Counter-Insurgency in India: Observations from Punjab and Kashmir

by Hamish Telford

INTRODUCTION

Over the past 15 years, India has experienced separatist challenges from a variety of ethnic and religious minorities - Sikhs in Punjab, Muslims in Kashmir and various tribal groups in Assam and other parts of the Northeast Frontier. Between 1983 and 1993, over 20,000 people were killed in Punjab, and since 1989 a greater number of people have been killed in Kashmir. While the insurgency in Punjab developed slowly between 1978 and 1984, it crumbled quickly between 1992 and 1994. The insurgency in Kashmir continues unabated. In fact, it seems to have been transformed into a proxy war with Pakistan.¹ These crises have exacted a serious toll on the Indian state.

Ethnic and religious conflict is not new in South Asia. Indian independence, in fact, was achieved amid one of the greatest religious conflicts of the century. The legacy of partition in India placed two issues outside the boundaries of acceptable political discourse. First, the government of India made clear shortly after independence that demands for the political reorganization of the Indian state along religious lines would not be entertained. Second, it made equally clear that separatist claims would not be tolerated.² Indeed, the Indian Constitution requires all political candidates and elected officials to swear that they will "bear true faith and allegiance to the constitution of India" and that they will "uphold the sovereignty and integrity of India."³

The insurgencies in Punjab and Kashmir over the past 15 years have seriously challenged these two fundamental rules of Indian political practice. As such, the government of India has moved forcefully to repress these movements. When the normal techniques of suppression failed to erode Sikh militancy in the early 1980s, the Indian government experimented with a number of ad hoc counter-insurgency strategies before developing a "successful" modus operandi in the early 1990s. The government of India has, by and large, pursued the same strategy of containment in Kashmir since 1989, but with less success. This modus operandi, which will be described below, was employed in each case with large social costs and without resolving the fundamental political issues. In Kashmir, furthermore, it has not even succeeded in terminating political violence.

COUNTER-INSURGENCY IN INDIA: TOWARD A MODUS OPERANDI

Over the past 15 years, the government of India has developed a counter-insurgency strategy by trial and error. This strategy has a number of components. First, where an insurgency materializes the state government is dismissed and direct rule is assumed by New Delhi. Second, the central government has tried to avoid open, high-profile counter-insurgency operations ever since the bungled operation at the Golden Temple in 1984. Third, the central government has generally refused to negotiate settlements to these crises, preferring instead to fight wars of attrition. Fourth, the problem state is flooded with security forces, which attempt to contain militancy within a small geographic area. Fifth, the security forces have shown a ruthless determination to eliminate militant leaders, while providing leniency for followers. Sixth, the government has been helped by the delegitimation of these movements caused by militant excesses; the security forces may well have helped facilitate this moral disintegration. Finally, when the government feels confident it is winning the war on the battlefield, it moves to restore the democratic process with enforced elections. This places a veneer of democratic legitimacy on the final push to eradicate militancy in troubled states.

The Constitutional and Legal Framework

India is a federal union, but the Constitution equips the central government with extraordinary powers, including the right to dismiss democratically elected state governments. This is the "normal" method for dealing with obstreperous state governments or the breakdown of law and order in particular states. It is made possible by Article 356 of the Indian Constitution, which is supposed to be employed only upon the assessment of the state governor that the constitutional machinery of the state has collapsed. In practice, the central government has been able to invoke Article 356 at will.⁴ It is supposed to be imposed only for six-month increments, although it can be renewed up to a maximum of three years. In the Punjab case however, the Constitution was amended frequently to extend Article 356 beyond three years.

When Article 356 is in effect, the state is ruled from New Delhi, under the authority of the president. Under President's Rule, the government may violate the constitutional freedoms normally guaranteed in Article 19 - the freedoms of speech, association and movement - in its effort to restore order. In fact, the 59th Amendment of the Constitution, effected on 30 March 1988, served to suspend Article 21 - the right to life - in Punjab. This repugnant alteration of the constitution was repealed by the 63rd Amendment, which took effect 6 January 1990, but for 21 months the security of the person was not guaranteed in Punjab, while the government sought to quell this "internal disturbance."⁵

In addition to these constitutional powers, the government of India has a wide array of repressive legislation at its disposal. The Armed Forces Special Powers Act (1956) permits the Army "to arrest suspects, conduct searches, and use lethal force" in "disturbed areas." The National Security Act (1980) "authorizes security forces to arrest and detain without warrant people suspected of undermining national security, public order, and essential economic services." The Unlawful Activities Prevention Act (1967) "empowers the government to ban any subversive organization, such as those advocating secession." The Terrorist Affected Areas (Special Courts) Ordinance (1984) "promulgated in 1984 with special reference to Punjab, provides for secret tribunals to try terrorists." Finally, the now-repealed Terrorist and Disruptive Activities (Prevention) Act (1985) provided the security forces with unprecedented powers of search and seizure. Under the TADA, furthermore, suspects were tried in camera and were presumed guilty.⁶ Harish Puri et al report that 14,457 people were "detained without trial" in Punjab under these ordinances up to 1993.⁷

President's Rule was imposed in Punjab in October 1983, and it remained in effect until February 1992, with the brief exception of the period from September 1985 to May 1987. When Article 356 was invoked in October 1983, the Congress government in Punjab was discredited, and the opposition Akali Dal, the party of Sikh nationalism, was marginalized. The two primary democratic forces in the state were thus sidelined. Sikh militants, however, were emboldened. Sant Jarnail Singh Bhindranwale, the fiery religious leader of the militants, had been directing his followers from the guest house of the Golden Temple since July 1982. In December 1983, just after the imposition of President's Rule, he shifted to the Akal Takht, the seat of Sikh temporal and spiritual authority, in the inner sanctum of the Golden Temple. He was joined, furthermore, by his armed supporters. The militants continued their campaign from the Akal Takht, apparently in the belief that they would not be attacked in a place of religious worship.

The political dynamics in Punjab at the time were such that the imposition of Article 356 was ineffective in containing Sikh militancy. The problem in Punjab did not lie with elected, democratic political parties. The culprits were extra-parliamentary militants, yet it was the government and political parties that were sanctioned. Legitimate political space was consequently closed, and the Indian government found itself facing a shadowy, underground enemy. While President's Rule was ill-advised and ineffective in Punjab, it continues to be the weapon of first strike for the government in these situations. The government seems to consider it necessary, if insufficient, to assume direct control of the state to restore law and order.

The Lesson of Operation Bluestar: Avoid High-Profile Military
Confrontation

Six months after Bhindranwale moved into the inner sanctum of the Golden Temple, Prime Minister Indira Gandhi ordered the army to storm the Temple and arrest Bhindranwale and his followers. Operation Bluestar, as the attack was code-named, was a military and political disaster. The army did not have the military capacity or the intelligence necessary to perform the operation. It deployed regular troops for the operation, and it initiated the attack with mortar fire in the middle of a crowded city. Bhindranwale and his key supporters were killed, as were about 1,000 innocent pilgrims. The Temple, furthermore, was heavily damaged. The larger Sikh community was devastated by Operation Bluestar. Six months later, Indira Gandhi was assassinated in retaliation by two Sikh bodyguards.⁸

After Operation Bluestar, it was decided that the regular army would not be employed in high-profile operations. The regular army was subsequently deployed to secure the state, while commandos were trained for special operations. The National Security Guards (NSG) was formed in 1985 as an elite anti-terrorist force, and to protect leading public figures, and "[i]n 1991, the Army raised a new force, the Rashtriyia (National) Rifles, specifically charged to deal with terrorism, rioting, and communal violence. By 1996-97, the new force had 40,000 members."⁹ As the last vestiges of militancy were being eradicated in Punjab in 1992-93, the army was deployed to provide a secondary security perimeter while the police engaged in most of the actual combat.¹⁰ In short, the government decided that counter-insurgency had to be as discreet as possible.

The Lesson of the Longowal Accord: Refuse to Negotiate a Settlement

After the assassination of Indira Gandhi, Rajiv Gandhi replaced his mother as leader of the Congress party and as the Prime Minister of India. In the general election the following month, he campaigned with rather chauvinistic rhetoric against the Sikhs but, after winning the largest majority government in the history of India, he felt confident to pursue a more conciliatory strategy. In summer 1985, Rajiv Gandhi struck a deal with Sant Harcharan Longowal, a moderate leader of the Akali Dal. The accord included the restoration of the democratic process and the transfer of Chandigarh - the joint capital of Punjab and Haryana - to Punjab alone. It was agreed in the accord that more contentious issues would be resolved later by commissions and future negotiations.¹¹ Longowal was assassinated within a month by Sikh militants, but the election proceeded as scheduled in September 1985. The Akali Dal, now led by Surjit Singh Barnala, won its first majority government and a degree of normalcy returned to the state.

As the date for the transfer of Chandigarh approached at the end of January 1986, the government began to hesitate, fearing a backlash from right-wing Hindus across north India, especially in the neighbouring state of Haryana. Ultimately, Chandigarh was not transferred on the date stipulated in the accord. The Chief Minister of Punjab, Surjit Singh Barnala, instantly became a lame duck and militancy was resuscitated. In April 1986, militants recaptured the Golden Temple and announced the "formation" of "Khalistan." This was the first time that the militants openly declared their separatist intentions. In 1987, Barnala was dismissed and President's Rule was re-imposed in Punjab.

After the Longowal accord failed, the centre evidently determined not to accept a negotiated settlement, lest they agree to promises under duress that later they might not wish to keep. The Punjab crisis certainly ended without a negotiated settlement. The government of India decided instead to fight a war of attrition. The government, until very recently at least, similarly refused to negotiate a settlement in Kashmir. In spring 2000, immediately following President Clinton's visit to India, it contacted some militant groups about the possibility of conducting peace negotiations.¹² On 24 July, the Hizbul Mujahideen announced a ceasefire and its intentions to negotiate a settlement with the government.¹³ Other militant organizations opposed the negotiations and sought to thwart them by engaging in a series of massacres that left more than 90 people dead.¹⁴ The talks collapsed within a week, when the Hizbul insisted that the government of Pakistan would have to be brought into the process.¹⁵ Nevertheless, the government of India announced a new ceasefire on 29 November to coincide with Ramadan; it subsequently extended its unilateral ceasefire for the month following Ramadan.¹⁶ Although winter is typically a time for depressed insurgency and counter-insurgency activity, this ceasefire may indicate a new willingness by the government of India to negotiate a settlement in Kashmir.

The Indian government, however, is likely responding to the imperatives of geopolitics, as opposed to a re-evaluation of its counter-insurgency strategy. While India's decision to test nuclear weapons in May 1998 was condemned by the international community, it did succeed in focussing world attention on South Asia. Since that time, the United States has shifted its foreign policy preferences from Pakistan to India. Pakistan was criticized for initiating the Kargil conflict with India in the summer of 1999, and it also lost international support after the military coup in October 1999. The United States has also been displeased by Pakistan's continued support of the Taliban regime in Afghanistan. In an effort to exploit Pakistan's current weakness, and to solidify its relations with the United States, the government of India has attempted to seize the initiative in Kashmir. However, by initially sidelining the democratic government in the state, the government must now negotiate with the militants. This is a direct consequence of the government's counter-insurgency strategy. The geopolitical significance of the Kashmir conflict likely makes this case the exception to the rule. In other words, the government of India is not likely to change its general disposition against negotiated settlements in crisis situations without geopolitical significance.

Overwhelming Security Presence and the Geographic Containment of
Militancy

If Indian leaders learned to avoid high-profile military confrontations, this did not stop them from flooding troubled states with an overwhelming number of security forces - regular army, state police, the Central Reserve Police Force and the Border Security Force. At the height of Sikh militancy, there were approximately 250,000 security forces in Punjab, a state of about 20 million people. Some 400,000-500,000 troops have been stationed in Kashmir, a state of only 5 million, although admittedly many of these troops are defending the Line of Control with Pakistan, not fighting Kashmiri militants. Indeed, the primary objective in both states was to seal the border with Pakistan, which was a point of refuge for Sikh militants, as it continues to be for Kashmiri militants. In Pakistan, Sikh and Kashmiri militants were safe from the Indian government; they could solicit the support of the Pakistan government and it was an excellent place to purchase weapons.¹⁷

While sealing the border with Pakistan was an obvious security objective, non-border troops were initially deployed rather haphazardly.¹⁸ Sikh militancy was always more concentrated in the northwest districts of Amritsar and Gurdaspur, and consequently the security forces were deployed so as to contain the insurgency in these areas.¹⁹ With the Pakistan border sealed on the west, and the mountains of Jammu in the north, the militants had nowhere to escape. At this point, it was only a matter of time before the militants were eliminated. In Kashmir, the fighting has been contained in the small Vale of Kashmir, north of Jammu and west of Ladakh, with the Line of Control in the north and west. The mountainous terrain, however, is much more difficult to patrol than the plains of Punjab. Nonetheless, the geographic containment of militancy in isolated areas has been a crucial aspect of India's counter-insurgency strategy. This strategy, furthermore, was congruent with the military's conventional "positional war" orientation.²⁰

Separation of Leaders and Followers

In Punjab, the security forces initially attempted to squash militancy indiscriminately. This did not succeed; it turned every potential militant into an actual militant. In its ruthless attempt to capture militants, the police frequently tortured the relatives and friends of suspected militants to obtain information on the whereabouts of the suspects. For many young men, the choice was to fight or be tortured. Furthermore, captured militants were routinely tortured and very often killed. There was thus very little incentive for militants to lay down their weapons. This tactic continued until the end of 1991, when K.P.S. Gill was reappointed as Director-General of the Punjab police.

After Gill assumed command, the police moved to separate militant leaders from their followers. Huge bounties were placed on the heads of leading militants. Rewards of Rs 25,000 to Rs 100,000 were standard, and a leading militant like Gurbachan Singh Manochahal was killed by the police with a three million rupee price on his head,²¹ as compared to the standard police salary of about Rs 2,500 per month. The police suddenly became highly motivated to capture senior militants. At the same time as the militant leaders were being hunted down, a tacit policy of amnesty, or at least leniency, was offered to the rank and file fighters. As senior militants were eliminated one by one, their followers could surrender. They were finally given a route out of the quagmire. The physical elimination of senior militants and leniency for the "small-fry" thus became another element in the counter-insurgency strategy developed by the Indian government.

The Delegitimation of Militancy

The Sikh separatist movement was never unified. Moderate Sikh nationalists were riven by patron-client networks and egotistical leaders. The militant nationalists were divided by personal vendettas from the inception of the movement. The unrelenting suppression of the movement instilled further distrust among militants, and caused a rapid factionalization among the militants. Over time, the Khalistan movement descended into thuggery. The militants increasingly engaged in robbery, extortion, rape, indiscriminate killings and ever-escalating terrorist attacks on innocent civilians. By 1991, Sikh militants were generally viewed as unprincipled criminal gangs. The brother of a leading militant quipped, "the reason that the Khalistan [land of the pure] movement failed was because the boys began working towards an Ujadistan (a land of ruin)."²² A similar pattern of debauchery and degradation has emerged in Kashmir. It is not clear if this moral disintegration was wholly self-inflicted, or if it was aided and abetted by security elements operating inside militant groups. While the security forces almost certainly infiltrated the militant movement to gather intelligence, the full scope of their activities remains largely unknown.²³

A new strategy has been employed in Kashmir: surrendered or captured militants have been redeployed by the security forces as counter-insurgents or "friendlies."²⁴ With 3,000 armed "friendlies" in the state, they are almost as strong as the militants. In Punjab, Gill deputized a number of low caste mazhabi Sikhs as "Special Police Officers." This tactic was designed as much to neutralize their incentive for joining the militants as it was for community protection.²⁵ These "police officers," however, were not former militants; indeed, mazhabi Sikhs were frequently the victims of militant attacks. Their loyalty was thus assured. In Kashmir, the recruits were initially fickle mercenaries whose allegiances switched each time a better offer was directed at them. This was a very risky strategy.

More recently, at least in the Jammu region of the state, the police have organized "village defence committees" (VDCs). The VDCs were initiated in 1995, but in the past two years they have mushroomed from 400 units to over 1600 units.²⁶

Since they are familiar with the area, VDC members have been providing the security forces with valuable information on the militants' movements and tactics. Thanks to the tips given by villagers, the number of militants killed by the security forces has almost doubled in the past year. [Consequently], [e]quipping the VDCs and the special police officers (SPOs) is now the top priority for the state police.²⁷

This community policing program, however, is not without its problems. In particular, the scheme seems to have heightened tensions between the Hindu and Muslim communities in the region. By and large, the Hindu communities have been receptive to the scheme, whereas the Muslim communities remain quite suspicious. Indeed, "there has been a persistent demand by local Muslim leaders to disarm the VDCs."²⁸ As such, the village defence program is likely to have little chance of success in the predominantly Muslim part of Kashmir where the fighting is concentrated.

The deligitimation of the Khalistan movement was a crucial variable in the termination of the Punjab conflict. How much the security forces precipitated this moral decline is unclear, but it seems evident that the government is also actively attempting to facilitate the factionalization and delegitimation of Kashmiri militancy. However, it is not clear that the factionalization of the insurgency makes it easier to contain. It may reduce the overall fighting capacity of the movement, and consequently frustrate the primary objective of the insurgency, but it may make it more difficult to eliminate the numerous fighting units this strategy creates. Many little gangs can inflict considerable damage, even if they cannot collectively achieve their objective of separating their respective states from India. It also makes a negotiated settlement more difficult because an agreement with one group might not be respected by others. Although the tactic is risky, it has become a part of India's counter-insurgency strategy.

Enforced Elections: Punjab

The Congress government of India went to extraordinary lengths to hold state elections in Punjab in February 1992, reversing a long-standing policy of not holding elections in disturbed areas. (Elections for the 13 parliamentary seats were held simultaneously.) The Punjab elections were originally scheduled to coincide with the general election in May-June 1991.²⁹ Most militants, however, opposed the election, fearing that it would give voice back to democratic Sikh leaders in the Akali Dal, who at that time were effectively marginalized. As the June poll approached, the level of violence in the state increased, culminating in two train massacres. On the eve of the poll, the Chief Election Commissioner cancelled the election, and rescheduled it for September.³⁰

While the Punjab election was cancelled, the Congress party won the general election, despite the assassination of Rajiv Gandhi. Narashima Rao, who emerged as the party leader after the death of Gandhi, became prime minister. As September approached, the new Congress government determined it was not yet ready to hold fresh elections in Punjab and pushed the election back again, until February 1992. As February approached, the militants once more opposed the prospect of an election and sought to thwart it. All of the mainstream factions of the Akali Dal, save one minor group, decided resolutely to boycott the poll. The government, however, was determined to hold the election. The state was swamped by 750 paramilitary companies and nine army divisions,³¹ comprising some 250,000 troops. Every candidate was assigned at least a 32-man security detachment.³² The main Akali leaders and several hundred party workers were detained under the Terrorism and Anti-Disruptive Activities (Prevention) Act to ensure that they could not disrupt the election. Under these Orwellian conditions, Punjabis were asked to go to the polls. Voter turnout was a meagre 21.6 percent, significantly less than a third of the normal Punjabi turnout of about 68 percent. In the twelve urban constituencies, the turnout was 38.3 percent, but in the 70 rural constituencies, where the militants held greater sway, the turnout was an abysmal 15.1 percent. In one constituency, the turnout was less than 1 percent.³

The Congress Party swept to power with rather exaggerated fanfare, given the electoral abnormalities. The Congress captured 87 of the state's 117 assembly seats with 43.8 percent of the vote. In other words, the Congress mandate emanated from less than 10 percent of the electorate. No other party captured more than nine seats. Beant Singh, an old Congress stalwart, was installed as the Chief Minister of the state. Beant Singh was a rough-hewn Sikh and, with his signature dark glasses, he was a rather sinister-looking individual. Notwithstanding the perversion of democratic norms, the election produced the result desired by the Congress government at the centre.

Beant Singh assumed the officer of Chief Minister just three months after the imposing but urbane K.P.S. Gill was reappointed as Director-General of the Punjab Police. Both men were Jats, the same caste as the majority of Sikh militants.³⁴ Gill rearmed the police and promised his men that he would shoulder all criticisms of police tactics. In short, he transformed the police into a disciplined and motivated fighting force.Beant Singh, with the support of the central government, gave Gill a free hand to quash militancy in the state. The presence of an elected government provided a facade of political normalcy in the state and allowed Gill to proceed with his mission without creating the impression that Punjab was a police state. The people accepted this charade because they were tired of the turmoil and the excesses of the militants. A Sikh journalist wrote in 1994 that "[w]hen [the] history of the fight against terrorism comes to be written, Mr. K.P.S. Gill could well be judged (by those partial to him) as one who restored order at the expense of law. But that seems besides the point now that the basic objective of peace has been largely achieved."³⁵ The depravity of the militants may well have been the government's saving grace.

Under Beant Singh, militancy was quelled in Punjab. Beant Singh, however, was assassinated on 31 August 1995, by a car bomb placed outside the state secretariat in Chandigarh, the capital of Punjab.³⁶ After Singh's death, the Congress government of Punjab stumbled along with a succession of hapless Chief Ministers until the state election of February 1997, when it was trounced by a resurgent Akali Dal. The voter turnout returned to its normal 69 percent. The Akali Dal, led by former Chief Minister Parkash Singh Badal, captured 75 seats, while its electoral partners, the Bharatiya Janata Party (BJP), won 18 seats. The Congress was reduced to 14 seats.

The Akali Dal's stunning victory and the Congress Party's ignominious defeat can be attributed to a couple of factors. The Congress seems to have paid the price for the strong-arm tactics employed by the police and security forces to terminate militancy. The near elimination of militancy, however, permitted the moderate Sikh nationalists to return to centre stage and it allowed the rural Sikhs to return to the polls. In this sense, the Congress was a victim of its own success. But its loss is perhaps more attributable to its failures. The fundamental political tensions that existed between Punjab and New Delhi in the 1970s were still unresolved. Consequently, the Akali Dal's old political platform was just as relevant as it was 20 years earlier. Finally, the national Congress party was no longer in power in the centre, and its reputation was in tatters, as its major leaders were charged with various counts of corruption. The national party was simply in no position to assist its Punjabi colleagues; indeed, it may well have become an electoral liability.

Gurharpal Singh has argued that the 1992 Punjab election was held "to fulfil a constitutional obligation," even if the cost was a "serious erosion of democratic legitimacy."³⁷ The elections were no doubt a democratic charade, but they were not held to meet a pro forma obligation. There was no legal requirement to hold the election - the central government could have continued to rule the state almost indefinitely through the emergency provisions of the constitution.³⁸ The purpose of the election was to provide a veneer of democratic legitimacy to the otherwise forceful suppression of Sikh militancy. While democratic normalcy largely returned to the state with the 1997 election, the 1992 exercise was nonetheless still a risky strategy.

Enforced Elections: Kashmir

By mid-1995, the Congress government in the centre was of the opinion that its election strategy in Punjab had been successful, and they began to plan for elections in Kashmir. The Vale of Kashmir, however, was more unsettled than Punjab and the people had a much greater antipathy toward the government of India. Most Muslim Kashmiris, unlike most Sikhs, do not identify themselves as Indian. This makes the Kashmir crisis much more intractable and the holding of elections in the state more problematic. Furthermore, fraudulent elections, especially the 1987 state poll, are thought to be a cause of militancy in Kashmir.³⁹ An enforced election was thus a risky proposition.

It was decided that elections for the six parliamentary seats would be held prior to the state assembly poll. The parliamentary elections were scheduled for May 1996, with the state election to be held in October. The conditions in May were so poor that polling for the six seats had to be spread out over three days. As with Punjab, the state was flooded with security forces, some 950 paramilitary companies and two army divisions, representing over 100,000 soldiers. Moreover, about 9,000 Urdu-speaking election officials were flown into the state to conduct the elections, with the promise of an extra month's salary and a half-million rupee life insurance policy.⁴⁰

The parliamentary election was boycotted by the National Conference, historically the governing party of Kashmir, and by the Hurriyat Conference, the political umbrella of the leading militant groups. The militants, as in Punjab, endeavoured to sabotage the elections. Without the National Conference, and with militant threats, the people of Kashmir - as with Punjab in 1992 - were inclined not to vote. Intelligence assessments for the Home Ministry suggested the voter turnout in the Vale of Kashmir could be as low as 10-20 percent, even worse than the farcical 1992 Punjab poll. Desperate to avoid an electoral embarrassment, the government seems to have ordered the police and army to ensure a good turnout. There were reports of soldiers moving from village to village with voter lists, rousing people to vote and threatening those who resisted.⁴¹ This strategy was moderately "successful." In the constituencies of Baramulla and Anantnag in the volatile Vale of Kashmir, the voter turnout was 35 and 43 percent respectively.⁴² When the ballots were counted, the Congress Party had won two of the three seats in the Vale of Kashmir and four of the six in the entire state. The next step was to conduct a state assembly poll.

By the time the state assembly poll was held in October 1996, the political landscape had changed dramatically at the centre. The Congress government was soundly defeated in the national election in May 1996; the BJP, the largest single party in parliament, was unable to form a government; and a thirteen-party coalition, led by the Janata Dal, assumed office with H.D. Deve Gowda as Prime Minister. The regional character of this motley coalition spawned considerable discussion about the federalization of the Indian party system.⁴³ This new climate may have emboldened Farooq Abdullah, the leader of the National Conference, to contest the state assembly poll. Abdullah's reputation was badly tarnished by allegations of massive electoral fraud in the 1987 state election, but his credibility was restored by his boycott of the May 1997 parliamentary poll.

While the militants and the Hurriyat Conference continued to oppose the election, the election proceeded "in what most independent observers concede was a free and fair election after the farcical parliamentary poll."⁴⁴ Voter turn-out ranged from 15 percent in some constituencies to as high as 60 percent, and the National Conference swept to power, capturing 59 seats in the 87 seat legislature. The Congress was reduced to seven seats, and the BJP took eight seats, all in Jammu, indicating a hardening of Hindu sentiment in that region of the state. While Farooq Abdullah's campaign for state autonomy may have struck a cord with some voters, reports from the state suggest that the National Conference's victory does not represent a whole-hearted endorsement of the party. Rather, the people may have simply "turned to what they recognized as the only credible political force to steer them out of the quagmire."⁴⁵

The election of the National Conference in Kashmir has not been a panacea for the troubled state, but neither has it been an unmitigated disaster. However, it seems that Farooq Abdullah has not taken advantage of his opportunity to capture the hearts and minds of the Kashmiri people. Furthermore, it would appear that the people of Kashmir have grown disillusioned with the democratic process. In the March 1998 federal election, voter turn-out in Srinagar was 30 percent; in the September 1999 federal election, voter turnout in the state capital dropped dramatically to 11 percent.⁴⁶ The electorate was almost certainly fearful of the resurgence of militancy in the summer of 1999 (see below), but they may also have come to the conclusion that the electoral process has not brought a solution to the troubled state. In the last four elections, the people have taken serious risks when they voted. The electoral calculus may now be that the weak return for voting does not warrant taking the risks of voting. In sum, "[d]espite three years of popular - at least nominally - government, the level of confidence among the people is abysmally low."⁴⁷

The final stage of India's counter-insurgency strategy is the holding of enforced elections. The government moves to this stage when it feels confident it has gained the upper hand on the battlefield. The strategy appeared to be "successful" in Punjab, but this might owe more to the fact that the partisan interests of the Congress governments of Punjab and India coincided. The same alignment of party interests does not exist between the centre and Kashmir. The strategy thus does not seem to be as successful in Kashmir as it was in Punjab.

COUNTER-INSURGENCY IN INDIA RECONSIDERED

Militancy has essentially been eradicated from Punjab, while the situation has been complicated in Kashmir by a large number of foreign mercenaries. In this sense, India's counter-insurgency strategy might be considered partially "successful." On the other hand, the strategy has had large social, economic and political costs. These counter-insurgency campaigns have exacted a heavy death toll. They have severely undermined the economic well-being of many families and retarded the economic development of the affected states. This strategy, furthermore, may not be appropriate in other states; it is not even clear that it was wholly successful in either Punjab or Kashmir. India's counter-insurgency strategy has been fundamentally apolitical. It has entailed serious violations of human rights, and it may have disturbed civil-military relations. Finally, these counter-insurgency campaigns may have weakened the democratic legitimacy of the Indian state.

The Human and Economic Cost of Insurgency and Counter-Insurgency

India's counter-insurgency strategy is premised on the assumption that a large state will eventually prevail over a small secessionist movement. The assumption is perhaps not incorrect, but there are large costs associated with such a strategy. The government of India has engaged in two successive wars of attrition. The death toll has been staggering. Over 50,000 people have been killed in Punjab and Kashmir combined. As Punjab and Kashmir are relatively small states, this high death rate has affected most villages and many families. Furthermore, since many of the victims were young men, many families have lost important breadwinners.

Each of these wars has dragged on for at least ten years at tremendous economic cost. In summer 1997, I. K. Gujral, the short-lived prime minister of India, forgave Punjab's debt of Rs 8,500 crores (about 3 billion USD), accumulated combatting militancy since 1984.⁴⁸ This was just the state government's debt; the central government would have assumed additional costs for its own counter-insurgency efforts. Similar costs will also have been incurred in Kashmir, perhaps even greater costs given the isolation of the state, difficult terrain and harsh (winter) climate.

There was also significant economic dislocation in each case. Punjab has remained the wealthiest state in India per capita but it is impossible to estimate how much investment the state lost during its "troubles." Punjab is still highly dependent on agriculture; its industrial development lags behind Gujarat and Maharashtra, the second and third richest states in the country. Kashmir, a rather impoverished state, has been decimated by the almost total collapse of tourism, formerly the state's primary industry. In each state, thousands of young men, participants who were not killed, have had their economic potential seriously depreciated by these decade-long conflicts. The social malaise that has pervaded each state has had incalculable human and economic costs.

Limited Utility of India's Counter-Insurgency Strategy

The Indian government has had to expend considerable energy combatting relatively small insurgencies in two of the country's smaller, landlocked, states. Indeed, militancy has been largely confined in each case to two or three districts. Militancy peaked in each state with no more than 6,000 lightly armed insurgents.⁴⁹ In each case, however, the government was required to deploy about 250,000 security forces, and still the disturbance has not been quelled in Kashmir. Furthermore, as the Punjab and Kashmir conflicts overlapped, half a million forces were engaged simultaneously in two internal disputes, leaving India's external defences vulnerable.

It is hard to imagine that India's counter-insurgency strategy would be successful in one of the country's larger states, especially a coastal state. Tamil Nadu, for example, has a population four or five times that of Punjab and a long coast line. A separatist insurgency in this state could not be defeated with the same tactics employed in Punjab or Kashmir. The Indian Army learned this lesson with its ill-fated peace-making mission to Sri Lanka in 1987.

Limited Success of the Strategy

It is not clear, in fact, how successful India's counter-insurgency strategy has been in either Punjab or Kashmir. The conflict in Kashmir continues, although perhaps perpetuated by foreign mercenaries.⁵⁰ In early 1997, there were still more than 3,500 armed militants in the state.³⁵¹ The presence of foreign fighters in Kashmir stands in contrast to the case in Punjab. Since 1991, as many as 1,380 foreign mercenaries have been killed in Kashmir by the security forces, and another 142 have been arrested.⁵² It was estimated that 1,500 foreign mercenaries were operating in the state in the last few years,⁵³ and it is speculated that as many more have crossed into Kashmir since summer 1999.⁵⁴

By all accounts, the crisis has escalated dramatically since the conflict with Pakistan in summer 1999. The government of India has reported that in the months following the clash with Pakistan, the number of violent incidents in Kashmir increased by 27 percent, "and attacks against the security forces by 95 percent."⁵⁵ In July 1999, militants in Kashmir launched their first suicide mission in the state, following the example of Hizballah in Lebanon and Israel, as well as of the Tamil Tigers in Sri Lanka.⁵⁶ This trend apparently continued throughout 1999, as militants "aimed at chipping away the morale of security forces."⁵⁷ The militants, furthermore, have continued their offensive into the winter, unlike previous years.⁵⁸

The successful hijacking of an Indian Airlines flight in December 2000 almost certainly raised the morale of the militants in Kashmir.⁵⁹ Two of the prisoners released in exchange for the hostages have surfaced in Pakistan and devoted themselves anew to the liberation of Kashmir. Maulana Masood Azhar, who is often described as a Muslim cleric from Pakistan, is the purported leader of the Harkat-ul-Ansar, an umbrella organization of the groups fighting for the separation of Kashmir from India. The hijackers also secured the release of Mushtaq Ahmed Zargar, alias Latram, who was the leader of the Al Umar Mujahideen, one of the original militant groups in Kashmir.⁶⁰ It is suspected that these groups have links to Osama bin Laden. Indeed, bin Laden reportedly has declared that "India and America are now our biggest enemies . . . all mujahideen groups in Pakistan should come together to target India . . . we are always ready to help the Kashmiri mujahideen."⁶¹ It is now feared that "bin Laden's jihad may revive the ideological basis of Kashmiri militancy."⁶² Indeed, it was reported in early 2000 that as many as 500 youth had crossed into Pakistan for training.⁶³

In the post-Kargil, post-Kandahar period, it appears that India's counter-insurgency strategy "is fast spinning out of control."⁶⁴ Indeed, the ability of militants to strike seemingly at will has created "the impression that they are more in control than the government."⁶⁵ India now faces two fundamental problems. First, the insurgency in Kashmir is deeply entrenched. This will have to be solved politically as well as militarily. But, second, this is no longer simply a domestic problem. The government of India has long resisted "internationalizing" the question of Kashmir, inasmuch as it does not want this issue solved by the United Nations or actors outside the region. However, it would now appear that India will not be able to resolve the situation in Kashmir without a peace settlement with the government of Pakistan. Indeed, as one Indian observer has noted, "Pakistan is not only a party to the Kashmir dispute, it is now a party in Kashmir as well."⁶⁶ Thus, from an Indian perspective, the greatest failure of the government's counter-insurgency strategy in Kashmir may be that it will be forced to make serious compromises in its relations with Pakistan.

Peace has largely returned to Punjab, but militant elements remain in the state and abroad. In fact, as late as 1997, the state was rocked by relatively frequent bomb attacks, including a train bomb which killed 38 people.⁶⁷ Furthermore, half a dozen small bombs were detonated in and around Delhi in October and November 1997, leaving at least seven people dead. While 1998 was relatively calm in Punjab, there were fears that militant groups were re-activating in 1999. In January and February 1999, a number of explosions occurred around the state.⁶⁸

India Today, citing anonymous intelligence sources, has claimed that about 300 militants, including two-dozen hardcore terrorists, were operating in Punjab in 1997.⁶⁹ Most of these militants were not on police lists, which have not been updated since 1993.⁷⁰ Moreover, militant groups are now making a concerted effort to recruit religiously motivated youth who do not have criminal records.⁷¹ The situation is further confused by the byzantine array of militant factions. At least nine groups were known to exist in 1997, ranging in size from 10 to 50 people.⁷² New militant tactics have confounded a rather demoralized police force. These groups appear to be much more clandestine than they used to be. They are organized with a more rigorous and compartmentalized cell structure such that the relatively few militants still operating in India are not known to each other. Furthermore, these groups have tended not to claim responsibility for their attacks, so as not to provide any clues for the police.

It seems that operations are now planned abroad by groups like the Babbar Khalsa International and the International Sikh Youth Federation but carried out by affiliated groups in India. The Babbar Khalsa tends to recruit from the community of illegal immigrants in the west and returns them to India with the necessary training and financial resources to continue the movement.⁷³ It is reported that the Babbar Khalsa, and other Khalistani militant groups, are supported by as many as 5,000 sympathizers living in at least 20 countries around the world. Most of the Khalistan militant groups also allegedly have operatives in Pakistan, where they have reportedly forged alliances with various Kashmiri separatist groups. The highly pluralistic region of Jammu, wedged between Punjab and Kashmir along the Pakistan border, has also become a significant operational base in India for both Kashmiri and Sikh militants.

Thus, seven years after militancy was thought to have been eradicated from Punjab, the police find themselves playing a cat-and-mouse game with an ever more elusive enemy. K.P.S. Gill, who thought he had won the battle against militancy, stated flatly in 1997, "there is no way we can stop terrorism from returning to Punjab."⁷⁴ This sad development is probably more attributable to the political failure of India's counter-insurgency strategy rather than any military shortcomings.

Failure to Address Underlying Political Issues

The long-term structural causes of militancy in Punjab and Kashmir are complex and vigorously debated.⁷⁵ While it may have been difficult for the government of India to foresee the underlying causes of militancy, it must accept responsibility for ignoring long-standing grievances in each state. These grievances stoked the fires of political unrest. While state politicians frequently raised their grievances with central authorities, they were repeatedly rebuffed. The state politicians consequently had nothing to show for their efforts, and thus were forced to return to their constituency empty-handed. It is little wonder that alienated youth lost patience with their traditional leadership and resolved to pursue a more radical course.

In Punjab, the Akali Dal presented a list of 45 demands in 1981.⁷⁶ Two issues were at the core of these demands - the transfer of Chandigarh and other territorial adjustments, and the division of waterways between Punjab, Haryana and Rajasthan.⁷⁷ Rajiv Gandhi promised to transfer Chandigarh to Punjab in January 1986, as part of his agreement with Sant Longowal, but he failed to deliver on his promise. Indeed, none of the salient aspects of the Longowal Accord came to fruition.⁷⁸ Four years after the end of armed hostilities, no progress has been made on any of these issues. The proximate causes of militancy in Punjab thus still exist. If the Akali government of Punjab is again unable to make progress on these issues with the centre, it may once again lose support in the Sikh community and open the door for renewed militancy. The Akali leaders did not help themselves with their incessant squabbling in 1999.⁷⁹

A similar pattern is evident in Kashmir. The fundamental issue concerns Kashmir's constitutional relationship to India. As a former princely state of British India, the Maharaja of Kashmir was required only to cede the powers of defence, foreign affairs and communications to the government of India, as stipulated in Article 370 of the Constitution. Sheikh Abdullah, the leader of the National Conference, refused to join India's Constituent Assembly and he attempted instead to negotiate a special status for Kashmir in India based on the terms of accession. Nehru tired of Abdullah's tenaciousness and orchestrated his downfall in 1953. The new leader of the National Conference, Baskhi Gulam Muhammad, quickly reached a constitutional agreement with the central government. Muhammad's 1953 agreement with the central government considerably reduced Kashmir's autonomy inherent in the terms of accession, although it still provided Kashmir with more autonomy than the other states in the Union.⁸⁰

Over the next 40 years, the autonomy of Kashmir was eroded further as the Indian government attempted to integrate the state into the Union on the same terms as the other states, contrary to the provisions in Article 370. After nearly a decade of bitter political conflict in the state, there is no indication that the government of India is ready to respect the provisions of Article 370. While the Janata Dal coalition was potentially more sympathetic to regional concerns than the other major parties, it did not restore the spirit of Article 370 during its term of office. On the other hand, the BJP government has not moved to eliminate Article 370 since it came to power in 1997, as it threatened to do when it existed in opposition.

Farooq Abdullah and the National Conference won the 1996 state election largely as a result of his promise to negotiate autonomy from New Delhi. Abdullah's inability to obtain autonomy from the centre threatens his fragile legitimacy. In early 2000, Abdullah renewed his efforts. He declared threateningly, "[i]f Jammu and Kashmir has to remain a part of India, autonomy has to be given. It's better the Centre start thinking that this has to happen to win the hearts of the people. They can dismantle me but the issue of autonomy will not disappear."⁸¹ However, it is exceedingly unlikely that the BJP government will accede to these demands.⁸² The underlying political issue in Kashmir thus remains unchanged, and the alienation of the Kashmiri people continues unabated.⁸³ Indeed, the alienation in the state may have deepened over the past decade as a result of the central government's oppressive counter-insurgency strategy. Once again, this strategy has failed to eliminate the root cause of militancy.

India's counter-insurgency strategy is decidedly apolitical: it assumes that militancy can be eliminated by force. Militancy in Punjab and Kashmir, however, has been politically motivated and as such it requires a political solution. Without a political solution, the long term prospects for peace appear tenuous, especially in Kashmir. While the political dynamics in Kashmir are much more explosive than they were in Punjab, the Akali Dal also requires a political settlement to ensure that it is not outflanked by militants in the future. While India's counter-insurgency strategy is apolitical, it has had political consequences. It has precipitated a delegitimation of the police, a weakening of civil-military relations and an erosion of India's democratic traditions.

Human Rights Violations and the Delegitimation of the Police

When K.P.S. Gill was reappointed as the Director-General of the Punjab police, it seems that there was at least a tacit understanding with the central government that human rights violations would be part of the price to pay for the restoration of order in Punjab. The Punjab police have been repeatedly accused of staging "fake encounters" with militants. There were also frequent accusations that militants were captured, tortured and then executed. Human rights violations similarly have been ignored in Kashmir. The government of India simply ignored these complaints, and it repeatedly refused to permit international human rights organizations to enter the country.

This wilful neglect of human rights may well return to haunt the government of India. In the 1980s, a number of Sikh intellectuals chose to advance the Khalistan cause by establishing human rights organizations. These organizations have been very adept at exploiting human rights violations to embarrass the government of India, both domestically and internationally. While the conflict has subsided on the battlefield, it seems that "Punjab Part Two" will be fought in the courts. Instead of AK-47s, the "litigation gun" is now firmly trained on the police.⁸⁴

Over 1,000 writ petitions have been logged against the police with the Punjab and Haryana courts, involving more than 2,000 police personnel. At present, "[t]hirty policemen are in jail, around 100 are out on bail and 140, including seven [superintendents], are facing prosecution."⁸⁵ There are also 85 active investigations against the Punjab police being pursued by the Central Bureau of Investigation and 91 judicial probes, including "cases involving partially identified or unidentified bodies, mass cremations, and disappearances from police custody."⁸⁶ A senior police officer fears that "[n]o less than one-sixth of the 70,000 strong force may find itself in the dock."⁸⁷ For a police force that "suffered 1,700 casualties between 1978 and 1993 and lost 800 family members in the war to keep Punjab in India, this seems like ingratitude at its worst."⁸⁸

The previous Congress government of Punjab funded a large war chest to provide the police with the best legal representation possible. The new Akali Dal administration is not as keen to protect the police, given the rather low esteem in which the force is viewed in the Sikh community. The Punjab Advocate General, G.S. Grewal, has said only that "[o]nce they are exonerated, the state is not averse to reimbursing the legal expenses incurred by the policemen."⁸⁹ The national Congress party is obviously no longer in a position to provide the police with immunity. While the Indian courts are notoriously slow, there is potential for the moral fabric of the police to unravel. The lower ranks, which cannot afford expensive legal council, may find it preferable to testify against their commanding officers rather than risk prosecution.⁹⁰

India, however, cannot afford to have the police command structure collapse. India is a highly volatile, diverse, developing society. Political disturbances are certain to be a recurring feature of the Indian landscape, and the security forces will be asked again to contain such movements. In defence of his force, K.P.S. Gill, now retired, has argued, "a mechanism must be found to obviate the legal harassment of those who put their lives at stake during low-intensity conflicts . . .. Otherwise, who is going to fight terrorism tomorrow?"⁹¹ Criminal immunity for human rights violations, however, is not the solution. Counter-insurgency efforts must be conducted within the letter and spirit of the law and under firm civilian political command. India's counter-insurgency strategy, which has relied on excessive force and disregard for human rights, is not sustainable in a political democracy, with constitutionally guaranteed liberties and an open legal system.

Civil-Military Relations

The Indian army, almost uniquely in the Third World, has remained under firm civilian control since independence. Rajesh Rajagopalan has noted that the Indian army developed its counter-insurgency doctrine during the conflict in Nagaland in the 1950s. He suggests that the main points of the doctrine include limits on the use of force, isolation of insurgents from the general population, military dominance of insurgency-affected areas and superiority of numbers.⁹² Although the army has engaged in a number of counter-insurgency operations since that time, it has not fundamentally revised its counter-insurgency doctrine. In short, the army does not seem to have responded to the modernization of insurgency movements, with their significantly improved weapons and communications equipment for the insurgents, and significant external support for the enemy, in the form of both regional sanctuary and international support networks. In short, the military has maintained its conventional "positional-war" orientation.

While India suffered a humiliating defeat against China in 1962, the Indian military performed admirably in conflicts with Pakistan in 1948, 1965 and especially the 1971 liberation of Bangladesh. The military thus developed a certain prestige in Indian society. The military has not engaged in an external war since 1971, notwithstanding the skirmish with Pakistan in summer 1999, but it has been employed frequently to suppress domestic uprisings. Between 1951 and 1970, the army was asked to suppress domestic strife on 476 occasions. By contrast, the military engaged in 433 domestic situations between June 1979 and 1984. "Most of these interventions were limited and . . . [i]n this sense, the military has acted in support of the political structure, providing the ultimate force in situations in which political solutions had failed and in which the police could not cope."⁹³ The increased deployment of the military for domestic peacekeeping has lowered its prestige in society, disturbed ethnic relations in the military and strained relations between senior military officers and civilian authorities.

The Indian military is a social institution. It is composed of individuals drawn from India's diverse communities. India's diversity is thus mirrored in the military, to a greater or lesser degree. When the military is asked to suppress a community-based insurgency, it is bound to disturb members of that community in the military. Soldiers are not divorced from their community and, if a counter-insurgency effort steps beyond eliminating terrorists to attacking the community or its cherished institutions, a soldier's loyalty may be called into question. The Punjab crisis is illustrative of this problem.

The Sikh community was regarded as the backbone of the British Indian army. While Sikhs only composed about 2 percent of India's population, the military was 25 percent Sikh at independence. Although the officer corps is still about 20 percent Sikh, the overall figure has dropped to about 12 percent, much to the consternation of Sikh political leaders. Sikhs in the military are as loyal as any other soldier, and they had little or no sympathy for the separatists in their community, but Operation Bluestar severely tested the loyalty of many Sikh soldiers.⁹⁴ Even K.P.S. Gill has said bluntly, "No one defends Bluestar. It was a grave mistake and everyone, every Sikh had the same reaction. There was a sense of outrage. Even I had it."⁹⁵

Three of the key generals who planned Bluestar were Sikh, as were four of the six battalion commanders who participated in the operation, and a fair number of Sikh troops were also involved.⁹⁶ The President of India, the constitutional Commander-in-Chief of the military, was also a Sikh. These loyal soldiers were instantly regarded as pariahs in their own community, and they lived with the threat of terrorist retaliation after the operation. Indeed, General A.S. Vaidya, who commanded Operation Bluestar, was assassinated on 10 August 1986, six months after he retired from active duty.⁹⁷ And President Giani Zail Singh was struck by a bullet when he toured the Golden Temple in the aftermath of the operation.

As news of the operation spread across India, over 2,500 Sikh recruits deserted the army in a desperate attempt to defend the Golden Temple.⁹⁸ Operation Bluestar placed every Sikh soldier in a moral quandary: they were asked to submerge their cultural identity for the sake of professional duty. Stephen Cohen has argued,

the Punjab crisis had one unprecedented impact upon the Indian armed forces. Given the evidence of the mutinies that occurred in June 1984, the temporary alienation of retired Sikh officers, and the close links between Sikhs in and out of the military, one can assume that no Sikh unit was fully trusted, especially in a situation that involved the Punjab itself . . .. The overall integrity of the Indian armed forces, especially the army, may have been badly, if temporarily, weakened.⁹⁹

The stress placed on the military's delicate ethnic composition during counter-insurgency operations is yet another reason to pursue political settlements to these problems.

The deployment of the military for counter-insurgency duty also strains the relations among senior military commanders and the political decision makers. The Indian military regards its primary task as the defence of India from external aggression, but "[a] third and occasionally more of the army is employed on internal security duties."¹⁰⁰ India and Pakistan share a long border, much of it over flat land with large civilian populations nearby. A large contingent of soldiers is thus required to guard the border. Furthermore, the defence of the isolated borders with China also requires an enormous logistical effort. The military regards domestic counter-insurgency duty as an unnecessary diversion from its primary task.¹⁰¹ The military has thus increasingly asserted that domestic insurgencies require political resolutions.¹⁰² Indeed, Lieutenant-General J.R. Mukherjee, commander of the 15th Corps in Kashmir, stated in July 2000, "I think it has been accepted by all that ultimately there would have to be a political solution" to the crisis in Kashmir.¹⁰³

The military's engagement in counter-insurgency places it in the middle of a political conflict between the centre and the disaffected region. As the military is necessarily an instrument of the centre, it stands to lose social prestige in disturbed regions. The military may well resent this loss of public support, especially if the political maladroitness of the centre has caused or perpetuated the conflict. Cohen has observed, "[s]ervice resentment over political incompetence, especially when that incompetence affects the conduct of war or the readiness of the military to fight a war, runs deep . . .. The armed forces, especially the army, are not passive observers of the deterioration and increasing lawlessness of Indian politics."¹⁰⁴ With all of the problems and crises facing the government of India, it can ill-afford a disgruntled military. Cohen has cautioned that the obedience of the military to civilian authority "will continue only so long as that authority is regarded as legitimate."¹⁰⁵ And therein lies the greatest danger of India's erstwhile counter-insurgency strategy.

Democracy in India

India's legitimacy is derived from its democratic traditions, which are rare among Third World states. Indeed, India's democratic history is more in keeping with western political traditions. Furthermore, the military in India has been wholly subservient to civilian authority. While India's security forces may be able to ensure the compliance of the population, the continued "use of force to suppress dissent, resolve social conflicts, and maintain order may lead to the erosion of legitimacy and undermine the capacity of the state to rule."¹⁰⁶ As Noorani notes, furthermore, "[d]emocracy is in peril when intelligence agencies mould policy and political decisions."¹⁰⁷ V.R. Raghavan has noted that "India's central paramilitary forces have expanded four times and more in the last 20 years." He continues, "that the Indian state is required to use such a large force, in an armed role against its own citizens, should be a sobering thought."¹⁰⁸ Raghavan also laments that all too often the central and state governments have been content to allow large portions of the country to be essentially governed by the military and paramilitary forces. Rajni Kothari has described the trend toward police rule and authoritarianism as "the state against democracy."¹⁰⁹

India's democracy is now under enormous strain. The fragmentation of the party system, unstable coalition governments, dubious exploitation of the constitution, massive graft and political corruption, the criminalization of politics and the general decay of many institutions have left many Indians deeply suspicious and distrustful of the political system. A pervasive cynicism has taken root across India, especially in highly disaffected states, such as Punjab and Kashmir. The citizens of Punjab and Kashmir first witnessed the dismissal of their democratically-elected governments. They then endured a decade of police rule followed by sham elections. The governments elected by these charades were the political puppets of New Delhi, not the representatives of the people. It would not be surprising if the citizens of these states have had their belief in democracy shaken and their faith in the central government diminished. A similar process of state delegitimation has occurred in Assam and the micro-states of the northeast frontier.

CONCLUSION

Punjab and Kashmir are relatively small states on the periphery of India. The country may well endure these crises, but it can ill-afford more crises. The government of India cannot alienate large portions of the population in successive states, if it wishes to maintain its legitimacy. The repeated use of the counter-insurgency strategy developed in Punjab and Kashmir could well undermine the democratic fabric of India. The long-term erosion of the political legitimacy of the Indian state is the real danger of India's counter-insurgency strategy.

India will likely suffer more separatist insurgencies in the years ahead. It is thus imperative that an effective counter-insurgency strategy be developed that avoids the pitfalls encountered by the efforts in Punjab and Kashmir. Force will undoubtedly be a part of this strategy, but it must conform to the letter and spirit of the law, if it is to obtain public acceptance. The frequent use of excessive force - extra-judicial executions and torture - has serious costs, in the short, medium and long-term. In the short-term, it enrages the local population and fuels militancy; in the medium term, it exposes the police to damaging legal suits, the possible unravelling of the command structure, the demoralization of the police and a possible reluctance to fight future insurgencies; in the long-term, it may cause the erosion of the state's political legitimacy.

India's erstwhile counter-insurgency strategy has been based overwhelmingly on the use of force; it has been almost wholly apolitical. The government of India has steadfastly ignored the political grievances that cause such crises. All too often, legitimate political issues have been allowed to fester. This leads only to greater complications. The Indian state must win back the hearts and minds of its disaffected populations instead of simply trying to enforce compliance. Winning the confidence of the people is the only real long-term solution to these protracted crises.

Endnotes

1. The battle of Kargil in June and July 1999 was, for all intents and purposes, an inter-state conflict between India and Pakistan.
Return to article

2. Paul Brass, The Politics of India Since Independence (Cambridge: Cambridge University Press, 1990), pp. 149-50.
Return to article

3. These provisions were effected by the Sixteenth amendment of the Constitution, which was adopted in response to Tamil nationalists who flirted with separatist notions in the early 1960s. The oath of office is contained in the Third Schedule.
Return to article

4. See B.D. Dua, Presidential Rule in India (1950-1974): A Study in Crisis Politics (New Delhi: S. Chand and Company, 1979). The incidents of President's Rule declined in the 1990s, as India adopted a more federal political culture.
Return to article

5. Robert Hardgrave Jr. and Stanley Kochanek, India: Government and Politics in a Developing Nation, 6th ed. (New York: Harcourt Brace Jovanovich College Publishers, 2000), p. 225.
Return to article

6. Ibid., pp. 223-25. The Terrorist and Disruptive Activities Act was repealed in 1995 after considerable protest from civil liberty groups, although those previously charged under the TADA Act are still being prosecuted.
Return to article

7. Harish Puri, et al., Terrorism in Punjab: Understanding Grassroots Reality (New Delhi: Har- Anand Publications, 1999), p. 116.
Return to article

8. In the days following Mrs. Gandhi's death, some 3,000 Sikh men were slaughtered by roving goondas in the streets of New Delhi. There is evidence that elements in the Congress party orchestrated some of this violence. The Delhi massacre furthered the alienation in the Sikh community.
Return to article

9. Hardgrave and Kochanek, India, p. 229.
Return to article

10. Shekhar Gupta and Kanwar Sandhu, "K.P.S. Gill, True Grit," India Today (Special North American Edition), 15 April 1993, p. 38. Gill was Director-General of the Punjab Police, who succeeded in finally terminating the Punjab insurgency.
Return to article

11. The other issues included the division of waterways for irrigation and some minor territorial adjustments.
Return to article

12. See Praveen Swami, "Strategic Shift?" Frontline 17, no. 11 (27 May-9 June 2000), http://www.frontlineonline.com/archives.htm; and Ramesh Vinayak, "Testing the Waters," India Today International, 22 May 2000, pp. 19-21.
Return to article

13. See Praveen Swami, "A Cease-fire in Kashmir," Frontline 17, no. 16 (5-18 August 2000), http://www.frontlineonline.com/archives.htm; and Ramesh Vinayak and Harinder Baweja, "Kashmir's Bloody Puzzle," India Today International,
Return to article

14 August 2000. 14. See Shujaak Bukhari, "Massacres Rattle Kashmir," The Hindu, Thursday, 3 August 2000, http://www.the-hindu.com.
Return to article

15. See Shujaak Bukhari, "Centre, Hizbul Commanders Hold Talks," The Hindu, Friday, 4 August 2000, http://www.the-hindu.com; B. Muralidhar Reddy, "Salahuddin Sets August 8 Deadline for Tripartite Talks," The Hindu, Friday, 4 August 2000, http://www.the- hindu.com; Praveen Swami, "Terror Unlimited," Frontline 17, no. 17 (19 August-1 September 2000), http://www.frontlineonline.com/archives.htm; Ramesh Vinayak, "Peace takes a Knock," India Today International, 21 August 2000.
Return to article

16. See Ramesh Vinayak, "Mission Kashmir," India Today International, 4 December 2000; Praveen Swami, "Another Cease-fire," Frontline 17, no. 24 (25 November-8 December 2000).
Return to article

17. The Afghan conflict has turned northern Pakistan into one of the best arms markets in the world for terrorists. See Mary Anne Weaver, "The Children of the Jihad," The New Yorker, 12 June 1995.
Return to article

18. When militancy resumed in Punjab in 1986, the government of India tried to cordon off the entire state. For a time, all visitors to Punjab were required to obtain special permission to visit the state and even transit documents were required to travel through the state. These provisions were eventually dropped and never even imposed in Kashmir.

19. Dilip Awasthi, "Terrorism: Expanding the Turf," India Today, 30 September 1991. Awasthi notes that when the security forces initially tightened their cordon on Amritsar and Gurdaspur some Khalistani militants escaped and sought sanctuary in neighbouring Haryana and the Terrai region of Uttar Pradesh.
Return to article

20. Rajesh Rajagopalan, "'Restoring Normalcy': The Evolution of the Indian Army's Counterinsurgency Doctrine," Small Wars and Insurgencies 11, no. 1 (Spring 2000).
Return to article

21. See Amnesty International, "Human Rights Violations in Punjab: Use and Abuse of the Law," (1991); and Kanwar Sandhu, "G.S. Manochahal: How the Trap was Laid," India Today (Special North American Edition), 31 March 1993, pp. 56-7.
Return to article

22. Gupta and Sandhu, "K.P.S. Gill: True Grit," p. 40.
Return to article

23. For a brief description of the police infiltration of militant groups in Punjab, see Puri, et al., Terrorism in Punjab, p. 131.
Return to article

24. Ramesh Vinayak, "Kashmir: Sliding into Gloom," India Today, 31 January 1996, p. 42.
Return to article

25. Gupta and Sandhu, "K.P.S. Gill: True Grit," p. 38.
Return to article

26. Ramesh Vinayak, "The New Vigilantes," India Today International, 11 October 1999, pp. 2-23.
Return to article

27. Ibid., p. 23.
Return to article

28. Ibid., p. 23.
Return to article

29. The caretaker government of Chandra Shekhar reportedly reached a secret agreement with certain militant factions to hold the election, if they agreed not to turn it into a referendum for Khalistan. See Gurharpal Singh, "The Punjab Elections of 1992: Breakthrough or Breakdown?" Asian Survey 32, no. 11 (November 1992), p. 990.
Return to article

30. The 1991 federal election was staggered over a number of weeks, during which time Rajiv Gandhi, the leader of the Congress party, was assassinated. This created a sympathy wave for the Congress in the subsequent balloting. By the time of the scheduled state election in Punjab, which the Congress had boycotted, it was apparent that the Congress would win the federal election, although the ballots had not yet been counted. While the Election Commission cited the violence in Punjab, including the death of 29 candidates, for the cancellation of the state election, Hardgrave and Kochanek suggest that "[t]he action [of the Election Commission] was taken in all likelihood at the behest of the Congress (I), which had boycotted the elections and that, had elections been held, would have been closed out of [Punjab] politically." India, p. 167.
Return to article

31. Ramesh Vinayak and Harinder Baweja, "Punjab Elections: Accelerating Alienation," India Today, 15 March 1992, p. 49.
Return to article

32. Gurharpal Singh, "The Punjab Elections 1992," p. 992.
Return to article

33. Vinayak and Baweja, "Punjab Elections," pp. 42, 44.
Return to article

34. Gill was once quoted as saying, "[w]hat is happening here in Punjab is purely between Jat Sikh (militants) and Jat Sikh (Punjab Police)." See Gupta and Sandhu, "K.P.S. Gill: True Grit," p. 37. Gill resolved to solve the problem of militancy in the time-honoured, "wild-west" tradition of Punjab - a fight to the death.
Return to article

35. Harcharan Bains, "Punjab: A New Phase," Hindustan Times, 5 October 1994.
Return to article

36. The Babbar Khalsa immediately claimed responsibility, although other militants are now thought to be responsible. The prime suspects are now thought to be Daljit Singh Bittoo and Jagtar Singh Hawara. See, Harinder Baweja, et al., "Punjab: Bitter Harvest," India Today, 16 June 1997, p. 39.
Return to article

37. Gurharpal Singh, "The Punjab Elections 1992," pp. 996-99.
Return to article

38. As mentioned above, the central government had little difficulty amending the constitution to extend President's Rule in Punjab beyond the normal constitutional limit of six months.
Return to article

39. Reeta Chowdhari Tremblay, "Nation, Identity and the Intervening Role of the State: A Study of the Secessionist Movement in Kashmir," Pacific Affairs 69, no. 4 (Winter 1996/97); Sumit Ganguly, "Explaining the Kashmir Insurgency: Political Mobilization and Institutional Decay," International Security 21, no. 2, (Fall 1996).
Return to article

40. Harinder Baweja, "Kashmir: A Risky Gamble," India Today, 31 May 1996, p. 109.
Return to article

41. Harinder Baweja, "Kashmir: Voting Under Coercion," India Today, 15 June 1996, pp. 68-70.
Return to article

42. Ibid., p. 68. Given the army's role in ensuring the voter "turn-out," we might better call it the voter "turf-out."
Return to article

43. Manoj Mitta and Smruti Koppikar, "Centre-State Relations: Changing Equations," India Today, 30 June 1996, pp. 80-83.
Return to article

44. Ramesh Vinayak, "Jammu and Kashmir: First Step into the Future," India Today, 31 October 1996, p. 68.
Return to article

45. Ibid., p. 68.
Return to article

46. See Ramesh Vinayak and Harinder Baweja, "Faltering Farooq," India Today International, 7 February 2000, pp. 10-16; see also, "India Today Editorial: Militancy's Second Innings," India Today International, 20 September 1999, p. 5. In the March 1998 federal election the BJP emerged again as the largest single party with 177 seats, as opposed to 161 seats in 1996. However, the BJP increased its allies from 26 to 73 for a total of 250 seats, which allowed it to form a minority government. A new election was called in the summer of 1999, when the BJP government lost a vote of confidence by a single vote. In the September 1999 election, the BJP was returned to power with a 296 seats, including allies. While the BJP still has 12 coalition partners, this government appears to be more stable.
Return to article

47. "Militancy's Second Innings," p. 5.
Return to article

48. "Timely Waiver," India Today, 4 August 1997, p. 17.
Return to article

49. It is difficult to estimate accurately the number of insurgents in each state. The Indian media frequently reports official or unofficial intelligence information on this matter, but the reliability of this information is uncertain. My estimate of 6,000 insurgents in each state is probably high; I just wish to note that to quell an insurgency of this size requires an extraordinary number of state security personnel.
Return to article

50. See Harinder Baweja, "War by Proxy," India Today International, 26 July 1999, pp. 32-33.
Return to article

51. Ramesh Vinayak, "Kashmir: Sliding into Gloom," India Today, 31 January 1997, p. 42.
Return to article

52. Vinay Kumar, "Swap, a Blow to Security Forces: Experts," The Hindu (On Line Edition), 2 January 2000, http://www.the-hindu.com.
Return to article

53. Ibid.
Return to article

54. Harinder Baweja and Ramesh Vinayak, "Reaction Plans," India Today International, 31 January 2000, p. 12.
Return to article

55. Harinder Baweja, et al., "New Targets," India Today International, 23 August 1999, p. 32.
Return to article

56. Harinder Baweja, "War by Proxy," p. 32.
Return to article

57. Ramesh Vinayak, "Valley of Fear," India Today International, 10 January 2000, p. 16.
Return to article

58. Baweja and Vinayak, "Reaction Plans," p. 12.
Return to article

59. After leaving Nepal, and touching down in Amritsar, Lahore and Dubai, the aircraft finally arrived in Kandahar, Afghanistan, the home base of the ruling Taliban regime.
Return to article

60. Ahmad Umar Sayeed Sheikh, who kidnapped four tourists in 1994, was the third man released by India. He is the least well-known of the three. He is a British national, who was born in Pakistan but raised in England, where he attended first class private schools before going to the London School of Economics. It is suspected that he is also a member of the Harkat ul-Ansar.
Return to article

61. Jason Burke and Harinder Baweja, "We Should Now Target . . .," India Today International, 4 October 1999, p. 29.
Return to article

62. Ibid., p. 33.
Return to article

63. Baweja and Vinayak, "Reaction Plans," p. 11.
Return to article

64. Ibid., p. 11.
Return to article

65. Vinayak and Baweja, "Faltering Farooq," p. 12.
Return to article

66. A.G. Noorani, "A Report on Kashmir," Frontline 17, no. 17 (19 August-1 September 2000), http://www.frontlineonline.com/archives.htm.
Return to article

67. Ramesh Vinayak and Rohit Parihar, "Punjab: Grim Signals," India Today, 21 July 1997, p. 32.
Return to article

68. Ramesh Vinayak, "Ominous Intent," India Today International, 15 March 1999, p. 28-29. It has been reported that the police recovered over 300kg of explosives in Punjab in 1997-98. See Ramesh Vinayak, "A SAD Affair," India Today International, 28 December 1998, p. 27.
Return to article

69. Vinayak, "Ominous Intent," p. 32; Ramesh Vinayak, "Militancy: Dwindling Menace," India Today, 28 February 1997, p. 59.
Return to article

70. Ramesh Vinayak, "Punjab: Return of Terror," India Today, 4 August 1997, p. 54; Vinayak and Parihar, "Punjab: Grim Signals," p. 33.
Return to article

71. Vinayak, "Ominous Intent," p. 28.
Return to article

72. These groups include the Babbar Khalsa, the Khalistan Commando Force (Panjwar), the Khalistan Commando Force (Zaffarwal), the Khalistan Liberation Front (Lahoria), the Khalistan Liberation Front (Sekhon), the Khalistan Zindabad Force, the Dal Khalsa and the International Sikh Youth Federation. The hitherto unknown Shaheed Khalsa Force has claimed responsibility for the bomb explosions in Delhi in October-November 1997. It is believed that the movement is being sustained primarily by the Babbar Khalsa, which is particularly strong in the expatriate Sikh community abroad.
Return to article

73. A central feature of Gurdwaras (Sikh temples) is that they all have communal kitchens and sleeping quarters for any person who wishes to stay. Even if one cannot make a financial contribution, they are welcome to stay and contribute by working in the kitchen and around the temple. Illegal immigrants, who are unable to work in the larger community, tend to congregate at the temples. Gurdwaras are thus an excellent recruiting ground for militant groups. Indeed, it has been reported that a reward of 50 million rupees for the killing of Punjab's Chief Minister, Parkash Singh Badal, was announced in a Toronto gurdwara in January 1999. See Vinayak, "Ominous Intent," p. 29.
Return to article

74. Quoted from Baweja, et al., "Punjab: Bitter Harvest," p. 39.
Return to article

75. See Hamish Telford, "The Political Economy of Punjab: Creating Space for Sikh Militancy," Asian Survey 32, no. 11 (November 1992); Paul Brass, "The Punjab Crisis and the Unity of India," in Atul Kohli, India's Democracy: An Analysis of Changing State-Society Relations (Princeton, NJ: Princeton University Press, 1988); Joyce Pettigrew, The Sikhs of Punjab: Unheard Voices of State and Guerilla Violence (London: Zed Books, 1995); Shinder Purewal, Sikh Ethnonationalism and the Political Economy of Punjab (London: Oxford University Press, 2000); Robert Wirsing, India, Pakistan, and the Kashmir Dispute: On Regional Conflict and Its Resolution (New York: St. Martin's, 1994); Sumit Ganguly, The Crisis Kashmir: Portents of War, Hopes of Peace (New York: Cambridge University Press, 1997).
Return to article

76. Brass, "The Punjab Crisis and the Unity of India," p. 204.
Return to article

77. The territorial issues stemmed from the linguistic reorganization of Punjab and the creation of Haryana in 1966. The division of waterways is essential for irrigation in these highly agricultural states.
Return to article

78. Gurharpal Singh, "The Punjab Problem in the 1990s: A Post-1984 Assessment," Journal of Commonwealth and Comparative Politics 29, no. 2 (July 1991); Gurharpal Singh, "The Punjab Crisis Since 1984: A Reassessment," Ethnic and Racial Studies 18, no. 3 (July 1995).
Return to article

79. The Akali Dal has historically been beset with factionalism between its parliamentary and organizational wings, and also with the Shiromani Gurdwara Prabandhak Committee (SGPC), which is an elected assembly that manages Sikh temples and religious affairs. When the Akali Dal forms the government of the state, those "in" government obtain leverage over those "out" of government. Unfortunately, the "outs" have a dangerous habit of courting militants, or giving them tacit support, to counter-balance the power of the "ins." With the 500th anniversary of the Sikh religion and the 300th anniversary of the Khalsa (the dominant "order" within the religion) in 1999, there was fierce competition between the state government and the SGPC. See Ramesh Vinayak, "A Sad Affair;" Vinayak, "Agony and the Akali," India Today International, 8 February 1999, p. 30-31; Vinayak, "Clash of the Kirpans," India Today International, 22 February 1999, p. 14-15; Vinayak, "Lady Luck," India Today International, 29 March 1999, p. 25-26; and Vinayak, "Badal at Bay," India Today International, 15 November 1999, pp. 30-31.
Return to article

80. See Hamish Telford, "The Federal Process in India: From Domineering to Conflictual Federalism," in Elliot Tepper and John Wood, eds., Enriched by South Asia: Celebrating 25 Years of South Asian Studies in Canada, vol. 2 (Montreal: Canadian Asian Studies Association, 1994). See also A. G. Noorani, "Article 370: Law and Politics," Frontline 17, no. 19 (16-29 September 2000), http://www.frontlineonline.com/archives.htm.
Return to article

81. Quoted in Vinayak and Baweja, "Faltering Farooq," p. 12. Vinayak and Baweja argue that Abdullah has launched this latest campaign to extract more financial resources from the centre and to disguise his "non-performance" as Chief Minister.
Return to article

82. In late June 2000, Abdullah's National Conference government endorsed an "autonomy resolution," which sought to return Kashmir to its pre-1953 status whereby the state government would have jurisdiction for all matters except foreign affairs, defence and communications. Abdullah was motivated by the centre's moves to establish a dialogue with the militants. While the resolution was congruent with the provisions of Article 370, the central government rejected it. Addullah responded to the centre's stance by saying, "don't try and reduce me to zero when we are the only ones who swear by the Indian Constitution. At least I am not talking of jehad [holy war] and azadi [freedom] like the Hurriyat Conference." Quoted in "Don't Reduce me to Zero," India Today International, 10 July 2000, p. 19. See also Praveen Swami, "The Autonomy Demand," Frontline 17, no. 14 (8-21 July 2000), http://www.frontlineonline.com/archives.htm; and "Autonomy for Jammu and Kashmir," Frontline 1, no. 14 (8-21 July 2000), http://www.frontlineonline.com/archives.htm. By waiting 11 years to start negotiations, the government has placed itself on the horns of a dilemma. Democratic politicians like Abdullah have had their credibility reduced, while the militants have become politically entrenched. Abdullah has no truck with the militants, and consequently a settlement with Abdullah, or any other democratic politician, likely will not succeed in quelling the disturbance. This is a direct consequence of India's counter-insurgency strategy.
Return to article

83. Baweja and Vinayak report that "[t]here are critical issues like beefing up the administration, winning over the local population and resettling the Pandits [Kashmiri Hindus] that have been completely ignored." See "Reaction Plans," p. 12.
Return to article

84. Baweja, et al., "Punjab: Bitter Harvest," p. 39.
Return to article

85. Ramesh Vinayak, "Punjab Police: Haunted by the Past," India Today, 9 June 1997, p. 42.
Return to article

86. Ibid., p. 42.
Return to article

87. Ibid., p. 42.
Return to article

88. Baweja, et al., "Punjab: Bitter Harvest," p. 36.
Return to article

89. Vinayak, "Punjab Police: Haunted by the Past," p. 43.
Return to article

90. Baweja, et al., "Punjab: Bitter Harvest," p. 36.
Return to article

91. Vinayak, "Punjab Police: Haunted by the Past," p. 43.
Return to article

92. Rajagopalan, "'Restoring Normalcy,'" pp. 53-54.
Return to article

93. Kuldip Mathur, "The State and the Use of Coercive Power in India," Asian Survey 32, no. 4 (April 1992), pp. 345-46.
Return to article

94. Some retired Sikh officers lent their support to the Akali Dal in the early 1980s and a few actively joined the militants, including Major-General Shubeg Singh, a decorated soldier from the 1971 campaign against Pakistan. Shubeg Singh organized the Mukhti Bahini insurgency in Bangladesh and he is thought to be responsible for the militant fortification of the Golden Temple.
Return to article

95. Interview, India Today (Special North American Edition), 15 April 1993, p. 43; emphasis added.
Return to article

96. Shekhar Gupta, "Operation Bluestar," India Today, 30 June 1984, p. 27.
Return to article

97. "The Killing of a General," India Today, 31 August 1986.
Return to article

98. Stephen Cohen has argued, "[m]any Indians were surprised at the Sikh mutiny, but they should not have been. All Indian army regiments bind their members with a religious as well as civil oath, a practice that the British merely carried over from earlier Hindu, Sikh, and Muslim armies. Sikh soldiers are accompanied into battle by unit priests and the Granth Sahib [the Sikh holy book]. The perceived desecration of their religion signalled a call to arms because of their oath, not in contradiction to it." Stephen Cohen, "The Military and Indian Democracy," in Atul Kohli, ed., India's Democracy: An Analysis of Changing State-Society Relations (Princeton, NJ: Princeton University Press, 1988), p. 135; emphasis original.
Return to article

99. Ibid., p. 137.
Return to article

100. V. R. Raghavan, "Internal Security's Diminishing Returns," The Hindu, Friday, 4 August 2000, http://www.the-hindu.com.
Return to article

101. Cohen, "The Military and Indian Democracy," p.110.
Return to article

102. Rajagopalan, "'Restoring Normalcy,'" p. 59.
Return to article

103. Quoted in Noorani, "A Report on Kashmir."
Return to article

104. Cohen, "The Military and Indian Democracy," p. 123.
Return to article

105. Ibid., p. 143.
Return to article

106. Mathur, "The State and the Use of Coercive Power in India," p. 337.
Return to article

107. Noorani, "Report on Kashmir."
Return to article

108. Raghavan, "Internal Security's Diminishing Returns."
Return to article

109. Rajni Kothari, The State Against Democracy: In Search of Humane Governance (New Delhi: Ajanta Publications, 1988).
Return to article