Man is the measure of all things.
A remarkable argument, which predates Hart's seminal analysis of the Fermi paradox, suggests mankind is probably alone. The argument relies on there being a number of "difficult steps" on the road to the development of a technologically advanced civilization. Examples of potentially "difficult steps" that we will discuss later include the genesis of life, the evolution of multicellular animals and the development of symbolic language. The precise details of the steps are, however, unimportant here. The argument simply requires there to be a number of critical yet unlikely steps on the road to intelligence. (The eminent evolutionary biologist Ernst Mayr once listed over a dozen of these "difficult" steps.178 Other scientists have suggested the number might be even greater, particularly if certain physical and astronomical coincidences are added to the list.) Some of the evolutionary steps we call "difficult" may not be hurdles at all. We think of a particular evolutionary step as difficult if it occurred only once in Earth's history; but some steps probably could be taken only once — the competition they stimulated would have made a second occurrence redundant. On the other hand, some steps may have been genuinely improbable. For example, if a particular critical step required several otherwise worthless mutations to take place at the same time, then it makes sense to regard the step as a fluke.
Now consider a remarkable coincidence, which lies at the heart of the argument below.
On the one hand, the lifetime of our Sun is about 10 billion years. The period over which it can sustain life-bearing planets may be less than this — some astronomers believe the future evolution of the Sun will cause Earth to become uninhabitable in another 1 or 2 billion years, so the "useful" lifetime of the Sun could be as little as 6 or 7 billion years. On the other hand, Homo sapiens arrived on the scene when the Sun was about 4.5 billion years old. These two timescales — the lifetime of the Sun and the time for the emergence of intelligent life to appear around the Sun — are certainly within a factor of 2 of each other, and could even be within a factor of 1.3 of each other. The near equality of these timescales is really quite incredible. The two timescales are determined by factors that, either individually or in combination, would seem to have nothing to do with one another. The Sun's lifetime is determined by a combination of gravitational and nuclear factors, while a combination of chemical, biological and evolutionary factors determines the time of emergence of intelligent life. We live in a Universe in which timescales span a vast range: many subatomic pro cesses occur on timescales as short as 10~10 seconds, while many astronomical processes occur on timescales as long as 1015 seconds. The typical times of certain other processes are even more extreme. The likelihood that two completely independent timescales have almost the same value is remote. How can we explain this observation without resorting to coincidence?
One solution would be if the evolutionary timescale is much smaller than 4.5 billion years. Suppose the typical time for the evolution of intelligent life on an Earth-like planet is just 1 million years. The coincidence of timescales would lessen — but at the expense of making the probability of mankind's recent emergence vanishingly small. After all, if we could have emerged just 1 million years after the Earth cooled, then why do we not observe the Earth to be 1 million years old? At the very least, why do we not observe it to be 2 million years old, or 3, or 4? Why did it take 4.5 billion years for us to appear? This is not a good solution.
The other solution requires the evolutionary timescale to be much longer than 4.5 billion years. This accords with Mayr's suggestion of a number of difficult steps in the development of intelligence — "difficult" in this sense meaning that, on a given viable planet, the typical time for a step to occur is long (perhaps longer than the present age of the Universe). If several difficult steps must be taken, then we would not expect to be here at all!
Most people, upon hearing this second solution, dismiss it on the same grounds as the first solution: the probability of mankind emerging recently is small. But the two situations are not equivalent.
Consider the ensemble of all possible universes. (Whether you consider these universes as somehow "real" or as some sort of mathematical idealization is up to you.) In some universes, unlikely things will occur; a chain of improbable events will happen. In some universes, due to the blind workings of chance, the set of difficult steps leading to intelligence will happen. And it is precisely such a universe an intelligent species will observe — with themselves in it. In other words, we can ignore the possible universes in which we do not exist — since by definition they do not exist for us. We must observe those universes in which the difficult steps have occurred and led to us. Now we can ask: Of all the universes that exist for us, when are we most likely to emerge, given that we can only emerge some time in the 10-billion-year total lifetime of the Sun? (Or, if it happens to be the case, the 6- to 7-billion-year useful lifetime of the Sun?) A simple calculation shows that if there are 12 difficult steps, then the most likely time of emergence is after 94% of the star's available lifetime has passed.
Our observations seem to be consistent with the results of this simple calculation. If the Sun were able to sustain life on Earth for 10 billion years, then mankind emerged after roughly 50% of the time available had elapsed. However, if, as some astronomers believe, the Sun can sustain life for only another billion years or so, then mankind emerged after roughly 83% of the time available. This is impressively close to the expected time of arrival.
The Most Probable Time of Emergence of a Communicating Civilization
Suppose there are n difficult steps on the road to the development of a civilization capable of interstellar communication. And suppose these steps must take place over the lifetime L (in years) of a star. A straightforward calculation shows that the most probable time of emergence of a communicating civilization is given by the expression L/(21/n). If there are a dozen difficult steps, so n = 12, then the most probable time of emergence is 0.94L. The calculation does not determine exactly when an intelligent species will emerge; just that the median time of emergence, if there are 12 difficult steps to negotiate, is 94% of the star's lifetime.
Finally, we come to the key point. Merely because we have selected universes in which we exist (and how could we select any other type of universe?), we cannot infer that other intelligent species exist. We have to be here because we observe ourselves to be here; but the existence of aliens must contend with probabilities, and the odds are not good. Another calculation makes this clear. If there are a dozen difficult steps to negotiate on the road to high intelligence, then even under generous assumptions there is only one chance in a million billion of there being another intelligent species in our whole Universe. No wonder we do not observe them!
Suppose there are n difficult steps on the road to intelligence and each step typically requires d years to occur. Furthermore, suppose there are p viable planets, each of which could have supported life for t years. The number of intelligent species out there is given by the expression p x [t/(n x d)]n. Let us be generous and suppose every star in every galaxy possesses a viable planet; so p « 1022. Let us be even more generous and suppose every planet has been viable for about the age of the Universe, so t « 1010 years. However, d must be long: that, after all, is what makes the step difficult. So let us suppose d « 1012 years — 100 times the age of the Universe. Finally, let us suppose as before there are a dozen difficult steps, so n = 12. If we plug these numbers into the expression above, we find the number of intelligent species out there is 10~15.
This type of argument for the non-existence of ETCs was first presented by Brandon Carter.179 It is known as an anthropic argument. (We have met anthropic ideas before in this book: the doomsday argument of Gott and Hart's suggestion regarding the improbability of life's genesis have anthropic overtones. We will meet other examples.) Carter's use of the term "anthropic" was perhaps unfortunate, since it implies mankind is somehow necessary. All that is needed for the argument to work is that intelligent observers — any intelligent observers — self-select their Universe. However, in this Universe it is we who make the observations.
The status of anthropic reasoning in science is contentious. Some view it as an abdication of the scientists' responsibility to provide explanations. For example, Smolin's idea of natural selection acting on whole universes (see page 57) is an attempt to move away from anthropic reasoning. Nevertheless, many respectable scientists have employed anthropic ideas in an attempt to explain several features of the Universe that seem to be "just right" for the evolution of life; if certain physical constants possessed only slightly different values, then we would not be here. Stars would not shine, or the Universe would have collapsed in on itself in a fraction of a second, or heavy elements could not form, and so on. The fact of our existence can perhaps, in some way, make sense of these observations. (But I think one can equally argue that these "explanations" are essentially trivial.)
There are several types of anthropic reasoning, corresponding to several anthropic principles each with different shades of meaning. According to Carter, the weak anthropic principle (WAP) is that "what we can expect to observe must be restricted by the conditions necessary for our presence as observers." The WAP seems almost tautologous. The strong anthropic principle (SAP), on the other hand, is more contentious: "the Universe (and hence the fundamental parameters on which it depends) must be such as to admit the creation of observers within it at some stage." Barrow and Tipler, in a classic book, also discuss the final anthropic principle (FAP), which they define as "intelligent information-processing must come into existence in the Universe and, once it comes into existence, it will never die out."180 The mathematician Martin Gardner, in his inimitable way, calls this latter version the completely ridiculous anthropic principle (crap).
It is interesting that Tipler expanded upon the notion of the FAP in a book entitled The Physics of Immortality.181 He considered the far future of the Universe, and was lead to a concept not unlike Teilhard de Chardin's Omega Point. His work showed that, if the Universe collapses in a Big Crunch, then a future intelligence would find it possible to perform an infinite number of computations. Every being who ever lived could be "resurrected" as a computational simulation. According to his interpretation of the FAP, the Universe must be such that it allows this infinite amount of information processing. Now, although Tipler's ideas were attacked as being altogether too speculative (and too overtly religious), his hypothesis at least had the virtue of being falsifiable. He made a definite, testable prediction: the Universe is closed and will collapse on itself. Recent observations, however, seem to indicate that the Universe is not only open, it is expanding more rapidly as it ages. Tipler, it appears, was wrong; his interpretation of the FAP seems disproven. Perhaps one day soon we will discover signals from extraterrestrials, or even receive a visit from them. Such a discovery would cast in doubt the WAP and SAP. I leave the reader to decide whether such a discovery is probable.
Was this article helpful?