He was in London to plug his book, Connected: The Surprising Power of Our Social Networks and How They Shape Our Lives, and he discussed how he’d arrived at his main idea – that many things in our lives are socially mediated, including depression, stress, obesity, emotions, and death – and what it might mean for society.
Christakis said first came up with the idea of investigating the effects our social networks when studying the “widow effect” – that partners in a relationship often die soon after each other. He found that those at some remove also suffer stress (he came across this when a friend of the partner of someone whose parent had a terminal disease said they were concerned: the ripple of stress spread out to “three degrees of separation”).
He started to investigate social networks, and found a lot of different phenomena were mediated through networks. Most famously, his analysis of obesity showed that having obese friends increases the likelihood of your being obese by up to 45%, and this too is effective at up to three degrees of separation – that is, if your friend’s friend’s friend is obese, you are more likely to be obese.
There might be lots of reasons for this – you might all drink in the pub together, or hang out in the same fast food joints, or live in the same gym-free neighbourhood, or …
We are social beings: we are influenced by our friends and their behaviour, and vice versa.
Christakis also demonstrated that emotions also spread contagiously, and postulated that there was an evolutionary benefit to this, going so far as suggesting that we are adapted to networks. He reckoned that 40% of our network structure was down to genetic influence – that is, the number of friends or contacts we have, and the closeness of those contacts, is partly under genetic control. (Spoiler: theories suggesting evolutionary benefits of social adaptations are notoriously tricky and more or less impossible to prove: they might be neat ideas, but they are not particularly scientific…)
He reckoned our social networks influence the following behaviours and emotions:
- drug use
That is quite a broad list, though it is possible to see the social side to most these points. Christakis described this as “a connected life”.
It was interesting, but it actually seemed a lot too neatly tied up. Life seems more rough and unregulated than Christakis described. Much of his work has been done in closed networks, if only to be able to map and measure all the connections. Educational institutions seemed to be common networks to work in. He showed network maps (like the one above – a map of my Facebook network, created by Touchgraph); in closed networks, the people at the periphery look like lonely souls, but of course no network is closed: those people have friends and connections outside, in other networks. These weren’t taken account of.
The book may answer some of my unease – I haven’t read it yet, so I can’t say. I think there are a lot of implications in how we manage communications and develop policy: in answer to questions, Christakis said it might be more effective to target policy on specific nodes – people – in a network in order to reduce, say, drug taking rather than everybody in the network. These poeple sound a lot like the mavens described by Malcolm Gladwell in The Tipping Point – those individuals who are key to spreading new ideas, just as there are carriers responsible for spreading infectious diseases.