Do Locusts Understand Prime Numbers?
Dennett uses a proposed explanation of cicadas’ 13- and 17-year cycles to distinguish exploiting a mathematical property from understanding it. Natural selection need not perform arithmetic for a prime-numbered cycle to offer an advantage.
THE IDEA
41. DO LOCUSTS UNDERSTAND PRIME NUMBERS?
Just to drive home the point about the independence of numbers and numerals, free-floating rationales and represented reasoning, consider the case of the seventeen-year locusts. In 1977, Stephen Jay Gould wrote an insightful and admiring chapter about the curious fact that cicadas (such as the “seventeen-year locust”) have reproductive cycles with lengths, in years, equal to prime numbers—thirteen years, or seventeen, but never fifteen or sixteen, for instance. “As evolutionists,” Gould (p. 99) says, “we seek answers to the question, why. Why, in particular, should such striking synchronicity evolve, and why should the period between episodes of sexual reproduction be so long?” The answer—which makes beautiful sense, in retrospect—is that by having a large prime number of years between appearances, the cicadas minimize the likelihood of being discovered and later tracked as a predictable feast by predators who themselves show up every two years, or three years, or five years. If the cicadas had a periodicity of, say, sixteen years, then they would be a rare treat for predators who showed up every year, but a more reliable source of food for predators who showed up every two or four years, and an even-money gamble for predators who got in phase with them on an eight-year schedule. If the length of their reproductive period is not a multiple of any lower number, however, they are not worth “trying” to track for any species that isn’t lucky enough to have exactly the same periodicity (or some multiple of it—the mythical Thirty-Four-Year Locust-Muncher would be in fat city).
It should be clear that the soundness of this explanation (which may not yet be established) does not depend on any hypothesis suggesting that locusts understand arithmetic, let alone prime numbers. Nor does it depend on the process of natural selection understanding prime numbers. The mindless, uncomprehending process of natural selection can exploit this important property of some numbers without having to understand it at all. For another example: neither bees nor Mother Nature need to understand the geometry that declares the hexagon the ideal shape for cells in a honeycomb. Many more examples of evolution’s mathematical competence-without-comprehension could be cited.
When to use it
Understanding the evolution of life cycles in periodical cicadas and other species.
Explaining how adaptive traits can be the result of selection pressures even without conscious design.
Teaching the concept of mathematical properties being exploited by natural processes.
Analyzing predator-prey dynamics in terms of periodicities and synchronization.
FROM THEORY TO EVERYDAY LIFE
In practice
Seventeen-Year Locusts
Periodical cicadas have cycles such as 13 or 17 years. In the proposed explanation, this reduces regular overlap with hypothetical predators on cycles of 2, 3, or 5 years. Dennett explicitly notes that the explanation may not be established; his point is that such an advantage would not require the cicadas to understand primes.
Bee Honeycomb Hexagons
Neither bees nor evolution need to understand geometry to exploit the hexagon as the ideal shape for honeycomb cells, demonstrating competence without comprehension.
A MOMENT TO REFLECT
Put it to work.
Can you think of other biological phenomena where natural selection exploits mathematical properties without understanding them?
How does this example illustrate the difference between 'knowing' a fact and 'using' a fact in evolution?
Why might it be misleading to say that nature 'understands' prime numbers?
References
- Chapter on cicada cycles (likely from Gould's book) — Stephen Jay Gould (1977)
CONTINUE THE CONVERSATION
Source texts
- Intuition Pumps and Other Tools for Thinking