Social network theory really emerged in 1965 with Stanley Milgram‘s experiment into how many hands a letter addressed to someone you don’t know would have to pass through to get to its destination. He reckoned it was six – through it does seem that he was fairly approximate with his methodology – and so was born the (urban myth?) that we are only six people away from everyone on the planet. When other social scientists began to look at society in terms of the webs of relationships that seemed to make it up there began a forking in the sociological literature. There were those who saw society as either as a bunch of atomised individuals or as a structure that glued people together. Famously, Anthony Giddens put them together in a fancy way which led to his theory of ‘structuration’. Social network theory though came at the problem of understanding from a different direction, influenced by communication and information theory, where the metaphor of a network came to be used to describe such structures as telephone systems which had nodes or points on them that communicated with other points. For social network theorists the metaphor was apt to describe humans in society – a bunch of nodes which tie together to form networks. Study the network and you study society.

One aspect in understanding networks is fundamental to the way that social network theory has begun to be applied to everything from obesity to book buying (see Christakis and Fowler’s Connected and Duncan Watts’ Six Degrees as examples). In a group of two there is one connection. In a group of three there are four connections. With four there are sixteen connections …. i.e. the number of connections increases much much quicker than the number of ‘nodes’. This is what is known as the network effect. It’s self-generating: add a node and the number of connections increases really quickly and at times it becomes an explosion.

Social network theory, then, started to look at society in terms of the connections between people rather than people themselves. That’s why so many social network ‘graphics’ have thousands of points connected up to each other. With the explosion of the web came an explosion in the application of social network theory. It’s not difficult to see why. The inernet is ‘the network of networks’ and with the World Wide Web a protocol to navigate those networks through hyperlinks, the social network theorists have more ‘data’ than they know what to do with. In the nineties the theory was taken on by internet business gurus who talked about viral marketing and  the tipping point – the moment when the network effect would kick-in and see an explosion of interest/revenues/fame etc. More lately it’s been taken on by the government with their idea that by nudging people towards life-style changes you can bring about enormous social change. Of course, you’ve got to nudge the right people – the right nodes on the network in order that the behavioural changes propagate.

Networks though have been at the heart of understanding of cultures for a very long time. After Clifford Geertz identified the work of understanding culture as understanding ‘webs of significance’, the patterns of belief, habit and behaviour that constitute culture, anthropologists have done some really significant work not only on exotic, sun-kissed islands, but also in grotty neighbourhoods amongst drug takers and criminals. In fact the word ‘network’ was, before the work of social network theorists, used to refer most often to those hidden cabals where a secret handshake identified you as part of the network. The old boy network may not have a secret handshake but it is recognised by those in it and it certainly predates the internet.

The problem of doing this kind of work now though is the amount of data that is available and the cultural encouragement to make connections – to draw the patterns that are in the data. I was drawn into it a little when I was trying to explore a group of people who had just begun to blog from Cuba. It was tempting to use various computer programmes to map out the ‘nodes on the network’ and see the connections between them. But ultimately I was more interested in the nodes and I became suspicious that the connections I was seeing were tempting me towards patterns that didn’t really exist. That’s the danger. And it’s a danger that emerges everytime there is a conspiracy theory that suggests connections where connections don’t exist. It’s great fun to plot your ‘friendship’ network on Facebook but dangerous to draw too many conclusions from it. That’s what happened when it was suggested in the media that the ‘epidemic’ of suicides in Bridgend, Wales was the result of a simmering network fed by Facebook. As it emerged in the investigation, it wasn’t, but the temptation to generate the theory from the data was great – and it was a cheap story for the  tabloids to pedal.

So, some conspiracy theories do feed off the amount of data now available to plot patterns from it where no patterns exist.