What is constructor theory and what does it seek to accomplish?
«When you have ruled out the impossible, what remains must be the truth, however improbable it may sound.» Arthur Conan-Doyle had his famous detective Sherlock Holmes say this sentence. However, this sentence could also be used to summarize the basic principle of constructor theory, which the well-known quantum physicist David Deutsch and the Italian physicist Chiara Marletto have been developing for about ten years now. Deutsch first presented her principles in 2012, and since then it has raised great expectations, but of course has also been met with many skeptics. This is not because it provides the ultimate theory about the fundamental forces of the universe. It is not a competitor to general relativity or quantum physics. Deutsch and Marletto have not yet found out anything with it that we do not already know.
Why is it nevertheless one of the most exciting new developments in physics in the last fifty years? Because it seeks to turn on its head the way we make physical laws and look at nature through the eyes of physics. The usual approach is to figure out the physical law of how an object evolves from state A to state B. The kinematics of an object is the law of how it evolves. For example, kinematics tells us how fast a cyclist gets from one place to another depending on his speed. Quantum physics tells us the same for an electron, and general relativity does the same for planets or rockets. Constructor theory, on the other hand, sees the world completely differently. It considers a change from A to B as a task. But more important are the counterfactuals: They determine whether the execution of a task is possible. Only if it is possible, i.e. not forbidden by a counterfactual, there is a constructor that brings the system from A to B. IT people who are familiar with object-oriented programming may find the concept familiar. In fact, Deutsch has borrowed from information theory here.
What would a practical application look like? Let’s consider a water wheel that revolves, transporting the water it is powered by to the top – in other words, a classic perpetual motion machine. Now we can find reasons why it cannot work: Frictional losses, for example. That would be the classical approach. If we rebuild the perpetual motion machine, we need a new explanation. But we also know a suitable impossibility, a counterfactual: The second law of thermodynamics says that the entropy of such a system decreases with time, so that it CAN’T remain permanently in the same state. Thus we can exclude all such machines at one stroke. Deutsch looks for first applications of the principle in quantum information theory. Marletto has proposed, among other things, a constructor theory of life which explains the origin of life without recourse to chance or higher powers.
In general, the constructor theory might be helpful if we have not yet understood the nature of the dynamical processes involved in getting from A to B at all. Gravity is such a case. We can measure and use it, but we also know that neither Newton’s nor Einstein’s theories of gravity fully describe reality. Constructor theory, say its proponents, would be able to draw new insights solely from a collection of the counterfactuals, the impossibilities, without needing an understanding of the process itself. In a way, it behaves as we do as spectators of a magician: We don’t know what the magician is doing with his white dove, but we try to guess a theory of the magic trick just performed («He must be hiding the dove first in his sleeve, then in his top hat») by means of impossibilities known to us (the magician can’t create a dove out of nothing, it’s too big to put in his mouth or ear, and he certainly won’t kill it on stage).
Whether the constructor theory succeeds in scientific breakthroughs, one will see. In any case, it can be exciting to look at a well-known field from a completely different angle.