There has been much talk in the run-up to the elections about holding national debates between parties. Such debates force candidates to reveal their platforms. Voters are then better informed about the choice they are about to make, and we can hope that such information-sharing will cause parties and candidates to agree on which issues are important.
At least, they should disagree less as a result. Indeed, economic theory argues that differences in beliefs are purely a function of mutual ignorance. As all necessary information is shared, disagreements, at least about what is important, should vanish.
But in fact public disagreements do persist, even as information is revealed. For example, while it may be best for us to move beyond caste- or religion-based politics, this is not necessarily a view shared by everybody. This is clear insofar as political parties still gain plenty of mileage through polarised ideologies.
Thus a mutual lack of information is not all. Sometimes inherent biases in views can cause disagreement to remain even when all necessary information is made available. Merely getting both sides to talk may not always be enough. A new research paper* builds a model to explain this, and shows that the underlying structure of society is crucial in reconciling or magnifying these initial biases.
Societies, the authors say, are of three types. The ideal society is fully integrated. In this event, views do in fact converge as information is revealed, and eventual disagreement is limited. The opposite case is that of complete fragmentation. Here, every man is an island, and nobody knows anyone else’s beliefs. Public disagreement is potentially the highest in this case.
These represent the best and worst, and neither of these scenarios is probable. The most realistic case is an intermediate one that the authors call segregation. Here, society is composed of groups. Beliefs are shared by people within a group, but not necessarily between individuals of different groups.
This is reflective of much of India’s caste-based politics, where leaders are able to unite a caste or sub-caste under one ideology, but in a manner that makes their collective views more opaque to everybody else. The Yadavs may unite under Lalu Prasad, but few who are not Yadavs may understand their motivations.
The fewer the groups there are, the more the polarisation there might be. This is also evident in, say, the Punjab-Sind regionalization in Pakistani politics, or in the ideological conflicts between insular Middle East regimes and the West. The same facts may be interpreted differently by both sides, the difference arising largely out of the respective biases.
The counterintuitive result in the paper is that when societies are highly segregated in this way, increased communication between groups may actually hurt. It can distort biases, causing initial ones to be amplified, or even create biases where none existed before. The distortion effect is greatest on minority groups. While opinions may converge within the group, there is still an inability to adequately process external information (e.g., the views of other groups). The more insular the group, the starker is the consequence. Collective beliefs may form within the group that are radical when viewed from the outside.
The authors don’t blame this on extremism or a failure of reasoning. Instead, this is the result of segregation. Put differently, the moderating influence of other people’s views is absent. Social integration (in the sense of a better understanding of others’ beliefs across the population) may result in lesser public disagreement, especially in large populations.
The fact that India remains a moderate democracy may in part be attributed to this. It is not that radical views do not exist. Instead it is perhaps that the process of interaction with so many other social groups via political compulsions ends up suppressing or quelling such views. Radicalism is then limited mainly to insular groups, and loses its sting when exposed to public scrutiny. A case in point is the BJP’s record in power versus its rhetoric while in opposition. This also suggests a host of policy prescriptions, whether in domestic politics or international affairs. Bring on the debate.
*Sethi, Rajiv and Yildiz, Muhamet (2009), “Public Disagreement”, MIT Working Paper 09-03. Available at http://ssrn.com/abstract=1345484
Madhav Raghavan is a doctoral student of economics at the Indian Statistical Institute, New Delhi. He can be found at http://www.isid.ac.in/~madhav8r.