the term "phoneme" misused by English professors. Depressing. I would be truly (and happily) shocked if James Crippen's claim were correct that "the majority of English profs working in the US probably were required to take some sort of linguistics course during their graduate studies." Would that it were so. My program certainly doesn't, and neither do any of the other ten-odd well-known programs I've looked at.

  37. Aaron Davies said,

    June 19, 2008 @ 12:33 am

    Since no one's explained, continental drift was an early theory that involved continents floating around on a layer of water. Plate tectonics is the proper term for the modern version.

  38. dauthie said,

    June 19, 2008 @ 2:48 am

    Andrew Goldstone, I totally agree with yout that interpretation–and things like metaphors–are part of every knowledge-producing endevour. But does science admit that much of what they do is "mere" interpretation?

    As for syntax and diction: I really don't see why we need to seek the advice of linguists to analyze them. What might a traditional literature scholar be doing wrong in analyzing syntax? And as for phonemes, I dunno, but why the heck should an English lit professor have a good idea of what it is? Both of those are serious questions, btw.

    As for Bill Benzon's post and your comments, at my university (I'm a grad student) we have a professor who is indeed interested in cognitive science, especially "Theory of the Mind." I totally respect such explorations, but, I must say, I am still in no way convinced by what they are doing. Why have we kept looking for redemption from science for so many decades. In order to finally become a "science" ourselves, and then make lots of money? Haha.

    But, again, I respect such approaches, and maybe I'll see the light some day.

    And after reading my first post, I feel I should apologize to Mark: It really wasn't my intention to call linguists or yourself "arrogant." It's just that recently I've been surprised by attacks against lit people from numerous other sources–from the political blogger Matt Yglesias, for example. Last night when I read your posts, it seemed to me like you were giving us advice from above on how we could improve ourselves, as if we were in need of such advice. In any case, I apologize if my post sounded antagonistic.

  39. Another New Haven inhabitant said,

    June 19, 2008 @ 6:44 am

    It's not true that the Yale English department is disinterested totally in linguistics; there are undergraduate classes in the English and Literature departments that touch upon linguistics.

    Prof Roberta Frank teaches Ling 150 ( = Engl 150), Old English where students read Beowulf, among other texts. According to the syllabus (you are welcome to search for "ling 150") the first few weeks cover linguistics-related topics such as pronunciation (probably some phonemes introduced), inflections, strong and weak verbs, prepositions, etc, before moving onto the texts. Another search shows that Prof Frank also taught a class focusing on readings in Old Norse, which I presume will involve some brandishing of linguistics terminology.

    In addition, Prof Stanley Insler taught Litr 150, Sanskrit Classics in Translation, in fall 2007. I couldn't get to the syllabus, but since Litr 150 is also credited as Ling 111, I presume (having taken neither Prof Frank's nor Prof Insler's classes) that there is substantial linguistics-related content in that class too.

    Of course, whether these classes are popular with English or Literature majors, and whether transformational grammar is taught, are entirely different matters.

  40. Another New Haven inhabitant said,

    June 19, 2008 @ 6:45 am

    I forgot to add that Profs Frank and Insler are professors in the Linguistics Department.

  41. Gary said,

    June 19, 2008 @ 9:03 am

    I think that the original comment came fom a universe in which the phrase "transformational grammar" is a half-remembered term that others would call Paul de Man-ian deco9nstruction.

  42. Charles Hartman said,

    June 19, 2008 @ 10:18 am

    As a professor of English at a smallish college who has tried to teach himself some linguistics, I find that the autodidact impulse is hobbled less by shifts in prominence among larger theoretical modes and methods than by fluidity in the terminology. Though "phoneme" looks (to my ignorant eye) relatively stable, "foot" — which in any case means something crucially different from what we lit types were brought up to mean by it — is up for grabs between different groups of linguists who approach metrics from different angles. "Stress" and "accent" name innumerable conflicting concepts. Pullum & Huddleston call terms like "preposition" into question in certain contexts. The boundaries of "syntax" seem to be a perpetual battleground. I may be wrong about all of these, but since I'm an eager student, that's my point.

    I'm not under the impression that the terminological difficulties represent some wild immaturity in the half-century-old scientific field of modern linguistics (though it's interesting that physicists appear to create new terms more often than they redefine old ones). I'm just pointing out a practical difficulty for someone who agrees deeply that the study of literature needs a more rigorous understanding of language than it usually gets. For example, I can't agree with dauthie at all that literary scholars are unlikely to make mistakes that matter with regard to syntactical analysis, or that a definite concept of phonemes is irrelevant to understanding how poetry works. But for somebody trying to get rigorous and detailed knowledge of the terms and tools from books, it's not easy.

    Mark L's post the other day mentioning Vanderslice & Ladefoged (talk about great names!) led me to that 1972 article. It's fascinating. But for an outsider to understand why it was (I gather) long ignored and now might be seen as useful is difficult. "Strong" and "heavy" (which I'd encountered previously in Kiparsky, Hansen, etc.) may not be as bad as "stress" and "accent," but it's clear to me that a linguist would encounter those terms, in this article, with a long list of implicit footnotes in mind that a non-linguist doesn't have so readily available.

    By the way, since Reed-Kellogg diagramming can't represent any connection between sentences like "John kicked the dog" and "The dog was kicked by John," and since it went out of fashion in most grade schools around the time Transformational Grammar came along, I've always assumed TG killed off R-K. Untrue? I hold no brief for R-K (and since it never got spelled out very rigorously it, too, is pretty hard to teach oneself), but my few students who learned it — mostly in Catholic schools, it seems — are a lot less clueless than others about nouns and subjects and the very idea that a sentence *has* a structure.

  43. Nathan Myers said,

    June 19, 2008 @ 1:52 pm

    Aaron: Thank you. So, geologists changed the fluid and the name. "Plate tectonics" is more dignified than "continental drift", which must be what matters. "Big Bang", as a name, started as ridicule, but stuck. I suppose cosmologists are used to ridicule, or should be if they aren't still.

    Diagram that last sentence.

  44. jamessal said,

    June 20, 2008 @ 5:28 pm

    "I am very grateful to you, by the way, for your posts of long ago on metrics; I used them in preparing to teach meter to my discussion-section students."

    Anyone have the link to that post? I'd greatly appreciate it.

  45. Mark Liberman said,

    June 20, 2008 @ 5:55 pm


    "An internet pilgrim's guide to accentual-syllabic verse, 7/6/2004.

    Maybe also "A Thanksgiving discussion", 11/22/2007.

  46. jamessal said,

    June 20, 2008 @ 8:43 pm

    Thanks so much! What you do here is amazing.

  47. T. Cullen said,

    June 22, 2008 @ 5:19 am

    Sorry for my ignorance of URI. I don't really care whether it is called URI or URL, I was just being a pain in the ass for no reason.

    As for my statement about English departments being the last holdouts of TG, I should have clarified that I was referring to English departments where linguistics are wedged in, and more specifically, where applied linguistics is stuck. And of course, when I say applied linguistics, I mean TESL, which somehow has become interchangeable in many peoples' minds. As I mentioned in my posts, I was ill and not thinking clearly, but I doubt that matters much. I should have said something more along the lines of:

    Transformational grammar, in its mid-60s incarnation, seems to be taught in TESL classes, whereas in real linguistics courses, the ideas of Chomsky and whatever the term of the the day is for his theories are still discussed. I must admit to being one of the lowly people who had to attend a university where the "Applied Linguistics" professors where placed in the English department. I should also say, I make no claims about knowing what goes one in English classes that do not pertain to TESL and applied linguistics. So, I hold no illusions about clandestine factions of TG guerrillas holding secret meetings in the English department lounge after hours. I am fairly certain that most of the non-TESL/Applied linguistics profs couldn't give a rat's ass about TG.

    Of course, what goes on in Stanford or Yale, I can only read and dream about.

    One last note: I received an MA in TESL several years ago when the program was quite young in the English department I studied in. In those days, the TESL students were looked at with either puzzlement or outright disdain by the non-TESL English grad students. However, upon returning to the same school a number of years later, I was quite amused to discover that the majority of the students in the English graduate programs were in TESL or Applied Linguistics. This seems to be the case because that is how the English department made money. Hordes of TESL teachers-to-be were paying to go to the school because they new there was money to be made out in the real world. The same is not always the case for poets or literary analysts. The truly funny thing, though, was that the Lit and Creative Writing profs and students still looked down on the TESL students. I won't even attempt to guess why, but the disdain was palpable.

    All of this is from my own very limited experience, so I won't attempt to generalize to other schools or programs. However, I still found freshman comp teachers who couldn't identify a passive sentence, even though they hated them with all their being.

  48. T. Cullen said,

    June 22, 2008 @ 5:29 am

    Please excuse the terrible punctuation, occasional incorrect spelling, and the haphazard use of capitals. I will once again claim fatigue as an excuse, even though I don't really try as hard when writing for blogs. No one cares about those things online anymore and I must admit, I don't either most of the time.

    PS – "friend" as a verb may develop into common usage, as languages do change constantly, but I still hate the sound of it. It reminds me too much of words like "task" as a verb. There are perfectly good words to use instead and yet we turn the noun into a verb, for what reason, I don't know. Of course, Shakespeare was really guilty in this regard, too.

  49. David Marjanović said,

    June 22, 2008 @ 6:40 pm

    continental drift was an early theory that involved continents floating around on a layer of water.

    Not a layer of water, but a layer of what is today called oceanic crust, as opposed to continental crust. This has turned out to be wrong — oceanic and continental crust lie next to rather than on top of each other.

  50. Ellie S said,

    June 23, 2008 @ 9:52 am

    Commenting a bit late on the end of a rather long string…

    I graduated from Iowa State University out of a Linguistics (cross-disciplinary studies) Program where most of my professors came out of the English Department. While most (if not all) of those English-Linguistics crossovers were familiar with TG, I clearly perceived an anti-Chomskian vibe, especially from my advanced grammar professor.

    Now I am an ESL teacher in China (which up until graduation, I swore I'd never do) and I teach oral English to graduate students majoring in Linguistics. One of my student's has Chomsky for an English name. Another student confronted (and I mean confronted) me on why I'm not more interested in TG. It's really all they seem to teach around here.

  51. Nathan Myers said,

    June 23, 2008 @ 7:44 pm

    T. Cullen: I guess you meant to say "yet we we verb nouns."

  52. diodotos said,

    June 23, 2008 @ 7:46 pm

    On linguistic approaches to literature, and especially to metrics, may I mention the work of Nigel Fabb (MIT Linguistics PhD), especially Language and Literary Structure: The Linguistic Analysis of Form in Verse and Narrative (Cambridge 2002), Linguistics and Literature (Blackwell 1997), and the group-authored textbook Ways of Reading: Advanced Reading Skills for Students of English Literature, coauthored with Martin Montgomery, Alan Durant, Tom Furniss and Sara Mills (first ed Routledge 1992, third edition Routledge 2006).

    The last two volumes come with exercises for students. "Ways of Reading" especially should be much better known and more widely used, imho. It has sections on Techniques and Problem-Solving, Language Variation, Attributing Meaning, Poetic Uses of Language, Narrative, and Media Texts.

    Also, Meter in Poetry: A New Theory by Nigel Fabb and Morris Halle has been announced for publication at the end of the summer.

RSS feed for comments on this post