In my previous article, I said that we need to pay careful attention to the role of institutions in determining the course of Indian history. Apart from their intrinsic interest, institutions play an important role in mediating the tensions between social forces and in integrating conflicting modes of social analysis.

Suppose Ram takes up a position in an IT company. I could explain his behaviour in two ways, one focused on his individual choice and the other focusing on large-scale social changes. I might argue that Ram joined the IT industry because he was always good at mathematics and IT was a natural choice of a well-paying profession for a mathematically talented young man. An alternate explanation might involve the claim that structural changes in the Indian and global economy and government incentives are making it possible for IT industries to hire a vast number of bright engineering graduates.

The first explanation is a contingent one; it is geared towards Ram's peculiar talents and desires. The second explanation is deterministic; it seeks to uncover a statistical regularity in the behaviour of whole class of people. The success and failure of a particular institution, say, an IT company, is a combination of contingent factors such as the opportunities spotted by the founder of that company, and deterministic factors such as the general demand for outsourcing in the west.

Determinism and contingency are competing sources of explanation for political and social processes. Determinism suggests that social and political events and changes are inevitable consequences of the deep structures of Indian and world society. Consider the following question and two potential answers:

Q: Why did George W. Bush invade Iraq?
A1: Because the US wants to control the supply of oil.
A2: Because the Bush family had a personal vendetta against Saddam Hussein.

The first explanation is a deterministic explanation; it explains an individual event, i.e., the US invasion of Iraq in terms of the long term strategic interests of the United States. In the deterministic account, it scarcely matters who is the president of the United States since the underlying desire for control of energy resources will remain constant from administration to administration. The latter explanation is contingent, reducing a war between nations to the personal feud between two families. The truth of the matter in the Iraq war seems to be a combination of determinism and contingency: surely the US is interested in controlling Middle Eastern oil, but there also seems to have been a genuine personal enmity between the two ruling families.

Like most other absolute distinctions, both determinism and contingency are at play in complex events. However, we can discern a rule of thumb: the events caused by those distant from us or those who are immensely more powerful are likely to be seen as deterministic while events caused by people close to us and relatively equal to us are likely to be seen as driven by personal and contingent factors. The reporting of local politics is far more personality driven than the reporting of national and international politics.

There is, of course, a compelling rationale for the local to be personal. We know our friends, neighbours and colleagues intimately. When important events happen in the family or the workplace we can reliably attribute these events to the beliefs and desires of particular individuals:

Q: Why did X not marry Y?
A: Because her parents did not want her marrying a person of [ABC] caste.

Contrast that contingent explanation, based on the psychology of X's parents, with the following set of arguments:

Q: Why is it that sixty years after independence, most marriages in India are arranged?
A: Caste is still the most important social reality in India and jatis are endogamous.

Superpowers such as the United States, and long-standing social institutions such as caste are deemed to be so powerful that they can dictate affairs; determinism follows from the power of the institution. Nevertheless, as the above example shows, determinism and contingency do not contradict each other: X may not have been allowed to marry Y because of her parents' individual prejudices, but those prejudices may themselves be dictated by institutional factors. The personal is political, and the political is personal.

Events caused by those distant from us or those who are immensely more powerful are likely to be seen as deterministic while events caused by people close to us and relatively equal to us are likely to be seen as driven by personal and contingent factors.

 •  Nation and nation-state
 •  Privatisation of behaviour
 •  Democracy as vaccine

As every Bollywood director knows, when we depict individual people and families, contingency becomes more compelling but as we move to larger and larger groups of people, deterministic explanations have more value. Human society is both bottom-up, with individuals dictating institutional priorities and top-down, with institutions determining individual beliefs.

On the face of it, it appears that institutions are on the side of order and determinism. By reducing risk and imposing order, institutions such as the family, caste and nation increase the scope of deterministic explanations of human behaviour. I may die tomorrow or live to be a hundred years, but Indian men of a certain age and caste have a well defined mortality rate. As an institution, life insurance makes sense precisely because mortality rates are a predictable statistic. Similarly, we have courts because we want our disputes to be resolved uniformly and predictably, so that individual citizens do not impose their idea of justice on their enemies.

Nevertheless, contingent factors often influence the evolution of institutions. The Indian constitution would have been a rather different document if Dr. Ambedkar had not been the chairman of the constitution drafting committee. The linguistic division of states was precipitated by the death of Potti Sreeramulu. In these two events and on numerous other occasions, including violent events like the 1984 and post-Godhra riots, the response or lack thereof of the Indian state was mediated by a contingent factor mixed with underlying structural factors.

Any study of Indian institutions, whether of the largest institutions such as the nation or an intimate institution such as the family, has to pay careful attention to structural as well as contingent factors. In my opinion, a good Indian institution is one that operates according to a deterministic schedule but is also robust enough to accommodate various contingencies. A purely mechanical, deterministic institutional process will collapse instantaneously in the chaos of India.

In the west, institutions have the capacity to impose order on everyone. Disputes about the outcome of presidential elections are decided by the courts. In Europe and North America, the rules of the game, i.e., the structure of the courts, legislature and executive are well determined. What is contingent is the way people play the game and the outcome of that game, i.e., the winners and the losers. When the rules of the game are fixed for everyone, institutions can settle into an umpiring role rather than being active political agents. In India, we believe, often correctly, that the rules of the game aren't settled. In such a situation, every institutional actor is a political agent.

Imagine the following thought experiment: suppose at some future date, the Indian Supreme Court upholds the Cauvery Water Disputes Tribunal decision. If the verdicts does not boost Karnataka's waters, can we expect that the sitting chief minister of Karnataka will hold a press conference saying "oh well, we hoped for a better response but we are going to abide by the court's decision." Or, alternately, if the verdict reduces Tamilnadu's share of the waters, will the CM of that state say, "Oh, well. We had hoped to retain our rightful share, but the court has given its verdict, and we will now settle for a lower volume of water"?

Not a chance. Instead, no matter what the verdict, we are more likely to hear views about the regional bias of the justices and grumblings about how the views of this minister or that one, or this coalition partner or that one helped them put pressure on the judiciary. There will be self-immolations and riots by activists on both sides of the border (strangely, these are usually quite far from the border itself!). The central government will constitute yet another committee to 'resolve' the matter while everyone in power hopes that the crisis disappears for the rest of their reign.

In such situations, the lack of a deterministic, well defined, mutually acceptable and constitutionally sanctioned means of conflict resolution seems to hurt the country. However, the openness to contingency is as much a strength of our politico-social system as it is a weakness. The disputes in this country, whether over water or religion, are real. If interest groups feel that the rules of the game are unalterable, they may prefer to exit the system altogether and resort to violence from the outside. Kashmir is a prime example of a situation where the inflexibility of the rules of the game - in this case, the integrity of the Indian union - has lead to unbearable violence.

In contrast, reservations and the openness of the electoral system have convinced Dalits and others who have suffered discrimination that they can alter the rules of the game in their favour. Contingent events - such as the death of Potti Sriramulu - have often given the state and other Indian institutions the rationale for altering the rules of the game. We should profitably contrast the flexible chaos of our institutions with the iron order of the west. Whatever else we might say of our political parties, we do not label them the democratic and republican wings of the corporate party. The management of diversity requires a sensible mix of order and chaos.