Interdisciplinarity and Cognitive Science

Christopher D. Green

York University

Presented at the University of Toronto Interdisciplinary Symposium on the Mind

30 January 2004

The title of this Session, "Reflections," gives me fairly wide latitude to talk about whatever I please. So please pardon me if I take some advantage of that opportunity and emphasize the "Interdisciplinary" part of this conference's theme more than the "Mind" part. Rather than making a case for a particular area of the content in Cognitive Science, today I would like to make a plea for Cognitive Science as an institution -- reminding you of some of its frailties in the administrative context of the modern university, and asking you to act with me to help shore it up a bit.

The attempt to forge an interdisciplinary career -- such as one in cognitive science -- remains problematic for scholars and scientists alike in universities almost everywhere. Researchers who embark on studies of the interdisciplinary, multidisciplinary, transdisciplinary, or pluridisciplinary kind (pick your favorite term), risk a number of less-than-optimal professional outcomes, most of them stemming from the disinterest or even distrust of the conventional members of any of the disciplines from which they draw inspiration.

First, of course, is the problem of getting a permanent position at all. Job descriptions are almost always the outcome of competing interests among the members of a single departmental discipline, and rarely is there much room for advocates of extradisciplinary factors in that struggle. As a result, interdisciplinary individuals, who rarely line up well with the demands of disciplinary job descriptions, can have a tough time in the highly competitive market for obtaining an interview and ultimately, a job. Of course, many interdisciplinarians learn early on to "play it straight" when going for a job, "massaging" letters of intent and CVs to make it seem as though they are one of "the tribe."

Even when this strategy is successful, however, interdisciplinarians are then faced with five-to-seven more years of "playing it straight" in order to get published in the "right" journals, invited to the "right" conferences, and ultimately obtain the holy grail of tenure in a field where they may feel only half at home. By this time many have either failed to "make the grade" by the lights of the discipline that has hired them (or have decided to stop trying), or they have become victims of a kind of intellectual Stockholm syndrome, having come to identify with their disciplinary "captors" and deciding to play it straight for the rest of their careers. The question of when one should "come out" as an interdisciplinarian can be a tricky one indeed.

In a few places, interdisciplinary studies of one sort or another, have gained departmental status -- effectively moving from the foggy and uncertain realm of "interdiscipline" to that of "discipline," but this still remains the exception much more than the rule. Cognitive science is, perhaps, better off than a number of other "interdisciplines," in no small measure because it has an agreed-upon name and there are a few academic journals and more than a few books that use that name in their titles. And then, of course, there was the early interest in the field by the U.S. military, but that is the topic of a whole other talk… Still, there are many schools in which cognitive science has little place (apart from the trendy self-descriptions of some otherwise fairly conventional disciplinarians), and there are many others in which "cognitive science" is, in the main, just words on a recruiting poster. Even right here at Toronto, tribulations within the undergraduate cognitive science program have, in no small part, been caused by the hesitation of the university to establish permanent professorial positions in the field. These are the kinds of things that make young aspiring interdisciplinarians run for the cover of well-established research programs, rather than striking out into new and innovative territory. In short, disciplinary institutions, at least in this matter, hamper rather than facilitate, the acquisition of knowledge.

There is currently a year-long web conference on the question of interdisciplinarity being hosted by the French Centre Nationale de la Recherche Scientifique (http://www.interdisciplines.org/interdisciplinarity). Among the individuals participating who would be known to cognitive scientists are Dan Sperber, Howard Gardner, and Toronto's own Ian Hacking. One of the interesting claims repeatedly made in the articles there is that, contrary to the university's stated and traditional goal of fostering new and innovative thinking, interdisciplinarity fares much better in industry than it does in the academy. This is in no small measure because industry doesn't much care about the scholarly apparatus of journals, conferences, departments, tenure, granting agencies and the like but, rather, hires the team of people it needs, regardless of where they come from, in order to get a particular problem solved. If it takes a psychologist, a computer scientist, a neuroscientist, and philosopher to get it done, so be it.

Universities, by contrast, rarely operate in this way. Problems to be solved are rarely imposed from the outside (applied professional programs such as medicine and engineering notwithstanding); they are developed from the inside by the faculty members themselves. But faculty members are not really at liberty to study anything they want. They must consider factors such as the likelihood of the relevant granting agencies funding such research, the availability of publication vehicles such as conferences and journals accepting such research as being within their scope, the likelihood (come time for tenure or the next promotion) of their own university departments recognizing such research as being a contribution to their discipline (rather than just to knowledge as a whole), and so on.

In essence, understanding all of this amounts to recognizing that the distribution of disciplines in the academy, as it now exists, is not simply a matter of our having found the most "natural" or "efficient" epistemic dividing lines in the broad spectrum of human knowledge and then arranging our university structures to mirror those. Disciplinary boundaries are at least as much about institutional administrative structures as they are about the "proper" division of knowledge. Psychologists should be more aware of this than most because of their own discipline's history.

Because of the dominance of Wilhelm Wundt's Leipzig laboratory in the textbook accounts of the history of psychology, the battle that experimental psychologists had to wage in the early 20th century in order to achieve the level of an officially recognized discipline in the German university system is not well known. The story is told in the first few chapters of Mitchell Ash's masterful book, Gestalt Psychology and German Culture (1995). During the end of the 19th century and the start of the 20th, experimental psychology was regarded as a subdiscipline of philosophy. The justification for this was the oft-made claim that experimental psychology had the power to usefully address important philosophical questions that that had long eluded resolution, not only in the philosophy of mind but also in fields like epistemology, aesthetics, and even ethics. As chairs in philosophy became available due to the death or retirement of older professors, they often went to individuals with training in experimental psychology. At first, there was little objection to this because the experimental psychologists were also trained in the full range of philosophical disciplines and thus were seen as members of the philosophical community, able and willing to teach students and to make original contributions in a variety of philosophical areas. As time went on, however, and experimental psychology began to acquire a large literature of its own and an array of investigative techniques to be learned, individuals whose training was almost wholly in experimental psychology began to acquire philosophical chairs as well. These individuals had neither the expertise nor the interest to teach students in traditional philosophical methods and content areas, nor were they prepared to contribute to these areas of study themselves. Thus, they came to be seen as a threat by traditional philosophers, taking up positions that the philosophers, perhaps understandably, believed should be reserved for members of their own guild. Unfortunately for traditional philosophers, they did not have control over the appointments to these positions, which were made instead by government officials. (Those of you who have seen my documentary on the hiring of James Mark Baldwin and James Gibson Hume at Toronto in 1889 will know that roughly the same arrangement existed here at about the same time.) In 1912 matters came to head when a petition was circulated among professors throughout Germany and Austria demanding that philosophical chairs no longer be given to experimental psychologists. It was signed by 107 professors. Of the 66 professors occupying philosophical chairs at that time, 27 (41%) signed. If one eliminates the 12 philosophy professors who were, at the time, experimental psychologists, the proportion of philosophers who signed rises to fully one half. Wundt himself, then over 80 years old, was forced to leap to the experimentalists' defense, publicly castigating the traditional philosophers for putting their "property rights" (as he put it) ahead of the good of their field of knowledge (Ash, 1995, pp. 46-47). Partly because of the objections of the traditionalists, but also due to the onset of World War I, it was a good long time before the solution of creating separate psychology departments was agreed to and implemented in Germany and Austria.

In the North America, things were somewhat different, in no small measure because the American and Canadian university systems were still quite immature at the time -- having only just developed serious graduate schools very late in the 19th century -- and so there was more room for growth and innovation than there was in Germany's centuries-old system.

But now, in 2004, we find a problem with respect to cognitive science and other "interdisciplines" that is in many ways similar to that the then-interdiscipline of experimental psychology experienced in Germany in the early 20th century. But there is one crucial difference; one that is even more detrimental to interdisciplinarians. Because departments, for most intents and purposes, do their own hiring now, they do not have to go to the extreme of firing petitions off to government officials in order to maintain their "property rights" as Wundt so aptly put it more than 90 years ago. They are simply at liberty to ignore applicants who don't fit into their traditional idea of what constitutes a member of their discipline, and to deny tenure to those who do manage to slip through the front gate.

If one examines the procedures that are now employed with the aim of rendering tenure and promotion decisions more "objective," one can almost immediately see where a critical problem lies. The members of tenure committees in psychology, I would venture to say, almost never actually read the published work of the candidates whose futures they are deciding. This is, at least superficially, for good reason -- psychology is so broad in its scope that most of us do not know what constitutes good and bad work in subdisciplines of the field apart from our own. Instead they rely on the putatively "expert" opinions of a few reviewers and they rely on some numbers: (1) counts of the articles published in peer-reviewed journals, (impressions of the relative status of those journals, sometimes bolstered by their "impact factors"), (2) counts of the number of times the articles have been cited by other members of the discipline (of course, if an article has been cited as a prime example of how not to investigate a topic, that article remains a "tick" in its favor if no one on the committee actually reads the citing article), and (3) the number and value of the research grants the candidate has received (along with impressions of the perceived status of the source).

For someone who works at the center of an established subdiscipline, this procedure can work relatively well. The journals in which one "should" publish are well-known to the community as a whole (as are the journals one would turn to should one's work not be accepted by the "top" journals). The appropriate granting agencies are well known, as are their success rates and standard granting amounts.

For someone who is interdisciplinary, however, this approach can be quite unfair. Other "experts" in the interdiscipline may be difficult to locate, there are few or no journals catering to the field the research is pioneering, so s/he has to publish in vehicles that are not well-known (and which are often perceived by the conventional disciplinarians of the tenure committee as "questionable" journals). The "impact factors" of these journals and citation counts of the articles published therein are low because there are not as many people working in that area and so not as many citations in total. (Just for instance, the three best journals in the history of psychology are Journal of the History of the Behavioral Sciences, History of the Human Sciences, and History of Psychology. The first two have impact factors of .267 and .207 respectively. By comparison, Can. J. Exper. Psychol. is 1.182(~5x); JEP:LMC is 2.443 (~10x). The third major history of psychology journal is not even indexed by ISI's Journal Citation Reports.

At least in psychology departments they still occasionally debate the importance of such numbers when making tenure and promotion decisions, allowing at least for the theoretical possibility of taking the special circumstances of interdisciplinarians into account. I have heard tell that in some economics departments, however, they aspire to rank order all the journals in the field and then simply weight and sum the publication records of tenure and promotion candidates.

For a true interdisciplinarian this kind a system is disastrous. The official criteria, even if developed in consultation with a group of the "best" people in each discipline (perhaps because it was developed in consultation with them), can only penalize people for doing anything other than continuously writing conventional journal articles explicitly designed for the sensibilities of the most venerable journals in the most traditional subdisciplines. While this may make for a great deal of high-quality conventional research (and there is something to be said for that), it allows almost no room for people to explore new combinations of ideas drawn from different intellectual domains. What is more, it makes it virtually impossible to explore new ways of presenting research, such as, just for instance, the various scholarly World Wide Web sites I have created over the last few years, or the video-documentary-linked-to-footnotes-and-primary-source-transcriptions format that I have been developing recently. All this work -- whether good or bad -- would simply fail to be considered at all because it did not correspond to an appropriate tick-box in the list of "officially sanctioned" forms of expression.

Let me say as an aside that I am not complaining about my own current situation. I have been extremely lucky in that York has treated my esoteric ways quite tolerantly, though even there, I have made it a point to see that everyone knows the kind of thing I am doing and to "talk up" my justifications for doing so. That way, my style of work is a surprise to no one come committee time. Most other interdisciplinarians, however, do not have it so lucky. Either they aren't strategic enough to properly "prepare the ground" as I do, or they do so and no one cares. Here, I want to speak out for them.

Although I am, in all likelihood, preaching to the converted here, I would like to close with a plea to those in this audience. We interdisciplinary cognitive scientists should move to foster interdisciplinary research in our own home institutions. This means doing things like seeing that job descriptions are written so that interdisciplinary cognitive scientists have a chance at getting interviews, seeing that after-the-fact discussions of interviews are not completely dominated by parochial subdisciplinary sensibilities about who is "one of us" and who is "one of them." When it comes time to revise tenure and promotion regulations, we should try to insert language that makes it possible for interdisciplinarians to pass muster. (One thing I did recently was to argue for the inclusion of the phrase "electronic publications and productions" in the list of things the tenure committee is obliged to consider as part of its evaluation of candidates' research records. This particular wording may not be perfectly suited to the problems of cognitive scientists, but you get the idea.) Try to break down artificial, bureaucratic disciplinary boundaries where possible. If you're a psychologist, attend the philosophy, linguistics, and anthropology colloquia from time to time. Invite like-minded folk there to attend the psychology department colloquia. Encourage your students to do the same. Universities are divided up into departments for historical and administrative reasons. Try not to let those issues interfere with your intellectual development.

As a final comment, please note that I have not intended, in this talk, to claim that all interdisciplinary work is "good" and all disciplinary work is "bad." Indeed, people can generate extremely poor interdisciplinary work because there is an opening here to treat "other" disciplines superficially. (This is often signaled when claims are made that begin with the phrase "Philosophers believe…" or "Psychologists have shown…" Which philosophers? Which psychologists? What was their argument or evidence? Every discipline is contentious and diverse). Instead, all I am trying to say is that interdisciplinary work, such as that in cognitive science, deserves a fair hearing. Many modern disciplines (psychology perhaps chief among them) got their start as an interdisciplinary study. You can either assume that the disciplinary distribution we have now will be so for all time, or you can try to assist in bringing to light the new disciplines of the 21st, and even the 22nd, centuries.