Breaking the Spell: Religion as a Natural Phenomenon

There is no single proprietary code, parallel to the four-element code of DNA, that can be used to anchor meme-identity the way gene-identity can be anchored for most practical purposes. This is an important difference, but one of degree. If the current trend of language extinctions continues at its present pace, in the not-so-distant future every person on earth will speak the same language, and it will then be difficult to resist the temptation (which should still be resisted!) to identify memes with their (now practically unique) verbal labels. But so long as there are multiple languages, to say nothing of the multiple media in which nonlinguistic cultural items can be replicated, we are better off to keep strictly to the abstract, code-neutral understanding of a meme as a “package of information,” bearing in mind that, for high-fidelity replication to occur, there must always be some “code” or other. Codes play a crucial role in all systems of high-fidelity replication, since they provide finite, practical sets of norms against which relatively mindless editing or proofreading can be done. But even in the clearest cases of codes, there are often multiple levels of norms. Suppose Tommy writes the letters “SePERaTE” on the blackboard, and Billy “copies” it by writing “seperate.” Is this really copying? The normalization to all lowercase letters shows that Billy is not slavishly copying Tommy’s chalkmarks but, rather, being triggered to execute a series of canonical, normalized acts: make an “s,” make an “e,” etc. It is thanks to these letter-norms that Billy can “copy” Tommy’s word at all. But he does copy Tommy’s spelling error, unlike Molly, who “copies” Tommy by writing “separate,” responding to a higher norm, at the level of word spelling. Sally then goes a step higher, “copying” the phrase “separate butt equal”—all words in good standing in the dictionary—as “separate but equal,” responding to a recognized norm at the phrase level. Can we go higher? Yes. Anybody who, when “copying” the line in the recipe “Separate three eggs and beat the yolks until they form stiff white cones,” would replace “yolks” with “whites,” knows enough about cooking to recognize the error and correct it. Above spelling and syntactic norms are a host of semantic norms as well.

Norms can both hinder and help replication. The anthropologist Dan Sperber (2000) has distinguished copying from what he calls “triggered production” and has noted that in cultural transmission “the information provided by the stimulus is complemented with information already in the system.” This complementing tends to absorb mutations instead of passing them on. Evolution depends on the existence of mutations that can survive the proofreading processes of replication intact, but it does not specify the level at which this survival must occur. A brilliant cooking innovation might indeed get corrected away by an all-too-knowing chef in the course of passing on the recipe, but other “errors” might get through and replicate indefinitely. Meanwhile, the correction of other varieties of noise at other levels, responding to spelling norms or others, must be ongoing, in order to keep the copying process faithful enough so that multiple exemplars of each innovation can be tested against the environment. As Williams puts it, “A given package of information (codex) must proliferate faster than it changes, so as to produce a genealogy recognizable by some diagnostic effects” (1992, p. 13). Recognizable, that is, to the unfocused, independently varying environment, so it can yield probabilistic verdicts of natural selection that have some likelihood of identifying adaptations of projectible fitness.

Just how big or small can a meme be? A single musical tone is not a meme, but a memorable melody is. Is a symphony a single meme or is it a system of memes? A parallel question can be asked about genes, of course. No single nucleotide or codon is a gene. How many notes or letters or codons does it take? The answer in both cases tolerates blurred boundaries: a meme, or a gene, must be large enough to carry information worth copying. There is no fixed measure of this, but the bountiful system of case law on copyright and patent infringements indicates that verdicts on particular cases form a relatively trustworthy equilibrium that is stable enough for most purposes.