Two weeks on from the worst attack on Indian soldiers in Kashmir for decades, the dust is starting to settle. Many Indian politicians, press, and analysts had struck a relentlessly hostile note, demanding that New Delhi take (in their view) long overdue military action against the Pakistani terrorists who routinely conduct cross-border attacks, and the Pakistan Army's intelligence service, which has for decades aided and abetted them. There had also been unprecedented levels of interest in alternatives to military action, ranging from covert warfare to abrogation of the historic Indus Waters Treaty. But it is now clear that this government, like both its two immediate predecessors, is aiming to lower expectations of an overt strike, while focusing instead on a coordinated campaign to isolate Pakistan diplomatically. 

Immediately after the Uri attack on 16 September, a senior leader of the ruling Bharatiya Janata Party (BJP) demanded severe retaliation: 'for one tooth, the complete jaw'. 'Pakistan is a terrorist state', tweeted the home minister, 'and it should be identified and isolated as such'. 'Every Pakistan post through which infiltration takes place should be reduced to rubble by artillery fire', argued the retired Brigadier Gurmeet Kanwal. 'Action against Pakistan is an imperative for national morale', echoed retired Lieutenant General HS Panag, urging a '(Kashmir)-centric limited war'. An intelligent debate unfolded in sections of the print media, with more cautious voices like Pratap Mehta, Sushant Singh, Ajai Shukla, Suhasini Haidar, Arun Pakash and Manoj Joshi expressing doubts over both the feasibility and desirability of punishing Pakistan with force. 

On 24 September, after a three-day party conclave in the southern city of Kozhikode, Prime Minister Narendra Modi gave a nuanced speech that indicated where the government stood. He made no reference to retaliation or punishment, challenged Pakistan to a thousand-year war to eradicate poverty, and sharply distinguished between ordinary Pakistanis and the government 'making you fools in the name of Kashmir'. Yet he also promised that India would 'leave no stone unturned to isolate Pakistan in the world' and that 'we will ramp it up and force you to live alone in the world'. Indeed, India has already ramped it up, with aggressive speeches at the UN General Assembly by a junior diplomat, who called Pakistan the 'Ivy League of terrorism', and then the foreign minister herself on Monday.

Modi also hinted at his reasons for restraint: 'the whole world recognises India to be the world's fastest growing major economy'.

Even if India could strike at Pakistan, doing so would place at risk its wider economic and diplomatic objectives, including rapid development and heightened integration into key global institutions like the Nuclear Suppliers Group (NSG) and the UN Security Council. It is crucial to understand that this restraint is not itself a fixed policy, but a function of several factors. Some are broadly static, such as the economic and diplomatic stakes that might be threatened by a regional war. Others, however, are more contingent. One of these factors is the state of Indian military and intelligence capabilities, which are likely to adapt over time in line with funding and political attention. Such deep-seated reform (which the Economist discussed briefly last week) will take years.

Another factor, however, is the trigger itself. The Uri attack killed 18 soldiers, but this has to be understood in the context of the death of 135 members of Indian security forces this year alone, including 64 in Kashmir. It may be harsh to say so, but Uri is not an attack on the scale of Mumbai in 2008, and so the serious risks that accompany air strikes or ground warfare are not realistically on the table. What this does suggest is that Modi's homilies on economic growth and war on poverty could well be set aside when (not if) we see the next mass casualty attack, political assassination, or similar provocation from Pakistan-backed actors. Restraint now does not guarantee restraint in the future; it may even make it less likely, as public pressure accumulates. 

More broadly, the Uri attack has catalysed Indian debate over non-military deterrence and coercive options. On Monday, Modi chaired a meeting, attended by both his foreign secretary and national security advisor, on the Indus Waters Treaty, a landmark 1960 agreement that governs how India (the upper riparian state) shares river water with Pakistan. The IWT has endured through several wars and crises, and the government is claiming that the treaty itself isn't under review. But it has suspended talks in the Permanent Indus Commission, the treaty's joint dispute resolution body, and declared that it will go from using 11,000 to 18,000 megawatts of power from the western rivers, which will have an impact on Pakistan. All this is a clear signal that even the most durable of India-Pakistan agreements is at risk should attacks recur. Indian critics have, however, pointed out that India is the lower riparian with respect to China and is seeking similar agreements with other neighbours like Bangladesh and Nepal, so abrogation could set a dangerous precedent for New Delhi.

Elsewhere, India's former national security advisor (2005-11) MK 'Mike' Narayanan argued in The Hindu that restraint in the past had 'greatly added to India's prestige', and that Indian forces were anyway incapable of 'spectacular raids' using special forces. 'India's best option', he suggested instead, 'would be to engage in cyber sabotage and cyberwarfare, hiding behind the plausible deniability available in such attacks. Our capacity in this area is considerable, and it should be possible to...bring Pakistan to its knees'.

This seems unduly confident. As Arun Sukumar explains in an excellent brief, India has fairly limited offensive cyber-capabilities and has more to lose from any change to the 'stability of cyberspace in South Asia'. Moreover, it seems highly unlikely that Pakistan, a nation that lost 1000 civilians to terrorism last year, would be crippled by virtual damage. Indian hackers 'once took down Pakistan Railways' website', noted one Pakistani on Twitter, wryly. 'Nobody brought it back up for at least a year'. 

This curious ferment of ideas, some more coherent than others, is a reflection of Indians' growing frustration at their apparent impotence in the face of major attacks in 2001-02, 2008, and now 2016. This problem is intractable. Pakistan's relationship to non-state armed groups is a fundamental part of its security strategy, linked both to Pakistan's revisionist aims in Kashmir and its broader insecurity around a rapidly rising India. This strategy has contributed both to the ravaging of Pakistan in militant violence for over a decade and to growing alienation from the US. If these costs have not altered Pakistan's calculus, India would have to impose one larger still.

Yet any steps that do so, such as the seizure of Pakistani territory or the degradation of the Pakistani army, would catalyse support for Pakistan's army and anti-Indian jihadist groups, generate costs for India in excess of the original terrorist provocations, and eventually bring nuclear weapons into play. Slow-burning, non-kinetic options (like weaponising water, or supporting Pakistani separatists) might limit the third of these risks, but not necessarily the first two. And yet, although smaller steps that reduce the likelihood of escalation (from angry speeches all the way up to symbolic air strikes in disputed territory) may mollify Modi's restive domestic constituents, they won't make a dent on Pakistan's calculus. Addressing state sponsorship of terrorism is a fiendishly difficult problem, as Israel has found with respect to Iran, and only coercion and diplomacy in parallel are likely to work over the longer term.

 Photo: Getty Images/Hindustan Times.