I wrote this elsewhere about a talk I went to in the Edinburgh International Science Festival in April 2008.
Of three lectures I went to at the Science Festival, it was the last which was most interesting: I actually got excited about it, perhaps because it was covering new(ish) stuff for me, things I didn’t know much about. I almost wished I had become an evolutionary-neurologist-psychologist (add your –ology to the list here).
The speaker was Robin Dunbar who is now apparently at Oxford. (I can’t help wondering why Liverpool University still hosts his website, but that must be a whole different discussion.) I hadn’t heard of Prof Dunbar before, but I am interested in human evolution and why we evolved as we did, and the lecture topic – What Makes Us Unique? – sounded my kind of bag. (By the way, that website describes Dunbar as an evolutionary anthropologist – an -ology I hadn’t included in my list…)
When I arrived, he was explaining how unique we weren’t – how, essentially we share 99% of our DNA with chimpanzees – or, according to New Scientist, 95%, or 96% or 98.77% – or whatever the figure really is – how, basically, we our very similar to our chimpanzee cousins, at least in terms of DNA.
Dunbar wanted to show how this was down to his “social brain” hypothesis: how, as a result of our complex social world, our brains had developed to manage and cope with this. What made us unique, he argued, was our ability to delve into the imagination, through our “theory of mind”, taking in religion, storytelling (one and the same, some might argue) and imaginative creation – music and dance. Dunbar himself told a wonderful story.
He started by looking at the brains of primates against other animals, and found that neocortex ratio – the ratio of the neocortex to the rest of the brain – was linked to the size of social groups. (You can look at Dunbar’s original 1993 paper here.) He was at pains to point out that this was not a quantitative relationship, but a qualitative one. Across a range of species, larger social groups were represented by a higher neocortex ratio – the more complex the social set up, the larger the neocortex.
In primates, that is. Because in other animals, there was no such relationship; and the social groups were different, too. The social structures in other animals were based around bond-pairs – those animals forming strong pairs bonds had big brains, and those forming life-long pairs had the biggest brains. Herd or flock size didn’t seem to come into it.
Dunbar had two hypotheses as to why brain size and complexity in most animals was related to the structure of pair bonds. Firstly, it could be due to the costs associated with getting it wrong: you need a big brain to sort out the wheat from the chaff and avoid costly catastrophic errors – either the risk of cuckoldry or the risk of partners (and offspring) having low reproductive success. The second was that maintaining pairs is expensive – the need to synchronise behaviour within the pair is difficult, and this requires big brains. He reckoned it was the latter: if it was simply to cost of getting the decision right, the size of the brain would stop being important once that decision had been made, so that there would be no need for a large brain, whilst maintaining a pair through behaviour would require thee brain to be used the whole time.
So why were primates different? He put this down to the need to maintain higher bonds within groups – not just pair bonds – which results in the variety of social styles Dunbar and others observed. He used different baboons as an example, some showing social grooming, some building coalitions within the troop and some using tactical deception; and all have relatively big brains.
He extrapolated from human brain size to propose that the size of human social groups would be about 150 (ok, it isn’t a quantitative relationship but …), and then he looked at ways to test this hypothesis.
Looking through the literature on human group size, he found lots of evidence. In different cultures, round about 150 seems to be about the optimum size for social groups. He gave a lot of examples: medieval villages, Amish communities, “clans” within anthropology (a concept which he said was poorly defined – it only existed in the minds of anthropologists – but wherever there was work into clan size, the size of the clan was about 150) and hunter-gatherer groups. Hutterite communities also fit the bill, and when asked, they say that above that number a village needs police – the social bonds break down and rules become necessary. He also used the more modern case of W.L. Gore – the makers of Goretex. When Bill Gore established the company, he decided that no location should have more than 150 people in it: he felt that this the maximum number of people that one could recognise, and he wanted to keep the hierarchy flat, communications fast and decision making communal: so he stuck at 150. (I saw a talk about the culture and structure at W.L. Gore a while ago – it certainly seems to work.)
All in all, Dunbar felt that 150 felt like the number of people in which one would feel like “a member of a real community”, and above this, the community would fragment in some ways. This was the boundary below which you would know the members of the group as people themselves, rather than simply unknown “others”.
He had done some research into this, working with modern (real-life) social networks. The number of close friends one has is (generally!) about five, then we form close relationships with another ten or so, are on good relations with a further thirty five, and know – really know, that is – another hundred or so. The relationship of 5:15:50:150 gives relationships a scaling ratio of about 3. Colleagues of his had also looked at these in fractal terms – there was a fractal peak at about 5.2 – and they got very excited. (But I can’t remember why…)
Within primate groups, the limits to group size seem to be down to the time spent grooming: grooming keeps the group together. Grooming release endorphins in the groomed’s brain, and this experience binds the group together. But if people spent the time grooming to sustain a group size of 150, more than 40% of our time would be spent in grooming activity (apparently, only 20% actually is).
Dunbar’s theory is that language changes the nature of grooming: it lets us communicate more widely. And some communication releases endorphins within our brains, too: laughter, music and dance, and religions and other rituals. Dunbar postulated that we evolved laughter first (indeed, some primates laugh), then singing and dancing – laughter being essentially musical (apparently…) and lastly rituals.
“Traditional” religions are highly bonded communities, where the ritual is rigid, repetitious and fixed. These rituals stimulate the release of endorphins in those taking part. (Interestingly, the Anglican tradition apparently release much less endorphin than other, more charismatic or ritualistic religions.)
Dunbar suggested that human storytelling is another form of ritual, which built into religion – which, fundamentally, is about telling stories focussing on beliefs. Stories bind communities, forming a common understanding: we communicate by telling stories, our arts are about telling stories (at least until we invented abstract art!); and storytelling can be repetitious, building into the ritual until the original story is forgotten.
Storytelling took us back to the “theory of mind”. Most people can think of other’s thoughts to five orders: “I think…” is the first order, “I think that you think …” is the second order; “I think that you think that he thinks…” is the third order. It is hard to conceptualise examples above this: Dunbar did it by examining the play Othello. Iago makes Othello think that Desdemona loves Cassio, who reciprocates. This goes to four orders (Iago – Othello – Desdemona – Cassio – Desdemona); Shakespeare’s genius is that he conceives this (fifth order) and the effect it will have on the audience (sixth order).
Dunbar also said that religions and rituals become binding at high orders of “theory of mind”: “I think that you think that we believe … in the relationship we have with god”. Through the relationship with god(s), the high orders of theory of mind are reached.
The size of the brain enabled the evolution of language, communication and the cognitive power to build these pictures of relationships. Other primates can be shown to use a theory of mind: monkeys at the first level, apes at the second; humans commonly use the fifth level. There is a correlation between primate brain size and the order of theory of mind these animals utilise.
Dunbar concluded that the cognitive constraint was based on brain size, and this had enabled humans to create the virtual world of our imagination – our theory of mind – and build our relationships: we can comfortably cope with 150 relationships. And that exploring these ideas from an evolutionary perspective enables us to ask some interesting questions.
I am not sure I fully understood it, but it was a fascinating lecture.