In 2008 India Was Attacked and Did Not Strike Back. In 2025 It Took Fifteen Days

In November 2008, ten gunmen killed 166 people in Mumbai. India did not fire a shot in reply.

In April 2025, gunmen killed 26 people at Pahalgam in Kashmir. Fifteen days later Indian missiles were landing on targets inside Pakistan.

Those two facts belong to the same country, the same dispute and, in political terms, almost the same generation. Between them sits a change in doctrine that was never announced as a doctrine, was never written down as a white paper, and has probably done more to shape this part of the world than anything either government has said out loud.

This is the story of how India’s answer changed, told through five attacks. It is not a scoreboard. It is an account of how a state’s decision-making moved.

2001: mobilise everything, use none of it

On 13 December 2001, gunmen attacked the Indian Parliament complex in Delhi. Britannica attributes the attack to Lashkar-e-Taiba operating with Jaish-e-Mohammed. Parliament was in session.

India’s response was Operation Parakram: the largest military mobilisation the country had undertaken since 1971. Divisions moved to the western border. They stayed there for most of a year. And then, gradually, they went home.

No strike was made. What the mobilisation achieved is still argued over: some hold that it extracted diplomatic pressure on Pakistan at a moment when the United States needed Pakistani cooperation in Afghanistan, others that it cost a great deal of money and a significant number of lives in accidents and mine-laying for no strategic gain.

What is not argued over is the operational lesson, and Indian planners clearly absorbed it. Moving half a million men takes weeks, announces your intention to everyone including the adversary, and then leaves you with only two options: an all-out war you do not want, or a climbdown everyone can see. It is an instrument with no middle setting.

2008: the answer that was not military at all

Then came 26/11, and the most striking thing about India’s response is what did not happen.

There was no strike, no mobilisation, no ultimatum backed by force. The government of the day judged that a military response risked a war between two nuclear-armed states over an attack carried out by a non-state group, and chose not to take it.

What India did instead was build institutions: a federal investigating agency, regional deployment for its counter-terror force, a coastal watch that had not previously existed. That rebuild is a story in itself, and by most measures it worked. India has not had another attack of that scale on a metropolitan centre.

But restraint has a political cost, and it was paid. The decision not to respond militarily in 2008 became, over the following decade, the thing that any Indian government would be measured against. Whatever one thinks of it, that argument is the hinge on which everything after 2014 turns.

2016: the raid was not new, the press conference was

In September 2016 an army camp at Uri was attacked and a large number of soldiers were killed. Days later, the Indian Army announced that it had carried out surgical strikes: ground raids across the Line of Control against launch pads on the other side.

Here is the part that matters, and it is easy to miss. Cross-LoC raids were not an Indian invention of 2016. Both armies had operated across that line for years, quietly, and neither talked about it. What changed in 2016 was that India held a press conference.

Pakistan denied that the raids had happened as described. That denial is itself part of the design: an adversary that denies an event does not have to retaliate for it, which is exactly the escape hatch a publicised raid leaves open. India got to demonstrate a response domestically; Pakistan got to say nothing had occurred. Both could stop there.

It was, in effect, the middle setting that Operation Parakram had lacked.

2019: across the international boundary

On 14 February 2019 a suicide bomber drove an explosives-laden vehicle into a convoy of the Central Reserve Police Force at Pulwama in Jammu and Kashmir. Some 40 security personnel were killed. It was the deadliest attack on Indian forces in Kashmir in decades.

Twelve days later Indian aircraft struck at Balakot, and this was a different thing from 2016. Balakot is not in Pakistan-administered Kashmir. It is in Khyber Pakhtunkhwa, across the international boundary. India had used air power against a target inside undisputed Pakistani territory for the first time since 1971.

What the strike actually destroyed has never been settled, and this article will not pretend otherwise. Indian officials said a large Jaish-e-Mohammed training facility had been hit and a substantial number of militants killed. Pakistan said the aircraft had dropped their payload on an empty hillside and produced journalists at the site to make the point. Independent readings of satellite imagery published afterwards disagreed with each other. Britannica, notably, describes the target only as “what it said was” a Jaish-e-Mohammed camp. That hedge is the correct one and we are keeping it.

The following day Pakistani aircraft crossed in reply, an Indian pilot was shot down and captured, and was returned within days. Both governments declared the outcome a success. Then it stopped.

2025: the treaty, and then the missiles

On 22 April 2025 gunmen opened fire on tourists at Pahalgam, in Anantnag district. Twenty-six people were killed. A group calling itself The Resistance Front claimed responsibility and then retracted the claim. The Indian government linked the attack to Lashkar-e-Taiba.

India’s first move was not military, and it was arguably the more significant one. The Cabinet Committee on Security placed the Indus Waters Treaty of 1960 in abeyance, in the government’s own words until Pakistan “credibly and irrevocably abjures its support for cross-border terrorism”. It closed the Attari land crossing and withdrew visa exemptions for Pakistani nationals.

The Indus treaty had survived three wars. Suspending it moved the confrontation onto ground that had been treated as off limits for sixty-five years.

On 7 May, India struck multiple targets in Pakistan and Pakistan-administered Kashmir with missiles. Pakistan launched a retaliatory operation. What followed was short, sharp, and included the first use of drones by each country against the other. A ceasefire ended it on 10 May.

Both governments made extensive claims about what they had destroyed and what they had lost. Those claims contradict each other, and we are not going to adjudicate them here, because no account written now can. Note the pattern instead: an attack in April, a formal treaty suspension within days, missile strikes within a fortnight, and a stop inside four days.

What actually changed

Line up the five and the movement is unmistakable.

The time between attack and answer collapsed. Ten months of mobilisation in 2001. Nothing in 2008. Days in 2016. Twelve in 2019. Fifteen in 2025.

The geography expanded. Across the Line of Control in 2016. Across the international boundary in 2019. Multiple sites deep in Pakistan in 2025.

The response became loud on purpose. In 2016 the announcement was the innovation. By 2025 the naming, the briefings and the imagery were part of the operation rather than a footnote to it.

And the non-military levers came off the shelf. The water treaty is the clearest example of India using something other than force and other than diplomacy, and it may end up mattering longer than any strike.

The question nobody has settled

Has it worked? There are two serious answers and this magazine is not going to pretend one of them is obviously right.

The case that it has: since 2016 the cost to Pakistan of an attack traceable to groups on its soil has stopped being purely diplomatic. India has demonstrated four times that it will respond in ways that are visible, and each round has ended without a wider war. Deterrence that has been tested and held is not nothing, and there has been no repeat of 26/11.

The case that it has not: attacks have continued through the whole period, including Pahalgam itself, which happened after nine years of the new approach. Each exchange has begun from a higher rung than the last. Two nuclear-armed neighbours have now normalised a pattern where missiles and drones are an ordinary response to a terrorist attack, and every round depends on both sides choosing to stop, quickly, while a great deal is happening at once. That has worked every time so far. “So far” is doing a lot of work in that sentence.

Both readings are held by serious people in India. Anyone who tells you the question is closed is selling something.

The part that is not in doubt

Underneath the doctrinal argument there is a plainer fact, and it deserves the last word.

In every one of these episodes, the people who died first were not soldiers making strategy. They were a shift of CRPF men on a road at Pulwama. Commuters at a Mumbai station. Tourists on a meadow at Pahalgam, and a local pony handler who worked there.

India has spent twenty-five years getting better at what it does afterwards. It has not yet solved the thing that keeps making an afterwards necessary. Those are two different problems, and only one of them has been the subject of all this ingenuity.