Semiotic systems with duality of patterning and the issue of cultural replicators

Two major works in recent evolutionary biology have in different ways touched upon the issue of cultural replicators in language, namely Dawkins' Selfish Gene and Maynard Smith and Szathmáry's Major Transitions in Evolution. In the latter, the emergence of language is referred to as the last major transition in evolution (for the time being), a claim we argue to be derived from a crucial property of language, called Duality of Patterning. Prima facie, this property makes natural language look like a structural equivalent to DNA, and its peer in terms of expressive power. We will argue that, if one takes seriously Maynard Smith and Szathmáry's outlook and examines what has been proposed as linguistic replicators, amongst others phonemes and words, the analogy meme-gene becomes problematic. A key issue is the fact that genes and memes are assumed to carry and transmit information, while what has been described as the best candidate for replicatorhood in language, i.e. the phoneme, does by definition not carry meaning. We will argue that semiotic systems with Duality of Pattering (like natural languages) force us to reconsider either the analogy between replicators in the biological and the cultural domain, or what it is to be a replicator in linguistics.