To clarify, that quote comes from John Peacock, the reviewer of The Cosmic Landscape for American Scientist.
What he's saying is that physics, like art, is partly a function of the human aesthetic response, because empirical evidence only constrains theories allowed in physics, but does not determine it. I would agree with him only partly, because a large part of what Peacock includes under "aesthetics" is a common preference for simplicity, which can be formalized in various ways (e.g. Kolmogorov complexity).
But even taking into account the formalizable common preference for simplicity, I think there is still a role for aesthetics to play in physics. I can think of two arguments for this. One is to say that the general preference for simplicity in our aesthetic response is a result of the limited complexity of our minds. In other words, everything we consider beautiful must be relatively simple, but not everything simple has to be considered beautiful. Thus the most beautiful theory is not necessarily the simplest one.
The other argument is that all of the formalizations of complexity leave a free parameter, for example the choice of universal Turing machine in algorithmic information theory. Our aesthetic choices can therefore be encoded into this free parameter.
- Re: Multiverse concepts in string theory Wei Dai
- Re: Multiverse concepts in string theory Saibal Mitra