The assumption "life-supporting planets elsewhere must entail alien intelligence" is not confined to UFO enthusiasts. We have already discussed the probability gulf between a potentially life-supporting planet and the emergence of multicellularity. If there are Earth-like planets, if they have characteristics compatible with life, if life chemically similar to ours ever began on them and if that life evolved with symbiosis-driven complexity increase, then maybe multicellular organisms evolved elsewhere in the universe. No step in this argument follows with great likelihood from the previous one. But the steps that follow are even less likely.
Let us take "intelligence" to mean an ability to respond to stimuli in novel (not pre-programmed) ways in order to solve real-world problems effectively. This behavioural definition makes no presumptions about internal processing of data by the organism. We shall use the more specific phrase "human intelligence" to signify a capacity (a) for detailed internal modelling of the perceived world and (b) to pass on information and material to future generations by non-biological means. This two-part definition is not behavioural. It embraces both human mental powers - our ability to relate what we see, hear and touch to stored memories via language and mental pictures, and culture - the passing on of tools, techniques, knowledge and social form to our children by teaching them, rather by genes.
We shall discuss this topic further in the remaining chapters, but two points are immediately obvious. First, human intelligence presupposes intelligence (e.g. culture is primarily a store of once-novel problem-solving techniques); but intelligence does not imply the inevitability of human intelligence. An organism might have novel problem-solving capability without anything akin to human mentation or culture. Second, when people talk about intelligence on alien worlds, what they mean, in our terminology, is human intelligence: language, culture and mental powers. "Human intelligence" in our sense of the phrase need not be confined to Homo sapiens. In some respects it might, for instance, be shared by other apes. But the phrase is convenient; it signifies the sort of "intelligence" that humans have, without implying that it is necessarily unique to our species.
39 We use this word for convenience to mean any set of qualities that can distinguish "intelligent behaviour" from reflexive, pre-programmed or invariant responses to stimuli. It could be misleading; the set of qualities in question need not be the same in all animals, or always the same in the same individual or species. To suppose that "intelligence" denotes any thing-in-itself is an error. Here, we use the word as shorthand for a variable, open-ended set of descriptions of behaviour.
With all this in mind, let us consider the likelihood that a species with human intelligence will evolve, granted a planet with a thriving population of multicellular eukaryotes or their equivalent. On Earth it took roughly half the sun's expected life-span for human intelligence to emerge. How much more quickly could it have happened? It is hard to imagine human intelligence evolving in as little as - say - one tenth of the time it has taken ours, considering the number of evolutionary steps involved and their individual probabilities. So the statement "evolution of human intelligence takes a significant fraction of a star's life-span" could be generally true. This implies that although the universe might contain human intelligences with longer histories than ours, there cannot be any with vastly longer histories.
The fact that our sun is only one-third to one-half the age of the universe is not particularly relevant here. Life-bearing planets, indeed planets of any sort, could only come into being around second-generation stars like the sun; that is, stars made from the remains of stars that have already burned themselves out. The elements necessary for life - carbon, nitrogen, oxygen, phosphorus and heavier elements such as iron - are only made in stars nearing the ends of their lives. Therefore at least one generation of stars had to form, exist for a few thousand million years and finally explode before second generation stars with planetary discs could be formed and the universe could begin to bear life. The notion of a human intelligence that has survived for thousands of millions of years is therefore nonsensical, even if we ignore the fact that all species become extinct in much shorter times than that.
These arguments place general limits on the likelihood of human intelligence on other planets, but they do not rule it out. However, there are further arguments that cast serious doubt on the possibility. Suppose multicellular eukaryotes evolve on a planet several thousand million years before the star's intensifying radiation sterilises it - as happened on Earth. Would human intelligence necessarily, or probably, appear on that planet?
For organisms to behave "intelligently" in our sense they must be animallike - able to move around to obtain food. How likely are animal-like creatures to evolve from primitive multicellular organisms? The probable answer is "not very". The time interval between the first Ediacara fauna and the first (Cambrian) animals on Earth was roughly the same as the time interval (around 600 million years) between the first animals and the present day. In other words, it did not happen quickly. This suggests that it was unlikely. However, suppose animal-like creatures do evolve. Will some sort of intelligence appear? This step is more probable: the ability to cope effectively with novel situations would surely be advantageous in any world of animals. But a major hurdle follows: given intelligence, would human intelligence evolve?
Gould and others would ask: why should it? Dinosaurs of little brain thrived for 140 million years. Mammals have "ruled the Earth" for less than half that time. Modern humans have existed for less than 1% of the mammalian era. Where was intelligence, let alone human intelligence, before the dinosaurs? All the dominant species of the Earth's past survived without anything akin to human intelligence. Even if we accept Kauffman's view of evolution rather than Gould's, it is very hard to see human intelligence as one of a choice of likely patterns. Ernst Mayr pointed out that during animal evolution, vision of some kind has evolved no fewer than twenty separate times, flight four separate times and human intelligence once. Vision has great survival value for animals; flight has considerable survival value for some animals - hence the instances of "convergent evolution". But what is the survival value of human intelligence (as opposed to intelligence in general)? It cannot be very great or it would have evolved more than once. Human intelligence seems to be a freak phenomenon, very unlikely to be repeated in any alien evolutionary system.
In summary: given multicellular organisms, animals seem unlikely to evolve. If animals do evolve, then some form of intelligence is quite likely. But given intelligent animals, human intelligence is extremely unlikely. And as we reasoned earlier, the probability that multicellular organisms exist anywhere else in the universe is not very high to begin with. Reasoning from the evidence available to us, therefore, we are forced to conclude that human intelligence is an extraordinary and probably unique freak of nature, so there is almost certainly no human intelligence anywhere in the universe except on our own planet.
Was this article helpful?