The Army Chief Upendra Dwivedi’s “Smart Power” Speech Was a Confession, Not a Doctrine

On 19 May 2026, at the Manekshaw Centre in New Delhi, General Upendra Dwivedi gave the headline writers exactly what they came for. Operation Sindoor, a year on. The 22-minute strike window. Nine targets across Bahawalpur, Muridke, Sialkot and Pakistan-administered Kashmir. The deliberate halt at 88 hours. All of it dutifully reported the next morning, and forgotten by the one after.

The most consequential sentence in that speech was about none of those things. It was about India. And it read less like a statement of doctrine than an admission the Army does not often make from a public stage.

Speaking at a Centre for Land Warfare Studies seminar titled “Security to Prosperity: Smart Power for Sustained National Growth,” Dwivedi argued that the boundary between security and prosperity has effectively dissolved. He reached for the familiar DIME framework, the diplomatic, informational, military and economic instruments of statecraft, and said it now needs a fifth element, technology, plus a “whole of nation” approach to make the other four move together.

Read that again. When a serving Chief of Army Staff stands at a think-tank podium and calls for the nation’s instruments to act as one coherent whole, he is telling you, plainly, that they currently do not. The call is the diagnosis. Everything worth analysing in the speech follows from that single inversion.

The 22-Minute Strike Was the Easy Part

The operational facts of Sindoor are now in the public record. The strike on the night of 6–7 May 2025 answered the 22 April Pahalgam attack, which killed twenty-six civilians. It was a tri-service action against nine terror-linked targets, executed with precision-guided munitions, inside a defined 22-minute window, and then stopped after 88 hours.

Dwivedi framed that halt as “smart power in its most complete expression”: the discipline to know which instrument to apply, at what intensity, and when to stop. It is a clean story. It is also where the analysis has to slow down, because the clean story conceals the harder question.

Choice or constraint?

A halt is restraint only if continuing was a real option. If India stopped because it chose to, the 88-hour ceiling demonstrates escalation control, the ability to hold the upper hand even as pressure builds. If India stopped because continuing was becoming difficult, whether through magazine depth, the arithmetic of sustained precision strikes, diplomatic pressure, or the management of Pakistani nuclear signalling, then the halt was a constraint wearing the costume of a choice.

From outside, we cannot know which it was, and honesty requires saying so. But the distinction is not cosmetic. The entire logic of escalation dominance rests on one assumption: that the option to go further actually exists. If that option was thinner than the official framing suggests, an adversary will read the constraint before India’s own commentary does. Posture is what you signal. Capability is what survives contact.

Doctrine is the theory that connects the two. Operation Sindoor showed India can compress the first strike into twenty-two minutes. It said almost nothing about what happens in the eighty-ninth hour of a crisis that does not end on India’s schedule.

Striking is a solved problem. Sustaining the position after the strike, on terms you set, is the unsolved one.

Two Vulnerabilities, Not One Supply Chain

Dwivedi offered two illustrations of how interdependence has become coercion: the Strait of Hormuz, which he called a zone of active contestation, and semiconductors, whose selective availability he described as a tool of leverage.

He also noted, correctly, that global defence spending has crossed 2.7 trillion dollars, more than the entire UN budget for the Sustainable Development Goals.

It is tempting to file both examples under “supply-chain risk” and move on. That would be an analytical error, because they are different vulnerabilities with different clocks and different cures.

Hormuz is a vulnerability of geography, and it is immediate. India imports the large majority of its crude, and depending on the year, roughly 40 to 50 percent of it transits Hormuz; that share has risen recently as Indian refiners trimmed Russian purchases.

The cushion is thinner than the headline storage figures suggest. The dedicated Strategic Petroleum Reserve covers only about nine to ten days of imports.

Counting the commercial stocks held by state oil companies, total national storage reaches roughly seventy days, but a sustained chokepoint disruption bites into an economy long before tanks run dry.

This vulnerability is not fixed by another fighter squadron. It is fixed by storage, bypass routes, and diplomatic reach in the Gulf. The answer does not sit with the armed forces at all.

The semiconductor vulnerability is structural, and it is long. India has finally moved past the announcement stage: the Tata–PSMC fab at Dholera is under construction with first silicon targeted for late 2026, and Micron’s assembly and test facility at Sanand was inaugurated in February 2026. But the relevant detail is the node.

These plants are built for 28-nanometre and larger geometries, the mature nodes that run cars, appliances and industrial electronics. That is sensible industrial policy.

It is not sovereignty over the advanced silicon that sits inside seekers, AESA radars and modern datalinks. Those chips still come from abroad, and India’s own roadmap places domestic leading-edge production somewhere around the middle of the next decade. What comes from abroad can be withheld at precisely the moment it is most needed.

One vulnerability needs an answer by next monsoon. The other will resist solution for a decade. Collapsing them into a single talking point is rhetorically convenient and strategically dangerous.

DIME Is Breaking Down, and Adding a Letter Will Not Fix It

The deeper problem with the speech is not the content of the prescription but what the prescription reveals. A state can wield national power “as one coherent act” only if it has a spine to coordinate it. India does not have that spine, and the gap is older and more domestic than any external threat.

Start with the most authoritative tell, the one that sits at the top of any honest evidence hierarchy: published doctrine. India has no public national security strategy. There is no document that states the country’s threats, ranks its priorities, and sequences its resources accordingly.

This is not the same as having no thinking; the thinking exists, scattered across ministries and service headquarters, each optimising for its own mandate. What is missing is the standing machinery to fuse it under pressure.

This is why “add technology to DIME” misreads the problem. The DIME construct assumed that diplomacy, information, military force and economics were separable boxes, each managed by its own institution. Technology dissolves that assumption.

A single satellite constellation supports military targeting, civilian communications, and the timing layer that financial markets run on. A single undersea cable carries both a nation’s data and the heartbeat of its payments system.

A single class of chip sits in a missile’s guidance section and in a consumer handset. Technology does not stand beside the four pillars. It runs through all of them at once. So the real question is not where to insert a “T.”

It is who, in a real crisis, sits at one table and reads a shock that strikes diplomacy, information, economy and military readiness simultaneously. Today there is no clear answer. “Whole of nation” is the polite name for that empty chair.

The Fragmentation Is Indian, Not Chinese

There is a reflex in Indian strategic writing to explain every domestic shortfall through Beijing. China did X, therefore India failed to do Y. It is the easy frame, and because it is easy, it has become the default. It is also, often, wrong.

India’s fragmentation is not a reaction to China. Its roots are internal, and they predate China’s rise by decades.

The acquisition system routinely takes the better part of a decade to move a platform from sanction to induction, a pathology with no connection to the Line of Actual Control. The industrial base is fractured, with DRDO, the defence public-sector undertakings and private industry behaving more as rivals than as a single production ecosystem.

The budget itself is the constraint that gets least attention: revenue expenditure, dominated by salaries and pensions, steadily crowds out the capital expenditure that buys new capability, so even agreed modernisation competes for a shrinking slice.

And the services pursue their own institutional interests, which is why theaterisation, the single most important structural reform on the table, has stalled in debate for years rather than moving to implementation.

None of this was authored in Beijing. Treating China as the cause is a way of deferring the diagnosis. If the real driver is internal, the cure is internal too. The external threat is a factor in the timing. It is not the origin of the disease.

Long Wars Are Won in Factories

Ukraine settled an argument that two decades of expeditionary, precision-centric thinking had let drift. Long wars are not decided by the courage of the first echelon. They are decided by the depth of the magazine and the throughput of the factory behind it.

Artillery consumption ran into the millions of rounds a year, and even the most advanced Western industrial bases struggled to keep pace with demand they had stopped planning for.

Dwivedi’s emphasis on industrial capacity and research systems is the right instinct. The trouble is that India’s record here is unforgiving, and recent rather than historical.

A 2017 audit by the Comptroller and Auditor General found that for roughly 40% of the 152 critical ammunition types, war-wastage reserves stood at only about ten days of intense fighting, against a stated norm of forty days and a minimum acceptable level of twenty.

The worst shortfalls were in artillery and tank ammunition, precisely the natures that decide a sustained land fight. Emergency procurement powers were subsequently delegated to the Vice Chief, and the picture has reportedly improved since.

But the structural point survives every revision. Buying stock is one thing. Producing it continuously, for months, while under fire, is another. The first is solved with money. The second is solved only with industrial capacity that exists before the war begins.

Make the scale human. Sustaining an armoured regiment through a Ladakh winter is not a question of valour but of tonnage: fuel hauled up hundreds of kilometres of fragile road, in temperatures where hydraulic and lubricating fluids stop behaving predictably, week after week, before a single round is fired.

India lost in 1962 less to tactics than to that logistics chain. Kargil in 1999 was won with artillery, but even then the adequacy of shell stocks was a live worry. The lesson keeps arriving, and keeps being half-learned.

India today sits in the most dangerous phase of all: independent enough to make large claims, not yet independent enough to fight a long war alone. Critical engines, advanced electronics and key sub-systems still arrive from abroad. “Atmanirbhar Bharat” is a compelling slogan and an incomplete reality, and the gap between the two is exactly where an adversary would aim.

China Is the Mirror, Not the Cause

Respecting an adversary’s capability is not pessimism; it is the precondition for serious analysis. China has to be read concretely, not as a slogan, and the comparison is sobering.

The scale is the place to begin. The People’s Liberation Army Navy is now the largest in the world by hull count, with over 370 ships and submarines per the Pentagon’s most recent China Military Power Report.

The fleet behind the fleet matters even more: the Office of Naval Intelligence estimates Chinese shipbuilding capacity at many times that of the United States, on the order of half of total global capacity.

In a long war, that is the decisive metric. The side that can build hulls, shells and spares faster is the side that endures, and the relevant comparison is not the Indian Navy’s tonnage in isolation but the tonnage Chinese yards can launch in the time it takes India to commission a single major combatant.

Then there is the plateau. Over the past decade China has laid down a dense lattice of roads, rail, airfields and dual-use border settlements across Tibet, which means it can now move and sustain forces at altitude through the year, at speed and in mass. That is the logistics depth whose absence broke India in 1962. The advantage has changed hands.

And beneath both sits the deepest point. China’s fusion of technology, industry and military, its formally declared Military-Civil Fusion policy, is precisely the “whole of nation” integration Dwivedi was asking for. The difference is that in China it already exists as state architecture, while in India it remains an aspiration voiced from a podium.

Hold the discipline here, or the analysis slides back into its old habit. China is not the cause of India’s internal weaknesses. It is the clearest mirror of them. Placed alongside Beijing, India’s gaps look sharper, but those gaps would persist with or without China across the Himalayas. What China demands is not a faster trigger but the capacity to last.

This is also why the two-front problem cannot be met with a single template. Pakistan weaponises instability: the sudden crisis, the border flare-up, the proxy. Its strength lies in the shock, not the duration, and against it India needs the fast, precise response Sindoor demonstrated.

China plays the long structural game across technology, industry, the sea lanes and the frontier, and against it India needs endurance. The exhaustion built into India’s position is that Pakistan keeps seizing the attention while China quietly builds the architecture that will matter a decade out. Mistaking the urgent for the important is the costliest error available.

India and the PLA, by Axis

A number in isolation says nothing. Squadron strength, fleet size, a budget line: each becomes analysis only against adversary force structure, in a specific geography, under realistic conditions.

The balance below is read across the axes that decide a sustained campaign, not the headline counts that decide a news cycle.

Figures are weighted by source reliability: India’s Union Budget 2026–27 and PRS analysis; the US DoD China Military Power Report; US Office of Naval Intelligence estimates on shipbuilding; the 2017 CAG audit on ammunition reserves.

Hull and aircraft totals are approximate, since trackers differ on what counts as a front-line combatant, and flagged figures should be reconciled against the current IISS Military Balance before publication.

The Test Is Not Whether India Can Strike

“Read the world as it is, not as you wish it to be,” Dwivedi told his audience. The instruction applies to his own speech.

Sindoor proved India can organise a strike inside twenty-two minutes. But modern coercion does not run on a 22-minute clock. It runs for weeks, and it arrives from several directions at once: cargo held up at a port, insurance premiums climbing, markets wobbling, a disinformation surge, and somewhere in the middle of it, a small incident on the line.

In that kind of crisis, the question is not whether India can act. It is whether the Indian state can act as one when the petroleum ministry, the finance ministry, the services and the foreign office all have to move in the same direction at the same time.

And underneath even that lies the question the speech circled but never named, the one no published document currently answers. On the eighty-ninth hour, who holds the authority to decide?