Authoritarian elements in democracy
I remember Kanshi Ram, the founder of the Bahujan Samaj Party (BSP), speaking to fellow Dalits at election rallies. As a graduate student then writing a dissertation on the BSP, I heard him speak at many such rallies. Kanshi Ram was nothing if not a democrat. He led a revolution for the empowerment of Dalits and other excluded subaltern groups within India’s democratic framework.
But his egalitarian, democratic ends were achieved through a hierarchical style of representation. At rally after rally, he stood on the dais, with a pen tucked in the pocket of his shirt and a faintly irritable expression on his face, and lectured the throngs of young men before him on what they needed to do to obtain power. It was not an empathetic address but a paternal or a school-masterly one. It helped that he was older than most of his flock. He was, not incidentally, called Sahib (Sir).
A hierarchical style of representation is common among politicians in India. The difference is often in the style of hierarchy, not the fact. One is spoilt for choice between the imperial style of Indira Gandhi, the different but no less imperial style of Narendra Modi,1 the didactic style of Charan Singh,2 the dictatorial style of Bal Thackeray, the regal style of Vasundhara Raje,3 the patronal style of Karunanidhi, and the ‘elder-sisterliness’ of Mamata Banerjee.4 As Kanshi Ram’s example highlights, it is found in even the unlikeliest places, as much among the leaders of subaltern groups as among those backed by dominant groups.
Hierarchy is more commonly associated with authoritarian than democratic government. It does not, of course, amount to authoritarianism. But it is one element in a package of elements associated with authoritarianism. Others include personality cults, a circumvention of institutions, a separation between rulers and ruled, a principle of selection based on appointment or heredity or wealth or something other than popular choice, centralization of power, an arbitrary use of that power, suppression of dissent, rulers who are unconstrained or unaccountable, a suspension of rights and so on.
Democracy, by contrast, is a system associated with elements such as political equality, a principle of selection based on free and fair elections, rule of law, a pluralistic distribution of power, freedom to dissent, rulers who are both constrained and accountable, a guarantee of fundamental rights and so on.
India has many democratic elements: unusually competitive elections; plural centres of power, given both the large number of parties in India’s party system and the multiple levels of government, spanning national, regional and local politics; a vocal and critical media; a remarkably open political system in which the fluidity of electoral coalitions acts as a catalyst for bringing new groups into politics; frequent turnover not only between the parties but also the groups who are in power; and, since the introduction of panchayati raj, a tenacious form of grassroots democracy. What is more, a strong and vibrant democratic culture that has sprung up around these structures and practices.
But India’s democracy also has several authoritarian elements. This does not mean that India is an authoritarian regime. But it does mean that any reckoning of democracy in India has to take the admixture of authoritarian elements into account. The Oxford English Dictionary defines the word ‘admixture’ as ‘something mixed in with something else; a minor ingredient.’ And the often minor authoritarian ingredients mixed into Indian democracy can become major in particular spaces or time periods.
Consider the example of personality cults. The latest example of a cult of personality is the one that has sprung up around Prime Minister Modi. But there are many more. Mayawati, the former Chief Minister of the state of Uttar Pradesh, erected statues to herself while in government. The city of Kolkata is strewn with larger than life posters and cut-outs of Mamata Banerjee. Perhaps the biggest personality cult of them all was the one that enveloped Jayalalithaa, who was worshipped by her supporters as Adhiparasakthi or the ‘omnipotent female power’.
In fact, even those leaders who self-consciously avoid hierarchy consciously cultivate a cult of personality. Take Laloo Prasad Yadav, for example, who purposefully challenges hierarchy as a mode of relating to his constituent public. His style is jocular and affectionate, emphasizing his shared background with subaltern groups, not his superior knowledge and qualifications. But it was under his reign that the government owned Bihar Textbook Corporation published a Class VII textbook with a lesson that described Laloo ‘as the priceless jewel of the soil of Bihar... whose birth can be compared with that of the Enlightenment of Lord Buddha under the Bodhi Tree.’5 Similarly, Arvind Kejriwal’s face and name has become practically synonymous with the Aam Aadmi Party.
Then there is the politics of heredity and family. This is widespread too. 22% of the MPs, 33% of India’s current chief ministers, and the leaders of 36% of parties in Parliament have a dynastic background. And many of those representatives who avoid hierarchy and personality cults, such as Akhilesh Yadav or Omar Abdullah, usually because they are too young to draw on them successfully, relied instead on heredity for their entry into politics.
Now consider the limits on dissent. India has a remarkably broad colonial era sedition law which makes practically any action (speech, writing, signs, or any other visible representation) that is critical of the government (more specifically, ‘which brings or attempts to bring into hatred or contempt, or excites or attempts to excite disaffection towards the government’) punishable by a maximum sentence of life imprisonment.6 The Supreme Court later limited the application of the law only to cases where an incitement to violence had occurred. But both the central and the regional governments, led by parties across the political spectrum, have invoked this law freely and arbitrarily. The most recent, almost farcical example, is the opposition’s demand that BJP MP Tarun Vijay be charged with sedition for his prejudiced remarks on race and colour.7
In recent years, this law has also begun to be used to stifle large-scale protests. In 2012 and 2013, the police under Tamil Nadu’s AIADMK-led government charged 9000 people in Tamil Nadu with sedition for protesting a nuclear power plant.8 In 2016, the BJP-led national government charged JNU student Kanhaiya Kumar with sedition for supposedly shouting anti-India slogans. Hardik Patel, the then 22 year old who led an agitation of the Patel caste in Prime Minister Modi’s home state of Gujarat in 2015, was also charged with, and jailed for, sedition.9 Sedition charges of this sort rarely result in a conviction – but they are now being used as a routine tool to suppress protest through incarceration and harassment.
Finally, consider the ‘emergency powers’ given to the central government by the Indian Constitution. A national level emergency has not been declared since 1975, and constitutional amendments now make the declaration of such an emergency more difficult. The declaration of President’s Rule in individual states has also become significantly less frequent in the past two decades than in the 1970s and 1980s. However, to the extent that these laws remain on the books, and can easily be invoked by elected governments, the shadow of a constitutionally sanctioned authoritarianism continues to be closely linked to the practice of democracy in India.
Further, another form of ‘routine emergency’ in the form of the Armed Forces Special Powers Act (the AFSPA) is in effect in Jammu and Kashmir, Assam, Arunachal Pradesh, Nagaland, Manipur and Mizoram.10 The act empowers the government of these states or the central government to declare all or part of the territory of these states a ‘disturbed area’. In areas declared ‘disturbed’, security forces are empowered to shoot to kill, arrest without warrant, search without warrant, and enter or destroy property. All areas in all these states are not declared disturbed all the time. But as long as the AFSPA remains in effect in these states, fundamental rights can be suspended in any part of these states with relative ease.
This list of authoritarian elements which coexist with democratic ones suggests a change in the way in which we think about democracy and authoritarianism in India. So far the standard interpretation of India is as a mostly democratic country with occasional authoritarian episodes. The Emergency was one such episode, with the blame laid primarily at the door of Indira Gandhi. Present day India, some have suggested, is approaching a second such episode, with the blame laid primarily at the door of Prime Minister Narendra Modi.
But authoritarianism in India is not an episode that can be located in one person, either Gandhi or Modi, or in one demarcated period, either past or present. It is a continuous strain, mixed in with democratic elements, and found continuously in varying strength across time periods, regions and leaders in India.
The balance between authoritarian and democratic features may shift from time to time. Certainly, it shifted decisively towards authoritarianism during the Emergency. There are obvious excesses that happened during the Emergency that have not happened since, including large-scale detentions and arrests, and severe censorship imposed on the media. But there are lower key authoritarian elements that have remained persistently in the background both before and after, sometimes coming to the foreground as well.
Importantly, and paradoxically, these elements coexist with what is also a real and undeniable proliferation of democratic elements. It should not be surprising, then, that this oxy-moronic coexistence also shows up in the data we have on what citizens think. According to a 2013 survey on India done by CSDS-Lokniti, 44% of those who prefer democracy to other forms of government also prefer a strong leader who dispenses with elections and Parliament.11
Further, authoritarian and democratic elements do not just coexist in India’s democracy. They shape each other. In the example that I started with, Kanshi Ram used hierarchy in the service of democracy. The essays in the issue are replete with many more such examples. The essays by Michelutti and Vaishnav suggest that criminal ‘bosses’ in local politics provide service and are accountable in many ways to their constituent publics. In the essay by Ikegame, the guru and other traditional leaders use hierarchy to achieve democratic ends in elections. The essay by Harriss on Jayalalithaa notes that the personality cult around her has gone hand in hand with a strong welfarist agenda. In other work, I have suggested that dynasties can bring about social inclusion in some respects.12
Democratic features in turn shape authoritarian ones. The criminal bosses, patrons, top down reformers, gurus and dynasties discussed here all have constraints on their power based on elections – and also have to operate in a plural marketplace. Their constituencies are not captive. They have the option to go elsewhere. This also shapes and constrains their own behaviour. I will return in the concluding section of this essay to the implication of this coexistence and interaction between democratic and authoritarian elements for how we think about democracy, in India and globally.
India is by no means unique. All democracies have a cache of authoritarian elements, although the content and size of this cache, and the balance between democratic and authoritarian elements varies across space and time. In other democracies too, in a fact that is often ignored, authoritarian features not only coexist with but also shape the democratic elements. To illustrate, I will focus on the two democracies invoked most often in discussions of the classic model of liberal democracy: the United Kingdom and the United States.
The United Kingdom currently has many of the classic elements associated with democracy: constitutionally protected rights, a constrained and accountable government, a free press, rule of law, competitive elections and so on. But, as a monarchy with dedicated space for the aristocracy in the upper house of parliament, it also legitimizes the principle of heredity, and negates the idea of political equality. One could argue that the monarchy and aristocracy are constrained by, and now support, a democratic system. But that only proves the point that authoritarian and democratic elements often interact and shape each other in the countries that we think of as classic cases of democracy.
Further, democracy inside Britain developed in a way that was closely linked to authoritarianism outside, in the form of colonialism. The two were linked in at least three ways. First, the idea of democracy as the rule of ‘we the people’ rests on the idea of exclusion, for there can be no ‘we’ without a ‘they’. Colonialism, by identifying outsiders, helped solidify the sense of Britishness. Second, and relatedly, the idea of democracy as it developed in Britain rested on the notion that some people were fit for it while others were not. There is still an archway in India’s Rashtrapati Bhavan – the former Viceroy’s Palace – with the inscription, ‘Liberty will not descend to a people. A people must raise themselves to liberty. It is a blessing that must be earned before it can be enjoyed.’ This legitimized empire – and in turn strengthened the sense of collective distinctiveness that legitimized democracy in Britain. Third, colonialism helped to produce the wealth which is associated with the flourishing of democracy in Britain and elsewhere.13
Now, consider the example of the United States. Many commentators have suggested that Donald Trump’s election as President has brought authoritarianism to America.14 And indeed, his style of leadership does carry several hallmarks of authoritarianism: it is whimsical and arbitrary, concentrates authority in his person and family while running down institutions, represents an oligarchy, and relies for legitimacy on an ideology of racial exclusion. But while Trump has begun to shift the balance between democratic and authoritarian features in the US, it would be hard to argue that he created authoritarianism in America. Many of the authoritarian elements he is accused of existed in some form prior to his coming to power.
The tendency towards family-based politics in the US, for example, predates Trump. His main opponent in the presidential election, Hillary Clinton, is the wife of a former president. One of his opponents in the Republican primary, Jeb Bush, is the son and brother of other former presidents. Dynasticism also persists in the US Congress and in executive positions.15 Trump’s version of family-based politics is different in the sense that his family members have been appointed to non-transparent and non-accountable advisory positions rather than running for elected office. But this is not the first time an American president has appointed family members to such positions. Bobby Kennedy was his brother John F. Kennedy’s closest adviser before he ran for elected office. In fact, the ‘Federal Anti-Nepotism Statute’ which prohibits a public official from appointing relatives to positions in his agency –which Trump had to circumvented by arguing that his ‘personal staff’ was not covered by the law – has been nicknamed the ‘Bobby Kennedy law’.16
Oligarchy also predates Trump: a 2014 study of public opinion surveys by political scientists at Princeton and Northwestern found that ‘majorities of the American public have little influence over the policies of the government while the rich exercise an effective veto on many issues.’17
The racial exclusion that characterizes Trump’s appeal has also been a recurrent strain in American democracy – and in democracies more broadly. As David Scott Fitzgerald and David Cook-Martin write, in a study of democracy and racism in twenty-two countries including the US, ‘Racism has been the cultural frame that allowed inferences about people’s morality and capacity for democratic participation from their appearance or cultural practices.’18 In the US, it justified racist immigration policies in the 19th century, and the disenfranchisement of African Americans until well into the 20th century.
This exclusivist strain coexists with other inclusive ones and the balance between them shifts from time to time. But it is an old strain that Trump has taken to a new, dangerous level by threatening to build a wall keeping immigrants out, not something that he has introduced to American politics for the first time.
None of this is to say that the UK or the US are authoritarian regimes, any more than India is. Quite the opposite. They have many democratic elements and the balance of democratic elements is still greater than the authoritarian ones. But these elements coexist and influence each other.
In the circumstances it should not be surprising that public opinion polls find that 50% of Britons have ‘authoritarian’ views, correlated with the vote on Brexit,19 and that 44% of white Americans have ‘authoritarian’ worldviews, correlated with the vote on Trump.20 As in India, these views coexist with support for democratic government. And this combination of democratic and authoritarian public attitudes is found elsewhere as well. Hale points in his essay in this issue, for example, to survey evidence in Russia that shows over 31 per cent of the population thought that both democracy and a strongman leader were ‘good fits’ for Russia.
The standard way of thinking about democracy and authoritarianism in political science and outside, is in typologies. The most common typology is a dichotomous one, which distinguishes between democracies and authoritarian regimes. Other typologies include trichotomous ones which distinguish between ‘liberal democracies’, ‘illiberal democracies’ and ‘authoritarian regimes’, or between ‘democracies’, ‘hybrid regimes’ and ‘authoritarian regimes’. Levitsky and Way take this proliferation of categories further, unpacking the different types of ‘hybrid’ regimes, and focusing on one kind in particular – competitive authoritarian regimes.21 There have also been attempts to introduce continuous measures for democracy, including a 10-point scale constructed by the Polity Project or Freedom House’s 100 point score.22 But these continuous measures are inevitably collapsed for heuristic purposes into discrete categories.
These typologies are not simply of academic interest. They come with value judgments attached, according to which liberal democracies are best, illiberal democracies second best and authoritarianism the worst. They confer legitimacy on regimes classified as democratic – even better, liberal democratic – and on the actions taken by their leaders in the name of democracy. They affect policy: for example American foreign policy treats democratic countries differently from non-democratic countries. And they affect self-perception. The fact that India is usually classified as a democratic country, in contrast to its historically authoritarian neighbours, is a major source of national pride.
The problem with these typologies, however, is that they come with blind spots. In typological thinking, countries belong to one and only one category. This prevents us from acknowledging the coexistence of democratic and authoritarian elements and their interaction in many countries.
Consider the case of dynastic politics. This is a pervasive feature in modern democracies. But because dynastic politics is seen as the antithesis of democracy – associated with authoritarian or pre-democratic forms of rule – and because typological thinking forces us to categorize countries as democratic or authoritarian but not both, this prevented us from acknowledging for a long time the coexistence, and interaction, between dynastic and democratic politics. We either underplay the role of heredity in shaping democracy – treating the British monarchy or House of Lords, for instance, as somehow intrinsically democratic – or study heredity as a separate subject – as in the case of dynastic politics in the US Congress, which stimulated several standalone research papers – without integrating it into a theory of democracy.
Similarly, we study British colonialism as a phenomenon that is separate – geographically and analytically – from British democracy without acknowledging the role that one played in shaping, and shoring up, the other. Empirical studies of the effect of British colonialism on democracy typically investigate the effect that British colonialism had on sustaining democracy in the colonies, not on the mutually reinforcing relationship between colonialism and democracy in Britain itself. In the US, we study oligarchy in America as something separate from democracy or as the anti-thesis of democracy: the standard head-line is that America is an oligarchy, not a democracy.23 But in fact, America combines both oligarchic and democratic elements in the same regime.
Changing the classifications of individual countries does not fix the problem. Consider again the case of the US, usually classified as a liberal democracy. Now, with Trump in charge, many commentators have changed the classification of the US from a ‘liberal’ to an ‘illiberal’ democracy.24 But downgrading the contemporary US into the category of ‘illiberal democracy’ downgrades the US present while upholding, even glorifying, the category of liberal democracy – and simultaneously glorifying the US past. American democracy, this implies, may be illiberal now, but it was unsullied by authoritarian features earlier – and in that earlier golden period, it fit the category of a liberal democracy. But my point here is that there is no democracy that is unsullied by authoritarian elements: the category itself, and the typology that contains it, both bear questioning, not simply the placement of countries within that category and typology.
It’s the same with India. The classification of present-day India as democratic rather than authoritarian has produced a literature that celebrates its successes in electoral politics – broad based turnout, competitive politics, anti-incumbency – to the point of fetishization, without integrating into this literature other equally valid insights about authoritarian elements, such as heredity, coerciveness, the personalization of power, limits on dissent, and militarization which also characterize Indian democracy. Further, the sharp separation between the Emergency as an ‘authoritarian’ period and the rest of India’s post-independence history as ‘democratic’ blinds us to the way in which authoritarian features have continuously characterized and shaped India’s democracy.
Classifying India as an illiberal democracy does not solve the problem either. It reifies the category of a ‘pure’ liberal democracy while at the same time placing India outside that category. But it is not the case that a class of relatively ‘pure’ democracies exists, and India has failed to pass muster. In fact, any work that creates a special category in which to place India – including my own previous work which classifies India as an example of a special class of democracies termed ‘patronage democracies’25 or Partha Chatterjee’s work which classifies democracy as it is practiced in India as an example of ‘political society’26 – is problematic to the extent that by introducing a separate category in which to place India, they uphold at the same time categories such as ‘liberal democracy’ or ‘civil society’ which India is believed to deviate from. Part of what I have tried to suggest in this essay is that the category of liberal democracy itself is problematic, to the extent that it blinds us to the routine presence of authoritarian elements even in those countries which are routinely classified as ‘liberal democracies’.
When it comes to democracy and authoritarianism, then, it is better to think in admixtures rather than typologies. ‘Admixing’, as I noted above, refers to ‘the action of adding an ingredient to something else.’ The ingredients I refer to here are the authoritarian elements which are mixed into a democratic regime. The admixture of these authoritarian elements does not make democratic regimes authoritarian. Nor does it imply that all democratic regimes are the same. Democratic regimes can and do vary in which and how many authoritarian elements they contain, and in the balance of authoritarian to democratic elements. This balance often does not tip over into authoritarianism.
But thinking in admixtures allows us to acknowledge and measure this variance – and place democratic regimes on a continuum according to the ratio of authoritarian to democratic elements.27 This way of thinking is easily measurable, perhaps more precisely measurable than dichotomous or trichotomous typologies. At the same time, the advantage it has over typologies is that it allows us to acknowledge the presence of authoritarian elements, and also theorize about their interaction with democratic elements, which a typological form of thinking does not permit.
The intuition behind this suggestion is not entirely new. One related statement comes from Ayesha Jalal who wrote in 1995: ‘Far from representing a new and sharp dichotomy, democracy and authoritarianism are reflective of ongoing struggles between dominance and resistance. Without blurring the distinction between them it is important to acknowledge that they may frequently overlap irrespective of the formal designation of polities and states as democratic or authoritarian. It seems more apt to view democracy and authoritarianism as both antithetical and interdependent historical processes, coexisting in tension while at the same time each informing and transforming the other.’28
But the concept remained under-developed in Jalal. It was too quickly overwhelmed by historical detail. Further, and more problematically, Jalal also dismissed the many aspects of participatory democracy in India, suggesting in the process that India was not that different from Pakistan or Bangladesh.
I have tried to develop the concept of admixture in a fuller and more precise way. Further, the idea as I develop it here acknowledges India’s democratic elements as fundamental. Recognizing the existence of authoritarian elements does not negate the presence or import of democratic ones. Quite the opposite: they coexist and can, paradoxically, even strengthen each other. Recognizing this does not collapse the distinctions between democratic and authoritarian regimes in South Asia or outside, but renders them more precisely.
*I am grateful to the Harvard Academy and Jorge Dominguez for inviting me to organize a panel which explored some of the themes developed in this essay, and to Uday Chandra, Brian Min, Rama Mantena and the contributors of the essays in this volume for helpful insights.
1. Kanchan Chandra, ‘Emperor Modi Ended the Gandhi Dynasty’, Foreign Policy, 8 August 2016. http://foreignpolicy.com/2016/08/08/emperor-modi-killed-the-gandhi-dynasty/. http://foreignpolicy.com/2016/05/18/all-hail-queen-banerjee/; Kanchan Chandra, ‘Authoritarian India’, Foreign Affairs, 16 July 2016.
2. Yogendra Yadav, quoted in Sunil Khilnani, Incarnations: A History of India in Fifty Lives. Farrar, Straus and Giroux, New York, 2016.
3. Anastasia Piliavsky, ‘Elective Aristocracy or Political Representation the Indian Way’, Talk given at UCL on 9 October and in Madison on 18 October 2014.
4. Kanchan Chandra and Saroj Nagi, ‘All Hail Queen Banerjee’, Foreign Policy, 18 May 2016. http://foreignpolicy.com/2016/05/18/all-hail-queen-banerjee/.
6. http://ncw.nic.in/acts/THEINDIAN PENALCODE1860.pdf; http://www.bbc. com/news/world-asia-india-37182206
7. http://www.hindustantimes.com/opinion/black-hole-of-prejudice-tarun-vijay-only-reflects-our-coloured-view-on-race/story-F9Bw4arr2EaDpeSLrzQDMK.html; http://indianexpress.com/article/india/for-his- remarks-opposition-says-book-tarun-vijay-for-sedition-4608015/
8. http://www.bbc.com/news/world-asia-india-37182206; http://indianexpress.com/ article/india/india-news-india/kudankulam-nuclear-plant-protest-sedition-supreme-court-of-india-section-124a-3024655/
9. http://www.hindustantimes.com/india-news/hardik-patel-s-six-month-exile-over- sedition-cases-ends-to-reach-gujarat-today/story-9PqCY3atiRvtaaWn7jKSfM.html
10. Sanjib Baruah, ‘Routine Emergencies: India’s Armed Forces Special Powers Act’, in Nandini Sundar and Aparna Sundar (eds.), Civil Wars in South Asia: State, Sovereignty, Development. Sage, New Delhi, 2014 and http://www.india-seminar.com/2010/615/615_sanjib_baruah.htm
11. Lokniti-Centre for the Study of Developing Societies (CSDS), Democracy in India: A Citizen’s Perspective, 2013. Chapter 1, Figure 1.5.
12. Kanchan Chandra (ed.), Democratic Dynasties. Cambridge University Press, New York, 2016.
13. Adam Przeworski, Michael Alvarez, Jose Antonio Cheibub and Fernando Limongi, Democacy and Development. Cambridge University Press, New York, 2000.
15. https://www.brookings.edu/opinions/ political-dynasties-an-american-tradition/
18. David Scott Fitzgerald and David Cook-Martin, Culling the Masses. Harvard University Press, Cambridge, 2014.
20. http://www.vox.com/2016/3/1/11127424/trump-authoritarianism. Support among black Americans is ‘higher’ but the measurement on this may be a matter of error. See Efren O. Perez and Marc J. Hetherington, ‘Autoritarianism in Black and White’, Political Analysis 22, 2014, pp. 398-412. Also see Marc J. Hetherington, Authoritarianism and Polarization in American Politics. Cambridge University Press, New York, 2009.
21. Steven Levitsky and Lucan Way, ‘The Rise of Competitive Authoritarianism’, Journal of Democracy 13(2), April 2002.
22. https://freedomhouse.org/report/fiw-2017-table-country-scores; http://www. systemicpeace.org/polity/polity4.htm
24. See for instance Fareed Zakaria, the author of an influential 1997 article on ‘Illiberal Democracies’, discussing the US in 2016 in https://www.washingtonpost.com/opinions/america-is-becoming-a-land-of-less-liberty/2016/12/29/2a91744c-ce09-11e6-a747-d03044780a02_story.html?utm_term=. 1118485a8700
25. Kanchan Chandra, Why Ethnic Parties Succeed: Patronage and Ethnic Head Counts in India. Cambridge University Press, New York, 2004.
26. Partha Chatterjee, The Politics of the Governed. Columbia University Press, New York, 2004.
27. Such a continuum could also cover authoritarian regimes, thus replacing the typology altogether with a continuum which measures the ratio of the democratic elements. But, since I have not covered the democratic elements in authoritarian regimes at all here, that is a subject for a different essay.
28. Ayesha Jalal, Democracy and Authoritarianism in South Asia. Cambridge University Press, New York, 1995.