Social Evolution Forum
FIND sef:
Not a Feather Shy of Magnificent

I recently was sent a link to this really wonderful video about birds of paradise. Watch the video, you won’t regret it. The birds of paradise are out of this world:


(from Australian Geographic image source)


(image source)

But beautiful coloration and weird mating rituals are found, although in not quite as extreme form, in many other birds, as well as other groups of animals. Almost invariably it is the male that wears the gaudy suit:


(image source)

The bright red one is the male, the subdued grey one is the female. She has a nice and functional tail, while he has to negotiate the brambles with a ridiculously long and easily damaged tail feathers. Why is it always the male that goes to such extremes in order to obtain a mate? A pretty good answer is offered by the standard evolutionary theory, sexual selection, which goes back to Charles Darwin. Basically, being brilliantly colored, or having a very long tail (like peacocks), and being able to dance for hours is a very accurate signal of male quality. You may lie to the girl which you met in the bar about your bank account, but if you have crooked or rotting teeth in your mouth, she will know you can’t afford a good dentist. And bird females know this, too. So they choose brightly colored males, and then the evolution pushes this to ridiculous extremes.

In humans, by the way, males and females are relatively similar (in the biological jargon, the degree of sexual dimorphism is quite low). In the evolutionary precursors of the genus Homo the males were much larger than females, like in chimps and, especially, gorillas. The most likely explanation is that in humans between-male competition was greatly reduced because of pair-bonding and the need for males to provide food for their mate while she nurses the baby.

But this lack of significant sexual dimorphism refers just to biology, and humans have culture. A human male doesn’t need to grow bright feathers or a long tail, he can put them on. And we sure do.


(image from badgerinvaders)


Detail from Tavolette di San Bernardino, 1473. Gallerie Nazionale dell’Umbria (photograph by the author)

But unlike in birds, it is not always the male who is the most gaudily dressed.In fact, it is rare for the male costume to be more colorful than the female’s. I looked through a whole lot of plates in The History of Costume(Braun & Schneider – c.1861-1880) and here’s one that seems to fit the avian model best:

PLATE122BXChinese in Malaysia, late 19th century

Most of the time, if men wear colorful and elaborate dress, so do women:

PLATE70BXGentleman and lady of the court of Louis XIV

And when men tone down the colors, so usually do women. So a century after the ‘decadent’ fashions of Ancien Regime, we have this:

PLATE89BXUpper-class dress (1815-1820)

The cultural variability, and how fast it can change with time, is extraordinary. So what kind of theory do we have that can explain such variation and its dynamics? Is there anything in cultural evolution approaching the rigor and empirical support that the theory of sexual selection has in biology?


Join the discussion


  1. I am sure that the solution should be found in the ritual theory (the tradition of Durkheim-Goffman-Collins). Parade costumes are always significant part of self presentation in definite rituals (including opera perfomance, salons, home parties, pickniks, etc). Each ritual includes also ‘sacred’ symbols (including principles of great variety, say, ‘to be honest and modest’). Following these symbols is the direct way to confirm or develop own membership position. Symbols and rituals change because of multiple macro-events and processes (including revolutions, wars, economic crises, changes in class and political structures). It is usual also for new generation to establish new style of self-presentation for only reason to DIFFER from rejected previous generation (in youth subcultures ‘generations’ change each 3-5 years!). Elements of each new style appear just while be taken from given ‘menus’ and that’s why they are substatially accidental. I don’t think that the theory of evolution can explain great variety and fast changes in fashion on global and even on national scale. Maybe only on micro-level there is some real competition and smth like evolutionary laws operates.

    • Peter Turchin says:

      I don’t think it’s as fast as 3-5 years, but there is definitely a ‘generational’ effect – the next generation or cohort has a tendency to reject the values of the previous one, which sets up a back and forth cycle (short skirts, long skirts, then back to short skirts). There must be studies that measured this -fashion is a multi-billion industry! But there is also a longer cycle. So in my example of going from very cloroful Baroque costumes to subdued Neoclassical ones (note the parallel with architecture styles), this is not just a singel trend, but actually a cycle, which tends to be synchronized with the secular cycle, by the way. So we have elegant Gothic – Flamboyant Gothic – Renaissance – Baroque – Neoclassical – etc.

      • Peter, I meant definitely slyle changes of teens. They are really brief: new cogorts of 14-16 need to prevail domimance of ‘old hats’ of previous ‘generation’ of 20-25. That’s why ANY new style gets mass support. Your idea of long cycles fits classical series of history of arts but I cannot see deep laws and processes that play role of substantional determinants for these longue dureee shifts. Do you see them? Please, share with us.

        • Peter Turchin says:

          Sure I do. During the disintegrative phases of the secular cycle competitive spirit prevails, so people compete in the lavishness of their clothes (and many other things). Conspicuous consumption reigns the ball. During the integrative phases the emphasis is on cooperation, so the prevailing norms it to dress the same and in a modest manner. There you go.

          • Russia in terrorist period 1870-80, USSR in period of GULAG 1930-50, China in period of cultural revolution, Cambodia in times of Pol Pot – all they are charaterized by rather modest, uniform clothes. Do you really thing there were periods of social integration?

        • O.Voron says:

          Nikolai, ‘Russia in terrorist period 1870-80, USSR in period of GULAG 1930-50, ….– all they are charaterized by rather modest, uniform clothes.

          Russia in terrorist period 1870-80 – If you call the Parisian chic of the time modest and uniform, then yes :)

          USSR in period of GULAG 1930-50 – Soviet elites always tried to follow the current fashion. As for the rest, there was not much available, but they did their best. Not much quality and style, but certainly no uniformity.

  2. stephenduplantier says:

    Yesterday was Mardi Gras in New Orleans, Rio, Venice, etc. The masking and ostentatious display is maybe first of all, an enjoyable performance for the performer and the observers. Mardi Gras masking has a variety of motivations but the dominant one is best understood through Bakhtin’s idea of the carnivalesque–the subversion of authority through chaotic, rule-breaking and reversals of the day-to-day order of things. Mardi Gras is also very sexy: New Orleans and especially the Rio carnaval prove that. I like the idea that politics can be subverted and sexuality can rule the day all in one very enjoyable event. Birds of Paradise, for all their exotic beauty, do not subvert the dominant paradigm of the New Guinea rain forests.

    • Peter Turchin says:

      There is more to the theory of carnival. One suggestion I saw was that by inversing social relations it allowed the lower classes to blow off steam, which brought the society in better balance. This hypothesis suggests that there was a functional reason at the level of the whole society, not just personal gratification.

      • O.Voron says:

        There is something in it. In Republican Rome during Saturnalia, if I remember right, there was one day when slaves and their masters traded places. Just for one day.

  3. Martin Hewson says:

    “So what kind of theory do we have that can explain such variation and its dynamics? Is there anything in cultural evolution approaching the rigor and empirical support that the theory of sexual selection has in biology?”

    I cannot resist such a challenge! Here’s my attempt:

    Human dress and adornment is partly signalling to potential mates (as with birds). But it is also partly signalling group membership and status (a kind of sub-group membership). So male-female similarities of dress reflect their common membership of groups, while fast changes reflect the arms-race of status signalling.

    Or, something like that!

    • Peter Turchin says:

      Makes sense to me. Basically, as is clear from the variety of comments above, there are multiple mechanisms that may sometimes work at cross-purposes, resulting in pretty complex dynamics. Short cycles nested within long cycles, divergent trends in male and female clothing, etc.

      • I think gets to the heart of it. A lot of traditional sociology, rhetoric rather than science or statistics, is not compatible with a rigorous Darwinian analysis. This analysis will require the, in Enlightenment terms, atomisation of the part of the continuum of the material universe which is culture; making manipulable what it is fallacious to portray as discrete. Only when we begin this division of culture into bits which have some sort of stability and integrity, so that they are robust enough to survive our study of their behaviour, will we start to make real progress.

  4. Tim Tyler says:

    Culture is part of biology. Sexual selection is largely a form of deliberative selection. Humans apply deliberative selection to mate choice – and to cultural variation. The topic is an old one. It is sometimes called “artificial selection”.

  5. Ross David H says:

    Here’s my stab at it. Fancy dress, especially but not only on the part of the male, suggests both that the person is well off, and that they haven’t needed to face violent (or even particularly physical) situations lately. This, in turn, suggests that they may not be able to (both violence and hard physical labor are things you’re unlikely to do well if you’re new to them). So, what does this suggest to a female?

    If the environment is non-violent, then it suggests you have money, and who really cares if you’re not able to handle physical violence. In fact, in very peaceful environments, the biggest risk (of violence) to the woman is her mate, so an incapacity for physical violence is a good thing.

    On the other hand, in times of anxiety, people (including women) want someone who can handle themselves in case of trouble. Moreover, extravagant dress is more likely to invite trouble. So, this would explain the tendency for extravagant dress to be more common among the upper class, even in modern western societies where those of relative lower income can still afford lots of bright colors and lace. Black leather jackets or camo are more indicative of an ability to handle yourself in a fight; eye liner, glitter, and lace indicate the opposite. The lower income/class is more likely to find themselves in a situation of high anxiety (economic recession, high crime, etc.), so they are less likely to favor fashions that suggest you have not had any practice at handling physical/violent situations.

    In times of great self-confidence, with little fear of social unrest, crime, or other violence, then the fashions would tend to get more extravagant.

    I have no idea if the theory I’ve just outlined matches with the empirical evidence. Perhaps someone else here with a knowledge of fashion history can chime in with that?

    • Peter Turchin says:

      What is interesting is that while your observations make a lot of sense for the way things are today, several of these relationships were inversed in the past. For example, it was violent men who wore gaudy clothes – military men wore much more colorful uniforms in those days before khaki and camo became necessary for survival on the battlefield. While lawyers, merchants, and bankers wore sober, dark clothing. Additionally, until very recently (in historical times) it was the upper class who were much more prone to violence than lower classes (except for vagabonds, bandits, etc). How do you like the idea of a ‘criminal overclass’?

  6. Bruce Lepper says:

    Is there anything in cultural evolution approaching the rigor and empirical support that the theory of sexual selection has in biology?

    No, I’m sure there isn’t, especially if discussing something as fast-changing as fashion. However, your question brings to mind Herbert Spencer’s work. Take this quote from Dr Ronald Bolender’s site – link to

    “For Spencer, evolution is the master process of the universe, and it revolves around movement from sim­ple to complex forms of structure. As matter is aggregated—whether this mat­ter be cells of an organism, elements of a moral philosophy, or human beings—the force that brings this matter together is retained, causing the larger mass to differentiate into varying components, which then become integrated into a more complex whole.”

    So, according to Spencer there is always a certain “force” ( which is the sum of lots of forces, as Turchin notes above) which propels say, long skirts to short skirts, and as long as this force retains enough strength you get different versions of short skirts, people will experiment around the style, and you get a full-blown short skirt fashion, or a 4×4 automobile fashion (or a culture, if you like).

    The quote continues:
    “This complex whole must sustain itself in an envi­ronment, and as long as the forces that have aggregated, differentiated, and integrated the “matter” are sustained, the system remains coherent in the environment. Over time, however, these forces dissipate, with the result that the basis for integration is weakened, thereby making the system vulnerable to forces in the environment.”

    These could include, for example, the next generation rejecting the dominant skirt length simply because their parents wore them, or global environmental dangers convincing the adults to think twice before buying gas-guzzlers.

    The quote continues:
    ” At certain times, these environment forces can revi­talize a system, giving it new life to aggregate, differentiate and integrate, whereas at other times, these forces simply overwhelm the weakened basis of integration and destroy the system. Thus, evolution is a dual process of build­ing up more complex structures through integration and dissolution of these structures when the force driving them is weakened.”

    Spencer’s schema can be applied to fashion fads or empires. If I’m not mistaken, parts of his theory seem similar to conclusions reached in recent work on the rise and fall of civilizations.

  7. Hiroko Inoue says:

    Thank you for the beautiful pictures and video! They are amazingly beautiful.
    It is not an evolutionary theory, but Georg Simmel has a theory of fashion and points out its cycle over time. In his theory, the dynamic is vertical and happens across different social classes with some time lag among them. According to him, only the upper class is affected by the latest fashion. The inferior class imitates their immediate above, and the fashion of the elite class filter down the class pyramid. Eventually, the imitation forms the market with supply of low-cost copies. When the lower class adopts the elite’s fashion and everybody is wearing affordable copies, the fashion is no longer a marker of upper class. Then the upper class people start to wear something new to distinguish themselves. And, this forms cycle, and it repeats.
    This classic cycle of fashion as class differentiation became out of date with the advent of mass, but it looks like the similar cycle is happening at individual/group level. Herbert Blumer points out that fashion is a central mechanism in forming social order in a modern type of world, and the operation of mechanism will increase. It looks like it is now global. From evolutionary thinking, I am wondering what would be the selective advantages for the groups that have faster cycle of fashion trend.

  8. Peter Turchin says:

    Hiroko, I did not realize that Simmel had a theory about fashion. But the logic that you describe makes a lot of sense. So we have ‘biased transmission’ of cultural elements (fashion) from higher-status individuals, meanwhile those same higher-status individuals are trying to stay ahead of the masses by inventing new symbolic markers. But this mechanism doesn’t necessary lead to cycles. It is more likely to set up a runaway process in which the fashions become increasingly more expensive.

  9. Jurhum says:

    Not necessarily, the elite borrow and modify from the classes below them as well, see baggy clothes that rich rappers wear for e.g. or jeans for the modern day entrepreneurs.