top of page

How humans create bonded communities: the challenge for humanism

By Robin Dunbar

Robin is Emeritus Professor of Evolutionary Psychology at the University of Oxford and a Humanists UK Distinguished Supporter of Humanism. He was a speaker at Humanists UK's Annual Convention in Liverpool. In this article he explains how social species like humans manage to live in large groups without resorting to homicide, and how religion is particularly good at creating bonded communities. The challenge for humanism lies in fostering similar bonds without involving religion.

The single biggest problem that we, and our monkey and ape cousins, have had to cope with during the course of our evolutionary history has been managing the stresses of living in very large social groups. These stresses are universal in that they afflict all mammals – and for all I know, birds as well – but my knowledge of the bird world is a great deal more limited. These stresses arise from what I sometimes call 'the London commuter problem'. There you are, crammed into your underground carriage twice a day, jostled and jabbed by all the other harassed and stressed folk struggling to get past you to find a seat or even just somewhere with a handhold where they can stand not too far from the exit they’ll need. There’s nothing malicious or intentional in all this, but having your toes innocently stepped on twice a day adds cumulatively to your stress levels.

Violent deaths increase as human group size increases

The bigger your group, the worse it gets. The consequences appear in two different places. One is that stress destabilises the reproductive endocrinology of female mammals, ultimately causing complete infertility. In mammals in general, it is not possible for more than about five females to live together in a group without all becoming infertile. The other is that psychological stress tends to spill over into physical violence, especially among males. Homicide rates in human hunter-gatherer societies increase steeply with living-group size – so much so that when group size reaches about ninety people, all deaths would be due to homicide and the group would rapidly become extinct. As a result, no hunter-gatherer tribe lives in groups with more than about fifty people. If a group gets bigger than that, it will soon split into two separate groups.

Of course, we and our primate cousins all live in larger groups than these particular limits. But that’s because we have found ways to defuse the stresses so as to allow more individuals to live together. In fact, monkey, ape, and human evolutionary history can be viewed as a long-running attempt to find ways to break through the glass ceiling that stress imposes on group size. Each solution defers the problem, but only for a bit – providing a breathing space that allows us to occupy a more stressful habitat without falling prey to the pressures. Then we hit the buffers again, and have to find another solution to allow us to increase group size still further when we need to.

Monkeys and apes are willing to devote large amounts time to social grooming.

Intensely-bonded alliances

The main solution that we and the primates have latched onto to solve this problem is the formation of intensely-bonded alliances based on friendships. These create the psychological environment that allows us to hold back on our selfish personal interests so as to create just enough space to allow the other members of our group to get their fair share of the metaphorical dinner table. Psychologically, this is no mean feat. It requires a neurologically very expensive capacity known as mentalising – the ability to understand others’ mindstates, to see the world from their point of view as well as our own. That is important because it allows us to decide whether someone trampling on our toes, or kicking sand in our eyes as they pass by, is doing it deliberately or by accident. And then we have to add enough thinking power to decide, in the light of that conclusion, how to respond: do you remonstrate with them (or, adopting a strategy of offence being the best form of defence, clock them one) or do you hold back and, as the song says, let it be?

The second cognitive skill that underpins this is what psychologists refer to as the inhibition of prepotent responses – the ability to back off from your own selfish demands so as to allow everyone else to have their fair share. If you can't do that, your ‘friends’ are not going to be willing to hang out with you. They’ll soon abandon you and find someone else who is more generous. This requires a lot of hard cognitive work – as we all well know when confronted with that last slice of delicious cake on the plate when everyone else happens to be out of the room. Self-control, or the ability to inhibit behaviour through sheer willpower, correlates with brain size in primates – and humans are better at it than other species.

The Dunbar Number Appears often in Human Affairs
Humans can maintain 150 close relationships

Skills of diplomacy

In short, it is all about the skills of diplomacy. Diplomacy – knowing when to stand your ground and when to smooth things over – is what makes it possible for us to live in large social groups. These skills are very sophisticated, are cognitively very demanding for the brain to process, and take a very long time to learn. The evidence from developmental psychology suggests that it takes the first twenty-five years of life to learn them effectively – or at least as effectively as any of us ever manage! Contrast that with the acquisition of language: by age five, a child has pretty much mastered everything you need to know to hold a conversation and process the complexities of language. Sure, they don’t have the vocabulary of an adult and they will still make the odd grammatical mistake, but they are pretty much fully-functional on the basics and could get by quite happily in life if they progressed no further. The human social world is the most complex thing in the universe, mainly because it is so unpredictable, and involves so many different individuals each with their own interests and objectives in life, that you cannot legislate for it with simple genetic rules of behaviour. It requires endless practice and in-your-face experience.

Social grooming triggers the endorphin system

So much for the framing of the problem. The issue that looms in the background here is the mechanisms that allow us to behave in ways that don’t cause social groups to break up and disband. The evolution of self-control is one component, as is mentalising, and both are handed to us by our species-specific genetic endowment. But there is another suite of behaviours that we use to create the sense of bondedness, trust and obligation that allows us to behave kindly to those we live with rather than robbing them of their hard-earned food or sleeping sites. Monkeys and apes create these bonds with each other through social grooming. Although a largely pointless activity (there are only so many ticks, lice and bits of vegetation that one can remove from someone else’s fur), monkeys and apes are willing to devote large amounts time to it. Some of the most intensely social species devote as much as twenty per cent of their entire waking day to grooming each other. And we still do a lot of it, though, having long since lost our fur, it now mostly takes the form of stroking, cuddling and hugging.

Social grooming of this kind triggers the endorphin system in the brain via a very specialised neural system whose receptors in the skin respond to one stimulus, and one stimulus only – light, slow stroking at exactly three centimetres a second (or about the speed of hand movements during grooming…. and stroking). Endorphins are part of the brain’s pain management system. Chemically they are very closely related to morphine and other opiates, and they create the same sense of relaxation, calmness and trust that induces a strong sense of bonding and obligation. As a result of this cheap chemical trick, we are willing to behave altruistically towards anyone with whom we engage in these activities. When asked for a favour, we do it at the drop of a hat without even thinking about repayment.

How to trigger the endorphin system virtually

The problem with grooming, however, is its intimacy and the fact that it is very time-consuming. That limits the number of people we can bond together. To allow us to live in larger groups, we have discovered a suite of behaviours that allow us to trigger the endorphin system in other people virtually – without needing to touch them. These are, in the order in which we probably acquired them, laughter, singing (without words), dancing, feasting, telling emotionally-charged stories and the rituals of religion. We have shown that all of these activate the endorphin system and, through that, create a strong sense of bonding to those with whom we do these activities. For better or for worse, religion is a particularly effective mechanism, capable of bonding very large numbers of strangers – providing they engage regularly in the rituals.

Religion ticks the boxes needed to create a bonded community

Remarkably – or, perhaps, more because it was designed to do this – religion ticks most of the boxes needed to create a bonded community, even of strangers. It offers a cultural story/myth of why we all belong and what our grand programme is. And it exploits most of the other buttons in this array of bonding behaviours – singing, dancing, feasting together, sad-and-triumphant storytelling, and even occasionally some laughter (as in the infamous ‘Toronto blessing’)… all ramped up with ecstatic experience – a mysterious, heady (literally), mystical sense of being in direct communion with the mind of god (which seems to happen when you get a very high dose of endorphins, or morphine for that matter).

The challenge for humanism

Which leaves me to end on a reflective note. In the light of this, the problem facing humanism is how to create that same sense of belonging and bonding that leads to the same sense of commitment and obligation without involving the religion bit. It’s not as easy as you might think.

Further reading

Human Evolution (2014) and How Religion Evolved (2022), both by Robin Dunbar.

103 views2 comments

Recent Posts

See All


Aug 07, 2023

Surely, in the case of being confronted with that last slice of delicious cake on the plate when everyone else happens to be out of the room, the best strategy would be to eat the last slice of delicious cake, as all those outside the room will not be there to see you eat the cake?

Only the penultimate person who left the room would know they had left the slice of cake on the plate when they left the room but for all the others, they would have no way of definitively knowing whether you or the last person to leave the room had consumed the slice of cake.

As far as competing with religion is concerned, I think…


David Good
David Good
Aug 03, 2023

Robin I found this an excellent article. I believe we have some events in modern society that have created bonded communities of strangers. I'm thinking of things like organised sport and music. Fans share a common bond from their enjoyment, even obsession, with their favourite football team or musical artist. They participate in mass rituals of both despair and enjoyment and develop a common ethos of what it means to be a fan. When a national team plays in a major sporting event the bonding spans large parts of the nation. There are many similar activities in modern life. Whilst these activities might not fully repalce the social organisation that religion once provided, they might point a way forward in replac…

bottom of page