Breaking the Spell: Religion as a Natural Phenomenon

Darwin himself proposed words as an example: “The survival or preservation of certain favoured words in the struggle for existence is natural selection” (Descent of Man, 1871, p. 61). Billions of words are uttered (or inscribed) every day, and almost all of them are replicas—in a sense to be discussed below—of earlier words perceived by their utterers. Replication is not perfect, and there are many opportunities for variation or mutation in pronunciation, inflection, or meaning (or spelling, in the case of written words). Moreover, words are roughly segregated into lineages of replication chains; for instance, we can trace a word’s descendants from Latin to French to Cajun. Words compete for airtime and print space in many media, with words going obsolete and dropping out of the word pool, while other words spring up and flourish. We discover conTROVersy going to fixation in some regions and CONtroversy going to fixation in others, while the original meaning of “begs the question” is supplanted in some quarters by a variant. The detectable historical changes in languages have been studied from one Darwinian perspective or another since Darwin’s own day, and a great deal is known about patterns of replication, variation, and competition in the processes that have yielded the diverse languages of today. Indeed, some of the investigative methods of modern evolutionary biology, in bio-informatics, for instance, are themselves descended from pre-Darwinian researches conducted by paleographers and other early scholars of historical linguistics. As Darwin noted, “The formation of different languages and of distinct species, and the proofs that both have been developed through a gradual process, are curiously the same” (1871, p. 59).

Words, and the languages they populate, are not the only culturally transmitted variants that have been proposed, however. Other human acts and practices that spread by imitation have been identified as potential replicators, as have some of the habits of nonhuman animals. The physical substrates of these media are various indeed, including sounds and all manner of visible, tangible patterns in the behavior of the vector organisms. Moreover, behaviors often produce artifacts (paths, shelters, tools, weapons,…signs or symbols) that may serve as better exemplars for the purposes of replication than the behaviors that produce them, being relatively stable over time, and hence in some regards easier to copy, as well as being independently movable and storeable—like seeds in this regard. One human artifact, the computer, with its prolific copying ability, has recently provided a distinctly new substrate, in which both deliberate and inadvertent experiments in artificial evolution are now burgeoning, taking advantage especially of the emergence of gigantic networks of linked computers that permit the swift dispersal of propagules made of nothing but bits of information. These computer viruses are simply sequences of binary digits that can have an effect on their own replication. Like macromolecular viruses, they travel light, being nothing more than information packets including a phenotypic overcoat that tends to gain them access to replication machinery wherever they encounter it. And, finally, researchers in the new field of Artificial Life aspire to generate both virtual (simulated, abstract) and real (robotic) self-replicating agents that can take advantage of evolutionary algorithms to explore the adaptive landscapes they are situated in, generating improved designs by processes that meet the three defining conditions while differing from carbon-based life-forms in striking ways. While at first glance these phenomena may appear to be only models of evolving entities, thriving in modeled environments, the boundary between an abstract demonstration and an application in the real world is more easily crossed by these evolutionary phenomena than by others, precisely because of the substrate-neutrality of the underlying evolutionary algorithms. Artificial self-replicators can escape from their original environments on researchers’ computers and take on a “life” of their own in the rich new medium of the Internet.