Academia should forget artists and embrace artisans

The more I read first drafts of working papers, the more I pick up on scholars’ diverse skills: some authors are really good at coming up with an interesting question; some can really execute a rigorous research design; some have a flare for composing elegant prose; and some are really good at organizing a manuscript. But few – very few!! – are good at all those things combined. In yet another piece in my saga to save academia from itself, I say scrap the expectation that a single scholar must struggle to be a creative, methodologist, writer and editor. An alternative system is possible.

The “artist model”

The humanities and some social sciences seem to have an irrational attachment to the “artist model of research”, in which we expect a lone genius to produce a brilliant idea, execute it flawlessly, write it understandably, and package it persuasively. It’s as if we were painters locked up in our basements, or singer-songwriters waiting to be discovered by a record label: we price solitary obsession, the commitment to purity of vision, and an ability to be both a specialist in a very narrow topic and a jack-of-all-trades. Don’t let conferences and peer review fool you: this is about me doing my thing, and the rest of you can go to hell!

Ironically, we have become a community of individualists, competing with each other for the greatest artistic achievement that only a dwindling group of people – ourselves – will ever see or understand. Not only that, we are often proud of this inward-looking artistic isolation: just like musicians who “sell out” by signing up with big labels, all too often we criticize researchers who publish public interest books, accusing them of simplifying our beautiful and arcane science for the sake of “impact” (which we suspect is code-word for “selling books”). It’s as if we yearned to be misunderstood.

The funny thing about this is that the artist model is not the only one, let alone the best one. The natural sciences usually go for the lab model, in which the project lead gets the money and sets the agenda, competent postdocs execute it, and grad students (cheap labor) carry out the experiments and collect the data. But – as someone who has family in the hard sciences – I can tell you that this system can easily degenerate into exploitation and free-riding, with co-authors constantly being added to research products in which they had only a tangential role.

I would actually look outside academia for a new model: film-making.

The “artisan model”

There are some film-makers out there who do write their own scripts, light and decorate their sets, shoot all scenes, edit them together, and even compile a soundtrack for their movie. But they are few and far in between: with a few exceptions who may be considered legitimate geniuses, these lone creators tend to be amateurs. Instead, the film industry realized a long, long time ago that many different skills go into producing a quality movie, and that individual members of the team may be left to focus on their specialized task, as long as they are all guided by a common purpose. As much as we like to think of auteur directors and screenwriters, the film industry is one of artisans or craftsmen. (Love Stanley Kubrick? Almost all his classic films were based on novels, were shot by talented cinematographers, and scored by genius composers; not to speak of the great actors who starred in them. Can you imagine Dr. Strangelove without Peter Sellers? Exactly. Remember that next time you list it among your favorite “Kubrick films”.)

Taking a cue from film-making, an “artisan model of research” would also recognize the merits of specialized skills, and strive to produce a finished product that was indeed greater that the sum of its parts. Instead of a lone author struggling with concepts, research design, and editing, a research project should combine people fulfilling one or more of the following roles:

  • The Producer. Someone has to be in charge to make sure that everything works out, to ensure that everyone is doing their job. My apologies to those who crave “freedom” or “autonomy”, but it’s not a coincidence that countries, businesses, and even social movements tend to have hierarchies. We already have that in the form of the research director, the program director, or the project lead, so it would be just a small step to get that person to manage the production of written research outputs.
  • The Creative. We were told time and again in gradschool that good research begins with an interesting puzzle, which begs the question: why are there so  many absolutely boring pieces of research out there? It’s one thing to do what Thomas Kuhn would call “normal research”, another one altogether to fall into the trap of pursuing mildly incremental – and incrementally boring – contributions that fail to capture anyone’s attention. The fact is some researchers are better at asking questions than others, but they may lack the skills or drive to shepherd those questions down the path to research findings. That’s fine: give them a task at which they can excel.
  • The Methodologist: Just like there are people who love to ask “what if?”, there are many researchers out there who thrive on consistency and testing: they are the born methodologists among us, who keep asking questions about omitted variables, selection bias, internal validity, or robustness. It is a travesty that our disciplines sometimes waste these individuals’ talents by asking them to also come up with questions of their own – resulting, more often than not, in high-quality studies of largely irrelevant questions. Of all of us they are already artisans – proud ones! – so give them room to contribute instead of asking them to be what they are not.
  • The Communicator. Even when a paper asks an interesting question, even when it articulates it through a powerful analytical methodology, the most frequent sin committed by social scientists is one of boredom. No wonder: ask people to constantly hone their conceptual and methodological skills and their ability to communicate effectively will erode gradually but inexorably. With some exceptions, as we become better “researcher-artists” we tend to communicate with an increasingly reduced community of like-minded people, our artistic peers. This makes for boring papers that are a slog to read, and also to diminished policy impact and access to funding due to a failure to persuade non-specialists. Good communicators – both written and spoken – are essential to good research.

And there you have it: these are the four roles that any research has to satisfy. Perhaps you are a brilliant individual who can wear multiple hats at once: maybe you truly are an “research auteur” who can produce, create, analyze and communicate. Kudos to you. The rest of us? We probably excel at two or even just one of those four skills, and outside of that we try to improvise and get by with some help from low expectations. Research teams structured around artisans – producer, creative, methodologist, communicator – would produce higher-quality research, of greater significance, and with a larger chance of impacting the outside world.

I for one would welcome an academic community with fewer artists and more artisans.