WE have witnessed an eventful couple of weeks in India-Pakistan ties. Indian Prime Minister Modi surprised everyone with his impromptu stopover in Lahore. But even before one could fully digest this, Pathankot offered a stark reminder of the challenges ahead.

In these twin developments lies one crucial takeaway: dialogue is an absolute compulsion.

Above all, Lahore signified a hard-line leader’s grudging acceptance that dialoguing can’t be wished away. Modi’s flip-flop policy on Pakistan since he came into office was only consistent in its aim to play hardball with Islamabad. But his attempt to ignore Pakistan (as it seemed to many here) was never going to work, for the same reason it didn’t for his predecessors: the presence of nuclear weapons has made war suicidal and the world powers are acutely sensitive to developments that may increase the risk of an India-Pakistan crisis.

This compulsion was at play when president Musharraf, the architect of the Kargil war who proudly talked about Kashmir and only Kashmir as army chief, was forced to take a U-turn once he became responsible for the state and its people — not just the army. This reality is also what led prime minister Vajpayee to offer friendship to Pakistan after a 10-month long military stand-off in 2001-02; got prime minister Manmohan Singh to come back to dialogue after Mumbai; and has led Prime Minister Nawaz Sharif to champion talks with India consistently.

Dialogue is an absolute compulsion for India and Pakistan.

This reality is as permanent as it gets. With nuclear weapons in the mix, the no-dialogue option can do little more than make you look bad and force you to seek a face-saver to return to the negotiating table sooner or later.

To be sure, dialogue without the will to resolve issues won’t get us far. But even as an exercise in grandstanding, there is immense utility to talking in crisis-prone environments. It provides a way to avoid extreme negatives by creating an option to talk constructively at tense moments or to use communication channels to blow off steam.

Pathankot reminded us of its relevance. Lahore (and the positive signals between Delhi and Islamabad that led to it) and the communication it allowed post-Pathankot must be credited for the non-crisis it ended up being — at a time when the Indian media had defaulted to a war footing and their Pakistani counterparts were making things worse by insinuating a false flag operation.

Importantly, this is a generalisable observation. In every crisis since Kargil, the media has been far ahead of officialdom in demanding blood. But each time India and Pakistan have had a dialogue mechanism in place prior to the crisis-triggering event; they have managed to keep a lid on things with little difficulty. The exception was the 2001-02 stand-off where a far less spectacular incident than Mumbai, or even Kargil, resulted in an eyeball-to-eye ball confrontation.

I’d be the first one to admit that there is no guarantee that this pattern will hold. There is enough evidence from other contexts that exposes the limits of generic dialogue and the goodwill it generates.

Dialogue specifically aimed at crisis prevention and mitigation must therefore be taken beyond the generic level to actual coordination — even in the absence of any hope of mega breakthroughs at the strategic level.

What we require is direct engagement between ISI and RAW on terrorism prevention and post-terrorist incident evidence-sharing. No diplomacy, no bureaucracy but real intel-to-intel talk and coordination aimed at sharing real-time information to help avoid terrorist attacks and at coming clean or demonstrating action against those responsible if an attack occurs.

How can this work when these states perceive each other to be the actual perpetrators? My response is that if the use of non-state and subversive actors as a policy tool — in any form — is not going to be given up by these states, no protocol can work. But by getting fixated on this aspect, India and Pakistan have left a gaping hole to exploit for truly autonomous third-state actors with huge incentives to push them into war.

To my mind, the most probable scenario for a future India-Pakistan war is a Daesh-type entity playing catalyst. And they are far more likely to cause a catastrophic outcome as unlike state-sponsored groups constrained by their patrons, they will aim for the biggest bang possible. Intelligence coordination could well be the difference between an amicably managed non-crisis and an uncontrollably escalated one in such a situation.

Modi’s Lahore visit and Pathankot are the latest reminders to both sides that trying to dodge dialogue is counterproductive. Both sides would do well to internalise this reality so that they get out of the habit of trying to find a way around talking every now and then, wasting precious time before returning to the inevitable.

The writer is a foreign policy expert based in Washington, DC.

Published in Dawn, January 5th, 2016