The meme’s eye view of language

Abstract for Evolang IV Leipzig April 2004


Why did language evolve? The vast majority of theories assume that the ultimate beneficiary of language must be the individual human speaker or that speaker’s genes, with functions including increased cooperation, improved hunting skills, sharing of useful information, sexual selection and so on. I shall argue that it may, or may not, be the case that early language was of benefit to the people who spoke it. It certainly appears to benefit humans now, but this does not necessarily mean that this is why it evolved. I shall explore the alternative possibility that language evolved for the benefit of memes, not genes.

As soon as our hominid ancestors became capable of imitating sounds, a new replicator was let loose and memes (information copied between people with variation and selection) began to compete with genes. These memes then drove the genes to create ever better meme machines with brains that were not only larger but specifically better at copying the most successful sounds.

Basic principles of replicator theory may help explain how this process could give rise to language: successful memes are those of high fidelity, fecundity and longevity; copy-the-instruction is better than copy-the-product; digitisation increases fidelity; combining small arbitrary units increases fecundity, and so on.

This approach to language evolution may be tested e.g. by simulations and by experiments with imitating robots. It provides a dramatically different view of language – as a parasite turned symbiont, evolved at great cost to its host and its host’s planet, and now moving on to new hosts in the form of computers and the internet. I shall consider whether this view provides any advantages over traditional gene-based theories.