Two close observers of Indian nuclear policy recently suggested that official thinking about India's nuclear strategy may be moving in a radical new direction, towards a first-use or even a first-strike strategy. Until now, it had been assumed that Indian nuclear policy would be retaliatory rather than pre-emptive, and that it will be focused on countervalue (i.e., the adversary's cities) rather than counterforce (the adversary's nuclear forces) targets.

For India, both a first-use and a first-strike strategy (and they are not the same) are the wrong choices for the simple reason that they will be ineffective in achieving either its wartime or deterrence requirements, in addition to embroiling India in an unnecessary nuclear arms race, and is likely to lead to dangerous crisis instability to boot. But, before considering these issues, an equally important question needs to be asked: how credible are the claims that Indian nuclear strategy may be changing?

Vipin Narang argued in a presentation at the Carnegie International Nuclear Policy Conference that comments made by senior Indian officials — specifically an essay written by Lt. Gen. B.S. Nagal, former commander of India's Strategic Forces Command (SFC), the "personal" comments made by former Defence Minister Manohar Parrikar, and most importantly, in the chapter on India's nuclear doctrine in the recent book by Shivshankar Menon, the former National Security Advisor — suggest that India may launch a nuclear attack first if and when it believes Pakistan is ready to cross the nuclear threshold and that this might take the form of an attempted full counterforce strike against Pakistan. Narang points to other pieces of evidence that India is working towards this strategy: the focus on MIRVs (Multiple Independently-targetable Re-entry Vehicles), missile defences, missile accuracy, readiness, and numbers. Shashank Joshi backs up Narang, arguing that these comments by former senior officials are "indicative of the fluid, elusive nature of nuclear strategy, as well as a more uncertain security environment and growing confidence in Indian capabilities."

How credible are the claims that Indian nuclear strategy may be changing?

Neither Narang nor Joshi claim these as settled changes. Joshi suggests that Menon's words are "more likely a warning, than an indication of imminent shifts," while Narang is more categorical that this is "where India may be heading, and certainly wants to head." But it is difficult to judge whether former officials are outlining their personal views or reflecting an internal debate when they write, considering that even ministers feel little compunction in pontificating about personal views on critical nuclear issues in public. But there is little indication in the writings of these officials that they have grappled with the serious problems that comes with a first-strike or a first-use strategy (outlined below), which suggests these are not well thought or researched policy positions in an internal debate. That these are positions taken by former officials means that they need to be seriously debated, but that is a far cry from seeing it as indicative of the direction of official policy.

A more serious problem is that there is no clear evidence that India is attempting to develop the capabilities that it needs for such a strategy. Each of the indicators that Narang points to could have other explanations. For example, India's BMD programme is over two decades old, starting well before India's 1998 nuclear tests. It is possible, of course, that whatever its origins, India now considers its BMD programme as part of a damage-limitation first-strike strategy. But it is difficult to make this assumption without better evidence of the link between the BMD programme and any change in India's nuclear employment strategy.

A more credible piece of evidence is India's push for greater accuracy of its missile systems. Obviously, there is little need for accuracy if India were considering hitting only cities and so the push for greater accuracy could be seen as presaging a counterforce strategy. Similarly, MIRVing of missiles may also appear to be less important for a minimal deterrent force that targets only cities. But as Narang himself pointed out in an earlier essay, many of these capabilities are being developed by the Defence Research and Development Organization (DRDO) without political sanction. While such unauthorised development (or at least the DRDO boasting about them) could possibly force political decisions later, there is little indication of anything of that sort happening yet.

Narang also points to India developing more missiles, but there is little indication of any spurt in the numbers of India's missiles. Moreover (as I point out below) any first-strike or even first-use strategy would require India to have a nuclear arsenal far larger than Pakistan's, whereas what exists is a significant nuclear imbalance in Pakistan's favour. Not only do we not see any dramatic growth in Indian nuclear forces, we have not even heard any expression of concern by any Indian official about this imbalance, which suggests that Indian officials are not particularly concerned about it. This by itself is an important indicator that India is not considering a counterforce strategy, at least in any systematic manner.

Still, it is important to reflect on the implications of a counterforce first-use or first-strike strategy, if only because it is now the central issue in the Indian nuclear debate. But first, there has to be some clarity about terms and concepts. What Shivshankar Menon is suggesting, it appears, is not so much a "first strike" strategy, but a first-use one that may be entirely or partially a counterforce strike. But all counterforce strikes are not "first strikes". Even an entirely counterforce first attack is not a "first strike" strategy. Traditionally, first-strike is a bolt-from-the-blue attack, with no warning, which means that the targeting side does not have to worry about the adversary dispersing its mobile systems and bombers. This is the assumption that Keir A. Lieber and Daryl G. Press make, for instance, in their well-received study on US nuclear superiority. But, all the discussion in the debate about Indian first-strike, both from Menon and Nagal as well as in the responses to them, refer to India attacking either after a Pakistani nuclear attack or in the context of an imminent Pakistani nuclear attack. Neither fit the pattern of a 'bolt from the blue' attack because in a crisis situation, Pakistan will already be on alert and has potentially already dispersed its nuclear weapons. (Even if they had not planned to do so until now, this Indian debate will certainly force their hand). This is thus better described as first-use than as first-strike.

There has to be some clarity about terms and concepts. What Shivshankar Menon is suggesting, it appears, is not so much a "first strike" strategy, but a first-use one that may be entirely or partially a counterforce strike.

Even a true surprise first-strike strategy, which may be possible with a relatively smaller arsenal, is simply not viable for India because it does not have the necessary superiority vis-à-vis Pakistan in nuclear warheads. The numbers problem becomes even more onerous if India is considering a first-use strategy, after Pakistan is already alerted.

Considering the requirements of a first-strike strategy against Pakistan will illustrate the problems that India faces. Pakistan has dozens of ballistic missiles of varying ranges which are deployed, according to one assessment, in seven to eight garrisons. Pakistan's F-16s, Mirages and J-17s fighter-bombers, which are generally suspected to have nuclear missions, are based or have been spotted at about half a dozen different Pakistan Air Force (PAF) airbases. That makes about fifteen targets. If some of the most important command and control sites are included (the three service headquarters and the Strategic Plans Division) as well as key facilities associated with nuclear weapons (the assessment by Kristensen and Norris, cited above, mentions nine facilities, which is a good starting point: Gadwal, Khushab, Chasma, Nilore, Kala Chitta Dhar, Fatehjung, Tarnawa, Taxila, and Wah) plus a couple of the Corps headquarters that might be in charge of Pakistan's TNWs, an extremely conservative estimate suggests around 30 counterforce aimpoints.

India is currently thought to have about 100 to 110 nuclear warheads, deployed on a mix of missiles and aircraft. Considering that China has a much larger nuclear arsenal and is also a much larger country, it would not be implausible to assume that India devotes a larger part of its arsenal for China oriented missions than Pakistan oriented ones. But even if it is assumed that India needs only half of its arsenal for China oriented missions, that leaves just about 50 to 55 warheads for these thirty-odd counterforce target aimpoints. Calculating the number of warheads per aimpoint is complicated, with variables including the hardness of the target and the reliability of the warhead and delivery vehicle. Lieber and Press, for example, suggest that the US needs anywhere from one warhead (for nuclear weapons storage facilities) to as many as seven warheads (for Russian silo-based and land-mobile missiles) depending on the type of target.

 In India's case, there is little information about reliability, warhead yields or target hardness. Still, assuming (very) generously that India needs only two warheads per aimpoint, it needs at least 60 warheads even for such a truncated target list. Of course, Indian decision-makers will also need to keep some weapons in reserve to target any surviving Pakistani nuclear assets and to retaliate if Pakistan attacks India with these. If we assume just 30 reserve warheads, India needs a total of about 90 warheads just to conduct a surprise 'splendid' first-strike against Pakistan, which will leave India with barely two dozen warheads to deter China.

In India's case, there is little information about reliability, warhead yields or target hardness.

The numbers problem would be insurmountable in a first-use strategy of the kind discussed by Menon and Nagal, which is in the context of an India-Pakistan crisis, when Pakistan can be expected to have already dispersed its nuclear forces. For example, even targeting Pakistan's air-based nuclear weapons will be difficult. India will have to target all 25 to 30 airbases and airfields from where Pakistan could deploy its nuclear-capable bombers, instead of just half-a-dozen airbases that it could target in a true surprise first-strike. This does not exhaust the list of bomber-basing options for Pakistan, of course. Pakistan has over one hundred airports and it has even practiced landing aircraft on highways to disperse them in times of crisis. In addition, Pakistan's missiles are road-mobile, and it is unlikely that India will be able to find them after they are dispersed. Thus, a true surprise first-strike attack may be possible if India dramatically increased the size of its nuclear arsenal and achieved significant numerical superiority over Pakistan. But a counterforce attack on Pakistan in a crisis, after it is alerted, is simply not feasible even if India had a larger arsenal than Pakistan because India will not succeed in hitting more than a handful of Pakistan's nuclear weapons, which will achieve little even as a damage limitation exercise.

Pakistani army surrender guns in Dhaka, 1971

There are additional problems with moving towards a first-use strategy. Such a strategy requires India to achieve significant numerical superiority, which will lead to an arms race because Pakistan will be forced to respond, and it can depend on China to help out. More worryingly, it could also worsen crisis instability if both sides are worried that the other might launch first, a condition Thomas Schelling described decades ago as the 'reciprocal fear of surprise attack.'

Finally, the key question is about whether such a strategy will better serve to deter Pakistan because Menon and Nagal appear to be more interested in winning a nuclear exchange than in deterring it. While there are significant credibility problems in threatening to launch a massive nuclear attack on Pakistan's cities in response to a Pakistani TNW attack, the credibility problems in threatening a counterforce attack on Pakistan is greater because of the level of numerical superiority it requires, which India does not have and is unlikely to achieve. Non-credible threats dilute India's deterrence, and moving towards a counterforce strategy will only worsen India's nuclear deterrence credibility problem.

The proposals made by these former officials require consideration and should be debated. But it is unlikely that these proposals reflect official thinking or that they suggest India may be considering a first-use or first-strike strategy.

• Defence and Security
• Indian Defence
• China
• India
• Neighbourhood
• Counterforce
• First Strike
• First Use

The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.