What's new



Mar 21, 2007
Reaction score
United States


The nuclearisation of South Asia can be analysed from different angles.

Ejaz Haider
May 28, 2023

Exactly 25 years ago, on May 28, Pakistan conducted five nuclear tests (a sixth was done on May 30). The tests were a response to India’s five nuclear tests, conducted on May 11 and 13. Both governments, after the tests, declared a moratorium on further testing. A quarter century since May 1998, they have stuck to the moratorium, though neither has signed the Comprehensive Test Ban Treaty (CTBT).

The nuclearisation of South Asia can be analysed from different angles. For our present purpose, however, the main question is simple: while there’s general consensus among security experts and nuclear establishments within India and Pakistan that nuclear weapons are a deterrent, should we consider that proposition a truism for now and the time to come?

Having analysed these issues for a long time and advocated the deterrence framework, I have come to believe that such unerring faith in deterrence might increasingly be misplaced for a host of reasons. This is also true for other nuclear weapons states (NWSs). We should be less sanguine about deterrence holding in all situations. Indeed, there’s now growing literature about the world being “on the cusp of a Third Nuclear Age” where a number of factors are likely to make the old belief in deterrence increasingly problematic if not entirely untenable.

This article primarily deals with India and Pakistan, a nuclear dyad in conflict. But some of the observations here are also applicable to other NWSs, including those whose possession of nuclear weapons have been “legitimised” by the Nuclear Non-Proliferation Treaty (NPT).

Today marks 25 years since Pakistan became a declared nuclear weapons state. Although conventional wisdom holds that Pakistan’s nuclear parity with India has served to deter a serious conflict between the two countries, with our peculiar history and geography and the emergence of new technologies, is this faith misplaced?

What follows below is not an exhaustive discussion or analysis of the new factors but seeks to draw the attention of the informed generalist to them. After listing those factors, I intend to focus the discussion on India and Pakistan.


In a 2021 article for European Journal of International Security, titled Strategic non-nuclear weapons and the onset of a Third Nuclear Age, Andrew Future and Benjamin Zala describe the Third Nuclear Age as the combination of Second Nuclear Age thinking — i.e., deployment of Strategic Non-Nuclear Weapons (SNNWs) — “with the return of the kind of major power competition associated with the First Nuclear Age.” This combo, as experts have begun to note, is highly dangerous.

So, what are SNNWs? Broadly, they are weapons with the “ability to engage targets at the strategic level of warfare, where the adversary’s sources of national power are located.” They can be kinetic and non-kinetic.

The kinetic category includes the increasingly more sophisticated conventional precision-strike capabilities: cruise and ballistic missiles, hypersonic glide vehicles and unmanned aerial vehicles (UAVs). This category also includes anti-satellite weapons and missile defences. Missile defences, by denying the adversary strategic strikes, can offset the deterrent balance and force adversaries into using overwhelming preemptive force. The ongoing Russo-Ukraine War has seen the employment of these SNNWs.

The second, non-kinetic category, includes cyber attacks (which can be mounted with strategic effect) and electronic-warfare capabilities (dominating the electromagnetic spectrum can not only degrade or deny the environment for the adversary but achieve strategic results for the side dominating the spectrum). Another aspect is information warfare and influence operations.

Again, we are witnessing the use of info war by both Ukraine/NATO and Russia. Misinformation and disinformation campaigns, as is now widely known, serve to undermine trust in public institutions and governments. They are also essential tools in influencing perceptions.


While the SNNW technologies are disrupting the battlefield by getting increasingly more accurate and sophisticated, there’s the additional problem of cross-tech/cross-platform integration that is concentrating and increasing firepower devastatingly.

Developments in artificial intelligence (AI) are also likely, by most accounts, to take the human out of the decision-making loop. This is not hypothetical, since multiple such systems are being tested and prototypes developed. Another problem is the presence of SNNWs in a nuclear environment, such as between a nuclear dyad.

Since they “constitute an employable and credible weapon system that can engage the sources of enemy power directly, skipping the tactical and operational levels of warfare”, nuclear adversaries can use them thinking that they could keep the conflict below the nuclear threshold. Also, in the case of NWSs there’s no way of knowing whether these weapons carry conventional warheads.

Cross-platform integration and the introduction of AI — what we can call the Internet of Military Things — is changing the concept of the decision-making speed and what the militaries call the kill chain (it also throws up many ethical issues). Up until now, the kill chain was a series of processes, executed sequentially. With AI, we are now looking at overlapping some of those phases and completing them in parallel, to reduce the execution time of the chain.

In other words, the weapon is fired first, the find and fix processes overlap the flight time, and the final target designation is sent to the weapon in flight, through a SATCOM channel. Now bring AI into this and combine it with hypersonic missiles and you get an idea of how much time will be reduced on the kill chain.
In August 2020, the Defense Advanced Research Projects Agency of the United States conducted a simulated experiment. It pitted a Heron-developed AI pilot against a top F-16 human pilot. The AI pilot beat the human pilot in a 5-0 sweep in dogfight and manoeuvres.

According to DARPA, the trials were designed as a risk-reduction effort for its Air Combat Evolution (ACE) programme to flesh out how human and machine pilots can share operational control of a fighter jet to maximise its chances of mission success. The overarching ACE concept is aimed at allowing the pilot to shift “from single platform operator to mission commander”, in charge not just of flying their own aircraft but managing teams of drones slaved to their fighter jet.


 As nuclear weapons states outside the framework of the NPT, Pakistan and India have certain legal and normative responsibilities | AFP

As nuclear weapons states outside the framework of the NPT, Pakistan and India have certain legal and normative responsibilities | AFP

While some of these technologies are expensive, others are not. The cost of many will steadily come down further as they proliferate (drones are a case in point, as is cyber and digital expertise). At this point, there are almost no regulatory frameworks for a number of emerging technologies.

Integration and the digital environment has also brought other risks. In a January 2018 report, the Royal Institute of International Affairs warned that US, British and other nuclear weapons systems are increasingly vulnerable to cyber-attacks. This is far from an abstract threat.

In 2010, the US Air Force lost contact with a field of 50 Minuteman III Inter-Continental Ballistic Missiles (ICBMs) at Warren Air Force Base in Wyoming for an hour, raising the terrifying prospect that an enemy actor might have taken control of the missiles and was feeding incorrect information into the nuclear command-and-control networks.

Earlier, in 2016, Andrew Futter wrote a paper for the Royal United Services Institute, titled ‘Cyber Threats and Nuclear Weapons: New Questions for Command and Control, Security and Strategy.’

Futter says: “There are two significant implications of this for nuclear weapons management and nuclear strategy: first, increasing complexity, particularly through computerisation and digitisation, raises the risk of normal accidents within the nuclear enterprise; and second, complex systems used to manage nuclear forces contain inherent vulnerabilities, weaknesses and bugs that might be exploited or manipulated in a variety of different ways by hackers.”

This is not all. In a January 2023 edited volume, The Fragile Balance of Terror: Deterrence in the New Nuclear Age, professors Vipin Narang and Scott D Sagan point to what they call the “declining confidence in deterrence.”

Gathered in the book are some of the top world experts trying to analyse the presence of nuclear weapons in a new age: growing prevalence of personalist dictatorships; incomplete or incorrect information; states facing multiple nuclear adversaries (regional subsystems) and operating in a novel information environment where misinformation can be maliciously planted or otherwise spread rampantly; states possessing small nuclear arsenals which they fear may not be reliable or survivable or over which they may not retain firm command and control.

As they put it: “Each of these factors alone, and especially in combination, generate risks that our standard strategies of nuclear deterrence are simply unequipped to manage or address.”

This is just a bird’s-eye view of a much larger and complex body of literature, a corpus that is increasing in volume, not as an exercise in alarmism but in analysing the emerging, uncertain trends.


Some might say that it will be a while before some of these advanced technologies will get to these shores. They are wrong.

UAVs are already here, as are missile defences and precision, long-range artillery. India is already seeking to bolster its anti-ballistic and cruise missile defences, developing MIRVs (multiple independently-targetable re-entry vehicles), hypersonic glide vehicles (reducing the already very short missile flight times) and fielding SSBNs (ship submersible ballistic nuclear), essentially, nuclear-powered submarines which can carry ballistic and cruise missiles with nuclear warheads.

These systems can both be non-nuclear and nuclear — i.e., they can carry tactical and strategic warheads. Coupled with non-kinetic strategic options (cyber, electronic and information warfare), they extend the same set of problems to India and Pakistan. In fact, given the contiguity of this nuclear-armed conflict dyad, the chances of accidents, miscalculations and bluster are even higher.

There are a number of additional risk factors as far as India and Pakistan are concerned: they are locked in a conflictual model; they have had multiple armed confrontations; both are nuclear-armed with growing arsenals and capabilities; since 2014, but more so since 2016, there is no dialogue framework between the two (covert channels notwithstanding); multiple military and non-military confidence-building measures (hotlines etc) are hardly used to offset a crisis, though they are used after a crisis begins to unfold; and the two sides’ respective positions on outstanding disputes are almost mutually exclusive.

This environment is known. But it also feeds into another set of problems that increases the likelihood of a conflict:
a: India’s stated doctrine of limited conflict;

b: India’s movement away from the declaratory no-first use policy, as contained in its 2003 doctrine, to what can now only be described as a non-stated first-use policy; and

c: India’s poor record of handling weapon systems and platforms, which increases the risk of accidental conflict.


 The remains of a missile fired from India which landed near Mian Channu on March 9, 2022 | Reuters

The remains of a missile fired from India which landed near Mian Channu on March 9, 2022 | Reuters

India’s doctrine of limited war has a high probability of a spiral. That story goes back to India’s full-scale mobilisation in 2001-02. The Indian political and military leadership drew two lessons from that mobilisation: the requirement for a forward-leaning posture, and developing strategies to punish Pakistan in limited engagements that do not escalate — i.e., assuming there is a band in which India can act militarily and punitively.

If India were to play within that band, goes the theory, it would be very difficult for Pakistan to use nuclear-level deterrence, because such escalation would be considered highly disproportionate and would draw international opprobrium.

This was the start of India’s ‘Cold Start’ doctrine. Cold Start is a limited-war strategy designed to seize Pakistani territory swiftly without, in theory, risking a nuclear conflict.

As an explainer in The Economist put it, the strategy envisaged “having nimbler, integrated units stationed closer to the border [that] would allow India to inflict significant harm before international powers demanded a ceasefire. By pursuing narrow aims, it would also deny Pakistan a justification for triggering a nuclear strike.”

This thinking is flawed and rests on the assumption that, in any first or possibly second round, India will gain an upper hand, forcing Pakistan to either escalate or blink. If Pakistan does escalate, India believes the international opinion will be on India’s side and Pakistan will have to climb down.

But events may not follow India’s script, as was evident in February 2019. Pakistan’s aerial retaliation and complete domination of that point of conflict forced India to threaten missile strikes, a big no between a nuclear dyad. International pressure, Pakistan’s promised response and Islamabad’s decision to return the downed Indian pilot served to de-escalate the situation.

But things could have got worse: what if India hadn’t missed the target in Balakot and killed seminary children; what if, under pressure, Pakistan had actually engaged and destroyed the Indian targets that were selected for engagement? What if India had followed through on its missile strikes threat, forcing Pakistan to retaliate with greater numbers?

At every point along that trajectory, many things could have gone wrong. That they didn’t could possibly be traced back to the chance factor that the Indian strike package missed its target.

At the politico-strategic levels, relations remain tense, with little to no dialogue between the two sides since August 5, 2019, when India unilaterally and illegally revoked the autonomous status of Jammu and Kashmir, in violation of UN Resolutions. Past experience suggests that deteriorating relations result in a higher probability of conflict between the two sides.


India’s 2003 Nuclear Doctrine declares that India is wedded to no-first use (NFU) — i.e., India will not be the first to use nuclear weapons, unless it is attacked with nuclear weapons. The NFU declarations, as experts have widely argued, are political, not operational statements.

In a 2019 article for International Security, Christopher Clary and Vipin Narang argued that “Indian officials are [increasingly] advancing the logic of counterforce targeting, and they have begun to lay out exceptions to India’s long-standing no-first-use policy to potentially allow for the preemptive use of nuclear weapons.” There is also other evidence that India has come a long way since the 2003 doctrine and is no longer tied down by NFU operationally.

Additionally, as is evident from debates on NFU, it is very difficult to verify a state’s NFU declaration. For instance, Li Bin, a professor at China’s Tsinghua University and an expert in nuclear strategy, argues that, without knowledge of a state’s size of nuclear force, its force composition, the accuracy of its weapons (for counterforce targeting) and conventional force strength, it is not possible to verify such a declaration.

The operational insignificance of NFU is also evident from what Nato officials found in the classified documents on Warsaw Pact war plans after the fall of the Berlin Wall. French diplomat and strategic expert Therese Delpech, wrote that “…military records of the Warsaw Pact that fell into German hands demonstrated beyond doubt that Russian operational plans called for the use of nuclear and chemical weapons in Germany at the onset of hostilities, even if Nato forces were using only conventional weapons.”

In view of the evolving Indian military thinking, Pakistan has no option but to consider India’s 2003 NFU as merely a declaratory policy without military significance.


On March 9, 2022, at 6:43pm, the Air Defence Operations Centre of the Pakistan Air Force (PAF) picked up and began tracking a high-speed flying object which, after flying for some time inside Indian airspace, sharply changed course and manoeuvred towards Pakistani territory. It violated Pakistan’s air space and ultimately fell near Mian Channu in Pakistani Punjab’s Khanewal district.

The next day, the Director-General Inter-Services Public Relations (DG-ISPR) briefed the media on the incident. The DG-ISPR noted that “It is important to highlight that the flight path of this object endangered many international and domestic passenger flights…as well as human life and property on ground. Whatever caused this incident to happen… it, nevertheless…reflects very poorly on their [India’s] technological prowess and procedural efficiency.”

The Indian government kept silent for two days after the incident before publicly confirming that one of its missiles had been accidentally launched while undergoing “routine maintenance” and that it had crossed into Pakistan.

On March 15, six days after the incident, speaking in the Indian parliament, Indian defence minister Rajnath Singh admitted to an “inadvertent release of a missile… during routine maintenance and inspection.”

He stated that: “It was later known that the missile fell in Pakistan’s territory. The incident is regrettable. But it’s a relief that no losses happened. I’d like to inform the House that the Government has taken this matter very seriously and [an] official order for a high-level probe has been given.”

Singh’s statement raised a number of technical and strategic questions, justifying Pakistan’s demand that any inquiry into the launch must involve a team of technical experts from Pakistan.

Pakistan’s statement also listed seven fundamental questions: the exact type of the missile/projectile that was accidentally launched; what was the missile’s programmed flight path; how and why it had veered from its flight path, stressing, as an Indian analyst termed “justifiably”, that “such an incident had the possibility of precipitating a prospective crisis if Islamabad had opted to react, as it was completely unaware that the missile was unarmed”; was the missile fitted with a self-destruct mechanism (if not, why not and if it were, why and how the device had malfunctioned); and what technical measures and procedures, if any, does India have in place to prevent accidental missile launches during “routine maintenance.”


In addition to the technical questions — answers to which are vital for a number of reasons, including strategic stability, safety and security — the important question is whether India informed Pakistan that a missile had been accidentally launched.

From Singh’s statement that India “later learnt” that the missile landed in Pakistani territory, it is obvious that, while the missile was in flight and even after it had gone off course, India did not inform Pakistan, either because it failed to monitor the missile’s flight or deliberately withheld the information.

No matter which way one cuts it, whether the operators were incompetent or complicit, the incident is indicative not just of India’s inability to handle sensitive technology, but also of the deeply worrying communication gaps between India and Pakistan, a contiguous nuclear dyad locked in a conflictual paradigm.

Serious questions have been raised by analysts within and outside India about multiple incidents and accidents that have happened over the years during equipment and systems handling.

As noted by one Indian analyst, “the Indian military services over the last several years have witnessed many high profile tragedies and mishaps.” During PAF’s Operation Swift Retort on February 27, 2019, while there was a dogfight going on over the skies along the Line of Control, Indian ground air defence shot down one of its own Russian Mi-17V5 ‘Hip’ medium-lift helicopters, killing six service officials and one civilian.

In two other incidents, “India’s indigenous Arihant [nuclear] submarine [was left] out of commission for many months in 2018; and a fire and explosion on board an Indian Kilo-class submarine in 2013…killed 18 crew members.”


As nuclear weapons states outside the framework of the NPT, Pakistan and India have certain legal and normative responsibilities. Their civilian nuclear programmes are under the International Atomic Energy Agency (IAEA) safeguards.

India, which has a 123 Agreement with the United States, has been trying to become a member of the Nuclear Suppliers Group and is already a member of the Missile Technology Control Regime. Pakistan, for its part, has constantly argued against discriminatory approaches to non-proliferation and disarmament. While supporting efforts on the CTBT and Fissile Material Cut-Off Treaty (FMCT), Pakistan continues to argue for a standardised approach to these multilateral arrangements at the Conference on Disarmament.

It has also flagged the requirement of regulating the emergence of new technologies, including hypersonic missiles, lethal autonomous weapons, cyber security, military uses of artificial intelligence and quantum computing.

There are two sets of problems in finding common ground. One is, of course, the nature of relations between India and Pakistan and the disputes that underpin it. The other is India’s perception of itself as a regional power that must find a place at the high table.

A good example of that, among others, is India’s bid for a permanent seat at the United Nations Security Council (UNSC). This desire, coupled with India’s tense relations with China, means that India is unwilling to accept a broader bilateral arrangement with Pakistan that it might consider as hampering the development of its conventional and nuclear military capabilities in relation to China.

The second problem is the desire by the United States to partner India and look at New Delhi as a net security provider in this region. This strategic approach is not just about India’s market and economic potential but also the US’ competition with China. These problems, going forward, will only exacerbate further.


Emerging technologies are increasing the risks in nuclear environments. For India and Pakistan the risks are much higher for the reasons cited above. There are a number of military confidence building measures in place between India and Pakistan but they have, at best, been partially respected.

As the situation stands, India does not seem to have much appetite to engage with Pakistan. Pakistan, on its part, has shown the desire to resume the normalisation process but has preconditioned it with India’s reversal of its August 5, 2019 decision that revoked the special status of Jammu and Kashmir.

In the foreseeable future, there does not seem to be much space for a serious dialogue on nuclear risk reduction measures even as risks continue to grow.

The writer is a journalist interested in security and foreign policies. He tweets @ejazhaider

We thank and applaud the brilliant scientists and engineers who worked tirelessly to achieve the Pakistani nuclear capability.




Top Bottom