« back to table of contents

Indian Federalism and the Conduct of Foreign Policy in Border States
State Participation and Central Accommodation since 1990
By Rafiq Dossani and Srinidhi Vijaykumar

The role of subnational units (states, provinces, cantons, lander) in international affairs is a growing subject in the literature on federalist affairs (see note 1). Scholars of political science have traditionally seen the conduct of foreign policy as the exclusive domain of the national government. This would seem an especially apt observation about India’s federalist system. The Indian constitution has given the center particularly strong powers – so strong, in fact, that some have described it as “quasi federal” because of the lack of autonomy it affords to the states.2 Yet, there is an increasing consensus that the states have not been shy of foreign policy advocacy. Some have argued that the era of coalition government has increased such advocacy and, potentially, influence, especially in the context of globalization and economic reform and liberalization.3

This paper considers the role of Indian border states in the conduct of foreign policy toward their transnational neighbors and asks whether coalition governance results in more power generally or to some state actors more than others. In particular, we explore whether the effectiveness of a state’s foreign policy advocacy depends on that state’s position in the coalition. Effectiveness may also be influenced by the type of advocacy – on ethnic issues, for example, as opposed to economic ones – and by constitutional limits.

That India’s constitution grants greater power to the center than to the states has been well documented in scholarly literature.4 A few key points highlight this center-oriented position. The constitution clearly delineates powers to the states, the center, and those over which the states and center have concurrent adjudication. However, whenever there is any conflict over the laws on the Concurrent list, the national parliament laws prevail over those of the state legislative assemblies.5 In addition, according to article 248 of the constitution, all residuary powers are given to the Union, unlike in the United States, where residuary powers go to the states. In practice, this has meant that the central government controls most of the powers of taxation (an important exception is agricultural income tax, which is a state subject).6 This has led the states to negotiate revenue sharing and other financial allocations among themselves and with the center through centrally supervised bodies, such as the planning commission and the finance commission. Finally, the center appoints the governor of each state, a conventionally powerless position that nonetheless can and has been used on occasion to influence state policy and politics, such as the dismissal of state governments.

Despite the immense powers granted to the center, states are extremely important actors in the Indian political system. States retain sole or primary constitutional authority over education, agriculture, law and order, health, welfare, and local government. Central leaders may make policy on these and other concurrent subjects, but they may find that the states do not always implement the center’s recommendations. In particular, finance and planning commissions havemet with resistance from powerful state leaders.7

In a discussion of state advocacy in foreign affairs, “involvement in foreign policy” can involve different degrees of influence. For example, the central government may informally consult a border state government to understand the impact that placing troops in the area prior to offensive action would have on local opinion. The outcome of such consultations may influence the central decision about timing, location, and the like, but the state government would not be considered to have played a strongly influential role.  On the other hand, if the state government advocated against war and was found to have changed the central government’s position on the subject, then such a role would be considered strongly influential. Jenkins argues against overstating the states’ level of influence, and makes important qualifications about various scholars’ works.8 He points out that Indian states’ increasing exposure to and involvement in international affairs does not necessarily translate into autonomy in conducting foreign policy. He also argues that although states interact increasingly with multilateral institutions (such as the WTO), the result has been “to take domestic policy to an arena – the intergovernmental negotiations within the WTO – where India’s subnational authorities have difficulty gaining access.”9 In other words, Jenkins suggests that states influence foreign economic policy, but only within a limited framework and amid a complicated network of state politicians, national politicians, and bureaucrats. This paper’s argument, as discussed above, counters this thesis as it explores the ways that states have successfully influenced foreign policy.

Power Sharing in India: A Framework

Kincaid coined the phrase “constituent diplomacy” to define the international activities of subnational groups. He asserts that constituent diplomacy is on the rise, and that there has always been some state activity in international affairs. Indeed, he says that “there is an ambiguity in federal polities about the status of constituent governments in world affairs and about the authority of the general government to act unilaterally in foreign affairs.”10 Kincaid delineates a number of actions which would traditionally be thought of as belonging exclusively to the national government, but some of which, he argues, had at times involved the participation of subnational units. His list includes the authority to:

  1. Declare war
  2. Build and maintain armed forces
  3. Conduct relations with foreign nations and international organizations
  4. Appoint and receive diplomatic and consular officials
  5. Conclude, ratify, and implement treaties
  6. Assure that treaty law is supreme over any federal or constituent laws that conflict with it
  7. Control entry and exit across national borders
  8. Acquire or cede territory11

Indian states are constitutionally prohibited from engaging in the first through sixth items on the list, as well as the eighth. And yet, there are uncertainties – areas where the center would prefer state participation, and areas where the state is pushing to participate despite center unwillingness. The authority to “conduct relations with foreign nations and international organizations” (item three), for one, has become increasingly ambiguous in recent years. For example, the center encourages the states to negotiate energy agreements with foreign companies. A recent and oft-cited example of this is Maharashtra’s power purchase agreement with Enron Corporation of the United States. Even more interesting is item seven, which concerns border security. The Indian constitution does not specify who has the authority to control entry to and exit from the nation, beyond empowering the center with authority over all foreign relations matters. As state parties have increased their voice in government, they have demanded an increased role in controlling migration into their states. Nowhere is the ambiguity of federal jurisdiction more apparent than in the case of border patrol. In West Bengal, the state police and Border Security Forces (a centrally-controlled paramilitary group) often work together to control the movement of migrants and goods from Bangladesh. The final authority over controlling the border is extremely unclear: the states have the authority to maintain law and order and to provide ration cards (a key form of identity) to each “citizen,” but the central government has the (far stronger) authority to conduct relations with Bangladesh and protect national security.

Sridharan argues that states have gained influence in foreign affairs advocacy because of two trends which have devolved more power to the state: the increasing regionalization of Indian politics, and economic liberalization. Through economic liberalization, there is an environment in which states can play a key role in foreign economic policy, by seeking foreign direct investments and promoting foreign trade. For example, states participate in negotiations with foreign investors and attempt to influence those with the WTO. Indian economic reform, Sridharan posits, has aligned politics more closely with economic welfare. Combined with globalization, this gives state-level politicians an interest in accessing global economic opportunities. Key examples of state interaction with external agencies are in energy reform and agriculture negotiations.12

In addition, Sridharan argues that coalition governance gives regional parties the opportunity to influence the formulation of foreign policy, and to advance policies of specific interest to their states. The trend toward regionalization was marked by the debacle of the Congress party in the 1996 election.13 What has emerged is a multiple party system, in which regional and local parties play a large role. The age of coalitions at the center,14 which began with the formation of the BJP-led National Democratic Alliance (NDA) in 1998, has allowed state-focused rather than nation-focused parties to access a share of national power, which they can then use to further foreign policies that help their own states. In the May 2004 elections, voters elected the Congress Party to power, showing that coalitions were not a preserve of non-Congress parties. The Congress formed a coalition called the United Progressive Alliance, which consisted of parties such as the Dravida Munnetra Kazhagam (DMK), the People’s Democratic Party (Kashmir), Rashtriya Janata Dal (Bihar),and was supported from the outside by the left parties. Sridharan says that the increase in state advocacy “has occurred at a time when no single national party is able to win a majority without traversing the high road to a state capital in order to forge an electoral alliance. The space for the regional parties to assert themselves at the national level opened up as the one-party dominant system came to an end in Indian politics.”15

Sridharan’s argument suggests that increased state participation in foreign policy ought to be a general practice across states. This paper addresses the question of whether coalition governance truly opens a space for the participation of all regional parties, or if it only gives certain sets of states the scope to engage in extra-constitutional activities (ECAs). It may be true that all states undertake ECAs because coalition governance reflects a weakness on the part of the center, or, as Sridharan puts it, because the center cannot assume state cooperation as a given. Alternatively, it may be that all states do not engage in more ECAs when there is a coalition government because states whose ruling factions are part of the national coalition prefer not to destabilize the center through such actions. States outside the coalition would be subject to harsh repercussions (for example, President’s Rule) if they undertook the same activities. Another possibility is that only states within the coalition actually engage in ECAs because they are not afraid of central backlash, given their ability to threaten the stability of the government. And yet another possibility is that states outside of the coalition actually undertake more ECAs because they have no voice in the coalition government and are therefore are less likely to get what they want in any case.

The possible outcomes are summarized as the following four hypotheses.

H1a. State parties undertake more ECAs when part of the central coalition.
H2a. States parties undertake more ECAs when they are not part of the central coalition.
H1b. States parties undertake fewer ECAs when part of the national coalition.
H2b. State parties undertake fewer ECAs when not part of the national coalition.

Sridharan’s work suggests that H1a and H2a ought to hold true and, therefore, that H1b and H2b do not hold true. As argued below, it is also possible for H2a and H1b to hold true while H1a and H2b do not hold true.
Table 1 presents a framework of Indian states’ new foreign policy activities since 1990, the year that marked the start of the decade of economic reform and globalization in India. We separate activities by constitutionality, and by two types that exemplify Kincaid’s items three and seven above, viz., cross-border economic activities and cross-border control of co-ethnic populations, respectively. In most cases, control of co-ethnic populations has had implications for India’s national security. India shares a border with Pakistan in Kashmir, Punjab, Rajasthan and Gujarat in the north and west; with China and Nepal in the north-central states of Uttar Pradesh and Bihar; with Bangladesh in the eastern state of West Bengal and the small northeastern states to the east of Bangladesh; and with Sri Lanka in Tamil Nadu in the south (a littoral border). Due to commonalities in ethnicity, religion, and stages of economic development, only some border states have an incentive to undertake foreign policy activity, notably Kashmir, West Bengal, Tamil Nadu, and the northeastern states.

Table 1: States’ border-related foreign policy activities since 1990

Constitutional, economic

Constitutional, co-ethnic



State’s ruling party is part of central coalition

- Water treaties, West Bengal 1996 16
- Cross-border trade policy, West Bengal, 1999–2003; Northeastern States and Sikkim, 2000 17

- Intervention in Sri Lanka, Tamil Nadu, 2000 18

State’s ruling party is outside central coalition

- Cross-border bridges, roads in NE states, 2003, West Bengal, 2002–04 19

- Joint border patrol, West Bengal, 2004 (see below).
- Release of suspected militants in Kashmir, 2002 (see below).
- Interference with border security forces, Kashmir 2002–03, West Bengal 2004 (see below).
- Issuance of identity cards to those crossing the border, West Bengal up to 2003, (see below).
- Influence on deportation policy, West Bengal 1998 (see below).

As Table 1 shows, given constitutional limits, there are few economic or co-ethnic focused cross-border activities that states can engage in that are actually constitutional. However, this has not stopped state parties from involving themselves in these policies. The table shows that the majority of states’ increased advocacy in foreign policy has occurred in co-ethnic matters by states whose ruling party is outside the central coalition. State parties within the coalition, although they have engaged in extra-constitutional economic policy, are reluctant to undertake ECAs that influence co-ethnic affairs. For example, the ruling parties in Tamil Nadu and Kashmir between 1998 and 2002 belonged to the ruling coalition at the center during a time of considerable turmoil in Sri Lanka and Kashmir, but these did not undertake ECAs to regulate co-ethnic matters.

We use case studies of West Bengal and Kashmir during the reign of the National Democratic Alliance at the center (1998–2004) to analyze the hypotheses regarding the relationship between a state party’s position in the coalition and its ECA. During this time, the ruling party in West Bengal, the Communist Party of India (Marxist) CPI(M), was in opposition at the center and ruled in the state. In Kashmir, the National Conference was in power at the state level until 2002 and was a party in the ruling coalition at the center. From 2002 and until the national elections in 2004, the People’s Democratic Party (PDP)-Congress coalition ruled in Kashmir and was in opposition at the center.

Border Policy on the East: The Bengali Connection

The policies of West Bengal have been shaped by the state’s unique geography, sandwiched between the rest of India and Bangladesh. Bangladesh was formed in 1971 after it fought a war of partition with Pakistan. Although initially friendly, relations between India and Bangladesh worsened in the mid-1970s because of domestic opposition to Bangladeshi policy, which was perceived as being over-dependent on India.20 Bangladesh then went through a series of coups and military governments, which increased anti-Indian sentiment.

The two countries’ concerns reflect their different geopolitical aspirations. As Kathryn Jaques puts it, India and Bangladesh have a “fundamental difference in the foci of their foreign policies. Indian concerns are typified by the broader fear of Bangladesh’s potential to produce destabilizing conditions in the subcontinent which, in the long term, could invite external meddling and perhaps, ultimately, the disintegration of the Indian Union. …By contrast, Bangladeshi foreign policy has been moulded by the fear of India’s regional hegemonistic designs.”21 In describing India’s security interests in Bangladesh, Padmaja Murthy has argued that the four principal Indian concerns are:

  1. The porous Indo-Bangladeshi border, which allows insurgents in Northeastern Indian states to take refuge in Bangladesh.
  2. The massive illegal immigration that occurs through the border, which has implications for the politics of West Bengal—that is, new immigrants have tended to support the ruling state party, the CPI(M).
  3. The pro-Pakistani elements holding influential positions in the Bangladeshi government, a concern given India’s volatile relations with Pakistan.
  4. The strong political and defense links that Bangladesh has with China and Pakistan.22

It is worth noting that two of these issues specifically concern border policy, and by extension, regional and internal Indian stability.

West Bengal has a complex position in the India-Bangladesh equation. West Bengal is the only state to have been ruled by the Communist Parties for over three decades (Kerala is the other state with significant communist influence). It shares nearly a 2,200 km border with Bangladesh.

Some believe that West Bengal’s government has no role whatsoever in border issues such as immigration and deportation, and that its responsibilities begin only after migrants cross over the border.23 Others think that increasing joint operations between state and center are essential for security, and that the state government should bear a large part of the burden of dealing with cross border issues.24 This next section of this paper examines the ways the state has attempted to influence policy, and argues that, even in areas where the center disapproves of a state’s actions, the state has still successfully pursued its own policy – an indication of the beginning stages of autonomy. These trends were visible when the CPI(M) was in opposition to the central coalition during the NDA years, and shows that coalition governance affords opportunities to opposition states parties to pursue more autonomous policies.

Illegal Immigration

The population density and poverty of Bangladesh, combined with its poor treatment of Hindus since the rise of the Bangladesh Nationalist Party in 1978, have created immigration problems of epic proportions. Since the formation of Bangladesh in 1972, approximately 15 million illegal immigrants have entered India from Bangladesh, both Muslim and Hindu. Of those, over five million reside in West Bengal.25  According to one Indian politician, “The six border districts of West Bengal, viz. South 24 Parganas, North 24 Parganas, Nadia, Murshidabad, Malda, and West Disnajpur; four districts of Bihar…; and ten districts in Assam… have become ‘extensions of Bangladesh.’”26 The massive influx of immigrants brings with it the danger of separatist tendencies. The Bangladesh Liberation Organization has already called for a separate homeland for the non-Muslim minority in Bangladesh and those forced to migrate to India.27 To date, there has been no significant record of violent action in support of this idea in West Bengal. Under the NDA, allegations of increasing Pakistani-inspired terrorist infiltration into West Bengal from Bangladesh have been common. India’s past problems in managing insurgency in the northeast are also alleged to have increased, due to easy sanctuary that insurgents have found in Bangladesh, via the West Bengal border.

According to the Foreigners Act of 1947, the center has the right to control the movements of foreigners, to “make provision, either generally or with respect to all foreigners … for prohibiting, regulating, and restricting the entry of foreigners into India, or their departure therefrom or their presence or continued presence therein.”28 Although the Foreigners Act requires the cooperation of state officials and police, the power of regulation clearly lies with the center. And yet, on numerous occasions, the center has wanted to hand the matter off entirely to the states. Thus, in 2003 then-deputy Prime Minister (and Home Minister) L.K. Advani, in a reference to immigration via Bangladesh, noted that “Over 11,500 Pakistanis have entered the country with regular papers and passports but have overstayed. There is no reason why our states should be soft on them. Immediate steps should be taken to identify them, locate them, and throw them out. They [the states] should launch drives to detect and deport these foreigners.” Advani further noted that the center’s powers to detect and deport illegally residing foreign nationals had been delegated to states and Union Territories.29 The center has obviously encouraged states to play a larger role in immigration policy despite its own strong mandate to do so. In many of the principal efforts taken to control illegal immigration – border patrol, fencing, deportation, and the issuance of national identity cards – the state has only an uncertain constitutional mandate to play a role. With respect to border patrol, according to Item 1 on the State List of the Indian Constitution, states are responsible for all matters of public order and policing. But according to Item 10 on the Union List, the center is responsible for all matters relating to foreign affairs. The Foreigners Act clearly gives the center authority over deportation, and there is little constitutional basis for the state to be at all involved in the multi-purpose national ID card system. And yet, the center has encouraged states to participate in all these issues.

Such central encouragement for state participation in immigration policy came at a time when the center was led by the NDA. Historically under the Congress party – but even more so under the NDA – tension has arisen between the center and West Bengal over whether the state government allows illegal immigration to increase its voter bank. Critics of the CPI(M) say that officials turn a blind eye to illegal immigration because those who enter tend to vote communist. According to critics of the BJP, such statements often have fascist overtones, and they further observe that the NDA makes these claims in order to stir communal tension—the implication being that most of those who enter illegally are Muslim, whereas in fact, the larger number are Hindus. Given this background, it is all the more surprising, (and likewise indicative of a weakening at the center) that the national government has allowed the state government scope to pursue its own border policies.

Border Patrol

The Border Security Forces (BSF), a central paramilitary organization, plays a large role in border patrol, principally in West Bengal, Kashmir, and the Northeastern states. Founded in 1965, it is responsible for protecting India’s land border during peacetime and for preventing transborder crimes, and it has an extensive intelligence network.30 The BSF is called a paramilitary force because it occupies a hazy area between being part of a police force and being a professional military organization. Since the 1990s, when insurgency levels increased, the BSF has also been involved in anti-insurgency and counterterrorism activities. The state government has very little control over the BSF, even though the BSF strictly monitors villages along the border and has been accused of unprovoked firings and human rights violations. After an incident on Uttam Saha in the Raiganj district—in which the BSF took over a ten-kilometer belt along the Indo-Bangladeshi border, and allegedly engaged in highhanded and violent behavior toward the villagers—the West Bengal government asked the center to review the organization’s role.

There are signs, however, that the state has successfully controlled the BSF through other avenues. A West Bengal high court recently forced the BSF to hand over five of its members in a criminal proceeding, saying that if the BSF wanted to take charge of the accused for trial, they could take “appropriate legal steps before an appropriate forum.”31 The center protested, saying that the BSF Act – written in 1968 to clarify the basic duties of the force – had empowered the BSF to handle all investigations within its own organization. Despite these sentiments being well known, the High Court granted jurisdiction to the state government. The state government asserted its autonomy through its judicial process in early 2004, and experienced no backlash from the center.


West Bengal has been remarkably successful at influencing deportation policy. The state has always been careful about this issue because of the ethnic background shared between the peoples of West Bengal and Bangladesh. It is easy, the state argues, for an Indian Bengali to be wrongfully accused and mistakenly deported. The strategies employed by the Union and state regarding illegal immigration have increasingly coincided, but even when there was discord, the state showed signs of autonomy. In 1998, when the Maharashtra government attempted to deport around eighty Bengalis from New Delhi through West Bengal into Bangladesh, the West Bengal government protested, having long demanded that Maharashtra inform it in advance about any deportations. In this instance, the West Bengal government asserted that some of the immigrants were bona-fide Indians, and again demanded that it be given advance notice because the deportations would be happening on its soil. Despite protests from the center that the Maharashtra government had followed the law and acted in accordance with the Foreigner’s Act, the West Bengal government prevailed when the Union home ministry accepted three of its major demands. Those were first, that Maharashtra consult the West Bengali governmentbefore deciding to push back any immigrants; second, that the state government be given thirty days to check if any potential deportee was actually from West Bengal; and third, that the deportation be carried out only after the mode of pushback was discussed.32 In this case, the state was able to exercise a degree of autonomy and regulate a central policy being undertaken on its territory.

Kashmir: State Involvement, Legitimacy, and Peace?

Now we turn to another border state, but one with altogether a different place in Indian political thought and culture. In this section, we see that the state coalition government of Mohammad Sayeed Mufti has undertaken a remarkable number of ECAs even though, from 2002 to 2004, his government was not a part of the central coalition.

The issue of Kashmiri autonomy has long been a central concern of both Indian and Pakistani foreign policy.  The two countries have fought three wars over the region in the past half century since independence.  The people of Kashmir began to play an increasingly important role in the conflict in the 1980s. Before this point, the conflict in Kashmir was generally associated with cross-border infiltration from Pakistan and denial of democratic processes by India—both external factors. Since the mid 1980s, however, the Kashmiri conflict has assumed an internal dimension, as insurgents within the state have waged guerrilla war, and various factions have called for independence. In the past decade,separatists’ groups demanding independence from the Indian Union have been on the rise, along with others who demand a return to pre-1953 levels of autonomy.

Kashmir has only recently, in 2002, had what most cite as its first fair and free elections since 1987. These have facilitated a more “federal” relationship, instead of one in which the center exercises direct rule or rule by proxy. James Manor, in his article about the viability of the Indian federalist system, argues that one of the principal reasons that relations between New Delhi and the states have tended to remain manageable is the existence of political institutions that “can still make the politics of bargaining work.”33 He points out that “Political competition has a number of different outlets. Not only are there elections for national and state legislative assemblies; there are also positions of influence available in three tiers of decentralized, elected councils, and in numerous quasi-official boards, cooperatives. …The existence of so many opportunities to capture at least some power persuades parties and politicians to remain engaged with elections and logrolling, even when they are defeated in some arenas.”34 This lack of political competition characterized elections in Jammu and Kashmir between 1987 and 2002. Previously, there had been no viable alternative to the National Conference (NC), which was backed by the center. In the 2002 elections, the People’s Democratic Party (PDP) performed well and, in alliance with the Congress party, formed the government, taking over from the NC.35 Despite the boycott of the elections by the independence-seeking Hurriyat and other parties, it finally seemed that democracy was at work.

The legitimacy of this government means that it has the potential to exercise more real autonomy from the center, even though the Congress rules both in coalition with the PDP in Jammu and Kashmir and at the center. Indeed, the slogan of the PDP-Congress government has been to apply the “healing touch” locally – in other words, to work for the best interests of the Kashmiris and to counter negative influences from the Indian center. With policies such as the release of militants and the disbanding of the Special Operation Group, the Chief Minister has taken steps in this direction. These trends are similar to those in West Bengal, with the state beginning to participate in ECAs even when the state party is not in a coalition at the center.

In order to understand the constitutionality of state actions, a brief review of the critical constitutional agreements pertaining to Kashmir is necessary. The most important ones include Article 370, the Constitutional Order (1950), the Delhi Agreement (1952), the Jammu and Kashmir Constitution (1957), and the Kashmir Accords (1975).
In October 1947, Article 370 was introduced and adopted by the Constituent Assembly. Article 370 shields the state from certain national legislation and also gives the state powers to legislate on matters of land settlements, rights to property, immigration, and political titles – powers denied to other states.36 It gave the national parliament substantive rights in three areas only: defense, foreign affairs, and communications. These same issues were covered in the Instrument of Accession.

The Constitutional Order of 1950 was issued by President Rajendra Prasad, and affirmed the relationship outlined in Article 370. In July 1952, Nehru and Sheikh Abdullah signed the Delhi Agreement, in which they discussed Kashmir’s position in greater detail. They did not, however, come to a consensus on several important issues, including a separate chapter on Fundamental Rights of State, the eventual jurisdiction of the Supreme Court, and the Emergency Powers. The Constitutional Order of 1954 extended jurisdictional power to the Parliament on almost all the issues in the Union List, and applied the Fundamental Rights of the state constitution. Between 1953 and 1986, forty-two constitutional amendments were passed which increased the power of the center.

In 1975, Sheikh Abdullah, emerging from a twenty-year imprisonment, signed the Kashmir Accords with Indira Gandhi, which essentially confirmed all the constitutional changes that had occurred over the past two decades.37 In the past decade, Kashmiris and others have increasingly clamored that puppet national assemblies unduly gave away their constitutional rights, and that there should be a return to pre-1953 constitutional status. Others have advocated for complete independence and still others for joining Pakistan.

This constitutional altercation is part of a much larger dispute about Kashmir’s place in the Indian Union, and the amount of autonomy the state will have. As a state party, the PDP has acted on three specific issues that affect border policy. Most notably, they have demanded that the separatist Hurriyat group be included in dialogue with the center, that the militants be released, and that the Special Operation Group be disbanded. The NC, which was part of the NDA coalition until 2002, was less effective in this regard.

Dialogue with the Hurriyat

Part of the PDP-Congress Common Minimum Programme, written in 2002, is “to request the Government of India to initiate and hold, sincerely and seriously, wide-ranging consultations and dialogue, without conditions, with the members of the legislature and other segments of public opinion in all three regions of the state.”38 Although this objective mentions no names, it was a clear indication that the government wanted to include the Hurriyat in negotiations. The Hurriyat, founded in 1992 and then factionalized into the All India Hurriyat Conference (AIHC) in the late 1990s, demands a referendum in which Kashmiris could be given the opportunity to leave India without necessarily joining Pakistan. In other words, Kashmiris would vote for an independent state. It is certainly within constitutional boundaries for the state to make such a request, but by no means was it necessary for the center, eventually, to agree to it.

The NC repeatedly made this request, but to little avail. The center did offer to include the AIHC in dialogue in 2000, around the time that Farooq Abdullah, then chief minister of Jammu and Kashmir, introduced an Autonomy Bill that essentially demanded a return to the pre-1953 situation. It is believed that the center then offered to enter negotiations with the Hurriyat as a way of marginalizing the NC.39 Although Farooq Abdullah and the NC were important partners in the coalition, they were also constrained by their dependence on the center. The PDP-Congress coalition has a legitimacy that the NC lacked, and is therefore freer to pursue its own agenda. Even though it is outside the central coalition, it has been far more successful in opening a broad dialogue. At the end of 2002, Advani promised talks with the Hurriyat, which began and were continued by the Manmohan Singh government until they stalled in August 2004.40 In this way, the state government exercised greater autonomy about who it wanted included in negotiations over the state’s future and the future of Indo-Pak relations.

Release of Militants

One of the areas in which the state has exercised extensive autonomy is in the release of militants, both local insurgents and cross-border infiltrators. As part of its common minimum programme, the state government pledged to “release all detainees held on non-specific charges, those not charged with serious crimes and those who have been held on charges that are such that the period they have spent in jail exceeds their possible sentence.”41 Directed mostly toward militants of the Hurriyat and Hizb-ul-Mujahideen (an organization dedicated to the integration of J&K with Pakistan), this statement clearly signaled the administration’s disapproval of the now-discarded Prevention of Terrorism Act (POTA) and the wide-reaching detention powers it granted.  Although the state is responsible for law and order,42 the center is responsible for “Preventive detntion for reasons connected with Defense, Foreign Affairs, or the security of India.”43 Therefore, neither the state nor the center has clear-cut constitutional authority on this issue.

When the PDP came to power, and began releasing militants, the center specifically complained that the state government had been “acting unilaterally despite being sent a letter by the Union home secretary suggesting that decisions with a bearing on the country’s battle against terrorism be taken after consulting the Center.”44 The center, however, did allow the state to continue in this process, because (according to many) it wanted to allow the state time to stabilize and did not want to be seen as de-legitimizing a newly elected government. After fiery debates and exchanges in October and November 2002, in December, the state agreed to put the cases before a screening committee, before it released the militants. According to one article, “Under pressure from coalition partner Congress and the Center, the PDP-led Jammu and Kashmir government has agreed to release militants only after a screening committee examines the cases first. The five-member committee—including a Central nominee—will reportedly be headed by Mr. D.S. Singh, finance commissioner, home [Jammu and Kashmir]. …Though the screening committee will be constituted, it will not be an impediment in the release of a number of detainees whose prison terms under the Public Safety Act are to expire shortly.”45 Although the center interfered and disapproved, the state still accomplished its primary objective.

Disbanding the Special Operations Group

Another bone of contention between the state and center has been the disbanding of the Special Operations Group (SOG). The SOG was founded in 1995 as a counter-insurgency wing of the Jammu and Kashmir police (unlike the BSF or Central Reserve Police Force, which are paramilitary groups). It consisted of non-Muslim, non-Kashmiri recruits, as well as some former militants, and was apparently formed to “create the impression that the counterinsurgency effort had local support.”46 In the past few years, it has become increasingly unpopular, with reports of human rights abuses, torture, and humiliation of civilians. The army and police force, on the other hand, believe that the intelligence and manpower that SOG provides have proven essential in the fight against terrorism and militancy.

To understand the constitutionality of Mufti’s decision to disband the SOG, it is important to recall that the SOG was a state creation under the NC. Because the SOG was a state-run organization and because it dealt with law and order, it may be argued that it was under the state’s jurisdiction according to Items 1 and 2 on the State List of the Indian Constitution. However, because it was engaged with anti-terrorist actions, the center would have had grounds to protest its disbandment. The center did indeed request a “‘revival of the SOG in some form or the other’ by arguing that normal policing” was not adequate to deal with the extraordinary terror threat faced in the state.47 Despite these pressures, the Mufti government has not backed down. It disbanded the SOG in early 2003 and has refused to re-establish it. Here again is an example in which the center was justified in intervening, but instead allowed the state to have autonomy.

Concluding Discussion

In this paper, we have shown that state-level ruling parties (the CPM and the PDP-Congress) that undertook foreign policy-related actions – even when they were not part of the central ruling coalition – have been successful. By contrast, we have also shown that a state-level party (the NC) that was part of the ruling coalition was unsuccessful at foreign policy. These findings support hypotheses H2a and H1b, that coalition governance creates a space for opposition parties at the center to play a role in national politics through their activities at the state level, but not for ruling state parties that are also part of the national coalition. Although we have not presented it in detail, the case studies are further supported by evidence against hypotheses H1a and H2b (see Table 1) about the passive roles of the ruling state parties in Tamil Nadu and Kashmir in Sri Lanka’s and Pakistan’s affairs, respectively, at a time when these parties belonged to the national ruling coalition (1998–2004).

As noted above, our case studies focused on foreign policy advocacy with neighbors (Kincaid’s #7). However, they have not dealt with the full range of possible challenges to federal authority. We have pointed out that Indian states have also conducted negotiations with international organizations (Kincaid’s #3).  In this respect, Sridharan validates her analysis with the example of energy negotiations by the Andhra Pradesh government. Note that these negotiations occurred while the ruling party in Andhra Pradesh supported the central government. Yet several states – such as Karnataka, whose ruling party, the Congress Party, was not part of the central coalition between 1998-2004 – also entered into such international agreements on energy which support Sridharan’s hypotheses (H1a and H2b).

A resolution to the puzzle of why co-ethnic activities, but not economic activities, are the focus of states’ ECAs clearly needs further work. A possible explanation is that the constraints of economic activity are well codified,and as a result, that checks and balances exist. For example, if a state signed a water-sharing treaty with a neighboring country without central approval, it would have no mechanism to enforce the treaty in the absence of a convertible currency and without the participation of centrally regulated banks that would transmit funds. On the other hand, co-ethnic activities, such as the release of militants, are non-codified in the sense that they do not involve bodies that the center regulates or controls.

If, as many believe, India is to have coalition governments at the center for many years, our findings have significant implications for the trajectory of states’ autonomy. A change of guard at the center, which would moderate the central government’s policies over time, could result in a change in ECAs at the state level, but might not bring about similar moderation in state-level policies.


1     See Brian Hocking, ed., Foreign Relations and Federal States, (London: Leicester University Press, 1993); and Jans J. Michelmann and Panyotis Soldatos, Federalism and International Relations—The Role of Subnational Units, (Oxford: Clarendon Press, 1990).

2     U.C. Jain and Jeevan Nair, Encyclopedia of Indian Government and Politics, vol. 7, Centre-State Relations, (Jaipur: Pointer Publishers, 2000); and Robert L. Hardgrave, Jr. and Stanley A. Kochanek, India—Government and Politics in a Developing Nation, 4th ed., (San Diego: Harcourt Brace Jovanovich, 1986), p. 44.

3     Kripa Sridharan, “Federalism and Foreign Relations: The Nascent Role of the Indian States”, Asian Studies Review, vol. 27, no. 4, December 2003; and Pratap B. Mehta, “India: Fragmentation amid Consensus”, Journal of Democracy, vol. 8, no. 1 (1997), pp. 58–59.

4     See Jain and Nair, Encyclopedia of Indian Government and Politics, pp. 47–55; and Hardgrave and Kochanek, India—Government and Politics in a Developing Nation, pp. 115–18.

5     Jain and Nair, Encyclopedia of Indian Government and Politics, p. 48.

6     Agriculture is a state subject, hence agricultural income tax is also a state subject. States have usually chosen not to levy a tax, so that agricultural income is generally tax-exempt. Even in this area, the center has acted to reduce state power. For instance, income from plantations has been exempted from the definition of agricultural income and been subjected to non-agricultural income tax levied by the center.

7     Jain and Nair, Encyclopedia of Indian Government and Politics, pp. 209–10.

8     Rob Jenkins, “India’s States and the Making of Foreign Economic Policy: The Limits of the Constituent Diplomacy Paradigm,” Publius, vol. 33, no. 4, fall 2003.

9     Jenkins, “India’s States and the Making of Foreign Economic Policy: The Limits of the Constituent Diplomacy Paradigm”, p. 69.

10     John Kincaid, “Constituent Diplomacy in Federal Polities and the Nation-state: Conflict and Co-operation” in Michelmann and Soldatos, Federalism and International Relations—The Role of Subnational Units, p. 64.

11     Kincaid, p. 67.

12     Sridharan, “Federalism and Foreign Relations”, pp. 468–71.

13     Several major political corruption scandals during the Rao-led Congress Party government from 1991 to 1996 contributed to the party’s worst ever electoral defeat in the 1996 elections.

14     Coalitions at the state level have been common for many decades.

15     Sridharan, “Federalism and Foreign Relations”, p. 469.

16     See Sridharan, “Federalism and Foreign Relations.”

17     See Times of India articles “W. Bengal Wants Border Trade with Sikkim,” 17 September  2003; “N-E CMS Wants Border Trade Restored,” 21 June 2000; “NE States To Press for Sub-Regional Ties,” 13 June 2000. See also “Basu for Bangla Trade Corridor,” The Statesman, 20 December 1999.

18     See Sridharan, “Federalism and Foreign Relations.”

19     See “ADB Seeks Remake of N-S Board” The Economic Times, 12 May 2002; and “W. Bengal Wants Border Trade with Sikkim,” The Times of India. 17 September 2003.

20     Rekha Saha, India-Bangladesh Relations, (Calcutta: Minerva Associates, 2000), pp. 55–67.

21     Katherine Jaques, Bangladesh, India, and Pakistan. (New York: St. Martin’s Press, Inc., 2000), pp. 17–18.

22     Padmaja Murthy, Managing Suspicions: Understanding India’s Relations with Bangladesh, Bhutan, Nepal, and Sri Lanka. (New Delhi: The Institute for Defense Studies and Analyses, 2000), p. 8.

23     See “Illegal immigration: Need for good policy and timely action,” The Statesman, 11 May 2004. For opinions from state actors on the need for center action, see “Cong no to refugee status for intruders,” Times of India, 6 November 2001; and Saha, India-Bangladesh Relations, p. 174.

24     See L.K. Advani’s comments in “Heat on illegal immigrants,” The Statesman, 8 January 2003.

25     Pathania, Jyoti M. “India & Bangladesh – Migration Matrix”, South Asia Analysis Group [accessed January 4, 2005],

26     See “Neighbour or invader” in, February 23, 2003, p. 1.

27     Saha, India-Bangladesh Relations, p. 174.

28     Foreigners Act, 1947.

29     “Heat on illegal immigrants,” The Statesman,8 January 2003.

30     Government of India, Ministry of Home Affairs, Border Security Force, [accessed August 20, 2004], <>.

31     “State to probe BSF rape case,” Times of India, 12 March 2004.

32     See “State gets its way on deportations,” The Statesman, 25 September 1998.; and “Illegal migrants—Maharashtra has the right to deport,” The Statesman, 15 September 1998.

33     James Manor, “Making Federalism Work,” Journal of Democracy, vol. 9, no. 3 (1998), pp. 21–35.

34     Manor, “Making Federalism Work,” p. 23.

35     The NC still won the largest number of seats of any one party in the state legislature.

36     Vernon Hewitt, Towards the Future? Jammu and Kashmir in the 21st Century, (Cambridge: Granta Editions, 2001), pp. 142–43.

37     “The Autonomy Demand”, Frontline, vol. 17, issue 14, July 8–21, 2000, [accessed August 29, 2004], <>.

38     Government of Jammu & Kashmir, “Elements of a Common Minimum Programme for a Congress-PDP Coalitions Government in Jammu and Kashmi,” [accessed August 10, 2004],  <>.

39     See “Jammu leaders cool to center’s talk offer,” Times of India, 9 April 2000; and “India: Separatism or Autonomy: NC,” TheHindu, 23 June 2000; and “Government to consult NC before talks with J&K Ultras,” The Economic Times, 26 December 2000; and “Hurriyat has no role in Indo-Pak talks: BJP,” The Hindu, 26 May 2001.

40     “Keep the peace: New govt has its job cut on the internal security front,” The Economic Times, 14 May 2004.

41     “Elements of a Common Minimum Programme for a Congress-PDP Coalitions Government in Jammu and Kashmir” (part 3).

42     According to Items 1 and 2, Seventh Schedule, State List, Indian Constitution.

43     See Item 10, Union List, Seventh Schedule, Indian Constitution.

44     “Center, Congress in war of words” Times of India, 27 November 2002.

45     “Ultras to be freed after screening,” The Statesman, 6 December 2002.

46     “Behind the Kashmir Conflict – Abuses by Indian Security Forces and Militant Groups Continue,” Human Rights Watch Report, July 1999, [accessed September 3, 2004], <>.

47     “Mufti rules out SOG by another name,” The Economic Times, 2 June 2003.