Posted on January 9, 2011 - by David

By Any “Memes” Necessary: Dennett’s Distortion of Culture

I’ve probably spent an inordinate amount of time over the last year thinking about “memes.” (Perhaps this is evidence that these parasitic mind viruses do in fact exist). Unsatisfied with my first critique, I hope to offer something more valuable here.

I may be wrong, but I get the feeling not that many social science types take the memetic view of culture seriously enough to respond to it – they smirk, or shrug it off, and go about their business. But with the amount of public attention “memes” have received, I think this ambivalence is a mistake. Students of culture who believe they have something better to offer ought to speak up.

Over the last 20 years, Daniel Dennett has probably been the strongest advocate of the memetic perspective, which grew out of Richard Dawkins’ book, The Selfish Gene (1976). Some of Dennett’s more recent thoughts are found in a 2009 article called ‘The Cultural Evolution of Words and Other Thinking Tools.’ I’ll focus my attention here, but refer to his other work as well.

What is culture?

(see Liah Greenfeld’s view here)

Dennett offers no explicit definition of culture here, but two that can be extracted from the article are “behavioral-perceptual transmission” and “transmission by replication of non-genetic information.” Obviously, “behavioral-perceptual transmission” plays an important role in the survival of many individual organisms and the continuation of many different species. So what, according to Dennett, distinguishes the human, “hyperpotent variety of cultural evolution” from transmission of learned behaviors in other species?

As Richerson and Boyd (2006) show, just as the standard information highway, the vertical transmission of genes, was optimized during billions of years, the second information highway from parents to offspring had to evolve under rather demanding conditions; however, once this path of vertical cultural transmission had been established and optimized, it could be invaded by “rogue cultural variants,” horizontally or obliquely transmitted cultural items that do not have the same probability of being benign. (The comparison to spam on the internet is hard to avoid.) These rogue cultural variants are what Richard Dawkins (1976) calls memes, and although some of them are bound to be pernicious—parasites, not mutualists— others are profound enhancers of the native competences of the hosts they infect. One can acquire huge amounts of valuable information of which one’s parents had no inkling, along with the junk and the scams.

This passage begs the question, in what respect are these “rogue” bits of culture “variants” from the “behavioral-perceptual transmission” we see in other species? The answer is found in his comparison of a termite castle and Gaudi’s ‘La Sagrada Familia,’ where he writes that “the design and construction [of Gaudi’s church] could not have proceeded without elaborate systems of symbolic communication”

What are “memes”?

If Dennett is saying that humans are distinguished from other animals by their dependence on symbolic transmission, then we are in agreement. But this still leaves the question: what, exactly, are “memes”?

In his 1991 book, Consciousness Explained, Dennett quotes Dawkins’ definition of the meme as a “unit of cultural transmission, or a unit of imitation.” (202) This article calls memes “cultural items that replicate with varying amounts of input from intelligent vectors.” If we try to synthesize a definition of “meme” by combining these statements with the implicit definition of culture I refer to above, (“transmission by replication of non-genetic information”), we can say that a meme is a replicating or replicable unit of non-genetic information. But this differs in two important ways from the “elaborate systems of symbolic communication” upon which Dennett correctly states that the construction of Gaudi’s church depends. First, there is no requirement that the “meme” be symbolic in nature, and second, culture is assumed to be fundamentally composed of discrete, self-replicating units or entities. To grant any one symbol an independent existence as a self-replicating unit is to remove it from the context in which it has significance and by which it renders its effects. So if the memetic perspective obscures the distinction that Dennett initially draws out, what, if anything, does it clarify or contribute?

Words as “memes”

What are words? They are not just sounds, or marks, or even symbols. They are memes (Dawkins 1976; Dennett 1991, 1995, 2006). Words are that subset of memes that can be pronounced.

Dennet calls words “our paradigmatic memes” and tells us that they “have an identity that is to a considerable extent language-independent”:

Like lateral or horizontal gene transfer, lateral word transfer is a ubiquitous feature, and it complicates the efforts of those who try to identify languages and place them unequivocally in glossogenetic trees. English and French, for instance, share no ancestor later than proto-Indo-European (see Fig. 2) but have many words in common that have migrated back and forth since their divergence (cul-de-sac and baton, le rosbif and le football, among thousands of others). Just as gene lineages prove to be more susceptible to analysis than organism lineages, especially when we try to extend the tree of life image back before the origin of eukaryotes (W.F. Doolittle, this volume), so word lineages are more tractable and nonarbitrary than language lineages.

It seems like all “lateral word transfer” really means is that throughout history, individuals and societies which speak different languages have come into contact with each other and shared words. And the fact that a word is found in more than one language does not mean its identity is “language-independent” either, it just means there is an even wider range of linguistic contexts in which it can be used and understood.

After this less than compelling argument for words as “memes,” Dennett goes in a somewhat different direction:

Words have one feature that has a key role in the accumulation of human culture: They are digitized. That is, norms for their pronunciation permit automatic—indeed involuntary—proofreading, preventing transmission errors from accumulating in much the way the molecular machines that accomplish gene replication do.

But these norms and the automatic correcting Dennett is talking about are not features of individual words; they come from the symbolic system of a language. Of course, he knows this, and writes:

… when you acquire language, you install, without realizing it, a Virtual Machine that enables others to send you not just data, but other virtual machines, without their needing to know anything about how your brain works.

Dennett’s computer analogy, the “Virtual Machine,” is the symbolic system of a particular language. Again we come back to context; the meaning of a word changes with the context in which it appears, with time, and from place to place. This may seem like a trivial observation, but I make it repeatedly because it is the fact which most obviously challenges the idea of discrete, self-replicating units of culture.

Why “memes”?

One of the goals of the meme concept is to unify culture and biology by attempting to demonstrate that natural selection governs not only biological evolution, but the cultural process as well. He writes in Consciousness Explained:

Meme evolution is not just analogous to biological or genetic evolution, not just a process that can be metaphorically described in these evolutionary idioms, but a phenomenon that obeys the laws of natural selection exactly. The theory of evolution by natural selection is neutral regarding the differences between memes and genes; these are just different kinds of replicators evolving in different media at different rates. (202)

This of course depends on whether “memes” exist (in the kind of concrete, material sense in which the language used to talk about them suggests they exist). But Dennett cleverly dodges the demand to prove this existence by instead suggesting that genes might not be as concrete as we tend to think:

Genes, according to George Williams (1966, p. 25) are best seen as the information carried by the nucleotide sequences, not the nucleotide sequences themselves, a point that is nicely echoed by such observations as these: A promise or a libel or a poem is identified by the words that compose it, not by the trails of ink or bursts of sound that secure the occurrence of those words. Words themselves have physical “tokens” (composed of uttered or heard phonemes, seen in trails of ink or glass tubes of excited neon or grooves carved in marble), and so do genes, but these tokens are a relatively superficial part or aspect of these remarkable information structures, capable of being replicated, combined into elaborate semantic complexes known as sentences, and capable in turn of provoking cognitive, emotional, and behavioral responses of tremendous power and subtlety.

I’m no geneticist, but I’m fairly certain that a nucleotide sequence is not merely a superficial token, arbitrarily related to the information it carries, the way that a word, (as a symbol), is arbitrarily related to its referent. The information is literally embodied in the nucleotide sequence. Nonetheless, this sleight of hand on Dennett’s part critical to advancing his argument past questions of the meme’s existence.

Now, taking the existence of memes for granted, the next step is to argue that their “selection” is due to their own fitness – they may or may not enhance the reproductive fitness of their hosts. He contrasts this with the “traditional wisdom – ‘common sense’ – according to which culture is composed of various valuable practices and artifacts, inherited treasures, in effect, that are recognized as such (for the most part) and transmitted deliberately (and for good reasons) from generation to generation.” He writes:

The key improvements, then, of the memetic perspective are its recognition that:

1. Excellently designed cultural entities may, like highly efficient viruses, have no intelligent design at all in their ancestry.

2. Memes, like viruses and other symbionts, have their own fitness. Those that flourish will be those that better secure their own reproduction, whether or not they do this by enhancing the reproductive success of their hosts by mutualist means.

“Memes” and Dennett’s ‘Intentional Stance’

If the tautology in number 2 above, (“those that flourish will be those that better secure their own reproduction”), makes “memes” sound an awful lot like intentional agents, a look at Dennett’s philosophy should explain why. He wrote a book called The Intentional Stance in 1987, but he has been working with the idea it contains for the last four decades. Here’s how he describes it in this summary of ‘Intentional Systems Theory’ from the Oxford Handbook of the Philosophy of Mind:

The intentional stance is the strategy of interpreting the behavior of an entity (person, animal, artifact, whatever) by treating it as if it were a rational agent who governed its ‘choice’ of ‘action’ by a ‘consideration’ of its ‘beliefs’ and ‘desires.’(1)

But Dennett is pretty clear that this is more than just a strategy; he tells us that “anything that is usefully and voluminously predictable from the intentional stance is, by definition, an intentional system.”

Where on the downward slope to insensate thinghood does ‘real’ believing and desiring stop and mere ‘as if’ believing and desiring take over? According to intentional systems theory, this demand for a bright line is ill-motivated.(7)

Seeking one’s own good is a fundamental feature of any rational agent, but are these simple organisms seeking or just ‘seeking’? We don’t need to answer that question. The organism is a predictable intentional system in either case. (9)

It’s one thing to argue that taking the intentional stance might help us describe certain aspects of the cultural process; when we consider how the internal logic of a system of symbols may predict the behavior of individuals and groups for whom that system is important, this is in some sense what we are doing. But Dennett’s claim that cultural evolution is governed by natural selection is dependent on a much more generous application of this kind of thinking – chopping the cultural process into little pieces and treating them as intentional agents (their intention being simply to replicate themselves).

The memetic view contains two intentional systems: the intention of the meme is to replicate itself at any cost, while the intention of the host remains to replicate its genetic material. There are memes that help this process and memes that hurt it, and a multitude of more or less neutral cultural trappings in which we are dressed along the way toward death. A meme is classified as a parasite, mutualist, or commensal, based on its effects on the reproductive fitness of its host. But why should we be committed to such an impoverished view of our existence? By Dennett’s own account, culture transformed our species, just as it transforms each new member that acquires it. He writes in Darwin’s Dangerous Idea:

… it cannot be “memes versus us,” because earlier infestations of memes have already played a major role in determining who or what we are. The “independent” mind struggling to protect itself from alien and dangerous memes is a myth.

It is no accident that the memes that replicate tend to be good for us, not for our biological fitness…, but for whatever it is we hold dear. And never forget the crucial point: the facts about whatever we hold dear – our highest values – are themselves very much a product of the memes that have spread most successfully. ( 364-365)

We truly are cultural beings; our own intentional states are inseparable from the historical, symbolic process that happens inside our brains. To be fair, Dennett knows how much context matters.  He wrote in this piece from 1998, that “the environments that embody the selective pressures that determine [memes’] fitness are composed in large measure of other memes.” But trying to explain culture by asking, as he suggests, “the cui bono question” (who benefits?), and answering, “our memes,”  means ignoring a critical, if easy to miss, fact revealed in the quote above: we are a species that has values and holds things dear. Dennett clearly values “mutualistic memes” – the culturally driven development of “technology and intelligence” which has made surviving so much easier for our species – but seems not to fully appreciate the observation anthropologist Clifford Geertz makes in his essay, ‘Ethos, Worldview, and the Analysis of Sacred Symbols,’ (1957):

The drive to make sense out of experience, to give it form and order, is evidently as real and as pressing as the more familiar biological needs. And, this being so, it seems unnecessary to continue to interpret symbolic activities-religion, art, ideology-as nothing but thinly disguised expressions of something other than what they seem to be: attempts to provide orientation for an organism which cannot live in a world it is unable to understand.

‘Traditionalists’ made of straw

Dennett makes a point of claiming that what he considers a naive, “economic model” of culture, “where possessions, both individual and communal, are preserved, repaired, and handed down,” “is for the most part uncritically adopted by cultural historians, anthropologists, and other theorists.”  He believes the fact that “many of our most valuable cultural treasures have no identifiable author and almost certainly were cobbled together by many largely unwitting minds over long periods of time” presents the “traditionalist” with a serious problem:

Nobody invented words or arithmetic or music or maps or money. These apparent exceptions to the traditional model are typically not seen as a serious problem. The requirement of intelligent authorship can be maintained by distributing it over indefinitely many not-so-intelligent designers whose identities are lost to us only because of gaps in the “fossil record” of culture. We can acknowledge that many of the improvements accumulated over time were “dumb luck” accidents that nevertheless got appreciated and preserved. With these concessions, the traditionalist can avoid acknowledging what ought to seem obvious: These excellent things acquired their effective designs the same way plants and animals and viruses acquired theirs—they evolved by natural selection, but not genetic natural selection.

First, (whether or not anyone cares about maintaining “intelligent authorship”), what Dennett labels here as “concessions” actually describe the cultural process better than just saying “they evolved by natural selection”; how, after all, can he claim that that this evolution happens “the same way,” when the crucial set of facts is of a radically different (symbolic) nature? Second, it’s possible for an individual to have an impact on the collective cultural process without intending to and without being aware of it – I don’t see how this scenario supports Dennett’s claims. But most importantly, I return to the Geertz, because his view of culture is conspicuously not like the “traditionalist” straw man Dennett sets up. He wrote this in ‘Religion as a Cultural System,’ (published 10 years before Dawkins invented the “meme”):

So far as culture patterns, that is, systems or complexes of symbols, are concerned, the generic trait which is of first importance for us here is that they are extrinsic sources of information. By “extrinsic,” I mean only that–unlike genes, for example–they lie outside the boundaries of the individual organism as such in that intersubjective world of common understandings into which all human individuals are born, in which they pursue their separate careers, and which they leave persisting behind them after they die. By “sources of information,” I mean only that–like genes–they provide a blueprint or template in terms of which processes external to themselves can be given a definite form. As the order of bases in a strand of DNA forms a coded program, a set of instructions, or a recipe, for the synthesis of the structurally complex proteins which shape organic functioning, so culture patterns provide such programs for the institution of the social and psychological processes which shape public behavior.

Notice that “to make sense out of experience” and “provide orientation,” it is not necessary for “systems or complexes of symbols” to be some carefully curated set of goods. Nor does Geertz assume that culture patterns will necessarily be positive, or work to enhance genetic or reproductive fitness – it’s possible for an order-creating system to be reprehensible and disastrous, (take Nazi ideology for example). It’s a shame that Dennett pretty much discounts the work of all previous cultural theorists for the sake of a rhetorical device, rather than at least attempt to use someone like Geertz as a jumping off point.

But Geertz’s work gets at the heart of the problem with memetics; it may function as an explanation for the phenomenon of culture, but I think any attempt to use it in a robust analysis of empirical events must involve dropping its most characteristic feature – the idea of individual units of culture attempting to replicate at any cost and governed by natural selection. The following comes from ‘Person, Time, and Conduct in Bali’(1966):

One cannot run symbolic forms through some sort of cultural assay to discover their harmony content, their stability ratio, or their index of incongruity; one can only look and see if the forms in question are in fact coexisting, changing, or interfering with one another in some way or other, which is like tasting sugar to see if it is sweet or dropping a glass to see if it is brittle, not like investigating the chemical composition of sugar or the physical structure of glass. The reason for this is, of course, that meaning is not intrinsic in the objects, acts, processes, and so on, which bear it, but–as Durkheim, Weber, and so many others have emphasized–imposed upon them; and the explanation of its properties must therefore be sought in that which does the imposing–men living in society. The study of thought is, to borrow a phrase from Joseph Levenson, the study of men thinking; and as they think not in some special place of their own, but in the same place–the social world–that they do everything else, the nature of cultural integration, cultural change, or cultural conflict is to be probed for there: in the experiences of individuals and groups of individuals as, under the guidance of symbols, they perceive, feel, reason, judge, and act.

Symbols, if they are to be understood, cannot be divorced from their function/use in creating order and meaning for individuals and groups. This is why, I believe, Dennett admits in the summary of chapter 12 of Darwin’s Dangerous Idea, that “the prospects for elaborating a rigorous science of memetics are doubtful,” though he maintains that “the concept provides a valuable perspective from which to investigate the complex relationship between cultural and genetic heritage.” From what I’ve read, none of Dennett’s claims which hold water are original, or require the memetic perspective.

I’ve tried my best to be thorough here, but it’s impossible to cover everything. As always, comments are welcome.