Sunday, September 23, 2012

Act and Potency in Physics

A while back, BeingItself commented on this post by quoting Ed Feser:

The Aristotelian theory of act and potency is the classic example of such a piece of middle ground knowledge. It is grounded in the basic empirical datum, the fact of change. But it is not a description of this or that particular change or this or that particular kind of change but rather of all change as such. Hence while empirically grounded it is not subject to falsification by theorizing in physics, chemistry, etc., because the phenomena dealt with in all such theorizing, since they all involve change, implicitly presuppose the theory of act and potency.

BI then asked,

Do you think that is correct? Do physicists presuppose this metaphysics?

This is a great question, and I thank Mr. Mosis for bumping it back to my attention. I really have no idea what the answer is. Although I've read a fair smattering of philosophy of science and philosophy of physics, I can't recall ever coming across a discussion of the metaphysical assumptions underlying physical theories - not even in discussions of the foundations of quantum mechanics, where some sort of metaphysical groundwork would really be useful. As Feser points out in the linked article,

[F]or most of its history the philosophy of science was essentially concerned with questions about the methodology of science, the logical structure of scientific theories, the meaning of scientific assertions, and the like...
Feser says there has recently been a return to metaphysical issues, and links to some relevant books. Not having read these, though, I'm going to plunge in and try to give you an idea of how physicists go about their business.

(I suspect one reason philosophers avoid these issues is the fear of having the rug pulled out from under them by physicists. Anything any philosopher wrote on time, for instance, probably sounded remarkably silly after Einstein revolutionized our concepts of space and time.)

Let's start with classical mechanics. (Quantum mechanics is a whole different ball of wax, and its interpretation remains very controversial.) The basic entities we work with are:
  • Space - the three-dimensional continuum that is the stage on which everything takes place.
  • Time - is what keeps everything from happening all at once.
  • Point particles -  the actors on the stage. They are idealized objects that have a location in space at a given time. 
  • Fields - extended entities that exist at every point in space. They are produced by (certain types of) particles and they exert influence on  (certain types of) particles.
(Of course, after special relativity we had to stop talking about space and time and start talking about a unified spacetime as the stage on which everything takes place. And after general relativity, we had to get used to curved spacetime. But the basic picture remains the same.)

We then proceed to write equations for how the particles and fields change in time. (Newton's laws for motion of particles, and Maxwell's equations for the changes of fields, for example.) Here the possibility of change is implicit in the fact of a time coordinate: there is no reason to assume that what obtains at one time also obtains at some other time. So, my first reaction to BI's question was to think that of course there's nothing like act and potency in physics. We just talk about change as a fundamental fact about the universe.

When I read Feser and his Aquinas quotes on actuality and potentiality, it seemed as though they were saying that the actual is real, and the potential was sort of real, too, as if it were sitting around somewhere waiting to be pushed into actuality. This seemed like a quaint, medieval idea that had no relevance to modern physics. But the more I thought about it, the more I began to realize that we actually have an approach in physics that very closely resembles this. That is the idea of a state space.

The state space is the collection of all possible states of the system. For a point particle, the state consists of its position and its velocity, or state of motion. For a single particle moving in one dimension, the state space consists of the two-dimensional plane (x,v). The particle's motion is then envisioned as a path in this plane. So, we do indeed think of the possible states of the particle as sort of existing "out there", and the (actual) particle as moving from one of these states to another.

So is Feser right - does modern physics presuppose the notions of actuality and possibility? I'm not convinced. For one thing, the whole notion of state space wasn't developed until the 19th century (by Hamilton, I think). And for another, the state space picture doesn't remove the need for a time coordinate. As I said before, the existence of time already allows for the possibility of change. And we can't even talk about the state space without the notion of velocity, which itself requires time. So it seems to me that the concepts of time and change are metaphysically prior to those of potentiality and state space.