The Kargil episode was spawned by a set of unrelated circumstances, the first being Jehangir Karamat's resignation. Too professional and far-sighted a general not to see where the flow of events would lead, he was unwilling to satisfy fundamentalist urgings in and outside the army. Squeezed from below by adventurous corps, commanders and the lunatic fringe, he chose the only option of a mature soldier.
Of greater interest is the timing of the go-ahead to cross the LoC probably August/September '98. The decision arose out of the gravest misperception Islamabad has made since '71 that India's conventional superiority has been negated by Pakistan's bomb. Such misperceptions were reported by State department officials, the Pentagon, the American nsc, think-tanks like Brookings, the Stimson Center and any number of foreign journalists who live in Delhi and cover Pakistan, as early as October '98.
The parity lobby has undoubtedly recreated the quantitative analysis of the Korean and Vietnam wars where a side armed with nukes was fought with impunity by the other side with conventional weapons. The analysis runs thus. All wars are set off by a Reactionary Power (R) against the status quo power (Q). If the destruction R and Q can inflict on each other are measurable on a scale of 1 to 10, and 10 represents N-holocaust, then Kargil represents, say, 2. If R (Pak) can inflict damage on Q (India) to the tune of -2, will India respond with nukes? If it did, net loss to R would be (-10). In its return strike, Q too would be destroyed (-10). So, say the Pakistani ostriches, why would Q want to trade a loss of -2 for a loss of -10, just for the satisfaction of imposing a loss of -10 on R? Ergo, R can needle Q in small bursts of aggression (up to -2) and Q won't respond with nukes. Hence parity. What induced this logic? A range of faulty signals from India created not so much by bad N-strategy, but an absence of any strategy: conventional or nuclear.
Nuclear deterrence is always posited in a scenario. The new South Asia scenario was created as early as '87, when the Pak bomb became operational. It ran thus. Pakistan would try to internationalise Kashmir and escalate at the LoC.
India would retaliate conventionally and Pakistan would respond. Eventually Indian conventional power would force Pakistanis back from the LoC and ibl until a mythical geographical line was reached called the 'nuclear threshold'. Defined first by Sundarji, nuclear decision-making enters a grey area here. When India weaponised overtly in May '98, it should've had in place a strategy to conduct 'deterrence signalling' through the command and control system. By not articulating any strategy, by not setting up a C3I system, by not promulgating its missile system's accuracy, by not defining minimum deterrence, Islamabad has had to make some wild guesses all inaccurate. Not surprising, considering the intellectual standing of generals like Musharraf. Their hope, wild and improbable, is that India understands the nuclear threshold is the LoC!
India should've long ago made it clear it was prepared to fight 'through' a conventional war that it had no belief in an N-threshold. The logic is clear. Pakistan needles India as in Kargil. Score Pak (0) India (-2). India ripostes, Pak -2, India -2. Indian conventional weight begins to bear and the N-threshold is reached, Pak (-6), India (-2). Now the shoe is on the other foot: why would Pakistan exchange a loss of (-6) to launch a nuclear attack on India (-10), when the return strike would raise Pakistan's loss from (-6) to (-10)? An Indian nuclear strategy tailored for Pakistan should have been created for this scenario, transparently promulgated and signaled to Pakistan through the creation of a specific type of arsenal. But where are the Indian strategists?