A culturally transmitted design can, in this way, have a free-floating rationale in exactly the same way a genetically transmitted design does. The boatbuilders and boat owners no more need to understand the reasons why their boats are symmetrical than the fruit-eating bear needs to understand his role in propagating wild apple trees when he defecates in the woods. Here we have the design of a human artifact—culturally, not genetically transmitted—without a human designer, without an author or inventor or even a knowing editor or critic.12 And the reason the process can work is exactly the same in human culture as it is in genetics: differential replication. When copies are made with variation, and some variations are in some tiny way “better†(just better enough so that more copies of them get made in the next batch), this will lead inexorably to the ratcheting process of design improvement Darwin called evolution by natural selection. What gets copied doesn’t have to be genes. It can be anything at all that meets the basic requirements of the Darwinian algorithm.13
This concept of cultural replicators—items that are copied over and over—has been given a name by Richard Dawkins (1976), who proposed to call them memes, a term that has recently been the focus of controversy. For the moment, I want to make a point that should be uncontroversial: cultural transmission can sometimes mimic genetic transmission, permitting competing variants to be copied at different rates, resulting in gradual revisions in features of those cultural items, and these revisions have no deliberate, foresighted authors. The most obvious, and well-researched, examples are natural languages. The Romance languages—French, Italian, Spanish, Portuguese, and a few other variants—all descend from Latin, preserving many of the basic features while revising others. Are these revisions adaptations? That is, are they in any sense improvements over their Latin ancestors in their environments? There is much to be said on this topic, and the “obvious†points tend to be simplistic and wrong, but at least this much is clear: once a shift starts to emerge in one locality, it generally behooves local people to go along with it, if they want to be understood. When in Rome, speak as the Romans do, or be ignored or misunderstood. Thus do idiosyncrasies in pronunciation, slang idioms, and other novelties “go to fixation,†as a geneticist would say, in a local language. And none of this is genetic. What is copied is a way of saying something, a behavior or routine.
The gradual transformations that turned Latin into French and Portuguese and other offspring languages were not intended, planned, foreseen, desired, commanded by anyone. On rare occasions, a particular local celebrity’s peculiar pronunciation of a word or sound may have caught on, a fad that eventually turned into a cliché and then into an established part of the local language, and in these instances we can plausibly identify the “Adam†or “Eve†at the root of that feature’s family tree. On even rarer occasions, individuals may set out to invent a word or a pronunciation and actually succeed in coining something that eventually enters the language, but in general, the changes that accumulate have no salient human authors, deliberate or inadvertent.
Folk art and folk music, folk medicine, and other products of such folk processes are often brilliantly adapted to quite advanced and specific purposes, but, however wonderful these fruits of cultural evolution are, we should resist the strong temptation to postulate some sort of mythic folk genius or mystical shared consciousness to explain them. These excellent designs often do owe some of their features to deliberate improvements by individuals along the way, but they can arise by exactly the same sort of blind, mechanical, foresightless sifting-and-duplicating process that has produced the exquisite design of organisms by natural selection, and in both cases the “judging†is harsh, austere, and unimaginative. Mother Nature is a philistine accountant who cares only about the immediate payoff in terms of differential replication, cutting no slack for promising candidates who can’t measure up to the contemporary competition. Indeed, the tin-eared and forgetful singer who can hardly carry a tune and forgets almost every song he hears but can remember this one memorable song contributes as much quality control to the folk process (by replicating this classic-in-the-making at the expense of all the competing songs) as the most gifted tunesmith.