In lieu of an abstract, here is a brief excerpt of the content:
Perspectives in Biology and Medicine 43.2 (2000) 227-242
[Access article in PDF]
The Meme Metaphor
Mark Jeffreys *
A Short History of the Metameme
In 1976, Richard Dawkins published The Selfish Gene, arguing that natural selection must be understood from the perspective of the genes because they alone--not individuals, not groups, not species--are the replicators on which selection ultimately acts. He also stressed that the concept of the replicator, not the gene per se, is the key to this understanding: "The gene, the DNA molecule, happens to be the replicating entity which prevails on our own planet. There may be others" [1]. In his concluding chapter, Dawkins speculated that human culture might actually be hiding another such replicator, which he dubbed the "meme," a neologism combining memory and mimetic with gene to suggest "a unit of cultural transmission, or a unit of imitation." [1]
Dawkins was by no means the first or last theorist to speculate about an entity akin to a social gene [2]. However, his coinage has proved the catchiest, and the meme-gene analogy as he presented it is not only memorable, but ideologically appealing: on the one hand, it holds out the tantalizing prospect of an elegant, universal theory of cultural evolution; on the other, it evades genetic determinism by offering a parallel cultural process with interests of its own. The meme provides a second replicator, which, though as "selfish" as any replicator, is at least independent of the interests of our selfish genes. As a "just-so story," it comforts us: "So that's why we're different from the other animals. So that's why we feel like we exercise a measure of precarious control over our bodies. So that's why we do such counterproductive things and feel so ill-at-ease in our own skins. We're the product of a conflict of interest between our memes and our genes."
Still, the concept of the meme didn't really blossom until the 1990s. Its championing by cognitive philosopher Daniel Dennett in a pair of books (1991's Consciousness Explained and 1995's Darwin's Dangerous Idea), coupled [End Page 227] with its natural suitability to the age of the Internet and the World Wide Web (in which ideas began more and more to appear as if they had lives and purposes entirely of their own), turned the meme from an illustrative analogy into the all-purpose widget of a pseudoscientific, cyber-cottage industry of evolutionary culture theory [3, 4]. As Cambridge anthropologist Robert Aunger, one of the meme's advocates, recently observed, "the study of memes has even become the subject of the obligatory academic journal. Indeed, all the claptrap surrounding a growing academic industry is in place. It's a bona fide memetic epidemic!" [5]
But is it yet or could it ever be a memetic science? Articles in the Journal of Memetics, now in its third year and available online, of course, continually quarrel over the definition of a meme and the possibilities of a methodology, but present virtually no data or original research of any kind [6]. Even a flashy Time article on memetics, published in April 1999, dutifully noted Stephen Jay Gould's dismissal of memetics as a "meaningless metaphor" and H. Allen Orr's description of it as "silly . . . co*cktail party science" [7].
Certainly, the meme is a metaphor, but to accuse it of being meaningless gets us nowhere. The question is whether or not it is a scientifically useful metaphor. Science perforce seeks to explain the unknown in terms of the known, rather than in terms of additional unknowns. The advantage of a good scientific metaphor is that the base domain is better understood than the target domain and so can be used as a template. As Dedre Gentner has pointed out, "the most important feature of a scientific metaphor is that the base domain be well understood and fully specified." [8] The history of domestic breeding, for instance, was far better understood in Darwin's England than was the natural history of species, and his metaphoric mapping between those domains...
