Sen, ‘The Idea of Justice’, Chapter 16, ‘The Practice of Democracy’

This chapter covers three empirical issues relating to democracy: 1) the connection between democracy (or public reasoning—Sen seems to use these terms interchangeably) and famine, 2) the connection between democracy and economic development and 3) the promotion of tolerance toward minorities. In what follows, I will first restate Sen’s account of democracy (given in the previous chapter), as this is relevant to his interpretation of the data he provides in chapter 16. Second, I will outline his discussions of each of the three topics he takes up, and, third, I will raise a few questions about the causal connections he proposes in the discussion of famine.


Sen views democracy as not merely the presence of elections and ballots, but as “government by discussion,” which includes “political participation, dialogue and public interaction (326).” He believes that an unrestrained media is especially important to the functioning of democratic societies, for a number of reasons, one of which plays a central role in his discussion of famines: a free press, Sen tells us, contributes to human security by giving a voice to the vulnerable and disadvantaged and by subjecting the government to criticism. (More on this

Democracy and Famine

In 1982, in an article in The New York Review of Books, Sen made the observation that “no major famine has ever occurred in a functioning democracy with regular elections, opposition parties, basic freedom of speech and a relatively free media (even when the country is very poor and in a seriously adverse food situation) (342).” Further, while India was under autocratic British rule, famines were regular occurrences; once India achieved democratic self-rule famines ceased. (Apparently, Sen’s observation about democracy and absence of famine was initially met with a fair amount of skepticism. Now it is widely accepted.) Sen infers from the observed correlation that democracy prevents famine. He offers two reasons in support of this inference. First, democratic governments are accountable to their citizens and subject to uncensored criticism from the media. So, in order to maintain power, democratic governments have a strong incentive to eradicate famines. (Indeed Sen argues later in the chapter that the famine case is really an instance of a broader phenomenon whereby democracy advances human security by giving political incentive to rulers to respond to vulnerable citizens.) Second, because of the informational role of the free press, democratic governments are likely to know about the plight of citizens and therefore about the need for amelioration. By contrast, authoritarian regimes, which suppress public discussion, may be simply uninformed about the severity or extent of a famine and fail to provide assistance for that reason.

Democracy and Development

There is a received view, held by both defenders and critics of democracy, which says that democracy is inimical to economic development whereas authoritarianism is not. Sen rejects the received view, arguing that under the proper understanding of “democracy” and “development” there is little evidence to support it. First, Sen views political liberties and democratic rights as constitutive of development—development, in his view, should not be understood narrowly in terms of e.g., economic growth. In this case, of course, democracy guarantees development. Second, even if we emphasize the economic growth aspect of development, it turns out that the received view is based upon “selected cross-country comparisons,” specifically between East Asian countries on the one hand, and India on the other. More comprehensive cross-country comparisons, Sen says, do not support the received view. (We are not given any details, here, as to which additional countries should be included in the comparisons.)

Minority Rights and Toleration

Democracy understood as public reasoning, Sen argues, can accommodate the problem minority rights that arises under a regime of majority rule. He notes that the formation of tolerant values is essential to democracy so that those in the majority have a desire that their society guarantee minority rights. At the same time, however, the formation of tolerant values is partly achieved through democracy by means of open discussion of the various affiliations citizens have. That is, “democracy can…assist in bringing out a greater recognition of the plural
identities of human beings”—it can aid in citizens in seeing one another as something more than a member of the majority or minority.

Democracy and Famine Prevention

In his discussion of the Bengal famine of 1943, Sen notes that an official report of the British colonial government to the British parliament claimed the number of deaths per week to be about 1000, when in fact it was approximately (a staggering) 26,000 per week. According to Sen, this dissembling and/or misperception (which was accompanied by the gross failure of rulers to respond to the crisis) was enabled largely by severe restrictions placed by the British on Indian media and by the British-owned media’s self-imposed silence on the famine. This restriction of the press, Sen claims, greatly hindered public discussion about the famine in Britain and India. Therefore rulers were not subject to public pressure and so had no incentive to provide relief. (“The rulers,” Sen says, “never starve.”)

Sen’s account suggests that only if rulers either suffer themselves or experience political pressure, are they moved to respond. And, since they never suffer, they are moved to respond to famines only if they experience political pressure. This strikes me as over-simplified. Given the amount of suffering and death caused by the Bengal famine (which would be obvious to any statesman who deigned to pay the least bit of attention) the neglect of the British government seems almost diabolical. Can we explain the government’s indifference simply by appeal to the absence of a threat to its authority contrived by suppressing the media? It seems that there must have more going on. If thousands of British expats were dying of starvation in Calcutta, it seems likely that the government would have responded more appropriately even if the media were severely restrained and the information flow therefore minimal.

Moreover, Sen notes that in famines only a very small proportion of the population is affected—much less than 10%. Political pressure from this group alone would not be enough to force a democratic government to respond. It is the pressure from the non-suffering members of society that makes the difference. But if government officials in democracies don’t care about the
starving unless they are threatened with a loss of power, why do members of the population who are not starving care about the starving? It seems that if compassion or solidarity (or whatever) moves non-starving citizens to advocate for famine victims, it would move government officials to respond to the famine. This leads me to think that either 1) there is something else going on in democracies (besides “excellent political incentive to act supportively”) that explains their tendency to respond to famines or 2) there is something else altogether that explains the tendency for governments to respond to famines and this factor is somehow contingently connected to democracy.

Be Sociable, Share!
This entry was posted in Books, Posts, Reading Group and tagged , , , . Bookmark the permalink.

3 Responses to Sen, ‘The Idea of Justice’, Chapter 16, ‘The Practice of Democracy’

  1. Cynthia, I have not actually been following the Sen reading group very closely, but there is actually some literature that reviews the famine-democracy link recently and finds support for Sen’s assertion, with some exceptions:

    *Myhrvold-Hanssen, T. L. 2003. Democracy, News Media, and Famine Prevention: Amartya Sen and The Bihar Famine of 1966-67.
    *Neumayerb, E. & Plümpera, T. 2009. Famine Mortality, Rational Political Inactivity, and International Food Aid. World Development 37 (1) , 50-61.
    *Rubin, O. 2009a. “The Merits of Democracy in Famine Protection – Fact or Fallacy?” European Journal of Development Research 21 , 699-717.
    *Rubin, O. 2009b. “The Niger Famine: A Collapse of Entitlements and Democratic Responsiveness.” Journal of Asian and African Studies 44 , 279-297.

    Sen’s point about information is stronger than you perhaps make it seem: bureaucracies in authoritarian governments have lots of incentives to falsify information that goes to the top leadership, as do all bureaucracies, but in democracies there are alternative sources of information (the free press). It is not so much that top leaders in non-democratic countries don’t care (though they sometimes do not), as that, on average, even if they care, they both fail to have sufficient information about the true dimensions of the problem and cannot be easily “shamed” into action by the common knowledge produced by a free media. (Mao is supposed not to have known much about the famines caused by the Great Leap Forward, for example, even though these famines killed millions of people, but even if he had known, it would have been difficult to shame him into action). The informational problems in non-democracies are severe enough that sometimes non-democratic rulers allow for freer media than might be expected; see the recent article by Egorov, Guriev, and Sonin (2009) in the APSR on “Why Resource-poor dictators Allow Freer Media”.

    Moreover, even if only 10% of the population is actually affected by famine (e.g., actually suffering from starvation) a much larger group is usually at risk, and politicians are more likely to respond to such groups (with some exceptions – see the literature above). It is not merely that some care about others starving, but that others see themselves as at risk of starving (many famines occur in extremely poor countries where many people have low food security to begin with) and hence are more likely to pressure governments to do something. (This mechanism may break down if, for example, the starving belong to a different ethnic group or are seen as otherwise distant from one’s own situation).

    Regarding the democracy/growth connection, though the literature is tangled, Sen is very much right: if you simply compare growth rates between democracies and non-democracies (using a measure of democracy like the Polity score, or like the Freedom House score) democracies grow on average more or less as fast as non-democracies, though non-democracies exhibit more variability (there are more growth disasters and more “miracles” among non-democracies; democracies grow with less variability). In fact, if you use the data on growth from the Penn World Tables from 1950 to 1990 and the Polity data on democracy and autocracy for the same period, democracies actually come out slightly ahead, though the result is not necessarily statistically significant given measurement errors. There are of course all kinds of issues about causality that nobody has been able to satisfactorily disentangle, but Sen is right that there is no evidence that democracies grow, on average, more slowly than non-democracies.

    The standard reference is probably Przeworski, Adam et al. 2000. Democracy and Development: Political Institutions and Well-being in the World, 1950-1990. Cambridge University Press, which finds no real differences between dictatorships and democracies in terms of growth, but more recent work supports the idea that democracy is in no way a disadvantage to growth, and may be actually beneficial. See, e.g., Gerring, John et al. 2005. “Democracy and Economic Growth: A Historical Perspective.” World Politics 57(3): 323-364. A recent meta-analysis of a variety of studies also found support for the idea that democracies are better for growth than non-democracies, though via indirect effects on human capital (Doucouliagos, Hristos, and Mehmet Ali Ulubasoglu. 2008. “Democracy and Economic Growth: A Meta-Analysis.” American Journal of Political Science 52 (1):61-83.)

    There are also theoretical reasons to expect that democracies would actually do better than most non-democracies in terms of simple economic growth; see, e.g., Bueno de Mesquita, Bruce et al. 2001. “Political Competition and Economic Growth.” Journal of Democracy 12(1):58-72, which summarizes their book The Logic of Political Survival. So Sen is on pretty solid ground.

    In fact, I don’t think the “received view” that democracies are bad for growth is really “received” any longer among economists and political scientists; even economists who are skeptical of a link between democracy and growth, like Barro or Acemoglu, don’t argue that democracy is bad for growth, or that dictatorships are better than democracies.

  2. I should add that the studies on growth and democracy I mentioned above use comparisons typically involving every country in the world with population over 500,000 people. These are not very limited comparisons between India and some East Asian countries, but very large-N studies, though of course the validity of such studies may be limited by all kinds of other things, e.g., the reliability of democracy or economic growth measures, mis-specification of statistical models, etc. Nevertheless, the balance of evidence seems to suggest, as Sen argues, that democracy is not bad for economic growth and development, and may even be good.

  3. Thanks for your very helpful elaboration of the issues Sen raises in chapter 16. It was very enlightening to me and hopefully to others who generally operate in the domain of theory.

Leave a Reply