Ph.D. students I meet for the first time are sometimes taken aback when I ask them with a smile “What’s your problem?” I then clarify, “What problem are you addressing in your research?” Even then, they are often still bemused. That’s because many research students think that their task is to research a “topic”. Later, they conclude that their job is to “answer a question”.
However, for the most part, Ph.D. students, like many other researchers, should be driven by problems of understanding. There should be some situation, state of affairs, data and/or capability that causes in them deep cognitive unease. The situation calls for cognitive work to solve the problem. It’s the motivator for research.
I ask the foregoing questions of students for several reasons. It’s mainly out of curiosity about their research. But I also want to plant in them a seed: that problems should drive their research. Further, this moves the conversation forward. One can talk endlessly and progressively about problems. Talking about topics is not nearly as productive.
I don’t expect a precise answer to the “What’s your problem?” question from a new Ph.D. student. To discover and specify a problem worth spending several years on is tough! If the problem is really new, it consists in a historical contribution to knowledge. Even if it is not new, it counts as knowledge building. (One can build knowledge.) Discovering what is not known, what is misunderstood, and why that matters can qualify as a discovery to report in a Ph.D. thesis. (A good experiment can do this by disconfirming a prediction derived from a theory.)
So, students do need to focus their attention on problems. If they don’t know that they should be trying to formulate their problems, then we can do a service to them by letting them know, whether explicitly or implicitly.
The primacy of problems is not my discovery. The two pre-eminent philosophers of science of the 20th century, Karl Popper and Imre Lakatos, are quite clear about the role of problems. (See for example Popper’s Realism and the aims of science.)
Problem-driven enquiry is the approach the designer-stance takes. The designer-stance is a systematic way of conducting Artificial Intelligence [AI] research, which sheds light on existing and possible minds. You can get a sense of the role of problems in Teach theses, a document about writing Ph.D. theses. David Chapman’s “How to do research at the MIT AI lab” is also relevant. Even if you’re not into Artificial Intelligence, these documents will help you understand how AI research is conducted. It’s different from many other sciences. But many disciplines outside of AI stand to gain from knowing how AI is done. (Not that there is just one way to do AI research.)
This post can be read as follow up to the one I wrote yesterday about Learning in Depth. There I mentioned that Learning in Depth assigns a topic to students. So it is prima facie topic-driven inquiry. I believe it is a very useful approach for grade school. However, it needs to be combined with problem-driven research. Problem-driven research should become dominant in high school, and super-dominant in adulthood.
Being on the look-out for problems, knowing one’s way around one’s problems, stating them, analyzing them, keeping track of them and systematically addressing them: These are essential to knowledge-building and personal progress.
My upcoming book provides several tips to help you use knowledge and information technology to build knowledge and develop yourself.
[Update: my Cognitive Productivity book has been released].