Information Physics

1995


We’ve been thinking about physics in entirely the wrong way. Spacetime considerations dominate our thought processes about physics. What I have in mind is an information-dominated approach to physics.

I begin by asking, what is the universe? My answer is, "that which is knowable". In other words, we begin by defining physical reality in terms of information.

My next postulate is that the total amount of information in the universe is finite and constant. This springs from the second law of thermodynamics. If you could increase the total amount of information in the universe, then you could run heat engines forever on their own exhaust. The Uncertainty Principle is a localized version of this postulate.

Inext observe that time can be used to increase the amount of available information. Measure a system once and you’ve got some information; measure it again some time later and you can pool the two observations to yield even more information. The greater the time difference between the two observations, the more information you can extract by pooling the data. It’s like capital drawing interest and the analogy is much tighter than you might first imagine.

This observation implies that you could increase the total stock of information on a system without limit. This violates my second postulate. Which brings me to a fascinating conclusion:the universe acts in such a way as to foil such efforts. For example, when you bring the Uncertainty Principle to bear on the two-part experiment, it turns out that you really can’t get away with the trick, because the uncertainties in the original measurement will always make the second measurement more difficult to make accurately.

This gives us a much cleaner way of explaining the Big Bang. If you think in terms of an explosion lots of energy, expansion, violence things get very complicated. But think of it solely in terms of information. Take all the mass in the universe and concentrate it in a single point. That’s a high-information state; everything in the universe is precisely specified. If it remains in that state for even an instant, then the total amount of information about the universe is greatly increased; ergo, in order to conserve information, it has to start spreading out into other states.

We like to think of gradients in assessing physical behavior. Energy flows down a gradient. But information itself is the gradient; it does not flow. The information structure of the universe does indeed evolve as the universe expands. The information itself isn’t spreading out or equalizing; it’s acting in such a way as to remain constant. Think of a star not as a body of matter but as an expanding sphere of light with a massy core. This new system maintains constant information content. Place a planet inside the sphere, blocking part of the expansion, and you’ve got an information imbalance; translate some of that information into living systems and you’ve corrected the imbalance. Le Chatlier’s Principle may be more fundamental than we ever imagined.

This is not an idea; it is only the germ of an idea.