UNDOING THE ANTHROPIC PRINCIPLE: THE SUITABLE CONDITIONS PRINCIPLE, OR, HOW A TOY DROVE THE DESIGN OF THE UNIVERSE

I have a toy that consists of a hollow, transparent sphere about 10 cm across into which you can place a number of 5mm diameter white beads. There are 60 of these beads, each imprinted with a number from 1 to 60. One button on the machine makes it stir them vigorously. A second button on the apparatus causes a scoop to move so that it picks up exactly one of the beads and presents it for viewing.

There is nothing to make the device choose one bead over another, and the beads are nearly identical — in any case, their differences are not such to make one more likely to be chosen than its mates. The device was made for games of chance, and it can be considered a good randomizer. My friend Julie ran the machine 100 times, each time with a full load of beads. Here is the output:

06, 31, 42, 29, 33, 05, 26, 01, 05, 28,
22, 32, 59, 59, 09, 57, 16, 46, 12, 13,
16, 25, 45, 14, 12, 38, 37, 51, 10, 34,
21, 10, 09, 35, 21, 23, 60, 09, 04, 33,
12, 32, 32, 13, 28, 11, 54, 46, 58, 33,
25, 07, 09, 02, 19, 60, 52, 23, 29, 48,
52, 35, 18, 13, 57, 45, 15, 24, 28, 24,
05, 59, 03, 03, 45, 22, 48, 53, 27, 18,
49, 01, 59, 37, 17, 51, 36, 33, 09, 41,
04, 43, 06, 39, 31, 60, 32, 06, 17, 41.

The probability p that we would get exactly this sequence is 1/(60100) or, approximately,

           p = 1.5 * 10^-178

Each turn of the machine takes about one second; to produce another sequence of this length therefore takes at least 100 seconds (actually a lot longer because of mixing time). To have a 50-50 chance of getting this same sequence again, you'd have to turn out sequences on the order of 10¹⁷⁸ times, which would take at least 10¹⁸⁰ seconds. The age of the universe is currently thought to be less than 20 billion years; it is less than 10¹⁸ seconds old. We can be pretty confident that even if I were to have the machine generate sets of 100 numbers for the next few billion years, the sequence shown above would not be likely to turn up again.

If the force of gravity had been ever so slightly different, if the coefficient of friction of the surface of the balls varied from what it was, if the charge on the moving electrons that drove the motor were barely different from what it is, or if any of a number of other physical laws or constants had been changed, the sequence would have come out differently. In other words, the result of this experiment depended on the nature of the universe, the interplay of its laws, the value of its constants, and its unique history. For example, if our planet were too hot, the plastic of which the machine was to be made would melt, and the experiment could not have been carried out. Can we therefore conclude that "all the particular laws and regularities in nature are united in a single principle law: Somewhere in the universe this machine must create this particular sequence"?

This conclusion is peculiar. It argues that because of a unique result to an experiment we can conclude that behind the laws of the universe is the need to evoke this result. Yet the conclusion is a paraphrase of the last sentence of an article which attempted to justify the anthropic principle. (Kuzin, A. "The anthropic principle." Quantum, January/February 1999)

The anthropic principle is the teleological belief that the universe was "tuned" to make life — in particular, human, conscious life — inevitable. Teleology is causality worked backwards, a later event influencing a prior event, or that somehow a physical system "knows" to head toward a "desired" outcome.

In his article, which is typical of those that argue for the anthropic principle, Kuzin began by saying, "The discovery of the vastness of the Universe has led to a fundamental problem: Does a human being mean anything in this immense Universe?" As a scientist, I cannot justify the term "immense". Sizes, of course, are relative. We can say that the sun is far larger than a rabbit. But is the sun large? It is small relative to our galaxy. Is the rabbit small? It is large relative to an atom.

The impetus to make the universe revolve around us has ancient roots. One of humankind's great achievements was to figure out that our earth is not the center of the universe, not even the center of our solar system. More recently we have come to understand that humans are not the end point of evolution, and are not a goal toward which previous life forms have striven. We should not make a similar mistake and base our judgment of the absolute size of things on the size of a human. In other words, there is no inherent measure of bigness, and the universe is neither immense nor tiny. Even if the universe is infinite (and it is not clear how we'd ever determine that), then we might wonder if there are still larger infinities, as there are in mathematics.

But that is a quibble. Kuzin's opening sentence poses the seemingly deeper question of whether a human being "means" anything. What are the implications of a human being having a "meaning"? In fact, what is the "meaning" of any object? For an object x to have a meaning implies that x encodes, is a symbol for, or represents, something else, say y. The meaning of the red dot on a map in the park means that you are standing at the place represented by the position of the red dot on the map. The red dot and its location are an object x, which represents the location of the map itself in the park. That location in the park is y. For the red dot to have meaning to us means that we can realize that y is represented by x. The map, the dot, and even the park have been created by us for us; without us, there is no meaning to the red dot. In the terms of Kuzin's article, we must ask, what entity is doing the interpreting of the "meaning" of a human being?

There is a lot more that can be said about the question of meaning (a library's worth of books on the subject have been written), but even this elementary view suffices to reveal the difficulties with the anthropic approach. Meanings stand outside things in themselves. When a male coyote marks its territory with urine, the urine (in itself a means of getting rid of metabolic waste products) becomes a symbol for "I live here; keep away" to other male coyotes. What is a human being a symbol of, and to whom is the signal directed? Kuzin does not specify, and without such a specification his words are empty. Nobel laureate Stephen Weinberg wrote, "The more the universe seems comprehensible, the more it also seems pointless" (Weinberg, S. "The first three minutes" 1977). Increasing knowledge does indeed make the universe seem pointless, that is, not having any intrinsic meaning. This is exactly what one should expect because things don't have any intrinsic meaning. In other words, the universe is pointless; which means only that the universe and human life are not symbols that have some external meaning because (as far as we know) there is no external agent to whom they represent something. Such agents have often been hypothesized but none have been demonstrated to exist.

Kuzin says that "the anthropic principle is the child of a mental experiment. In this experiment we assume some change in the natural laws and then see whether or not a human could exist in the modified world." Unfortunately for his thesis, my experiment with the ball-choosing machine and the sequence that it brought into existence is just as valid. There is no difference between choosing

(a) human beings (or, as he puts it, creatures with "freedom" and "intelligence"; terms he admits are not presently defined by the natural sciences) and

(b) my machine and its output sequence

as the "goal" of the "evolution of the universe."

We also have to be very careful when we use "evolution" in conjunction with physical systems. In biology, the term "evolution" describes the arrival and extinction of species via natural selection; a physical process based on the increased survivability of a reproducing creature that is better adapted to its local environment because of an inheritable characteristic. Changes in systems which do not have a selective mechanism, such as the secular changes in our galaxy, for example, are not evolution in the same sense. In astronomical examples, "evolution" means merely a change over time. In common parlance, we find a third usage. The everyday term "evolution" implies an improvement or advance in some regard. We must be very careful in using the term "evolution" in scientific discourse, and make sure which meaning of the word is intended. In particular, it is important to remember that in biology, evolution does not necessarily lead to increasing complexity or "higher" life forms, but only to a better fit to the current environment. In the case of many parasites and cave-dwelling species, they are simpler (a "lower" life form) than their ancestors -- for example, when the cave-dwelling descendants no longer have eyes).

When Kuzin speaks of the "evolution of the universe" toward a goal, he is not using either technical meaning, but the common parlance meaning, which applies to neither physical objects nor biological systems.

A large part of Kuzin's article, as with many other arguments in favor of the anthropic principle, is devoted to what is known about the sequence of events that make up the history of the universe (which he spoke of metaphorically as the "evolution" of the universe). For example, "we see that due to a long chain of ‘coincidences'... much carbon is produced [by stars], which is so important for life in the Universe" as if what seems to be a low-probability event can be explained only by its eventually being "needed" to create humans. Kuzin also points to the limited temperature range needed for organic reactions as something unlikely to occur. However, as my opening example shows, unlikelihood does not justify teleology. Once something has happened, the probability that it happened is 1. What Kuzin sees as essential for intelligent life reveals his anthropocentric bias and perhaps a lack of imagination as to the variety of possibilities for other forms of sentient beings. I think it foolishly narrow to assume that they would have to be something chemically similar to ourselves. The universe has always had a great capacity to overturn our prejudices and guesses, I see no reason to think that the same cannot happen with respect to possible life forms.

Many who support the anthropic principle base their argument on a number of physical constants whose values are currently unexplained. Kuzin observes that we have more equations (representing physical laws) than we have universal constants. This fact might point, as he believes, to some potential further unification of these laws. He does not stop to consider other alternatives, such as that there might be other essential constants; perhaps there are whole classes of phenomena as yet undiscovered. His observation leads him -- like other supporters of the anthropic principle -- to venture that "the very structure of the natural laws hides some extremely important principle. At present we don't know how to describe this in mathematical language... Everything we presently know is just the consequence of this main principle." But he has presented no basis for thinking there is such a principle. Then he makes another leap beyond logic, and continues, Fquantum As I have already noted, the final sentence can, with equal justification (or, rather, equal lack of justification) be replaced with, "All the particular laws and regularities in nature are united in a single principle law: Somewhere in the Universe a little plastic machine must create this particular sequence."

Put this way, the absurdity of Kuzin's argument for the "strong anthropic principle" becomes apparent. Brandon Carter (who first put forward the idea of an anthropic principle) posited what is called the "weak anthropic principle": "Our location in the Universe is necessarily privileged to the extent of being compatible with our existence as observers." which is accepted by many who reject the strong form.

There is something to object to, however. The problem is with the word "privileged" which I believe has unnecessary connotations (I'll get back to the word in a minute), and also with the way in which the statement is presented. Rejecting solipsism (as we usually do in science), I start by pointing out the obvious: We do exist and therefore conclude that the universe is such that our existence is possible. I call this "the anthropic instance". The anthropic instance is a special case of the "suitable conditions principle": If something exists, a set of conditions suitable for its existence also exists.

Carter's "anthropic principle" is merely an anthropic instance. To see this, rewrite Carter's statement to read more neutrally, "Our existence as observers is compatible with the conditions that obtain at our location in the universe." The phrase "as observers" can now be seen as irrelevant; in fact, we could say of a rock on Mars: Its existence as a rock is compatible with the conditions that obtain at its site on Mars. Or, to use Carter's language, "The rock's location in the universe is necessarily privileged to the extent of being compatible with its existence as a rock." There are places (e.g. the interior of a star) where the rock could not exist. A rock is hard to anthropomorphize, and the weak anthropic principle loses its emotional impact when applied to an inanimate object. There is nothing special about human existence per se as opposed to the existence of other objects.

Just what does "privileged" mean in a physical context? An empirically meaningful definition is what is required in a discussion of physics. If "privileged" means "special", that is, distinguishable from any other location in the universe, then in what way is our location special -- other than the happenstance that we are here and that the suitable conditions principle holds? As the example of the rock on Mars shows, every macroscopic object that exists is unique and privileged in that sense. There is a more subtle problem with the word "privilege" however. Privileges are usually conferred, implying an agency that can grant privileges. This connotation of "privilege" leads to the same problem as that of "meaning" above.

Even the weak anthropic principle appeals only to our emotions. It, too, vanishes into the trivial "suitable conditions principle" when analyzed. The New York Times (28 Oct 2003 p D4) records Dr. Weinberg as saying that "those who argue against it [the anthropic principle] recognize that it may be unavoidable." I must disagree, it is not unavoidable, it is untenable.

SUMMARY

The anthropic principle extols a pre-scientific viewpoint and defies logic. We start with the uncontroversial “suitable conditions principle” (if a macroscopic object exists, then a set of conditions suitable for its existence also exists ). It is clear that nothing can be logically concluded about the reasons for the conditions being as they are from the object's existence. Applying this principle to ourselves we obtained the "anthropic instance": We exist and therefore conditions suitable for us exist.

That's as far as we can get. We cannot logically conclude the converse as some supporters of the anthropic principle do (the necessity of our existence). Nor can we conclude that the conditions had to be such as to force our existence (which is the strong anthropic principle). The anthropic principle is merely a fashionable expression of the feeling that the universe revolves around us.