Interdisciplinarity

Timothy Burke has some interesting thoughts about the College of the Atlantic, which represents a real effort to build interdisciplinarity on an institutional level. "Interdisciplinary" is the buzzword of the moment in large swathes of academia, and the College of the Atlantic, which doesn't have departments and works very hard to make connections between disciplines, is sort of the apotheosis of the interdisciplinary movement.

Toward the end of his post, Burke relates a story from earlier in his career:

When I was briefly at Emory at the start of my career, I was in a workshop on interdisciplinarity. After many of the junior people there duly celebrated interdisciplinary study, a very wise, interesting senior professor, a classicist, stirred himself. "You guys," he said, "don't have a beef with disciplines. You have a beef with departments. Everybody's interdisciplinary in some respect in their scholarship. It's departments that cause the problem by raising the barriers to interaction and discussion".

To some degree, I agree with this-- a lot of the problem with interdisciplinary work has to do with the structure of academia, not the content of the disciplines. But we shouldn't overlook the fact that there are very real differences in the ways that scholars in different disciplines go about their business. The most extreme example, and the one that causes the most angst when discussing merit evaluations, has to do with book publishing-- scholars in the humanities and social sciences are expected to produce scholarly monographs, but that's almost unheard of in the natural sciences and engineering. This makes it very difficult to construct a system that appropriately rewards both humanists and scientists for their scholarly production.

But there are significant differences even within the sciences:

The one that comes to mind immediately is Physical Review Letters. PRL is probably the top physcis-specific journal, and it has a very hard four-page limit for articles-- all the figures, tables, text, and references must fit in four journal pages. This means that the very best work in physics is crammed into short four-page articles.

This isn't the case in all disciplines-- in many areas, the most prestigious journals don't have such tight page limits, and the length of a paper can serve as a rough measure of the effort that went into the work. It's a little difficult to get people to understand that a four-page PRL can represent a much greater effort than a fifteen-page Journal of Subfield Physics article.

And, of course, there's the extreme example of computer science, where conference presentations are the prestigious and important measure of scholarly output, and journal articles mean relatively little. At least we have Science and Nature to help normalize differences between natural science fields-- CS is just bizarre...

There are also significant differences in the way different disciplines do research. In chemistry, a lot of the equipment seems to be shared-- there are a few lab-specific instruments, but for the most part, everybody uses the same NMR machine, the same UV/VIS spectrometer, and so forth. The effort goes into making the samples that are tested, while the big expensive equipment is owned by the department as a whole.

That doesn't work so well in physics. Every bit of major equipment in my lab is a custom job-- if one of my lasers dies, I can't just borrow one from a colleague, I need to rebuild or replace that one. It means that the time scale for building a lab is different, and the way that we spend start-up money is radically different.

So, while a lot of the barriers to interdisciplinary work are structural and departmental, we shouldn't lose sight of the fact that there are real, tangible differences that have to do with the nature of the subjects being studied. It's not as simple as eliminating departments-- which Burke says as well, but I think the problems run deeper than the sort of thing he talks about in his final paragraph.

Categories

More like this

I know this has come up before, but is PRL really that much better than other journals? In my field, at least, I find their standards incredibly bizarre and certainly not representative of what I would consider the "best" papers out there.

You have a $59K toroidal pseudoencabulator. A Chemistry colleague borrows it to collect data for a paper of his own. Does the NSF, having grant funded its purchase, smile upon you for efficiency or file charges of embezzlement of laboratory funding?

Similarly, in CS, everything interesting is in 10-15 page conference papers. Journal articles take too long to come out...

Aaron asks: is PRL really that much better than other journals?

That is definitely subfield dependent. In many subfields you are absolutely expected to publish your best stuff in PRL--indeed, I have heard rumors that certain departments will only count your publication if it was in PRL (or perhaps Nature or Science). In other fields, such as my own, the differential is definitely much less, in part because the audience is significantly smaller: people will go looking for your PRL if they know about it, but for historical reasons they are much more likely to see your letter if you publish in Subfield Physics Letters instead. Specifically, it has always been relatively easy to read papers in that journal and its companion Journal of Subfield Physics, since the publisher of those two journals has had a longstanding practice of selling cheap individual subscriptions. It probably does have a significant negative effect on my subfield that we rarely publish in PRL, but I don't have any hard statistics to back my intuition on that.

My WAG on the topic is that it depends on the relative need for longer papers. In my subfield there are significant results that cannot be adequately explained in four journal pages, so longer papers are relatively important. (This is in addition to the historical factors.) I have no idea whether this is true for your field.

By Eric Lund (not verified) on 09 May 2008 #permalink

I think the department/interdisciplinary problem goes much earlier than faculty advancement issues. It's a major issue for faculty hiring. If a person truly spans two departments, that means they don't fit well in either. Sometimes schools have searches for joint faculty, but generally you need to convince a single department that you are primarily of that group to get a job. This continues in the grant application process. Even with the interdisciplinary lip service, most review committees have a specific field's focus and don't positively evaluate ideas that partially leave that focus.

Aaron: I know this has come up before, but is PRL really that much better than other journals? In my field, at least, I find their standards incredibly bizarre and certainly not representative of what I would consider the "best" papers out there.

Yeah, but look at the people who are in your field...

I obviously can't speak for all subfields, but back when I was in grad school, the local attitude was that if it didn't at least have a chance of getting into PRL, it wasn't worth writing it up. I published one Phys. Rev. A as a grad student, because somebody else scooped us on the main result, and my immediate supervisor thought it would be good for me to write one longer paper. Other than that, we generally wrote things up for PRL, and would settle for a Rapid Communication in PRA. By the time I left, the BEC stuff was all aimed at Science or Nature.

I think there's definitely a difference in status between PRL and other PRA journals, in the subfields I'm familiar with. There's also a significant step between Phys. Rev. journals and Journal of Subfield Physics papers, particularly when it comes to evaluating publications in job searches.

"[in chemistry] The effort goes into making the samples that are tested, while the big expensive equipment is owned by the department as a whole."

At Union, for the most part, yes. By and large? I'm not so sure. Personally, my research is all about building the better piece of equipment to analyze the chemicals (or the biological cells, or the hunk of metal...) which makes most of the equipment I've worked with custom and homemade.

bsci's comment in #5 points to the need for some clarity in what we mean when we talk about interdisciplinarity. In my experience, the most common goal in the sciences is to bring together a variety of disciplinary specialists to tackle a complex problem that spans traditional field boundaries. For humanists and social scientists, interdisciplinarity more often involves an individual working between traditional fields. Both are important and challenging, but the changes to traditional administrative structures needed to support one model versus the other are very different. Sometimes people try to split these concepts using words like multidisciplinary or transdisciplinary, but more often people just use interdisciplinary without noticing the overloaded meanings.

By Anon dean (not verified) on 09 May 2008 #permalink

MRW: "[in chemistry] The effort goes into making the samples that are tested, while the big expensive equipment is owned by the department as a whole."

At Union, for the most part, yes. By and large? I'm not so sure. Personally, my research is all about building the better piece of equipment to analyze the chemicals (or the biological cells, or the hunk of metal...) which makes most of the equipment I've worked with custom and homemade.

That's my impression from listening to a lot of student talks, so I'm not 100% sure it's accurate. I get the same basic sense from a lot of the chem blogging I've read, too, but again, I can't swear that that's representative.

A physical chemist at Williams once started a seminar by giving the joke definition that a physicist is someone who buys a sample and builds an apparatus to study it, while a chemist is someone who buys an apparatus and then makes a sample to study with it. A phyisical chemist, he claimed, was thus someone who built an apparatus and made a sample to study with it.

Which, I suppose, makes a chemical physicist someone who buys a sample and buys an apparatus to study it with...

Anon dean: In my experience, the most common goal in the sciences is to bring together a variety of disciplinary specialists to tackle a complex problem that spans traditional field boundaries. For humanists and social scientists, interdisciplinarity more often involves an individual working between traditional fields. Both are important and challenging, but the changes to traditional administrative structures needed to support one model versus the other are very different.

This is an excellent point. Thank you.
It also illuminates the comment "Everybody's interdisciplinary in some respect in their scholarship." in Burke's story-- that's much more true for humanists than scientists.

MRW might be an analytical chemist. That is what a grad school friend did during his PhD work, and continued to do in the medical industry.

My observation, from a past that included a fair bit of time sliding down the razor blade of cross-, multi-, or inter- disciplinary science, is that my CC comes closer to that model than anything at any university except for one odd teaching intensive "school" I know about. The reason is that we don't have disciplinary departments. There is a chemist next door, a biologist across the hall, and a volcanologist plus two calculus profs not much farther away. We talk shop every day. I know how they teach thermo. I know the specific procedures that are emphasized in calculus (such as how they teach u substitution). I know what day they get to Stoke's theorem, so I can review Gauss' Law from that perspective.

That just does not happen in a university, and it is mostly because of the department structures that dominate life in that system. You are who you talk to every day.

As the Anon Dean points out, you have to find a work-around at a university (like a Materials laboratory shared by chemists, physicists, and engineers) because it is not built into the administrative system. And when those places exist, a very real concern of untenured faculty is that they will be viewed as NotPhysicists or NotChemists because they have become contaminated by those associations. Will their outside letters understand what is going on?

MRW is indeed an analytical chemist.

I certainly agree with Chad's broader point about different disciplines, and I think he has fairly assessed research by most chem. students at Union (I was one of them a couple years before Chad got there). I'm just being nitpicky because I think that the statement is a bit skewed towards the type of research that generally goes on in an undergrad chem. department, and is not representative of the research that goes on in many analytical or physical labs at research institutions.

Personally, I have no interest in synthesizing anything with a more complicated recipe than my dinner. In fact, in grad school, I don't think I even did anything more complicated than making powdered lemonade.