The punctuated equilibrium of culture

By Razib Khan | June 29, 2011 12:32 am

John Winthrop, ~1600. Mitt Romney, 2008 – image credit, Jessica Rinaldi

Recently Megan Mcardle had a post up where she expressed curiosity as to why “futurists” circa 1900 had a tendency not to imagine revolutions in clothing style which might have been anticipated to occur over the next few decades. You also see see this in Star Trek in the 1960s, where faux-future fashion was clearly based on the trends of the day, from the beehive hair to miniskirts. So I thought this comment was of interest:

I don’t know the answer, but I don’t know that they were wrong to do it. Keeping fashions exactly the same as the present generally winds up with more in common with the actual future than deliberate “future” fashions. A fair number of men still wear ties, and on rare occasions a few even wear tailcoats; rather fewer wear silver jumpsuits.

There have been a few counters to extreme fashions in media SF: “Blade Runner”‘s lead wore the same trenchcoat as his noir forebears; “Babylon 5” went for modified business suits and moderate variations on military uniforms; the “Battlestar Galactica” reimagining was pretty much straight conservative turn-of-the-millennium wear despite being in a far different time. How have those worn versus the approach taken by “Star Trek” or the 2015 segment of the “Back to the Future” movies?

I’m not sure that I accept this case as airtight, but this is certainly true in the specifics. Though I just saw some clips of Running Man for the first time on Youtube, I viewed Blade Runner a few years back for the second time and was struck by how undated it was in regards to fashion sense. At least in a very noticeable manner. It got me to thinking of the nature of cultural evolution even then.

If you read this weblog with any closeness you know that I have a pretty orthodox skepticism of supra-individual level natural selection biologically. By this, I mean that the unit of selection above the level of the individual is often difficult to maintain because of the technical obstacles. To be concrete, consider that the vast majority of genetic variation between two adjacent human demes is to be found within them, not across them. As the power of selection is proportional to the genetic variance which it has to work with that means that lower level selective forces are far more powerful all things equal. All things are not always equal, so that doesn’t render higher level selection impossible, but it does load our die in terms of expectations.

But I think it is somewhat different when it comes to culture. As I said above most of the genetic variation between two spatially adjacent demes is partitioned within them, not across them. So it might be reasonable for allele A to have frequency of 45% in deme 1 and 55% in deme 2. There’s a difference, but you would probably not want to make inferences of a person’s identity in a particular deme by their allelic state.

Now move the toy model to culture. One can easily imagine language A having frequency of 95% in deme 1 and 5% in deme 2. Even accepting that most populations are at a dialect continuum so that this doesn’t mean that much, there will be cases when populations are verging upon another group which speaks a totally unintelligible language. The dialect of Ionia was different from that of Aeolia, but there was a clear and distinct difference between these Greek dialects and the languages of the Thracians. The term “barbarian” derives from the confused incomprehensible babbling of foreigners to Hellenes.

A real life example of the salience of culture and the relative triviality of genetics in selection would be what occurred in Southeastern Europe in the 1990s. The Serbs, Croats, and Bosniaks (Bosnian Muslims) emerged out of the same milieu of South Slavic tribes which pushed into the Balkans in the wake of the collapse of the Byzantine frontier in the 6th century. Their languages are mutually intelligible. But, because of religious differences these three groups have radically different cultural orientations. The Croats are Roman Catholics who face West. The Serbs are Orthodox Christians who turn to Russia. And the Bonsiak identity is a product of the Ottoman period, when a substantial fraction of South Slavs and Albanians converted to Islam but did not assimilate to a Turkish identity.

Massacres all across the former Yugoslavia were perpetrated based on cultural cohesion predicated on these tribal markers and civilizational affinities. The biological difference between these groups was marginal at best. The stark substantive religious difference between Serbs and Croats is far more recent than the original conversion to Christianity by the Western and Eastern Roman churches, because the chasm between what became Roman Catholicism and Eastern Orthodoxy really escalated in the medieval period and later the Ottoman years. As for the Bosniaks, they seem to have emerged out of an environment of marginal South Slavic highlanders within the last five centuries.

The power of culture to be buffeted by natural selection is contingent upon the ability of memes to rapidly homogenize and differentiate through conformity toward alternative norms. Differences once non-existent can quickly harden and seem eternal and inflexible. Biological evolution constrained by the parameters of Mendelian genetic inheritance is by nature somewhat more gradual and less protean.

I’ve been focusing on space, but why restrict it to that? The chart to the left shows what I have in mind: a change in the parameter of space or time when comparing two populations is only relevant if you’re talking about inter-demic competition. In purely descriptive terms they’re probably somewhat interchangeable. Imagine two groups which are basically disjoint on a quantitative cultural trait. I mean that they don’t overlap at all. To give an example, how about the number of times you have to pray per day. Population 1 could hold to a religion which suggests that 10 short prayers has the most merit, while population 2 adheres to a religion which asserts that 3 short prayers has the most merit. If you assume that religion is enforced top-down so that everyone in the population conforms, not anthropologically unrealistic, then one can conceive of a scenario where there are a few deviants, but the distributions don’t really overlap. The two populations occupy different locations, so there’s perfect correlation between cultural form and position in space. But the same can occur with time, as cultural forms evolve.

This is where punctuated equilibrium comes to mind. I’ve gone on the record as thinking this model is overrated. It’s a fine enough descriptor of the fossil record and change on the scale of eons, but I don’t see much mechanistic interest because I believe evolution is mostly scale independent. And yet I wonder if it might be a much more useful description for what happens to cultural forms. There are long periods of stasis and change on the margins, followed by radical breaks from the past. I have read that the standard business attire of elite males in the modern world derives from the suits preferred by Calvinist Northern Europeans. For various reasons I think this is an over simplification, but it seems obvious that phenotypically the modern man of power resembles a Roundhead more than a Cavalier in public. In fact the modern man resembles the 5th century Frank in trousers far more than the 3rd century Gallo-Roman in toga. The substantive distance in a trait value can be far greater over small notionally time periods because of the wild swings in the rate of cultural change.

The observation that the fossil record manifests discontinuities and shifts due to the coarseness of its measure of evolutionary change is interesting, but ultimately I don’t know how actionable it is in follow up projects. It’s a scaffold, not a pathway. But I do think that a better quantitative measure of the nature of cultural change can be very useful, because it can spotlight the “hinges of history.” In the past three generations it seems that the years from 1965-1970 were particularly important in all manner of ways. But is that so? And if it is so, why?

  • Markk

    Culturally, say in your fashion example, there are technological changes that could very well cause discontinuities. In colder wetter climates when central steam heat came around I think you would see quite a large difference in clothing, and I think you do. In hot, humid climates certainly when air conditioning came around clothing styles changed to match, as well as population and population activities. Not too many siestas in the American Southwest, and evening meals are way earlier than they used to be. I think these kinds of cultural (technology changes are cultural changes right?) discontinuities from technology should somehow be reflected in any model.

  • Gil

    While it may not be accurate, given that this observation is made via the filter of the media, I’ve always thought it interesting that men worldwide seem to have more western dress, such as suits (particularly for men in power), and women worldwide seem to have more local dress.

    Similarly, I was surprised to see in footage of the ‘Arab spring’ in Egypt and Libya how the young men there were dressed: much like young men here in t shirts and jeans adorned with adidas logos.

  • Razib Khan

    While it may not be accurate, given that this observation is made via the filter of the media, I’ve always thought it interesting that men worldwide seem to have more western dress, such as suits (particularly for men in power), and women worldwide seem to have more local dress.

    this is true. i recently saw a muslim family enjoying the sun. all the males were in t-shirts and jeans or shorts. the women were covered up in a culturally expected manner.

  • Razib Khan

    #1, good point btw. thinking on it.

  • supersnail

    hey razib, could you point me to source on why evolution is scale independent?

  • Clark

    I think the gender differences of dress are particularly fascinating. Male fashion frankly is much more stable. And when there are changes it tends to be towards normalizing fashion. (i.e. bow ties even in tuxes falling out of favor for the already dominate long tie) What constitutes radical male fashion is a slight change in color or a slight raising or shortening of the ankle cuff of pants.

    Contrast this with female fashion where it changes much more rapidly with far more radical changes. When I watch movies about adults from say the 30’s through 60’s or 80’s men look noticeably different but not *that* noticeably different. Women on the other hand in film or TV look very dated. (The 70’s were a slight divergence due to the initial play of polyester leisure suits – but you could argue that was due to the combination of new technology and limits on using that technology)

    I think that what is going on is partially a reflection of the different sort of power relations. I’d expect that within realms where women’s power becomes more like men you’ll start to see a lot more conservative tendencies. (I think you already see that in business dress with the slight move in hair style) With regards to other cultures not already western I think you see men wanting to “protect” the women from the western culture but embracing it themselves. As women get more autotomy they’ll embrace it. (I think you see that among the different groups even in Iran)

    What I don’t know is whether that range of variability within women’s fashion applies within other cultures – just within the already established limits of that culture. I know I’ve seen that in some African cultures but I don’t know if it is a general trend. (i.e. you may have a social taboo perhaps primarily enforced by males to be non-western but a wide range of options they can work within outside of the western fashion) I just don’t know enough about Islam to know if or how that would work with the way females are covered. I’d expect (but may be completely wrong) that in private or with other women we’d still find fashion as a signal of place, power and so forth.

  • Clark

    BTW – I vaguely recall reading an article on the timing and social networks of women’s fashion comparing it to the flight characteristics of flocks of birds. i.e. a similar mathematics at work. This was obviously for western fashion. I tried to find it but can’t seem to locate it anywhere. (And I’ve no idea about the validity of the model)

  • Razib Khan
  • ohwilleke

    One way to think about culture is as a hierarchical structure. Some branches are at the very root of the tree from which the overall culture is constructed, some are mid-level, and some are at the very top.

    Top level branches in the tree (e.g. narrow neckties v. wide neckties) change frequently. Indeed, one of the important way that art historians date portraits is from the micro-fashion details in vogue at different points in the target time period.

    Mid-level branches in the tree (e.g. neckties v. no neckties for men), change less frequently, but upset not only the mid-level branch, but every subordinate top level branch detail that flows from them.

    Base branches in the tree (e.g. may women wear pants), change infrequently, but have profound implications for a large share of all of the branches in the tree.

    The 1965-1970 time period, probably most notable historically because it threw out the very basal cultural assumptions that discrimination on the basis of race and gender was appropriate. This upset myriad other cultural assumptions that were subordinate to these understandings.

    Fundamentals of selective pressures from a changing environment (e.g. air conditioning, clothing prices) would act on each cultural choice point until it flipped, and where those selective pressures act mostly on the base choices without much relevance to the subordinate branches of that choice (e.g. slavery v. not slavery without much importance to the details of slave codes), you would expect immense strain in the system followed by dramatic, revolutionary change (e.g. abolition in a Civil War).

    If I were to start out to make a mathematical model of this kind of situation and didn’t want to get too carried away in drawing cultural component hierarchy trees, I’d assign each cultural choice a number N that would reflect the extent that the choice was foundational or superficial, with N chosen such that N=log(frequency of change in that cultural component). Thus, I’d expect the same kind of stochastic pattern that you see in earthquake magnitude and flood severity.

    Also, like earthquakes, you would expect to routinely see rather severe aftershocks whose number, duration and severity is proportionate to the magnitude of the original peak shock.

  • Ian

    In hot, humid climates certainly when air conditioning came around clothing styles changed to match,

    Not if my father is correct. He grew up wearing worsted wool suits in the humid tropics (at least for work and parties) at a time when air conditioning was exceedingly rare. If you look at older pictures, people wore suits regularly until the 1970s, when lighter weight “shirtjack suits” became popular. By the late 1980s only businessmen wore suits any more. I would argue that, at least in Trinidad, the dominant signal was the 1960s (including the Black Power movement) and the decline of the suit (in Trinidad) probably owes more to Nehru and Nkrumah than air conditioning.

  • Ian

    Hmm…apparently “shirtjack” is West Indian usage. See here, if the Google Books link works for you.

  • ohwilleke

    “In hot, humid climates certainly when air conditioning came around clothing styles changed to match,”

    While the traditional Anglo-Italian suit of Europe and the Northeast is a dark died wool, the traditional suit of the American South and some Latin American jurisdictions to its South is a light gray or white lightweight linen (Dukes’ of Hazard fans can recall “Boss Hog”, Perry Mason also favored them).

    The British who were not bound by the mystique of their own fashion rules, as a former boss of mine famously liked to recall when justifying his own coiture, conquered the world in shorts. But, many colonial subjects, erring on the side of caution in emulating the British model wore suits.

    The American West, which was settled more by Yankees than Southerners, originally favored the Northeastern style three piece dark wool suit even as temperature rose in its arid summers in Southern latitudes. Conceptually, the dress shirt worn beneath the suit which not surprisingly got pretty nasty in the hot weather, was conceived of as a form of underclothing at that time.

  • Mercy

    @Clark, in addition to women’s business fashion being relatively conservative (but not nearly as conservative as men’s) men’s casual fashion is relatively experimental, about as much as women’s for the lower classes/teenagers- compare a mods to punks, shellsuits to tracksuit bottoms, the long hair of the nineties to the currently supreme high and tight. But from what I can tell upper class men have stuck to the same loafers and casual suit sort of things for a while.

    Which makes me think you’re prediction about women’s fashion becoming more conservative is right- probably the main factor is the knock on effect of having to wear formal suits to work, which limits what you can do with hair/body mods/etc – it’s easier to have a fairly conservative style so your wardrobes can overlap a little and you don’t need to learn two vocabularies. Upper class people go to a lot of events and places where they are expected to dress formally and that bleeds through into their casual wardrobe. Male students only dress formally to crash bullingdon club parties, so they are free to experiment as much as women.

  • Douglas Knight

    I may be missing the point, but it seems worth pointing out that the Roundhead picture in wikipedia is from the 19th century and thus, perhaps, not reliable. The Cavalier pictures are a mix of contemporary and 19C.

  • Clark

    Mercy, I think there is experimentation among even males in youths. But I think if you look at the typical American High School you’ll find that a surprisingly few number actually do this. The vast majority honestly don’t dress that different than they did in the 60’s. There will be stylistic variations primarily due to what stores actually sell. But if anything I think there’s probably much less innovation among the youth than in that odd period of 1964 – 75. I think that was just an exception due to a lot of innovation culturally. What’s so interesting is that rather than reaching a new equilibrium we tended to revert to what was going on prior to the 60’s. Only with fewer hats. (Much to the disappointment of bald men everywhere although after a couple of decades shaved heads became pretty mainstream so the loss wasn’t that big of a deal)

    This is a bit surprising when you look at the evolution of fashion in prior centuries. Typically each century was radically different from what went on before. Now even when the youth innovate typically it’s just for a period from 16 – 24. After which they adjust to societal norms. The norms vary regionally. For instance in San Francisco there’s a big influence of Indian dress for some reason. Go to Utah and you’ll find a stereotypical “suburban look.” You have an “inner city look” (where arguably there is a lot more innovation in male fashion although that’s tended to stabilize more the last 6 years or so)

    It’s kind of interesting honestly. Were I to make a prediction it would be that women’s fashions will stabilize more and we’ll have a small range of casual appearance and an even smaller range of professional appearance. Much as happened with men.

    The biggest innovations right now, as you mentioned, are tattoos and piercings. I’m not sure how those will end up. (Personally I find a lot of piercings pretty ugly – especially on the face – but obviously a lot of people disagree with me) My guess will be that it’ll become more conservative with time and that we’re in the experimentation phase since the early 90’s. I’d bet it’s already peaked and we’ll end up with more piercings as mainstream than in the 80’s but with far fewer than what you might see on the streets of San Francisco. We’ll see however.

  • Justin Giancola

    Clark, I think you are overlooking that most of the variation in clothing was with men until the 20th century. Before that women were fairly limited to varieties of dresses: different patterns, colors, material, but typically full length garb of some sort. Maybe this is outside of what you are focusing on though?

    I think what really killed it was men just got really fucking stiff. I think the two world wars and then the cold war- stiffer and stiffer. America vs. Communism- whole lotta stiffs – power moved away from Europe- less stiff fashion wise. There is this Bush song Everything Zen I love to quote in certain matters like this where the line goes “there’s no sex in your violence”. These forces seem to be seperating or compartmentalizing off idealogies to fragment peoples lives into profession life, private life, family life, sex life, public life, religious life, web life. A lot of these don’t overlap as much today as I think they might have once had. Web life is a new addition to hide/escape in and that starting to have more and more of an eye on you.
    This is a crude way of breaking down some weird progression of militarism that follows the Elightenment.

    And then a big battle is today in acknowledging homosexuality people are always eager to create clear lines of what is straight style / this is “questionably un-masculine”.

    I think the tatoo/piercing thing is a misguided/over-done aspect in a trend to loosen up. Without getting too into it I feel there’s a connection between militarism and tribalism, I think a lot of people aren’t as idividual as they think.

  • Justin Giancola

    Also mass production and disconnection with job people aren’t dressing specifically for their gig as much and would have to rely solely on their own creativity, and may not want to stand out.


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar