Many people are familiar with the concept of the ‘meme’: units of cultural material that are transmitted throughout populations and evolve in a manner somewhat analogous to genes. Richard Dawkins coined the term and the general idea , prompting some measure of scientific activity, including a journal , devoted to the study of memetics. However, memes and memetics never gained much traction amongst social scientists and philosophers, and ‘meme theory’ currently enjoys essentially no credibility as a scientific theory.
In this post, I will explain why that is, and I will point to some alternative, sounder approaches to thinking about and studying the way knowledge and practice diffuse and evolve throughout societies .
Evolutionary epistemology and social science seem to go well together. Theories spread, transform, grow, live and die – ‘are selected for’ – not just in science but in society as a whole. Nor is this pattern of diffusion, evolution, and selection restricted to substantive or propositional content (i.e. to claims about the world); music, language, food and fashion, technology all seem to fit this model throughout periods of history. If we want to understand why that is, we must answer two questions:
- What is the thing that is being transmitted/diffused, selected-for, and transformed? That is, what is the ‘genetic unit’ of cultural evolution?
- What mechanism(s) are responsible for cultural evolution? That is, what kinds of recurring processes lead to the units of cultural evolution spreading and changing?
Meme theorists seem to offer these answers:
- Memes are independent units of cultural information, such as ideas, behaviours, or theories, that move between human hosts and influence what those hosts do, thereby causing changes in their environment.
- Memes ‘leap from brain to brain’  by somehow generating imitation.
These answers, as I will explain, are not very good.
The notion of independent, self-replicating units of cultural data is both conceptually and empirically problematic. Conceptually, it appears to rest on dubious ontological foundations; that is, it seems to be a very strange kind of thing. Memes are not cognitive phenomena, according to Dennett, although they clearly can produce something cognitive (beliefs). They are contained within human beings, and so they are not social structures or systems, unless we conceive of structures and systems in highly reductionist terms. So what are they? Perhaps they are just a convenient shorthand for a bunch of other stuff, and are not meant to refer to something real? But if that’s the case, then (i) they are not analogous to genes, which we probably think are real and (ii) we only have reason to use the concept of a meme at all if it provides considerable empirical value.
It doesn’t, though. Provide empirical value, that is. As any anthropologist or sociologist will attest, culture isn’t made up of little, discrete bits of behaviour or knowledge. It is this big, inter-subjective, inter-related mess of interacting and continually changing practices, tastes, dispositions, and interpretations oriented around social life. It exists in holistic ways, with one particular bit of culture only making any sense when placed within the context of the larger whole. It’s not just that culture could be a ‘memeplex’ , but that culture is a web of symbols and meanings  surrounding us and making us who we are even as we continually recreate it through our actions . Hence dividing it into independent units deprives us of our ability to appreciate culture as something emergent, and completely ignores the way that culture is not only something that seems to dwell within us but also constitutes the actual social environment in which we live and act . Whatever empirical value we get from keeping the concept of the meme must be counter-posed to the enormous empirical value we lose by adopting a concept that is unsuited for appreciating vast and relevant parts of culture and social life.
Not only that, but the mechanisms of evolution proposed by meme theorists seem either trivial or absurd. Nothing simply leaps from brain to brain; people imitate other people due to processes of socialisation and influence. These can be such things as direct peer pressure, in-group solidarity, coercion, persuasion, observation, or adaption. To name a few possible candidates. These mechanisms range from the level of individual psychology to society-level structural influences, and thus do not appear to correspond to anything remotely similar to the ways in which genes engineer the machinary of their own reproduction; again, we must appreciate cultural or social evolution by taking into account emergent structures and systemic wholes, as well as their component parts. The reductionist approach of memetics just won’t do the conceptual and empirical job.
Of course, if we were left with no alternatives, we might decide that memetics is good enough. Luckily, though, we have alternatives. Better alternatives. So many alternatives, actually, that there is a robust debate among actual social scientists over where and when one alternative is better than another for a given problem or area of social life. For example, one of the most popular approaches is to conceive of culture as made up of symbols. Associated with hermeneutic theorists  such as Wilhelm Dilthey, Max Weber, Clifford Geertz, and Paul Ricoeur, and semiotic theorists  such as Ferdinand de Sasseure and Roland Barthes, this approach offers a much more helpful way of thinking about what culture is, ontologically or cognitively. Nevertheless, this approach makes it difficult to understand how culture spreads and changes. To address this, theorists such as Anthony Giddens, Pierre Bourdieu, Margaret Archer, Jeffrey Alexander, and Charles Tilly have treated the symbolic structures of culture as existing in a mutually-constituting relationship with the individual actions and practices of people as they go about interacting with one-another and living their lives. Practices concatenate or chain together in various ways to produce emergent changes at the structural or system level, which circle back to the level of individuals as their environment changes accordingly, in a dialectic process stretching back into history.
Drawing more explicitly from evolutionary theory, evolutionary epistemologists such as Karl Popper or Donald Davidson have suggested that theories spread and are selected-for in a way somewhat analogous to Darwinian evolution, where more empirically successful (and perhaps more accurate) theories win out over less successful ones. And pragmatist philosophers such as Charles Sanders Peirce and John Dewey have used evolutionary metaphors to produce highly influential metaphysical theories  and theories of mind and action , which have formed the basis for more recent attempts to theorise change and innovation based on something like spontaneous mutation or novel synthesis, such as by Hans Joas .
It may seem daunting to look at this long list of names and theoretical traditions in thinking about what culture is and how it changes, but are you really going to stick with memes out of laziness? This is all material that can be covered in an introductory course in sociology or anthropology with enough detail to make it possible to talk about culture or society without resorting to unhelpful or incoherent Darwinian metaphors. And if taking such a course is not feasible, buying and reading a textbook surely is. I’ll even take questions by email.
Tl;dr almost everything conveyed by the term ‘meme’ can be conveyed by the term ‘practice’, ‘approach’, or ‘tradition’, and what cannot be conveyed with those terms can be conveyed with other slightly more complex vocabulary associated with an actually credible theory in the social sciences.