There are things of which we have such a clear picture that it takes a second glance to reveal their true intricacy. Social groups are such phenomena. Often, for example, we are quite off the mark in our estimation of what holds a certain community of human beings together. As a result, time and again, we presume to know why one group harmonises with another–or how and why crises arise between other groups.

In fact, our supposed knowledge is based for the main part on particular phenomena, such as the conflicts in Northern Ireland or in Israel/Palestine, or–to take a successful model of the coherence of a nation with multiple "nationalities"–Switzerland. In individual cases the reasons for the functioning or non-functioning might arise from historical circumstances, for example, or be structured along economic, cultural or religious lines. It would be erroneous, however, to infer patterns from observations of these cases.

Methodology: not universal enough

It might sound odd, but even for the social sciences, the patterns underlying social interactions are often the subject of controversial debate, and it is unclear precisely how disparate collectives correlate. To date, there has been no reliable scientific methodology with which to standardise the analyses of such interactions. Methods are needed that can track down the irritating grit in the mysterious motor or pinpoint the lubricants. As Lars-Erik Cedermann, ETH Professor of International Conflict Research, has argued. "...the method should be able to function without the stop that thinking within the borders of a country entails–because the importance of borders in our age of globally active networks has rapidly diminished."

A co-initiator of the transatlantic research initiative on complex networks: Lars-Erik Cederman, ETH Professor of International Conflict Research.

The long-term goal of TAICON (1), an initiative with new approaches to investigate and analyse complex organisations and networks, is to promote the development of such methods. The initiative was driven by Cederman and David Lazer, Professor of Political Science at Harvard (2). TAICON is to be a platform where three communities meet: complexity researchers and network specialists, social and natural scientists, and researchers from both sides of the Atlantic. "Nowadays, computer models enable us to simulate far more accurately highly complex systems, such as social clusters are. We have to take advantage of this opportunity," says Cederman, who has used computer simulations to research, amongst other things, the rise and fall of nations.

Frank Schweitzer, Professor of Systems Design in the new ETH Department Management, Technology, and Economics (3) since 2004, is also investigating complex collectives. The current research focus of his team is the application of the theory of complex systems on the dynamics of organisations. Apart from the development of formal concepts, Schweitzer also sets great store on quantitative modelling and computer simulations. He will be closely involved with TAICON.

Non-blinkered thinking

Lars-Erik Cederman's intention is to set the abstract, goal-limited macro variables, such as "nation" or "economic situation", in relation to one another without according too much weight to the perceptions or interests of individuals. Large amounts of detailed data would have to form the basis of this analysis. "Geographical information systems and spatial statistics, for example, could supply the building blocks needed to create a realistic picture of a particular field of interest," says Cederman. Among social scientists, however, physical space as a scientific category carries a heavy mortgage: In Nazi ideology the perceived lack of "Lebensraum" was turned into one of the driving forces behind the extermination campaigns in Eastern Europe. While bearing this in mind, Cederman nevertheless considers that it is time we overcame our inhibition towards what is an unquestionably important criterion in the analysis of the interaction between groups.

Conflicts, markets, transport

Space, time and (group) identity-these key concepts will mean a lot of work in the exploration of social networks. Cederman is convinced that the results of such research might not only serve to illuminate conflict mechanisms of entire regions-like the Caucasus, but also advance our understanding of the functioning of markets or transport systems. Moreover, the thrust forward is driven by an impulse inherent in research itself: "We want to help close the divide between classic social science approaches and methods established in the natural sciences," says Cederman. As such, is TAICON a self-test of the scientific community? "Yes," he answers, "and I am very curious to see how it pans out".