j_doyne_farmer's picture
Professor of Mathematics, Oxford University; Director, Complexity Economics, Institute for New Economic Thinking, Oxford Martin School; Co-founder, Prediction Company
What do these discarded questions tell us?

The road of knowledge is littered with old questions, but by their very nature, none of them stands out above all others. The diversity of thoughtful responses given on the Edge forum, which just begin to scratch the surface, illustrates how progress happens. The evolution of knowledge is a Schumpterian process of creative destruction, in which weeding out the questions that no longer merit attention is an integral part of formulating better questions that should. Forgetting is a vital part of creation.

Maxwell once worried that the second law of thermodynamics could be violated by a demon who could measure the velocity of individual particles and separate the fast ones from the slow ones, and use this to do work. Charlie Bennet showed that that this is impossible, because to make a measurement the demon has to first put her instruments in a known state.

This involves erasing information. The energy needed to do this is more than can be gained. Thus, the fact that forgetting takes work is essential to the second law of thermodynamics. Why is this relevant? As Gregory Bateson once said, the second law of thermodynamics is the reason that it is easier to mess up a room than it is to clean it. Forgetting is an essential part of the process of creating order. Asking the right questions is the most important part of the creative process. There are lots of people who are good at solving problems, fewer who are good at asking questions.

Around the time I took my qualifying examination in physics, someone showed me the test that Lord Rayleigh took when he graduated as senior wrangler from Cambridge in 1865. I would have failed it. There were no questions on thermodynamics, statistical mechanics, quantum mechanics, nuclear physics, particle physics, condensed matter, or relativity, i.e. no questions covering most of what I had learned.

However, the classical mechanics questions, which comprised most of the bets, were diabolically hard. Their solution involved techniques that are no longer taught, and that a modern physicist would have to work hard to recreate. Of course, in a field like philosophy this would not have surprised me — it just hadn't occurred to me that this was as true for physics as well. The physicists in Rayleigh's generation presumably worked just as hard, and knew just as many things. They just knew different things. After overcoming the shock of how much had seemingly been lost, I rationalized my ignorance with the belief that what I was taught was more useful than what Rayleigh was taught. Whether as a culture or as individuals, to learn new things, we have to forget old things. The notion of what is useful is constantly evolving.

The most important questions evolve through time as people understand little bits and pieces, and view them from different angles in the attempt to solve them. Each question is replaced by a new one that is (hopefully) better framed than its antecedant. Reflecting on those that have been cast aside is like sifting through flotsam on a beach, and asking what it tells us. Is there a common thread that might give us a clue to posing better questions in the future?

When we examine questions such as "What is a vital force?", "How fast is the earth moving?", "Does God exist?", "Have we seen the end of science?", "Has history ended?", "Can machines think?", there are some common threads. One is that we never really understood what these questions meant in the first place. But these questions (to varying degrees) have been useful in helping us to formulate better, more focused questions. We just have to turn loose of our pet ideas, and make a careful distinction between what we know and what we only think we know, and try to be more precise about what we are really asking.

I would be curious to hear more discussion about the common patterns and the conclusions to be drawn from the questions that have disappeared.

J. DOYNE FARMER, one of the pioneers of what has come to be called chaos theory, is McKinsey Professor, Sante Fe, Institute, and the co founder and former co-president of Prediction Company in Santa Fe, New Mexico.