The Puzzle of Indian Democracy: A Consociational Interpretation

Author(s): Arend Lijphart

Source: The American Political Science Review, Vol. 90, No. 2 (Jun., 1996), pp. 258-268
Published by: American Political Science Association
Stable URL: https://www.jstor.org/stable/2082883
Accessed: 08-12-2018 11:30 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Political Science Association is collaborating with JSTOR to digitize, preserve and
extend access to The American Political Science Review

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2 June 1996

The Puzzle of Indian Democracy: A C

AREND LIJPHART University of California, San Diego
I ndia has been the one major deviant case for consociational (power-sharing) theory, and its sheer size
makes the exception especially damaging. A deeply divided society with, supposedly, a mainly majoritarian
type of democracy, India nevertheless has been able to maintain its democratic system. Careful
examination reveals, however, that Indian democracy has displayed all four crucial elements of power-sharing
theory. In fact, it was a perfectly and thoroughly consociational system during its first two decades. From the
late 1960s on, although India has remained basically consociational, some of its power-sharing elements have
weakened under the pressure of greater mass mobilization. Concomitantly, in accordance with consociational
theory, intergroup hostility and violence have increased. Therefore, India is not a deviant case for consocia-
tional theory but, instead, an impressive confirming case.

India has long been a puzzle for students of compar- characterized by (1) grand coalition governments that
ative democratic politics. Its success in maintaining include representatives of all major linguistic and reli-
democratic rule since independence in 1947 (exclud- gious groups, (2) cultural autonomy for these groups, (3)
ing the brief authoritarian interlude of the 1975-77 proportionality in political representation and civil ser-
Emergency) in the world's largest and most heteroge- vice appointments, and (4) a minority veto with regard to
neous democracy runs counter to John Stuart Mill's vital minority rights and autonomy. In contrast, under
(1958, 230) proposition that democracy is "next to majoritarian winner-take-all democracy- characterized
impossible" in multiethnic societies and completely im- by the concentration of power in bare-majority one-
possible in linguistically divided countries.' And it con- party governments, centralized power, a disproportional
founds Selig S. Harrison's prediction (1960, 338), in line electoral system, and absolute majority rule-consocia-
with Mill's argument, of India's democratic failure tional theory regards stable democracy in deeply divided
and/or territorial disintegration: "The odds are almost societies as highly unlikely. In other words, consocia-
wholly against the survival of freedom and ... the issue tional theory maintains that power sharing is a necessary
is, in fact, whether any Indian state can survive at all." (although not a sufficient) condition for democracy in
The Indian puzzle is even more troublesome for conso- deeply divided countries.
ciational (power-sharing) theory. In contrast with Mill's Consociational theory has had a strong influence on
and Harrison's thinking, power-sharing theory holds that comparative politics, and it has spawned a vast litera-
democracy is possible in deeply divided societies but ture. Soon after it was formulated, Daalder (1974, 609)
only if their type of democracy is consociational, that is, spoke of "an incipient school" of consociationalism, and,
a few years later, Powell (1979, 295) proclaimed the
Arend Lijphart is Research Professor of Political Science, University theory "among the most influential contributions to
of California, San Diego, La Jolla, CA 92093-0521.
comparative politics." It has become a widely accepted
Earlier versions of this article were presented as seminar or confer-
ence papers at the Centre for the Study of Developing Societies, Delhi,
paradigm for the analysis of democracies that can be
on December 3, 1993; the Department of Political Science at Delhi regarded as the prototypes of power sharing, such as the
University, South Campus, on December 8, 1993; "Regime Transfor- Netherlands (Daalder and Irwin 1989; Mair 1994), Bel-
mation and Democratization in Comparative Perspective," University
gium (Huyse 1987; Zolberg 1977), Austria (Powell 1970;
of California, Los Angeles, May 20-21, 1994; and the Department of
Luther and Muller 1992), Switzerland (Lehmbruch
Political Science, University of California, Santa Barbara, May 15,
1995. A preliminary version was published as an occasional paper by 1993; Linder 1994; Steiner 1990), Lebanon (Dekmejian
the Rajiv Gandhi Institute for Contemporary Studies in New Delhi 1978; Messarra 1994), Malaysia (Von Vorys 1975;
(RGICS Paper No. 18, 1994). It is part of a collaborative and Zakaria 1989), and Colombia (Dix 1980; Hartlyn 1988).
comparative U.S.-Indian research project, directed by K. S. Bajpai and
And it has been used for the interpretation of many
supported by the Ford Foundation and the Rajiv Gandhi Foundation.
I would like to acknowledge the valuable assistance of the Centre for other political systems, from tiny Liechtenstein (Batliner
Policy Research in New Delhi and its director, V. A. Pai Panandiker, 1981) to the European Union (Chryssochoou 1994;
and of the Library of the India International Centre, New Delhi. For Gabel 1994; Hix 1994; Lindberg 1974); in all parts of the
helpful comments on earlier drafts, I am grateful to Kanti Bajpai, Paul
world, for instance, Canada (Cannon 1982), Venezuela
R. Brass, Pradeep K. Chhibber, Jyotirindra Das Gupta, Henry W.
(Levine 1973), Suriname (Dew 1994), Italy (Graziano
Ehrmann, Dipak K. Gupta, Thomas A. Koelble, Victor V. Magagna,
G. Bingham Powell, Jr., V. Ramachandran, Varun Sahni, Ashutosh 1980), Nigeria (Chinwuba 1980), Gambia (Hughes
Varshney, and three anonymous referees. 1982), Kenya (Berg-Schlosser 1985), and Sri Lanka
I Two other puzzles are posed by Indian democracy. The first is its (Chehabi 1980); and not only democracies but also such
survival despite widespread poverty and illiteracy (Dahl 1989, 253),
nondemocratic states as the former Yugoslavia (Gold-
which casts grave doubts on the hypothesized link between the level of
socioeconomic development and stable democracy, further weakened man 1985; Vasovic 1992) and the former Soviet Union
by the fact that several other Third World democracies have by now (Van den Berghe 1981, 190-91). Furthermore, conso-
established stable democratic rule (e.g., Barbados, Botswana, Costa ciational democracy has been proposed as a normative
Rica, Jamaica, Malta, Mauritius, and Papua New Guinea). The
model for many ethnically divided countries, and it had
second, which I shall discuss later, is Myron Weiner's (1989, 9) "Indian
a decisive influence in the shaping of South Africa's 1994
paradox," that is, "the far more puzzling contradiction between India's
high level of political violence and its success at sustaining a demo- power-sharing constitution (Huntington 1988; Lijphart
cratic political system." 1994; Worrall 1981). Given its prominent status, conso-

258

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2

ciational theory has received a commensurate amount of its negative significance for consociational theory was
criticism (e.g., Barry 1975; Halpern 1986; Horowitz relatively mild.
1985; Taylor 1992), but it has successfully held its own, This argument can be taken much farther, however,
partly by rebutting its critics and partly by incorporating on the basis of a more thorough examination of the
many of the critics' concerns (Lehmbruch 1993; Lijphart Indian case. The evidence clearly shows that India has
1985, 83-117; Steiner and Dorff 1980). always had a power-sharing system of democracy, espe-
Nevertheless, consociational theory has remained vul- cially strongly and unmistakably during its first two
nerable on one major count: the glaring exception of decades of independence, from 1947 to 1967, but con-
India to its otherwise unblemished empirical validity. tinuing, albeit in somewhat attenuated form, after about
Indian democracy has worked despite the fact that, 1967. As Indian democracy has become less firmly
according to the usual interpretation (Pathak 1993, 36; consociational, intergroup tensions and violence have
Weiner 1989, 78), the Indian political system devised by increased. If this reinterpretation is correct, as I shall try
the founding fathers was patterned after the majoritar- to demonstrate, then India is no longer a deviant case
ian and adversarial Westminster model. B. K. Nehru for consociational theory and, in fact, becomes an im-
(1986, 74) writes that the Indian mind was "completely pressive confirming case.3
conditioned to believing that whatever was British was
best" and calls it no wonder that the Indian Constitution THE FOUR ELEMENTS OF
is but an "amended version" of the 1935 Government of POWER-SHARING IN INDIA
India Act. And Paul R. Brass (1991, 342) states that "the
consociationalists ... consistently ignore the experience Indian democracy has clearly exhibited all four of the
of India, the largest, most culturally diverse society in the defining characteristics of power sharing also found in
world that has ... functioned with a highly competitive the other prominent examples of consociational systems:
Canada from 1840 to 1867 (strictly speaking, a conso-
and distinctly adversarial system of politics." A theory
ciational predemocracy), the Netherlands from 1917 to
with only one disconfirming case comes close to perfect
1967, Lebanon from 1943 to 1975 and again after the
validity, of course,2 but one cannot simply shrug off a
1989 Taif Accord, Switzerland since 1943, Austria from
deviant case that looms as large as India's huge democ-
1945 to 1966, Malaysia since 1955 with a temporary
racy, with its 900 million inhabitants.
breakdown from 1969 to 1971, Colombia from 1958 to
In the admittedly rare attempts to come to terms with
1974, Cyprus from 1960 to 1963, Belgium since 1970,
the Indian exception, consociational scholars have con-
Czecho-Slovakia from 1989 until the 1993 partition of
ceded that India's democracy is, in line with the usual
the country, and South Africa according to its 1994
interpretation, mainly majoritarian because of the fre-
interim constitution (Lijphart 1977, 1992, 1994; Olson
quency of one-party majority cabinets, the highly cen-
1994).
tralized federal system that K. C. Wheare (1964, 28)
considers only "quasi-federal," and a highly dispropor-
tional electoral system that has regularly enabled the Grand Coalition
Congress Party to win parliamentary majorities without
Government by grand coalition can take many different
ever winning a majority of the popular vote. Yet, they
forms. The modal form is an inclusive cabinet coalition
have claimed that India is not completely majoritarian,
of ethnic, linguistic, or religious parties, as in the Aus-
citing Rajni Kothari's (1970, 421) description of the
trian, Malaysian, and South African power-sharing sys-
Indian political system as a "coalitional arena," akin to a
tems, but there are many other possibilities. One entails
grand coalition, and the autonomy for the major linguis-
the formation of grand governing coalitions in sites
tic groups provided by the coinciding linguistic and state
other than the cabinet, such as the Dutch pattern of
boundaries of India's federal design, and they have
permanent or ad hoc "grand" councils or committees
equivocated between calling India nonconsociational with much greater influence than their formal advisory
(Lijphart 1977, 181, 225) and semiconsociational (Lijp- role. Another entails grand coalitions in cabinets, de-
hart 1979, 513; Powell 1982, 215). In other words, the fined not in partisan terms but more broadly in terms of
argument was that, while India remained a deviant case, the representation of linguistic or other groups in a
predetermined ratio; for instance, Belgian cabinets have
rarely been coalitions of all significant parties, but they
2 Three other counterexamples mentioned by Powell (1979, 296) are
have been ethnically "grand" because of the constitu-
Sri Lanka, Trinidad, and the Philippines, but the first two are cases of
majority "control" instead of genuinely democratic majority rule with tional rule that cabinets must consist of equal numbers
alternating majorities, in Ian Lustick's (1979) sense of the term. of Dutch-speakers and French-speakers. Yet another
Lustick argues that power sharing is not the only method that can option entails neither cabinets nor parties: the alloca-
maintain stability in divided societies; the alternative is a system of
tion of top governmental offices-such as the presi-
control in which a dominant group uses its superior power to keep the
other group or groups subordinate. In control democracies, power is dency, prime ministership, and assembly speakership in
almost permanently in the hands of the majority group (Sinhalese in Lebanon, and the presidency and vice-presidency in
Sri Lanka, Africans in Trinidad, and, until 1972, Protestants in Cyprus-to specified ethnic or religious groups.
Northern Ireland), and the minorities are excluded from power and
The Indian case adds even greater variety. Its main
often discriminated against. In the case of the Philippines, it is doubtful
that we can speak of a true deeply divided society, and, in any case,
democracy broke down in 1972 and was not restored for many years 3~ India obviously remains deviant in terms of Mill's and Harrison's
(see Lijphart 1985, 103). nonconsociational thinking, mentioned earlier.

259

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 The Puzzle of Indian Democracy June 1996

vehicle for grand coalition is the cabinet, which is not an cies: (1) federal arrangements in which state and linguis-
exceptional form, but the unique aspect in India is that tic boundaries largely coincide, thus providing a high
cabinets are produced by the broadly representative and degree of linguistic autonomy, as in Switzerland, Bel-
inclusive nature of a single, dominant party, the Con- gium, and Czecho-Slovakia; (2) the right of religious and
gress Party. In a seminal article, originally published in linguistic minorities to establish and administer their
1964, Kothari (1989, 21-35) tried to analyze the Indian own autonomous schools, fully supported by public
party system from the comparative perspective of the funds, as in Belgium and the Netherlands; and (3)
distinction between one-, two-, and multiparty systems. separate "personal laws"- concerning marriage, di-
He found that the intermediate category of one-party vorce, custody and adoption of children, and inheri-
dominance provides a reasonably good fit but that tance-for religious minorities, as in Lebanon and
Indian one-party dominance is still quite different from Cyprus. Indian democracy has had all these three forms,
the authoritarian type in a country like Ghana. The the last two from the very beginning and linguistic
Congress Party's location in the center means that minor federalism since the 1950s.
parties surround it on all sides. These, in turn, which The British colonial rulers of India drew the admin-
Kothari (1989, 22-23) calls "parties of pressure," per- istrative divisions of the country without much regard for
form the role of preventing the ruling "party of consen- linguistic or cultural cohesion. The Congress movement
sus" from straying too far from "the balance of effective was opposed to this policy and committed itself to a
public opinion." Hence, he assigns a separate conceptual thorough redrawing of the boundaries along linguistic
category to India's party system, uniquely occupied by lines; from 1921 on, it also based its own organization on
the Indian case: the "Congress system."4 One important linguistically homogeneous units, the so-called Pradesh
conclusion that emerges from this classificatory exercise Committees. Jawaharlal Nehru and other Congress
is to highlight the vast differences between the Congress leaders had second thoughts, however, and the Constit-
system, with virtually permanent rule by a centrist party, uent Assembly, following the advice of its Linguistic
and the Westminster-style two-party system, with alter- Provinces Commission, decided not to incorporate the
nation in office by right-wing and left-wing parties. linguistic principle into the new Constitution. Pressures
The second major conclusion is that the Congress from below forced a complete change of policy in the
system has served as the foundation for a consociational 1950s. After the State of Madras was divided into the
grand coalition. Despite never winning a majority of the separate Tamil-speaking and Telugu-speaking states of
popular vote in parliamentary elections, the Congress Tamil Nadu and Andhra Pradesh in 1953, the States
Party has been balanced in the political center and has Reorganization Commission embraced the linguistic
encompassed "all the major sections and interests of principle and recommended drastic revisions in state
society" (Kothari 1989, 27). Prior to independence the boundaries along linguistic lines in 1955. These were
Congress was already an internally federal organization quickly implemented in 1956, followed by the creation of
with a high degree of intraparty democracy and a strong several additional states in later years.
penchant for consensus. This "historical consensus" Linguistic federalism has not fully satisfied the minor-
Kothari (1989, 23, 51) writes, was successfully trans- ities' desire for autonomy and security. The balance of
formed into a "consensus of the present," and he comes power in the Indian federal system was asymmetrical in
close to using consociational terminology in describing favor of the central government from the beginning, and
Indian democracy as a "consensus system which oper- further centralization has occurred from the late 1960s
ates through the institution of a party of consensus," on, a subject to which I shall return below. As a result,
namely, the Congress Party. Crawford Young (1976, many states have been demanding greater autonomy.
314) makes the same point in explicitly consociational The special autonomous status constitutionally granted
language: "Lijphart's theory of consociational democ- to Kashmir, the one Muslim-majority state, was in
racy has application to the Indian pattern of integration. practice also soon undermined, and smaller linguistic
... At the summit is a national political elite who are minorities without statehood have agitated for the cre-
committed to reconciling differences through bargaining ation of new states. But the leadership's initial fears that
amongst themselves." The combination of the Congress linguistic federalism would strengthen fissiparous ten-
Party's inclusive nature and political dominance has dencies have not been realized, and, in retrospect, the
generated grand coalition cabinets with ministers be- policy is regarded as a success by most observers. As
longing to all the main religious, linguistic, and regional consociational theory would have predicted, the "ration-
groups. alizing [of] the political map of India" has made lan-
guage "a cementing and integrating influence" instead
of a "force for division" (Kothari 1970, 115; see also
Cultural Autonomy
Banerjee 1992).
Cultural autonomy for religious and linguistic groups The crucial feature of educational autonomy is not
has taken three main forms in power-sharing democra- just the minorities' right to set up and run their own
schools but the ability to make this right effective
4A further comparison with Japan, not yet so obvious in the early through full government financial support of these
1960s, reveals the additional contrast between India's centrist Con-
schools. Dutch and Belgian religious minorities had to
gress Party and Japan's right-of-center Liberal Democrats. Mexico's
fight hard to obtain this right, and, while full educational
Institutional Revolutionary Party (PRI) is probably the closest parallel
to the centrist Congress Party, except that it does not operate in a fully autonomy was granted in the Netherlands in 1917, it was
competitive democratic setting. not instituted in Belgium until 1958. In India, however,

260

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2

the constitution provided this right from the outset. circumvent these disproportional effects. For instance,
Article 30 states that "all minorities, whether based on despite Malaysia's plurality elections, the interethnic
religion or language, shall have the right to establish and coalition has succeeded in guaranteeing a nearly propor-
administer educational institutions of their choice" and, tional share of parliamentary seats to the minority
more important, that "the State shall not, in granting aid Chinese and Indian parties by giving them the coalition's
to educational institutions, discriminate against any ed- exclusive nomination in a number of districts.
ucational institution on the ground that it is under the In India, too, power sharing has managed to coexist
management of a minority, whether based on religion or with the plurality electoral system inherited from the
language." British. One reason is that plurality does not disfavor
Separate personal laws for Hindus, Muslims, and geographically concentrated minorities, and India's lin-
smaller religious minorities already existed under British guistic minorities are regionally based. Another is that
rule, and they were carried forward and sometimes the Congress Party's repeated manufactured majorities
amended or replaced by similar new laws in independent have not come at the expense of India's many minorities
India. Examples are the 1955 Hindu Marriages Act, the due to its special status as the "party of consensus,"
1956 Hindu Succession (that is, inheritance) Act, the which has been deliberately protective of the various
1937 Muslim Personal Law (Shariat) Application Act, religious and linguistic minorities. Indian cabinets, which
the 1939 Dissolution of Muslim Marriages Act, and the have been mainly Congress cabinets, also have accorded
1872 Indian Christian Marriage Act (Fyzee 1964; Engi- shares of ministerships remarkably close to proportional,
neer 1987). These statutes were enacted by parliamen- especially given the constraint of only about twenty
tary majorities but, when intended for one of the minor- positions usually available, to the Muslim minority of
ities, were drafted in conformity with the minority's about 12% and even the much smaller Sikh minority
wishes. For instance, after the controversial 1985 Shah (roughly 2%), as well as to the different linguistic groups,
Bano decision by the Supreme Court (involving the right states, and regions of the country (Pai Panandiker and
of a divorced Muslim woman to financial support from Mehra 1996). In addition, a special feature of the
her former husband), a new Muslim Women (Protection electoral law guarantees the so-called Scheduled Castes
of Right on Divorce) Act was adopted in 1986, largely in (untouchables) and Scheduled Tribes (aboriginals) pro-
line with the wishes of the Muslim Personal Law Board. portional shares of parliamentary representation by
And the new 1993 Christian Marriage Act was proposed means of "reserved seats," that is, seats for which only
by the government after extensive consultations with and members of these groups are allowed to be candidates.
the final approval of all Christian churches, albeit only Finally, these scheduled groups and the so-called Other
reluctant endorsement by the Roman Catholic Church. Backward Classes have benefited from other quotas-
The Constituent Assembly explicitly considered the so-called reservations-with regard to public service
question of whether separate personal laws ought to be employment and university admissions (Mehta 1991;
continued in independent, democratic India. An amend- Prasad 1991; Srinivasavaradan 1992, 105-33).5
ment to the draft constitution was proposed that would
have ended this form of religious autonomy: "The Union
Minority Veto
or the State shall not undertake any legislation or pass
any law ... applicable to some particular community or The minority veto in power-sharing democracies usually
communities and no other" (cited in Luthera 1964, 83). consists of merely an informal understanding that mi-
Significantly, such a clause was not included in the norities can effectively protect their autonomy by block-
constitution. A year later, Law Minister B. R. Ambed- ing any attempts to eliminate or reduce it. The major
kar, replying to accusations of discrimination on the exception is countries in which one or a few minorities
ground of religion during a parliamentary debate, again face a solid majority (such as Belgium, Cyprus, and the
emphatically endorsed the principle of minority personal former Czecho-Slovakia), and the minority veto is for-
laws: "The Constitution permits us to treat different mally entrenched in the constitution. India has a numer-
communities differently and if we treat them differently, ical Hindu majority of about 83 percent, but the Hindus
nobody can charge the Government with practising are so thoroughly divided by language, caste, and sect
discrimination" (cited in Luthera 1964, 86). that they do not form a political majority. A good
example of the informal veto in Indian politics is the
1965 agreement by the central government that Hindi
Proportionality
would not be made the exclusive official language with-
In accordance with the principle of proportionality, the out the concurrent approval of the major non-Hindi
normal electoral system in power-sharing democracies is speaking regions, in effect giving a veto to the southern
proportional representation (PR). The plurality (first- states, which had opposed dropping English as a lan-
past-the-post) and other majoritarian methods have the guage of administration. The provision works best if the
tendency to overrepresent majorities and large parties
and to discriminate against smaller minority parties, as 5 Clearly, the consociational interpretation does not fit India's caste
well as the corollary tendency to create artificial parlia- conflict as well as it fits the linguistic and religious divisions. In the
early years, an accommodation with the Scheduled Castes was reached,
mentary majorities for parties that fall considerably
but further accommodation with the backward castes came about only
short of winning popular vote majorities, what Rae
later and mainly in parts of southern India. Especially in northern
(1967, 74-77) has called "manufactured majorities." It is India, where there has been little intercaste accommodation, caste
not impossible, however, for power-sharing systems to conflict is the most serious (see Frankel 1988).

261

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 The Puzzle of Indian Democracy June 1996

minority veto does not have to be used very often in mained basically consociational rather than "not conso-
order to protect minority rights and autonomy, and this ciational at all."
has been the case in India. No attempts have been made
to reverse linguistic federalism, and, while opposition to
INDIA'S POWER-SHARING SYSTEM:
educational autonomy has been increasing, no govern-
HOW MUCH OF A SURPRISE?
mental actions to weaken or abolish it have been under-
taken. The one clear instance of the actual use of the Categorizing India as one of the consociational democ-
minority veto occurred in the mid-1980s in connection racies, completely on a par with the other well-known
with the separate personal laws: The Muslim minority cases, is a novel interpretation, although several scholars
saw the Supreme Court decision in the Shah Bano case have identified particular instances of power sharing in
as an attack on Muslim personal law, and it succeeded in India (even Brass 1991; Kothari 1989; Young 1976; see
vetoing this decision by persuading the government to also Hardgrave 1993; Weiner 1969). What needs to be
propose, and parliament to enact, a law reversing the emphasized, however, is that, from the perspective of
court's judgment. consociational theory, the adoption of power sharing by
The one respect in which India does seem to differ India and its maintenance for nearly half a century is not
from the other consociational democracies is that power at all unexpected or surprising. For one thing, consocia-
sharing was not instituted by a deliberate and compre- tional theory places great emphasis on the contribution
hensive agreement, such as the 1917 Pacification in the of prudent and constructive leadership in the develop-
Netherlands, the 1943 National Pact in Lebanon, the ment of successful power-sharing systems. Jawaharlal
1945 Grand Coalition accord in Austria, and the Ma- Nehru is an almost perfect example of such leadership.
layan Alliance of the early 1950s. But not all consocia- He was prime minister from 1947 until his death in 1964,
tional democracies have been established by a compact during the heyday of Indian power sharing. Kothari
of this kind of comprehensiveness and intentionality; in (1976, 15-16) comments that in India "it is essential that
Belgium and Switzerland, for instance, power sharing the institutional system provides for widespread diffu-
developed in a slow step-by-step fashion over more than sion of power. That this happened to a significant degree
a century, and Daalder (1974) has argued that even the under Nehru, and that this trend even appeared to grow
Dutch Pacification should be seen as merely one step in stronger in the later part of his career, is a tribute
a long incremental process. This means that India's [mainly] to the democratic values, vision, and self-
incremental and sometimes haphazard development of confidence of one man." That Nehru was not a fully
power sharing is somewhat unusual among consocia- convinced consociational thinker is shown by his initial
tional democracies but not at all unique. opposition to the principle of linguistic federalism. But
In the face of overwhelming evidence concerning the his leadership combined firmness and self-confidence
consociational character of India's democratic system, with flexibility and tolerance, and he unfailingly re-
how can we explain the explicit and complete rejection spected and promoted the internally democratic and
by Brass (1991, 342-43) of the applicability of consocia- federal nature of the Congress Party. Even on the issue
tional theory to India? Brass claims that India is not at of linguistic federalism, he turned out in the end to be a
all a consociational democracy and, on the contrary, has consociational practitioner. In Kothari's (1970, 157)
"functioned with a highly competitive and distinctly words once again, "Nehru's understanding of the con-
adversarial system of politics." One explanation is that sensus framework represented by the Congress was
he defines power sharing in much too narrow terms. His better than that of most of his contemporaries," al-
main point is that India has had a variety of interethnic though he operated "more on the intuition of a prag-
and intercommunal as well as monoethnic parties and matic politician than on any intellectual grasp of the
sometimes coalitions among these. The implication is logic of the system."
that only cabinet coalitions of monoreligious and mono- Furthermore, consociational theory tries to explain
ethnic political parties deserve to be regarded as grand the probability that power sharing will be instituted and
coalitions, which is obviously incorrect in view of the maintained in divided societies in terms of nine back-
great variety of forms that grand coalitions can assume. ground factors that may favor or hinder it. Since most of
Moreover, by focusing exclusively on parties and coali- these conditions are favorable in India, it is again not
tions, Brass completely ignores the evidence with regard very surprising that consociational democracy was estab-
to autonomy, proportionality, and the minority veto. lished and has worked quite well. The following brief
Brass (1991, 343) concedes that India "has adopted review of the nine factors rates India on each; the two
many consociational devices, some permanently, some most important factors are listed first.
temporarily," but he fails to see that together these (1) The most serious obstacle to power sharing in
devices add up to a fully consociational system. Com- divided societies is the presence of a solid majority that,
pared to India, the other consociational democracies do understandably, prefers pure majority rule to consocia-
not have any additional or stronger methods of power tionalism; this factor was mainly responsible for the 1963
sharing. The final explanation of Brass's disagreement failure of the Cypriot consociational system, for in-
with my interpretation may be that he focuses on India's stance. As indicated earlier, India's numerical Hindu
more recent democratic experience, when its consocia- majority is internally divided to such an extent that the
tional character has not been as strong as in the first two country consists of minorities only.
decades, a subject that I shall treat at greater length (2) The second major factor is the absence or pres-
later. But even in more recent decades India has re- ence of large socioeconomic differences among the

262

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2

groups of a divided society. In India, there are disparities a separate "nation," but this challenge was effectively
of this kind among regions and, hence, among linguistic removed by the 1947 partition.
groups, as well as and more important between Hindus (8) If groups are geographically concentrated, then
and Muslims. But even the latter difference is not as federalism can be used to promote group autonomy.
great as is often assumed. In a country such as India, Although India's religious groups are territorially inter-
where illiteracy is still quite high, literacy rates are good mixed, the geographical concentration of linguistic
indicators of different levels of socioeconomic develop- groups has made India's highly successful linguistic
ment. In rural areas-and India is still mainly rural- federalism possible. Hence, on balance, a positive rating
there is very little difference in the literacy rates of is justified.
Hindus and Muslims; in urban areas, about two-thirds of (9) Traditions of compromise and accommodation
Hindus are literate compared with one-half of Muslims foster consociationalism. The Indian National Congress
(Sharif 1993). Linguistic-regional variations in socioeco- was a movement based on consensus before it became
nomic development are mitigated by the fact that the the party of consensus in 1947. More generally, too, as
poorer Hindi-speaking areas have historically exercised Austin (1966, 315) writes, "consensus has deep roots in
more power in the central government than the rest of India. Village panchayats traditionally reached decisions
the country, similar to the trade-off between the eco- in this way.... Indians prefer lengthy discussions of
nomically dominant Chinese and the politically domi- problems to moving quickly to arbitrary decisions."
nant Malays in consociational Malaysia (Esman 1972, In sum, India rates favorably on seven of the nine
25). Finally, socioeconomic differences within religious conditions for power sharing, or on eight if we accept
and linguistic groups are so much larger that they Weiner's reasoning. These include the two most impor-
overshadow intergroup disparities. tant factors. Among the other consociational democra-
(3) If there are too many groups, then negotiations cies, such a favorable predisposition is matched only by
among them will be too difficult and complex. India, with Switzerland and the Netherlands. Perhaps it would have
its extremely large number of groups, including fourteen been more surprising if India had not adopted and
major languages, receives an unfavorable rating on this maintained a power-sharing system!
factor.
(4) If the groups are of roughly the same size, then
THE WEAKENING OF POWER-SHARING
there is a balance of power among them. India's division
AFTER THE LATE 1960S
into very many minorities, without any clearly predom-
inant groups, achieves such a rough balance. Indian power sharing from independence to the present
(5) If the total population is relatively small, then the can be divided into two periods: the two decades after
decision-making process is less complex (Dahl and Tufte 1947, when consociationalism was full-fledged and com-
1973, 40). Since India is the world's second most popu- plete, and the period beginning in the late 1960s, when
lous country, there appears to be no doubt that its score power sharing continued but in slightly weaker form.
on this factor should be negative. Weiner (1989, 35-36) How can we account for this shift? Generally speaking,
suggests, however, that India's success in sustaining the main reason for the decline (and sometimes failure)
democracy despite growing tensions and violence can be of power-sharing systems is an inherent deep-seated
explained, first, in terms of its federal system (essentiallytension. Political leaders have to perform a difficult
a consociational explanation, because India's linguistic balancing act between compromises with rivals and
federalism is a key element of its power-sharing system) maintaining the support of their own followers, both
and, second, in terms of the size of the country, which activists and voters. Pleasing other elites will tend to
means that much of the conflict remains localized and displease their own supporters, and vice versa, and the
does not directly endanger the central authority. Wein- search for compromise is a time-consuming task that
er's second argument is also highly plausible and sug- may lead to a degree of immobilism, which is also likely
gests that the relationship between size and the chances to discontent supporters, who expect and demand effec-
for power sharing is curvilinear instead of linear; as size tive and decisive government action.6 It is therefore
increases, conditions for power sharing worsen initially, easier for political elites to share power successfully if
but beyond a certain critical point the tendency is their followers are relatively passive and deferential, as
reversed. shown in particular by the Dutch case (Lijphart 1968,
(6) External dangers promote internal unity. The long 139-77). This also means that strong pressures from
struggle against British colonial rule was such a unifying below will increase the elites' tendencies to concentrate
factor in India, as was the 1962 war with China. The wars and centralize power rather than to share it.
with Pakistan had the potential of inflaming internal The weakening of power sharing in India after the late
Hindu-Muslim tensions but did not produce this nega- 1960s fits this explanatory framework very well. As many
tive effect. scholars have pointed out, the 1960s marked the begin-
(7) Overarching loyalties reduce the strength of par-
ticularistic loyalties. Indian nationalism, powerfully stim- 6 Other possible causes of the decline of power sharing are the
ulated by the Indian National Congress in the period emergence of new and unforeseen problems, such as the international
crisis that can explain much of the collapse of Lebanese power sharing
before independence, has been such a unifying force
in the 1970s (Lijphart 1985, 91-92), and the improvement in inter-
(Khilnani 1992; Masselos 1985; Suntharalingam 1983). group relations by successful power sharing to such an extent that
The only serious challenge came from the Muslim full-fledged power sharing becomes superfluous, as in the Austrian and
League, which claimed that India's Muslims constituted Dutch cases after 1966 and 1967, respectively.

263

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 The Puzzle of Indian Democracy June 1996

ning of mounting democratic activism by previously 1986 Muslim Women (Protection of Right on Divorce)
quiet groups, especially the middle peasants (Brass 1990; Act necessarily entailed a wholesale condemnation of
Frankel 1988; Kohli 1990; Rudolph and Rudolph 1987). personal law; many critics objected mainly to the specific
The resulting pressures for more decisive and less con- provisions of the new law, calling it "a primitive anti-
sensual government action have prompted greater con- woman bill" (Iyer 1987, xvi). But the Supreme Court
centration and centralization of power, especially in the judgment in the Shah Bano case explicitly called for the
Congress Party and the federal system. Four factors elimination of separate personal laws and their replace-
contributed to this weakening. ment by a "uniform civil code," arguing in a clearly
First, under the leadership of Indira Gandhi, who anti-consociational vein that "a common civil code will
became prime minister in 1966 (after the brief interreg- help the cause of national integration by removing
num of Lal Bahadur Shastri, who succeeded Nehru after disparate loyalties to laws which have conflicting ideol-
his death in 1964), the Congress Party was transformed ogies" (cited in Engineer 1987, 33). The reversal of the
from an internally democratic, federal, and consensual court's decision gave new ammunition to the foes of
organization to a centralized and hierarchical party. separate personal laws.
According to Varshney (1993a, 243), "Nehru had used In an examination of the claim that minorities enjoy
his charisma to promote intraparty democracy, not to more rights than the Hindu majority, Sharma (1993, 102,
undermine it, strengthening the organization in the 106) argues that it is valid as far as the minorities'
process. Indira Gandhi used her charisma to make the educational autonomy is concerned: Their schools are
party utterly dependent on her, suspending intraparty not "subject to governmental control in the way similar
democracy and debate, and weakening the organization institutions run by the majority community are. The
as a result." In very similar terms, Das Gupta (1989, 71) minorities in this respect do in fact enjoy rights not
describes the new Congress Party as "less a national available to the majority community." He concludes that
institution of interest reconciliation than a central orga- "this in effect means that the majority community sub-
nization for mobilizing endorsement for the leadership sidizes the educational system of the minority commu-
and its hierarchical apparatus." It has remained a nities." Sharma captures the growing criticism of minor-
broadly inclusive party, but less by means of representa- ity educational autonomy very well, including the
tion from the bottom up than by representativeness from tendentious argument that it is the "majority," instead of
the top down. society as a whole, that does the subsidizing. One way to
Second, the federal system, never highly decentral- solve the problem would be to make educational auton-
ized, was centralized even more. One instrument was the omy available to any group, regardless of its majority or
increasingly frequent use of so-called President's Rule minority status and regardless of whether it is a religious,
for partisan purposes. The founding fathers had given linguistic, or any other kind of group, such as a group of
the central government the right to dismiss state govern- people espousing a particular educational philosophy
ments and to replace them with direct rule from the like Montessori. Instead of such an improvement of the
center for the purpose of dealing with grave emergen- system along consociational lines, as in the Netherlands,
cies, not foreseeing that the central government "would for instance, the prevailing tendency among the critics is
resort to devices intended to safeguard unity and cohe- the anti-consociational one of abolition.
sion for undermining democratically elected [state] gov- The Bharatiya Janata Party (BJP) has made itself the
ernments and seeking to diminish their role and impor- main mouthpiece against the government's alleged pan-
tance" (Arora and Mukarji 1992, 8). President's Rule dering to minorities, what its leader L. K. Advani calls
was invoked ten times before the end of 1967 but 66 "minorityism" (Varshney 1993a, 252). The BJP, usually
times in the only slightly longer period from 1968 to described as a "Hindu nationalist party," is clearly anti-
early 1989 (Kathuria 1990, 339). Like the centralization consociational, and its growing strength represents a
of the Congress Party, the similar trend in the federal major potential danger to power sharing in India.7 The
system is often attributed to Indira Gandhi. It would be 1991 state elections brought the BJP to power in India's
wrong, however, to interpret these trends primarily in largest state, Uttar Pradesh, with one-sixth of the coun-
terms of the-admittedly starkly contrasting-leader- try's population, as well as in Madhya Pradesh, Rajast-
ship propensities of Nehru and his daughter. For one han, and Himachal Pradesh. After the imposition of
thing, they can be explained more convincingly in terms President's Rule and new elections in November 1993,
of the structural tensions inherent in power sharing. For the BJP retained control only in Rajasthan, but it also
another, Indira Gandhi's two main successors reverted won the election in the union territory (and capital city)
to a less confrontational and more consensual style of of Delhi. In the February 1995 state elections, it ex-
leadership (Rajiv Gandhi intermittently and P. V. Nara- tended its influence from the northern Hindi-speaking
simha Rao more consistently) without, however, undo- heartland to the western part of the country by winning
ing either the party's or the federation's centralization. elections in Gujarat and, allied with the Hindu funda-
The third source of weakness is that the pressures mentalist party Shiv Sena, in Maharashtra.
from below have specifically included calls for the abo- The fourth and final source of weakness derives from
lition of crucial consociational rules put in place by a combination of the inherent tensions of power sharing
power-sharing compromises: separate personal laws, mi- and the unique Indian form of grand coalition, based on
nority educational autonomy, and Kashmir's constitu-
tionally privileged (although no longer actually imple- 7 The BJP also can be called a majority-control party in the sense that
mented) autonomous status. Not all the criticism of the Lustick (1979) uses the term control; see note 2.

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2

the predominance of a broadly representative party. All noted by scholars not explicitly belonging to the conso-
the pressures from below make it especially difficult to ciational school. For instance, Weiner (1989, 11) writes
maintain broad support for a party explicitly committed that "conflict management has become more difficult
to power sharing and minority rights. The Congress with the decline of the Congress party organization and
Party has never won a majority of the popular vote, and the weakening of the federal structure." Varshney
in 1967 its plurality fell to only slightly more than 40%. (1993b, 17-18) finds it "not surprising that the attempt
It lost the 1977 and 1989 elections outright, and because by the post-Nehru leadership of the Congress party to
it gained a mere plurality of seats in 1991 it could only centralize an essentially diverse and federal polity has
form a minority cabinet. In fact, the 1989 and 1991 co-existed with some of the worst stresses that the polity
election results show that India has shifted from a has experienced, including the insurgencies} in Punjab
dominant- to a multiparty system. The shift in the and Kashmir."
effective number of parliamentary parties-the number
of parties in parliament weighted by their size (Taage-
A RETURN TO FULL-FLEDGED
pera and Shugart 1989, 77-91)-is instructive in this
POWER SHARING?
respect: The eight elections from 1951 to 1984 yielded
eight manufactured majorities (seat majorities won with- A final piece of evidence about the close fit between the
out vote majorities) and an average effective number of Indian case and consociational theory is provided by the
2.2 parties, typical of either a two-party or dominant- proposals for political and constitutional reform. If the
party system; the elections in 1989 and 1991 failed to consociational interpretation of India's democracy is
produce a majority party, and the average effective correct, that is, if the survival of Indian democracy can
number of parties increased to 3.8, clearly a multiparty be explained by its power-sharing character and if its
system. increasing turbulence after the 1960s can be explained in
These weaknesses do not signify that power sharing terms of the weakening of power sharing, we should
has ended or is ending in India. Congress Party cabinets expect these proposals to have two characteristics. First,
have continued to be broadly representative, and non- all or most of them should be aimed at strengthening the
Congress cabinets have been only marginally less so consociational aspects of the political system. Second,
during their two brief periods in power. Federalism has given the growth in intergroup tensions and violence and
weakened but is far from dead, and the principle of the growing opposition to the very principle of power
linguistic federalism is very much alive. Minority educa- sharing, they can be expected to call for far-reaching
tional autonomy and separate personal laws are under reform with a sense of urgency.
attack, but they have so far survived, along with the Both expectations are correct. Although there is no
minority veto and the proportionality principle. vigorous public debate about or widespread demand for
The above description of continued, although weak- political change, Indians who do call for reform have in
ened, power sharing in India fits consociational theory mind drastic measures, indeed. For instance, Abid Hus-
in two other respects. The theory states that power sain (1993, 11) asserts that India's "deformed polity" is
sharing is a necessary condition for the survival of "in need of drastic surgery." In a volume entitled
democracy in divided societies; indeed, Harrison's Reforming the Constitution, others have called for "fun-
(1960, 338) dire prediction of India's democratic failure, damental changes" (Reddy 1992) and "major amend-
quoted at the beginning of this article, is not shared by ments" (Vira 1992) to the constitution, or even for the
any knowledgeable observer of Indian politics today (see election of a new constituent assembly (Malaviya 1992)
especially Varshney 1995). At the same time, while that should draft an entirely "new constitution" (Rao
Indian democracy is quite stable in this fundamental 1992) as the foundation for a "Second Republic" (Jais-
sense, the weakening of power sharing should be ex- ingh 1992). Significantly, the substantive thrust of all but
pected to be accompanied by increases in intergroup one of the major reform plans is in the direction of
tensions and violence, which clearly has been the case in stronger power sharing. The one exception, which is only
India. The official figures, which tend to be on the a partial exception, is the frequently voiced suggestion
conservative side, on Hindu-Muslim violence in the that India should adopt an American-style or French-
1954-85 period presented in P. R. Rajgopal's (1987, style presidential system (Pathak 1993; Rao 1992; Sathe
16-17) study show an alarming trend. When the first five 1991, 37-38; Trehan 1993; see also Noorani 1989). From
years (1954-58) are compared with the last five years the consociational perspective, the problem with presi-
(1981-85), the number of violent incidents rose from dentialism is its concentration of executive power in the
339 to 2,290, the number of persons killed from 112 to hands of one person, who, in a divided society, is
2,350, and the number of persons injured from 2,229 to inevitably a member of one particular group; power
17,791. This trend, Rajgopal observes, "shows no signs sharing requires joint rule by the representatives of all
of being reversed." Indeed, in the aftermath of the major groups in a collegial decision-making body, ideally
destruction of the mosque at Ayodhya in December provided by cabinets in parliamentary systems. The most
1992, rioting in many parts of India led to about 1,200 prominent and detailed presidentialist proposal for In-
deaths in one month, and more than 600 people were dia, however, put forward by B. K. Nehru, explicitly
killed in anti-Muslim rioting in Bombay in January 1993 recognizes this disadvantage and tries to compensate for
(Hardgrave 1993, 64-65). it by recommending a special form of presidentialism,
The causal link between the weakening of power used in Nigeria and also recommended by Horowitz
sharing and these problems of governance has also been (1985, 635-38) for ethnically heterogeneous societies

265

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 The Puzzle of Indian Democracy June 1996

elsewhere. Nehru's (1992, 138) proposal is to "divide the larger than a bare majority will be formed. In the case of
country into four zones- east, west, south, and north- India, even a narrow coalition of parties elected by PR is
and require a successful candidate for the Presidency not likely to be based on at least a popular majority, which
only to get an overall majority of the votes cast through- means that it would be more broadly based than any
out the country but also a specific, relatively small, Indian cabinet so far.
percentage of votes in all the zones, before he can be One reform that PR almost certainly would preclude
declared elected." This would ensure that the winning is a return to the "Congress system," which Kothari
candidate has at least a minimum of support in regions (1989, 304-6) appears to favor and which, it should be
other than his or her own.8 noted, is also consociational in orientation, with either
The other major reform proposals, entailing the the Congress Party itself or another party becoming the
strengthening of the federal system and the adoption of new party of consensus. Without a majority-manufactur-
a proportional representation (PR) election system, are ing electoral system, it would be difficult for such a party
all fully consonant with power sharing. There appears to to develop. But it is unlikely anyway that a new party of
be almost universal agreement that India's federal sys- consensus could form without the advantage of the
tem should be decentralized; this is the tenor of the 1988 unique historical circumstances of 1947, when the ruling
report of the Sarkaria Commission on Union-State party emerged from an enormously effective and suc-
Relations, which Mukarji and Mathew (1992, 280) call cessful national liberation movement, and without Jawa-
"conservative but constructive," since they and other harlal Nehru's unusually high quality of leadership.
reformers would prefer to go much farther. An espe- Moreover, instead of helping the moderate and centrist
cially interesting proposal by Mukarji and Arora (1992, Congress Party, first-past-the-post might well bring an
270) is to establish a three-level federalism, with each
anti-consociational party like the BJP to power with a
state becoming a federation, or even a more radical
manufactured majority.
multilevel federalism. They call such a system a "cascad-
ing federalism: a federation of federations." One reason
this kind of reform is so attractive is that the Indian
states are inordinately large; not counting the seven CONCLUSION
union territories, the average population of the 25 states
The big puzzle of Indian democracy-its survival despite
is about 35 million, larger than California, the most
the country's deep ethnic and communal divisions-is
populous state in the United States.
solved by the consociational interpretation presented in
Another and more straightforward solution to the
this article. India has had a power-sharing system of
problem of unwieldy state size would be to increase the
democracy during its almost fifty years of independence,
number of states. Kothari (1976, 81) suggests about 40,
and an especially full and thorough form of it during its
and a detailed proposal by Khan (1992, 108-22) speci-
first two decades, displaying all four of the essential
fies 58, 6 of which would be carved out of the huge state
elements of power sharing as clearly as Austria, the
of Uttar Pradesh, with a population of almost 150
Netherlands, Switzerland, Lebanon, Malaysia, and the
million. Similarly, Kashyap (1992, 32-33) recommends
other well-known examples of consociational democ-
the creation of "50 to 60 States of almost equal size." A
racy. That newly independent India embraced power
considerable increase in the number of states also offers
sharing and has maintained it ever since is not even very
an opportunity for further fine-tuning of linguistic ho-
surprising, because most of the conditions found to be
mogeneity.
conducive to it in these other countries are also favor-
Finally, many reformers have proposed the adoption
able in the Indian case. After the late 1960s, as a result
of PR for parliamentary elections (Bhambhri 1971;
Nehru 1992). The German system, which combines of greater mass mobilization and activation, power shar-
first-past-the-post elections for half the parliamentary ing became less strong and pervasive, evidenced by the
seats with overall proportionality for all seats by means centralization of the Congress Party and the federal
of list PR, is the most frequently mentioned specific system, the decline of the Congress Party's electoral
suggestion (Hegde 1986, 107; Seth 1971; Singh 1986, strength, the attack on minority rights, and the rise of the
120; see also Vanhanen 1987). PR is based on the BJP. As consociational theory would have predicted,
consociational principle of proportionality, and, as com- Indian democracy has remained basically stable, but the
parative studies of democratic systems show, it is con- weakening of power sharing has been accompanied by
ducive to multiparty systems, which in turn are condu- an increase in intergroup hostility and violence. Concern
cive to broad multiparty cabinets (Lijphart 1984), about these trends is reflected in the consociational
although there is no guarantee, of course, that coalitions thrust of the major proposals for political and constitu-
tional change by reform-minded Indians.
The consociational interpretation of India strengthens
8 A serious drawback of the Nigerian system, used in 1979 and 1983 (in
our understanding of the Indian case by providing a
which the winner needs a nationwide plurality plus at least 25% of the
vote in no fewer than two-thirds of the states) is that it can easily result theoretically coherent explanation of the main patterns
in none of the candidates being elected. This is not a problem in and trends in its political development. Furthermore,
Nehru's (1992, 137) plan because he proposes the indirect election of it strengthens consociational theory by removing the
the president-by an electoral college of national, state, and local
one allegedly deviant case and by showing that, instead,
legislators-in which repeated ballots can be conducted until a winner
emerges. Of course, consociationalists would still prefer a broadly the crucial case of India is unmistakably a confirming
representative collegial executive to a broadly supported presidency. case.

266

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 American Political Science Review Vol. 90, No. 2

REFERENCES Graziano, Luigi. 1980. "The Historic Compromise and Consociational

Democracy: Toward a 'New Democracy'?" International Political
Arora, Balveer, and Nirmal Mukarji. 1992. "Introduction: The Basic Science Review 1 (3): 345-68.
Issues." In Federalism in India: Origins and Development, ed. Nirmal Halpern, Sue M. 1986. "The Disorderly Universe of Consociational
Mukarji and Balveer Arora. New Delhi: Vikas. Democracy." West European Politics 9 (April): 181-97.
Austin, Granville. 1966. The Indian Constitution: Cornerstone of a Hardgrave, Robert L., Jr. 1993. "India: The Dilemmas of Diversity."
Nation. Bombay: Oxford University Press. Journal of Democracy 4 (October): 54-68.
Banerjee, Ashis. 1992. "Federalism and Nationalism: An Attempt at Harrison, Selig S. 1960. India: The Most Dangerous Decades. Princeton:
Historical Interpretation." In Federalism in India: Origins and De- Princeton University Press.
velopment, ed. Nirmal Mukarji and Balveer Arora. New Delhi: Hartlyn, Jonathan. 1988. The Politics of Coalition Rule in Colombia.
Vikas. Cambridge: Cambridge University Press.
Barry, Brian. 1975. "Political Accommodation and Consociational Hegde, Ramakrishna. 1986. Electoral Reforms: Lack of Political Will.
Democracy." British Journal of Political Science 5 (October): 477- Bangalore: Karnataka State Janata Party.
505.
Hix, Simon. 1994. "Approaches to the Study of the EC: The Challenge
Batliner, Gerard. 1981. Zur heutigen Lage des liechtensteinischen Par- to Comparative Politics." West European Politics 17 (January): 1-30.
laments. Vaduz: Verlag der Liechtensteinischen Akademischen Ge- Horowitz, Donald L. 1985. Ethnic Groups in Conflict. Berkeley:
sellschaft.
University of California Press.
Berg-Schlosser, Dirk. 1985. "Elements of Consociational Democracy Hughes, Arnold. 1982. "The Limits of 'Consociational Democracy' in
in Kenya." European Journal of Political Research 13 (March):
the Gambia." Civilisations 32 (2): 65-92.
95-109.
Huntington, Samuel P. 1988. "One Soul at a Time: Political Science
Bhambhri, C. P. 1971. "Electoral Reform and Party System in India: A
and Political Reform." American Political Science Review 82
Plea for Proportional Representation." In Elections and Electoral
(March): 3-10.
Reform in India, ed. Subhash C. Kashyap. New Delhi: Institute for
Hussain, Abid. 1993. India: Challenges and Changes. New Delhi: Rajiv
Constitutional and Parliamentary Studies.
Gandhi Institute for Contemporary Studies.
Brass, Paul R. 1990. The Politics of India Since Independence. Cam-
Huyse, Luc. 1987. De verzuiling voorbij. Louvain: Kritak.
bridge: Cambridge University Press.
Iyer, V. R. Krishna. 1987. The Muslim Women (Protection of Rights on
Brass, Paul R. 1991. Ethnicity and Nationalism: Theory and Compari-
Divorce) Act, 1986. Lucknow: Eastern Book Company.
son. New Delhi: Sage.
Jaisingh, Hari. 1992. "Time for a Second Republic." In Reforming the
Cannon, Gordon E. 1982. "Consociationalism vs. Control: Canada as
Constitution, ed. Subhash C. Kashyap. New Delhi: UBSPD.
a Case Study." Western Political Quarterly 35 (March): 50-64.
Kashyap, Subhash C. 1992. "Constitution, Its Working and Need for
Chehabi, H. E. 1980. "The Absence of Consociationalism in Sri
Reforms." In Reforming the Constitution, ed. Subhash C. Kashyap.
Lanka." Plural Societies 11 (Winter): 55-65.
New Delhi: UBSPD.
Chinwuba, Felix Aneze. 1980. "Consociationalism as an Approach to
Kathuria, Harbir Singh. 1990. President's Rule in India, 1967-89. New
Political Integration: The Case of the Federal Republic of Nigeria."
Delhi: Uppal.
Ph.D. diss., Tulane University.
Khan, Rasheeduddin. 1992. Federal India: A Design for Change. New
Chryssochoou, Dimitris N. 1994. "Democracy and Symbiosis in the
Delhi: Vikas.
European Union: Towards a Confederal Consociation?" West Eu-
Khilnani, Sunil. 1992. "India's Democratic Career." In Democracy: The
ropean Politics 17 (October): 1-14.
Unfinished Journey, 508 BC to AD 1993, ed. John Dunn. Oxford:
Daalder, Hans. 1974. "The Consociational Democracy Theme." World
Oxford University Press.
Politics 26 (July): 604-21.
Kohli, Atul. 1990. Democracy and Discontent: India's Growing Crisis of
Daalder, Hans, and Galen A. Irwin, eds. 1989. Special Issue on
Governability. Cambridge: Cambridge University Press.
"Politics in the Netherlands: How Much Change?" West European
Kothari, Rajni. 1970. Politics in India. Boston: Little Brown.
Politics 12 (January): 1-185.
Dahl, Robert A. 1989. Democracy and Its Critics. New Haven: Yale
Kothari, Rajni. 1976. Democratic Polity and Social Change in India:
Crisis and Opportunities. Bombay: Allied.
University Press.
Dahl, Robert A., and Edward R. Tufte. 1973. Size and Democracy. Kothari, Rajni. 1989. Politics and the People: In Search of a Humane
Stanford: Stanford University Press. India. Delhi: Ajanta.
Das Gupta, Jyotirindra. 1989. "India: Democratic Becoming and Lehmbruch, Gerhard. 1993. "Consociational Democracy and Corpo-
Combined Development." In Democracy in Developing Countries: ratism in Switzerland." Publius 23 (Spring): 43-60.
Asia, ed. Larry Diamond, Juan J. Linz, and Seymour Martin Lipset. Levine, Daniel H. 1973. Conflict and Political Change in Venezuela.
Boulder: Lynne Rienner. Princeton: Princeton University Press.
Dekmejian, Richard Hrair. 1978. "Consociational Democracy in Crisis: Lijphart, Arend. 1968. The Politics of Accommodation: Pluralism and
The Case of Lebanon." Comparative Politics 10 (January): 251-65. Democracy in the Netherlands. Berkeley: University of California
Dew, Edward M. 1994. The Trouble in Suriname, 1975-1993. Westport, Press.

CT: Praeger. Lijphart, Arend. 1977. Democracy in Plural Societies: A Comparative

Dix, Robert H. 1980. "Consociational Democracy: The Case of Exploration. New Haven: Yale University Press.
Colombia." Comparative Politics 12 (April): 303-21. Lijphart, Arend. 1979. "Consociation and Federation: Conceptual and
Engineer, Asghar Ali, ed. 1987. The Shah Bano Controversy. Hyder- Empirical Links." Canadian Journal of Political Science 12 (Septem-
abad: Orient Longman. ber): 499-515.
Esman, Milton J. 1972. Administration and Development in Malaysia: Lijphart, Arend. 1984. Democracies: Majoritarian and Consensus Pat-
Institution-Building and Reform in a Plural Society. Ithaca: Cornell terns of Government in Twenty-One Countries. New Haven: Yale
University Press. University Press.
Frankel, Francine R. 1988. "Middle Classes and Castes in India's Lijphart, Arend. 1985. Power-Sharing in South Africa. Berkeley: Insti-
Politics: Prospects for Political Accommodation." In India's Democ- tute of International Studies, University of California.
Lijphart, Arend. 1992. "Democratization and Constitutional Choices
racy: An Analysis of Changing State-Society Relations, ed. Atul Kohli.
Princeton: Princeton University Press. in Czecho-Slovakia, Hungary, and Poland," Journal of Theoretical
Fyzee, Asaf A. A. 1964. Outlines of Muhammadan Law. 3d ed. London: Politics 4 (April): 207-33.
Oxford University Press. Lijphart, Arend. 1994. "Prospects for Power-Sharing in the New South
Gabel, Matthew J. 1994. "Balancing Democracy and Stability: Consid- Africa." In Election '94 South Africa: The Campaigns, Results and
ering the Democratic Deficit in the EU from a Consociational Future Prospects, ed. Andrew Reynolds. New York: St. Martin's
Perspective." Presented at the Joint Sessions of Workshops of the Press.
European Consortium for Political Research, Madrid. Lindberg, Leon N. 1974. "The Political System of the European
Goldman, Joseph Richard. 1985. "Consociational Authoritarian Poli- Community." In Politics in Europe: Structures and Processes in Some
tics and the 1974 Yugoslav Constitution: A Preliminary Note." East Postindustrial Democracies, ed. Martin 0. Heisler. New York: David
European Quarterly 19 (June): 241-49. McKay.

267

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms
 The Puzzle of Indian Democracy June 1996

Linder, Wolf. 1994. Swiss Democracy: Possible Solutions to Conflict in Seth, J. D. 1971. "Towards a New Electoral Law." In Elections and
Multicultural Societies. New York: St. Martin's Press. Electoral Reforms in India, ed. Subhash C. Kashyap. New Delhi:
Lustick, Ian. 1979. "Stability in Deeply Divided Societies." World Institute of Constitutional and Parliamentary Studies.
Politics 31 (April): 325-44. Sharif, Abu Saleh. 1993. Some Socio-Economic and Demographic
Luther, Kurt Richard, and Wolfgang Muller, eds. 1992. Special Issue Aspects of Population According to Religion in India. Bombay: Centre
on "Politics in Austria: Still a Case of Consociationalism?" West for the Study of Society and Secularism.
European Politics 15 (January): 1-226. Sharma, Arvind. 1993. "Minority vs. Majority Rights." In Perspectives
Luthera, Ved Prakash. 1964. The Concept of the Secular State and on the Constitution, ed. Subhash C. Kashyap. Delhi: Shipra.
India. London: Oxford University Press. Singh, L. P. 1986. Electoral Reform: Problems and Suggested Solutions.
Mair, Peter. 1994. "The Correlates of Consensus Democracy and the New Delhi: Uppal.
Puzzle of Dutch Politics." West European Politics 17 (October): Srinivasavaradan, T. C. A. 1992. Federal Concept: The Indian Experi-
97-123. ence. New Delhi: Allied.
Malaviya, S. P. 1992. "Case for a Constituent Assembly." In Reforming Steiner, Jurg. 1990. "Power-Sharing: Another Swiss 'Export Product'?"
the Constitution, ed. Subhash C. Kashyap. New Delhi: UBSPD. In Conflict and Peacemaking in Multiethnic Societies, ed. Joseph V.
Masselos, Jim. 1985. Indian Nationalism: An History. New Delhi: Montville. Lexington, MA: Lexington Books.
Sterling. Steiner, Jirg, and Dorff, Robert H. 1980. A Theory of Political Decision
Mehta, Piarey Lal. 1991. Constitutional Protection to Scheduled Tribes Modes: Intraparty Decision Making in Switzerland. Chapel Hill:
in India: In Retrospect and Prospects. Delhi: H. K. Publishers. University of North Carolina Press.
Messarra, Antoine Nasri. 1994. Thorie generate du systeme politique Suntharalingam, R. 1983. Indian Nationalism: An Historical Analysis.
libanais. Paris: Cariscript. New Delhi: Vikas.
Mill, John Stuart. 1958 [1861]. Considerations on Representative Gov- Taagepera, Rein, and Matthew Soberg Shugart. 1989. Seats and Votes:
ernment. New York: Liberal Arts Press. The Effects and Determinants of Electoral Systems. New Haven: Yale
Mukarji, Nirmal, and Balveer Arora. 1992. "Conclusion: Restructuring University Press.
Federal Democracy." In Federalism in India: Origins and Develop- Taylor, Rupert. 1992. "South Africa: A Consociational Path to
ment, ed. Nirmal Mukarji and Balveer Arora. New Delhi: Vikas. Peace?" Transformation 17: 1-11.
Mukarji, Nirmal, and George Mathew. 1992. "Epilogue: Federal Trehan, Virender M. 1993. "Need for Change." In Perspectives on the
Issues, 1988-1990." In Federalism in India: Origins and Development, Constitution, ed. Subhash C. Kashyap. Delhi: Shipra.
ed. Nirmal Mukarji and Balveer Arora. New Delhi: Vikas. Van den Berghe, Pierre L. 1981. The Ethnic Phenomenon. New York:
Nehru, B. K. 1986. Thoughts on Our Present Discontents. New Delhi: Elsevier.
Allied. Vanhanen, Tatu. 1987. "What Kind of Electoral System for Plural
Nehru, B. K. 1992. "A Fresh Look at the Constitution." In Reforming Societies? India as an Example." In The Logic of Multiparty Systems,
the Constitution, ed. Subhash C. Kashyap. New Delhi: UBSPD. ed. Manfred J. Holler. Dordrecht, Netherlands: Kluwer.
Noorani, A. G. 1989. The Presidential System: The Indian Debate. New Varshney, Ashutosh. 1993a. "Contested Meanings: India's National
Delhi: Sage. Identity, Hindu Nationalism, and the Politics of Anxiety." Daedalus
Olson, David M. 1994. "The Sundered State: Federalism and Parlia- 122 (Summer): 227-61.
ment in Czechoslovakia." In Parliaments in Transition: The New Varshney, Ashutosh. 1993b. "India's Democratic Exceptionalism and
Legislative Politics in the Former USSR and Eastern Europe, ed. Its Troubled Trajectory." Revised version of paper presented at the
Thomas F. Remington. Boulder: Westview Press. 1990 annual meeting of the American Political Science Association,
Pai Panandiker, V. A., and Ajay Mehra. 1996. The Indian Cabinet and San Francisco.
Governance of India. New Delhi: Konark. Varshney, Ashutosh. 1995. "The Self-Correcting Mechanisms of In-
Pathak, Bindeshwar. 1993. "Facets of the System: Presidential vs. dian Democracy." Seminar (Delhi) 425 (January): 38-41.
Parliamentary." In Perspectives on the Constitution, ed. Subhash S. Vasovic, Vucina. 1992. "A Plea for Consociational Pluralism." In The
Kashyap. Delhi: Shipra. Tragedy of Yugoslavia: The Failure of Democratic Transformation, ed.
Powell, G. Bingham, Jr. 1970. Social Fragmentation and Political Jim Seroka and Vukasin Pavlovic. Armonk, NY: M. E. Sharpe.
Hostility: An Austrian Case Study. Stanford: Stanford University Vira, Dharma. 1992. "Inadequacy and Impracticability of the Present
Press. Constitution: Case for Major Amendments." In Reforming the
Powell, G. Bingham, Jr. 1979. Book review of Arend Lijphart's Constitution, ed. Subhash C. Kashyap. New Delhi: UBSPD.
Democracy in Plural Societies. In American Political Science Review Von Vorys, Karl. 1975. Democracy Without Consensus: Communalism
73 (March): 295-97. and Political Stability in Malaysia. Princeton: Princeton University
Powell, G. Bingham, Jr. 1982. Contemporary Democracies: Participa- Press.
tion, Stability, and Violence. Cambridge: Harvard University Press. Weiner, Myron. 1969. Party Building in a New Nation: The Indian
Prasad, Anirudh. 1991. Reservation Policy and Practice in India: A National Congress. Chicago: University of Chicago Press.
Means to an End. New Delhi: Deep & Deep Publications. Weiner, Myron. 1989. The Indian Paradox: Essays in Indian Politics.
Rae, Douglas W. 1967. The Political Consequences of Electoral Laws. New Delhi: Sage.
New Haven: Yale University Press. Wheare, K. C. 1964. Federal Government. 4th ed. New York: Oxford
Rajgopal, P. R. 1987. Communal Violence in India. New Delhi: Uppal. University Press.
Rao, S. Ramachandra. 1992. "Plea for a New Constitution." In Worrall, Denis. 1981. "The Constitutional Committee of the Presi-
Reforming the Constitution, ed. Subhash C. Kashyap. New Delhi: dent's Council." Politikon 8 (December): 27-34.
UBSPD. Young, Crawford. 1976. The Politics of Cultural Pluralism. Madison:
Reddy, K. Brahmananda. 1992. "Case for Fundamental Changes." In University of Wisconsin Press.
Reforming the Constitution, ed. Subhash C. Kashyap. New Delhi: Zakaria, Haji Ahmad. 1989. "Malaysia: Quasi Democracy in a Divided
UBSPD. Society." In Democracy in Developing Countries: Asia, ed. Larry
Rudolph, Lloyd I., and Susanne Hoeber Rudolph. 1987. In Pursuit of Diamond, Juan J. Linz, and Seymour Martin Lipset. Boulder: Lynne
Lakshmi: The Political Economy of the Indian State. Chicago: Uni- Rienner.
versity of Chicago Press. Zolberg, Aristide R. 1977. "Splitting the Difference: Federalization
Sathe, Vasant. 1991. National Government: Agenda for a New India. Without Federalism in Belgium." In Ethnic Conflict in the Western
New Delhi: UBS. World, ed. Milton J. Esman. Ithaca: Cornell University Press.

268

This content downloaded from 59.92.7.196 on Sat, 08 Dec 2018 11:30:23 UTC
All use subject to https://about.jstor.org/terms