Over at Uncertain Principles, Chad writes:
He [Dawkins] rubs me the wrong way when he talks about science (I wish I could find the old post somebody on ScienceBlogs did about the vapidity of the whole "meme" thing), and really gets up my nose when he talks about religion.
I don't know if he's referring to my critique of the meme concept, but I figured it gives me an excuse to resurrect this post about memes from the old site:
I've mentioned in several posts that I can't stand the word meme. I suppose it's time I explain myself. (You can find several definitions of the word meme here). It was first coined by Richard Dawkins. According to Dawkins, a meme is "the new replicator, a noun that conveys the idea of a unit of cultural transmission, or a unit of imitation." Here's some reasons why I think the word meme is a bad idea:
- Funny, I just used the word idea. Call me old-school, but the word idea would seem to be "a unit of cultural transmission." Why make up a new word? Although, using the word meme might be a useful chick magnet (n = 1).
- Memes are analogous to genes. Ok, so what's a gene? I count at least four different meanings of the word gene. A gene can be a transcribed region of protein-coding DNA (a locus). A gene can be shorthand for a genetically heritable trait. A gene can be a particular bunch of nucleotides (e.g., regulatory regions). And a gene can be multiple loci that segregate together, such that they act as one gene, even if each locus is transcribed independently. So, what's a meme like?
- What is the mechanism of acquisition of memes? In other words, are memes particulate? When Darwin formulated his theory of natural selection, one huge problem he faced (and was unable to resolve) was the mechanism of inheritance. At the time, most people, including Darwin, thought that blending inheritance was the primary mode of inheritance: cross a tall pigeon with a short one, and you get lots of intermediate pigeons. This made Darwin's theory hard to accept, since how could a population ever diverge? What rescued Darwin's theory was particulate inheritance (usually ascribed to Gregor Mendel). Here, discrete units of inheritance (or genes) are sorted into each offspring. Thus, selection on a particular variant of a gene (an 'allele') can cause those alleles to increase in frequency, enabling divergence. So the question I have is how are memes acquired? Are they particulate? If you're a Hegelian (or you just like dialetics even if you're not), ideas might actually blend. Oh dear.
- Memes are based on Dawkins' replicator-interactor dichotomy, which is a stupid dichotomy. One of Dawkins' 'contributions' is the idea of replicators and interactors. Replicators are things that make copies of themselves, whereas interactors are the things replicators use to ensure further replication. Hence, the idea (meme? snark alert!) that 'a cell is simply a gene's way of making another gene.' Guess what? A gene is also a cell's way of making another cell. Decomposing the cell into DNA and everything else is foolish because it ignores that cells (and organisms) reproduce as wholes. I won't even broach the subject of maternal effects (the status of the maternal cell or organism influences the fitness of the offspring independent of genes). The replicator concept ignores all the cool interactive things DNA does: viral integration in the genome, plasmid integration, and gene conversion (just to name a few). Any meme concept based on such a flawed misunderstanding of biology is highly questionable.
- The meme concept is vitalistic. The meme concept implies that information is separate from the organism, in the same way, some think the mind is divorced from the brain (and rest of the body). Biological structures contain information, but are not separate from it (e.g., DNA and maternal effects, alterations during transcription and translation, etc.). Where exactly is the meme? Humanity is not a giant server onto which software is loaded. In other words, 'information' in the biological context is not independent of the organism (the computer geeks never remember that thingees called computers actually run the software).
- The information of a meme is not context independent. Two people with the same meme ("Bush is a good man") can react in completely different ways. A Republican, upon hearing this, will fall to his knees and abase himself before his personal savior. A Democrat will project vomitous fluids. The effects and acquisition of ideas do not appear to be independent of the other ideas held by the subject. Once again, what exactly is the nature of the meme?
- The meme is a conflation of cause-and-effect. Because Dawkins engages in this silly divide between replicators and interactors, he incorrectly makes genes the causal entity in natural selection. Evolution in its most reductionist form is the change in gene frequencies; however, the change in gene frequencies (or the frequency of a single gene) is not the target of the process of natural selection. The target of selection is usually a polygenic trait or an organism-level trait (e.g., "speed"). By the same logic, it is not clear that the adoption of a 'meme' is due to the 'selfishness' of the meme itself. For example, men and women have done remarkably stupid things to impress members of the opposite gender, even when they have known that said things are stupid. Here, the desire to impress is what is favoring the 'meme.' Dawkins claims that many ideas, such as religion, are the equivalent of viruses, in that they infect people to their hosts' detriment (a view of religion with which I disagree). By focusing on memes as causal units, he makes understanding human cognition that much harder.
So, I don't like 'memes.' The word doesn't add much, obscures important phenomena, is imprecise, and is vitalistic. Now, if we could do something about "paradigm shifts"...
- Log in to post comments
I see you've been infected by the "memes are vapid" meme.
Do you have the permalink to the original edition of this post? If I remember correclty, there were some interesting comments there.
Yes, this is exactly the piece I was thinking of.
Thank you-- I knew I had followed a link to it from somebody on ScienceBlogs, but as you might imagine, searching the site for "Dawkins meme" turns up a lot of things that aren't really useful.
Dawkins writes well, but I disagree with many of his points, and your critique of "meme" is spot on. I have not read his latest diatribe against religion yet, but I've read several articles about it, and I think I have a different idea of what religion is than the idea Dawkins has. There is no question that some use religion to help them gain power, money, and/or multiple wives, but that is not what religion is about at its core. I think Religion is about figuring out how you relate to the world. Atheism is simply another type of religious belief, and it can be used to justify immoral actions just as easily as belief in god can. For example, in some communist states, the official state atheism was used to persecute people for their religious beliefs. I believe in god, I think that humanity's spiritual impulses are generally a good thing, but I long ago came to the conclusion that if the only reason you can give for an action is because god wants it that way or because the rules of some "-ism" say it must be so, then you are probably wrong.
I have a discussion of many of these claims here [PDF]. But I think the vitalism claim is overdrawn. There's no need for memes to be seen as vital agents.
This post cheers me immeasurably. Hail the Mad One.
Memes do not exist.
But of course most eukaryotes do NOT reproduce as whole organisms...I, for one, was only able to reproduce as 50% of my genes (however defined). I study organisms for a living--whole animals, not cells or molecules--and I take a back seat to no one in my conviction that the whole organism is the fundamental unit of biology, but there really is a distinction to be drawn between genotype ("information") and phenotype...a distinction with fuzzy blurry edges when you look real close, but still.
Go back and reread that chapter of the Selfish Gene--Dawkins was simply trying to draw some parallels between biological and cultural evolution. Are there no valid parallels to be drawn? If a meme is a unit of cultural inheritance (and given the ambiguities pointed out in the very useful concept of the "gene" I think that's a good-enough working definition), then it's not, I think, the same concept as an "idea."
Also, in fairness to Dawkins (who's smarter than me, and as far as I can tell smarter than you, and who does, I'm sure, appreciate these subtleties and ambiguities about inheritance both biological and cultural), it has been others, not him, who have stretched the whole memetics thing farther than it ought to be stretched. Dawkins keeps writing in an interesting and thought-provoking way mostly about biology--and, sometimes, religio (is that perhaps what you really don't like about him?).
religio + missing n = religion
I actually like the word "meme" a lot, but I understand it as a whimsical rather than a rigorous analogy to "gene". In that sense, it has a more precise meaning for me than the generic "idea" does. IMO the problem arises when people try to fast-track this perfectly useful coinage into the foundation of a science analogous to genetics. A meme isn't a gene. It's an abstraction of an abstraction, and any similarity a meme has to a gene is severely constrained by the fact that we can't map a meme to a molecule.
Of course, I'm also an unrepentant neologism-lover; if language couldn't grow, we'd still be communicating via grunts and primitive foot gestures. :-) Dawkins may have been taking himself and his coined word more seriously than is warranted (it's been years since I read the source), but that doesn't mean the word is completely useless as an informal term.
I don't know why this is bothering me so much--I really don't care that much about the meme concept--but your "critique" is of a caricature. Nobody, certainly not Dawkins, has ever claimed that a meme is precisely analogous to a gene in every detail, nor that cultural evolution exactly mirrors biological evolution. So what if cultural inheritance includes blending effects? So what if there is no good analog to maternal effects? Are you trying to criticize a simplistic concept of the "gene"? Because you're not addressing what Dawkins means by a "meme."
My daughter was practicing a schoolyard rhyme/hand-dance thing at the bus stop this morning, and I think that's an excellent example of a meme. It's not the same thing as an "idea." It passes kid to kid (not, to be sure, by reproduction per se), and it evolves as "mutations" are introduced and replicated. And yes, it is information that exists, in some sense, independent of my daughter's brain.
I just think your critique is extremely wrong-headed, and now I'll shut up.
CCP,
A quick response, because the topic requires a post all of its own. I think a fundamental problem Dawkins scientific writings suffer from is that he took G.C. Williams ideas about genic selection and took them to a ridiculous extreme. There is an entire literature within the philosophy of biology that deals with the fallacy of the replicator/interactor dichotomy. It's not my view of the concept of the "gene" that's simplistic; it's Dawkins. So if I have any bias against Dawkins, it's due to his science, not his thoughts on religion (which I find irrelevant).
Fair enough. That's really what I thought was going on (that crack about religion was probably better directed at Chad Orzel)...your post is really directed at Dawkins's gene concept, not "memes" at all. And I agree with you about genes! But don't mung up the meme concept (for which, as Julie Stahlhut points out, Dawkins had much less grandiose intentions) with an extremist genist viewpoint.
I maintain again that a.) cultural artifacts do something a lot like evolving, b.) something like "genes" do indeed serve as heritable units of information in sexually reproducing systems, and therefore c.) it's reasonable to draw the analogy and call units of cultural inheritance something; Dawkins chose "memes." It's not the same thing as an "idea"! A meme CAN be an idea--that's one type--but a meme can also be a nonsense rhyme, a melody, a catchphrase, a dance craze, a quotation, a way to use a screwdriver to open a beer bottle, etc. Show me someplace where Dawkins claims that a meme is exactly analogous in every way to his (or anybody else's) concept of a DNA-based gene.
OK, NOW I'll shut up.
Memes do not exist.
They belong in the pseudo science list with multiverses, String Theory, and SETI.
I think CCP's distinction between an 'idea' and a 'meme' is valid and do not agree with you that these two words can be interchanged without changing meaning of the sentence using it.
Let's remind oursevles of what an idea is. From the Oxford English Dictionary, the definition of 'idea' is:
1. a thought or suggestion about a possible course of action
2. a mental impression
3. a belief
4. (the idea) the aim or purpose
(Interesting to note that the orgin is Greek: 'form, pattern')
A whistled tune is not an 'idea' (but it could be described as a pattern). When someone hears and mimics this tune, has 'an idea' been passed along? I don't think so. Do you?
'Wassup!'. If this is an Idea, it's a bad one ;-). No, of course this isn't an Idea, it's just annoying sound that means 'hello' and has done the cultural rounds passed along from idiot to idiot (including me) via TV initially and then p2p, via email, blogs, phone calls, etc.. Is 'Wassup! a belief? Or a thought about a possible course of action even? No, this is a good example of how a unit, a pattern or a 'meme' has been immitated and gone from one empty head to another.
As CCP points out, an idea *can be* a meme - but that doesn't mean a meme = idea. The latter is a subset of the former. I don't see this being a difficult concept (def: an abstract idea) to grasp (or, a meme to infect you with).
Memes are not the same as ideas because ideas are not defined as being socially transmitted. The enormous popularity of thee meme concept illustrates the need for such a term. Memes are no more vitalistic than genes are. Genes are not "context independent" either. I could go on - but you probably get the point.