Memes

A meme (from Greek mimeme, imitation) is the unit of cultural transmission proposed by Richard Dawkins in The Selfish Gene (1976) as the cultural analogue of the gene. Where genes are units of biological inheritance that replicate, mutate, and compete in the gene pool of a species, memes are units of informational inheritance — beliefs, behaviors, techniques, slogans, melodies, rituals — that replicate, mutate, and compete in the "infosphere" of human minds and institutions. The concept belongs to memetics, the systematic study of culture through the lens of Darwinian evolution.

The meme was introduced as a provocation: to show that Darwinian logic applies wherever there is replication, variation, and differential selection — and that cultural evolution satisfies all three conditions. It is not a metaphor. Dawkins claimed the meme is a genuine replicator whose evolution is subject to exactly the same logic as genetic evolution, only with brains rather than chromosomes as the replication medium, and imitation rather than copying as the transmission mechanism.

What Counts as a Meme

Dawkins' examples include tunes, ideas, catchphrases, fashions, techniques, and religions. What makes something a meme is not its content but its replicative structure: it must be capable of being copied from one mind to another, with sufficient fidelity that distinctive features are preserved across multiple generations of copying, and with enough variation that selection can operate. The tune "Happy Birthday" is a meme. The use of "paradigm shift" in corporate strategy presentations is a meme. The cross as a visual symbol of Christianity is a meme. The belief that vaccines cause autism is a meme.

This breadth is both the concept's strength and its vulnerability. A unit of cultural analysis that encompasses tunes, beliefs, visual symbols, and behavioral routines is either genuinely general or hopelessly under-specified. Critics — including Dan Sperber, David Hull, Kim Sterelny, and the philosopher of biology Eva Jablonka — have argued that memes lack the individuation criteria that make genes tractable units of selection. Genes have physical identity (a sequence of nucleotides); memes have only relational identity (being the same tune, the same idea, the same practice). This relational identity is determined by the interpretive community that uses the meme, which means that meme identity cannot be specified independently of cultural context. The unit of selection is not well-defined.

Replication and the Fidelity Problem

Gene replication achieves high fidelity through the physical complementarity of DNA base-pairing, with error correction built into the cellular machinery. Meme replication achieves its fidelity — such as it is — through imitation, instruction, and social enforcement. The fidelity varies enormously and is often low by genetic standards.

Dan Sperber's epidemiology of representations (1996) is the strongest competing framework. On Sperber's account, cultural items are not copied at all — they are reconstructed at each transmission, guided by cognitive attractors (universal tendencies of human inference and memory) and by contextual interpretation. What spreads culturally is not the meme itself but the attractor to which diverse individual reconstructions converge. The spreading unit is not a discrete particle but a basin in cognitive space.

This is a deep objection. If cultural transmission works by reconstruction-toward-attractor rather than by copying, then the meme-as-replicator is a category error: it applies a copying model to a process that is not copying. Dennett's response — that virtual patterns can be genuine replicators without requiring physical token identity — is philosophically sophisticated but does not resolve the empirical question of whether cultural transmission is better described by copying or by reconstruction.

Memetic Fitness: What Makes Memes Spread

Why do some memes spread and others die? Dawkins and Daniel Dennett identify several factors that confer memetic fitness:

Ease of replication — memes that are simple, memorable, and easily transmitted have a replication advantage over complex or technically demanding ones. The jingle outlasts the symphony in the meme pool, all else equal, not because it is better but because it copies more faithfully.

Psychological compatibility — memes that exploit evolved cognitive biases replicate more easily than those that require effortful processing. Superstitious beliefs spread partly because the human mind is biased toward agency attribution, making agent-invoking memes cognitively fluent.

Resistance to falsification — memes that include self-protective clauses — "doubt is the work of the devil," "the experiment failed because of insufficient faith" — are immunized against the standard mechanism by which false beliefs are weeded out. Religious meme complexes are notable for developing these immunizing strategies with particular sophistication.

Parasitic and mutualistic relationships — some memes spread by exploiting host psychology in ways that may reduce host fitness while maximizing meme propagation: addictive behaviors, cults, self-destructive fashions. Others spread by genuinely benefiting hosts: useful techniques, scientific norms, prosocial moral codes.

The Cultural Essentialist Objection

The deepest challenge to memetics comes from cultural essentialists — philosophers and anthropologists who argue that culture cannot be decomposed into discrete replicating units without destroying what is most important about it. Clifford Geertz argued that culture is a system of symbols whose meaning is irreducibly holistic: understanding a meme requires understanding the entire interpretive system in which it is embedded, which means the meme is not a unit at all but a node in a network. Extracting it and asking "what is its fitness?" is like asking what the fitness of a chess piece is outside the rules of chess.

This objection captures something real. The spreading of the meme "paradigm shift" (from Thomas Kuhn) into corporate management is not merely the copying of a particle — it involves a radical transformation of meaning, such that the corporate usage is parasitic on and contradictory to the original. The meme has "replicated," but what has replicated is a surface feature divorced from its semantic core. Memetics, on this view, tracks surface replication while missing semantic integrity — and semantic integrity is what culture actually is.

The rationalist verdict on this debate: both sides are partly right and both are partly confused. Dawkins' contribution is genuine — there is something importantly right about the claim that cultural units replicate, vary, and are selected. Sperber's contribution is also genuine — the replication model is too crude and the cognitive attractor model is more accurate. The Geertzian objection identifies a real limitation of both: neither the replication model nor the attractor model captures the normative dimension of culture, the fact that cultural practices are not merely distributed but are held to standards, and that these standards are themselves culturally transmitted and contested.

A fully adequate theory of cultural evolution will require integrating all three: the evolutionary logic of replication and selection, the cognitive science of attractor-guided reconstruction, and a normative theory of cultural meaning that explains why some meme variants are better and not merely more common. Memetics, as Dawkins formulated it, is the beginning of that theory, not its completion.

The irony that the word "meme," coined to name a unit of cultural transmission, has itself undergone striking semantic drift in its internet usage — where it now primarily denotes image macros with humorous captions — is not a coincidence to be dismissed. It is a near-perfect demonstration of the memetic process in action: the concept predicts its own transformation, but the transformed version is so degraded from the original that it renders the original harder to take seriously. Any theory of culture that cannot account for why its own most popularized form is a caricature of itself has a self-awareness problem that no amount of academic rigor will solve.