Many people have written about collaboration between disciplines. Recently, I was re-reading Richard Fenyman's anecdotes, from Surely You're Joking Mr. Feynman, and was reminded of his story about why he decided always to avoid inter-disciplinary meetings: however seductive the idea might be, it doesn't work out well!
All the same, there are some good arguments for continuing to encourage academics from different disciplines to collaborate. Solving some immediate problems in renewable energy, for example, requires the cooperation of people from different backgrounds. Companies consist of many people from different disciplines, more or less working together. If academic institutions are to contribute effectively to society in an age of increasing specialisation, it makes sense to at least experiment with inter-disciplinary collaboration. As Feynman noted, however, there are some real difficulties when academics from different background work together, some more fundamental than others; sketched observations from my own experience of inter-disciplinary work follow.
Firstly, how the project fits into the administrative structures of the institution is important.
One cross-disciplinary effort that I was involved with was depicted in hierarchical diagrams as an arrow sweeping across the traditional structure, subject only to the president of the institution. While this may have accurately reflected a certain amount of freedom to operate, the arrow also happened to occupy the same position in the hierarchy that you might expect a secret police to occupy. It looked suspicious! For some academics, the project probably looked a bit too much like an attempt to interfere with or plunder their work. This was never the intention, but it might've looked that way to some; understandably.
In fact, the arrow was essentially meaningless; it conveyed no useful information to us, or to anybody else, about what our role was. In the end, it would have been much easier if the project and its team had been firmly placed in the traditional academic structure from the outset.
This is the first thing that helps the long-term success of a project; it's important to clearly locate the project and the people in the institution's academic structure. The more ambitious its location, the more consistent, coherently articulated, support it will need from senior management.
A second observation is that the project must offer the academics involved some tangible benefits. It's hard to make such collaborations work, but especially with only words of encouragement and ideas.
It seems that small-scale, grassroots collaborations, with one or two enthusiastic academics and some initial funds were the most successful. Ambitious top-down efforts, in general, failed to generate momentum and attract funds. If a top-down effort is to succeed, it needs an awful lot of weight behind it: the people in charge need to be on board. This can be very hard to do well.
So the second lesson is that seed funds help to support nascent efforts that can grow; failed, ambitious proposals often fail and serve to discourage participants.
A third thing that hampers cross-disciplinary efforts can be a lack of a clear vision. A clearly identified project leadership is also essential. A clear vision must shared by the institution management and the academics. If management has no fixed ideas and if the project has no internal mechanism for setting a clear direction, it will eventually fail.
So thirdly, have a clear vision, clear leadership and a well-informed, engaged management.
A fourth obvious, controllable, challenge is the financial uncertainty that's sometimes associated with uncertain management aims. Performance is almost certainly compromised if projects have no clear long-term funding model.
Have well-defined funds in place to support the participants for the duration of the work.
A fifth observation is that luck, and lack if it, also plays a big part. One period of most cross-disciplinary activity, perhaps not entirely by coincidence, coincided with the onset of the recent great recession. At least one major government backed research award was cancelled, quietly.
Luck and its timing, as always, are factors that are, unfortunately, mostly beyond control.
Finally, working out how to communicate is essential; different disciplines have different languages and orthodoxies. Things can be difficult enough between engineers from different backgrounds. A gulf of incomprehension can stand between the engineer and the social scientist.
A real effort must be made to communicate and understand collaborators from different disciplines. This is not trivial.
In general it's hard to make changes in large organisations. In his book Innovation and Entrepreneurship, Peter Drucker, a management thinker, cautions against attempting to create an entrepreneurial spirit without changing the underlying practices of an organisation.
This may be true about any kind of innovation in an organisation, or discipline, and it's perhaps the fundamental difficulty that needs to be understood at the outset of any inter-disciplinary project. It may be that continuing to attempt inter-disciplinary cooperation is one way to effect this underlying kind of change.