I attended a talk by a sociologist, Noah Friedkin (UCSB), discussing the structure of influence in networks and groups and his findings strike me as interesting for our discussion of multilateralism.

Friedkin has done experiments in which people are asked for some assessment and then get a chance to interact and then make a post-discussion assessment. A general finding in groups of 3 or more is that the second assessment almost invariably is within the bounds set by the initial assessment. Whether people modify their initial assessments or , whether they achieve a consensus or not, the set of second judgments falls within the outer bounds of peoples’ initial position.

This result does not hold in dyads. In the interaction between two people, a substantial portion of the time there are post-discussion outcomes of both consensus and disagreement that are outside the bounds of the initial judgments of the people. This is puzzling on two counts. First, why would two people take one position initially, and then have one or both of them take a subsequent position outside the range established by their initial positions. Second, why should this occur almost 20\% of the time in dyads when it is an incredibly rare event in triads and tetrads. Friedkin only offered conjectures, but he did say that the influence dynamic, the process by which two people try to reach a consensus, differs from that of groups of 3 or 4. One speculation is that the critical difference is that in trying to get agreement between two, any one person who tries to get convergence by asserting their dominance simply raises the hackles of the other person and the result can be post-discussion assessments that are outside the original bounds. In contrast, it is easier to find a focal point or it is easier to be influenced by a set of people than by just one.

One can draw implications from this result for our discussion of multilateralism and bilateralism. Multilateralism entails reaching a consensus in groups larger than 2. In contrast, what we sometimes mean by unilateralism is a state acting on its own to deal with an adversary. The US has, for example, insisted on multilateral talks with North Korea rather than accepting the North Korean demand for direct negotiations. The US has agreed to join the EU-3 in nuclear discussions with Iran. The implications of Friedkin’s work on social influence processes is that social influence is more readily achievable in larger-N settings (small groups of 3-4) than in bilateral ones.