Are memes just the extended phenotype?

March 11, 2007 10:46 PM

How are memes anything other than an extension/expression of the extended phenotype?

And I mean "phenotype" in the generalized-Richard Dawkins-sense. It seems the meme meme is taken so seriously by so many, but I can't figure out how they are anything other than an interesting thought experiment/model.

Could someone please step off the meme-wagon and explain it to me--are memes anything other than an extension/expression of the extended phenotype?

I'm confused about this, so I hope I'm not sounding like an idiot--please forgive me if I do. Be gentle.

posted by whatgorilla to science & nature (20 comments total)

The point of the meme is just that ideas are subject to the same pressures and evolutionary processes that biological traits are. For instance, religion that encourages proselytizing will be more "successful" than one that does not, similar to how an extremely fertile organism will tend to outcompete one which is less fertile (all else being equal).

The word phenotype generally refers primarily to biological traits. You could say that ideas are really manifestations of biological processes, but saying memes are just phenotypes is like saying the ideal gas law is just Newtonian mechanics.
posted by aubilenon at 11:08 PM on March 11, 2007

However, the big difference is that biological evolution is Darwinian, whereas memetic evolution is Lamarckian.
posted by Steven C. Den Beste at 11:52 PM on March 11, 2007 [2 favorites has favorites]

In my opinion memes are basically bullshit, it really made me pissed off that Dawkins had to ruin "The Selfish Gene," with this ridiculous concept tacked on at the end.

The main problem with memes is that they are impossible to define. A meme can be anything from a sentence, to a song, to a concept or a preference. Furthermore the way memes are transmitted from one person to another is also impossible to know. Is it only necessary for you to hear a statement for the meme to be passed on to you? Do you have to remember it? Do you have to believe it? And as Den Beste pointed out above it is possible that memetic evolution is lamarckian, meaning that it is basically impossible to use what we know about biological evolution to say anything interesting about memes.

Basically there are two different meanings of evolution, the first is evolution in a strict biological sense which requires that some sort of genes get passed on and are expressed as phenotypes. These genes will only be changed by random variation and natural selection. The other meaning of evolution is merely change over time. So one can say that music, history or skate boarding has "evolved". This is the level that memes are at, they are not the equivalent of genes. I could go into a rant about how culture has allowed humans to make great changes in how they live free from evolutionary principles, but this comment is way to long anyway.
posted by afu at 12:34 AM on March 12, 2007

Some parts of culture can be viewed as part of the extended phenotype. The difference between those cultural elements and memes lies in how they reproduce.

A cultural practice such as kin affiliation gives a reproductive advantage to the genes of people who carry it. That makes it part of the extended phenotype. Those cultural elements reproduce by being passed on from parents to children, along with the genes.

Memes reproduce by being passed from person to person, independently of genes. A meme succeeds by being good at transmitting itself, independently of whether it is good for the genes of its "host".

That's not the same as Lamarckian evolution. Lamarck theorized that learned changes could be passed biologically from parents to children. Memes rely on communication, not inheritance, to reproduce.

That's the theory. My personal feeling is that memes are useful today only as a metaphor: what if we looked at culture as a process that behaves like biological evolution? My impression (experts, please correct me if I'm wrong) is that we still don't have a complete theory of biological evolution that rigorously defines the necessary and sufficient conditions for evolution to occur, and quantifies the behavior of the evolutionary process. So it's not surprising that the meme theory doesn't have those kinds of rigorous definitions either. Take it as a potentially useful perspective for viewing certain types of cultural behaviors.
posted by fuzz at 1:53 AM on March 12, 2007

My personal feeling is that memes are useful today only as a metaphor: what if we looked at culture as a process that behaves like biological evolution?

You are correct. There is a lot of talk about memes, but very little serious academic work and none (that I am aware of) that makes it more than just an interesting idea. I think it was the philospher Mary Midgley who commented that - according to meme theory - ballet routines mutated, reproduced, spread and "infected" dancers, but it was unclear what this meant or whether it was useful in any way.

To be fair to Dawkins, I think it is his followers and vocal fans who have been the most bullish about memes. I'm thinking here about Susan Blackmore, who I saw speak on memes a few years ago. A question from the audience challenged that meme theory was essentially unprovable and non-falsifiable. She got very flustered and insisted there were experiments that proved memes existed - she just couldn't think of any offhand.

we still don't have a complete theory of biological evolution that rigorously defines the necessary and sufficient conditions for evolution to occur, and quantifies the behavior of the evolutionary process

I'd say that we do, but it's an arguable point. There's a lot of math and statistics you can do with evolution. However, "memes" are so different and ephemeral that none of this is applicable. What's the mutation rate of an idea? For that matter, what is an idea?
posted by outlier at 2:23 AM on March 12, 2007

"Meme" is a fancy and rather imprecise term for the very elaborately theorized concept of "culture." Biologists hate to admit the force of culture on human evolution, and want to reduce it to a positive variable rather than a determinant of human social organization, feelings, and expressions. Hence they come up with useless, reductionist concepts like "meme."

You want to know from "culture," it's a different literature.
posted by fourcheesemac at 4:56 AM on March 12, 2007

"Meme" is a fancy and rather imprecise term for the very elaborately theorized concept of "culture."

Rather, it's a fancy and imprecise term for units of culture. And what, then, is "culture"? Ask a hundred different anthropologists (trained at a hundred different schools) and you'll get a hundred different answers. Little "c" culture itself, for all of its elaborate theorization, still doesn't have a unified definition because itself is a fancy and imprecise term referring to everything extraphenotypic about people. The most well-known definition is the one given by E. B. Tylor in the late 1800s, a period not exactly known for its good social science. Since then, "culture" has been an integral part of anthropology, for better or for worse.

So, OP, "phenotype" refers specifically to the expression of genetic traits, and culture is not at all controlled by genes, and so memes, however poorly defined they may be, are not extensions of the phenotype.
posted by The Michael The at 5:31 AM on March 12, 2007

outlier: There is a lot of talk about memes, but very little serious academic work...

... most likely because the concept is so imprecise as to be inaccessible for serious research.

AFACS, fourcheesemac has it pretty much nailed: "Meme" is a fancy way to talk about units of transmissible meaning that are deemed, in a particular context, to be worthy of passing along. What the meme "wants" is irrelevant; people pass along a meme, or don't. Why they do that (or don't) is not a property of the meme. It is, rather, a product of the circumstance, which most notably includes the person -- the actor -- passing along the meme, as well as many other factors.

As with genes, Dawkins in his casual usages conflates the driver with the car: You can't drive to Montreal unless you've got something to drive in. You could ride there in a bus, or on a tortoise, but you can't drive there unless you're driving something. Something similar is true for genes and organisms: The genes do not get anywhere unless there's a body to facilitate that.

The concept of the meme is so imprecise that it's very confusing, IMHO -- and made more so by the fact that it does not appear to be confusing -- to try to draw analogies with genetics. It's one of those cases where a simple analogy has proven so powerful that it permeates our zeitgeist (yes, I realize I'd describing a meme) to the extent that it shapes how we are able to talk about ideas. The problem is that this idea has tainted the way we can talk about it. We can't discuss the transmission and natural selection of ideas without invoking Dawkinsian memes, any more. It's a pain in the ass, because the Dawkinsian conception of the meme is essentially rooted in magical thinking. Not that Dawkins would ever admit that.

But, probably better to ignore me and go read Edward O. Wilson on the same subject. He may well be completely wrong, too, but at least he's a hell of a lot smarter than Dawkins.
posted by lodurr at 8:03 AM on March 12, 2007

In Edge 198, Scott Atran explains very clearly The Trouble With Memes (scroll down, around the last fourth of the ginormous page):

For genes, there is an operational definition: DNA-encoded units of information that dependably survive reproductive division, that is, meiosis (...). In genetic propagation, information is transmitted with an extremely high degree of fidelity. In cultural propagation, imitation is the exception, not the rule; the typical pattern is of recurrent, guided transformation.

A few years ago, trying myself to come to grip with memes, the best I could do was a metaphor (which has its own limits):

Memes are best thought as waves or ripples on a surface of water: you can observe one, photograph it, surf on it or have your boat rocked by it but it doesn't have any autonomous existence. You can't measure them out of context because they are contextual manifestations, not autonomous units. Exactly as waves are manifestations of a temporary interference between water and wind (there can be other factors), memes are a manifestation of a temporary interference between memory and knowledge.

Memes are a fun concept to play with, but as soon as you are trying to give it a scientific ground, it becomes a misconception and, if you insist, a fallacy.
posted by bru at 8:14 AM on March 12, 2007

I agree with TMT and 4cheese, memes are the reductionist approach to culture. I don't really buy into the utility of the theory, but one venue where there has been some intriguing application is archaeology, where the extended (physical) phenotype a.k.a. "material culture" is the main dataset, and, therefore, memetic theories have been very focused on the "tangible meme" in a historical-explanatory sense not terribly different from a palaeontological one. A good expression of memes in archaeology is a book by my former advisor, reviewed here
posted by Rumple at 8:40 AM on March 12, 2007

Memes are an extension/expression of the extended phenotypes, in that the expression of the human phenotype gives rise to the environment in which memes exist and memetic evolution occurs. But to say they're just that would be like saying that genes are just an extension/expression of chemistry.

One of Dawkin's points in The Selfish Gene is that we'd expect to see evolution by natural selection in any environment that featured variation, selection, and heredity. The idea of memes arose through his thinking about what other environment featured such a thing.

I highly recommend Susan Blackmore's The Meme Machine for an exploration of memetics.
posted by Zed_Lopez at 10:57 AM on March 12, 2007

in a particular context, to be worthy of passing along. What the meme "wants" is irrelevant; people pass along a meme, or don't. Why they do that (or don't) is not a property of the meme.

I am no expert, but I think that is contrary to the basic tenets of those who write about memes "seriously." Richard Brodie, author of Virus of the Mind, says that "ideas will spread not because they are "good ideas", but because they contain "good memes" such as danger, food and sex that push our evolutionary buttons and force us to pay attention to them."

Some of the objections to the notion of memes appear to be the usual complaints of hard scientists confronted with the messy world of the social sciences. If nothing else, there is room for some delight in the effectiveness of the concept of "meme" as a meme itself. It may not be the stuff of Darwin, but it has certainly took hold in the popular imagination.
posted by Lame_username at 11:35 AM on March 12, 2007

Zed_lopez: One of Dawkin's points in The Selfish Gene is that we'd expect to see evolution by natural selection in any environment that featured variation, selection, and heredity. The idea of memes arose through his thinking about what other environment featured such a thing.

The problem is that Quantitative Genetics depends on some very specific properties of how variation, selection, and heredity work. Many of those properties don't seem to apply to "ideas" which we know from work in cognitive psychology are ephemeral and reconstructed on an ad hoc basis.

lodurr: We can't discuss the transmission and natural selection of ideas without invoking Dawkinsian memes, any more.

Memes are only necessary if you are using something like Darwinian natural selection as your framework. Quite thankfully, only a few people feel that is the best way to explain the transmission of concepts and behaviors within a culture. You can read thousands of pages on communities of practice, social networks, Vygotskian activity theory, situated cognition, information transmission theory, and linguistics without ever having to parse the ugly word "meme."
posted by KirkJobSluder at 3:27 PM on March 12, 2007

This thread is filled with misinformation. Evolution and criteria for evolution have been described in rather general terms in mathematical biology. Richard Dawkins' Idea of Universal Darwinism is encoded into a family of equations known mainly by the ubiquituously used replicator equation, applied in evolutionary game theory to biological situations as well as economic and strategic interactions. All that is required for evolution is replication.

Despite the many opinions to the contrary, there is no solid argument against memetics. Although it is correct that the definition is of a meme has not stabilized, it would be rather presumptuous to dismiss the entire framework based on that alone. Gregor Mendel had no idea was a gene was when he conducted his pea-plant experiments but that did not stop him from observing the evolutionary patterns in action. (Similarly, mathematical concepts are often researched for decades before the final definitions are laid down.) Technology has not advanced enough to allow us to see how information is stored in the brain, and hence a final definition of what the units of culture are will have to wait (if indeed culture is transmitted discretely).

Some authors, such as Dan Sperber, propose continuous models rather than the discrete model of memetics. These models are on as weak of footing as is memetics. If there was a sufficient model of culture that fit all the observable evidence and was falsifiable then there would be no controversy.

In terms of expressive power though, nothing comes even close to memetics. Memetics has not been made into a predictive theory yet and hence has not been scientifically tested. It is probably possible with current mathematics and rich sources of data such as the blogosphere to try to model memes as replicators, but it appears no one has seriously made the effort.

From what I have read, most people tend to dislike the idea because it challenges their own beliefs regarding free will and religion, and hence they would rather deny it than consider the alternative. People like to think that they are in control of their own minds and actions. Additionally, many objections are poorly thought out or easily dismissed as inaccurate. (For example, see Gould's objections.) Often opponents attack authors rather than ideas, such as the comment about Susan Blackmore above; memetics is not its authors, and to attack it on those grounds is a logical fallacy and deceptive. (Note that one could view these attacks as memetic defense mechanisms....)

I encourage the readers of this thread to consult more legitimate sources, such as Daniel Dennet's "Darwin's Dangerous Idea" and "Consciousness Explained". For a much higher level of discussion, see "Darwinizing Culture: the status of memetics as a science." Make your own well-informed decision.

Other corrections:
(1) Memetic evolution is not Lamarckian.
(2) Humans do not live free of evolutionary pressures -- pathogens are constantly affecting the distribution of genes in the gene pool, and there are many other factors at play. (We may not be running from saber-toothed tigers, but we are still very much under the influence of natural selection.)
(3) Imitation is a major mechanism in the transmission of culture and has been used to create mathematical models for the acquisition of language. (Even Chomsky's Universal Grammar has been studied evolutionarily.)

It is worth noting that memetics has been independently conceived multiple times, for instance recently by physicist Aaron Lynch, whose book "Thought Contagions" was nearly published before he ever encountered the idea of a meme or any of Dawkins' Work.

To actually address the question, all of human culture is part of our extended phenotype. Every road, canal, building, satellite, and institution is all part of our effect on the environment. Memes (if they exist) exist in our minds and hence are an effect of our existence, although computers and intelligent machines may change that someday...
posted by mharper3 at 9:24 PM on March 12, 2007 [1 favorite has favorites]

The Michael The: I am an anthropologist. I've read hundreds of definitions of culture, and published an extensive one myself. So the lecture is a bit misdirected.

As for: culture is not at all controlled by genes, and so memes, however poorly defined they may be, are not extensions of the phenotype.

I'll debate that. Culture is "controlled by genes" in the very precise sense that it is dependent on abstract symbolic capacities of the species, and language especially, that are the direct consequence of genetic adaptation. More than that, culture has been driving (or accelerating, or altering) biological evolution on earth since it emerged. It has even evolved to the point where it has enabled humans to intervene directly in the genome of any species on earth, though this is an extension of tens of thousands of years of animal husbandry and agriculture and selective human breeding.

The consensus, among those hundreds of definitions by anthropologists, is that culture is a biologically determined faculty of the human species, that has altered our biological destiny. This is not to say its ideational and abstract aspects are superstructural, but in Kroeber's famous term, they are "superorganic." Culture is real. By virtue of the fact that it is real it also must reside in the material world. It resides in the material world in the form of words and social institutions and ultimately in the form of genetically encoded, species-specific capacities of the human body, brain included.
posted by fourcheesemac at 9:22 AM on March 13, 2007

mharper3: All that is required for evolution is replication.

...and some relatively stable unit of information that is passed between generations. It's this "relatively stable unit of information" that is most problematic when applied to cognitive psychology.

Despite the many opinions to the contrary, there is no solid argument against memetics. Although it is correct that the definition is of a meme has not stabilized, it would be rather presumptuous to dismiss the entire framework based on that alone...

Even when there are other frameworks that 1) have better explanatory power, and 2) have better reliability and validity?

People have been tracking units of behavior, ideas, and language through cultural spaces for over a century. You can use behavior, you can use linguistic signs, you can scale the recognition->adoption of concepts. All three of these have a rich body of empirical research behind them.

Until someone quits fooling around and develops a memetic theory that has predictive power for educators, and communications professionals, they are not obligated to abandon better theories for memetics.

Technology has not advanced enough to allow us to see how information is stored in the brain, and hence a final definition of what the units of culture are will have to wait (if indeed culture is transmitted discretely).

Which is a red herring and you know it. Neither Darwin nor Mendel had any idea about the physical coding of genes. Mendel's genetics, (and the messier quantitative genetics that followed) could be derived simply by observing the results of genes. Likewise, we have a rich understanding of how behaviors and concepts work simply by watching them in action.

In terms of expressive power though, nothing comes even close to memetics. Memetics has not been made into a predictive theory yet and hence has not been scientifically tested.

Gee, talk about a contradiction in two sentences. Memetics has no expressive power, because it is not a predictive theory and has no scientific evidence. In contrast, the evidence for other frameworks (although imperfect) is more than enough to make them more useful, and more worthy of exploration.

I mean, for pete's sake. Even naive behaviorism as woefully inadequate as it is does a better job of explaining learning than memetics.

From what I have read, most people tend to dislike the idea because it challenges their own beliefs regarding free will and religion, and hence they would rather deny it than consider the alternative.

No I dislike the idea because it's had 30 years to produce something useful, and by your own admission, it has utterly failed to do so. In contrast, Social Networks theory has exploded over the last 30 years and produces practical tools.
posted by KirkJobSluder at 9:38 AM on March 13, 2007

there is no solid argument against memetics.

I'll bite. What's the solid argument for "memetics," whatever that is?

Nothing I have seen written on the subject of "memes" has been anything more interesting than the most rudimentary (and context-free) semiotic models for culture, language, and communication. It's a poor man's (or a crude determinist's) version of culture theory. If we knew the basic unit of culturally transmitted information to the purported degree of specificity implied in the concept of a "meme," it would be another story. I can offer lots of candidates, none of which exclude the others. To begin with, I offer up the phoneme, the word, the sentence, and the concept. Then I suggest the institution, the social role, and the unit measurement of material exchange. Or perhaps we can investigate the amplitude of microvoltage electiral pulses in the brain, or the composition and admixture of various neurotransmitters. Finally, I suggest adenine, thymine, cytosine, and guanine. Lots of sciences have thought about the problem of information transmission in human evolution and in evolution more generally, in very different ways. "Meme" is one of the more pseudoscientific and reductive concepts to emerge from this effort.
posted by fourcheesemac at 2:50 PM on March 13, 2007

I had a wonderful experience the other day where I realized what a meme really is while discussing the creation of the Corporation.

The Corporation is a meme. It is an idea, a concept which has evolved over time into something which has all the rights of a sentient being.
posted by DecemberRaine at 4:43 AM on March 14, 2007

DecemberRaine: You might enjoy a documentary, released in 2004, called The Corporation. You won't hear the word meme in it, but it's right along the lines of your comment. Specifically it will describe how corporations attained the status of 'legal-person' and the consequences.

fourcheesemac: I did not say that there was any argument for it. The word is not the meme because the word is just a representation of a meme. Same for 'the sentence' and similarly for the 'phenome'. A meme is not the same a representation, just as a musical piece is not the same as its representation as sheet music. You go on to suggest possibilities which would be collections of memes or further superstructures, not yet feasible, or clearly intentionally nonsensical. (To get you started understanding what level the meme is at, I suggest that you think about bacterial plasmids, which are replicators having both horizontal and vertical transmission and can consist of a single gene or several genes.)

KirkJobSluder: Your first mistake began immediately when you claimed that evolution requires "...some relatively stable unit of information that is passed between generations." Perhaps then you can answer the following question: where is the information stored in a prion, a pathogen consisting solely of protein? There is no nucleic acid in a prion, and if you say that the information is stored in its three dimensional structure I expect that you will give as an exact specification of how it is stored as you demand that memetics specify about memes.

While you are at it, maybe you can also inform us where the information is stored in an RNA virus? (By which of course we are speaking of the kind that are merely naked RNA strands, which typically occur as plant pathogens, not RNA retroviruses.) Is the information the molecule itself? That would force recognition that a meme could also be its information.

For yet another example, consider Spiegelman's monster, an RNA strand approximately 200 bases in length that was produced from an original strand of length 4500 by allowing it to replicate freely in a nutrient rich laboratory environment until it reduced to the smallest length possible to maintain replication. Where does it store its information?

"I mean, for pete's sake!"
posted by mharper3 at 4:30 PM on March 17, 2007

Thanks for all the answers--I guess my issues with the meme meme are justified...in so much as there's no general consensus here. A professor friend, Dr. Sam Kimball, just finished a book of evolutionary literary theory called The Infanticidal Logic of Evolution and he expressed his belief that the meme meme was utter BS.

PS--here's more about his book:
"Infanticide is difficult to contemplate, and yet the Greek and Judeo-Christian traditions repeatedly allude to it. Why? Why does Kronos eat his children? Why does Agamemnon sacrifice his daughter Iphegenia at the outset of the Trojan War? Why does Laius pierce the ankles of his newborn son Oedipus and order him to be left to die? Why is the defining test of Abraham his willingness to sacrifice his son? Why is it important in the life of Jesus that Joseph's departure from Bethlehem is followed by HerodÂ’s slaughter of the innocents? Why in Revelation 12:1-6 does the Â"great portentÂ" involve a pregnant woman in front of a dragon who is about to devour her newborn? What can it mean that Adam and Eve eat the fruit of the Tree of Knowledge when everywhere else Â"fruit of the wombÂ" means child?
According to The Infanticidal Logic of Evolution and Culture, to answer requires confronting the problem that existence costs (the two words are cognate). The problem is that, since living things must shift more of their energy costs onto the world, including other living things, than other living things are able to divert or impose upon them, there is simply no life that is not a death threat to some (other) life, hence to countless possible futures.

Developing this premise by distinguishing between infanticide as empirical phenomenon and the infanticidal as the logical form of phenomena that may or may not eventuate in an infant's death, Kimball demonstrates in his opening chapters that evolution proceeds infanticidally. The key concepts of evolutionary theory—reproductive fitness, adaptation, survival, mutation, and extinction—all presuppose the general infanticidism of the evolutionary economy.

In the seven central chapters of his study, Kimball shows how the apprehension that the material conditions of existence are inescapably infanticidal is central to the most cherished, heritage-defining narratives of the Judaic, Greek, and Christian traditions. Genesis, the Odyssey, Oedipus the King, and the New Testament accounts of the Eucharist attempt to name what Western culture has resisted—that the infanticidal is the condition of the possibility of being, indeed, of the sacred.

What then becomes of the human future? In his final chapter, Kimball turns to the films Star Trek, The Matrix, Terminator 2, and Alien Resurrection for their depictions of humans threatened by their technological or hybrid progeny. Only Alien Resurrection, however, faces the infanticidal cost of humankind's victory over the life of the alien and the alternative futures this life figures. It thus rediscovers the question that, Kimball claims, echoes throughout western culture: How to heed the inconceivable responsibility these (infanticidally) lost futures impose on the living?"
posted by whatgorilla at 8:40 PM on March 19, 2007