Here is a fundamental fact about the human animal: newborns are everywhere the same but adults are hugely diverse. A human infant can develop into a culturally competent adult in any society, anywhere on earth, despite obvious and massive variation in the social and cultural environments into which they are born.
This is a highly distinctive feature of our species. Traditions exist in some other animal societies, but in no other is there such ready flow of beliefs, artefacts, knowledge, thought, opinion and judgement. Humans are so soaked in culture from the day they are born that by the time they are adults, they are all different. Anthropologists often make this point with words like ‘embedded’, ‘contingent’, ‘specific’ and ‘distinct’. Margaret Mead, one of the pioneering figures of early anthropology, expressed the point crisply: different humans, and different human groups, are each “experiments in what could be done with human nature”.1 Each life a data point about what it is to be human.
From the 17th century onwards, global exploration made salient to European and American scientists the extraordinary extent of human diversity. A new science was hence born, called anthropology, from anthropo- meaning ‘human’. Human variation was the central object of enquiry, and the ambition was to develop a natural science continuous with the other sciences.
What was learned is diversity. Take something quintessentially cultural and supposedly familiar: let’s say ‘marriage’. If the environment in which you live is anything like mine, then the meaning of ‘marriage’ you’re familiar with includes the idea of a commitment to sexual monogamy, between two unrelated people, that is socially enforced one way or another. A notion of ‘love’ might also play some important role. But if you look more broadly you will learn that other humans, from other times and other places, talk and think of ‘marriage’ slightly differently. For instance, it sometimes includes a degree of sexual polygamy on one side or the other, or both. Other sources of variation include whether marriage to cousins is taboo, encouraged or neither; whether remarriage is allowed or even possible; the degree to which sexual relations outside of marriage are tolerated or even supported; whether homosexual marriage is possible; whether there even is a word that could translate as ‘love’ in the local language; and whether payments to families on one side or the other form part of the marriage contract. All of this varies, and the different social arrangements typically translated into English as ‘marriage’ are in fact a family (excuse the pun) of similar but different types of social relation. The only truly universal thing we can say about ‘marriage’ is not much more than we would say about an ordinary business contract: it’s a socially recognised union, normatively endorsed, between two or more individuals.2 The lesson is that when you compare across people and see similarity on the surface, there is often still a great deal of diversity underneath.
And yet there is also unity. This might seem obvious and uncontroversial to you, a globalised citizen of the 21st century, but it was not so obvious to many citizens of 19th century colonial powers, for whom the most salient fact about the inhabitants of newly discovered lands was indeed how very different they seemed. Vice versa too, I’m sure. So while it is true that all human groups have beliefs and practices that outsiders find unusual or intriguing, sometimes even bizarre or absurd, there are also important things we all share.
Can we square the circle? Can we understand unity and diversity together, within a natural-scientific framework? Thinking about these issues forces us to confront the most basic question of any science: What even is this thing we are studying? How, then, should we conceive of culture, if we are committed to continuity with other sciences?
Many early anthropologists were influenced by evolutionary ideas stemming from Darwin (and by Whig history stemming from colonial power). They proposed that human societies each pass through the same stages, as if on a ladder. The western European societies of the time were always placed highest, and discoveries of other, different human groups were treated as discoveries of previous stages of human social evolution.
Subsequent generations repudiated this approach. In Franz Boas 1883 went to Baffin Island to study the different effects of light in the Arctic. While there he came to appreciate the environmental challenges of living in such a place, and the extent to which the everyday survival of Inuit depended on specific expertise and technology that he himself struggled with. Boas began to see how even something as supposedly objective as ‘expertise’ is inherently context-dependent. What even counts as ‘knowledge’ depends on where you are and what you find yourself doing. Pushing this insight as far as it would go, Boas began to argue that humans are so much individual creatures of their own time and place that to even make ‘comparisons’ between them is to miss the very thing that makes us human in the first place.3
To help learn more about humans and their diversity, anthropologists often do ‘participant observation’: they go and live in and among the community they study, in order to develop a close and intimate familiarity with them and their practices. After all, if your goal is to better understand how others see the world, then what better method could there be than to become one of them yourself? There are obvious difficulties and limitations with this approach but the basic idea has an appealing coherence. Anthropologists who, having done this, then reflect on what they have become and write down what they have learned are doing ethnography: ethno- meaning people and -graphy meaning writing. Ethnographers have, over the course of the past 150 or so years, compiled an enormous amount of valuable descriptive data about human culture and human diversity.4 This knowledge has repeatedly validated Boas’ key point: that human lives are, in a real and fundamental sense, incomparable with one another.
Recognising this truth has caused anthropology to gradually morph. Many contemporary anthropologists are simply indifferent or, worse, actively skeptical about the ambition of being a science. They aim to understand the ‘other’ (other people, other cultures), and others can only be properly understood on their own terms. Which in turn means that to compare, to contrast and to measure—to do all the things of ‘science’—is to exactly miss the point. Context and circumstance is how humans live, and that must, therefore, be how we understand them. This is not a resolution of unity and diversity, as it is a rejection of the challenge.
One of the most influential pieces of writing in this vein is ‘Thick description: Towards an interpretative theory of culture’, by American anthropologist Clifford Geertz.5 First published in 1973, it has been required reading for anthropology students ever since, leading to substantial influence over the discipline. Geertz’s main idea is that a natural science of culture is neither possible nor desirable. I do not agree with this, but Geertz is very insightful about the depth and seriousness of the problem. He appreciates—far more than most self-identifying scientists do!—the real and difficult challenges of approaching culture as a project in natural science.
It is good practice to engage with the best version of what you disagree with. Philosopher Daniel Dennett puts it well, “when you want to criticize a field, a genre, a discipline, an art form… don’t waste your time and ours hooting at the crap! Go after the good stuff or leave it alone”. I endorse this attitude completely, and Geertz’s writing is certainly the good stuff. It’s a pleasure to read and it makes transparent how serious and nuanced the problem really is. His conclusions are mistaken, but Geertz describes better than anyone else what we must face up to, if we are serious about the goal of understanding the human in a natural-scientific way.
Here is Geertz’s challenge in a nutshell: What is a wink?
You could say a wink is a rapid contraction of one eyelid. That’s true enough but it’s also nowhere near sufficient, because an involuntary twitch is also a rapid contraction of one eyelid, and the difference between a wink and an involuntary twitch is enormous. Mistake one for the other and social mishap will occur. So it’s not good enough to describe what is happening when someone winks just in terms of what is physically perceptible. A wink is more than a rapid contraction of one eyelid. A wink is a rapid contraction of one eyelid done voluntarily, for someone, with communicative intent but often covertly, and according to some socially-acquired norm.6

Or is it? Sometimes we perform overt, parody winks, as a way to jest or to mock. Or fake, not-really-covert winks, to knowingly mislead others into believing conspiracy is afoot. Perhaps some of these not-really-covert winks were practiced in advance, without an audience? Maybe you just tried some winks while reading this, just to feel how they differ? (Did you?!) There are also winks that don’t involve the rapid contraction of anyone’s eyelid, like this one ;-) It seems all the things that are supposed to make a wink a wink aren’t needed after all, not exactly.
A wink is a wink when people who know what a wink is say it’s a wink. The conspiratorial wink, the mock wink, the practice wink, the emoji wink and all the other winks out there in the world: what makes them winks is just the fact that they are understood as winks by wink-understanders: which is you, me and anyone else with the requisite social and cultural knowledge. Winks are like risotto. Risotto is risotto just when risotto-understanders say it is. (And how do we know who is a risotto-understander? Risotto-understanders are risotto-understanders just when people who know who risotto-understanders are say those people are risotto-understanders. Risotto-understander-understanders!)
Also, mutatis mutandis, marriage, science, sacrifice, grammar, coffee, football, Marxism, Bolivia, transubstantiation, goblins, books, computers, banks, lullabies, selfies, stone tools, designer chairs, War & Peace, Balinese cockfights, the Māori haka, collateralised debt obligations, the International Space Station—all of it! None of these things are what they are unless people who we think know what these things are say this is what they are.
The matter is especially complex for some of the most quintessential cases (marriage, transubstantiation, science, grammar), which involve many people, each of whom has beliefs about others’ beliefs, and whose actions are the consequence of their own interpretations of seeing others doing similar things in the past. Most people learn what marriage is, or science or grammar or transubstantiation, just by observing what others do and say. So, explains Geertz, “right down at the factual base, the hard rock, insofar as there is any, of the whole enterprise, we are already explicating: and worse, explicating explications. Winks upon winks upon winks”. The picture gets complex mighty fast. You cannot track all the explications and re-interpretations.
One of the most famous anthropologists of 20th century was Claude Lévi-Strauss. His most oft-quoted line is about how it’s not so much that “men think in myths…” but rather that “..myths operate in men’s minds without their being aware of the fact”. This line is from a book written originally in French, La Pensée Sauvage. The title is widely translated into English as The Savage Mind but this misses a double meaning that is present in French and evocative of what culture is really like: for ‘pensée sauvage’ means both ‘untamed thinking’ and ‘wild pansies’. Lévi-Strauss perceived a deep similarity between the development of thought in human societies, and the growth of wild flowers. I find this an evocative image: thought, culture and society as a slightly out-of-control garden, with some emergent order but also contingency and chaos, massive variety, and constant processes of growth, cross-polination, rebirth and death. Many people think Thought Untamed would have made a far better title for the English version.
The lesson is this: that if we are sincere in our pursuit of a scientific perspective on culture, then we must acknowledge that interpretation is what human worlds are made of! Interpretation is how we interact with one another, and how we construct the social and cultural worlds in which we all live. We all do it, all of us, all the time. You’re doing it right now as you read these words, trying to infer what it is I mean. Geertz put it this way: the human is an animal “suspended in webs of significance he himself has spun”.
And this is how uniform newborns become diverse adults. Humans decide in the moment and usually without conscious thought that this is a real wink and that is not. Also this is a goblin, that is the lindy hop and those people are getting married. We interpret and re-interpret countless times every day, every hour, even every minute, and these interpretations in turn inform our own future behaviour—which is itself interpreted by others.
This fact poses an important challenge for the scientific project. How are we to pursue a natural science of culture, if culture is all interpretation and webs of significance? This is a difficult question and its import is, frankly, not taken seriously enough by many of the psychologists and biologists studying culture today. In Part II of this essay I’ll summarise Geertz’s pessimistic answer—and then, in Part III, my more positive take on the prospects for a natural science of culture.
Mead, M. (1928/2001). Coming of Age in Samoa. Mariner Books Classics.
See e.g. Fortunato, L. (2015). Evolution of marriage systems. In: J. D. Wright (Ed.), International Encyclopedia of the Social & Behavioral Sciences, 2nd ed., Vol. 14 (pp. 611-619). Elsevier.
An excellent group biography of Boas and those he influenced (Margaret Mead, Ruth Benedict, Ella Deloria, Zora Neale Hurston) is King, C. (2020). Gods of the Upper Air: How a Circle of Renegade Anthropologists Reinvented Race, Sex, and Gender in the Twentieth Century. Doubleday.
For instance, the Ethnographic Human Relations Area Files (eHRAF) is an online database of ethnographic collections covering all aspects of cultural and social life. See also, from the pre-internet era, the Ethnographic Atlas.
Geertz, C. (1973). Thick description: Toward an interpretive theory of culture. In: The Interpretation of Cultures: Selected Essays (pp. 3-30). Basic Books.
By using the example of winking, Geertz builds on a thought experiment first developed by the philosopher Gilbert Ryle.