The seven pillars of friendship

Our social relationships and friendships are the number one factor for happiness and mental well-being, far more so than money. Most of us are well aware of this. And yet we live in an age when we feel increasingly lonely and alienated and when social anxiety levels are higher than ever before in history. The importance of research in this area can therefore hardly be overstated.

One of my favourite evolutionary psychologists, Robin Dunbar, has got a brand new book published: Friends: Understanding the Power of our Most Important Relationships (2021). Dunbar himself made some great contributions to understanding friendships from an evolutionary point of view, most famously with his social brain hypothesis and Dunbar’s Number (the maximum number of meaning friendship is restricted to roughly 150 people).

Dunbar’s book contains a lot of findings that are intuitive, like extroverts tend to have more friendships, whereas introverts tend to have fewer, but more intimate ones. George Washington’s famous quote comes to mind: ‘Be courteous to all, but intimate with few, and let those few be well tried before you give them your confidence. Some of the most exciting findings of the book can be found in chapter 10: Homophily and the seven pillars of friendship.

Homophily may at first glance seem quite trivial: birds of a feather flock together. Duh. But this goes much deeper than skin-deep. Deep into our genes:

Compared to strangers, the people the subjects chose to be friends with had significantly more in common genetically. They shared about one percent of their genome — about as related as fourth cousins. Most often, friends shared genes related to sense of smell, the authors found .

Basically, this means, you are genetically more related to a random stranger you made friends with on a trip than a random person in your neighbourhood. So, how do we choose our friends? Dunbar found seven pillars of friendship:

The seven pillars are language (or, better still, dialect), place of origin, educational trajectory, hobbies and interests, worldview (religious/moral/political views), musical tastes and sense of humour. The more of these we share in common with someone, the stronger the relationship between us will be and the more altruistic we will be to each other.

The first three aren’t very surprising. Worldview is what has started to polarize us in the past decade or so and has received a lot of attention. Musical tastes and sense of humour were the two most surprising findings. And yet music and laughter are probably the oldest means for bonding in humans. What is really surprising about it it, that we have different tastes that seem to make us like fewer people who have different tastes.

What’s missing in Dunbar’s book are overarching principles that connect the single pillars. An obvious one is culture, which connects place, language, and interests. Another one is personality which links interest, worldview, musical taste and humour. If I may make a guess you will find few conservatives with a liking for David Bowie or Nirvana and Monty Python or Stephen Colbert when it comes to music and humour, respectively.

When I do the Myers-Briggs personality test with my students I usually get a picture like this:

Typically students with not much more than one or two different traits (like introversion) will sit next to each other. Empty seats are usually found between two very different types and loners are most typically introverted intuitive types. I have argued before that these types are evolutionarily derived from our ancestral modes of subsistence: foraging (N), farming (SJ) and herding (SP). For simplicity’s sake, I’ll discuss only the first two groups. How would Dunbar’s seven pillars differ in both groups? Based on well-researched traits such as tightness (farmers) and looseness (foragers) (cf. Michele Gelfand) as well as my distinction of more in-group social (nationalism/farmers) vs outgroup social (universalism/foragers) we can infer the following differences:

These differences explain some interesting phenomena, e.g. why people in rural areas (dominated by conservatives) generally have a stronger dialect or accent than people in urban areas (dominated by liberals). In fact, it looks like dialectization in rural America is increasing while at the same time decreasing in urban America, according to the eminent linguist William Labov .

For more information on evolutionary type theory check out my book Foragers, Farmers and Pastoralists : How three tribes have been shaping civilization since the Neolithic

Originally published at on May 22, 2021.