A Critique of the Search for Extraterrestrial Intelligence
Can SETI Succeed? Not Likely.
(This originally appeared in The Bioastronomy News, vol. 7, no. 3, 1995.)
By Ernst Mayr
What is the chance of success in the search for extraterrestrial intelligence? The answer to this question depends on a series of probabilities. I have attempted to make a detailed analysis of this problem in a German publication (Mayr 1992) and shall attempt here to present in English the essential findings of this investigation. My methodology consists in asking a series of questions that narrow down the probability of success.
How Probable Is It That Life Exists Somewhere Else in the Universe?Even most skeptics of the SETI project will answer this question optimistically. Molecules that are necessary for the origin of life, such as amino acids and nucleic acids, have been identified in cosmic dust, together with other macromolecules, and so it would seem quite conceivable that life could originate elsewhere in the universe.
Some of the modern scenarios of the origin of life start out with even simpler molecules--a beginning that makes an independent origin of life even more probable. Such an independent origin of life, however, would presumably result in living entities that are drastically different from life on Earth.
Where Can One Expect To Find Such Life?Obviously, only on planets. Even though we have up to now secure knowledge only of the nine planets of our solar system, there is no reason to doubt that in all galaxies there must be millions if not billions of planets. The exact figure, for instance, for our own galaxy can only be guessed.
How Many of These Planets Would Have Been Suitable for the Origin of Life?There are evidently rather narrow constraints for the possibility of the origin and maintenance of life on a planet. There has to be a favorable average temperature; the seasonal variation should not be too extreme; the planet must have a suitable distance from its sun; it must have the appropriate mass so that its gravity can hold an atmosphere; this atmosphere must have the right chemical composition to support early life; it must have the necessary consistency to protect the new life against ultraviolet and other harmful radiations; and there must be water on such a planet. In other words, all environmental conditions must be suitable for the origin and maintenance of life.
One of the nine planets of our solar system had the right kind of mixture of these factors. This, surely, was a matter of chance. What fraction of planets in other solar systems will have an equally suitable combination of environmental factors? Would it be one in 10, or one in 100, or one in 1,000,000? Which figure you choose depends on your optimism. It is always difficult to extrapolate from a single instance. This figure, however, is of some importance when you are dealing with the limited number of planets that can be reached by any of the SETI projects.
What Percentage of Planets on Which Life Has Originated Will Produce Intelligent Life?Physicists, on the whole, will give a different answer to this question than biologists. Physicists still tend to think more deterministically than biologists. They tend to say, if life has originated somewhere, it will also develop intelligence in due time. The biologist, on the other hand, is impressed by the improbability of such a development.
Life originated on Earth about 3.8 billion years ago, but high intelligence did not develop until about half a million years ago. If Earth had been temporarily cooled down or heated up too much during these 3.8 billion years, intelligence would have never originated.
When answering this question, one must be aware of the fact that evolution never moves on a straight line toward an objective ("intelligence") as happens during a chemical process or as a result of a law of physics. Evolutionary pathways are highly complex and resemble more a tree with all of its branches and twigs.
After the origin of life, that is, 3.8 billion years ago, life on Earth consisted for 2 billion years only of simple prokaryotes, cells without an organized nucleus. These bacteria and their relatives developed surely 50 to 100 different (some perhaps very different) lineages, but, in this enormously long time, none of them led to intelligence. Owing to an astonishing, unique event that is even today only partially explained, about 1,800 million years ago the first eukaryote originated, a creature with a well organized nucleus and the other characteristics of "higher" organisms. From the rich world of the protists (consisting of only a single cell) there eventually originated three groups of multicellular organisms: fungi, plants and animals. But none of the millions of species of fungi and plants was able to produce intelligence.
The animals (Metazoa) branched out in the Precambrian and Cambrian time periods to about 60 to 80 lineages (phyla). Only a single one of them, that of the chordates, led eventually to genuine intelligence. The chordates are an old and well diversified group, but only one of its numerous lineages, that of the vertebrates, eventually produced intelligence. Among the vertebrates, a whole series of groups evolved--types of fishes, amphibians, reptiles, birds and mammals. Again only a single lineage, that of the mammals, led to high intelligence. The mammals had a long evolutionary history which began in the Triassic Period, more than 200 million years ago, but only in the latter part of the Tertiary Period-- that is, some 15 to 20 million years ago--did higher intelligence originate in one of the circa 24 orders of mammals.
The elaboration of the brain of the hominids began less than 3 million years ago, and that of the cortex of Homo sapiens occurred only about 300,000 years ago. Nothing demonstrates the improbability of the origin of high intelligence better than the millions of phyletic lineages that failed to achieve it.
How many species have existed since the origin of life? This figure is as much a matter of speculation as the number of planets in our galaxy. But if there are 30 million living species, and if the average life expectancy of a species is about 100,000 years, then one can postulate that there have been billions, perhaps as many as 50 billion species since the origin of life. Only one of these achieved the kind of intelligence needed to establish a civilization.
To provide exact figures is difficult because the range of variation both in the origination of species and in their life expectancy is so enormous. The widespread, populous species of long geological duration (millions of years), usually encountered by the paleontologist, are probably exceptional rather than typical.
Why Is High Intelligence So Rare?Adaptations that are favored by selection, such as eyes or bioluminescence, originate in evolution scores of times independently. High intelligence has originated only once, in human beings. I can think of only two possible reasons for this rarity. One is that high intelligence is not at all favored by natural selection, contrary to what we would expect. In fact, all the other kinds of living organisms, millions of species, get along fine without high intelligence.
The other possible reason for the rarity of intelligence is that it is extraordinarily difficult to acquire. Some grade of intelligence is found only among warm-blooded animals (birds and mammals), not surprisingly so because brains have extremely high energy requirements. But it is still a very big step from "some intelligence" to "high intelligence."
The hominid lineage separated from the chimpanzee lineage about 5 million years ago, but the big brain of modern man was acquired less than 300,000 years ago. As one scientist has suggested (Stanley 1992), it required complete emancipation from arboreal life to make the arms of the mothers available to carry the helpless babies during the final stages of brain growth. Thus, a large brain, permitting high intelligence, developed in less than the last 6 percent of the life on the hominid line. It seems that it requires a complex combination of rare, favorable circumstances to produce high intelligence (Mayr 1994).
How Much Intelligence Is Necessary To Produce a Civilization?As stated, rudiments of intelligence are found already among birds (ravens, parrots) and among non-hominid mammals (carnivores, porpoises, monkeys, apes and so forth), but none of these instances of intelligence has been sufficient to found a civilization.
Is Every Civilization Able To Send Signals into Space and To Receive Them?The answer quite clearly is no. In the last 10,000 years there have been at least 20 civilizations on Earth, from the Indus, the Sumerian, and other near Eastern civilizations, to Egypt, Greece, and the whole series of European civilizations, to the Mayas, Aztecs, and Incas, and to the various Chinese and Indian civilizations. Only one of these reached a level of technology that has enabled them to send signals into space and to receive them.
Would the Sense Organs of Extraterrestrial Beings Be Adapted To Receive Our Electronic Signals?This is by no means certain. Even on Earth many groups of animals are specialized for olfactory or other chemical stimuli and would not react to electronic signals. Neither plants nor fungi are able to receive electronic signals. Even if there were higher organisms on some planet, it would be rather improbable that they would have developed the same sense organs that we have.
How Long Is a Civilization Able To Receive Signals?All civilizations have only a short duration. I will try to emphasize the importance of this point by telling a little fable.
Let us assume that there were really intelligent beings on another planet in our galaxy. A billion years ago their astronomers discovered Earth and reached the conclusion that this planet might have the proper conditions to produce intelligence. To test this, they sent signals to Earth for a billion years without ever getting an answer. Finally, in the year 1800 (of our calendar) they decided they would send signals only for another 100 years. By the year 1900, no answer had been received, so they concluded that surely there was no intelligent life on Earth.
This shows that even if there were thousands of civilizations in the universe, the probability of a successful communication would be extremely slight because of the short duration of the "open window."
One must not forget that the range of SETI systems is very limited, reaching only part of our galaxy. The fact that there are a near infinite number of additional galaxies in the universe is irrelevant as far as SETI projects are concerned.
Conclusions: An Improbability of Astronomic DimensionsWhat conclusions must we draw from these considerations? No less than six of the eight conditions to be met for SETI success are highly improbable. When one multiplies these six improbabilities with each other, one reaches an improbability of astronomic dimensions.
Why are there nevertheless still proponents of
one looks at their qualifications, one finds that they are almost
astronomers, physicists and engineers. They are simply unaware of the
that the success of any SETI effort is not a matter of physical laws
engineering capabilities but essentially a matter of biological and
factors. These, quite obviously, have been entirely left out of the
of the possible success of any SETI project.
Ernst Mayr is the Alexander Agassiz Professor
Emeritus, of Harvard University. Considered one of the leading
of this century, the 91-year-old Mayr is the author of about 650 papers
and 20 books. He is known for his work in ornithology and systematics
as a leader in evolutionary biology, he has written about the
of species, overpopulation, biodiversity, and, most recently, SETI.
ReferencesErnst Mayr, "Lohnt sich die Suche nach extraterrestrischer Intelligenz [Is It Worthwhile To Search for Extraterrestrial Intelligence?]," Naturwissenschaftliche Rundschau, vol. 45, no. 7, 1992, pp. 264-266.
Ernst Mayr, "Does It Pay To Acquire High Intelligence?" Perspectives in Biology and Medicine, 1994, pp. 150-154.
S. Stanley, "An Ecological Theory for the
Origin of Homo,"
Paleobiology, vol. 18, 1992, pp. 237-257.
Conditions That Must Be Met Likelihood of theNow read Sagan's response to Mayr's critique.