Terrorism Update
Show/Hide Search
    Click to Enlarge

Early Warning & Light Weapons
Tara Kartha*

The past decade has witnessed a rapid escalation in the intensity and number of conflicts and incidents of weapons-related violence in South Asia. In the early 1980s, conflict was limited to small militant movements in Kashmir and India’s Northeast, while Pakistan was only beginning to cope with the incidence of ethnic violence. Discontented elements relied largely on the .303 or a few pistols, and with these they had to be content. Today, militant groups have been able to access weapons of a higher technology and lethality than available with the police or even the armed forces of a country. At a secondary level, weapons have begun to diffuse outwards into society itself, fuelling ethnic, religious and sectarian conflict. Pakistan is reeling under a huge spread of more than five million guns of all kinds, while Bangladesh, Bhutan and Nepal are seeing a rising incidence of weapons-related violence. Various militant groups in India have, similarly, been able to inflict higher levels of casualties primarily due to access to rapid-fire weapons, explosives and rockets. In Sri Lanka, the war drags on with no end in sight, as the Tamil Tigers spring back to action, armed with newer and better weapons.

As the bureaucracy and political leaders react strongly to the rise of terrorism and sponsored violence, the average observer usually wants an answer to the one question that is absurdly simple at one level, and extremely complex in its answers at another – did the ‘authorities’ not learn early enough about the incoming weapons? If they did, why did they not react in time? The question is a logical one, and requires to be examined in detail. This paper tries to answer these and other questions, with the aim of providing a possible framework of analysis that would provide policy makers with a set of tools for early warning of impending violence. It takes the stand that, while even an unarmed mob can cause a certain degree of damage, violence escalates to the level of a national security issue when weapons are brought into a conflict situation. The police retreats in the face of such weaponry, which they are ill-prepared to counter, and the army is brought in. A conflict is born, that no one then really knows how to end. The paper has a narrow focus, and concentrates on the weapons aspect rather than the societal, political, or other factors that push a disagreement into a full-fledged conflict.

Light weapons and Conflict in South Asia

South Asian conflicts, have in the past, suffered from an excess of early warnings of impending friction, rather than any lack of these. This is especially apparent in the area of weapons build up, proliferation and diffusion. However, since little was done to avert the slide down the slippery slope of conflict, it is as near a perfect case as possible to study why the various actors who were monitoring the conflict – the journalists, bureaucrats, police forces and others – were unable to stop the slide. Yet, had these inputs been disseminated, assessed and acted upon promptly, thousands of lives would have been saved, and the present diffusion of weapons into society avoided.

Different academic and multilateral organisations are now working towards a ‘solution’ – be it control of further proliferation, retrieval of the existing excess, or confronting the problem from the point of view of conflict itself. Whatever be the persuasion of different analysts, the first lesson that is abundantly clear from ongoing efforts is that weaponisation of any internal disagreement, be it ethnic, religious or political, leads inevitably to a cycle of events that creates tremendous regional instability and adds to the further diffusion of global violence. Huge flows of refugees, trans-national crime, narcotics and weapons trafficking, and human rights violations are now typical of conflict areas. Ultimately, it is the state itself that is the prime victim, and when it fails, there is clearly no one involved who can claim advantage. The tragedy of states like Afghanistan, Kosovo, Angola or the Republic of Congo is testimony to the kind of havoc that a proliferation of light weapons can lead to. It has often been argued that large democracies like India can weather such a diffusion of weapons without it becoming a major security issue. This mindset has led to the relegation of the twin issues of militancy and weapons proliferation into the background. This argument fails completely on closer examination.

The ‘light weapons argument’

The proliferation of light weapons has led not only to human tragedy in terms of more than 30,000 lives lost, but has also had its impact on India’s defence and foreign policy. The armed forces have had to bear the brunt of militancy for more than a decade leading to troop demoralisation and exhaustion. This is true of any country that has had to go through sustained anti-militancy operations. Though it may wish to reduce troop strength, the army cannot consider such a move unless manpower-intensive anti-militancy operations are at least substantially reduced. This inevitably affects not only overall defence expenditure, but also how the monies allotted are spent – in short, less and less of the cake is available for redressing the crying need for new equipment, as the pay and pension bill expands. Foreign policy is affected as the government is invariably forced into a defensive posture on anti-militancy operations, while efforts at regional co-operation are often politically difficult to push through, as ‘enemy’ images are built up, even with regard to small states like Bangladesh. Constant accusations of cross-border assistance to militant groups and threats of hot pursuit do little to promote regional good will, which is undoubtedly the most desirable climate for a country that seeks to achieve developmental goals.

At another level, it must be realised that a large democracy espousing secular values is immensely vulnerable to sponsored militancy and terrorism.

Firstly, India’s very size makes border management an immensely challenging task. The borders cannot be ‘sealed’, regardless of what the media regularly reports. Arising from this, it follows that a weaponised neighbourhood is an immediate threat to the nation’s security.

Secondly, while espousing the values of secularism, these secular values are constantly put under test by the threat of religious and ethnic groups demanding more representation or autonomy. Weapons infusion into hundreds of pockets of dissension imposes an unbearable strain on national resources.

Thirdly, and again arising from this, such proliferation of weapons inevitably leads to a curtailment of democratic rights in affected areas (the imposition of the Armed Forces Special Power Act and the Disturbed Areas Act are just two instances of such constraints). The basic value systems of a democratic polity are consequently put under strain, even as policy at vital levels is adversely affected.

Why early warning?

Light weapons obviously affect security in multifarious ways. It is, consequently, logical to seek to limit their spread as quickly and efficiently as possible.

This, however, is easier said than done. The first lesson that control analysts working on these issues have learnt is that once weapons are out of state hands, the difficulties of getting them back are simply enormous. While there has been some limited success in ‘gun buy-backs’ in the US and Australia, the fact remains that the weapon will not be exchanged for money or land, nor be given up as long as the owner sees a higher profit in continued ownership, or sees it as necessary for his continued existence. Where security is non-existent or is at risk, as is the case in most conflict areas, any moves at retrieval of weapons will fail. This unfortunate lesson is coming across more and more clearly in the conflict zones of Africa where militancy has been replaced by rampant crime (South Africa), or where fresh eruptions of violence have taken place after one phase of conflict has formally ended (Congo, Angola). In Bosnia, heavily armed peacekeepers have so far been able to keep the peace, but there is as yet no guarantee that matters will change once these peacekeepers move out. The whole problem, therefore, acquires a certain surreal aspect. Weapons worsen conflict, but as long as conflict or an unstable peace continues, weapons will not be given up for any reason.

Given that weapons limitation or what are generally called ‘micro-disarmament measures’ have so far not been very successful, it would appear that a state under threat of conflict, or geographically situated in an area of weaponisation, would find it far more cost effective to set up a system of early warning that would prevent and effectively interdict even the first movements of weapons or prevent a slow arms build up, rather than try to retrieve weapons that no one wants to give up.

Simply put, while an early warning system cannot hope to keep track of every weapon that comes in, it aims at quick identification of initial infusion of weapons and arms build up, identification of the actors, routes and possible sources of such a build up, and the creation of a framework that effectively nips this phase of conflict in the bud – thereby allowing the state the breathing space to institute conflict resolution measures. Thus, while the ultimate aim is to ensure de-escalation of tensions in a potentially volatile area, early warning is predicated on a more narrow framework than conflict resolution in that it does not seek to go into the root causes of a conflict, though political and social issues would be knit in as part of the early warning indicators.

A word of caution is required here. Early warning is not aimed at allowing for effective police action alone. To serve as a continuing and dynamic process, early warning has to be used to trigger a host of measures that are aimed at social, political, and economic measures that would proceed side by side with weapons retrieval.

Early warning is, therefore, a process aimed at the early detection of arms flows and build up, with different instruments and tools used to shore up this process. These ‘instruments’ have so far been limited to inputs from intelligence sources. But the basic argument of this paper is that, while this is a substantial input, it is far from adequate. The case studies offered below serve to underline not only this fact, but also the fact that there are always other units of early warning – the press, analysts, and others – which are routinely ignored. The first objective of this paper is to identify such instruments of early warning that are ordinarily available to the public and to policy makers in conflict situations.

Having identified the actors, who in the past provided early warning, a second objective is to cull possible tools that could be used routinely as a set of early warning indicators.

A third objective is to provide a methodology for seamless information dissemination – that is, the movement of the signals provided by early warning actors using standard early warning tools, to the policy makers.

A fourth objective is the analysis of this information, and lastly, the identification of the precise inputs (or departments) that would be able not only to deal with weapons flows on the ground, but also to prevent further movements – through foreign policy, economic ‘rewards’, etc. The selection of these measures, however, remains outside the scope of the present paper, which is restricted to conflict prevention measures.


While all countries in South Asia suffer from various degrees of light weapons proliferation, it is important to note here that the problem manifested itself in different ways in each state. Pakistan’s role as a conduit for the US Central Intelligence Agency (CIA) weapon supplies in the Afghan conflict led to a massive societal diffusion of weapons into the hands of ethnic, religious and sectarian groups that has caused upwards of 2,000 deaths a year in Pakistan itself. India had to deal with two militant movements in its West, mainly abetted by Pakistan, and at least four major insurgencies in its Northeast. Sri Lanka has had to suffer a more than 14-year long war that has pitted the army against one of the most deadly militant groups in the world, the Liberation Tigers of Tamil Eelam (LTTE). Political and student violence in Bangladesh has been fed by weapons acquired from a growing black market. Nepal and Bhutan have both become shelters for militant groups, even as each is now experiencing rising levels of internal violence.

Across these many and varied upheavals, however, the ‘indicators’ are not all that different. In short, all of these problems are associated with the movement of deadly weapons – rocket propelled grenades, mortars, assault rifles and explosives – into the hands of various non-state actors and ‘uncivil society’: criminals, terrorists and others who thrive on the fringe of these conflicts.

Defending the cause of ‘civil society’ in each of these theatres are the police, bureaucrats, analysts, and a variety of non-governmental organisations and players. Some of these have been more alive to the dangers than others, but few, if any, were at all inclined to focus on the weapons issue. Attention remained clearly concentrated on conventional categories of factors, such as the actors and various national and international political issues (foreign assistance, fundamentalism, political parties, etc.), by and large ignoring the weapons factor – and this is in spite of an overwhelming number of indicators that specifically tied weapons to conflict escalation or formation. The result was that the sudden rise of violence consequent upon weapons proliferation, in most instances, took the state by surprise and resulted in what is termed ‘state paralysis’.

In attempting to outline a possible model for early warning, it is necessary to first flag the broad indicators, which are essentially those that could be used to identify ‘potential hot spots’ (situational warning). The specific indicators are more short-term and deal with arms build up and signals of approaching conflict (storm warning). An assessment of the role of various actors is given, followed by an analysis of the strengths and weaknesses of each of these actors and why early warning did not follow. Finally, a schematic model of early warning knits together these various institutions to deal with future situations.

The outline indicators (situational warning)

  • In all cases there was a strong historical precedent of covert arming. For instance Pakistan had been supporting Afghan rebels since 1973 with military aid.1 Similarly, Pakistan and India had already fought three major wars over Kashmir, which continued to be a festering issue. In India’s Northeast, rebellion had been carrying on since independence, with Chinese support and sanctuary in Burma. Sri Lanka’s ethnic dispute also began with independence, and had always found an ‘echo’ among the Tamils of South India. With few resources to draw upon, all these had been essentially small affairs and, therefore, had limited objectives.
  • The near certainty of weapons proliferation and availability as the scale of the Afghan adventure magnified and weapons supplies increased. To the East, the lowering of conflict in Cambodia was pushing weapons outwards.
  • In the case of conflict in India, a certain predictability in Pakistan’s strategy could also have warned analysts of increasing strife. The Pakistani army has always leaned heavily towards the use of irregulars2 (in 1947 and 1965) where arms were sent in to foment an uprising, which was to be followed by the invasion by regular forces. This was again the formula adopted in Kargil.
  • Another indicator was that all conflict areas had been hot beds of political tension and were obvious ‘magnets’ for weapons of any kind. Viewing the problem retroactively, the Sindh Chief Minister noted "A law and order situation had been developing in Sind since 1982-83… the situation has deteriorated mainly because of the massive availability of arms"3 In (Indian) Punjab, vituperative politics and economic slowdown had caused a general discontent, while in the Kashmir Valley, an almost complete indifference to governance and all pervasive corruption had led to the religious right’s demands for a change.
  • One clear signal which should have activated alarm bells among conflict studies analysts was the steady rise in narcotics cultivation in Afghanistan and Pakistan noted by the United Nations Drug Control Programme (UNDCP) as well as the US State Department. What is especially noteworthy is that, with the withdrawal of Superpower aid, militant leaders and their backers appear to have turned to refining heroin to sustain their operations. The involvement of the Pakistani army in such operations was hardly a hidden factor.4 In the Indian Punjab, the slow rise of militancy was preceded by an increase in narcotics seizures,5 and in 1984, operations against gold and silver smugglers taking place in a different sector demonstrated that these contraband smugglers were now showing considerable interest in narcotics as well.6 (The term "narco –terrorism" was not yet part of the international security lexicon.) In fact as a later report notes, Pakistanis were apt to see narcotics as a resource – in fact comparing it to the oil of the Middle East – seeing it as a problem for the "decadent West" rather than Pakistanis themselves.7
  • The sudden withdrawal of aid (China to India’s rebels in the Northeast, US to the Afghans, India to Sri Lankan Tamils) should also have served as a natural indicator that these groups would then turn to other sources for funding and weapons, since the conflicts remained very much ‘alive’. In the event Afghan splinter groups turned to regional sponsors, some of whom in turn accessed the international weapons black and grey market. Others like the Tamil Tigers and Indias’s Northeastern rebels, turned naturally to the Southeast Asian arms bazaars.

These, then, were the overall indicators. However more specific data was available. These were clearly signals of an impending escalation – in other words, what have often been called the accelerating indicators. These indicators were noted by the agencies involved well before the actual beginning of the conflict.

Arms build up phasethe storm warning

  • Warning indicators of weapons flows by border forces and security forces: All conflict areas saw significant border activity at least two to three years before the actual beginning of violence.8 For example, illegal cross-border traffic along the Punjab border increased, with security forces apprehending some 45,650 persons trying to cross the international border illegally (1986-1992).9 In 1988 alone, some 8,421 persons were arrested in the Punjab and Rajasthan with arms and drugs under the relevant laws.10 In the Indian Northeast, the hostile border prevented significant seizures, though a large seizure of weapons in 1995 (more than 775 AK-56 rifles and other weapons) provided a warning that militants were preparing for a fresh assault.11 This border input is not specially seen in Pakistan, or other conflict areas, since the Pak-Afghan border (the Durand line) had, to all intents and purposes, ceased to exist as the CIA–ISI operation was mounted. However, seizures of weapons from Peshawar to major cities began to increase from the early 1980s. The ensuing rise in urban violence was seen as a ‘law and order’ problem rather than one of excessive weaponisation arising from policies that created wider dangers.
  • Rise of armed crime: This period also saw a rise in bank robberies, heists and a general rise in law and order problems associated with armed crime, as militants sought more funds to buy arms. Interviews with police officers reveal that the use of pistols and revolvers in such incidents was not associated with the student unrest and violence that was then hogging the headlines, but perceived as isolated robberies by armed gangs.12 A similar trend is noticed when examining the Sri Lankan conflict, where militants were forced to attack jewellery shops, banks and other centres. Significantly, such a rise was not apparent in Kashmir, pointing to a steady source of weapons from across the border. Obviously, payment was not a factor. Crime in fact followed several years of conflict, arising apparently from the slow inclusion of criminal elements, a trend that was also apparent in the Indian Punjab.
  • Snatching of weapons: This is again clearly apparent in Sri Lanka and the Northeast, where early operations were specifically targeted against police and paramilitary outposts, or individual platoons.13 In Sri Lanka, the dreaded LTTE got its first weapon from a police officer who was attacked. Subsequently, operatives were told to snatch any weapons from the site of an attack, with even children trained to lift weapons after an explosion.14 In Kashmir, a crowd would pelt the police with stones while, in the ensuing confusion, some .303s would be snatched. These, however, were hardly used, since the AK made its appearance early in this conflict.15
  • The first sightings of rapid-fire and other weapons of a higher technology: Until 1987, in Kashmir, according to the Director General of Police (DGP) of that State, there was no evidence of training and certainly no guns (emphasis mine).16 Between October 1988 and April 1989, more than 100 persons were arrested with weapons so strange to the police who were still equipped with .303s that "had to rely on captured extremists to demonstrate to us [the police] how they are used".17 By 1989, the cry on the streets was "Jo kare Khuda ka Kauf, utha le Kalashnikov" (All god fearing men should now pick up a Kalashnikov).18 In 1990, the first firing incidents on Central Reserve Police Force (CRPF) occurred. In the Northeast, the AK-56 was already in use in 1988, well before the conflict went into its most difficult phase (in 1990). The RPG made its appearance a year later, while large-scale use of grenades began in 1992. The appearance of the Dragunov rifle (1993) was concomitant with the appearance of trained Afghan mujahideen. Thus, the weapons seized were indicative of various shifts in the pattern and scale of conflict, and of the actors involved. Ideally, the response should have been able to pick up these indicators. In actual fact, the presence of the RPG was officially noticed in the Northeast only well after the mid-1990s. Even now, there are persistent rumours of the presence of anti-aircraft missiles, though these have yet to be confirmed.
  • Weapons seizures: Weapons seizures are another indication of possible shifts in conflicts, or of their escalation into a new plane of violence. The first seizures of anti-aircraft weapons from Tamil militants in November 1986 by the Tamil Nadu police should have spurred a national level endeavour to sensitise intelligence agencies about the movement of these new weapons. In fact, it was almost a decade later that the Tigers finally used these guns against Sri Lankan aircraft in 1995.19 In Kashmir, the first anti-aircraft gun was seized in August 1997, and a shoulder fired Chinese missile, the HY-2, was also seized the same year.20 This gave clear indications of a possible escalation. Seizures of explosives along the Indian coast are certain warnings of an impending strike against important targets. Careful monitoring of data could well have prevented one of the world’s worst terrorist strikes in Bombay in 1993.
  • The glut in the black market: The inland movement of pipeline weapons was clearly evident in Pakistan in the form of a sharp fall in the prices of popular bores in the black market in major cities. The cost of a TT pistol in Karachi, once priced at Rs 5,500 went down to less than Rs 3,000 in 1989,21 while that of an AK-47 fell from Rs 30-40,000 in the early 1980s, to Rs 13,000 in 1989 and further to Rs. 8000 by 1994,22 at the infamous bazaars along the North West Frontier Province (NWFP). The sensitivity of the black market was apparent when, following a ban on the public carrying of weapons, savvy dealers simply changed their wares to offer smaller and multi-barrel weapons for "rich people and politicians."23

A similar response is apparent in the Indian Northeast, where, at an earlier time the local black market could only offer the .38. It was soon flooded with a much larger selection (which appeared to have been sourced from China as well as Nepal). However, unlike the western border, prices remained high. The M-16 was available at Rs. 40,000, while the AK Chinese copy (M-22) was much more expensive at Rs. 150,000. A Sterling stengun could cost between Rs. 65,000 to Rs. 70,000 though by the end of 1996, almost all prices had gone up in response to rising demand.24

Sri Lanka’s black market was, however, a factor of secondary diffusion, creating a glut as the intensity of conflict increased. The sources appear to be deserters (28,000 from the army)25, weapons issued for self-defence to villagers, and former militants, who have their own weapons for self-defence and are suspected to be involved in the trade in weapons, with the government understood to be turning a blind eye.26

At the level of the international blackmarket, the first warning of an access to outside sources was apparent when the Tamil Tigers reneged on a hard won peace agreement after a ship arrived from Singapore carrying weapons (some 700 rifles, ammunition and explosives) had docked in spite of surveillance by the Indian Navy.27 A similar indication was apparent in the Indian Northeast where the appearance of weapons that were clearly sourced from the Thai arms market began to make their appearance (M-16, Chinese bottle grenades, AK-56).

Licensed weapons

Warning signals of an internal diffusion of weapons were the sudden loosening of the licensing system as the rush to acquire arms for protection – both by householders but mostly by politicians – began. Pakistani reports note that 41,600 guns licenses were issued between December 1988 and May 1992. Another 92,000 were said to have been issued in the seven remaining months of 199228 (when the mayhem in Karachi was at it highest).

A series of political assassinations in Sri Lanka led to the ruling party diluting licensing laws to political parties to defend themselves, thus, raising the level of weapons proliferation within the country. A public debate on the issue appeared to point to the fact that all political parties had been equally responsible for issuing licenses from the early 1980s.

Though all these indicators were apparent, there was hardly any reaction by policy makers. In the following paragraphs, we will examine the inputs from the various actors involved. Again, it appears that warning was more than available but was not acted upon.

Early warning inputs by intelligence agencies

It was hardly likely that the Pakistani intelligence agencies would "warn" of impending trouble in Pakistan since they were the nodal agency not only for the Afghan operation but were also used by General Zia and his successors to quell political opposition. Thus, weapons were supplied to various ethnic groups in an endeavour to "divide and rule."29 The only, otherwise, stable institution in the state – the Army – which might have put the brakes on the slippery slope of conflict could not provide the warning that the State so urgently needed since, while it was heavily involved in maintaining internal order and was making lists of people engaged in drug and arms trafficking, at another level it was apparently willing to use the proceeds from drug trafficking for covert operations.30 This dichotomy possibly eroded its ability to act proactively. The fact that both these institutions had an axe to grind in the conflict – in terms of both institutional and personal power and profit, effectively eroded their ability to assess the situation.

In India, the intelligence inputs in the case of Punjab seem to have been timely. The clear proxy war input was revealed in 1988 when details regarding Pakistan’s training of militants were published in the press.31 The drug connection was also apparent in early 198832 and was later underlined in 1993, when a CIA report revealed the close links between politicians and drug smugglers of Lahore who used Sikh militants as ‘mules’.33 This early warning appears to have prompted the government into constructing an electrified fence across the border and increasing surveillance. This input may well have been crucial to the quick winding down of the militancy (1990-1993). In this case, a close co-operation between the army, police and the civil administration was crucial in information sharing (which intelligence agencies generally loath to do) and, therefore, a quick proactive action.

An almost appalling lack of information sharing was apparent in Indian operations in Sri Lanka. It would seem that the Indian Army, which was inducted into peace keeping operations under the most difficult conditions, had little guidance on the kind of resistance they were likely to meet. This attitude was evident even in the mandate of the Indian Peace Keeping Force (IPKF), which required it to implement disarmament within 72 hours! A comment by the then Army Chief is illustrative. Queried by the Prime Minister on the Army’s opinion on possible Tamil Tiger intransigence, he is said to have commented that they had no wherewithal to... confront India and if they did, they would be neutralised within two weeks.34 In the event, the operation dragged on for four long years.

In Kashmir, interviews by the author clearly reveal that evidence of training and weapons buildup had been supplied by concerned intelligence agencies to decision makers at least two years in advance.35 However, the higher ups tended to see these as ‘alarmist’ scenarios (possibly due to the rather tired ‘foreign hand’ theory that politicians invariably pound during elections). Later, intelligence collection was handicapped by targeted attacks on lower-rung intelligence officers, police personnel and key decision makers. Thus, over time, the intelligence base practically vanished.

Some points are worth noting. In the Northeast and Kashmir, the police were initially apt to think of the militants as ‘our boys’ whom they were loath to imprison. Police sympathy heightened as New Delhi brought in President’s rule in an attempt to control the situation. There is clear evidence that police intelligence (which is basically street-level) was denied to the Army or paramilitary forces. Second, competition between different agencies was also clearly detrimental to the overall operation since local commanders were eager to get the credit to their own service. ‘Leakage’ of information of an impending raid was not uncommon. Third, though intelligence men filed their reports to the local head, this was usually diluted to make it more ‘acceptable’ and less ‘alarmist’. Fourth, during militancy, the agencies of the government itself are usually infiltrated. Thus, the tendency to maintain secrecy increases manifold, so that, quite literally, one hand of the state rarely knew what the other is doing36. Fifth, intelligence briefs tend to concentrate on the actual persons involved, (Pakistani, Chinese, etc.) rather than the weapons themselves. Thus, the importance – for example – of a sudden rise in ammunition seizures was usually lost on intelligence analysts. Army intelligence men have similarly tended to remain unimpressed by the light weapons issue.

Overall, it has been seen that a plethora of ground level intelligence was available in all cases. Such routine ‘int-reports’ (especially in Kashmir) usually warn of militant activity and attempts at incursions – a state of affairs that has been going on for more than ten years. The lack of reaction that is often noticed, therefore, may be due to inertia that is inevitable in sustained militancy operations.

Early warning by ‘think tanks’ and non-governmental bodies

Overall, it appears that the strategic and analytical community among NGOs were far more alive to an impending conflict than government bodies (at least at the higher levels). Significantly, the most notable warning in the case of Kashmir came from a research group and their publication of a document by a leading defence journal of India called "Op Topac".37 It outlined what many believed was a Pakistani plan of subversion. Written in July 1989, it was the result of meticulous research by a team of specialists presented in a narrative form. As a piece of early warning it is unique since the scenario that unfolded was almost identical as far as the strategy and tactics followed were concerned.

Another unique analysis was from the doyen of strategic studies in India, K. Subramanyam, the then Director of the Institute of Defence Studies and Analyses, who, in 1981, was warning of both the dangers of the Afghan war and the costs to Pakistan. He observed, "If arms can flow in one direction, they can flow in the reverse direction," and later warned that the increased supply of arms in one country could lead to a greater flow into other trouble spots on the globe.38

Think tanks in Pakistan had not been particularly involved in the past but individual analysts have courageously pointed out the dangers of a proxy war policy. I.A. Rahman noted that the repeated bomb blasts in Karachi and the NWFP were hardly a ‘law and order’ problem, but were linked to the Afghan problem. "Neither the authorities’ noise about vigilance against terrorists, nor President Zia ul Haq’s discovery of the Kashmir dispute as the second most important issue before the nation," he asserted, "has prevented a debate on the central cause of Pakistan’s predicament, namely it’s involvement in the Afghan conflict."39 This perception was not shared by a majority of Pakistanis that was affected by the violence – traders who downed shutters, for instance, saw it as a ‘law and order’ problem.

Again, the Church in the Northeast was considerably involved in conflict reduction, but it is not clear whether they were specifically aware of the weapons problem. Though NGOs like the International Red Cross and Medecins sans Frontiers were particularly involved in Sri Lanka and were willing to talk in private, few were at all comfortable about talking to the press or government agencies.

Overall, the relationship between the think tanks and the Press was two-way, and where the relationship was close, good analytical pieces were seen. However, the Press in India, and in some instances in Sri Lanka, remained unable to assess the weapons issue, though this has changed recently.

The Press as an early warning input

The Press, like the analysts, is quick to pick up information on the ground, or investigate stories that are supplied by intelligence agencies or police officers. In Pakistan, the Press was most energetic in projecting issues that the authorities wanted to ‘hush up’. The most courageous and diligent among these were the Karachi-based Herald and the Peshawar-based Frontier Post, both of which underwent considerable harassment later. The Herald’s story in September 1985 on the Army’s involvement in narcotics trafficking was one of the earliest bits of public information available regionally. The Press was also the first to warn of the consequences of an uncontrolled spread of Islamic fighters and their ‘outfitting’, as it was to note the sudden rise in the number of madrassas (Islamic schools) which imparted military training to their students.40 Foreign investigative journalists gave an excellent coverage of the Afghan conflict and the intermeshing of the narcotics and weapons trade with Pakistan,41 (which provided a great deal of data to researchers). However, a precise focus on the weapons issue was non-existent.

Similarly, in the case of Punjab, the Press functioned clearly as an early warning input and remained close to decision making circles and the police throughout.42 The Press in Southern Sri Lanka was fairly impartial in its reportage, but was dependent on government handouts for actual reports on the fighting. In the East, they were unabashedly the mouthpieces of the Tigers. Reports from the Indian Press were more analytical, with investigative journalists often providing an early warning intelligence of the mood of the Tigers. Journalists were among those who warned Indian decision-makers on the Tamil Tigers’ decision to renege on the Agreement, and the inadvisability of turning against them. This advice appears to have been disregarded.43

In both Kashmir and the Northeast, the contrast is marked. With targeted killings of pressmen and repeated threats, the local Press was completely cowed down, and, as the Governor had put it, "were reporting at gun point."44 Editors and journalists were one at admitting that they simply published what was given to them or alternatively filed ‘reports’ that were slanted heavily in favour of the militants, or simply published notices calling for strikes.45 One vernacular paper notes that local journalists were well aware of the coming of Pakistan-trained militants as well as the conviction engendered in the Valley that ‘independence’ was approaching (though no one had the slightest idea of how this was to be implemented). Pressmen noted that their representations to bureaucrats had been ignored, and that the state bureaucrats had been compromised well before the uprising began.46 In the Northeast, the sway of the militants was so complete (for a period of about two years) that a journalist was to observe –– "Its writ ran unchallenged...not a leaf stirred without its approval."47 The national Press and Television were unable to get ordinary Assamese even to admit the existence of the United Liberation Front of Asom (ULFA), while those who did do so (under anonymity) admitted that, far from running a parallel government, ULFA was the government.48 It was only later that some members of the Press were used by the government as mediators.

The Bureaucracy

It needs to be stated clearly that the bureaucracy is by its very nature not suited to be an early warning instrument. It can, however, be expected to respond swiftly to indicators and warnings by other actors. In South Asia, the bureaucracy tends to be so secretive as to make any real analysis difficult. Thus, the Press is usually confined to drafting stories from press releases. However, it would appear that bureaucrats work in unison (instead of being departmentalised) when there is strong leadership from above.

In Pakistan, the bureaucracy seemed to be divorced from reality. Often, it appeared that the Press was far ahead in its awareness of the situation than the government. For instance, a Special Committee which submitted a 77 page Report (December 1989) on violence in the Sindh province (which was receiving daily coverage) merely mentions the need to have a ban on unlicensed arms and to ‘eliminate’ the drug mafia;49 the remainder is devoted to the political and social history of the province. In another instance, as the country reeled under sectarian violence, a report to the Prime Minister alleged large-scale external funding of training centres by "friendly Islamic nations" and gave a detailed summary of the activities of religious groupings,50 almost ten years after the problem had surfaced and the issue had been widely commented upon in the Press. By this time prominent journals were already noting that "there seems little argument that Pakistan is currently the world leader in hosting international terrorist organisations."51

The power of the Pakistani bureaucracy, however, is not to be underestimated. Its members usually work closely with the army, and the fact that a civil servant, Ghulam Ishaq Khan, eventually became the President is one indicator of this authority. Nevertheless, decision making is considerably ‘politicised’ and slow in the extreme. In rural areas, decision making follows traditional tribal patterns, with an overlay of modern corruption. Similar to the Indian system, the bureaucracy can, if it so wishes, hasten policy making or bring it to a standstill.

Bureaucrats in India shared the Pakistani instinct of ‘hushing up’ unrest, rather than dealing with it with determination.52 In 1988, the Kashmir Governor’s address to the Legislature reflected little of the trouble brewing, noting that the last vestiges of ‘terrorism’ were being wiped out.53 The panic had that set in with the arrival of the first few AK-47s is well documented in an account by the then Governor of the state.54

In the Northeast, the state bureaucracies have either been co-opted or sidelined by terror tactics. The Report of the then Governor of Assam, D.D. Thakur, to the President55 is an indictment of a government paralysed – it speaks of Central assistance in the form of additional vehicles and arms remaining unutilised, and of training camps run in broad daylight and in public view.56 Army action in 1990 and 1992 led to the militants taking up residence in neighbouring Bangladesh, which proved to be a better outlet since the groups could easily visit Thailand and Bangkok from there. This possibility was apparently overlooked by the local bureaucracy that was lauding itself on ‘dealing firmly’ with militancy. When the ULFA reappeared there was little surprise among the locals, the Press, or police officials.

In Sri Lanka again, policy makers appear to have grossly underestimated the Tamil Tigers’ capability and commitment to their own homeland. When the Tigers were pushed to the wall by the IPKF, the Sri Lankan government armed the Tigers with several truckloads of weapons, ammunition and explosives.57 After having got the IPKF out, the Tigers again turned upon the hapless Sri Lankan army; the war continues to date.

In Punjab, matters were somewhat different. As noted earlier, it seemed that all concerned were working together – for instance, following a Prime Ministerial meeting in April 1988, a decision was made to ‘seal the border’. Various other measures to deny entry to terrorists and weapons from Pakistan were also defined. Moves were also made to step up border security in neighbouring states. The state government, with the strong backing of the Centre, set out to revitalise a highly demoralised police force, increasing its firepower, mobility and communications capabilities (all of which were more or less non-existent prior to the Operations initiated at this time). Police crackdown and enhanced border security ensured by the Army resulted in a dramatic reduction in militant violence and its complete cessation by 1993. Again, this one instance stands out for the (relative) success that is possible when early warning is combined with prompt action.


Assessing drawbacks and strengths

When assessing the inputs from various actors – such as the Press, bureaucrats and others, it must be noted that the problem of light weapons proliferation or references to the ‘Kalashnikov culture’ are very much a thing of the late 1990s, and, therefore, may not have elicited the same reaction in the 1980s, when the Cold War threat overshadowed most of these problems. This aspect needs to be duly noted. Even Pakistan, which had a severe problem with societal arming, was more inclined to see and discuss it as a law and order problem in Sindh, as a refugee problem in Baluchistan, and as a political problem in the Punjab. In New Delhi, Prime Minister Rajiv Gandhi, mired in political scandal and bereft of experienced advisors, was too preoccupied to note the trouble in the Valley till late 1989. The government itself was busy drafting condemnation of arms and training given to Kashmir separatists, even as New Delhi warned Islamabad of serious consequences of such actions.

Bureaucracies (not just in South Asia) are traditionally given to playing down a problem, assigning all kinds of roles to opposition parties, the ‘foreign hand’ and others. The heavily departmentalised approach in this region means that each authority will not share information with others which, this is most apparent in Pakistan while least apparent in Sri Lanka (where the bureaucracy is more compact). Bureaucracies have traditionally reacted only after receiving inputs from intelligence agencies or facing sharp criticism by the Press. It was only in the Indian Punjab that all systems – the government, the police, bureaucracy and the army worked together, thus, swiftly reversing a situation that threatened to go out of control. It is, in fact, the one success story in an otherwise dismal show by the respective governments. Some points that stand in an evaluation of these facts included the following:

  • In India, governmental bodies like the Border Security Force, were valuable inputs in warning of impending light weapons flows. The collated data, which is published as an open source, received some attention from the Press and did alert think tanks. Similarly, the Narcotics Control Bureau (NCB) produces relevant statistics, as do Governmental institutions like the Crime Records Bureau, which has noted the rise in crime in states neighbouring conflict areas. In Pakistan, data collection on the Karachi issue appears to be efficient. However, few countries have public access to date bases, which remain either inconsistent or at least two years behind time. None of the available inputs appears to have been factored into decision making.
  • In all of the instances related above, the Press provided strong early warning inputs, especially, news magazines which devote greater space to analysis and strong criticism of the government. However, given the nature of its job, the Press cannot continuously focus attention on a subject once its ‘news value’ diminishes.

  • Non-governmental think tanks in Sri Lanka and India, while producing quality analyses of impending crises, have, however, not been able to sensitise governments on issues where the dangers are not obvious – like light weapon flows. When these are presented as clear proxy war instances, the reaction is likely to be more encouraging.

  • Thus, bureaucracies reacted only when presented with terms that they were familiar with. The weapons proliferation issue in its many dimensions remains one that evokes little interest, and is seen as one of peripheral security interest. The relationship between think tanks, the Press and the bureaucracy remains sketchy, with the first two at the mercy of the government when in need of hard data.

  • Intelligence agencies have always provided early warning. However, since this is often a daily exercise, the significance of particular events in a mass of data tends to get lost. What emerges is a general picture of the intention of the protagonists (Pakistan/India, fundamentalists, etc.) rather than forecasting of possible outcomes. Since weapons trade needs a widely flung net for analysis, spot intelligence rarely helps. Moreover, inter-agency rivalry considerably erodes efficiency.
  • The beat policeman is usually aware of developments in his area, but these reports appear hardly ever to be carefully read until a conflict breaks out. Sympathy factor works strongly against reporting on militant activities. Corruption is also generally higher, and while a few may wink at arms induction, fewer still may actually realise its significance. At the higher level, police officers have proved efficient in tackling a situation when backed by the political leadership. However, once the weapons are on the streets, the police usually have to retreat and leave the situation to the army. Conversely, once the army is called in, the police usually refuse to part with information.
  • NGOs and religious organisations rarely consult the organs of the State, seeing their job as ‘separate’ from the conflict issues. For instance, the Church, in India’s Northeast, only became involved in certain areas when it was required to pay ‘tax’ to militants. In Pakistan, however, the clergy are part of the problem, while some NGOs have made themselves unpopular as intelligence moles; most prefer to "live with the weapons problem"58 rather than to assist the government in tackling it.

A strategy for the future: Creating a national early warning system

In sum, it appears that a wide variety of data is available from different sources for early warning of weapons flows, both in proxy war and in black market operations. The crux of the problem is integrating the whole to allow greater linkages between specialists, information gatherers and decision makers. ‘Specialists’ here refers to those whose profession, by its very nature, requires them to be aware of illegal weapons flows. Such actors are border forces (inclusive of coast guard, and air surveillance), immigration officials and customs. This is referred to as the "first circle" of surveillance. The second circle of information is represented by the police forces and internal intelligence (including narcotics), who monitor events within their districts or areas. The third circle includes the concerned ministries involved – in the South Asian case this would mean primarily the Home Ministry, the Defence Ministry and Foreign Ministry liasing. Each of these has its own intelligence sources, and these would provide additional inputs. The primary aim would be to speed up information between these circles.

At a second level are the inputs from NGOs, think tanks and the Press. The Press clearly has a greater reach than the others in creating awareness, which is its main strength, while think tanks have greater ability to provide analytical solutions and inputs (formal and informal) into decision making centres. The main strength of NGOs would be that they are often closer to the problem than others, though their advantage lies in their relative anonymity. Given that they would dislike approaching the government, a greater lateral integration between these three would also be needed to complement each other’s skills.

At a third level are the static systems, which are required to quickly analyse data. These are the central data collection agencies – in India they are called the Criminal Records Department – whose input would be invaluable for both levels. This system, which is more or less defunct in other South Asian countries (and non-existent in others), depends on accurate and timely information from the first and second circles of early warning.

The process

Since early warning is clearly a process the objective would be to speed up information gathering, dissemination and analyses between and within these levels identified above. This essentially means that the two circles of primary information need to utilise the Press at the earliest possible moment on weapons seizures, which, in turn, would assist the work of government think tanks, an analysis which would be immediately available to the concerned ministry. Data also needs to be fed into the third level (criminal records bureau) – if possible, in real time, thus, creating a data base on weapons movement, increase or decrease, which in turn would be made available to the Press, interested NGOs or think tanks. Relative changes in a given area or within a given group, therefore, would be immediately recognised. This information needs to be relayed back to the action takers (the police, security forces, narcotics bureau and revenue departments) who could act upon this constant update of information.

The third circle of information (ministries + intelligence), in turn, would be ‘sensitised’ by the above analysis of impending danger from a particular direction allowing intelligence agencies to focus their attention along one direction. This would also allow a paring down of intelligence assets to allow for greater specialisation and expertise, instead of the present diffuse set up. These bodies cannot be expected to divulge their activities – however, some yearly ‘risk assessment’ would help the other two circles immensely in providing a framework for activity.

At the apex of all this activity would be an early warning panel, made up of the chiefs of all the concerned departments including the Home and Foreign ministries, with a senior level politician in the chair to ensure co-ordination. Alternatively, an early warning panel could provide weekly briefings to a national security council (where relevant). The Press and think tanks should be part of selected sittings, where data is made public, or may be called in to provide expert, outside opinions. The emphasis would need to be on transparency, with the clear realisation that secrecy benefits the militant and the criminal – not the State. This process is hardly one dimensional, but a schematic representation is given below.

Ultimately, this process is based as much on new information technology, which includes not only the hardware, but organisational flexibility and speed. If the nodal group were to function just like any other government group, (i.e., based on rigid hierarchical orders) it would remain ineffective.

At the international level, such nodes of warning should be linked to UN groups set up for the purpose, which would have a direct link to the Secretary General. Links to relevant multilateral organisations would also be useful in attracting attention – especially in the case of State-sponsored war or irresponsible supplies of weapons. Denial of equipment, cut-off of aid, and other tools of multilateral persuasion could be applied.

Many a proposal that would at one time have been discarded as hopelessly optimistic is now at least being considered at the international level. Regional co-operation initiatives like the Organisation of American States (OAS), Convention and the European Union Code of Conduct are milestones along the way to greater awareness of the problem of light weapons proliferation as a threat to every country, regardless of its developmental status. At the United Nations level, the Disarmament Division is as much taken up with "micro-disarmament" issues as it is with conventional arms trade. The World Bank now has a specific unit to deal with post-conflict situations. On the realistic side, it is also important to note that there are a large number of organisations willing to get on to the bandwagon and that not all are effective or even desirable. However, the fact that the subject is on the international agenda is an encouraging sign.

The one lacuna that still exists is that, while international declarations say a great many things, there is as yet no legislation (barring the OAS Convention) that makes illegal arming a crime, much less proxy arming. The latter is clearly still an operation that a large number of states find useful. Until it is realised that proxy arming is, in the long-run, a zero sum game that benefits no one, measures at any level can hardly hope to succeed. Ultimately, the strengthening of norms against arms dealing should itself help in focussing attention on States which go against this. The State eventually, has to realise its own interest in either desisting from such practices, or as a victim, in focussing early attention on them. After all, no problem can be handled unless its potential dangers, opportunities and options are well understood.

There is little doubt that an early warning structure would be immensely useful to developing states – enmeshed, as they are, in diverse demands upon their limited resources. The costs to the State of weapons diffusion are not clearly quantifiable. Apart from the clear economic losses, costs of state forces, costs of refugee repatriation, etc., there can be no ‘quantity’ assigned to the loss of a child to its mother, or the gunning down of the helpless and the weak. Early warning is a tool that allows the State to fulfil its primary function – that of ensuring the safety and security of its citizens. Unfortunately, it seems to be a function that most States have forgotten.

  • Ms. Tara Kartha is a Fellow of the Institute for Defence Studies and Analyses, and a frequent contributor to a variety of journals on strategic and security issues. She is the author of the book, Tools of Terror: Light Weapons and India’s Security.
  1. See WIRSING, Robert G., Pakistan’s Security under Zia, 1977-78, Houndmills: Macmillan 1991. Also see, LAMB, Christina, Waiting for Allah, London: Hamish Hamilton, 1991.

  2. See COHEN, Stephen P., The Pakistani Army, New Delhi: Himalayan Books, 1984.

  3. Herald (Karachi), July 1993, p. 60.

  4. See Herald, 1985. Also see, RUBIN, Barnett, Fragmentation of Afghanistan, Yale: Yale University Press, 1995.

  5. Crime In India––1988, National Crime Records Bureau, New Delhi.

  6. Joint-operation by Revenue and Finance Ministries along the coasts of Gujarat, Tamil Nadu, Maharashtra, Kerala and also in New Delhi. The operation netted some 140 smugglers and goods worth Rs. 900 million in 1988.

  7. The Friday Times (Lahore), September 3, 1993.

  8. Between 1988 and 1990, over 1474 AKs, and over 1,000 mines and 2,994 grenades were seized by police and border forces in Kashmir. India, Government of, Ministry of Home, Annual Reports, 1989 and 1991.

  9. Crime in India––1992, New Delhi: National Crime Records Bureau.

  10. Crime in India––1988, New Delhi: National Crime Records Bureau.

  11. For a description of the operation see SINGH, Jasjit, ed., Light Weapons and International Security, New Delhi: Pugwash––IDSA, 1995.

  12. Interviews by author with police officials, now retired, who had earlier been at the helm of affairs during that period, July 1994.

  13. Here even a pistol was initially a highly valued possession, with only area commanders sporting these. Interviews with militants by author, July 1994.

  14. The killing of the Mayor of Jaffna and, subsequently, a police officer yielded the first light machine gun. See KARTHA, Tara, Tools of Terror: Light Weapons and India’s Security, New Delhi: Knowledge World––IDSA, 1999.

  15. Interview with a police official, January 22,1999.

  16. India Today (New Delhi), May 31,1989, p. 71.

  17. Ibid.

  18. India Today, December 31, 1989.

  19. The Tamil Tigers shot down two SLAF Avros in April 1995 and one Puccarra shortly after. The missiles were suspected to be SA-7’s, though this has not been confirmed. The Times of India (New Delhi), April 30, 1995.

  20. KARTHA, op. cit.

  21. View Point (Karachi), May 31, 1990, p. 8.

  22. For a price comparison, see Rahimullah Yusufzai, Newsline, October 1989. For 1994 prices, see SINGH, Jasjit, ed., op. cit.

  23. See KARTHA, Tara, "The Diffusion of Light Weapons in Pakistan", Small Wars and Insurgencies, Vol. 8, No 1, Spring, 1997, pp. 71-87.

  24. Interview with sector commander of ULFA (1988), April 28, 1996.

  25. Brig Tissa Jayatunga, Army spokesman, c.f. Asian Recorder (New Delhi), January 1-7, 1997, p.26156.

  26. There is no clear indication of this, but media persons, students and others repeated the allegation during the course of the author’s visit.

  27. SINGH, Lt Gen (Retd.), Depinder, "The IPKF in Sri Lanka" Trishul (New Delhi), 1992, p.77

  28. The figures for 1992 are not entirely clear. See Heraldi, 1990. Also see HUSSAIN, Navaid, in Dawn (Karachi), March 31,1996.

  29. Analysts in Pakistan believe that the ethnic group the MQM (Mohajhir Quami Movement), created by General Zia as a foil to the nationalist movement under Benazir Bhutto, was armed by the intelligence. When it became too strong, a splinter group was similarly armed.

  30. Nawaz Sharif, then in the Opposition, The News (Islamabad), September 13, 1994.

  31. Asian Recorder, June 17-23, 1988, p. 20066.

  32. Militants arrested revealed that weapons had come from a well guarded "bungalow" (large house) in Lahore (Pakistan) where large quantities of missiles, weapons and ammunition was regularly stored to be passed on to Sikh militants. The main contact man was a well known heroin smuggler. For more details, see Ibid.

  33. CIA Report as published in Friday Times (Karachi), September 13, 1993.

  34. Account of the then Indian High Commissioner to Sri Lanka. See DIXIT, J.N., Assignment Colombo, New Delhi: Konarak, 1998, p.143.

  35. Interview with officials by author, November 1998.

  36. These conclusions have been arrived at after repeated interviews with various heads of services in the concerned states during the period 1995 – 1997.Their names have been withheld for obvious reasons.

  37. Indian Defence Review, (New Delhi), July 1989.

  38. SUBRAMANYAM, K., "Arms for Afghan Rebels", Strategic Analysis (New Delhi), Vol. IV. No 10., January, 1981.

  39. For a collection of essays on the political state of the country, see RAHMAN, I.A., Pakistan Under Siege, Lahore: Rohtas, 1990.

  40. Dawn, January 30, 1995.

  41. For some excellent accounts, see LAMB, op. cit. Also see DUNCAN, Emma, Breaking the Curfew, London: Michael Joseph, 1989.

  42. In Punjab, major newsmagazines noted the fact that weapons had reappeared within the Golden Temple Complex (after the army action which cleaned up the area) and that the fight for "Khalistan" by violent means was hardly over, unlike claims by the bureaucrats that the situation was "normal". They maintained a spotlight on the activities of the religious and separatist heads, and noted that in spite of a clutch of laws passed to prevent such activity, it was obvious that weapons were once more part of the politico-religious scene. See Asian Recorder, April 29-May 5, 1988, p. 19989.

  43. Prominent among these were reports filed in Frontline (Madras) which were detailed and of a high quality. Particular mention needs to be made of reports filed by V. Jayanth and P. S. Suryanarayana.

  44. See for an account by the then Governor of the province JAGMOHAN, My Frozen Turbulence, New Delhi : Allied Publishers, 1991.

  45. Author’s interviews with reporters and correspondents of major weeklies and lithos. Srinagar, May 1998.

  46. Author’s interviews with local journalists of vernacular and English press. April 16, 1998. Names of papers have been withheld at their request.

  47. HAZARIKA, Sanjoy, Strangers in the Mist, New Delhi: Viking, 1994, p. 189.

  48. Newstrack, September 1990.

  49. Report of the Special Committee of the Senate, The Situation in Sindh, December 8, 1989. See Herald, January 1990, for an analysis of this report.

  50. Report detailed in The News, May 6, 1997.

  51. Jane’s Intelligence Review (London), March 1997, p.136.

  52. As a journalist notes, in New Delhi, the Prime Minister Rajiv Gandhi mired in political scandal and short of experienced advisors, was too preoccupied to give adequate attention to the trouble in the Valley and remained dependant on the inputs of a new inner circle members. SINGH, Tavleen, Kashmir: A Tragedy of Errors, Viking: New Delhi, 1995). p.109.

  53. Hindustan Times, February 27, 1988.

  54. JAGMOHAN, op. cit.

  55. Report of the Governor to the President of India, November 1990. Tabled in Lok Sabha on December 27, 1990.

  56. Ibid.

  57. The Report of the Special Presidential Commission of Inquiry probing into the assassination of Lt. Gen Kobbekaduwa, Government Publication Bureau, Colombo, Sessional Paper No IX- 1997

  58. As noted to author by NGO members in the Northeast, November 1998.





Copyright © 2001 SATP. All rights reserved.