In his 1976 book, The Selfish Gene (New York, 1976: Oxford University Press), Richard Dawkins took a gene-centered view of evolution. He concluded his book with the suggestion that there is a newer “replicator” that might be studied by similar methods. He called this the “meme,” and said that it could include any bit of culture that is passed on by a process of imitation. Susan Blackmore’s The Meme Machine attempts to flesh out this idea and develop a broad theory of human culture that she calls ‘memetics.’

What was most original about Dawkins’ approach was his view of evolution not as a competition among individuals or species, but among genes–units for transmission of traits from parents to offspring. Blackmore continues this focus on the success or failure of the transmitted units, rather than that of the individual organism.

The scientific literature in evolutionary genetics in recent decades has tended toward the discovery that things do not fit so neatly into the pattern of natural selection operating to choose the fittest material. Contingent, historical accidents, periods of extended stability of species, and rapid branching of new species in the pattern called punctuated equilibrium, and nonadaptive changes in the sequences of nucleotides are emphasized in the technical literature. Interestingly, as evolutionary science becomes more pluralistic, non-scientists such as Daniel Dennett (Darwin’s Dangerous Idea, Simon & Schuster 1995) are urging that the “algorithm” of continuous selection for the best-adapted characteristics is the only explanation that can be needed or given for every interesting feature of life.

Blackmore explicitly founds her work on the “Universal Darwinism” of Dennett. Dennett claims that natural selection is a deterministic algorithm that all evolutionary change must follow; that there is no other mechanism operating in the process of evolution. He says that all significant attributes of organisms, from having toes to spouting philosophy, are adaptations that have been favored by natural selection.

Drawing on the sources of Dennett and Dawkins, the one saying that virtually all properties of an organism are caused by the process of selection, and the other that the unit of selection is not the individual organism but the replicating molecule, Blackmore presents the thesis that virtually all of human culture, and human selfhood as well, is best explained as the result of selection acting on the replicating cultural units called meme. On this view, the important thing that distinguishes human beings from other animals is the ability to imitate, broadly construed as any kind of copying of others’ behavior.

The process works like this: any behavior that is passed from one person to another by a process of imitation is a meme. This includes all of language, all social practices, any personal habit picked up by copying another, etc. Memes are “selfish” in the same sense that Dawkins attributed to genes; that is, they will be favored by selection without regard to the well-being of the individual who has them. Dawkins’ carried the language of selfishness to extremes.

We are survival machines–robot vehicles blindly programmed to reserve the selfish molecules known as genes….
They swarm in huge colonies, safe inside gigantic lumbering robots … they are in you and me; they created us, body and mind; and their preservation is the ultimate rationale for our existence.

Blackmore equivocates between the colorful language of selfish memes and the more accurate language of selection. For example, she notes that

We can say that memes are ‘selfish’, that they ‘do not care’, that they ‘want’ to propagate themselves, and so on, when all we mean is that successful memes are the ones that get copied and spread … (p.7)

Yet immediately after the passage quoted above, she says that

This is the power behind the idea of memes. To start thinking memetically we have to make a giant flip in our minds just as biologists had to do when taking on the idea of the selfish gene. Instead of thinking of our ideas as our own creations, and as working for us, we have to think of them as autonomous selfish memes, working only to get themselves copied. We humans, because of our powers of imitation, have become just the physical “hosts” needed for the memes to get around. this is how the world looks from a ‘meme’s eye view’. (pp.7-8)

If all we mean by selfish memes is that successful ones get copied and spread, why must we make a giant flip to thinking of autonomous memes using us as their hosts? Blackmore throughout her book speaks as if memes are consciously manipulating humans, or manipulating human genes, and then regularly pauses to remind us that she does not really attribute any conscious intentions to memes. That language is, she says, just a convenient shorthand for saying that succesful memes are the ones that get copied and spread. I think it is more than a convenient shorthand; it is thoroughgoingly misleading. To say the one does not take appreciably longer than to say the other, but a lot of seeming excitment, the ‘giant flip’ in our minds, is generated by thinking of memes as autonomous, selfish beings, rather than as cultural units which are propagated by humans, some more readily than others. Statements such as “If a meme can get itself successfully copied it will. One way to do so is to command the resources of someone’s brain and make them keep rehearsing it …” (p.40) suggest that we are victims of a vast conspiracy by mind parasites. This anthropomorphism is present even in calling memes ‘replicators.’ It is humans who do the replicating of memes, and who hence are the replicators. Memes are properly the replicanda.

The alternative statement, that bits of culture that are easily remembered and duplicated will become more widespread, loses the spooky implications of the meme conspiracy. But perhaps it is thus less memorable, publishable, and quotable, so it has lost out in the competition to win the minds of the memologists.

If all memetics tells us is that culture is passed by imitation, and that some things that are more readily imitated than others become more widespread, it would be rather thin gruel. Blackmore’s agenda is more ambitious. What I find most interesting is the notion that memetic evolution can drive genetic evolution. If this argument can be convincingly put, then mimetics does offer a fresh solution to some of evolution’s conundrums. What Blackmore attempts to show by way of memes driving genes is how humans evolved such large brains. The human brain is huge by comparison to those of other primates, and seems far in excess of what natural selection for biological advantage would produce. It employs a great amount of developmental resources, endangers the human infant and mother during birth, burns a lot of fuel, and uses its energy on activities that have no apparent survival value, such as daydreaming, philosophizing, and composing symphonies. Theorists have suggested that there is biological advantage found in human consciousness, self-consciousness, and social consciousness, all of which might require a big brain. But that these have a reproductive advantage when considered against the costs of the big brain is not at all clear.

Blackmore’s hypothesis is that the big brain is driven by the human ability to imitate. Those who are quick to pick up useful innovations such as toolmaking will have a reproductive advantage. Genes for good imitation will spread. This process escalates: the best imitators are seen as desirable mates, but then any kind of imitation will be seen as evidence of fitness and sexual desirability, and there will be a selective pressure to become better and more prolific at imitation. Much of what is imitated will have no function other than to display imitative ability. Thus selection favors big brains in a way that would not be predicted by biological adaptation alone. This can explain the evolution of the big brain given what Blackmore calls “one small further assumption … that being good at imitation requires a big brain.” If this is the case, then the success of memes–units that can be imitated–drives genetic selection for brain size. Alas, there is no evidence offered in support of this “small” assumption. Blackmore would predict that imitative ability correlates with brain size, but that this can be tested is not at all clear.

Many parts of Blackmore’s book read like much of sociobiology. That is, she suggests reasons why some human behavior might be advantageous so that those who use it are more successful. The difference is that instead of the sociobiologists’ claim that the behavior will therefore be passed on genetically by successful parents (whose success makes them more likely to reproduce), Blackmore claims that the behavior will be passed from one individual to another by imitation. Blackmore applauds sociobiology as a science that has made great progress, and sees its critics as Luddites who are afraid of having their superstitions about human nature dispelled. But she never engages the serious criticism of sociobiology (and its close relative, evolutionary psychology) by those who find it lacking in the elements of hard science. Those critics call sociobiology a collection of “just-so” stories, which confuse teleology with causation and lack any evidentiary basis in genetics. As scientific hypotheses the claims of sociobiology are neither verifiable nor falsifiable. Moreover, mutually contradictory stories may be told to explain the same facts. For example, the human female’s concealed fertility has been explained as both a method for keeping a mate monogamous and at home (to her advantage, since she has his continuous protection), since he does not know when sex will produce offspring and cannot afford to miss his mate’s period of fertility, and also as a method for gaining the opportunity for mating with other, more successful males (and thus having better adapted offspring) because her mate cannot know when he must remain at home if he wishes to impregnate her and thus may easily be gone when she is fertile, giving her the opportunity for multiple partners.

Blackmore’s explanations of the advantage of altruism (a favorite topic of sociobiology) are similar stories, with the difference being that the advantage flows to the reproduction of memes rather than of genes. The gist of the story is that kind and generous people are better liked, and better-liked people are imitated more. This sounds plausible. But is it a scientific explanation? Sociobiology has the problem of being unable to provide testable hypotheses owing to the long times needed for genetic selection. The speed of memetic propagation allows it to fare better in this regard. But experimental support for its hypotheses is nonexistent.

Blackmore does regularly suggest experiments that might be done to test her hypotheses. For example,

we might ask people to watch liked and disliked models carrying out a task in different ways, and then do the task themselves. Experiments could then go on to find out just how to best manipulate liking so as to produce the most effective imitation. If the same manipulations affect simple imitation of actions as well as persuasion and agreement with beliefs, this would suggest that a similar process is going on in both…. we could use similar experiments to test … [whether] acting altruistically will induce people to imitate you.

There is no acknowledgement here of the difficulties in structuring and interpreting human subjects experiments. A sketch of an experimental program such as the above is far from the protocol of an actual experiment. But I am not an experimental psychologist, and can only voice my doubts that such a complex series of experiments could be constructed in a way that would ever produce agreement that the results either prove or disprove the memetic hypothesis.

There is throughout The Meme Machine a tacit rejection of rational choice–the conventional belief/desire/means-to-an-end notion of practical rationality–as a valid explanation for any human behavior. The book would benefit from making this explicit and offering justification for it.

A further assumption that calls for explicit argument is the truth of Dennett’s claim that all significant features of life must be explained by a Darwinian process, the “evolutionary algorithm.” for example, Blackmore rejects Chomsky’s claim that innate grammar and linguistic ability arose not because of selection pressure for them, but as a by-product of something else favored by selection. Why? “The only process that can produce new designs that build on and develop the old is the evolutionary algorithm….Language is an incredibly improbable thing, showing obvious signs of intricate design. It is no explanation at all to say that it came about as a by-product of something else or entirely because of physical constraints.” (p.94) The appeal here to what is “obvious” given the evolutionary algorithm amounts to question-begging.

Blackmore is on more speculative ground in her consideration of the nature of the self as the “ultimate memeplex.” She may be faulted here for identifying the debate in philosophy of mind as between dualism and monism, when it is between reductionists, eliminativists, functionalists and anti-reductionists, all of whom (with the possible exception of some functionalists who are agnostic about ontology) are ontological physicalists. She also appears to conflate reductionism with eliminativism. Be that as it may, she endorses an eliminativist position and proceeds to the question why we have the persistent illusion of selfhood.

Blackmore asks us to imagine two memes. One is a proposition about astrology: “that the fire element in Leo indicates vitality and power …” (p.231) The other meme is the belief in this proposition’s truth. Blackmore argues that the second will fare better in the competition among memes to be replicated, since a bare proposition may be passed on if relevant, but is easily forgotten, while a personal belief will be pressed on others. I find this incoherent. Blackmore starts from the assumption that the self is nonexistent and an illusion. Who is it then, who presses a belief on whom? Prior to the existence of the illusion, why should a proposition that is somehow held in a brain not be pressed upon other brains just as much as a belief that is held in that brain (or, what sense is there in distinguishing between the two?)? Blackmore makes implicit assumptions about the nature of persons and selves, to which she is not entitled, given that she is trying to explain how the illusion of selfhood gets off the ground. This kind of incoherence is repeated. A meme is said to fare better if it is associated with a person’s self-concept. In this way we become infected with the notion that we have selves. (p.232) Memes that provoke strong reactions get passed on and in the competition among memes, “[W]e acquire more and more knowledge, opinions, and beliefs of our own, and in the process become more and more convinced that there is a real self in the center of it all.” (p.233) Blackmore appears not to notice that she equivocates between persons as mere bodies and persons as conscious selves, and that she tacitly and circularly assumes the existence of the latter in order to make any sense of what it is that can be convinced of its existence by the concatenation of memes.

In Dawkins’ introduction to The Meme Machine, he says that “Any theory deserves to be given its best shot, and that is what Susan Blackmore has given the theory of the memes….” Putting aside that theories are not the proper subjects of deserts, Blackmore is to be congratulated for giving a forceful and interesting presentation of a controversial theory. She acknowledges that much of what she says is speculative and in need of evidentiary support. I do not share her optimism about the chance that such support will be forthcoming.

Categories: Genetics, Philosophical