Some Thoughts About the Anthropic Principle

Kevin Langdon


Reading ``The Anthropic Principle,'' by George Gale, an article on this currently fashionable scientific subject published in Scientific American, I am struck by the faulty reasoning which is so often encountered when science turns abstract.

The anthropic principle, in its most basic form, holds that things are the way we observe them to be because they have to be that way to produce observers.

This is a little like saying that it's a remarkable coincidence that both you and your mother were present at your birth.

It is obvious that what we see is constrained by the conditions necessary for us to be here to see it, but the question of the necessity of those conditions for other reasons is independent of this fact and is of vital concern for science.

Science says that objects fall toward the ground and planets move in their orbits around the sun because of the same universal law of gravitation; the discovery of laws underlying seemingly disparate phenomena permits very powerful abstractions which are useful in a variety of practical ways.

Whenever we are able to understand the fundamental laws behind the appearances we see, the scope of our abstract vision is increased. If we can understand why the universe has evolved in this particular way, we may understand whether or not the presence of observers in it is accidental in a different sense.

I am saying that it is possible that the universe was created in order to produce observers, for some prior ontological reason, but that this is a proposition to be subjected to the same impartial scrutiny as any other in science, and moreover one which we presently do not have the means to verify.

This does not mean that it cannot be verified; science has found ways to answer questions that were previously considered unanswerable many times in its history, but we must not draw conclusions in advance of actual verification. Unfortunately, many proponents of the anthropic principle do just that.

For example, Gale claims that reasoning about the anthropic principle proceeds according to an approach which he contrasts with ``the deductive method,'' which ``begins by specifying the initial conditions of a physical system and the laws of nature that apply to it; the theory then predicts the subsequent state of the system.''

This idea contradicts the principle, in the philosophy of mind, that the partial understanding from which one approaches the truth does not influence what the truth turns out to be, in any defined sphere. (There are many people who do not agree with this principle, but I find it inescapable.) If this principle is correct, there is nothing wrong with beginning from a subsequent state and reasoning about a former state.

Indeed, we do this all the time in analyzing practical affairs in the world. I see that it's wet outside and conclude that it's rained.

This same misunderstanding is manifested in another passage (Robert Dicke formulated the anthropic principle in 1961):

Deductive or predictive logic proceeds from a fundamental assumption to a derived result: the future is deduced from the past. The temporal flow of Dicke's argument is in the opposite direction. He cites a present condition (man's existence) as the explanation of a phenomenon grounded in the past (the age of the universe). Clearly his result cannot be interpreted as a prediction, since it would be a prediction of the past on the basis of that past's own future.

This passage shows the lapse in reasoning very clearly. The science of astronomy, which cannot do physical experiments on its objects of study (except in a--so far--very limited way within our own solar system), can nonetheless make predictions in advance of particular observations which serve to test the models astronomers construct. If this were not the case, the models of the early universe of which Gale writes throughout this article would not exist as scientific theories.

There is much more in the article, including an examination of the ``many worlds'' interpretation of quantum mechanics. It is very much worth reading, but the reader must be cautious about going out on a metaphysical limb.


Essays and Reviews

Polymath Systems Home Page