Tags
Many people have observed that you can model the growth and decline of ideas as an evolutionary process. They begin as mutations of other ideas; they have traits that help them to survive or reproduce, and other traits that might make it more difficult; the ones that are more fit tend to flourish, and the ones that are less fit die out. This observation is sometimes called ‘memetics.’
I’ve noticed that talking about “memetic fitness” can refer to one of two completely different things.
First, it can mean how many people believe an idea. I think this is the one people originally meant by “memetic fitness”. I believe ideas like “God doesn’t exist”, “you shouldn’t judge people for their sex lives as long as they aren’t hurting anyone”, “ideas can be modeled as an evolutionary process”, “you can save a human life for three thousand dollars”, and “the sky is blue”. I can try to spread these ideas for other people: for instance, I can share a link to GiveWell’s top charities. Ideas which tend to be more fit in this sense include ideas which fit into people’s preconceptions, true ideas, ideas that justify evangelizing to other people, useful ideas, and evidently by revealed preference ideas told in ALL CAPS BOLD on Tumblr followed by a Japanese emoticon. ٩(ര̀ᴗര́)
Second, it can mean how many people talk about an idea. To pick an obvious example: Elevatorgate. Elevatorgate is not a conventional “is” or “ought”, the way that the ideas I talked about in the previous paragraph are. Neither side in Elevatorgate is particularly conventionally memetically fit: it seems like everyone basically continued to have the opinions they started out with. And yet when it flared up I used to entertain myself by sneaking tangential references to Elevatorgate in my blog posts and seeing how long it took the comments to descend into Elevatorgate madness. In case you’re wondering, mentioning the word “elevator” in an entirely unrelated comment in an entirely unrelated post was enough to cause it to break out.
There is a sense, I think, in which Elevatorgate is absurdly memetically fit– not as a set of ideas but as a topic of discussion.
This is sort of the general case of toxoplasmosis of rage. Scott points out that the topics of discussion which are the most memetically fit topics of discussion are often the divisive ones that make everyone get angry and want to scream at each other. But I don’t think that all absurdly fit discussion topics fall into this category. For instance, Kim Kardashian has made her entire career out of being a very fit topic of discussion, even though she is not a particularly controversial person (everyone seems to pretty much agree that she sucks). And people can certainly viciously argue with each other about Keynesianism and monetarism, but in most social groups they are nowhere near as memetically fit a topic as anything ending in -gate. That’s probably because monetary policy is confusing and difficult and involves words a lot of people don’t know the meaning of, whereas nearly everyone has hit on someone or wanted to hit on someone and therefore feels entitled to have an opinion on the subject.
The problem with memetically fit topics of discussion is that they are unlike memetically fit ideas. For ideas, memetic fitness is at least correlated with truth. The most memetically fit ideas are ones like “trees exist,” which we don’t even think of as ideas because they’re obviously true. Also, if you disagree that memetic fitness is correlated with truth, then you have no reason to believe any of the things that you believe and are forced to become a radical skeptic.
There is no guarantee, however, that memetically fit topics of discussion are actually important. In fact, most memetically fit topics seem to be really unimportant. As I write this, my Facebook trending news stories are about a New York Post critic who thinks that women don’t get Goodfellas and a virtual reality headset shipping with the Xbox One controller. I would contrast this with examples of important, memetically unfit topics of discussion, but then I realized that they’re memetically unfit so I don’t know what they are.
Memetically fit topics of discussion tend to be divisive. They tend to be things that everyone feels qualified to have an opinion on (thus, social issues more than economic or foreign policy). They tend to be interesting or at least to give everyone Someone Is Wrong On The Internet syndrome. But they don’t tend to be particularly important, probably because important things probably affect a lot of people, and people are bad at dealing with things that affect a lot of people.
What do we do about this? I’m not sure. I myself tend to get into vicious arguments about things far less important and more interesting than things I don’t get into vicious arguments about. But at least we should maybe take a step back and say to ourselves “is arguing about the plight of sad nerdy guys who can’t get a date really the most important topic I could be arguing about?”
Thing of Things 
I don’t think arguing on the internet has much value besides being fun. By all means pick candidates based on economic, foreign, and environmental policy rather than their stance on gays, weed and abortions, vote in local elections, donate to AMF, etc. but beyond that sort of stuff, I don’t think most people can really do much about important memetically unfit ideas, and arguing about them on the internet doesn’t seem like it’d be a huge improvement over arguing about whether astronomers should be allowed to wear five guys shirts on elevators, especially if talking about important things will make people bored.
LikeLiked by 1 person
“By all means pick candidates based on economic, foreign, and environmental policy rather than their stance on gays, weed and abortions”
. . . Why? Do you think they won’t have much influence on the latter?
LikeLiked by 1 person
Roe v wade prevents US politicians from really affecting abortion law, and abortion law doesn’t really affect abortion rates all that much. Gay marriage, non-discrimination laws, etc . are important, but not nearly as important as keeping the planet habitable, reducing poverty, and avoiding wars. I might have underestimated the importance of weed legislation,but the drug war is still probably less important than actual wars
LikeLike
“Gay marriage, non-discrimination laws, etc . are important, but not nearly as important as keeping the planet habitable, reducing poverty, and avoiding wars.”
I’m . . . not sure I agree with that.
LikeLike
@Hainish
Why? That seems pretty straightforward: even the highest levels of homophobia, misogyny, and racism can be undone over the course of few decades, as the history have shown. Runaway greenhouse effect, on the other hand, is not, and recovering from severe economic crises can take just as much if not longer than recovering from going full right.
LikeLike
@Maxim, economic prosperity, peace between nations, and environmental regulations can also be undone over the course of a few decades, I think. (Also not sure that history shows quite what you say it shows.)
I’m not seeing the straightforward argument here.
LikeLike
@Hainish: It may be possible to undo economic prosperity, peace between nations, and environmental regulations over the course of a few decades, but the same cannot be said for economic adversity, war between nations and environmental transgressions. So, it makes sense to give greater weight to these while electing candidates (but not necessarily in general) because people who believe gay marriage is an abomination are going to die out eventually. (The case for abortion is a little stronger because it is a health issue and I’m uncomfortable pitting it against economic well-being).
LikeLike
Gay marriage directly effects only a vanishingly small number of people.
It is a totem for social liberals, not an important political policy.
LikeLike
@shaugahai, we will have to agree to disagree on that.
(Well, no, not really . . . but it saves time to pretend.)
LikeLike
“It may be possible to undo economic prosperity, peace between nations, and environmental regulations over the course of a few decades, but the same cannot be said for economic adversity, war between nations and environmental transgressions”
I dunno, the world got a lot better really quickly once World War II ended, and that lasted less than a decade…
LikeLike
The memetic fitness of a topic of discussion is correlated with its importance – the vast majority of memetically unfit topics of discussion are nonsensical and therefore unimportant (e.g. should Bob eat fifteen parrots).
LikeLike
It becomes an entirely different kettle of fish once you start asking “Should bob be allowed to eat fifteen parrots”, which is likely to come up soon once you start the original question. And since this one is ideologically divisive (Liberalism vs. Animal Protection) you’re likely to have a Godwin in less than five moves.
(I hope I didn’t start a flame war here…)
LikeLiked by 5 people
The thing about thinking about “trees exist” as “memetically fit” is that it reproduces a lot more by spontaneous generation than by an instance of it reproducing itself — you see a tree, and you realize trees exist a lot more than someone convincing you that trees exist.
A similar thing biologically is calling rocks “evolutionarily fit” — the pattern of “rock” certainly successfully appears a whole lot, but it just seems wrong to call that “evolution”. Rocks don’t produce rocks, and “trees exist” doesn’t really produce “trees exist”.
LikeLiked by 2 people
“Electrons exist” is more like it.
LikeLiked by 1 person
Sure it does. We tell our children that those tall brown-and-green things are “trees”, and they’re a kind of plant, and …
LikeLiked by 1 person
No, you’re talking about the belief that there are things that are called “trees”, not the belief that trees exist. In most parts of the world, the latter is acquired before you know the word because you can see the damn things. That belief is caused by trees, not by the fact that the parent has the belief. The latter is quite irrelevant to its production, and so this production is not really reproduction.
LikeLike
A recently demised commuter tabloid in my home city had a “Boring but important” column, buried a few pages in, for a handful of actual news stories amongst all the coverage of Kim Kardashian and twitter trends.
I imagine in the modern era there’s a nontrivial proportion of the world’s population who may not have actually seen a tree, or at least, not see one often enough in their lives to be as confident of it as most people are.
But regardless, its trivial to modify the example: “London exists”, “the Earth is round”, etc.
LikeLiked by 1 person
Maybe London exists, but does Bielefeld?
For those who wonder: http://en.wikipedia.org/wiki/Bielefeld_Conspiracy
LikeLiked by 2 people
“I imagine in the modern era there’s a nontrivial proportion of the world’s population who may not have actually seen a tree, or at least, not see one often enough in their lives to be as confident of it as most people are.”
A corollary to this is the way a meme can survive in a culture even in the absence of its referent, and that meme/word can then be applied to a new referent.
This is how “elk” came to be applied to Cervus Canadensis in the US, rather than to Alces alces, its original referent (as is obvious from the species name itself.). The word had persisted in English for centuries after the last speaker to see one had died, so it was available to apply to an animal that vaguely fit the semantic parameters. And then since “elk” was already in use when English-speakers came into contact with Alces alces again, they had to adopt the Narragansett or Munsee (?) word for the species. The Algonquian etymology of our word “moose” is “he strips off” after the way they eat.
LikeLike
tw violence
This is a bit removed from the central discussion of the post, but I’d like to comment on it anyway:
“[I]f you disagree that memetic fitness is correlated with truth, then you have no reason to believe any of the things that you believe and are forced to become a radical skeptic.”
I’m not sure this is right. Memetics predicts those ideas which survive and propagate will be those which confer advantages to their carriers. True beliefs usually have instrumental value—it pays to know which trees bear the most fruit, and which rocks make the best flints. But if a false belief offers some expediency that a true one does not, memetics predicts that the false one will win in a head-to-head competition.
We see this happening in religion. Many beliefs about ethereal theological topics—say, soteriology or transubstantiation—don’t affect our fitness one way or another, UNLESS we happen to live in a community with strong opinions on those topics. In such communities, there are environmental pressures to hold those opinions (e.g. heretics may be shunned, beaten up, imprisoned, or executed). When an epistemic agent recognizes those pressures, their motivated reasoning modules have strong incentives to tilt beliefs towards the received opinions on soteriology and transubstantiation, regardless of whether they convey reality.
Memetics predicts that our beliefs about medium-sized dry goods (e.g. “Trees exist,” “Peach trees produce the best fruit in such-and-such month”) will be widespread and mostly true. But it also predicts abstract and theoretical beliefs are subject to pressures besides than truth-preservation, especially coherence with working beliefs and the opinions of Right Thinking People. However, we can fine-tune our enterprises so that they do track truth, or something like it. The paradigm example is science, which since Bacon has maintained a memetic filter for empirical adequacy. Those ideas which survive and thrive in populations of scientists are those which pass repeated observational experiment. We see disagreement in scientific communities on those questions which can’t be resolved solely by empirical investigation (e.g. quantum mechanics, in which various “interpretations” all try to fit the same empirical data).
LikeLike
correlated does not mean perfect alignment. It just says that, everything else being equal (which is the crucial caveat) a true idea is more likely to survive than a false one.
LikeLike
IDK. I think that when a belief relies on community enforcement, it stops becoming memetic in its own right. The meme becomes “that other person must believe this” rather than “transubstantiation is real.”
LikeLike
An even better thing is to pay lip-service to the untrue but received wisdom and then so what you would have done anyway. It seems to me that most people who theoretically believe in transubstantuon (mostly Catholics, but some others) aren’t letting that affect their behavior all that much. They might even take offense if you asked them what it was like to be a cannibal, or what human flesh tasted like.
Yeah, at the margin their are true beleivers, but “cafeteria catholic” really is a thing.
LikeLike
Regarding topics that “everyone feels qualified to have an opinion on”, see also Parkinson’s law of triviality aka the bike-shed effect:
https://en.wikipedia.org/wiki/Parkinson's_law_of_triviality
“Bikeshedding” — endlessly debating the trivial while neglecting the significant, in order to feel like you’ve contributed — is often commented on as a problem in tech spaces, particularly open-source.
LikeLiked by 1 person
Memetic fitness is simply about the persistence of ideas. If an idea tends to spread, persist, and enhance the host’s fertility and robustness (on a long term basis), then it fulfills its evolutionary mandate. Analyzing the characteristics of a meme is an interesting evaluation exercise, but the only criteria that is universal is that it “works.”
LikeLiked by 1 person
Ideas don’t care about their hosts. An idea can reduce its host’s fertility and robustness (e.g. celibate priesthoods, suicide bombing) and still do quite well for itself.
LikeLiked by 4 people
“An idea can reduce its host’s fertility and robustness”
The host of an idea is the community it resides in, not just the individual members. An idea can very well result in the expenditure of individuals and increase the robustness of it community, and even fertility.
LikeLike
I think it is at least theoretically possible that there are virulent ideas that are bad for the community they’re in as well. (Racism? Communism?) While benefit for the community is correlated with memetic fitness, I think it’s a grave mistake to think of them as the same thing. (Compare: genes usually work for the fitness of the organism, but transposons exist.)
LikeLiked by 2 people
“I think it is at least theoretically possible that there are virulent ideas that are bad for the community they’re in as well. (Racism? Communism?)”
I hadn’t thought of that, but it’s absolutely true. I think racism operates between communities, not within them, but various kinds of collectivism and zealotry certainly do operate within communities.
” While benefit for the community is correlated with memetic fitness, I think it’s a grave mistake to think of them as the same thing. ”
Yes it is. So then the question is, why do groups perpetuate these memes? They must perceive some advantage. The only advantage I can see is that however counterproductive and maladaptive a meme may be, it feels good, and so perversely the worse the situation, the more attractive the toxic meme is. “Whiteness” and segregationism among poor Southern whites is an example of this, I think.
LikeLike
@Ginkgo, If the benefit provided by an idea is to signal group membership and clarify in-group/out-group distinctions, the idea needs to be costly (otherwise, the out-group will just copy/fake it, and it won’t work as a distinction). It seems plausible that at least some terrible memes persist because they perform such a signalling function.
LikeLike
On a evolutionary time scale, a meme will not persist if it kills off its hosts or fails to reproduce itself by perpetuating similar hosts (or their replacement offspring). You’re looking a too small a window of time in regard to memetics and may be confused with habituality, which is a short term phenomenon.
LikeLike
Protagoras,
” It seems plausible that at least some terrible memes persist because they perform such a signalling function.”
Exactly. This is a major impulse behind the generation of professional jargons and also of elite sociolects. The so-called “Received Pronunciation” or BBC variety of English arose in the early 19th century specifically as a form of class signaling.
LikeLike
He said terrible memes. What’s so terrible about specific varieties of English?
LikeLike
“I think it is at least theoretically possible that there are virulent ideas that are bad for the community they’re in as well.”
I can think of historical examples, but, like destructive plagues, they tend to flare up and then die out rather than persisting. (Basically, anything that gets you to start a war that you end up losing badly won’t do your community any favors – see Confederate States of America, the various Jewish revolts against the Roman Empire, etc.)
LikeLike
I’ve got my reservations about memetics (leans too heavily on genetics terminology to be truly accurate, and as such worries about evolutionary fitness rather than behavioural effectiveness too much. Both selected for, but different selection criteria) but ideas are certainly subject to selection pressures – utilility, reinforcement value, power. So broadly, the metaphor works.
On your thesis that divisive topics are memetically fit, I’d say that it’s not the divisiveness that’s reinforcing the reproduction of the argument, but rather the ingroup identification that comes with them – the reinforcement comes from the powerful, passionate agreement of Us as much as the passionate rejection from Them. Ingroup identification tends to increase in times of conflict, no reason that’s not true with ideas as well.
As a further example, take the fact that “the weather” (and other small talk) is an incredibly fit topic of conversation. Its simplicity and non-threateningness allows it to function as a social bonding thing.
LikeLike
I wonder what other types of memetic fitness there are. One that occurs to me: memetic fitness of frameworks or models of argument. For example, I suspect that people are much more likely to use, say, a reductio ad absurdum argument against some position if they’re previously exposed to that method and style of argument, even if they haven’t encountered the specific reductio argument that they’re making.
Some things seem to be on the border between memetic beliefs and memetic argumentation models. (For example: “correlation does not imply causation” can be a sophisticated argumentative counterpoint against a spurious causal argument, including more plausible causal explanations of the correlation (with making the CDNIC argument still being dependent on prior memetic exposure), or it can be a fairly mindlessly repeated memetic mantra, or it can be anywhere between the two extremes.)
LikeLiked by 1 person
An alternative or maybe complementary explanation about why people argue about stupid things is that they feel like the important things are difficult and realize they are not domain experts.
For example, in the mailing list of some developers working on inventing new programing languages, if someone asks if nominative or structural type systems are better, people may have interesting discussions as domain experts weigh on the topic and learn from one another. However if someone were to ask whether array indices should start with 0 or with 1, all hell would break loose as everyone would have a strong opinion.
Considering the understandable human desire to contribute and feel useful, it’s no wonder that people take opportunities to talk about easy stupid topics (“look at me, I’m contributing! I have things to say!”) and shut up for the important topics, when “adults are talking”.
LikeLiked by 1 person
Pingback: On purpose alone | Compass Rose