Genes, Memes, and Cultural Heredity

Abstract

ion of their autocatalytic role) as replicator accounts try to do. A new “heterocatalytic” account of (biological) genes consilient with this is richly elaborated by Eva Naumann-Held (1998). Heterocatalytic gene-like things picked out by the GE criterion in biology would include some but not all genes, some things which are not genes, and most often, heterogeneous complexes of both. In some domains (like cultural evolution) gene-like things may be picked out by GE criteria where there is arguably nothing picked out by autocatalytic criteria, or more commonly, where autocatalysis is such a distributed and diffuse process that there seems no point to try to track compact lineages through it. [How does a scientific theory make a copy of itself? VERY indirectly!] Economist Kenneth Boulding’s used to say that: “A car is just an organism with an exceedingly complicated sex life.”13 Like a technological virus, it takes over a complex social structure and redirects the resources of a large fraction of it to reproduce more of its own kind. Indeed, our economic system has fostered an environment – a “culture dish” in which the invention, mutation, and expansion of such cultural viruses is encouraged, many environmentalists would say, until it has assumed cancerous proportions. (See also Sperber 1995). Griesemer’s account dovetails naturally with generative entrenchment, and was designed to do so. If an adaptive structure meeting (1–5) is even minimally adapted to its environment or task, then modifications of more deeply GE’d elements will have higher probabilities both of being maladaptive, and of being more seriously maladaptive. These probabilities become more extreme – in the simplest models, exponentially so – either for larger structures (e.g., if they grow by adding elements downstream), or as one looks to more deeply entrenched elements in a given structure. Either change increases the degree of “lock in” of entrenched elements.14 Crucial for cultural evolution is the larger number of ways we have for modulating or weakening generative entrenchment temporarily or for some purposes so we can (occasionally) make deeper modifications and get away with it. (Wimsatt 1987, 2000). [Most of these ways for facilitating deep modification are not applicable for, or present only in much weaker versions for biology. Thus, as we already know – for this and several other reasons – cultural evolution generally proceeds much faster than biological evolution.] Selection acts on the structure as a whole, so parts of an adaptive structure are inevitably coadapted to each other, as well as to different components of the environment. So larger changes in this structure will have to meet more design constraints. Fewer changes can do so. So ever larger changes tend to have a rapidly decreasing chance of being adaptive. Mutations in deeply GE’d elements will have large and diverse effects, and thus are much more likely