India’s Nuclear Doctrine and Deterrence in South Asia

India tested in 1998 and declared a doctrine in 2003, and the five years between them are the whole problem in miniature. A state that has weapons must say what they are for, because weapons whose purpose is unstated deter nobody and reassure nobody. What India said was unusually restrictive — retaliation only, and only against nuclear use — and the question that has run ever since is whether a country can hold that line while its capabilities, its adversaries and its politics all move.

What a nuclear doctrine is and why states publish one

A nuclear doctrine is a public statement of the conditions under which a state would use nuclear weapons, the purposes those weapons serve, and the authority that would release them. It is not a war plan and it is not a targeting list; it is the declaratory layer that sits above both.

The three functions of a doctrine

• Internally, a doctrine guides the armed forces and reassures the public. Without a stated role, the services cannot size, deploy, exercise or budget for a nuclear force, and the political system cannot hold that force to account.
  • It tells the Strategic Forces Command what it is preparing to do and, just as importantly, what it is not preparing to do.
• Externally, a doctrine signals to adversaries. Deterrence works on the adversary’s calculation, so the adversary must be able to read the threat.
  • K. Subrahmanyam’s formulation is the sharpest here — deterrence is about perception rather than posture, and a posture nobody can interpret produces no perception at all.
  • A declaratory line also tells the adversary where the threshold lies, which is information the adversary needs in order to stay below it.
• Third, a doctrine reduces the risk of inadvertent war. Nuclear weapons are not conventional weapons scaled up, and a crisis between two nuclear states is a contest in mutual inference under time pressure.
  • Clarity about the release authority removes another, by telling the adversary that a local commander cannot start a nuclear war on his own initiative.

A doctrine is not a promise about the future. It is a device for making one’s own behaviour predictable enough that an adversary can safely choose not to escalate.

Declaratory doctrine and operational doctrine

• Declaratory doctrine is what a state says; operational doctrine is what its forces are configured to do. The two are never identical, and the gap between them is where most of the argument about India sits.
  • Declaration is cheap to state and impossible to verify — no adversary can inspect an Indian storage site to confirm that warheads are de-mated.
  • Operational doctrine is inferred from capability, deployment pattern, exercise behaviour and procurement, which is why canisterisation and MIRVing are read as evidence about intentions.
• A gap is not by itself deception. Capabilities are built for a range of contingencies and outlive the doctrine that authorised them, and hedging is a normal feature of any nuclear force.
• India’s gap runs in an unusual direction — the declaration is more restrained than the capability, the reverse of the more familiar case where a state talks bigger than it can act.

Ambiguity: the case for and against

	The case for ambiguity	The case for a declared doctrine
Deterrent effect	An adversary uncertain where the line lies must be cautious everywhere	An adversary who knows the line can be deterred from crossing it
Escalation control	Leaves room to respond proportionately without being trapped by a prior commitment	Removes the incentive to pre-empt, since neither side fears a bolt from the blue
Force requirements	Permits a smaller, opaque, undeclared arsenal at low political cost	Anchors the force to an assured second strike, which caps rather than expands it
India’s choice	The recessed deterrent of 1974–98 was ambiguity in practice	After 1998 a declared state had to declare a purpose

• India moved from ambiguity to declaration because testing removed the option. Once the weapons were overt, silence would have been read either as irresolution or as concealed war-fighting intent.
• Ambiguity is not free even for those who prefer it. It requires the adversary to assume the worst, and an adversary assuming the worst builds against the worst.

The two texts

India has produced exactly two doctrinal documents of any standing, only one of which is policy, and the distinction between them is frequently collapsed.

The first comprehensive statement of policy

• Evolution of India’s Nuclear Policy was tabled in Parliament in May 1998, days after the Pokhran II tests, and is the first comprehensive official statement of the Indian position.
  • Its central claim was that nuclear weapons are not weapons of war but instruments of self-defence and of insurance against coercion.
  • It framed the arsenal as an answer to being threatened, not a means of threatening — a formulation every later text has kept.
  • It is a policy rationale rather than a doctrine: it says why India has the weapons, not under what conditions it would use them.

The Draft Report of the National Security Advisory Board

• The Draft Report was released on 17 August 1999 by the National Security Advisory Board, convened under the National Security Council apparatus and working under the National Security Adviser Brajesh Mishra.
  • The board was a body of non-official experts — strategists, retired officers, scientists and academics — with K. Subrahmanyam as convenor, and its report was a recommendation to government, not a decision by it.
  • It was never adopted as policy. Foreign Minister Jaswant Singh stated that it was not a policy document of the Government of India but the work of a non-official group, released for public debate.
• What the draft said. Its architecture survives in the 2003 text, but its ambitions were larger.
  • Credible minimum deterrence as the organising principle, defined as a dynamic concept related to the strategic environment rather than a fixed number.
  • “India will not be the first to initiate a nuclear strike” — no first use in its plainest form, without the chemical and biological qualification.
  • Punitive retaliation sufficient to inflict damage unacceptable to the aggressor, the phrase later hardened into “massive”.
  • A triad of aircraft, mobile land-based missiles and sea-based assets, justified by survivability through redundancy, mobility, dispersion and deception.
  • Release authority with the Prime Minister or the designated successor, and an insistence on survivable and enduring command and control.
  • A force able to survive repeated attrition and still retaliate, which is the technical content of the word “credible”.
• Why the draft was controversial. The triad language implied an arsenal on the scale of Britain’s or France’s, which sat awkwardly with “minimum” and alarmed the non-proliferation community and India’s own fiscal realists alike.
  • The government let it stand as a discussion paper and, when it eventually stated its own position, compressed the triad language out.

The Cabinet Committee on Security review

The operative text is a press release. On 4 January 2003 the Cabinet Committee on Security met to review the operationalization of India’s nuclear doctrine, and the statement issued that day is the only doctrinal document ever approved by the Indian government. It is short, and its brevity is deliberate.

The eight elements of the 2003 doctrine, as stated:

1. Building and maintaining a credible minimum deterrent.
2. A posture of “No First Use” — nuclear weapons will only be used in retaliation against a nuclear attack on Indian territory or on Indian forces anywhere.
3. Nuclear retaliation to a first strike will be massive and designed to inflict unacceptable damage.
4. Retaliatory attacks can only be authorised by the civilian political leadership through the Nuclear Command Authority.
5. Non-use of nuclear weapons against non-nuclear weapon states.
6. In the event of a major attack by biological or chemical weapons against India or on Indian forces anywhere, India will retain the option of retaliating with nuclear weapons.
7. Continued strict export controls on nuclear and missile-related materials and technologies, participation in Fissile Material Cutoff Treaty negotiations, and continued observance of the moratorium on nuclear tests.
8. Continued commitment to the goal of a nuclear weapon free world, through global, verifiable and non-discriminatory disarmament.

• The same statement created the command architecture. It announced the Nuclear Command Authority with its Political Council and Executive Council, and directed the operationalisation of the arsenal, including the appointment of a Commander-in-Chief, Strategic Forces Command.
• The 2003 text is a press release, not an act, a treaty or a white paper, and India has never published an elaborated doctrinal document since — a fact both sides of the ambiguity argument cite in their favour.

What 2003 added to, and subtracted from, the draft

Element	1999 Draft Report	2003 CCS statement
Status	Recommendation by a non-official board, never adopted	Approved government policy, the operative text
No first use	Unqualified — India will not initiate a nuclear strike	Qualified by the chemical and biological clause
Chemical and biological weapons	Absent	Added — nuclear retaliation reserved against a major CBW attack
Retaliation	Punitive, inflicting unacceptable damage	Massive, inflicting unacceptable damage
Force structure	Explicit triad of air, land and sea legs	Triad language compressed out; no structure specified
Command	Release authority with the Prime Minister or successors	Formal Nuclear Command Authority with two councils
Non-nuclear states	Implicit	Explicit negative security assurance
Regime commitments	General	Export controls, FMCT participation, testing moratorium written in

• The chemical and biological clause is the single most consequential addition. It converts no first use from a pledge of non-initiation into a narrower one: no first use of nuclear weapons in response to anything other than a weapon of mass destruction.
  • This is not later erosion. It is in the founding text, and any description of India’s NFU as unconditional misstates the policy it describes.
  • The clause exists because India, having ratified the Chemical Weapons Convention and holding no chemical or biological arsenal of its own, needed some deterrent against those categories, and nuclear weapons were the only instrument available.
• Dropping the triad language was a retreat from specificity, not from capability. India has since built all three legs, but without ever committing itself in writing to a force structure against which it could then be measured.

What the doctrine says about how India sees itself

• The doctrine treats nuclear weapons as political instruments rather than military ones, and that premise explains every restrictive clause in it.
  • The weapons exist to deny others the use of nuclear coercion against India, not to win wars or to hold territory, which is why there is no battlefield role and no tactical yield in the Indian inventory.
  • The text also positions India as a responsible nuclear-armed state, prepared to build confidence-building measures in the region and to keep its regime commitments — restraint offered as evidence of fitness.
• The doctrine has been read as an expression of defensive realism. It assumes a threatening environment and answers it with the minimum capability that removes the threat, rather than with a bid for advantage.
  • Raja Menon’s reading goes further: India’s nuclear doctrine reflects not so much India’s strategic choice as India’s culture, a claim that the restraint is prior to the strategy rather than produced by it.
  • Rakesh Sood describes India as a reluctant nuclear-armed state, which he treats as its distinguishing feature — a reluctance rooted in a non-violent freedom struggle and in an early commitment to a nuclear-weapon-free world.
  • The tension in that self-description is real: India acquired a tradition of restraint in the use of violence while also acquiring the capacity to coerce, to be used when everything else has failed.
• The cultural reading has an obvious weakness. It explains why India chose restraint but cannot explain why the same state built a triad, MIRVed a missile and struck across an international border in 2019 and 2025.
  • The safer formulation is that restraint was chosen because it was cheap and useful, and that it will last exactly as long as those conditions do.

Credible minimum deterrence

Three phrases carry the entire doctrine, and each of them is doing more work than its brevity suggests. Credible minimum deterrence sets the size and shape of the force; no first use sets the condition of employment; massive retaliation sets the character of the response. They are mutually dependent, and an attack on any one of them is an attack on the whole.

What “minimum” means, and what it does not

• “Minimum” is not a number and was never meant to be one. The 1999 draft called it a dynamic concept related to the strategic environment, and no Indian document has ever specified a warhead ceiling, a delivery-vehicle count or a target list.
  • The size of the force is set against what the adversary has and can protect, not against an absolute standard of sufficiency.
  • Because the adversary’s arsenal grows, a minimum deterrent can grow while remaining minimum — which is either the concept’s central honesty or its central evasion, depending on who is arguing.
• The reference point is unacceptable damage, not force parity. India does not attempt to match China warhead for warhead and has never claimed to.
  • The requirement is that enough survives a first strike to inflict costs the adversary would not accept.
  • That converts the sizing problem from an arithmetic of arsenals into an arithmetic of survivability, which is why command survival and dispersal matter more to the doctrine than raw numbers.
• Minimalism is a choice against the Cold War model. The doctrine’s authors read the American and Soviet arsenals as a wasteful spiral driven by war-fighting requirements India had no reason to adopt.
  • Rajesh Basrur is the most systematic Indian exponent of nuclear minimalism, arguing that the political utility of nuclear weapons saturates very quickly, that beyond a small survivable force additional weapons buy nothing, and that the pressure to build more comes from organisational and political sources rather than from strategic logic.
  • On this reading, the numbers game is a category error: nuclear weapons deter by existing and surviving, not by counting.

Existential, minimum and assured deterrence

Concept	What the force must do	Force implications	Where it fits India
Existential deterrence	Merely exist and be suspected; the adversary cannot rule out use	Undeclared, unassembled, possibly untested	The recessed deterrent of 1974–98
Minimum deterrence	Survive a first strike and retaliate against a few high-value targets	Small, dispersed, de-mated, second-strike only	The declared Indian position since 1998–2003
Assured retaliation	Guarantee a second strike under all conditions, including decapitation	Survivable triad, hardened command, sea leg	Where India’s capability has actually moved
War-fighting / counterforce	Destroy the adversary’s forces, limit damage to oneself	Large, accurate, high-readiness, intelligence-intensive	Explicitly rejected in doctrine, debated in analysis

• The three are points on a continuum, not rival schools. Each demands more of the force than the one before it, and each therefore costs more and constrains the adversary less.
• India’s declaratory position sits at minimum deterrence; its procurement sits closer to assured retaliation. That is a defensible evolution — assurance is what makes minimalism credible — but it is also the opening through which the counterforce argument enters.

Why credibility and minimality pull against each other

• Credibility asks for more; minimality asks for less. A force credible enough to guarantee retaliation after a determined first strike must be larger, more dispersed and more ready than a force sized to a bare notion of sufficiency.
  • Every measure that raises survivability — canisterisation, mobility, sea basing, MIRVs, hardened command — also raises capability and readiness in ways an adversary reads as offensive.
  • Every measure that keeps the force minimal — de-mated storage, small numbers, low readiness — raises the adversary’s hope that a first strike might work.
• The doctrine resolves the tension by definition rather than by design. By making credibility a property of survivability rather than of size, it permits India to improve the force indefinitely without ever conceding that “minimum” has been abandoned.
• The Chinese comparison was always the reassurance. Shivshankar Menon has pointed out that China maintains no first use despite gross asymmetry with the United States, and derives credible deterrence from survivability rather than numbers — the same logic India applies to itself.

The scholars on the shape of the force

• Rajesh Rajagopalan grounds the small arsenal in India’s threat environment rather than in ethics. His argument is that nuclear weapons answer the insecurity of national survival, and that states with genuine existential threats — he cites Israel and Pakistan — need them in a way India does not.
  • India faces the fear of a nuclear attack, not the fear of extinction, and a force built to answer that fear can be modest.
  • He treats the logic of NFU as counterintuitive but sound: the strongest state in its neighbourhood, facing no existential threat, has little use for going first.
  • He is also blunt that India does not have the nuclear superiority a counterforce doctrine would require, facing two nuclear-armed adversaries who maintain a strategic partnership directed at it.
• Ashley Tellis described India’s arsenal in its formative phase as a force-in-being — neither a recessed capability nor a ready arsenal, but a distributed set of components held under civilian custody and assembled only on political direction.
  • The force-in-being is deliberately unready, trading reaction time for political control and low accident risk.
  • Its coherence depends on the assumption that India would have time to assemble after absorbing a first strike, which is exactly what canisterisation and the sea leg have since made unnecessary — and therefore obsolete as a description.
• The pattern across all three is the same. The Indian minimalists do not argue that a bigger arsenal is immoral; they argue that it is strategically unnecessary and politically expensive, and that the burden of proof lies on those who want one.

Minimum deterrence survives as a doctrine because it was never defined tightly enough to be violated.

No first use

No first use is the load-bearing commitment of the Indian doctrine and the one most persistently attacked. The argument for it is a chain rather than a slogan: each link makes the next possible, and pulling any one out weakens the rest of the structure.

The case for no first use

• It preserves clarity about the threshold. NFU tells adversaries precisely which act triggers nuclear retaliation, and everything short of that act stays in the conventional world.
  • Removing NFU would remove that clarity and thereby weaken deterrence rather than strengthen it, because an adversary who cannot locate the line cannot be deterred from approaching it.
  • Clarity also protects India from the reverse error — being manoeuvred into a nuclear response to a provocation it had never intended to treat as nuclear.
• It makes an assured second strike, not a first strike, the organising requirement. This is the link that caps the arsenal.
  • A first-use posture requires the ability to destroy the adversary’s forces before they launch, which demands numbers, accuracy and intelligence on a scale India has never sought.
  • A second-strike posture requires only that enough survive to retaliate, which is achievable with a small force and slows the arms race instead of driving it.
• It permits de-mated, disassembled storage. With no requirement to launch first, warheads need not sit mated to delivery vehicles on alert.
  • The force is therefore less accident-prone, with fewer opportunities for unauthorised, mistaken or mechanical release.
  • It is also less delegation-prone: a force that need not react in minutes need not push release authority down the chain, which is what preserves civilian control in practice rather than only on paper.
  • Manpreet Sethi sets out the practical dividend precisely — no expensive high-alert infrastructure, weapons kept disassembled, no elaborate system against unauthorised use, no forces on a hair trigger, and the onus of escalation placed on the adversary.
• It removes the incentive to pre-empt on both sides. If neither side expects a bolt from the blue, neither has reason to use its weapons early for fear of losing them.
  • A declared first-use posture would generate mutual suspicion and use-them-or-lose-them pressure between two hostile neighbours with short flight times.
  • First-use capabilities also demand research, development and infrastructure investment on a scale India has not made and would have to fund at the expense of conventional forces it needs more.
• It fits India’s declared self-understanding and paid diplomatic dividends. Restraint was the currency in which India bought its way back into the international nuclear order.
  • The commitment to restraint built India’s standing as a responsible nuclear stakeholder and eased its accommodation after 1998 — the argument Harsh V. Pant makes when he notes that the relatively small size of the arsenal ruled out a first strike anyway, so the pledge cost little and bought much.
  • It was part of what made the 2008 accommodation politically possible; a first-use India would have been a far harder case to argue in Vienna and in Washington.

The case against

• Bharat Karnad attacks NFU on grounds of institutional competence rather than morality.
  • His position is that no first use is relevant only for a country with extreme confidence in the survivability of its nuclear forces, which India does not have.
  • NFU, he argues, requires efficient crisis management, and crisis management is not India’s forte — the Indian bureaucracy, in his judgment, being manifestly incapable of it.
  • The force of the objection is that NFU transfers the initiative to the adversary and then requires the Indian state to perform flawlessly under the worst conditions it will ever face.
• Lt Gen B.S. Nagal, a former Commander-in-Chief of the Strategic Forces Command, attacks it on moral grounds.
  • His objection is that no first use deliberately exposes the population to a first strike India has the means to prevent, and that there is no logic in accepting large-scale destruction.
  • The related argument is that the Indian public has never been prepared for what NFU commits it to absorb, and that a government has no mandate to accept those casualties on its behalf.
• The counterforce-temptation argument is the most serious modern critique. Vipin Narang and Christopher Clary argue that India is acquiring — and in official language quietly licensing — the capability to strike Pakistan’s strategic forces pre-emptively.
  • Their evidence is capability rather than declaration: canisterised missiles that keep warheads mated and raise baseline readiness; MIRVs, which are classically a counterforce technology; and a transformation in intelligence, surveillance and reconnaissance since 2003, including high-resolution satellites and persistent unmanned aircraft able to track dispersed Pakistani assets.
  • They read Shivshankar Menon’s published formulation — that there is a grey area in which India might strike first against an adversary about to launch, and that the doctrine has far greater flexibility than it gets credit for — as a potential first-strike loophole inside the pledge itself.
  • Their conclusion is not that India has abandoned NFU but that it is being creatively reinterpreted rather than revised, and that this is the more dangerous route, because it erodes the pledge without replacing the reassurance it provided.
  • The reply is that capability is not doctrine: canisterisation and MIRVs are equally explicable as survivability and penetration measures against Chinese defences.
• The signalling argument is the weakest but the most politically potent: given the conventional disparity with China, publicly adopting first use would communicate power and resolve at no material cost.
  • It is reinforced by a belief in parts of the Western non-proliferation community that India has already effectively relinquished the policy, and that the declaration is a public position covering private intentions.
  • The reply is that India’s actual arsenal cannot support a first-strike doctrine, so the signal would be read as bluff and would forfeit the diplomatic standing restraint has earned.
• P.K. Chari made two distinct arguments that are often run together.
  • First, that nuclear doctrines need not be static and should be treated as a work in progress, revised when circumstances demand, rather than defended as articles of faith.
  • Second, and more damaging, that there is a structural asymmetry in what the two arsenals do. Pakistan’s weapons deter India across the entire spectrum — from proxy war through conventional war to nuclear use — while India’s deter only the nuclear use at the top of it.
  • On his reading India’s weapons are primarily political and Pakistan’s explicitly military, which means that under NFU India buys deterrence against the least likely contingency and none against the most likely one.
  • He also questioned whether NFU was ever appropriate given that the original purpose of the arsenal was deterrence of China, against whom India’s conventional inferiority is real in a way it is not against Pakistan.
• Lt Gen Syed Ata Hasnain offers the middle position that NFU is not cast in stone, that India can retain the declaration while preparing for the possibility of first use, and that flexibility is a matter of judgment in a crisis rather than of published text.

The record of official ambiguity

The pledge has been surrounded by ministerial and official statements that qualify it in public perception without amending it in fact. Read in sequence they are less a drift than a pattern: senior figures periodically test the language, and the government then confirms that nothing has changed.

Statement	Who and when	What it said	What followed
The Menon formulation	Shivshankar Menon, then NSA, in a 2010 speech	India’s doctrine is no first use against non-nuclear weapon states — a phrasing that, read literally, leaves nuclear-armed states outside the pledge	Never repudiated; later read by analysts as the doctrine’s principal grey area
The alternative doctrine	A non-official task force of strategic analysts, 2012	Produced a complete alternative nuclear doctrine for India	No official uptake; evidence of a live expert debate
The Saran speech	Shyam Saran, then chairman of the NSAB, 2013	Restated NFU and massive retaliation, and added that the label on a nuclear weapon used against India, strategic or tactical, is irrelevant	The authoritative rebuttal of the tactical-weapon loophole
The manifesto commitment	The 2014 BJP election manifesto	Promised to “revise and update” the doctrine to make it relevant to the challenges of current times, while maintaining a credible minimum deterrent	No revision has been undertaken in the years since
The Parrikar remark	Defence Minister Manohar Parrikar, 2016	“Why should I bind myself? I should say I am a responsible nuclear power and I will not use it irresponsibly”	Immediately described as a personal opinion, not government policy
The Rajnath statement	Defence Minister Rajnath Singh, August 2019, at Pokhran	India has strictly adhered to no first use, but what happens in the future depends on the circumstances	Widely read as conditionality; no doctrinal change followed
Official clarification	The Ministry of External Affairs, in reply to Parliament	No change has been made to the no-first-use policy	The formal position, repeated whenever the question is put

“The label on a nuclear weapon used for attacking India, strategic or tactical, is irrelevant from the Indian perspective.”

— Shyam Saran

• Saran’s 2013 speech is the most important of these and cuts the other way. By declaring the label irrelevant, he closed the loophole that Pakistan’s tactical weapons were designed to open, at the cost of committing India to a response many find incredible.
• The Rajnath Singh statement is routinely over-read. It was made at Pokhran on the anniversary of Vajpayee’s death and asserted adherence in the same breath as conditionality.
  • A pledge that “depends on circumstances” is, on one reading, simply a statement about sovereignty — no government can bind its successors — and on another an invitation to the adversary to discount the pledge entirely.
• The current position, stated honestly. No first use remains declared policy and has never been formally revised, by any government, at any point since January 2003.
  • The qualifications it carries — the chemical and biological clause, and the confinement of the pledge to attacks by nuclear weapons — are part of the original text, not later accretions.
  • What has changed is not the text but the surrounding commentary and the underlying capability, and it is the combination of the two that has made the pledge look softer than it reads.

India has not revised its no-first-use pledge. It has allowed the pledge to be discussed in a way that lets adversaries discount it, which is the same outcome at none of the diplomatic cost.

Massive retaliation

• Massive retaliation means a response disproportionate to the attack that provoked it — the deliberate refusal of a graduated reply.
  • The term comes from the American doctrine of the 1950s, which threatened to retaliate instantly, by means and at places of its own choosing, rather than match an aggression at its own level.
  • India adopted the vocabulary but not the American purpose: for the United States it was a way to extend deterrence cheaply over allies; for India it is a way to make a small second-strike force sound sufficient.
• The 2003 formulation was chosen over flexible response for three reasons.
  • Flexible response requires a war-fighting arsenal — graduated options, tactical yields, battlefield delivery, and the command arrangements to use them — which is precisely the force structure minimalism rejects.
  • Graduated options invite the adversary to test each rung, converting a threshold into a staircase and making limited nuclear use thinkable.
  • A single, massive, undifferentiated threat is cheap to declare and easy to communicate, which suits a doctrine whose whole purpose is to be understood.
• The standing objection is that it is not credible against limited use. If Pakistan detonates a low-yield weapon on its own territory against an advancing Indian formation, is India really going to destroy Pakistani cities and invite the destruction of its own?
  • No adversary, the objection runs, believes India would trade Delhi and Mumbai for a battlefield strike in the Thar desert, and a threat nobody believes deters nobody.
  • The problem is structural: the doctrine offers one response to every level of provocation, so any provocation below the level that justifies it is effectively undeterred.

The answers to the credibility objection

• Manpreet Sethi’s answer is that the objection misunderstands what deterrence requires. A threat need not be certain to be effective; it needs only to be possible enough that the adversary cannot afford the risk.
  • She argues that deterrence rests on capability and the survivability of command, not on the proportionality of the threatened response — India’s task is to make retaliation physically certain, not morally proportionate.
  • Her prescription is therefore at the level of strategy rather than doctrine: harden storage and command, deploy longer-range sea-based missiles, raise the visibility of the Strategic Forces Command, and keep responses to terrorism far below the nuclear threshold.
• The “assured retaliation” reformulation is the widely-canvassed compromise: replace “massive” with “punitive” or “assured”, keeping the certainty of retaliation while dropping the commitment to a fixed scale.
  • The gain is credibility across the whole range of provocations, since India would no longer be committed in advance to a response it might not choose.
  • The loss is the clarity that Saran’s formulation bought — a graduated threat re-opens the question of whether a small first use might go unanswered, which is exactly the door Pakistan’s tactical weapons are meant to walk through.
  • Vipin Narang has classified India’s posture as one of assured retaliation on the evidence of the force rather than the text, which is a description of where practice already sits.
• The deeper point is that the objection proves less than it claims. Deterrence failures are not decided by the elegance of a doctrine but by whether an adversary is willing to be the first state since 1945 to use a nuclear weapon.

Command, control and the force

The Nuclear Command Authority

• The Nuclear Command Authority was created by the same statement of January 2003 that announced the doctrine, and is the institutional expression of civilian control.
  • The Political Council is chaired by the Prime Minister and is the sole body that can authorise the use of nuclear weapons. No other body, civil or military, has that power.
  • The Executive Council is chaired by the National Security Adviser. It provides the inputs for decision-making and executes the directives given to it by the Political Council; it does not itself decide.
  • An alternate chain of command and succession exists so that the authority to retaliate survives the decapitation of the political leadership — the technical requirement without which no-first-use is a promise India could not keep.
• The Strategic Forces Command, established in 2003 under a Commander-in-Chief drawn in rotation from the three services, is the tri-service body that holds and would deliver the weapons on political direction.
  • Custody of the warheads has historically rested with civilian scientific agencies rather than with the armed forces, with the military holding the delivery systems.
  • This separation is the physical guarantee of the constitutional principle: the Indian military does not hold assembled nuclear weapons.
• Exclusive civilian control is the most distinctive feature of the Indian nuclear state and the sharpest single contrast with Pakistan.
  • It is also operationally costly: a force that cannot be released without political decision cannot respond quickly, which India accepts because no first use means it never needs to.

De-mated storage and what canisterisation did to it

• India’s historic practice was to keep the force de-mated and partly disassembled — fissile cores separate from weapon assemblies, weapon assemblies separate from delivery vehicles, each element under a different custodian.
  • This was the physical guarantee of no first use: a force in that condition cannot strike first and everyone knows it.
• Canisterisation has quietly dissolved that guarantee. A canisterised missile is sealed in a climate-controlled tube with the warhead already fitted, ready to launch on receipt of an order.
  • The Agni-5 is canisterised, and the sea leg is canisterised by necessity — a submarine cannot surface mid-crisis to assemble a weapon.
  • SIPRI’s own assessment is that the practice has blurred the line between stored and deployed warheads, which is why its 2026 estimate counts a small number of Indian warheads as deployed for the first time.
• The doctrinal significance is that readiness has risen without any change in declared policy. Canisterisation shortens the time from decision to launch, which is valuable for assured retaliation and equally valuable for pre-emption.
  • What is not in dispute is that the old physical proof of restraint is gone, and India’s assurances now rest on declaration rather than on demonstrable incapacity.

The triad in 2026

Leg	Systems	Status	What it contributes
Land	Prithvi short-range; the Agni series to Agni-5 and Agni-Prime	Mature; Agni-5 exceeds 5,000 km, is canisterised, and was tested with MIRVs in March 2024	Range coverage of all of Pakistan and all of China; the backbone of the force
Air	Mirage-2000, Su-30MKI, Rafale with gravity weapons	Operational, the oldest leg	Flexibility and visible signalling; vulnerable to modern air defences
Sea	INS Arihant (2016), INS Arighaat (2024), INS Aridhaman (3 April 2026); K-15 at 750 km deployed, K-4 at 3,500 km tested	Three boats commissioned; the missiles lag the platforms	The only survivable second-strike leg, and therefore the doctrinal keystone

• The Agni-5 MIRV test of March 2024 was the most significant technical development since the triad itself. A single missile carrying multiple independently targetable warheads multiplies the targets a small force can hold at risk.
  • Its defensive rationale is penetration of ballistic missile defences, particularly China’s, which is the justification India offers.
  • MIRVing also cuts against minimalism arithmetically: it raises warhead requirements per launcher at the moment the doctrine claims to be capping the arsenal.
• The commissioning of INS Aridhaman on 3 April 2026 is the arithmetical turning point for the sea leg. Continuous at-sea deterrence requires three boats — one on patrol, one in transit or workup, one in refit — and India now has three.
  • INS Arihant was commissioned in August 2016 and conducted the first deterrence patrol in November 2018; INS Arighaat followed in August 2024.

The candid assessment: the boats have outrun the missiles

• India’s submarines can go to sea, but they cannot yet threaten what they need to threaten. The operational missile is the K-15 at 750 km, which forces a boat to approach an adversary’s coast to be useful — the opposite of survivability.
  • The K-4, at 3,500 km, has been tested repeatedly but is not operationally deployed, and getting it combat-ready is the acknowledged priority.
  • The K-5, at around 5,000 km, remains under development. Until one of these enters service, the sea leg holds Pakistan at risk comfortably and China only marginally.
• India also has no nuclear attack submarines. Two SSNs were sanctioned in 2024 but are not expected before the late 2030s.
  • Without SSNs, India cannot reliably escort its own SSBNs or trail an adversary’s, which is the standard method of protecting a sea-based deterrent.
  • The comparison is stark: China fields over sixty submarines, around a dozen nuclear-powered, with projections of a far larger attack fleet by the mid-2030s.
• The honest summary is that India has assembled the platforms of a continuous deterrent ahead of the weapons that would make it one. The triad is complete on paper and incomplete in capability, and closing that gap is the single largest item of unfinished business in the force.

Ballistic missile defence

• India’s BMD programme began in 1999 and has proceeded in phases: Phase 1, using the Prithvi Air Defence and Advanced Air Defence interceptors, was completed around 2019 after a first successful endo-atmospheric intercept in November 2006.
  • Phase 2, using the AD-1 and AD-2 interceptors — AD-1 endo- and low exo-atmospheric, AD-2 high exo-atmospheric, was validated in tests in June 2026, giving India a claimed capability against missiles of intercontinental class.
• Missile defence is destabilising in a way its proponents rarely concede. A shield that might blunt a retaliatory strike devalues the adversary’s deterrent and pushes him to build more, MIRV more, and shoot earlier.
  • Pakistan has responded by developing the Ababeel, a medium-range missile it claims carries multiple independently targetable warheads, explicitly to saturate Indian defences — tested but not in service, and its MIRV capability is doubted by outside analysts.
  • The paradox is that a state committed to no first use has the strongest reason to want defences — and that acquiring them is what most persuades adversaries the commitment is insincere, because first strike plus missile defence is the classic damage-limitation package.
• India’s transparency practices cut the other way and deserve equal weight. India issues NOTAMs and maritime warnings before missile tests, and is bound by the 2005 agreement to pre-notify Pakistan of ballistic missile flight tests.
  • Their gap is cruise missiles, which the 2005 agreement does not cover — the omission exposed by the accidental BrahMos launch into Pakistan in March 2022, which Pakistan chose not to treat as an attack.

Deterrence in a three-cornered relationship

The first nuclear age was a Cold War dyad, and its arms control, its vocabulary and its stability theory were all built for two players. The second nuclear age has moved the centre of gravity from the Euro-Atlantic to the Indo-Pacific and put multiple players into one equation, where a war between two of them can draw in a third.

• South Asian deterrence is not a dyad but a chain: India sizes its arsenal against Pakistan and postures it against China; Pakistan builds against India alone; China builds against the United States and treats India as a secondary problem.
• Each pair is therefore playing a different game, and moves that are defensive in one dyad are offensive in another.

Pakistan: full-spectrum deterrence

• Pakistan’s declared posture is full-spectrum deterrence, an extension of its earlier “credible minimum deterrence” designed to cover the whole range of conflict from a limited conventional incursion upward.
  • The purpose is to deny India any level of conventional operation that Pakistan cannot threaten to answer with nuclear weapons — that is, to close the space beneath the nuclear threshold.
  • Pakistan has never accepted no first use and has repeatedly refused Indian proposals for a bilateral no-first-use agreement, because first use is the whole point of the posture.
• The Nasr, or Hatf-IX, is the instrument. It is a short-range battlefield ballistic missile of roughly 70 km with a low-yield warhead, tested from 2011 and presented as a counter to Indian armoured thrusts.
  • Its real function is arguably signalling rather than war-fighting: analysts treat the Nasr as an instrument for advertising a low threshold rather than a usable battlefield weapon.
• India’s answer is the Saran formulation: the label on a weapon used against India is irrelevant, and any nuclear use invites massive retaliation. Whether that answer is believed is the entire tactical-weapons debate.

Pakistan’s red lines and command structure

• The four thresholds attributed to Lt Gen Khalid Kidwai, long-time head of the Strategic Plans Division, remain the standard statement of Pakistan’s red lines, though they have never been formal doctrine.
  • Spatial — Indian conquest of a large part of Pakistani territory, conventionally glossed as a crossing of the Indus line.
  • Military — destruction of a large part of Pakistan’s armed forces, particularly its air force, or attacks on its nuclear installations.
  • Economic — strangulation, such as a naval blockade of Karachi or the stopping of the rivers.
  • Political — destabilisation or the fomenting of internal disintegration.
• The thresholds are deliberately vague, and that vagueness is the strategy: they give no “when, where and how”, which forces India to treat almost any large operation as potentially threshold-crossing.
  • The vagueness is also the weakness — red lines that cover everything discipline nothing, and each crossing without a nuclear response erodes them further.
• Pakistan’s National Command Authority was created in 2000 and formalised by statute in 2010, with an Employment Control Committee and a Development Control Committee, serviced by the Strategic Plans Division.
  • In form it is civilian-led, with the Prime Minister chairing; in practice the army has always been dominant, and the SPD has been an army body throughout.
  • Pakistan’s 27th Constitutional Amendment, passed in November 2025, restructured this further. The post of Chairman of the Joint Chiefs of Staff Committee was abolished and a Chief of Defence Forces created, held by the army chief, Field Marshal Asim Munir.
  • A new National Strategic Command was established under an officer nominated on the army chief’s recommendation, with the Strategic Plans Division still army-run.
  • The effect is to concentrate nuclear authority in one service and remove the inter-service checks the previous structure supplied — read by analysts as reducing, not raising, the reliability of Pakistani nuclear control.
• The A.Q. Khan network remains the reference case for command security. The clandestine transfer of enrichment technology to Libya, Iran and North Korea from within a state nuclear establishment demonstrated that the risk in Pakistan is not only unauthorised use but unauthorised proliferation.

	India	Pakistan
Declared posture	Credible minimum deterrence	Full-spectrum deterrence
First use	No first use, qualified for CBW attack	Reserved; NFU explicitly refused
Response doctrine	Massive retaliation to any nuclear use	Graduated, including battlefield use
Tactical weapons	None; label declared irrelevant	Nasr / Hatf-IX and short-range systems
Control	Civilian — Political Council chaired by the PM	Military-dominated; further concentrated in 2025
Stockpile, January 2026	190 warheads, 12 assessed as deployed	170 warheads, none assessed as deployed
Weapons deter	Only nuclear use	The whole spectrum, including India’s conventional response

• The stockpile figures matter because the old story has reversed. India’s arsenal now exceeds Pakistan’s, and the claim that the two are at parity no longer holds, though Pakistan continues to expand delivery systems — Shaheen-III, the MIRV-claimed Ababeel, the Babur-3 cruise missile.

The stability–instability paradox

• The paradox is the analytical key to South Asia. Where nuclear weapons make war at the top of the ladder unthinkable, they make conflict at the bottom of it safer, and therefore more likely.
  • Stability at the nuclear level purchases instability at the sub-conventional level, because each side knows the other will not escalate to the point of mutual destruction over a limited provocation.
• Pakistan’s use of the paradox is deliberate, not incidental. Its proxy war is conducted under a nuclear umbrella that is designed to deter India’s conventional retaliation.
  • The weapons function as a catalytic device, drawing international pressure onto India to de-escalate whenever a crisis begins, which converts every Indian response into a diplomatic cost for India.
  • This is the substance of the “thousand cuts” argument — Pakistan can inflict damage repeatedly without bearing the cost of a conventional war, because its nuclear weapons deter the war and its deniability blunts the attribution.
  • It is the clearest illustration of Chari’s asymmetry: Pakistan’s arsenal deters India’s conventional response, while India’s arsenal deters only a Pakistani nuclear strike that was never likely in the first place.

India’s weapons deter the least probable attack. Pakistan’s deter the most probable response. That asymmetry, not the warhead count, is the real imbalance in South Asia.

China

• China’s arsenal is expanding faster than any other state’s, from roughly 300 warheads in the late 2010s to about 620 in early 2026, with American official projections of over 1,000 by 2030.
  • The expansion includes roughly 320 new missile silos across three fields in Gansu, Xinjiang and Inner Mongolia — the largest single addition to the Chinese force ever undertaken.
  • China is also developing early-warning satellites and exercising a launch-on-warning or early-warning counterstrike posture, which shortens decision time and sits uneasily with its declared restraint.
• China maintains an unconditional no-first-use pledge — no first use of nuclear weapons at any time and under any circumstances — and is the only other nuclear-armed state with one.
  • Whether it is being hollowed out is genuinely contested. Expansion on this scale is compatible with a survivability rationale, since a larger and better-protected force is precisely what a no-first-use state needs against American counterforce capability and missile defences.
  • The counter-argument is that launch-on-warning is in tension with no first use in practice, because a state that launches on warning may launch on a false warning.
  • For India the pledge is doubly awkward: it is the model India cites for its own restraint, and its erosion would remove India’s best precedent.
• The asymmetry of direction is the central structural fact. China’s deterrent is designed against the United States; India’s is designed against China; Pakistan’s is designed against India.
  • Chinese planners do not size their force against India, so Indian increases do not reliably register in Beijing while they register immediately in Rawalpindi.
• Chinese assistance to Pakistan’s programme is part of the deterrence picture, not merely a bilateral irritant. China supported the development of Pakistan’s nuclear and missile capabilities from the 1970s onward, and the two states maintain a close strategic and military partnership.
  • The effect is to give China a low-cost method of tying down Indian forces and to make Pakistan’s arsenal partly a function of Chinese policy.
  • The China–Pakistan–India triangle dates in its military form to the 1963 boundary agreement over the Trans-Karakoram Tract, which made Kashmir a trilateral rather than bilateral question.

The two-front problem and what it does to “minimum”

• India is the only nuclear-armed state facing two nuclear-armed adversaries who are strategic partners and both of whom hold territorial claims against it. That is the fact that makes “minimum” hardest to define.
  • A force sized for assured retaliation against Pakistan is inadequate against China; a force sized against China is destabilisingly large from Pakistan’s perspective, driving Pakistani expansion that India must then answer.
• The working Indian solution is that the arsenal is sized against Pakistan and postured against China. Numbers are set by the smaller problem; range, survivability and penetration are set by the larger one.
  • Hence also the awkwardness of Indian denials: every China-facing improvement is a Pakistan-facing improvement, whatever India intends.
• India’s rebalancing is not only nuclear. Force deployment has shifted from a purely western orientation toward the north, border infrastructure has been prioritised, and external balancing through the Quad and partnerships has expanded — all of which are conventional answers to a problem the nuclear doctrine cannot solve.

The crisis record as evidence

Doctrines are tested by crises, and South Asia has supplied a steady series of them. The record is the only empirical evidence available about where the thresholds actually lie, and it has been read in opposite directions by people looking at the same events.

Before the tests

• Brasstacks, 1986–87, was a very large Indian exercise on the Pakistani border that Pakistan read as preparation for war, in an environment where both sides were opaque nuclear states.
• The compound crisis of 1990 combined the outbreak of insurgency in Kashmir with Indian and Pakistani mobilisation and heavy American diplomatic intervention.
  • Its significance is that undeclared arsenals were already shaping behaviour eight years before either state tested, which is the strongest evidence for existential deterrence in the region.

Kargil, 1999

• Kargil was the first armed conflict between two overtly nuclear-armed states, coming barely a year after both had tested.
  • Pakistan’s infiltration across the Line of Control was premised on the belief that nuclear weapons would prevent India from widening the war, which is the stability–instability paradox operating as a plan rather than as an observation.
  • India fought a limited war, refused to cross the Line of Control or the international border, and won a costly local victory — restraint that was itself a form of escalation control.
• Kargil is also the case for the doctrine. Both states had nuclear weapons, both knew it, and neither used them or seriously threatened to; the war stayed conventional and geographically confined.

Operation Parakram and the limited-war debate

• Operation Parakram, December 2001 to October 2002, followed the attack on Parliament and was India’s largest mobilisation since 1971 — roughly ten months of deployment ending without an operation.
  • It was an exercise in coercive diplomacy that failed on its own terms: Pakistan did not dismantle the infrastructure India demanded, and India absorbed the casualties, cost and political embarrassment of a mobilisation it could not use.
  • The proximate cause of the failure was time: Indian strike formations took weeks to reach their launch positions, by which point international pressure and Pakistani preparation had removed the option.
• Cold Start emerged directly from that failure. The concept — shallow, rapid, multiple thrusts by integrated battle groups held close to the border — was designed to give India a usable conventional option below Pakistan’s nuclear threshold and fast enough to pre-empt diplomatic intervention.
  • Pakistan’s answer was the Nasr, which is best understood as a direct doctrinal reply to Cold Start rather than as an independent development.

Surgical strikes, Balakot and Operation Sindoor

• The 2016 surgical strikes after the Uri attack were shallow ground raids across the Line of Control, publicly acknowledged, and were the first deliberate breach of India’s long policy of strategic restraint.
• Balakot, February 2019, went further: air strikes on a target in Pakistan proper, officially described as a non-military pre-emptive action, followed by an aerial engagement and the return of a captured Indian pilot.
  • Balakot established that airspace over Pakistan was no longer sanctuary, and that India would accept the risk of a two-day crisis to make the point.
• Operation Sindoor, May 2025, is the most consequential episode since the doctrine itself. The Pahalgam attack of 22 April 2025 killed twenty-six civilians; India struck on 7 May; hostilities ceased on 10 May; the Prime Minister addressed the nation on 12 May.
  • The operative formulation of that address was that India will not tolerate nuclear blackmail and will strike at terrorist infrastructure sheltering under it — set out as one of three principles, alongside responding on India’s own terms and refusing to distinguish between terrorist groups and the state that sponsors them.
  • India described the outcome as a “new normal”, meaning that a mass-casualty terrorist attack will now be answered by direct military action as a matter of course rather than as an exception.
• What Sindoor claims analytically. It asserts that India believes Pakistan’s nuclear threshold is far higher than Pakistan advertises, and that a substantial space exists for conventional operations beneath it.
  • If that reading is right, the stability–instability paradox has been turned around: India, not Pakistan, is now the state exploiting the space beneath the threshold, and Pakistan’s nuclear cover for sub-conventional aggression has been devalued.
  • Post-Sindoor assessments in Pakistan point the same way from the other side — Pakistani first use looked less credible after the crisis than before it, which is why the response has been accelerated modernisation, a push for a sea-based second strike, and appeals for external intervention.
• The risks of that reading are serious and should not be minimised. The claim that a threshold is higher than advertised can only be tested by approaching it, and the test is not repeatable if it fails.
  • Each successful crossing raises the level at which the next crisis begins, because deterrence is competitive and the losing side must restore its position.
  • Pakistan’s concentration of nuclear authority under a single service since November 2025, and its dispersal of short-range systems, both shorten the fuse on the side India is now probing.
  • And a doctrine of assured retaliation gives India nothing to do at the sub-conventional level, so the “new normal” is a conventional answer operating in a nuclear environment the doctrine does not describe.

Confidence-building measures and their decay

• The 1988 Agreement on the Prohibition of Attack against Nuclear Installations commits India and Pakistan not to attack each other’s nuclear installations and to exchange lists of them.
  • The lists have been exchanged every 1 January without interruption, through wars, crises and the collapse of every other channel — the most durable artefact of the relationship.
• The Lahore Memorandum of Understanding, February 1999, was the most ambitious attempt at a nuclear risk-reduction framework, committing both states to notify ballistic missile tests, to engage on nuclear doctrines and to reduce the risk of accidental use.
  • It was signed three months before Kargil and has never recovered from the demonstration that a summit could be followed immediately by an infiltration.
• Later measures narrowed the ambition to the technical. A hotline between the foreign secretaries was established in 2004; the 2005 agreement on pre-notification of ballistic missile flight tests requires at least three days’ notice; a 2007 agreement on accidents relating to nuclear weapons requires immediate notification of any accident.
  • These are implemented, but they are thin, unmonitored and narrow: the missile agreement excludes cruise missiles, and the hotline has not been used in the crises it exists for.
  • The BrahMos misfire of March 2022 exposed the cruise-missile gap exactly, and was defused by Pakistani forbearance rather than by any agreed mechanism.
• The regime has decayed because the political relationship carrying it has decayed. Composite dialogue has been suspended for over a decade, and CBMs are complements to a relationship rather than substitutes for one.
  • What is missing is exactly what a crisis needs: a working channel used under pressure, and any agreed understanding about what each side considers escalatory.

Is the doctrine viable?

The doctrine has now survived twenty-three years, four crises and a technological transformation of the force it governs. That is a considerable record, and it should be weighed before the criticisms are.

The case that it works

• It has held through every crisis since it was adopted. Balakot and Sindoor stayed conventional, as Kargil and Parakram had before it; no nuclear weapon has been used, brandished operationally or moved in a way that produced a nuclear crisis.
• It capped the arsenal. At 190 warheads India has, over five decades, built a force smaller than France’s and a fraction of China’s, and the second-strike organising principle is why.
• It produced international accommodation. The restraint the doctrine institutionalised is what made the 2008 waiver and India’s admission to the export-control regimes politically arguable.
• It keeps the weapons out of war-fighting. India has no tactical weapons, no battlefield doctrine, no pre-delegation and no service custody of assembled warheads — an unusual record for a state in a live territorial dispute with two nuclear neighbours.
• It is cheap. A second-strike force requires no alert infrastructure, no continuous readiness and no arms race, which frees resources for the conventional forces India actually uses.

The case that it is eroding

• Capability is drifting toward counterforce even if intent is not. MIRVs, canisterisation, improved intelligence and precise long-range systems together describe a force that could do more than retaliate, and adversaries plan against capability.
• Canisterisation and MIRVing sit awkwardly with de-mating and minimalism. The physical guarantees that once made India’s restraint verifiable have been engineered away in the name of survivability.
• Ministerial statements have made the pledge conditional in public perception. The text has not changed, but a pledge widely believed to be conditional does not deliver the reassurance an unconditional one does.
• Sindoor’s “new normal” asserts space that the other side denies exists. India claims room for conventional operations under a nuclear overhang; Pakistan’s declared posture is built precisely to deny that room, and neither claim can be settled except by testing it.
• “Minimum” no longer disciplines anything. Defined dynamically against two adversaries, one of whom is expanding faster than any state in the world, it imposes no ceiling that could ever be shown to have been breached.

What a review would actually have to decide

A revision is often demanded and rarely specified. The questions a genuine review would face are narrower and harder than the slogans suggest, and most of them are about capability and command rather than about the pledge.

• Whether to replace “massive” with “assured” or “punitive” — buying credibility across the spectrum at the cost of the clarity that makes the label of a weapon irrelevant.
• Whether the chemical and biological clause still earns its place, given that it is the qualification most often used to argue that India’s NFU was never really NFU.
• Whether to say anything at all about pre-emption, or to leave the grey area where it is and accept that adversaries will assume the worst about it.
• Whether “minimum” should be given content — a stated ceiling, a stated force structure, or an explicit statement that it is sized against assured retaliation rather than against arsenals.
• Whether to publish more rather than less. There is near consensus in the Indian strategic community that the doctrine should be periodically re-examined and that more should be disclosed about it, both to deter adversaries and to inform domestic debate.

The strongest argument against revision is not that the doctrine is perfect but that revision has costs and no obvious benefit: the diplomatic dividend of restraint is real and would be forfeited, while the military dividend of first use is speculative and unfunded. India’s realistic choice is not between doctrines but between working on strategy — survivability, command endurance, a deployed sea-based missile, conventional options far below the threshold — and working on words. The doctrine’s weaknesses are, almost without exception, weaknesses of capability and crisis management wearing doctrinal clothes.