Do Emergent Properties Exist?

Ecologists often debate the question of whether emergent properties, namely properties that exist only at a higher level of system organisation and cannot be deduced from the properties of the constituent components, really exist.

In the world of physics emergent properties are well known, especially in statistical physics. For over 100 years, at least since the days of James Clerk Maxwell, physicists have been concerned about the paradox of irreversibility and the direction of time.

The fundamental equations of classical physics are based on Newton's laws of motion, basically F=ma. Since a, the acceleration, is a second derivative, it has the same value if we replace the time variable t by -t, which means that we cannot tell whether time is running forwards or backwards. For example, if we take a movie of some physical process and run the movie backwards, what we see is physically reasonable.

But many processes are not irreversible, If we take a picture of a balloon being punctured and collapsing and run it backwards, the result is nonsense and looks impossible. More generally, how can we say that entropy increases (the second law of thermodynamics) when Newton's laws say that it can decrease as well as increase?

We can see this even on a very small scale. Although thermodynamics and statistical mechanics deal with vast numbers of molecules, we can study irreversibility on a billiard table. Billiard balls make up a very simple two-dimensional classical gas obeying Newton's laws. It is easy to "break" the balls, that is to say to scatter the initial triangular configuration of the "rack", but impossible to recover this neat pattern. A skilled player can bring two or three balls back into a neat configuration, but more than that is beyond the reach of human control. 

Developed and maintained by William Silvert.