How India failed to predict Kargil War

India Today

26-07-2025

(NOTE: This article was originally published in the India Today issue dated June 14, 1999)The relentless daily list of casualties shook the nation. By the end of last week, 57 Indian Army personnel had died and 203 grievously wounded as the armed forces struggled to push back the shock invasion of Pakistani- backed Islamic fundamentalists from the strategic heights around Kargil.advertisementThey had died doing their duty—protecting the territorial integrity of their country. But as thousands of Indian troops brave a withering storm of artillery and machine gun fire to repulse the intruders, the country is asking what really went wrong. Former chief of army staff General V.N. Sharma has no doubts: 'It is an intelligence failure, one of the biggest in recent times.'Defence experts are amazed why an intrusion of such magnitude that must have taken Pakistan several months to execute went undetected. Briefings and authorizations would have been made in Islamabad and Rawalpindi months in advance, field officers would have recruited and trained the force in the Baltistan heights of Pakistan Occupied Kashmir (POK), equipment and supplies would have been stocked in Pakistan army garrisons at Skardu and Astor before being sent to camps close to the Line of Control (LoC) where the battle now rages. Admits Lt-General H.M. Khanna, Northern Army commander: 'We are now positive the build-up began during and before winter for it requires extensive preparations.'
Clearly, all three principal agencies— Research and Analysis Wing (RAW), Intelligence Bureau (IB) and Military Intelligence (MI)—must accept their share of the blame for failing to anticipate the move. RAW, whose domain is external intelligence, should have ferreted out information from its network of agents and technical means even while the planning for the assault was going on some time last year. The IBand MI should have detected the arrival of the intruders across the LoC the moment it happened in February or March. But in Kargil there was a string of major lapses:RAW collects an enormous amount of information on Pakistan's activities in Kashmir through its Aviation Research Centre (ARC) headquartered in Sarsawa in western Uttar Pradesh, which flies special electronic intelligence and photo-reconnaissance aircraft. But in the case of Kargil, say insiders, these missions were only flown after the intrusion was discovered. RAW is also supposed to have humint or human intelligence, mainly its agents stationed in Pakistan and POK. Although RAW headquarters in Delhi had information as far back as September 1998 of a Pakistani plan to send infiltrators into Kargil, the agency failed to act on it.The army must shoulder a large share of the blame too. Its failure began from the time the Pakistanis started to build supply bases apparently along the LoC between January and March this year for the militants, who subsequently breached it and took positions on the strategic heights. Says a senior officer of 121 Brigade somewhat lamely: 'We were prepared for an infiltration, not an invasion.' Also, during winter army aviation choppers are used to detect any such movement. Significantly it was not done this winter.The IB with over 400 intelligence gathering personnel has even a stronger presence in the Ladakh region than RAW. In October 1998 an IB report from Leh spoke of preparations for Pakistani special operations in the Kargil sector. IB officers in the sector claim they shared the information not only with their headquarters but also with 121 Brigade stationed at Kargil. Yet neither the brigade nor IB headquarters followed it up with greater surveillance to see if the information was accurate. IB now says that its Leh office reported in June 1998 that there was a noticeable army build-up at Skardu in POK but none of the other agencies involved took cognisance of it.The most critical piece of information, it is now claimed, came from the G-Branch, the intelligence wing of the Border Security Force stationed in Srinagar, in October 1998 when it interrogated Azhar Shafi Mir, a Kashmiri militant, in Baramulla. He revealed that he was one among a group of 80 militants trained by a Colonel Shams for specialised road-sabotage operations focusing on the Leh-Srinagar highway. IB also interrogated him but both the agencies didn't act on the information. The BSF report reached its parent Home Ministry only on May 26, well after the battle for the Kargil heights had begun.There are other vital signs that seem to have been ignored. There was an IB report in October 1998 of a Pakistan Army survey of the area using Remotely Piloted Vehicles (RPVs) for reconnaissance. An Indian Army analysis in January concluded that RPVs had not been used and, therefore, missed the significance of the report. In February a Pakistani Mi-17 helicopter was seen over Indian territory near Kaksar in Kargil.advertisementThe army now says the chopper even opened fire but later went back and the conclusion was that the chopper had strayed into Indian airspace. Further, unlike previous years when little or no fire was exchanged between the two armies in Kargil in the winter months, there was a constant exchange of artillery fire this year. Officers now concede this could have been related to the unfolding of the Pakistani plan in Drass, Kaksar and Batalik.advertisementSo how is it, despite having so much evidence of the build-up, none of the agencies rang the alarm bell? Some critics charge that both RAW and IB officials are so busy safeguarding their necks that they usually drown headquarters with a torrent of useless information. Insiders say some of their reports are like astrological forecasts, suitably vague about the specifics and frequently issued as bureaucratic ploys to cover all eventualities. Separating the chaff is a herculean task and on many occasions genuine tipoffs go unheeded because agency officials have cried wolf once too often.advertisementWhile that may sound like a charitable explanation, the real problem is that many of these agencies work at cross-purposes with the hierarchical structure still ill-defined. Information from intelligence agencies moves up the chain of command, depending on its significance. Data useful for the military is sent directly to the three services while other significant information is sent to the Joint Intelligence Committee (JIC), concurrently the secretariat of the National Security Council, that works out of Sardar Patel Bhavan in Delhi. Its chairman reports directly to the prime minister's Principal Secretary Brajesh Mishra who also happens to be the National Security Adviser (NSA).advertisementThis is largely in theory. In reality both IB and RAW keep the most important pieces of information to themselves and provide them directly to the Prime Minister's Office. They also produce daily intelligence summaries which are bland bits of analyses that do not usually go into the specifics. In the case of the Kargil failure even if the army was lax in acting on the alarming reports, why didn't the RAW and IB chiefs take it up directly with the prime minister to whom they have routine access?One explanation that seems to have considerable currency is that the failure was because of the history of the area. The sector dominated by Shias had not supported militancy and there had been no known instance of infiltration through this inhospitable terrain. But as General Sharma points out, the job of an intelligence professional is to predict the unexpected.K. Subrahmanyam, a leading defence analyst, says that the failure in Kargil was not the lack of intelligence information but of assessing and interpreting the available data. Chairman of the JIC in 1977-79, he says it is not enough to collate information. It has also to be adequately processed and analysed and placed in context. 'More than 50 per cent of the CIA comprises 'white collar' analysts, the professorial types who attend seminars and spend time in universities,' he says.Good intelligence work, according to him, requires the mastering of 'inductive as against deductive logic'. All that RAW and IB do is collate information, deduce its implications and present it to the government. What is needed is a more complex exercise that 'requires you to think what you would do were you in your adversary's position'. This means a profound understanding of the opponent's psychology, the context of his politics and society.This is the kind of job the JIC was set up to do in 1963. But the reality is that for the past several years it has functioned with little guidance and with a key staff of just two dozen persons to undertake the task it was set up for. Emasculating the JIC is the consequence of the generally poor appreciation of the need for 'assessed and processed intelligence'. One reason for this has been the skewing of intelligence work to benefit the party in power rather than towards national security. This has given enormous clout to the agencies enabling them to get involved in areas outside their charter. According to former foreign secretary J.N. Dixit, RAW became an actor in influencing Indian policy towards Sri Lanka in the '80s rather than being an agency gathering intelligence and information. The Sri Lanka experience also deepened the distrust between the army and RAW which is supposed to provide it external intelligence.Despite its great importance as the eyes and ears of India's national security, RAW's management by the government seems to defy explanation. A 'steering committee' of top-level secretaries of the government is supposed to exercise some control over RAW and IB but this is more on paper. A few months ago, the government appointed A.S. Dullat, special director in the IB, as officer-on-special duty in RAW. Given his seniority it appeared that Dullat was being positioned to succeed Arvind Dave, the current chief who was denied the usual two-year extension of service on technical grounds and given a one year extension last May.But on April 30, the government citing a redressal of grievance petition from the next seniormost RAW official, R. Nagarajan, gave Dave a three-month extension and said the next government would decide on the appointment. This stop-go arrangement is hardly conducive for the proper functioning of the only agency that provides the country with external intelligence.India's lack of an intelligence culture is rooted deep in its history but the inability of its present leaders to find measures to rectify this compounds the problem. Three years ago when the armed forces put up a request for a dedicated surveillance satellite, something the Indian Space Research Organisation can easily fabricate, the project was stymied by the then cabinet secretary. 'He just did not understand what space surveillance was all about,' says one bemused army general. Besides adequate satellite surveillance, the country's intelligence services need to boost their electronic intelligence gathering. A single aircraft, that too a vintage Boeing 707, does the job today. The army needs to enhance its own surveillance capability in the form of additional RPVs, battlefield surveillance radars and wide-area thermal imaging sensors.Kargil is a massive systemic failure for which India is already paying a heavy price. The intruders have to be evicted because from their pickets overlooking the vital Srinagar-Leh highway they can play havoc. Given the disadvantage in heights, the battle to push them back is going to be long drawn and costly. The exchequer is already doling out an estimated Rs 15 crore daily. For every soldier who is pressed into battle, four others are needed to support him. Many more Indian soldiers will be killed in the war to regain the heights. It is therefore important to know why India's intelligence failed in Kargil, not just to punish the guilty, but also to recompense the dead and prevent such serous lapses in future.—with Ramesh Vinayak in LehSubscribe to India Today Magazine- EndsMust Watch