Replies

Some folks have asked about the meaning of the word chaos in the quoted article. In the context used by the author of the paper referred to in the article, the word "chaos" has a very specific, technical meaning. This technical meaning of the word "chaos" is different from the regular usage in everyday speech - for instance the crappy chaos we see all around us caused by Obozo and the blithering idiots in the Congress who support/enable him is one usage of the word, but the way we use the word 'chaos' in physics is different.

In physics, the evolution of a physical system is determined by the solution(s) of an equation (or a system of equations) called the equation(s) of motion. For most interesting physical problems, the equations of motion are differential equations. In the case of those physical systems the evolution of which is determined by linear differential equations, the solutions to the equations, irrespective of whatever other characteristics they may or may not exhibit, do not exhibit the property of chaos. In those cases for which the equations of motion are nonlinear differential equations, it is sometimes the case that the solutions will exhibit the property of "chaos." This technical term means something specific: Associated to the equations of motion of a physical system is a mathematical "space" - the "place" in which the solutions to the equations evolve. This space is generically referred to as "phase space." Solutions of the equations of motion yield what are "trajectories" in this phase space. There is a particular type of technically-defined quantity, called the "Lyapunov exponent," associated to different such trajectories in phase space. For a given system of equations of motion, there is actually a set of Lyapunov exponents, called the Lyapunov spectrum. For this spectrum there is a largest (sometimes unique) Lyapunov exponent. Depending on the value of this largest Lyapunov exponent, the system is called "chaotic" or not. Specifically, if the largest Lyapunov exponent is a positive number, then the system is "chaotic" (most of the time - this is a very technical subject with a large number of qualifications that are not possible to go into in a blog.)

So much for a (partial) mathematical characterization. What does it mean qualitatively if a system has a largest Lyapunov exponent that is positive, and the system is thus (usually) chaotic? Roughly speaking, if a system is "chaotic" in the technical sense of physical dynamics, it means the following: if you let the system evolve under a given set of initial conditions, you will see a certain behavior: call this the "default behavior.". If you "nudge" the system a "little bit" (this roughly means that you only slightly alter the initial conditions on the solutions to the differential equations of motion), and if the system is chaotic, the result will be a BIG change in that default behavior. On the other hand, if the system is NOT chaotic, a "little nudge" will produce only a little change in the behavior of the solution. (Solutions describing non-chaotic systems, such as all solutions to systems described by linear differential equations of motion, do NOT behave this way: if you nudge things "a little," the solutions also only change "a little.") Another way of describing a system that exhibits chaos is to say that, in a chaotic system, two trajectories in phase space that are initially "very close" (this can be defined mathematically rigorously) together, will separate from each other exponentially quickly as a function of time in phase space.