Why Cultural Evolution Is Real (And What It Is)

Cultural evolution represents an entire field of study. It has the potential, like biological evolution, to be a mechanism underlying and connecting many fields of study. This short introduction will pull together a few themes and compelling stories from this large field and present some of its concepts, mechanisms, and evidence—hopefully enough to increase the reader’s suspicion of the claim that cultural evolution is a “myth.”

John Gray, whose work I admire, unfortunately provides the perfect statement of social evolution denial in a recent essay:

Social evolution is just a modern myth. No scientific theory exists about how the process is supposed to work. There’s been much empty chatter about memes – units of information or meaning that supposedly compete with one another in society. But there’s no mechanism for the selection of human concepts similar to that which Darwin believed operated among species and which later scientists showed at work among genes. Bad ideas like racism seem to hang around forever, while the silly idea of social evolution has shown an awesome power to mutate and survive.

John Gray, Why capitalism hasn’t triumphed, November 8, 2014. Thanks to xenosystems for highlighting this quotation.

These claims are very similar to claims he made in a review of Matt Ridley’s book The Rational Optimist over four years ago:

Whatever political goals it is used to promote, the idea of cultural evolution is not much more than a misleading metaphor….Memes are just a pseudo-scientific way of talking about ideas, not actually existing physical entities. There is nothing in society that resembles the natural selection of random genetic mutations; even if such a mechanism existed, there is nothing to say its workings would be benign. Bad ideas do not evolve into better ones. They tend to recur, as racist memes are doing at present in parts of the world where economic dis­location is reviving hatred of minorities and immigrants. Knowledge advances, but in ethics and politics the same old rubbish keeps on piling up. The idea of social evolution is rubbish of this kind, a virulent meme that continues to reproduce and spread despite having been refuted time and time again.

John Gray, Review of The Rational Optimist, August 2, 2010.

While Gray is correct that cultural evolution does not have human well-being as its goal any more than biological evolution does, he is wrong to deny the existence of cultural evolution, particularly on the grounds that there is “no mechanism” by which cultural evolution can be demonstrated to occur. It seems that Gray did not, in the four years between the two essays, find time to learn more about cultural evolution, such as by reading the paper Matt Ridley linked in his reply to Gray’s review, Five misunderstandings about cultural evolution by Henrich, Boyd, and Richerson. This (excellent) paper offered by Ridley is a 41-page PDF, written in somewhat technical language; forty-one pages is perhaps asking a lot. But I find it useful to treat cultural evolution as part of my model of reality, and it is discouraging to see such useful concepts discarded without a fair hearing. (Steven Pinker’s take-down of group selection, The false allure of group selection, probably did not help matters, even though group selection and cultural evolution are not at all the same thing; see also Joseph Henrich’s response to Pinker.) My aim here is to provide an introduction to cultural evolution that is somewhat more respectful of my readers’ time than a 41-page PDF, but also to provide many seductive jumping-off points for further reading.

At the outset, here is an outline of what I mean by cultural evolution:

  • Human culture includes language, artifacts, music, stories, rituals, behaviors, and other information stored inside and outside of human brains.
  • Humans acquire culture through social learning. There are two major pathways of transmission: vertical, from parents to children only, or horizontal, between unrelated people. The mode of transmission is itself a major selective factor, shaping the content of culture.
  • Culture cannot survive and reproduce itself without humans; humans, after long dependence on culture, can no longer survive and reproduce themselves without it.
  • Cultures vary and change, and experience differential “reproductive” success: they spread or fail to spread, and their host populations grow or shrink.
  • Biological organisms only change by selection on random mutations. Culture, on the other hand, varies in part through intentional (non-random) human action. Despite this, it is rarely the case that humans fully understand all the functions of their culture, or predict how any changes they make will affect them and their descendants. (See, e.g., Alex Mesoudi, Foresight in cultural evolution.)
  • The relationship between humans and their culture is best modeled as a relationship between host and symbiote or between host and parasite, depending on the fitness cost extracted by culture, with humans as the host.

Humans Eat Culture, Not Food

Culture is not a luxury; it is life or death. Culture has likely been the most important part of the human selective environment for hundreds of thousands of years. Humans are not adapted to life without culture. Robert Boyd, Peter Richerson, and Joseph Henrich (also the authors of the aforementioned 41-page PDF) invite us to consider the “lost European explorer experiment” in order to understand the gravity of our dependence on culture for survival:

[The Lost European Explorer] experiment has been repeated many times when European explorers were stranded in an unfamiliar habitat. Despite desperate efforts and ample learning time, these hardy men and women suffered or died because they lacked crucial information about how to adapt to the habitat. The Franklin Expedition of 1846 illustrates this point. Sir John Franklin, a Fellow of the Royal Society and an experienced Arctic traveler, set out to find the Northwest Passage, and spent two ice-bound winters in the Arctic, the second on King William Island. Everyone eventually perished from starvation and scurvy. The Central Inuit have lived around King William Island for at least 700 years. This area is rich in animal resources. Nonetheless, the British explorers starved because they did not have the necessary local knowledge, and despite being endowed with the same cognitive abilities as the Inuit, and having two years to use these abilities, failed to learn the skills necessary to subsist in this habitat.

Boyd, Richerson, and Henrich, The cultural evolution of technology: Facts and theories. Citations omitted.
[Note: if you read only one paper about cultural evolution, this is a great choice – it’s fun and insightful, only 25 pages, and it has pictures.)

Humans rely on culture to survive not only in marginal environments, but even in environments rich with resources, such as the Australian outback. People rely on culture to obtain sources of food, and also to eat their food safely:

In 1860, two intrepid Victorian explorers named Robert Burke and William Wills set out on an expedition to cross the Australian continent from Melbourne in the south to the Gulf of Carpentaria in the north – a distance of 2,000 miles. They were successful in reaching the north coast, but on the return journey they both succumbed to starvation. Burke and Wills were educated modern men, but they did not know how to survive in the Outback. They were living on a plentiful supply of freshwater shellfish and a plant known as ‘nardoo’ that the local Aboriginals ate. However, both contain high levels of an enzyme that destroys vitamin B1, which is a vital amine (hence ‘vitamin’) essential for life. By ignoring the traditional Aboriginal method of roasting the shellfish and wet grinding and then baking the nardoo, which neutralizes the toxic enzyme, Burke and Wills had failed to capitalize on the ancient cultural Aboriginal knowledge. They did not die because of a lack of things to eat, but of beriberi malnutrition. Aborigines did not know about vitamin B1, beriberi or that intense heat destroys enzymes; they just learned from their parents the correct way to prepare these foods as children – no doubt knowledge that was acquired through the trial and error of deceased ancestors. Their cultural learning had provided them with critical knowledge that Burke and Wills lacked. As the two explorers’ fates show, our intelligence and capacity for survival depends on what we learn from others.

Bruce Hood, The Domesticated Brain. 2014.

Food processing is one of the easiest parts of transmissible, reproducing culture for us to observe. The nixtamalization process, an ancient American method of processing corn, increases the availability of another B vitamin (niacin). Like the shellfish and nardoo of the Outback, corn without its associated cultural processing cannot form the basis of a human diet; where corn traveled without nixtamalization, malnutrition followed.

As Hood notes, above, the people using these cultural food processing methods – and these methods, by the way, are memes – did not need to understand their utility in providing essential vitamins, or even need to understand what a vitamin is. Despite a complete lack of modern chemical knowledge, around the world in vastly different environments with different nutritional challenges, people found diets that supplied all the nutrients they needed in a safe way. And there were many casualties along the way. I wonder if KNM-ER 1808, a female Homo erectus from Kenya who lived 1.6 million years ago and may have died of an overdose of vitamin A, was one of these “deceased ancestors” whose “trial and error” eventually produced cultural solutions. Daphne Miller has posited (in The Jungle Effect) that our sudden switch en masse away from our carefully-evolved recent ancestral diets is behind the most common form of malnutrition in the world today, obesity.

Humans change their culture, but cultures also change humans on the genetic level; cultures themselves formed a major part of our selective environments. Lactase persistence, for instance, seems to have evolved only among dairying populations in Europe, Africa, and the Middle East. The purely cultural availability of milk changed the selective environment, favoring those with the genetic ability to take advantage of this novel source of food.

The Observable Evolution of Artifacts

It’s difficult to visualize the evolution of ephemeral cultural items such as behaviors, rituals, and stories. Focusing on durable artifacts, whose properties may be catalogued and compared, makes this process easer to grasp.

There is one extremely special kind of artifact whose unusual properties make it the perfect introductory example for illustrating the workings of cultural evolution: the chain letter. The chain letter (whether paper or electronic) consists of discrete units of information, almost like DNA, and causes humans to reproduce it. (Most artifacts, like arrowheads and carpets, do not share these properties.) The brilliant Daniel VanArsdale has assembled hundreds of paper chain letters, dating back to the 1880s; his Chain Letter Evolution (which, by the way, is worth spending many leisurely weekend hours immersed in) explains how these documents—literal self-replicators—evolved over the centuries, exploiting beneficial mutations that increased replication:

Apocryphal letters claiming divine origin circulated for centuries in Europe. After 1900, shorter more secular letters appeared that promised good luck if copies were distributed and bad luck if not. Billions of these “luck chain letters” circulated in the last century. As these letters were copied through the decades they accumulated changes from copying errors, offhand comments, and calculated innovations that helped them prevail in the competition with other chain letters. For example, complementary testimonials developed, one exploiting perceived good luck, another exploiting perceived bad luck. Using an archive of over 900 dated letters, predominant types are identified and analyzed for their replicative advantage. In 1935 the first money chain letter appeared, the infamous “Send-a-Dime,” which flooded the world within a few months. The motives and insight of its anonymous author are examined. A 1933 luck chain letter is shown to have provided a model for the Send-a-Dime letter, and this letter itself may have brought unexpected money in the mail to some senders in small towns. In the 1970’s a luck chain letter from Latin America that mentioned a lottery winner invaded the US and was combined on one page with a chain letter already circulating. This combination rapidly dominated circulation. In 1979 the postscript “It Works” was added to it and within a few years the progeny of this single letter had replaced all the millions of similar letters in circulation without this postscript. By examining hundreds of chain letters in the archive, evolutionary hypotheses are formulated that explain these and other events in chain letter history.

Dan VanArsdlale, Chain Letter Evolution

Within this small subset of artifacts that are made from information and that cause their own replication (given a population of literate hosts), the tracks of evolution are clearly visible. Most artifacts are not made of discrete units of information, and do not cause their own replication directly; rather, they are the result of intensive social learning of the behaviors required to replicate them. But ordinary artifacts too display the hallmarks of evolution.

In my outline, I mentioned that there are two (basic) pathways of transmission of culture. First, culture may be transmitted from parent to child only; this is vertical transmission (sometimes called phylogenesis). Second, culture may be transmitted between unrelated individuals; for our purposes (and speaking a bit loosely), this is horizontal transmission (sometimes called ethnogenesis). Exclusively vertical transmission results in a perfect, branching “tree-like” cladistic structure; horizontal transmission makes the phylogenetic tree look more like a tangled bush. Another important distinction is the mechanism of memetic transfer: transmission might either be particulate, “all-or-nothing discrete transmission of cultural traits,” or blending, “adopting the average value of a continuous trait from more than one model” (Alex Mesoudi, Cultural Evolution, 2011, p. 57).

The evolution of biological organisms is particulate, in that traits are passed by discrete units of DNA. Biological evolution also mostly relies on vertical transmission, with genes passed from parent to child. However, as Mesoudi notes in Cultural Evolution (p. 100), horizontal gene transfer is common among bacteria and plants, transmitted between unrelated individuals and even across species by viruses. Biological evolution appears to be more of a “tangled bush” than classical models would predict; cultural evolution, however, appears both more particulate and more tree-like than one might expect – that is, more like biological evolution.

In 2002, Tehrani and Collard’s study of Turkmen textiles from the eighteenth century to the present revealed a very tree-like pattern, indicating mostly vertical transmission (“phylogenesis”) and particulate, rather than blending, transmission.

The Turkmen textiles were produced in a context ideal for exclusively vertical transmission, especially prior to 1881: women learned weaving from their mothers, rarely traveled to other villages, and tended to marry within their tribes, reducing the chances for horizontal transmission or “ethnogenesis” to tangle up the phylogenetic tree. Is this result representative of other cultural data sets? Mesoudi, summarizing Branching versus blending in macroscale cultural evolution: A comparative study, by Collard, Shennan, and Tehrani (2006), says:

[I]f biological and cultural evolution really are fundamentally different processes, the former branching and the latter blending…then there should be a systematic difference in the tree-likeness of the biological and the cultural data sets. To test this prediction, Collard et al. collected twenty-one biological data sets that contained genetic, morphological, and behavioral data from a diverse range of taxa, including lizards, birds, hominids, bees, and primates. They also collected twenty-one cultural data sets, including Tehrani and Collard’s Turkmen rug patterns, O’Brien et al’s North American projectile points, other material artifacts such as Neolithic pottery decorations, plus nonmaterial data sets regarding food taboos, religious beliefs, and puberty rites. For each data set, Collard et al. calculated the number of homoplasies [changes due to independent invention or diffusion across lineages or groups], this time using the retention index (RI), which controls for differences in the number of species and characters and is particularly useful when comparing data sets, as was done here….[A]n RI of 1 indicates no homoplasies and a perfectly treelike evolutionary pattern, with lower RI values increasingly less treelike. The results showed that the biological data sets and the cultural data sets had remarkably similar average RIs of 0.61 and 0.59, respectively. Assuming that the biological data sets are primarily generated through branching speciation, this analysis suggests that cultural data sets, too, have been shaped by a similar branching process.

Alex Mesoudi, Cultural Evolution, 2011, p. 101. (Preface and first chapter available here. If you want to read just one whole BOOK about cultural evolution, Mesoudi is engaging and thorough, and only about 40% more technical than the present article.)

Cultural evolution, then, shares many traits with biological evolution. But, with the exception of the chain letters, what is the substance of memes? How can cultural evolution appear particulate and tree-like if there are no magical particles that code for each cultural trait? An intriguing solution is presented in that same 41-page PDF that no one is ever going to read: attractors. In this model, “it is the attractors that create quasi-discrete representations for selective forces to act on” in the absence of genuine discrete, DNA-like memestuff.

Attractors in Culture Space

Why do cultural data sets reflect evolution as “particulate” as biological evolution, when particulate mechanisms of transmission are extremely rare? Henrich, Boyd, and Richerson argue that the particulate nature is in the traits themselves, rather than the stuff of inheritance:

[I]nferential processes often systematically transform mental representations, so that unlike genetic transmission, cultural transmission is highly biased toward particular representations. Following Sperber (1996), we call the representations favored by processes of psychological inference (including storage and retrieval) “cognitive attractors.”

[C]ultural transmission does not involve the accurate replication of discrete, gene-like entities. Nonetheless, we also believe that models which assume discrete replicators that evolve under the influence of natural- selection-like forces can be useful. In fact, we think such models are useful because of the action of strong cognitive attractors during the social learning.

The reason is simple: cognitive attractors will rapidly concentrate the cultural variation in a population. Instead of a continuum of cultural variants, most people will hold a representation near an attractor. If there is only one attractor, it will dominate. However, if, as seems likely in most cases, attactors are many, other selective forces will then act to increase the frequency of people holding a representation near one attractor over others. Under such conditions, even weak selective forces (“weak” relative to the strength of the attractors) can determine the final distribution of representations in the population.

Henrich, Boyd, and Richerson, Five misunderstandings about cultural evolution, 2008 (a cleaner, shorter version of the paper, with in-line figures).

A major feature of cognitive attractors is that particulate cognitive information is less costly to hold and transmit than blended information – for example, it’s easier to model the moon as either purely a rock in space or purely a conscious entity than some combination of the two (Henrich & Boyd, On modeling cultural evolution: why replicators are not necessary for cultural evolution., 2002).

Attractors are forms that frequently appear to be the targets of convergence; they are indications of strong constraints in the spaces in which they appear. Donald E. Brown’s list of human universals (from his excellent 1991 book, Human Universals) provides a list of likely “attractors” in human space – products of cognitive, physical, and social constraints. When a feature is common to all human societies ever studied, such convergence is a strong indication of the presence of constraints. The human poetic line tends to be just around three seconds long – about the length of a human breath. The pentatonic scale is found all over the world and is encoded in 40,000-year-old flutes found in European caves made from the bones of vultures.

The ship’s rudder is another attractor; it has convergently evolved many times. As ships became larger and more complex, the rudder also changed; major advances in rudder technology were not able to be widely used, however, until all pieces of the puzzle were present (see, e.g., The development of the rudder, A.D. 100-1600: A technological tale, by Lawrence V. Mott, summarized in Boyd, Richerson, and Henrich, The cultural evolution of technology: Facts and theories). It is not difficult to see biological analogies to this process.

Human universals are far from a comprehensive list of attractors. Many stable attractors exist only at a very complex level of development that not all human groups have attained, for instance. But the particulate nature of the space of possible cultural solutions can explain the particulate nature of cultural evolution even in the absence of particulate “meme DNA.”

The Cultural Evolution of Massive Institutions

It is easy to see the processes of cultural evolution in the small scale, in projectile points, language, and carpets; but evolutionary processes are also visible in the macroscale. Peter Turchin has proposed that the unprecedented mega-empires beginning around 3000 B.C. were the products of cultural evolutionary constraints. A theory for formation of large empires presents a model in which large empires are constrained into existence by warfare and competition between nomadic herders and settled agrarian societies; as with agriculture crowding out hunting and gathering as a viable way of life, smaller-scale societies simply could not exist. It is not safety that created these giant engines of “cooperation,” but danger and necessity. He says:

Nomads are hard to tax, because they are skilled at fighting and can move themselves and their wealth much more easily than farmers can. Moreover, their chief product—livestock—cannot be stored easily, unlike the grain produced by agrarian economies. Thus, political organizations among nomads had to draw resources from the agrarian societies, by robbing the farmers, by extorting tribute from agrarian states, or by controlling trade routes….This argument suggests a reason why the sizes of agrarian states and nomadic confederations are correlated. As agrarian states in East Asia grew, nomads needed to cooperate on an increased scale to continue successful raiding, to present a credible threat to extort the tribute, or to impose favourable terms of trade. Additionally, larger and richer sedentary states possessed greater wealth that nomads could extract, thus enabling larger nomadic polities. Consequently, Barfield calls the nomadic confederations the ‘shadow empires’, their size mirroring that of agrarian states.

Thus, Barfield and Kradin suggest that the appearance of agrarian mega-empires explains the rise of nomadic imperial confederations. This is probably correct. However, if the presence of a large agrarian state produced larger nomadic confederations, should not the presence of a large nomadic confederation have similar effects on farmer societies?

Peter Turchin, A theory for formation of large empires, 2009. Citations omitted.

The religions of the Axial Age, on a scale much larger than the Dunbar-sized tribe, also reflect this process. Seth Abrutyn presents a cultural evolution theory of the evolution of large-scale religions, recognizing holiness and piety as a shared resource and “religious entrepreneurs” as memetic carriers, in one of my favorite papers, Religious autonomy and religious entrepreneurship: An evolutionary-instutitionalist’s take on the Axial Age.

The evolution of religion is one of the most interesting and ignored areas of cultural evolution. We tend to exaggerate the degree of rationality and conscious thought in human life, ignoring the enormous and influential contribution of magical thinking. Just as there are vital amines (vitamins) that human cultures must figure out how to supply us with, there are social and cognitive equivalents; modernity may do as poor a job supplying us with social belonging and ritual as it does supplying us with proper nutrition.

How Transmission Pathways Matter

In my outline, I mentioned that the transmission pathway – vertical or horizontal – matters a great deal for the content and friendliness of transmitted cultural items.

In biology, there is already support for this model. Parasitic entities like bacteria that are limited to vertical transmission – transmission from parent to child only – quickly evolve into benign symbiosis with the host, because their own fitness is dependent on the fitness of the host entity. But parasitic entities that may accomplish horizontal transmission are not so constrained, and may be much more virulent, extracting high fitness costs from the host. (See, e.g., An empirical study of the evolution of virulence under both horizontal and vertical transmission, by Stewart, Logsdon, and Kelley, 2005, for experimental evidence involving corn and a corn pathogen.)

As indicated in an earlier section, ancient cultural data is very tree-like, indicating that the role of horizontal transmission has been minimal. However, the memetic technologies of modernity – from book printing to the internet – increased the role of horizontal transmission. I have previously written that the modern limited fertility pattern was likely transmitted horizontally, through Western-style education and status competition by limiting fertility (in The history of fertility transitions and the new memeplex, Sarah Perry, 2014). The transmission of this new “memeplex” was only sustainable by horizontal transmission; while it increases the individual well-being of “infected carriers,” it certainly decreases their evolutionary fitness.

Parent-child transmission plays an increasingly limited role in cultural evolution. Horizontal transmission allows for the spread of cultural items that are very harmful to the fitness of host organisms, though they may (or may not) benefit the host organism in the hedonic sense. Indeed, the carefully evolved packages of culture transmitted for hundreds of thousands of years from parents to children are almost certainly too simple to solve the complex problems that moderns face.

Unselfconscious evolution is no longer up to the challenges of the fast changes of culture. As my hero Christopher Alexander says in Notes on the Synthesis of Form,

[T]he culture that once was slow-moving, and allowed ample time for adaptation, now changes so rapidly that adaptation cannot keep up with it. No sooner is adjustment of one kind begun than the culture takes a further turn and forces the adjustment in a new direction. No adjustment is ever finished. And the essential condition on the process—that it should in fact have time to reach its equilibrium—is violated.

Christopher Alexander, Notes on the Synthesis of Form, 1964, p. 56.

While unselfconscious evolution with limited foresight can no longer be relied upon to produce solutions of good fit, an understanding of evolutionary processes is essential to producing such fit. In particular, the constraints that have not changed for humans – nutritional, social, and cognitive – must be recognized, and old, hard-won solutions to these constraints must be seriously considered as essential building blocks of synthetic, selfconsciously-designed solutions to human flourishing.


7 thoughts on “Why Cultural Evolution Is Real (And What It Is)

  1. Goddamn it. I hate this post, because—besides being enviably brilliant—it gives me a few hundred pages of fascinating-sounding stuff I want to read. As if I didn’t have enough.

    To retaliate, I’ll mention a book that is somewhat relevant and that you might perhaps enjoy, written by Beth Preston, a friend and my major guide to the Western philosophical literature. It’s A Philosophy of Material Culture: Action, Function, and Mind. It’s not primarily about the evolution of artifacts, but that is one significant theme. She borrows from Ruth Millikan’s philosophical account of biological function, and applies it to human-created objects. Some of her examples are odd 19th century American kitchen implements, from her extensive personal collection thereof. Other themes are the importance of collaboration and improvisation in sensible activity, and their relationship with technologies. Here’s the abstract.

  2. Hmm, actually, reading her introduction (“Memoirs of a Material Girl”) in Amazon’s “Look Inside” gives a better sense than the tediously dry abstract provided by Routledge.

  3. Not to muddy the waters, but do you think the horizontal transmission of culture can make biological group selection more plausible under certain circumstances?

    • Interesting! I’m not sure I understand why horizontal transmission would change things – can you explain?

      The cultural layer, even under majority vertical transmission, ends up producing patterns that *look* like group selection, even if everyone is acting selfishly in their decision to adhere to group norms. The big “memetic selective sweeps” we’ve seen in the past few hundred years (by horizontal transmission) seem to me to look *less* group-selection-y, but I admit I don’t have a clear picture of the whole situation!

  4. The examples that come to mind are probably naive. But it seems that if a culturally facilitated belief, say a religious injunction concerning usury or polygamy, appears (or disappears) by dint of horizontal transmission, that selection for fitness enhancing traits bound to such cultural scaffolding could be informatively construed as group adaptive. As a result of a memetic sweep, magnet populations gain an adaptive advantage relative to other populations, even as defectors within the (arguably selected) populations ride free.

    I’ll be told that’s not genuine group selection and that’s probably true. But culture is a human bug; when it germinates beliefs that in turn shape behaviors that in turn impart evolutionary shifts; and when this is conspicuous at the level of populations, the emergent pattern looks so damn groupish that I wonder whether adherence to technically defensible individual adaptive models may be limiting our understanding of human evolution.

  5. Wow, these are so much more in depth than your twitter posts. Thanks for sharing it with me.

Comments are closed.