Isolation Mismatch

You won’t find the term “isolation mismatch” in any scientific dictionary. Isolation mismatches occur when two complex adaptive systems cannot be merged after evolving in isolation from each other. It is a generalization of a concept that you might find in a scientific dictionary, “Dobzhansky-Muller incompatibilities,” which cause isolated biological populations to become separate species. When ported into the realm of culture, isolation mismatches might explain how cultures (not species) can diverge and become incompatible. It might also account for our disconcerting human tendencies towards xenophobia or fear of other human cultures. But let’s consider biology first, then the cultural analogy.

Formation of new biological species usually involves isolation and independent evolution of the two populations. As one toad population evolves through natural selection, each of its novel genes is tried out in many toads and will necessarily be selected to work together with the other genes in its population. Any genes causing within-population mismatches are weeded out. But, novel genes in one toad population are never tested with novel genes in another isolated toad population. Between-population mismatches will not be weeded out and will gradually accumulate, becoming apparent only when the populations later come into contact. 

This is the dominant model (though not only one) of how new species form. Interestingly, where much of evolution consists of fine-tuned adaptation, the formation of species in this manner is not directly adaptive. It is an accident of isolation, though such accidents will always happen given enough time. There is, however, an add-on mechanism called reinforcement that is not accidental. When two partially incompatible populations come together, selection may directly favor reduced interbreeding. Individuals will have fewer mismatches and more viable offspring if they preferentially mate with their own type. For example, females of two closely related species of spadefoot toads have evolved to prefer the male call of their own type only in areas where the two species overlap.

Similar isolation mismatches may occur in other complex adaptive systems. Let two initially compatible systems evolve separately for long enough and they will accrue mismatches. This includes cultural systems where what changes is the cultural equivalent of genes, which Richard Dawkins calls memes. For example, languages split, evolve independently, and become mutually unintelligible. Mixing diverged norms of social behavior can also cause mismatches. Acceptable behavior by man towards another man’s wife in one’s own culture might prove to be fatal in another culture.

We are all familiar with isolation mismatches in technology. We see them when we travel between countries that drive on opposite sides of the road or have railroads of different gauges or have electrical systems with different voltages and outlets. There is no logical reason for the mismatches; they just evolved incompatibly in different locations. We are also familiar with isolation mismatches resulting from computer system upgrades, causing some programs that worked fine with the old system to crash. 

Of course, software engineers can usually catch and correct such mismatches before a program’s release. But they can be defeated if they don’t know or don’t care about a dependent program, or if they may have incomplete understanding of the possible interactions in complex code. I suggest that throughout most of human cultural history, these two conditions often apply; cultures changed without concern about compatibility with other cultures, and people did not understand how new cultural traits will interact.

Consider the adoption of maize by Europeans from the Amerindians. The crop spread widely because of its high yield but it also caused pellagra, eventually traced to a deficiency of the vitamin niacin. The Amerindians suffered no such problem because they dehulled their corn by nixtamalization, a process involving soaking and cooking in alkaline solution. Nixtamalization also happens to prevent pellagra, possibly by releasing the niacin in maize from indigestible complexes. An 18th-century Italian maize farmer who dehulled by his own culture’s supposedly superior mechanical methods got a very harmful isolation mismatch. If he had also adopted nixtamalization, he would have been fine, but he had no way of knowing. Only being more conservative about adopting foreign traits would have saved him.

So imagine a history of thousands of years of semi-isolated human bands, each evolving its own numerous cultural adaptations. Individuals or groups that allow too much cultural exchange with different cultures would experience cultural mismatches and have decreased fitness. So a process parallel to reinforcement might be expected to occur. Selection could favor individuals or groups that avoid, shun, and perhaps even hate other cultures, much as the female spadefoot toads evolved to shun males of the other species.  Xenophobic individuals or groups would be successful and propagate any genes or memes underlying xenophobia. For the same reason, selection could favor the adoption of cultural or ethnic markers that make it clear who belongs to your group, as suggested by anthropologist Richard McElreath and colleagues.

Many questions remain to be addressed. Is xenophobia selected by genic or cultural selection? Is it individuals or groups that are selected? How much isolation is required?  Did prehistoric human groups frequently encounter other groups with sufficiently different cultures? When might selection favor the opposite of xenophobia, given that xenophobia can also lead to rejecting traits would have been advantageous?

It should be stressed that no explanation of xenophobia, including this one, provides any moral justification for it. Indeed understanding the roots of xenophobia might provide ways to mitigate it. The mismatch explanation is a relatively optimistic one compared to the hypothesis that xenophobia is a genetic adaptation based on competition between groups for resources. If correct, it tells us that the true objects of our evolutionary ire are certain cultural traits, not people. Moreover, even those traits work fine in their own context and, like software engineers, we may be able to figure out how to get them to work together.