We hate memes, pass it on…

Vaughn at Mind Hacks has a short post, Memes exist: tell your friends (clever, Vaughn, very clever), which links to a couple of meme-related talks at TED. Daniel linked to a lot of the TED talks back in April (TED: Ideas Worth Spreading), but Vaughn focuses on videos of Daniel Dennett and Susan Blackmore, both of whom are ardent meme advocates.

I’ve watched both talks, more than a half hour of my finite lifespan that I will never get back (okay, I’ve wasted part of my finite life doing worse… I think), so I need to unburden myself. I think ‘memetics’ is one of the bigger crocks hatched in recent decades, hiding in the shadow of respectable evolutionary theory, suggesting that anyone who doesn’t immediately concede to the ‘awesome-ness’ of meme-ness is somehow afraid of evolutionary theory. Let me just make this perfectly clear: I teach about evolutionary theory. I like Charles Darwin. I have casts of hominid skulls in my office. I still think ‘memetics’ is nonsense on stilts on skates on thin ice on borrowed time (apologies to Bentham), as deserving of the designation ‘science’ as astrology, phrenology, or economic forecasting.

What’s hard for me to understand is that I LIKE some of Daniel Dennett’s work, and I can’t cite Dennett’s other work confidently when he has picked up a ‘meme franchise,’ and is plugging away with the ‘meme’ meme, making it appear that I’m down with this later material. Blackmore, on the other hand, is a reformed para-psychologist, so she’s, at worst, made a lateral move in terms of respectability. I get particularly irritated during her talk because I think she does an enormous disservice to Darwin’s Origin of Species, but I will try not to late my irritation show too much (even though our regular readers know I won’t be able to manage). I wasn’t going to really heap scorn on Blackmore until I read her own account of TED on the Guardian’s website; gloves are now off.

But I digress, back to the content of the concept and Vaughn’s comments…

Vaughn argues that:

The concept of memes is controversial, not least because it’s hard to see exactly what empirical predictions follow from the theory. Rather than a set of specific hypothesis, it’s really a different framework with which we can re-interpret aspects of culture.

What particularly annoys the critics is the idea that cultural ideas are subject to a Darwinian-style process of selection and (presumably) evolution.

As a critic of memes, I agree with Vaughn to some degree; for example, even Bruce Edmonds argued that memetics had produced limited empirical research in the last edition of the Journal of Memetics before it went into a deep sleep (from which someone is trying to awaken it. Shhh… let it sleep…).

But I would suggest that Vaughn’s single reason for critics’ annoyance severely under-represents the many, varied, and significant reasons that some of us loathe ‘memetics’ (I speak here only for myself, not for Daniel, who may not have the same severe twitch I develop around talk of ‘memes’). Worst of all, memetics sucks the air out of the room for a serious consideration of the ways that culture, knowledge, technology, and human evolution might be interrelated. That is, like a theory of humours and vapors in illness, it provides pseudo-explanations in place of just getting the hell out of the way of serious thought. Memeticists often, perhaps intentionally, seem to generate confusion between what they are doing and what Gerald Edelman christened ‘neural Darwinism,’ a very different discussion of the physiology of neural conditioning; it’s unfortunate guilt by association for the latter, which seems to be grounded in actual evidence.

So, why do I hate the concept of ‘ideas replicating from brain to brain.’ After all, I work on physical education and imitative learning; shouldn’t I be happy that memetic theory places such a premium on imitative learning? What is my problem!? Ah, let me count the problems… I’ll just give you 10 Problems with Memetics to keep it manageable.

1) Reifying the activity of brains

In spite of what some undergrad writers seem to think, the subject of a sentence matters, as does the verb. So when one says, like Blackmore, that ‘ideas replicate from brain to brain,’ we have to ask about the implications of this sentence. Is it plausible, controversial, or just plain daffy? To argue that ‘ideas’ are an agent that ‘replicates’ suggests several layers of reification that I think are profoundly crippling to memetic theory.

If ‘culture’ is already a bit of a reification (treating as a ‘thing’ a complex of heterogeneous behaviors or concepts), ‘meme’ is a kind of mondo-reification. It is attributing to a human behavior, not just a thing-like status, but causal thing-like status (‘the meme made me think that’ or ‘the meme made me do this’). If I come to you and say, ‘The doing made me do something,’ you’d look at me like I was a bloody idiot. But that’s what a ‘meme’ is; a reification of a human behavior posited as the cause of human behavior. Sure, there’s a loopy, pseudo-existentialist ring to arguing that the ‘doing made me do’ something, but it’s hardly a solid place to start a supposedly ‘scientific’ theory.

2) Attributing personality to the reification of ideas

This reification issue with memes turns into a full-blown hare-brained robot conspiracy theory by the end of Blackmore’s talk because of the characteristics she attributes to memes; that is, it’s one thing to reify a concept, it’s another thing to start attributing it a whole complex personality, drives, desires, and levels of different reification. For example, Blackmore says the following about ‘temes’ (technological memes, a new type of replicator even scarier than memes):

But it’s dangerous: temes are selfish replicators, they use us to suck up more resources to produce more computers and more things. Don’t think we created the Internet, that’s how it seems to us.

Uhhhh… Susan, so we didn’t create the Internet? ummmm… who did? While mildly problematic to reify memes, now she’s attributing the cause of concrete, observable things (networks of information flow through fibre optic cable with a very clear history) to a dubious second-order abstraction (not even the memes, but the temes of the memes). Worse still, is Blackmore suggesting that the Internet created us or that it threatens to take control of us? If so, then we are truly in the realm of daft, science fiction free-association, beyond simply the silliness of mistaking a metaphor for a reality. Now she’s afraid of her own metaphors.

3) Doesn’t ‘self-replicating’ mean replicating by one’s self?

There are problems with defining even a gene as a ‘self-replicating’ structure, like, if DNA is so self-replicating, why is it so chemically inert? In actuality, the reason we can pull DNA from old bones is because, in the absence of complex cellular mechanisms, DNA just sits there, kind of doing nothing. Oh, okay so it’s not soooo ‘self-replicating’ that it can actually, well, replicate by its, err…. self….

If defining gene as ‘self-replicating’ is playing a little free with the details, defining meme, as ‘self-replicating’ beggars the imagination it’s so stupid. That’s right, the ditty that you keep singing is really replicating itself in your brain…. Arguing this reveals so little understanding of how brains work, especially how hard it is for ANY pattern to repeat completely. That is, even repetitive action typically involves constant changes in patterns of neural activation; maintaining consistency requires constantly shifting neural resources, even slightly, to take account to changes even in the organism itself.

Moreover, ‘self-replicating’ means, by definition, replicating by itself. Has anyone, ever, anywhere, seen an idea ‘replicate’ itSELF? Although this may seem like a semantic point, I think it’s a bigger logical problem with reifying culture as ‘memes’ and then attributing agentive power to the memes (which, by the way, is a concept created by a person, Richard Dawkins, in spite of Dennett’s assertion that the meme has autonomy).

4) The term ‘meme’ applied to divergent phenomena

Calling an idea a ‘meme’ gets around the enormous problem of incommensurate phenomena in the same category. For example, during Dennett’s talk, he refers to the following as ‘memes’: very abstract general concepts (‘justice,’ ‘freedom’ — note: I’m not sure we’d even agree on the definitions of these), political-economic philosophies (‘communism’ — note: you couldn’t even get proponents to agree on a definition), even whole religions (Islam, Catholicism — note: again, you couldn’t even get all adherents to agree on what these are). In addition, in other sources, we’ve seen memeticists refer to single ideas, strings of idea, melodies, and a host of other things as ‘memes.’ Even the most cursory glance reveals serious problems of scale; is a meme a single idea, a chain of ideas, a system of ideas, or an entire worldview? Fortunately, if you just keeping saying ‘meme’ you might be able to ignore the fact that not everyone in the room agrees what you’re on about.

Blackmore tries to finess this problem: memetics is a ‘much maligned science’ (*cough* … science? errr… ‘theory,’ maybe…). A meme isn’t an idea, she says. No, it’s ‘that which is replicated.’ If the definition of ‘gene’ were this circular, would be in deep shit in terms of genetic medical research. Blackmore goes on to use the example of the way a chap in the audience has his glasses hung around his neck; IF he copied it from someone else, it’s a meme. (And if he didn’t copy it, it’s not a ‘meme’?) Then there’s a brilliant moment when she ‘can’t seen any interesting memes’ in the audience… but I thought memes were everywhere, that we were swimming in a sea of memes.

5) Could memes transfer stably?

If Blackmore actually studied cultural reproduction, that is, where culture is being taught, learned, and corrected, she might be more sensitive to the fact that transmission is fraught with ‘transcription’ errors (I hesitate to use the term ‘transcription’ because it dignifies the whole ‘meme=DNA’ metaphor which memeticists are abusing like a borrowed mule). Even teaching might demonstrate how utterly improbable it is that ANYTHING gets copied accurately (my wife is marking final exams as I speak, and trust me, she doesn’t think of humans as perfect idea-copying machines). Sure, my daughter may sing the Kraft Vegemite jingle after she sees the commercial, but is she really getting the melody right? We all know myriad examples of even songs and melodies winding up with garbled lyrics.

Without stable transmission (and no plausible mechanism of meme transmission to match DNA transcription), meme transmission would be a chaotic process, with no certainty that ‘successful’ memes would be transmitted stably. As Blackmore herself argues, for natural selection to work, you need heredity, and it’s not at all clear that cultural materials are ‘inherited’ with integrity. In fact, one could plausibly argue that one of the most difficult choke points for memes is transmission itself; almost like a form of ‘sexual selection,’ cultural, behavioral, or intellectual transfer will impose their own sorts of pressures. For example, if students routinely simplify complex ideas, the memetic transfer has a kind of simplifying drag.

6) A host will not evolve traits in order for parasite to benefit

Over at Ionian Enchantment, Michael Meadon also writes about the Susan Blackmore video, and he offers a number of serious criticisms, including the idea that memes drive brain development. As he writes:

The serious logical error comes in when she argues, amazingly, that humans have big brains in order to copy memes. That is, she argues there is a “memetic drive” favoring brains that are better at copying memes completely independently of genetic evolution. Language, on this view, is a parasite which we only later “adapted to”. How such a process is meant to operate I have no idea. Why would selfish genes altruistically code for proteins that build bigger brains to help selfish memes replicate? I can see how memetic evolution could take off as a by product of increased intelligence brought about by biological evolution; I simply can’t see how memetic evolution could cause larger brains to evolve in the absence of a biological fitness benefit.

That is, unless a trait is beneficial in natural selection to the host, the parasite is not going to get evolution to create a better host for its own benefit. If the meme is truly a parasite, then there’s no way that the human brain is going to grow for the good of the meme.

7) Trivial examples as analogy to ideological change

A recurring problem in memetics theory is triviality being used to explain serious issues. Although she’s attempting to be funny, Blackmore uses the example of folding toilet paper so that the end forms a point as an example of the global spread of a meme. This example is supposed to explain something serious, like the spread of a religion. The same thing with the example of an advertising jingle. These simplistic examples are then argued to be analogous to something like Christian conversion or the spread of capitalism, as if getting a jingle stuck in your head is like undergoing a major religio-ideological or political-economic social transformation.

Dennett compares memes to lancet flukes, a parasite that takes over the brains of ants so that it can use the body of the ant even though the behavior is suicidal for the ant. He then compares this to ‘dying for an idea,’ whether that ‘idea’ be communism, capitalism, justice, freedom, Catholicism, or Islam. Is ‘Catholicism’ really ‘an idea,’ like an advertising jingle or a concept (like ‘memes’), or is it really something a hell of a lot more complex, including a social system of status, a community, behaviours, multiple ideas, desires, modifications of basic emotions, and a host of other things? That is, is Catholicism (or Islam or communism) like a gene? As a Catholic school boy and an avid reader of Marx, I can, with some confidence, say that neither are ‘an idea’; they are a lot of ideas, behaviors, even social relations, with long histories, marked transformations, and whole social worlds connected to them. We talk about ‘dying for an idea,’ but it’s a sloppy metaphor for what is really much more complex.

8 ) Gradual cultural transmission not like infection

The metaphor of ‘infection’ is another one that gets used in memes, as it is clear that memes must have some sort of Lamarkian dimenions. Dennett, for example, argues that ideas, like the lancet fluke, are infectious. Is Dennett saying that the jingle from the Kraft Vegemite commercial has really taken over my brain and is forcing me to do its bidding? And if the Kraft Vegemite jingle has my brain, does that mean that it has eradicated the other memes trying to seize my brain? And will I commit suicide to reproduce the Kraft Vegemite jingle? Or will my exploding brain infect other people with the Kraft Vegemite jingle like some virulent melodic ebola virus?

Okay, maybe I’m mocking the metaphor of infection, but I think that the hooky jingle that gets stuck in our brain is actually a pretty rare phenomenon (in addition, usually a fair bit of creativity and craft has gone into producing that jingle). Most ideas don’t reverberate inside our heads like the unstoppable Kraft Vegemite jingle; usually, a single exposure to an idea is not adequate for it to become firmly lodged in our brains and perfectly reproduced. Religious conversion, for example, is not often an instantaneous event, like a bolt on the road to Damascus or a viral infection; rather, it is a gradual change, often in stages. Why? Because a religion is not a ‘single idea’ like an ideological gene. But given this gradual establishment of ideas, then all sorts of other factors are going to affect whether or not we end up accepting and reproducing ideas.

9) Objective ‘science’ inconsistent with normative judgments about memes

If memes are successful at reproducting in our brains, than how can they be abused or misused, as Dennett worries about in his talk. Can an ant ‘misuse’ a lancet fluke? If a meme were really like a gene, it wouldn’t matter which variant of the idea got reproduced (in part because, aside from transcription errors and mutations, genes transfer stably). There’s this strong stream of judgment in both Dennett and Blackmore that’s inconsistent with evolutionary theory.

Dennett, for example, argues that ‘toxic ideas’ are like the pathogens brought be explorers to the New World. He suggests radio stations are ‘memes’ (but aren’t melodies memes, so wouldn’t stations be like dirty door knobs?), and that Western memes are wiping out indigenous ideas around the world. It’s memes versus memes, not changed economic systems, uneven access to technology and media, and forms of symbolic domination. Dennett argues that ‘we’ (presumably all those attending TED or watching the podcasts) have an ‘immunity’ to all the ‘junk’ that ‘lies around the edges of our culture’; pornography ‘for us’ is a ‘like a minor cold,’ but for people without ‘immunity,’ pornography is potentially fatal.

Dennett asks, How do we tell the ‘good memes’ from the ‘bad memes’? We don’t have to worry because memetics, the ‘science of memes,’ is ‘morally neutral.’ Ideas are like the HIV virus; the way to deal with it is to do science, to understand how and why ideas spread. We should encourage the spread of ‘relatively benign’ versions of the most toxic memes… I find the account incoherent, fluctuating between supposed ‘scientific’ neutrality and extreme normativity. The problem is that, without any sort of social or political theory (it’s just memes versus memes out there), there’s really no logical ground for Dennett or Blackmore to make the judgments about good and bad memes that they seem so likely to make. It’s one reason that Blackmore’s critique becomes completely mad at the end of the talk — she has a sense of moral outrage, but she singles out robot ‘temes’ that might eradicate us as the great threat. In a world of profound economic inequality, inadequate educational systems, massive health problems, ecological dangers, and the like, her fear would be comical, if I weren’t livid.

The end of Blackmore’s talk descends into a chaos of wild ideas — self-replicating ‘temes’ (technological memes), life on other planets, robot replicators taking over for humans. The bizarre turn might appear unusual if you assumed Blackmore was a rational evolutionary theorist or psychologist, but remember, she started off as a para-psychologist.

10) Resistance to memetics is not ‘anti-Darwinism’; Darwinism not a religion

Finally, I resent the argument that Dennett makes, that all those who resist memetics are ‘anti-Darwinist’ or afraid of the implications of Darwin’s ideas. I’m more than comfortable with Darwin’s contribution to evolutionary theory, especially natural selection, and I think his works (not just Origin of Species) were remarkable, but I don’t adhere to ‘Darwinism’ as if it were a scholastic faith.

It’s not ‘Darwinism’ that I support, like it’s a cult or a form of thought that I must follow religiously; I believe that ‘Darwinism’ is only useful in that it is a theory that provides hypotheses to be tested, a powerful explanatory framework that explains some (though not all) phenomena. That is, when Dennett argues that some people are insufficiently Darwinist because they don’t want to apply ‘Darwinism’ to the world-wide web or Hoover Dam, I feel like he’s treating ‘Darwinism’ as a one-size-fits-all über-explanation. That’s not science — that’s a cult. In fact, most people who study evolution argue that there a LOT of things that must be added to ‘Darwinism’ to get modern evolutionary theory (like, say, ooooh, genes…). So for Dennett to argue that it’s sufficiently ‘Darwinist’ to believe in memes, he’s stuck with a massive problem; a) Darwin didn’t know about genes, and b) Darwin certainly never mentioned ‘memes.’ So, then, Darwin wasn’t sufficiently ‘Darwinist’?

But worse, does Blackmore really believe in ‘Universal Darwinism’ as she argues? Is she even listening to herself? Does Darwinism explain the culinary arts? Fractals? Astrophysics? The development of animation? Royal marriages? Reality TV? Which direction water goes down the toilet? I’m not just being facetious — Blackmore’s position is simply incoherent.

Even the notion that Darwin had the ‘best idea anyone every had’ (Blackmore’s line, not mine), that the idea ‘explains all design in the universe’ (again, NOT my line — the word ‘design’ alone makes me cringe), suggests that Darwin’s insight was a kind of creation or invention when I think it’s better to describe it as a perception, or an analysis, or an insight. Darwin’s discussion of natural selection was not just a ‘great idea’; it was a ‘great observation.’ Darwin was a scientist. That’s what made his thinking great. He observed closely and fit theories to data, accepting when observations required overturning accepting ways of thinking. In fact, ‘Creation’ was a ‘great idea,’ too. Sucks that it doesn’t line up with observable facts. But that’s the way the creationist cookie crumbles.

Susan Blackmore: an irony-free blogger

One irony in all this is that Susan Blackmore, a veritable fountain of fuzzy-headed ideas, goes after Jill Bolte Taylor for some bits in her account of surviving a stroke and paralysis on Blackmore’s own blog account of TED. (Daniel liked to some material on Jill Bolte Taylor by Leslie Kaufman in Wednesday Round Up #13.) Blackmore thinks Jill Bolte Taylor’s ideas… errrr, memes, are not good ones, but she’s worried that they’ve escaped into the congenial environment of the internet, where they’ll have an undeserved additional lease on life. She writes:

The disquiet I feel about this is, I think, that scientific ideas must compete to be accepted. Within science, and at scientific conferences, the valid ones win by experiment and peer review, and the false ones are weeded out. In the great wide word of the web, and with easy access to podcasts, false ideas may thrive because of fine presentation or moving emotional manipulation. Taylor’s was precisely that.

Obviously, Blackmore’s irony meter is on the blink; the Journal of Memetics declared its own demise in 2005, including a number of obituaries for the whole concept (note: the JoMemetics site currently has all of its links broken). As Blackmore herself says, sometimes even the critique of experiment and peer review isn’t enough to put to rest a lousy idea. Jill Bolte Taylor’s account is a personal reflection on an experience, in contrast; we can be forgiven if we cut her some semiotic slack, especially talking about such an intensely personal experience. In contrast, Blackmore calls for the scientific review which has repeatedly found the meme concept wanting. I’d say it’s sort of like the pot calling the fruit bowl, ‘black.’

Cartoon by the very funny, and very generous, Hugh MacLeod, gaping void cartoons, from the back of a business card, and he offers them under a Creative Commons license. This cartoon, ‘The Hughtrain,’ is great, but I’ll try to post some more over the coming weeks.

Published by


Trained as a cultural anthropologist at the University of Chicago, I have gone on to do fieldwork in Brazil and the United States. I have written one book, Learning Capoeira: Lessons in Cunning from an Afro-Brazilian Art (Oxford, 2005). I have also co-authored and co-edited several, including, with Dr. Daniel Lende, The Encultured Brain: An Introduction to Neuroanthropology (MIT, 2012), and with Dr. Melissa Fisher, Frontiers of Capital: Ethnographic Reflections on the New Economy (Duke, 2006). My research interests include neuroanthropology, psychological anthropology, sport, dance, human rights, neuroscience, phenomenology, economic anthropology, and just about anything else that catches my attention.

46 thoughts on “We hate memes, pass it on…

  1. Yeah it’s a good critique of memes. There is a huge amount of bad thought and fallacy in your own argument though.

    Ad hominems e.g. “she started off as a para-psychologist.” Who cares. Let your argument stand against her own argument, now who she once was. I’m sure you held crappy ideas years ago as well.

    The continuous use of rhetorical questions e.g. “But worse, does Blackmore really believe in ‘Universal Darwinism’ as she argues? Is she even listening to herself? Does Darwinism explain the culinary arts? Fractals? Astrophysics? The development of animation? Royal marriages? Reality TV? Which direction water goes down the toilet?” Just come out and state the premises of your own argument. Arguing rhetorically isn’t the place of science, it’s for politicians and snake-oil salesmen. Your argument can stand without this fluff.

    I’ll also take aim at your overarching thought on the theory itself: “like a theory of humours and vapors in illness, it provides pseudo-explanations in place of just getting the hell out of the way of serious thought.” Well no. Originally it was an argument from analogy, which is a type of inference. Some of the best critiques of memes actually attack the disanalogies and relevance to genes and memes, as that is where the weakness in the theory lies. Buying in on it that it *is* a pseudo-scientific theory is actually making a strawman out of what memetic theorists like Dawkins et. al. are doing. If anything they were hoping to make the jump from analogical-inference to inference to best explanation with good inductive reasoning, which to me, doesn’t look like it hasn’t happened yet (Blackmore doesn’t help in this regard).

  2. Ok makes a few points, but, I think I lost a couple of them in negative phrasing (I’m not sure what he means, for example, in the last sentence by saying that inductive reasoning ‘doesn’t look like it hasn’t happened yet’). Ok may very well be right, I’m just not exactly sure that I know precisely what he’s suggesting, so I don’t know if I should be patting myself on the back or not (probably not, judging from his tone).

    Look, this isn’t a scientific argument — it’s a blog post. And I’m not arguing with a scientific journal article; I’m arguing with (as Ok points out) an extended analogy, in this case the meme analogy as presented in a couple of TED talks. Dennett argues by analogy to lancet flukes and Blackmore by analogy to toilet paper origami. They use analogies; I use analogies.

    If I’m lucky, I get a draw, and this meme thing doesn’t make any further inroads. Unfortunately, with the two TED talks likely to get 100s of 1000s of views, and my post likely to get less than 1000, I’m fighting a losing battle (especially when Blackmore’s talk is being rated by viewers as ‘fascinating’ and ‘informative’; fortunately, ‘unconvincing’ is running third, so maybe I shouldn’t be completely pessimistic. Dennett’s ratings, however, give no cause for hope: ‘persuasive,’ ‘informative,’ and ‘fascinating.’ Great…).

    As I explain in the post, I’m not going to back off Blackmore, not merely because I believe she’s been presenting these same ideas about memes for a few years (well, the ‘temes’ thing is new), and has probably heard a lot of good arguments against them and continues undeterred. The principal reason I got so hacked off as to write this was her blog comment on Jill Bolte Taylor, which was not only ungracious and unfair, but profoundly hypocritical.

    Blackmore grants that Taylor’s account of surviving a massive stroke was ‘gripping,’ but then she says Taylor ‘spouted misleading gibberish’ about right/left brain differences (fair enough) and about her experience of being ‘out of body’ during the stroke. She argues that bad ideas must be weeded out by peer review in journals, by experiments, and by debate at academic conferences; I agree. However, some of what Blackmore is criticizing is Taylor’s subjective account of a near-death experience, a subject she herself has written on. That is, if you tell me you’re possessed, we can argue about the source of that experience, but I can’t really argue that you didn’t have the experience. And I hardly think that the ‘meme’ concept has stood up well to review, experiment, and debate.

    Blackmore’s own brief profile on The Guardian’s website and her personal website discuss that she ‘no longer works on the paranormal,’ so it’s hardly outrageous for me to point this out. She criticizes other people (such as Tony Blair) for believing in things without proof of their existence (uhh… .like memes?), and yet also admits her own difficult road to realizing that her early ideas about the paranormal were not well founded in a piece on The Edge website. In other words, I think she’s an inconsistent ‘sceptic’ (a term she uses to describe herself ‘psychologist and sceptic’).

    But the bottom line is that this is a weblog; it’s a fluid genre with its own demands. It’s not an academic article. I sincerely doubt that regular readers feel that I typically err on the side of overly-short postings. If anything, my pieces probably tend toward overkill, which makes them hard to read. A bit of spice helps a long post go down.

    And if I see red when I read about memes (or about ‘racial science’), I hope our readers understand. PZ Meyers gets to heap abuse on Ben Stein and other creationists because they make him irate; Keith Olbermann can be counted on to take a swipe at Bill O’Reilly; Mr. Blackman will inevitably pour out scorn on the frumpy and tacky. I’m going to pull out every rhetorical, metaphorical, and dramatic trick in my bag to try to drive a stake through the heart of this meme thing, knowing damn well that I’m likely doomed to failure.

  3. Greg, I remember getting the memes thing back in college–I worked with one professor who was a pretty big proponent of anything Darwin related. But it just struck me as so irrelevant that I haven’t really paid any attention to the whole area since then. Your critique brought out a lot of the things that make it a wrong path. In my mind, being integrative means NOT taking everything on board–memes fails precisely because it is over-reaching, reductive, and not even close to the truth. Memes don’t irk me like they do Greg; I could simply care less–and that’s another death call for memes…

    In college, I largely disliked memes because I thought they were simply bad evolutionary theory. A cute idea perhaps, a play on words, but just missing out on the basics–where is a concrete adaptive function? what are the actual mechanisms of reproduction and inheritance? where is the central role of phylogeny, or common descent? Greg reflects on many of these issues above, but let me repeat, memes fail even in their own area of origin, evolutionary theory

    In reading your post, I was also struck by two things, first the assumption of compulsion inherent in the memes idea (even the hard core addiction people at least consider the drug and a reward system). But compulsion has to also be constructed and enacted, at the minimum through brain/body/environment interactions, and in the world of human knowledge, through social and cultural dynamics that are much more interesting in themselves than memes. Peter Berger and Thomas Luckman’s book The Social Construction of Reality is one classic reading to get people started.

    Second, Dennett really seems to be speaking about the dangers and powers of knowledge, and I do find it deeply disturbing and problematic that he takes out the human dimension when he basically reduces knowledge to memes. Ethics, context, tradition, emotion, and many other things are lost in this move. And it is a move by Dennett that, once again, is anti-Darwinian. Our knowledge exists in the tangled bank of humanity, shaped so deeply by our biological and social and cultural phylogeny.

    The genes-and-memes of hardcore evolutionary psychologists makes the biology-culture split in academia about as deep and wide and extreme as you can make it. Greg aims to drive a stake through memes’ undead, bloodsucking vampire heart, rising again and again to suck the vitality out of good ideas. But for me, there’s no heart there to begin with, just a wide gulf of emptiness and irrelevance.

  4. Oh, and about Jill Bolte Taylor, I think one reason she really scared Blackmore is that she brought an actual brain to the presentation, and talked about actual changes in her brain (in this case, a stroke) as making differences in her experiences and her own sense of compulsion about ideas and action and the like–all in all, very threatening to a memist, who like a mime, is silent on these matters and thus trapped in a glass box impossible to escape from–they don’t even see it.

    As for some of her spiritual stuff, I find it relevant more from a cultural sense–there’s no denying Bolte Taylor has become a cultural phenomenon, and I find it quite interesting to ask why as an anthropologist that is so. Do I find it personally inspiring and/or good neuroanthropology? No. But certainly germane enough to include in what we discuss here, that yes.

  5. “like a theory of humours and vapors in illness, it provides pseudo-explanations in place of just getting the hell out of the way of serious thought”…

    Ok, mention “serious thought” on cultural evolution that isn’t suspiciously close to memetics or that stands on solid evidence.

  6. Oh, I’m not too into this stuff, but I’d say ‘serious thought’ was something more like the ‘dual inheritance’ or gene-culture co-evolution material (say Lumsden, Cavalli-Sforza, Feldman, Boyd, Richerson, Dunbar, and Laland or even E. O. Wilson); at least with this frameworks, the obvious problems of overly-facile analogy between gene and culture are mitigated by a more sophisticated model. I’m not overly taken with the formal mathematical modeling, but it’s more compelling than most discussion of memes by a country mile. At least they get away from the obsessive focus on ‘memes’ as independent cultural replicators, which I think is the fatal problem in the metaphor.

    Even someone like Jared Diamond has a more sophisticated account of technological and cultural change, even if anthropologists tend to have big problems with his sweeping accounts, often because of local perturbations in the patterns that he perceives.

    Personally, I’m more of a dynamic systems theorist who would be more interested in culture change than cultural ‘evolution’; in this, I think I find more inspiration in the work of folks like Tim Ingold borrowing from eco-psychology. But I could also cite interesting examples of findings in the evolutionary study of language — there, instead of the idea that bits of culture or ideas drive their own independent processes, we have a sense that body, behavior, cognition, and social communication are mutually supporting, generating new phenomena that end up affecting their development. Neuroanthropology itself is an attempt to talk about dynamic interactions between brain, perception, biological processes, cultural effects, social interactions, institutions, and the like.

    For me, this attempt to model a dynamic including other factors is already an improvement on the culture-begets-culture model of memetics as independent replicators.

  7. There is too much in this post to tackle all of it, though I wish I had the time to engage in a serious discussion about almost every point you bring up. I would like to tease out the example you repeatedly use, the sticky ear worm (Kraft Vegemite) phenomenon. You probably know that conventional information processing accounts of this phenomenon refer to rehearsal in a specific element of working memory, the phonological loop (Bladdeley’s theory of WM). As a person who has experienced such sticky melodies, you may have been impressed with their very stickiness. They can be irritating. They can continue even when you want them to stop. A neural tic, if you will. An extreme example I have often noted is that a melody that has been annoying me for hours will go away completely when I have to concentrate, e.g. in sitting an exam, and once the concentration goes, I emerge from the exam hall, and there it is again. Now, the conventional account of neural processing in WM has absolutely nothing to say about this at all. Rehearsal is mentioned, but the rhetoric suggests that someone, or some process, is doing the rehearsal. As a dynamicist (I take it you are comfortable with that appelation?), this phenomenon reminds one a process of reverberation. Gibson, the ecological psychologist, tried to develop a notion of reverberation to account for much of neural processing and he got nowhere. However, his insights into perception and action have persevered, and are currently seen as ground breaking. Perhaps, as we move towards fuller account of the dynamics of nervous system function, and of the dynamics of the interactions between nervous systems, bodies and their environments, we might improve on the information processing account to deal better with this topic. If that is correct, then it would certainly be coherent to consider a description of the sticky tune in dynamical terms, as a specific, quasi-periodic attractor, and that mathematical definition might be applied to my tune and to your tune, and transmission might be discussed in terms of the perturbation required to induce that attractor in another nervous system. None of this seems too fanciful to me. But when that work is done, would we not talk of the propagation of that attractor from person to person? And would that not look a lot like the meme?

    I didn’t refer to Darwin anywhere there. Nor do I need to. But it suggests that there might be a coherent and useful way of talking about the propagation of ‘ideas’ from one person to the next. The notion of ‘idea’ here needs huge amounts of work. Granted.

    I’d love to have a longer discussion with you on this.


  8. Susan Blackmore, a Zen practitioner, writes in an article called “Meme, Myself, I” (New Scientist, p40-44, 1999) that ‘self’ is an illusion and that meditation is a “meme clearing meme.” Perhaps you just need a little more meditation Greg. Hopefully, then, what doesn’t look like it hasn’t happened yet won’t never not be happening.

    Don’t you just love Zen! Clear Memes we must!

    If a meme falls in a …

    HAHAHA! Okay, I won’t take this joke any further! I think I just had one of those unexpected snort-laughs where you have to clean the coffee and muffin off the keyboard (in my case, a tea and biscuits – messy nonetheless!)

  9. The label “meme” has usefulness in regarding the spread of ideas.

    In fact the term illustrates perfectly how we have come full circle through evolution. Now we aren’t just “monkey see; monkey do” creatures, but can discuss the intricacies of embedded monkeyshine threads as well!

    Actually, memes are observable at work, and predictable,as well. It’s not all dreck, but I do love your term “meme franchise,” and I can see the church being built across the street, now.

  10. Greg wrote:

    5) Could memes transfer stably?

    […] Sure, my daughter may sing the Kraft Vegemite jingle after she sees the commercial, but is she really getting the melody right? We all know myriad examples of even songs and melodies winding up with garbled lyrics.

    Maurice Bloch (2000) makes the same point about the instability of cultural elements when attacking memetics. However, this overlooks the huge number of cultural elements that are, unlike those examples of severely degraded melodies, remarkably stable across geographical space and historical time. These elements are the stuff or our very social lives.

    Take the incredibly useful and widespread meta-pragmatic device ‘Over to you’. I acquired this English phrase many years ago and use it every now and again (not very often) to pass an imaginary mike to an interlocutor, e.g. when acting as a mailing list moderator, in a classroom situation, or in an academic meeting. According to the linguistic and media anthropologist Debra Spitulnik (1996), this phrase is also widespread in Zambia, probably originating in a popular radio programme by this name but used in all manner of social contexts – often in slightly modified vernacular variants – such as weddings, choir singing and letter writing.

    How would you, Greg, and other thread contributors, explain the tremendous stability of ‘Over to you’ and myriad other sociolinguistic devices?

    Over to you.


    Bloch, M. 2000, see

    Spitulnik, D. 1996, see http://www.media-anthropology.net/Spitulnik_SocCirculation.pdf

    1. Dear John —
      I get what you’re saying, but I hardly think that the stability of ‘over to you’ proves the whole edifice of memetics. Rather, than transfer of a phrase like ‘over to you’ relatively intact is an anomaly in cultural transmission, like the transmission of melodies.

      My point about the Vegemite jingle is that EVEN THIS example of relatively stable, formulaic, encapsulated information is STILL liable to corruption, partial transmission, variation, and the like, and this is the BEST CASE SCENARIO for memetics, let alone how it might contend with things that we know are less stable, formulaic, encapsulated, and ‘meme-ish,’ like, well, just about every other part of culture.

      The extraordinary stability of a formulaic phrase is hardly all that unusual, nor do I find it that intellectually troubling. Humans are extraordinary mimics and research on chameleon behaviour shows pretty clearly that we are subject to entrainment, non-conscious imitation, and a host of other phenomena that all point back to an inclination to copy each other.

      But again, I think that this is an extraordinarily thin example to build a theory of cultural transmission, although it’s an interesting case of cultural diffusion. We find lots of very cool examples in media and pop culture, from hip hop gestures and body postures showing up in photographs of kids around the world to the spread of meaningless pseudo-phrases in each generation of students’ writing (at the moment, I’m fighting a losing crusade against a growth in ‘as such’ used nonsensically), slang, melodies, etc. All of these are interesting, but they do NOT explain the transfer of religious ideas, class-based political ideologies, cosmologies and other dimensions of culture. This extension of the example of ‘over to you’ to a whole cultural theory is the sort of thing that memeticists do that drives me nuts.

      One example to show you what I mean: reggae and ‘Rastafarianism’ are huge in the part of Brazil where I worked. Young Brazilians would often say ‘hey mon’ to me when they could speak no other English, they wore dreadlocks, wore tri-colour Afrocentric symbols, and they were terribly into Bob Marley. All these elements of culture had transferred, some of them meme-like, some of them much more laboured and conscious.

      But Rastfarianism meant myriad things to people. One Rasta ‘leader’ in Brazil told me that they wore dreadlocks because that’s how Jesus had his hair, and his ‘Rastafarianism’ was a mix of born-again Christianity with advocacy of ganja smoking. Others thought Rastafarianism was from Africa, were stupefied when I translated Bob Marley song lyrics for them (which they were trying to sing phonetically), and seemed to practicing ‘Rastafarianism’ by free association, putting anything that they wanted to the outward symbols.

      Here’s a clear case where meme-like transfer of some elements had very little impact on key structures and meanings in culture. Don’t even get me started on ‘samba-reggae’ which, to my ears, bore virtually no resemblance to Jamaican reggae. Lots of superficial copying and transfer was happening with virtually no underlying change in values (except for a positive valuation of Afrocentric aesthetics and black people’s appearances, which were both salutary, but hardly Rastafari specific).

    2. But, John, I think your final question, about explaining the tremendous stability of these sorts of sociolinguistic devices is extremely interesting, and I suspect it has as much to do with the way we remember and the mechanisms we use to communicate.

      For example, I suspect that there is some way, perhaps even a motor memory sort of way, that the phrase itself is encapsulated so that it becomes a very automatized behaviour. I’m struck, for example, by the utter repetitiveness of the way that most people answer the phone; they don’t just say the same thing, they say it with very similar tone, rhythm, and timbre, as if they were pushing a kind of mental ‘play’ button on a chain of annunciation behaviour.

      Usually, I argue that descriptions of automatized behaviour are not as formulaic as people might think; for example, athletes’ motor behaviours show a kind of responsiveness and sensitivity to the environment, as well as kinesiological variability, that demonstrates that they are not a kind of automatized ‘program’ which get executed automatically.

      In contrast, phrases like ‘over to you’ or the repetitive greeting on the telephone DO show a kind of insensitivity to context OTHER THAN the cue that provokes them.

      So, yes, John, I agree that it’s a fascinating behaviour, and that it’s transmission is one of those case studies that would fit the model of a ‘meme,’ BUT I steadfastly disagree if you are suggesting (which I don’t think you are having reread your comment a few times) that ‘over to you’ is proof of the whole edifice of memetics.

      1. i think caller-id definitively changed the way people answer phones, because they have more information on the communication that is about to happen, and can prepare appropriately. in contrast, in pre-caller-id days, you didn’t know who was calling, so you had to adopt a tone/greeting that would work in a wide range of circumstances.

        using conventions as a proof of memetics is daft: for instance, we say hellos when we meet because it establishes a known starting point, or initialises a ritual, or sets expectations that we can agree on (a ‘hello’ means something different from a ‘who the f**k are you?’). other reasons: identities, signaling theory, coordination, efficiency… i mean, really, i suppose the width of a train track is a roman meme, battling it out in the 20th century and losing it to the nasty detroit car meme, but slowly coming back due to the public transportation & sustainability memes? at this point we may as well abolish all other disciplines and pack up the university departments.

        didn’t we all learn something in middle school about theories that don’t explain anything?

  11. Thanks a lot Greg, this will keep occupied for a bit!

    I’ve just reread my post and it does sound as if were defending memetics. Sorry, that wasn’t my intention. I’m coming to this issue from the anthropology of media and from an interest in questions of diffusion/appropriation (no appropriation without diffusion, and vice versa, is what I say), the epidemiology of representations (Sperber, Boyer), social change, etc.

    Some time ago I reviewed Aunger’s book on memetics for the journal Ethnos – I’ll look up this review and get back to you.

  12. I think that if you stop trying to make memes conform to genetic concepts, and accept that the idea of memes is metaphoric (as is apt, since memes are themselves primarily metaphors, being mostly linguistic), you’ll find that calling memes invalid because they don’t behave like their metaphoric starting point, genes, is itself invalid.

    I agree that the term/concept of ‘memes’ is used with erratic abandon by its enthusiasts.

  13. Dawkins presented memes as an analogy, and he is always prepared to point out where the analogy breaks down or fails, and how memes are not the same thing as biological genes.

    Maybe some people have taken that analogy and tried to elevate it beyond its scope, which is perfectly valid to criticize. But I think as an analogy it works perfectly well and is extremely useful.

    And even if we are going to pretend it’s a full-fledged social science theory, it’s no dumber than any of the other crap social theories I had to read about in college.

  14. That almost sounds ironic, lol. So does that. But, actually, pretty much all of Richard Dawkins’s ‘anthropology’ is made up of tautology or reification – and I say ‘Richard Dawkins’ anthropology through gritted teeth, as it’s mostly all re-hashed from older, far better writers, re-introducing errors in their thinking that have already been highlighted as errors by other great thinkers, for our benefit. Oh dear – our subject appears to be devolving. Is that ‘Darwinian’, I wonder?

    The only issue is that not enough academics are challenging the fact that he presents flawed social science as natural science – and he appears to be supported by an increasingly anti-academic set of undeducated but vocal fans who class dissent from Dawkins’ views as ignorance – ironically. Both natural and social scientists should be alarmed at this.

  15. Most of these criticisms are simply wrong or apply just as well to genetic evolution.
    1. Are you seriously suggesting that human behavior can’t have causal status? In other words, you think it is impossible that “I did this because he did that.”???
    2. May be valid, but isn’t relevant to the meme concept per se.
    3. No. Nothing replicates by itself. Things only replicate in the context of their environment and phenotypes are almost never faithful copies of their parents. This is true of both memes and genes, and it makes evolutionary biology complicated, but it doesn’t invalidate the theory.
    4. Yes, there are problems of scale with defining memes. These problems also exist in defining genes. Since evolutionary theory seems to predict many things well despite these problems, there is good reason to think they probably don’t actually matter. If they do, we’ll need to throw out most modern biology along with memetics.
    5. Yes, memes can transfer stably. See, for example, Cardoso & Atwell. DIRECTIONAL CULTURAL CHANGE BY MODIFICATION AND REPLACEMENT OF MEMES, Evolution, 2010.
    6. There is plenty of theoretical work on the dynamics of host-parasite interactions, which can often transform into mutualisms when the fate of the parasite is tightly bound to that of its host. So yes, a parasite can evolve to advantage its host.
    7. Yes, this is a trivial example; just as color change in the peppered moth is a trivial example of natural selection on genes. How is that a problem for the theory?
    8. As per 4. Viral infection is rarely instantaneous; you are much more likely to get infected via continuous, close exposure, and if you are in a susceptible state, which might be induced by other, previous infections. So the analogy holds.
    9. Why is objective science inconsistent with normative judgements? Are you saying people who study malaria shouldn’t also think it’s unfortunate?
    10. Agreed. Again, this doesn’t invalidate the meme concept.

    1. Replying to some of Fluffy’s counter-criticisms:

      4) No, there aren’t problems of scale in defining genes. We use other terms for things of different scale (ie. phenotypes, chromosomes, methylation patterns, transposons, etc…), and these larger and smaller scaled things are operated on by processes which are, in many cases, distinct from (or in addition to) the processes which operate on genes.

      6) The thing is is that ideas and words are not “parasites”, or any kind of organism or virus in the context of biology. As an analogy: Solar cycles, tides, volcanos, climate, and many other events can change people over evolutionary time periods. The environment can even “evolve” in a feedback cycle with people (witness farming, water wheels, extinction-driven changes in hunting practices). Just because our feedback mechanisms with the environment alter both doesn’t mean it’s accurate to draw an analogy between genes and environmental “units”.

      My larger disagreement with this was that Gregdowney’s description of Blackmore’s speech made it seem as if she thought the future possibility of the brain as a living space for memes was how and why memes were driving evolution. The only evolutionary changes that take part as a **RESULT** of the future environment are those made via husbandry – where mates and offspring are selected by an external agent such as a farmer or ant, with an eye toward selecting organisms which are expected to flourish in the presumed future environment.

      The hypothetical memes which would be driving the evolution of the brain would only be driving it to evolve to benefit those specific memes. Sure, unintended side effects could occur, but a bigger and better brain would be more likely to drive the original memes out of existence by altering their original habitat. So if memes want to self-replicate, why are they making it so much easier for competing, future memes to replicate as well? I don’t think you see this with the lancet flukes or other parasitic and viral illnesses, except with opportunistic scavengers the equivalent of hyenas.

      “So yes, a parasite can evolve to advantage its host.”

      Has there been work on explaining how memes evolve to benefit the humans they are supposed parasites of?

      8) The thing with this is, that like the hyenas, opportunistic infections can be completely unrelated to the original infection. With memes this is unlikely to be the case. “Catching” a meme of technological advance (ie. the “Singularity”) is unlikely to make a person more prone to catching the meme of Dispensational Christianity, in fact it is likely to have the opposite effect.

      This is part of the problem with analogizing memes to genes, then also analogizing them to parasites, other symbiotic organisms, or infections.

      9) I have issues with the Gregdowney’s wording here too. No malaria researcher is going to claim they’re using objective science. The science they choose to use will be predicated on eradicating the disease, and they’ll likely freely acknowledge this.

      Analogies serve a useful purpose when first learning a subject, but the for-instance analogy of a star system, or Earth and the Moon, toward an Atom and its electron cloud breaks down so fast it could actually be harmful in understanding. That’s the issue here.

      1. Well, apparently 8 and a closing parenthesis turns into a smiley face. The 8) is me responding to Fluffy’s 8th point.

  16. Fluffy, first of all you need to pull away from the genes/memes comparison a little. That being one of the biggest sources of the problems here. A gene is a factual entity, whereas a ‘meme’ is name applied to a process of copying, mistakenly thought to apply to a ‘thing’ – usually the copied thing. So the first thing you need to consider is the fundamental structure of how you even perceive the actual ‘existence’ of ‘memes’ in the first place – before you start arguing about them in factual terms.

    Listen carefully to people like Susan Blackmore, and you’ll see that they say non-sensical things such as ‘memes exist when they are copied’. A gene exists whether it is copied or not. Get it? A ‘meme’ isn’t the thing copied, it’s the process of copying – meaning that it’s actually a viewer-applied description of an event, not a copied thing in itself. If a meme has to be copied for it to become a meme, then it’s not all clear how the factual status and existence of something can become altered by a process which has no bearing whatsoever on the original item – i.e., if, say, a phrase is copied, what actual, factual change takes place to the original phrase to turn it from a non-meme in to a meme? Answer – nothing; ‘meme’ is a viewer application of a description of a process, not of any ‘thing’. The same can not be said of a gene, which is a gene, whether copied or not.

    1. Thanks, Evans —

      Yeah, when I was reading Fluffy’s comment, I just felt like, ‘You want me to reargue the whole @#$% thing?!’ Your response is more elegant than mine would have been — “application of a description of a process, not of any ‘thing’.”

      The metaphor itself is not the problem, but the entailments that come from the metaphor can make the metaphor nor terribly apt or even obscuring. Worse, when a metaphor that isn’t terribly apt starts to slip into a substitution, then we’re likely to get confusion. The meme thing, to me, is a way of talking about cultural evolution without reading too much about culture change, cultural systems, or previous cultural evolution; you just take a metaphor and run with it, cherry-picking examples along the way.

      But you put the point very elegantly.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s