On April 5, 2024, The Guardian published what remains the most consequential Western media investigation into India’s alleged campaign of targeted killings on Pakistani soil. The British newspaper claimed that Indian and Pakistani intelligence operatives had spoken on the record, anonymously, about a systematic program of assassinations carried out by India’s Research and Analysis Wing against individuals New Delhi considered hostile to Indian security interests. The investigation named nearly twenty killings since 2020, described an operational architecture running through sleeper cells in the United Arab Emirates, and quoted an unnamed Indian intelligence officer stating that such operations required approval from the highest levels of government. Indian media erupted. Pakistan seized the report as vindication. India’s Ministry of External Affairs dismissed every claim as false and malicious anti-India propaganda. None of these reactions addressed the central analytical question the investigation raised, which is not whether India ordered the killings but what the report’s sourcing, methodology, and evidentiary foundation actually proved and where its claims crossed from confirmed fact into informed inference.

The distinction matters because the Guardian investigation occupies a unique position in the broader narrative of India’s shadow war against terrorism. Before April 2024, the campaign of targeted killings in Pakistan existed in a peculiar informational twilight. Pakistani media reported individual shootings. Indian media occasionally celebrated them. Analysts tracked the pattern. But no major Western outlet had assembled the pieces into a single investigative narrative, sourced to intelligence operatives from both countries, and presented them to a global audience. The Guardian’s report did not create the shadow war. It placed it on the Western public record. That act of documentation, independent of whether every claim in the report survives scrutiny, changed the informational environment in which the campaign operates. Understanding what the investigation actually established, and what it did not, is essential to evaluating every subsequent development in the campaign, from the post-publication acceleration of killings to the international diplomatic fallout that followed.

This article reconstructs the Guardian investigation in granular detail, analyzes its source categories and what each can credibly attest to, identifies where the reporting crosses from established fact into intelligence-agency narrative, and adjudicates the competing interpretations of the report’s significance. The findable artifact embedded in this analysis is a source-reliability assessment examining the investigation’s sourcing architecture: what unnamed Indian operatives can plausibly know, what unnamed Pakistani operatives have incentives to claim, what the documentary evidence allegedly obtained from Pakistani investigators can and cannot demonstrate, and where each source category’s testimony overlaps or contradicts the others. The namable claim is direct: The Guardian’s investigation did not prove that India ordered the killings. What it proved is that unnamed intelligence operatives from both India and Pakistan believe India ordered the killings and were willing to tell a British newspaper. The evidentiary gap between intelligence operatives say and India ordered is the gap the campaign operates within, and that gap is analytically productive precisely because it reveals the strategic logic of deniability that makes the entire shadow war possible.

The Investigation Lands

The Guardian published its investigation under a headline asserting that the Indian government had ordered killings in Pakistan to eliminate terrorists on foreign soil. The timing was notable. India was weeks away from the first phase of its 2024 general election. Prime Minister Narendra Modi was campaigning for a third consecutive term. The investigation’s central claim, that RAW had been conducting targeted assassinations abroad with approval from the highest level of government, landed in an electoral context where Modi’s national security credentials were a defining campaign theme. Critics immediately accused The Guardian of timing the publication to influence the election. Defenders of the investigation argued that the timing was irrelevant to the sourcing. Both positions contain truth, and neither addresses the investigation’s analytical substance.

The report was not a single-source story. The Guardian described its sourcing as including intelligence operatives from India and Pakistan, documents shared by Pakistani investigators, witness testimonies, arrest records, financial statements, WhatsApp messages, and passport details. The newspaper stated that it had seen these documents but acknowledged they could not be independently verified. This acknowledgment is analytically significant. It means the investigation’s documentary claims rest on materials provided by one party to the conflict, Pakistani investigators, whose institutional interests in establishing Indian culpability are substantial and whose reliability as neutral documentarians of evidence is, at minimum, contestable.

The investigation’s most quoted claim came from an unnamed Indian intelligence officer who stated that after the Pulwama attack in February 2019, the approach changed to target elements outside the country before they could launch attacks or create disturbances. The officer added that the safe havens were in Pakistan, so the agency had to get to the source. He further stated that such operations needed approval from the highest level of government. A second Indian intelligence officer provided corroborating statements. The Guardian reported that these officers described RAW as having drawn inspiration from Israel’s Mossad and Russia’s KGB, and that the killing of Saudi journalist Jamal Khashoggi had been directly cited by RAW officials as a reference point. This last detail, the Khashoggi reference, is the most analytically provocative element in the investigation because it positions the shadow war not as a defensive counter-terrorism measure but as part of a broader embrace of extraterritorial lethal operations by intelligence services worldwide, a framing that transforms the legal, moral, and strategic analysis of every killing in the complete timeline of targeted eliminations.

The Source Architecture Deconstructed

Understanding what the Guardian investigation proved requires disaggregating its sources into categories and assessing what each category can credibly attest to. This source-reliability assessment is the analytical artifact that no other treatment of the investigation has produced, and it is the framework through which every claim in the report should be evaluated.

Source Category One: Unnamed Indian Intelligence Operatives

The Guardian spoke to at least two Indian intelligence officers. These sources are the investigation’s most consequential because they are the only ones who can plausibly speak to the decision-making process behind the alleged operations. An unnamed Pakistani official can assert that India is responsible, but only an Indian intelligence operative can describe the internal logic, the Pulwama trigger, the highest-level approval, the Mossad inspiration, from the perspective of the alleged perpetrator.

The credibility assessment for this source category requires examining three questions. First, could these individuals plausibly know what they claimed to know? If they were genuine RAW operatives involved in or aware of the program, then yes, their descriptions of the program’s origins, authorization chain, and operational philosophy would reflect inside knowledge. Second, what were their motivations for disclosure? Indian intelligence operatives speaking to a British newspaper about covert operations against Pakistani targets would be committing a serious breach of security protocols. Their willingness to speak suggests either institutional sanction (the disclosures were authorized or tolerated by superiors), personal disenchantment (the operatives disagreed with the program), or a deliberate strategic communication (the disclosure was intended to send a message to Pakistan, to the international community, or to both). Third, can their claims be independently verified? The specific operational details they provided, the Pulwama trigger, the highest-level approval, are assertions of internal government decision-making that cannot be confirmed through open-source reporting. They can only be corroborated by additional insider sources or contradicted by the absence of the pattern they describe.

The analytical position this article takes is that the Indian intelligence operatives, if genuine, are the strongest source category in the investigation but also the most strategically compromised. Genuine inside sources can describe what happened but may frame their descriptions in ways that serve institutional interests. An operative who says the program was inspired by Mossad may be accurately reporting internal conversations or may be constructing a narrative that positions RAW as a peer of the world’s most celebrated intelligence service. Both framings can coexist. The choice to reference Mossad rather than, say, the CIA’s rendition program or France’s targeted killings in the Sahel, is itself a framing decision that aligns the alleged program with the most defensible precedent rather than the most controversial ones. The analytical value of their testimony lies not in whether every detail is precisely accurate but in the general contours: that a program exists, that it was triggered by a specific event, that it required senior authorization, and that the operational architecture involves extraterritorial infrastructure and recruited local assets. These contours are consistent with the pattern evidence assembled across the entire shadow war series.

Source Category Two: Unnamed Pakistani Intelligence Officials

The Guardian cited unnamed senior officials from two separate Pakistani intelligence agencies. These sources asserted that Islamabad suspected Indian involvement in up to twenty killings since 2020. They pointed to evidence from previously undisclosed inquiries into seven of the cases. The Pakistani sources provided the investigative documents, witness testimonies, arrest records, financial statements, WhatsApp messages, and passport details that the Guardian described as the documentary foundation of its report.

The credibility assessment for Pakistani intelligence sources is different from the Indian category, and the difference is crucial. Pakistani officials have a direct institutional interest in establishing Indian culpability for the killings. Every killing of a designated terrorist on Pakistani soil raises an uncomfortable question for Islamabad: how were these individuals living in Pakistan in the first place? The individuals killed, from Shahid Latif of Jaish-e-Mohammed to Muhammad Riaz of Lashkar-e-Taiba to Paramjit Singh Panjwar of the Khalistan Commando Force, were all individuals that Pakistan had publicly denied sheltering. Attributing their deaths to Indian intelligence operations accomplishes two things simultaneously: it shifts attention from Pakistan’s role as a safe haven to India’s role as an alleged assassin, and it reframes Pakistan from a state that protects terrorists into a victim of foreign aggression.

This does not mean the Pakistani evidence is fabricated. Institutional motivation to reach a particular conclusion is not the same as fabricating evidence to support it. Pakistan may genuinely have apprehended individuals involved in the killings, and those individuals may genuinely have confessed to Indian handlers, and the financial trails may genuinely lead to UAE-based intermediaries. The question is not whether any of this happened but whether the evidentiary chain from apprehended shooter to named Indian intelligence agent to authorized government program is as clean as the Pakistani narrative suggests or whether it contains gaps that investigative methodology should flag.

The most specific claim in the Pakistani evidence concerns the killing of Shahid Latif in Sialkot in October 2023. Pakistani investigators said they had apprehended the shooter, Muhammad Umair, who was trying to flee the country. They named an Indian agent, Yogesh Kumar, as the orchestrator operating from a third country, probably the UAE. They presented documentary and financial evidence linking the operational chain. On the killing of Muhammad Riaz in Rawalakot in September 2023, they named a second apprehended shooter, Muhammad Abdullah Ali, and traced his recruitment to Indian agents Ashok Kumar Anand and Yogesh Kumar, who had used Telegram to contact him. They released passport details of the two named Indian agents. Foreign Secretary Muhammad Syrus Sajjad Qazi presented this evidence at a press conference on January 25, 2024, formally accusing India of extraterritorial and extrajudicial killings.

This evidence, if accurate, constitutes the strongest public case any government has made linking India to specific targeted killings in Pakistan. The named agents, the apprehended shooters, the financial trails, and the communication records form a coherent narrative of a kills-for-hire operation run from the UAE. The investigative question is whether coherent also means complete and whether the apprehended shooters’ confessions were obtained under conditions that would survive independent forensic scrutiny. Pakistan did not provide access to the apprehended individuals for independent interviews, did not subject the documentary evidence to third-party verification, and did not explain why, if it possessed such compelling evidence, it waited months before presenting it publicly. These gaps do not invalidate the evidence. They define the analytical space within which the evidence should be evaluated.

Source Category Three: The Documentary Archive

The Guardian stated that it had seen documents shared by Pakistani investigators, including witness testimonies, arrest records, financial statements, WhatsApp messages, and passports, but that these could not be independently verified. This caveat transforms the evidentiary status of the documentary archive from primary evidence to reported evidence: the documents exist, a journalist has seen them, but their authenticity, completeness, and chain of custody remain unverified.

The analytical weight of unverified documentary evidence is a familiar problem in investigative journalism. The Guardian has a strong institutional track record on intelligence reporting. The newspaper published the Edward Snowden disclosures on NSA surveillance, was part of the consortium that investigated the Pegasus spyware program, and has broken multiple stories on British intelligence operations that proved accurate despite initial government denials. This track record supports the inference that The Guardian’s editorial standards for publishing intelligence-related claims are higher than average and that the newspaper would not have presented the documentary claims without internal editorial review of the materials. But institutional track record is not the same as independent verification of specific documents in a specific investigation. A newspaper that got the Snowden story right is not necessarily right about every subsequent intelligence story, and the Pakistani documents deserve the same skepticism that any unverifiable source material demands.

The most analytically revealing detail in the documentary archive concerns the alleged recruitment mechanisms. Pakistani investigators claimed that RAW sleeper cells operating from the UAE paid millions of rupees to local criminals or impoverished Pakistanis to carry out the assassinations. In one specific case, investigators alleged that a twenty-year-old illiterate Pakistani had been hired by RAW through intermediaries while working for low wages in an Amazon packing warehouse in the UAE. He was allegedly paid 1.5 million Pakistani rupees, roughly equivalent to four thousand British pounds, to track down Shahid Latif, and was promised fifteen million rupees and his own catering company if he completed the killing.

This detail, if accurate, reveals an operational model that is simultaneously sophisticated in its compartmentalization and crude in its execution. The sophistication lies in the multiple layers of separation between the alleged state sponsor and the triggerman: Indian handlers in the UAE, intermediary facilitators, local recruits who may not have known whom they were ultimately serving. The crudeness lies in the recruitment of illiterate, impoverished individuals for assassination operations that require surveillance, target identification, and escape, skillsets that contradict the profile of an untrained laborer. This tension between sophistication and crudeness is itself analytically significant. It suggests either that the operational model was more improvised than the strategic narrative implies, or that the recruitment details have been embellished in the retelling, or that the operation relied on local facilitators who handled the actual tradecraft while the recruited individual served as a deniable triggerman. Each interpretation has different implications for assessing the modus operandi of the broader campaign.

The Pulwama Trigger and the Doctrinal Shift

Both Indian intelligence operatives cited the Pulwama attack of February 2019 as the catalytic event that transformed RAW’s approach to extraterritorial operations. On February 14 of that year, a suicide bomber driving a vehicle laden with explosives rammed a Central Reserve Police Force convoy on the Jammu-Srinagar National Highway in Pulwama district, killing forty paramilitary personnel. Jaish-e-Mohammed, the Pakistan-based militant organization, claimed responsibility. The attack occurred while Prime Minister Modi was campaigning for a second term, and the government’s response, the Balakot airstrikes twelve days later, became the defining national security event of the 2019 election campaign.

The Guardian’s Indian sources described the post-Pulwama doctrinal shift in terms that position it as a strategic evolution rather than a reactive escalation. The framing was preemptive: target elements outside the country before they can launch an attack. This framing is significant because it redefines the shadow war from a retaliatory campaign, punishing past attacks, into a preventive one, eliminating future threats. The distinction has legal, moral, and strategic implications. Retaliatory operations carry a different justification framework than preventive ones. Retaliation responds to a specific harm and invokes, however controversially, the logic of accountability. Prevention responds to a perceived future threat and invokes the logic of preemption, which requires intelligence assessments of threat levels, target-selection criteria, and proportionality judgments that operate entirely within classified decision-making processes.

The Pulwama trigger also provides a temporal framework that can be tested against the pattern evidence. If the doctrinal shift occurred after February 2019, then the pattern of targeted killings should show a clear inflection point around that date. The earliest killings attributed to the campaign, the Lahore car bomb that targeted Hafiz Saeed’s convoy and the killing of Zahoor Mistry, both occurred several years later, consistent with a program that required planning, infrastructure development, and asset recruitment before becoming operational. The two-year build-up described by Pakistani officials, who said India spent approximately two years establishing sleeper cells in the UAE before the killings began, would place the operational start around 2021, which aligns with the timeline of the first confirmed targeted eliminations. This temporal consistency does not prove the Pulwama trigger narrative is true, but it does establish internal coherence between the claimed origin story and the observed operational pattern.

The Mossad and KGB inspiration claim adds another dimension to the doctrinal analysis. If Indian intelligence operatives genuinely cited these agencies as models, it positions RAW within a specific tradition of state-sponsored targeted killing that has its own analytical literature, legal debates, and strategic assessments. Mossad’s targeted killing program, which Israel has conducted for decades against Palestinian militants, Iranian nuclear scientists, and others, operates under a framework of self-defense that Israel has articulated publicly even when denying specific operations. The KGB’s assassination programs, from the killing of Ukrainian nationalist leaders in the Cold War to the Salisbury poisoning in 2018, operated under a framework of state security that tolerated international condemnation as a cost of operations. The Khashoggi reference is the most disturbing element because Khashoggi’s killing was universally condemned as an extrajudicial murder by an authoritarian regime, not as a legitimate intelligence operation. If RAW officials genuinely cited Khashoggi as a reference point, it suggests either a morally unreflective institutional culture or a deliberate provocation by the source, designed to maximize the investigation’s impact by associating India’s program with the most notorious state-sponsored killing of the decade. The comparison between India’s shadow war and Mossad’s decades-long targeted killing program illuminates the doctrinal parallels and divergences in greater depth.

The UAE Connection and the Operational Architecture

The most operationally specific claims in the Guardian investigation concern the alleged role of the United Arab Emirates as the logistical hub for the killing program. Pakistani investigators described a network of Indian intelligence sleeper cells based primarily in the UAE, tasked with organizing the assassinations. The operational model, as described, involved several layers. RAW handlers allegedly met in multiple locations, including the UAE, Nepal, the Maldives, and Mauritius, to plan operations. These handlers then recruited intermediaries who in turn recruited local Pakistanis, some of them criminals and some impoverished laborers, to carry out the actual killings. Payments flowed through Dubai-based financial channels, with amounts ranging from hundreds of thousands to millions of Pakistani rupees.

The UAE’s role in this alleged architecture is geographically and logistically plausible. The UAE, particularly Dubai, hosts large Pakistani and Indian diaspora populations. Financial flows between Dubai and Pakistan are well-established through both formal banking channels and informal hawala networks. Dubai’s position as a commercial hub means that the movement of individuals between India, the UAE, and Pakistan generates limited suspicion compared to direct India-Pakistan travel, which would be closely monitored by both countries’ intelligence services. The selection of the UAE as a base for compartmentalized operations is consistent with standard intelligence tradecraft for managing agents in denied areas.

Pakistani investigators provided specific details on the operational chain for two cases. In the killing of Shahid Latif, the JeM commander shot outside a mosque in Sialkot in October 2023, investigators named an Indian agent identified as Yogesh Kumar and traced the recruitment chain through UAE-based intermediaries to the apprehended shooter, Muhammad Umair. Financial transactions linking the chain were allegedly documented. In the killing of Muhammad Riaz, an alleged LeT commander shot during Fajr prayer at a mosque in Rawalakot in September 2023, investigators traced the recruitment to Indian agents Ashok Kumar Anand and Yogesh Kumar. The killer, Muhammad Abdullah Ali, was allegedly recruited through Telegram, provided with weapons and ammunition, and failed in a first attempt on September 7 before succeeding on September 8. He was apprehended on September 15 while attempting to board a flight at Jinnah International Airport in Karachi.

Pakistan’s Foreign Secretary released the passport details of Yogesh Kumar, born on October 9, 1996, in Sriganganagar, Rajasthan, and Ashok Kumar Anand, born on July 1, 1958, in Kolya. The release of passport details is an unusual step that signals Pakistan’s confidence in this specific evidence. It is also a step that invites verification: if India’s intelligence agencies employ operatives under these names with these biographical details, the release of their passport information constitutes a significant operational exposure. If the names are cover identities, the passport details may lead nowhere. India’s refusal to address the specific individuals named by Pakistan leaves this question analytically open.

The recruitment of jihadists deserves separate analysis. Pakistani investigators claimed that Indian agents recruited jihadists to carry out some of the killings by telling them they were killing infidels. If true, this represents a remarkably cynical operational technique: exploiting the religious ideology of one set of extremists to kill another set of extremists. The jihadist recruits would have believed they were conducting religiously motivated killings while actually serving as deniable instruments of a foreign intelligence service. This detail is difficult to verify but is internally consistent with an operational model that prioritizes deniability. A jihadist shooter leaves a different investigative trail than a professional assassin. Law enforcement agencies investigating the killing would initially pursue jihadi violence as the motive, which delays or prevents the attribution of the killing to a state actor. The analysis of the modus operandi across all confirmed eliminations shows that the absence of any claim of responsibility is one of the campaign’s defining characteristics, and the use of recruited jihadists who themselves believe they are acting on religious conviction would explain this silence: the recruits would not know whom to credit, and their handlers would have no incentive to claim responsibility.

The Specific Cases: Latif, Riaz, and the Operational Granularity

The Guardian investigation’s credibility rests most heavily on the specific cases for which Pakistani investigators provided detailed operational reconstructions. Two cases, the killings of Shahid Latif and Muhammad Riaz, received the most granular treatment, and the level of detail in these cases determines whether the investigation’s broader claims about a systematic program are supported by evidence or extrapolated from limited data.

Shahid Latif was shot outside a mosque in Sialkot on October 14, 2023. Latif was a Jaish-e-Mohammed commander whom Indian security agencies identified as a key figure in the Pathankot airbase attack of January 2016, in which one civilian and seven Indian security personnel were killed. Latif had survived multiple previous assassination attempts, a detail that suggests either persistent targeting by the same campaign or a high threat level from multiple sources. Pakistani investigators claimed that the killing was orchestrated by the Indian agent identified as Yogesh Kumar, operating from a third country. The shooter, Muhammad Umair, was apprehended while attempting to flee Pakistan on October 12, 2023, three days before the successful killing. Investigators described a chain of command running from Indian handlers through UAE-based intermediaries to the recruited shooter, with financial transactions documented at each link. The level of detail in this case, the named handler, the apprehended shooter, the financial trail, constitutes the most specific publicly available evidence linking a particular killing to alleged Indian intelligence operations.

The Latif case is analytically significant for several reasons beyond its evidentiary specificity. Latif was not a minor figure. His connection to the Pathankot attack placed him on India’s highest-priority target list, and the broader analysis of Jaish-e-Mohammed’s operational structure places Latif within the inner circle of JeM’s Punjab-based operational command. Targeting a figure of Latif’s seniority requires intelligence on his daily movements, his prayer schedule, his security arrangements, and his network of associates. The mosque location is consistent with the pattern documented across the campaign: attackers exploit the one daily routine that is both predictable and socially undisruptable. A target who varies his daily schedule can still be tracked to the mosque because prayer times are fixed and publicly known. The selection of this vulnerability point across multiple killings is itself evidence of a coordinated operational doctrine rather than random violence. The fact that previous attempts on Latif had failed suggests that the targeting was persistent and iterative, with operational learning from each failed attempt informing the next. This persistence is inconsistent with opportunistic violence by criminal gangs or sectarian rivals and consistent with a directed campaign with sustained institutional backing.

Muhammad Riaz was shot during Fajr prayer at a mosque in Rawalakot, in Pakistan-administered Kashmir, on September 8, 2023. Indian media identified Riaz as a top commander of Lashkar-e-Taiba, though Pakistan did not confirm his alleged organizational affiliation. The Guardian investigation, drawing on Pakistani investigative materials, described a recruitment chain beginning with Indian agents Ashok Kumar Anand and Yogesh Kumar, who used Telegram to recruit Muhammad Abdullah Ali as the triggerman. Ali was allegedly recruited in the UAE, provided with weapons and ammunition, and tasked with locating and killing Riaz. A first attempt on September 7 failed. The successful killing occurred the following day. Ali was tracked by Pakistani law enforcement and apprehended on September 15 while attempting to board a flight at Jinnah International Airport in Karachi, seven days after the killing.

The Riaz case adds several analytically significant details to the operational picture. The use of Telegram for recruitment suggests a digital communication channel that provides both reach and a degree of operational security, though Telegram’s metadata and message retention policies mean that investigative agencies with access to the platform can reconstruct communication histories. The failed first attempt on September 7 is particularly revealing. Professional intelligence operations anticipate the possibility of failure and build contingency into the operational plan. The fact that the recruited shooter attempted the kill a second time, the very next day, suggests either desperate recklessness or instructions to complete the mission regardless of the increased risk. The rapidity of the second attempt is inconsistent with the careful surveillance and planning that characterized other killings in the campaign and may indicate that the outsourced execution model, using recruited amateurs rather than trained operatives, produces operational quality that varies significantly from case to case.

The apprehension of Ali at Karachi airport provides a critical intelligence window. If Ali’s confessional statements are accurate and were obtained under conditions that a court would recognize as reliable, they provide a direct link between a specific killing and named Indian agents. The chain from recruited shooter to Indian handler to government authorization is the evidentiary chain that the shadow war’s deniability architecture is designed to prevent. The fact that Pakistan was able to apprehend Ali and reconstruct the operational chain suggests either that the operation’s compartmentalization was imperfect, a realistic assessment for an operation conducted by recruited amateurs, or that Pakistani intelligence had been monitoring the killing campaign and was positioned to react quickly when a shooter could be identified. Both interpretations have implications for the campaign’s future operational security and for Pakistan’s capacity to disrupt future operations.

Beyond the Latif and Riaz cases, the investigation referenced approximately eighteen additional killings without providing comparable operational detail. Among the individuals named or referenced in the broader coverage surrounding the investigation were Paramjit Singh Panjwar, the Khalistan Commando Force chief shot in Lahore in May 2023; Saleem Rehmani, on India’s most-wanted list, killed in January 2022; and multiple other individuals associated with Lashkar-e-Taiba, Jaish-e-Mohammed, and Hizbul Mujahideen. The lack of specific operational detail on these additional cases means they are included in the investigation’s count on the basis of pattern similarity rather than independent evidence. Each case individually might be attributable to internal Pakistani violence, sectarian conflict, or criminal activity. The investigation’s argument is that the aggregate pattern, the consistent targeting of India-wanted individuals, the similar modus operandi, the absence of any claim of responsibility, is too coherent to be coincidental. This is a reasonable analytical position, but it should be distinguished from the evidentiary certainty available for the Latif and Riaz cases.

The operational granularity of these two cases also reveals a tension in the alleged program’s design. The use of recruited amateurs, whether poor laborers from UAE warehouses or individuals contacted through Telegram, introduces an inherent vulnerability: amateurs get caught, and when caught, they talk. A program that depends on deniability but uses recruitable, capturable, confessing triggermen is a program that accepts a significant probability of exposure. This suggests either that the program’s architects judged the risk of exposure as acceptable because the diplomatic consequences of attribution would be manageable, or that the operational model evolved pragmatically rather than by strategic design, with the emphasis on deniable distance from the triggerman outweighing the risk that the triggerman himself might be apprehended and compelled to reveal the chain of command. The post-investigation trajectory of the campaign, in which the killings continued and accelerated despite the exposure of exactly this chain of command, suggests that the first interpretation is closer to the mark: the program’s architects had already calculated that exposure would not produce consequences sufficient to justify halting operations.

The Electoral Timing Controversy

The Guardian’s decision to publish on April 5, 2024, weeks before India’s general election commenced on April 19, 2024, generated a controversy that in many respects overshadowed the investigation’s substance. Indian government officials and aligned commentators treated the timing as prima facie evidence of political motivation. The argument held that The Guardian, a newspaper editorially critical of Modi’s government on issues ranging from press freedom to minority rights, had weaponized an intelligence story to maximize electoral damage. The release of the investigation during the campaign period, when Indian election law restricts certain forms of political communication and when the electorate’s attention is focused on candidate narratives, amplified this suspicion.

The timing argument has surface plausibility but analytical weaknesses. Investigative journalism operates on its own timeline. Investigations are published when the sourcing is complete, the editorial review is finished, and the lawyers have cleared the text. The timing of publication is determined by when the investigation is ready, not by the political calendar of the country being investigated. The Guardian may have been aware that the publication would coincide with the Indian election campaign, but awareness of electoral timing is different from deliberate electoral interference. Major newspapers publish stories about governments during election periods routinely; the alternative, withholding completed investigations until after elections, would itself constitute a form of editorial interference in the democratic process.

There is a counter-argument that the timing actually weakened the investigation’s impact within India. Indian voters in the midst of an election campaign were primed to interpret any critical coverage of the Modi government through a partisan lens. The investigation’s substantive claims about RAW operations were immediately subsumed into the election narrative, with supporters dismissing it as foreign interference and opponents treating it as evidence of government overreach. The analytical substance, the sourcing architecture, the evidentiary gaps, the strategic implications, received almost no serious public attention in India. The political noise drowned out the analytical signal. If The Guardian intended maximum analytical impact rather than maximum political impact, publishing after the election would have been more effective. The fact that it published during the campaign period may simply reflect the investigative timeline rather than editorial strategy.

The timing controversy also distracted from a more consequential timing question: why did the Indian intelligence operatives choose to speak at this particular moment? Unnamed sources do not appear spontaneously; they decide to talk, and their decision reflects calculations about timing, audience, and impact. Indian intelligence operatives speaking to a British newspaper during an election campaign may have been acting with institutional sanction, using the investigation as a channel to signal capability and resolve to multiple audiences simultaneously: to Pakistan, to the international community, and to Indian voters. This interpretation is speculative but consistent with the pattern of strategic communication that intelligence agencies worldwide employ. The sources may have chosen The Guardian precisely because of its timing proximity to the election, knowing that the combination of intelligence disclosure and electoral sensitivity would generate maximum international attention.

Indian Media’s Coverage and the Domestic Information Environment

Indian media’s response to the Guardian investigation reflected the deep polarization of the country’s information environment. Government-aligned outlets, which constitute a substantial portion of India’s television news landscape, treated the investigation as an attack on national security and sovereignty. The framing was defensive: a foreign newspaper with a history of anti-India coverage was amplifying Pakistani propaganda to undermine the Modi government during an election. The specific claims of the investigation, the named agents, the financial trails, the apprehended shooters, received minimal analytical attention. The focus was on the messenger rather than the message.

Opposition-aligned and independent media outlets took a different approach but faced their own constraints. In an election environment where national security is a dominant campaign theme, treating the Guardian investigation as credible implicitly challenges the government’s counter-terrorism narrative. Some independent analysts engaged with the investigation’s substance, noting its strengths and limitations. But the broader Indian media discourse was dominated by the binary framing: either the investigation was a hit job or it was vindication. The analytical middle ground, that the investigation contained credible elements requiring serious assessment but also significant limitations requiring critical scrutiny, received comparatively little airtime.

The social media response was more revealing of public sentiment. On platforms where government supporters are well-organized, the Guardian investigation was met with a response that combined denial of the specific allegations with celebration of the alleged operations. The dissonance is analytically notable: the same commentators who dismissed the investigation as fabrication simultaneously celebrated the killing of terrorists in Pakistan, effectively denying the specific claims while endorsing the broader program they denied existed. This cognitive dissonance is not unique to India; it is a common feature of public responses to covert operations that enjoy popular support. Citizens want the operations to happen but do not want the state to be accountable for them. The evidentiary gap that protects the state from international accountability also provides domestic audiences with the space to maintain contradictory positions.

The media coverage also highlighted a structural asymmetry in the information environment. The Guardian investigation relied on sources from both countries but was published for a Western audience. Indian audiences accessed the investigation primarily through Indian media’s interpretation of it, which was colored by each outlet’s political alignment. Pakistani audiences accessed it through Pakistani media’s interpretation, which treated it as vindication. The original investigation’s nuances, its acknowledgment that documents could not be independently verified, its presentation of competing narratives, were largely lost in the secondary coverage. This information-laundering effect, in which an investigation’s careful qualifications are stripped away as it passes through successive layers of media interpretation, is a structural feature of international reporting on intelligence matters and should inform how the investigation’s impact is assessed.

India’s Response and the Rhetoric of Denial

India’s response to the Guardian investigation was swift, categorical, and consistent with its responses to every previous allegation of extraterritorial operations. The Ministry of External Affairs dismissed the report as false and malicious anti-India propaganda. The ministry reiterated a previous statement by External Affairs Minister S. Jaishankar that targeted killings in other countries were not the government of India’s policy. This formulation is precise and worth parsing. Jaishankar did not say India has never conducted targeted killings. He said targeted killings are not the government’s policy. The distinction between operational practice and declared policy is not academic in intelligence affairs. States routinely conduct operations that their declaratory policies deny. The United States denied conducting drone strikes in Pakistan for years after the strikes were publicly documented. Israel has maintained a policy of ambiguity about its nuclear arsenal for decades. The gap between what a state does and what it declares as policy is a structural feature of covert operations, not an anomaly.

India’s denial was not accompanied by any specific rebuttal of the investigation’s claims. The MEA did not address the named Indian agents, the financial evidence, the apprehended shooters, or the documentary archive described by The Guardian. The denial was wholesale: everything was propaganda. This approach is tactically effective in the short term because it avoids engaging with specific claims that might legitimize the investigation. But it is analytically unsatisfying because it leaves every specific allegation standing without rebuttal. If the named Indian agents are cover identities created by Pakistani intelligence, saying so would discredit the investigation’s most specific claims. If the financial trails are fabricated, demonstrating fabrication would undermine the documentary foundation. India chose not to engage at this level of specificity, a choice that is consistent with a state protecting operational security by refusing to discuss operational details but that also prevents external observers from evaluating the strength of India’s position.

The Indian media’s response was divided along predictable political lines. Government-aligned outlets treated the investigation as a hit job timed to damage Modi before the election. Opposition-aligned outlets treated it as credible journalism that raised legitimate questions about extrajudicial violence. Independent security analysts were more nuanced, noting that the investigation’s sourcing had strengths and weaknesses and that neither wholesale acceptance nor wholesale rejection was analytically justified. The investigation’s most consequential effect on Indian domestic discourse was not to change opinions about the shadow war, most Indians who followed the story had already formed positions, but to elevate the shadow war from a subject of specialized security commentary to a subject of mainstream political debate.

Pakistan’s Strategic Dilemma and the Reluctance to Acknowledge

The investigation exposed a strategic dilemma for Pakistan that no amount of diplomatic language can conceal. Every targeted killing on Pakistani soil forces Islamabad to answer a question it would prefer to avoid: why were India’s most-wanted terrorists living in Pakistan? The individuals killed, from senior commanders of Lashkar-e-Taiba and Jaish-e-Mohammed to Khalistan-linked operatives, were people whose presence on Pakistani soil Pakistan had publicly denied or minimized. Acknowledging that they were killed in targeted assassinations requires acknowledging that they existed in Pakistan, which contradicts years of Pakistani denials about sheltering designated terrorists.

This explains Pakistan’s delayed and selective response to the killing campaign. For months, as individuals on India’s most-wanted lists were shot by unknown gunmen across Pakistani cities, Islamabad said nothing. The silence was not ignorance; Pakistani intelligence agencies were tracking the killings. The silence was strategic: acknowledging the killings meant acknowledging the targets, and acknowledging the targets meant acknowledging that Pakistan had provided sanctuary to designated terrorists, a point that India, the United States, the Financial Action Task Force, and multiple international bodies had been making for years. The Guardian investigation broke this silence not because Pakistan chose to speak but because the newspaper’s independent reporting made continued silence unsustainable.

When Pakistan did finally speak, through Foreign Secretary Qazi’s January 2024 press conference, the language was carefully chosen. Qazi referred to the killed individuals as Pakistani nationals, not as terrorists, not as militants, not as designated operatives. The framing was sovereign violation, not counter-terrorism debate. By positioning the killings as violations of Pakistani sovereignty by a hostile foreign intelligence agency, Pakistan reframed the conversation from should these individuals have been protected to how dare India operate on our soil. The sovereign-violation frame is legally sound. Extraterritorial killings do violate sovereignty regardless of the targets’ status. But the frame also deflects from the underlying question: why a state that claims to be combating terrorism was providing sanctuary to individuals designated as terrorists by India, the United Nations, and multiple Western governments.

The analysis of Pakistan’s evolving counter-narrative traces how Islamabad’s position shifted from denial of the killings themselves to denial of the targets’ significance to formal accusation of Indian involvement, with each shift driven not by a change in evidence but by the increasing impossibility of maintaining the previous narrative as the body count grew and international attention intensified.

The Legal and Ethical Dimensions Raised by the Investigation

The Guardian investigation implicitly raises legal and ethical questions that neither the newspaper nor any of the governments involved has addressed systematically, but that shape the long-term strategic significance of the shadow war.

The legal framework for state-sponsored targeted killings on foreign soil is contested, inconsistent, and largely determined by political power rather than jurisprudential principle. The United States conducted thousands of drone strikes in Pakistan, Yemen, Somalia, and Afghanistan against individuals designated as terrorists, with minimal international legal consequence. Israel’s targeted killing program has been challenged in Israeli courts, debated by international legal scholars, and condemned by human rights organizations, but it continues. Russia’s poisoning of Sergei Skripal in Salisbury in 2018 and Saudi Arabia’s killing of Jamal Khashoggi in Istanbul the same year provoked international condemnation and diplomatic consequences, but neither resulted in formal legal proceedings against the responsible states.

India’s alleged program, as described in the Guardian investigation, falls into this contested legal space. If India is conducting targeted killings of designated terrorists on Pakistani soil, the legal analysis depends on several factors: whether the targets are legitimate military targets under international humanitarian law, whether the state of armed conflict between India and Pakistan (which is contested) provides legal cover for lethal operations, whether the principle of self-defense under Article 51 of the UN Charter extends to extraterritorial operations against non-state actors operating from a state that is unable or unwilling to address the threat, and whether the means employed (recruited amateurs, purchased assassins) fall within any recognized legal framework for the use of force.

None of these questions has a settled answer. The legal ambiguity is not accidental; it is the space within which states conducting targeted killings operate. The absence of clear legal prohibition, combined with the absence of clear legal authorization, creates a permissive environment in which states can act and deny, and in which the international community can observe and decline to act. The Guardian investigation did not resolve these legal questions, and it would be unreasonable to expect a newspaper investigation to do so. But by documenting the alleged program in detail, the investigation created a factual record that future legal proceedings, should they ever occur, would need to engage with.

The ethical dimensions are equally complex. The individuals allegedly killed in Pakistan were, with few exceptions, designated terrorists responsible for or complicit in attacks that killed dozens or hundreds of people. Shahid Latif was linked to the Pathankot airbase attack. Muhammad Riaz was identified as a Lashkar-e-Taiba commander associated with operations that killed Indian security personnel and civilians. Paramjit Singh Panjwar led a Khalistan separatist organization designated as terrorist by the Indian government. The ethical argument for targeted killings rests on the claim that these individuals posed ongoing threats that could not be neutralized through legal channels because Pakistan refused to extradite them, prosecute them, or even acknowledge their presence on its soil.

The ethical argument against targeted killings rests on several counter-claims: that extrajudicial killing, regardless of the target’s record, violates the rule of law and the right to life; that the recruitment of impoverished individuals as disposable triggermen raises distinct moral concerns about exploitation; that the pattern of killing targets at mosques during prayer exploits a sacred space and time in ways that offense extends beyond the individual target to the community; and that accepting the principle of extraterritorial targeted killing by one state necessarily legitimizes the same practice by every state, with unpredictable consequences for international order. The comparison with Israel’s decades-long targeted killing program provides the most relevant framework for evaluating these competing ethical positions, because Israel has engaged in the most extensive public and judicial debate about the legality and morality of targeted killings by a democratic state.

The ethical analysis is further complicated by the operational model described in the Guardian investigation. If the program relies on recruiting impoverished laborers from UAE warehouses and paying them the equivalent of a few thousand dollars to kill human beings, the moral calculus involves not only the target but the triggerman. The recruited killer who is paid four thousand British pounds to assassinate a designated terrorist and is then apprehended, confesses, and faces prosecution in Pakistan has been used as a disposable instrument by a state that will categorically deny his existence and provide no legal protection. The asymmetry between the state that orders the killing and the individual who carries it out is a moral dimension that the investigation’s operational details bring into sharp relief.

The Evidentiary Gap as Strategic Space

The central analytical contribution of this article is the identification of what the Guardian investigation’s evidentiary gap reveals about the strategic architecture of the shadow war. The gap exists between two established points: intelligence operatives from both countries say India is responsible, and the pattern of killings is consistent with state-sponsored operations. The gap is defined by what remains unproven: no Indian official has publicly confirmed the program, no independent forensic analysis has verified the Pakistani documentary evidence, and no third-party investigation has corroborated the specific operational claims.

This evidentiary gap is not an analytical failure. It is a strategic feature. The shadow war operates within this gap by design. The entire operational architecture described by both Indian and Pakistani sources, the use of UAE-based intermediaries, the recruitment of local criminals and jihadists, the multiple layers of compartmentalization, is designed to maintain deniability. Deniability is not achieved by preventing anyone from suspecting state involvement. The entire intelligence community, as the Guardian investigation demonstrates, already suspects and in many cases believes in state involvement. Deniability is achieved by preventing the accumulation of evidence sufficient for legal proceedings, diplomatic consequences, or international sanctions. The gap between belief and proof is the operational space in which the campaign functions.

This understanding reframes the Guardian investigation. The investigation did not close the evidentiary gap; it documented it. By assembling the testimony of intelligence operatives from both sides, the investigation showed that the gap is not a gap of knowledge but a gap of provability. Everyone involved, Indian intelligence, Pakistani intelligence, Western journalists, appears to know what is happening. No one can prove it to the standard required for formal international accountability. The investigation’s contribution is to make this gap visible and to demonstrate that the campaign operates within it deliberately.

The ISI nexus with terror organizations operates within a similar evidentiary gap. For decades, Pakistan’s intelligence services have maintained relationships with terrorist organizations while denying those relationships publicly. The evidence of ISI support for groups like Lashkar-e-Taiba and Jaish-e-Mohammed is overwhelming in intelligence circles but rarely sufficient for formal international legal proceedings. India’s shadow war, if the Guardian investigation’s broad contours are accurate, mirrors this structure: conduct operations through deniable intermediaries, maintain categorical denial, and rely on the evidentiary gap to prevent accountability. The symmetry is analytically productive because it suggests that both states have learned the same lesson about the relationship between covert action and deniability.

The Guardian’s Track Record and Methodological Assessment

Assessing the investigation requires examining The Guardian’s institutional credibility on intelligence-related reporting. The newspaper occupies a specific position in the landscape of global investigative journalism. It is not a tabloid that publishes unverified sensationalism. It is a broadsheet with a strong editorial tradition and a documented history of publishing intelligence stories that major governments initially denied and subsequently acknowledged as accurate.

The Snowden disclosures are the most relevant precedent. In June 2013, The Guardian, working with journalist Glenn Greenwald, published classified documents provided by former NSA contractor Edward Snowden revealing the scope of mass surveillance programs operated by the NSA and its Five Eyes partners. The British government threatened legal action. The United States government initially denied or minimized the programs. Over time, the disclosures were confirmed, official investigations were launched, and surveillance laws were reformed in multiple countries. The Snowden precedent demonstrates that The Guardian has institutional experience in handling classified intelligence materials and that its editorial processes for publishing such materials are rigorous enough to withstand government pressure and legal scrutiny.

The Pegasus Project provides a second relevant precedent. The Guardian was part of a consortium of international media organizations that investigated the NSO Group’s Pegasus spyware, revealing that governments had used the technology to surveil journalists, activists, and political opponents. The investigation relied on leaked data, forensic analysis, and anonymous sources within intelligence agencies. Its findings were subsequently corroborated by independent forensic laboratories and multiple government investigations.

These precedents support the inference that The Guardian’s institutional methodology for intelligence reporting is sound and that the newspaper’s editorial gatekeeping would require internal corroboration of claims before publication. They do not guarantee that every specific claim in the India-Pakistan investigation is accurate. The Guardian acknowledged this limitation by noting that the Pakistani documents could not be independently verified. This acknowledgment reflects methodological integrity: the newspaper is transparent about the limits of what it can confirm. The analytical weight of the investigation rests not on the certainty of every detail but on the credibility of the general pattern: multiple sources from multiple countries, describing a consistent operational program, providing specific details that are internally coherent and consistent with independently observable evidence, published by a newspaper with an institutional track record of accuracy on intelligence-related stories.

The Investigation’s Analytical Limitations

No assessment of the Guardian investigation is complete without identifying its analytical limitations, and this article’s commitment to adjudicating the evidence honestly requires naming them explicitly.

The first limitation is source dependency. The investigation depends heavily on unnamed intelligence operatives whose identities, roles, and motivations are unknown to the reader. Unnamed sources are standard in intelligence reporting, but they create an irreducible information asymmetry between the journalist, who has assessed the source’s credibility through personal interaction, and the reader, who must trust the journalist’s assessment. The Guardian’s editorial standards mitigate this asymmetry but do not eliminate it.

The second limitation is the absence of Indian rebuttal at a specific level. India’s categorical denial, while consistent with covert-operations protocol, means the investigation’s specific claims remain contested but not specifically refuted. A more detailed Indian response would allow external analysts to evaluate competing narratives. The absence of such a response means the investigation’s narrative occupies the analytical space largely unchallenged on specifics, which may overstate its evidentiary strength.

The third limitation is the Pakistani documentary evidence’s provenance. Documents provided by one party to a conflict, particularly an intelligence service with institutional interests in the investigation’s conclusions, require a degree of skepticism that the investigation’s framing does not always provide. The Guardian reported what the documents showed but could not verify their authenticity independently. This is an honest limitation, but it should not be minimized. In intelligence affairs, the fabrication or manipulation of documentary evidence is not rare; it is a standard practice of intelligence agencies worldwide.

The fourth limitation is temporal coverage. The investigation covered nearly twenty killings since 2020 but provided operational detail on only a handful of cases. The remaining cases are enumerated but not substantiated at the same level of specificity. This creates an uneven evidentiary foundation: the cases with specific detail are stronger candidates for attribution, while the cases without such detail rest on the pattern inference rather than independent evidence.

The fifth limitation is the investigation’s framing. By titling the report with the claim that India ordered killings, The Guardian adopted a framing that the evidence does not fully support at the standard of proof implied by that framing. Intelligence operatives say is different from India ordered. The former is a report of claims; the latter is a statement of established fact. The investigation’s evidence supports the former but its framing implies the latter. This gap between evidence and framing is common in investigative journalism, particularly on national security topics, but it should be identified as a methodological choice rather than an analytical conclusion.

Competing Interpretations of the Investigation’s Significance

The Guardian investigation generated three competing interpretations, each with different analytical implications for the shadow war’s future. Each interpretation reflects not just a reading of the evidence but a set of assumptions about the nature of state power, media credibility, and the relationship between covert action and public accountability.

Interpretation One: The Investigation as Vindication

Pakistan and a significant portion of the international commentariat treated the investigation as vindication of Pakistan’s allegations against India. This interpretation holds that The Guardian, a credible Western media organization with no institutional bias toward Pakistan, independently confirmed what Islamabad had been claiming. The unnamed Indian sources were the key: Indian intelligence operatives admitting the program existed, describing its authorization chain, and explaining its strategic logic. For proponents of this interpretation, the investigation settles the attribution question. India did it. The only remaining question is what the international community will do about it.

Pakistan’s diplomatic apparatus moved quickly to leverage the investigation. The Foreign Office issued statements characterizing India’s network of extraterritorial killings as a global phenomenon requiring coordinated international response. Pakistani commentators drew explicit parallels between the Guardian investigation and the Canadian and American allegations, arguing that the three cases together constituted an undeniable pattern of Indian state-sponsored violence across three continents. The framing was strategic: by linking the Pakistan killings to the Nijjar and Pannun cases, Pakistan positioned itself alongside two Western democracies as a victim of Indian aggression, a diplomatic alignment that serves Islamabad’s interests regardless of the significant differences between the Pakistani cases (designated terrorists on Pakistani soil) and the Western cases (political activists on allied democratic soil).

The analytical weakness of this interpretation is that it treats unnamed sources as equivalent to confirmed evidence. Intelligence operatives speaking anonymously to a journalist are not the same as officials testifying under oath, documents verified by independent forensics, or admissions in diplomatic communications. The investigation provides strong circumstantial evidence and compelling testimony but does not establish the kind of proof that would support, for example, a case before the International Court of Justice. The vindication interpretation also elides the uncomfortable question that the investigation implicitly raises about Pakistan: every named target was an individual whose presence in Pakistan contradicted Pakistan’s official claims about not sheltering designated terrorists.

Interpretation Two: The Investigation as Hit Job

The Indian government and government-aligned media treated the investigation as a politically motivated attack timed to damage Modi before the 2024 election. This interpretation holds that The Guardian, which has published critical coverage of Modi’s government on multiple occasions, assembled a narrative from compromised Pakistani sources and disgruntled or fabricated Indian sources to produce maximum political damage at the most sensitive electoral moment. The release of passport details of alleged Indian agents, from this perspective, was a Pakistani intelligence operation that The Guardian uncritically amplified.

This interpretation gained significant traction in Indian domestic discourse for reasons that combine legitimate concern with motivated reasoning. The legitimate concern is that intelligence-related investigations published during election campaigns can distort the democratic process, particularly when the allegations concern national security operations that a significant portion of the electorate might support rather than condemn. The motivated reasoning is that dismissing the investigation as a hit job avoids engaging with its specific claims. If every critical investigation of a government can be dismissed as politically motivated, investigative journalism becomes impossible by definition. The hit-job interpretation functions as an epistemological shield: it does not refute the evidence but prevents the evidence from being evaluated.

The analytical weakness of this interpretation is that it does not engage with the investigation’s specific claims. Dismissing an investigation as a hit job requires demonstrating that the evidence was fabricated or the sources were unreliable, not simply asserting political motivation. The timing of the publication, while notable, does not determine the accuracy of the content. Major intelligence stories are frequently published during politically sensitive periods, and the editorial decision to publish is distinct from the investigative validity of the reporting.

Interpretation Three: The Investigation as Documentation of the Known

This article’s position aligns with a third interpretation: the investigation documented what was already widely believed but had not been systematically reported in Western media. The targeted killings in Pakistan had been tracked by analysts, reported by Pakistani media, and privately acknowledged by intelligence officials from multiple countries long before The Guardian published its investigation. The investigation’s contribution was not to reveal a secret but to formalize knowledge, to transform intelligence-community consensus into public-record journalism. This interpretation holds that the investigation is neither definitive proof nor political fabrication but a significant act of documentation that changed the informational environment without changing the operational reality.

The strength of this interpretation lies in its explanatory power regarding the post-investigation trajectory. If the investigation had constituted a decisive exposure of a previously hidden program, two consequences would be expected: international pressure for accountability and operational caution by the alleged perpetrators. Neither materialized. The campaign continued and accelerated, and the international community remained largely silent. This outcome is explicable only if the investigation documented what was already known and already factored into the strategic calculations of all parties involved. The investigation changed nothing operationally because there was nothing new to change: the program existed, everyone knew it existed, and the investigation simply placed on the public record what had previously existed in the classified and analytical record.

The evidence for this interpretation is the shadow war’s trajectory after the investigation. If the investigation had constituted a decisive exposure, the campaign would have been expected to slow down, either because international pressure forced a halt or because operational security concerns required a cooling-off period. Neither happened. The campaign continued and, by multiple accounts, accelerated significantly in the months following the investigation. This post-investigation acceleration suggests that the exposure had no deterrent effect, which is consistent with the interpretation that the investigation documented what was already known and that the parties involved had already factored public exposure into their strategic calculations.

The International Response, or Lack Thereof

The Guardian investigation’s most analytically revealing consequence was not the reaction it provoked but the reaction it failed to provoke. The investigation was published in a major British newspaper with global reach. It alleged that a democratic government had been systematically assassinating people in a neighboring country. In comparable cases involving other democracies, such allegations have triggered diplomatic crises, formal investigations, and sustained international pressure. When Canada accused India of involvement in the killing of Hardeep Singh Nijjar in British Columbia, Canadian Prime Minister Justin Trudeau addressed Parliament and expelled Indian diplomats. When the United States indicted an alleged Indian intelligence operative for the attempted assassination of Gurpatwant Singh Pannun on American soil, the case generated sustained media coverage and diplomatic friction.

The Guardian investigation, by contrast, produced a remarkably muted international response. No government issued a formal statement condemning the alleged killings in Pakistan. No international organization announced an investigation. No diplomatic consequences materialized. The selective silence was itself an analytical data point: the international community treated the alleged killing of designated terrorists in Pakistan differently from the alleged killing of political activists in Canada and the United States. This differential treatment raises the question of whether the targets’ status as designated terrorists creates a different moral and diplomatic category, a question the analysis of the international community’s selective response to India’s shadow war addresses directly.

The silence also suggests that the investigation’s claims, while publicly novel, were not privately surprising to Western intelligence agencies. If the CIA, MI6, and other agencies had not previously suspected Indian involvement in the Pakistan killings, the Guardian investigation would have prompted intelligence assessments, diplomatic inquiries, and interagency consultations that would have generated some observable response. The absence of such a response is consistent with the hypothesis that Western intelligence agencies already knew, or strongly suspected, what The Guardian reported and had decided that the killings of designated terrorists on Pakistani soil did not warrant the diplomatic cost of confrontation with India. The calculus appears to have been that India’s strategic value as a partner in containing Chinese influence, as a market for defense equipment, and as a pillar of the Indo-Pacific strategy outweighed the reputational cost of ignoring credible allegations of extraterritorial killing by a democratic ally. This strategic arithmetic, unspoken but unmistakable, is itself a finding of the post-investigation analysis: the international community’s response to the shadow war is determined not by legal principle but by strategic calculation.

What The Investigation Actually Proves

Stripped to its analytical core, the Guardian investigation establishes the following with varying degrees of confidence.

It establishes with high confidence that a systematic pattern of targeted killings of India’s most-wanted individuals has occurred on Pakistani soil since approximately 2020. This was already documented by open-source reporting, but the investigation adds the confirmation of intelligence operatives from both countries.

It establishes with moderate confidence that Indian intelligence operatives believe RAW is responsible for these killings and that the program was triggered by the Pulwama attack and authorized at the highest levels of government. The confidence level is moderate rather than high because these claims rest on unnamed sources whose motivations and accuracy cannot be independently assessed.

It establishes with moderate confidence that Pakistani investigators have assembled documentary evidence linking specific killings to named Indian agents operating through UAE-based intermediaries. The confidence level is moderated by the fact that the documents could not be independently verified and were provided by an interested party.

It establishes with lower confidence the specific operational details, such as the recruitment of jihadists, the payment amounts, the Amazon warehouse worker turned assassin. These details are internally coherent but rest on the thinnest evidentiary foundation: unverifiable documents from a single source with institutional interests in the narrative.

It does not establish with any confidence that the Indian government formally authorized a killing program. The unnamed Indian source’s claim of highest-level approval is a statement of belief or knowledge by a single source, not documented evidence of a formal authorization. The distinction between an intelligence operative saying the program was authorized and documentary evidence of authorization is not pedantic; it is the difference between a credible claim and a proven fact.

This graduated assessment is not a retreat from analytical conviction. It is an exercise in precision. The investigation’s value lies not in proving every claim but in assembling sufficient evidence to move the attribution question from speculation to informed assessment. Before the investigation, the claim that India was systematically killing terrorists in Pakistan was a hypothesis supported by pattern evidence. After the investigation, it is an assessment supported by pattern evidence, insider testimony from both sides, documentary claims, and the absence of any specific rebuttal from the accused party. The gap between hypothesis and assessment is the investigation’s genuine contribution.

The Investigation’s Place in the Shadow War’s Arc

The Guardian investigation marks a structural transition in the shadow war’s public narrative. Before April 2024, the campaign existed in an informational space defined by Pakistani media reports, Indian social media celebration, and analytical commentary. After April 2024, the campaign exists in an informational space that includes a major Western investigative report, formal Pakistani government accusations, and the attention of international media and diplomatic establishments. This transition has not altered the campaign’s operational trajectory, the killings have continued and accelerated, but it has altered the strategic context in which the campaign operates.

The investigation created a benchmark moment that divides the shadow war’s history into two phases. In the pre-Guardian phase, the campaign’s existence was an open secret. Pakistani media reported each killing individually. Indian social media users celebrated the deaths of designated terrorists. Security analysts tracked the pattern and published assessments attributing the killings to Indian intelligence operations. But no authoritative Western media institution had assembled these individual data points into a coherent investigative narrative. The campaign existed in a space of plausible deniability that was locally transparent but internationally obscure.

In the post-Guardian phase, the campaign operates in full international daylight. Every subsequent killing is contextualized by the investigation’s claims. When an unknown gunman shoots a Lashkar-e-Taiba operative in Sindh or a Jaish-e-Mohammed commander in Punjab, the reporting automatically references the broader campaign, India’s alleged responsibility, and the Guardian investigation that formalized these allegations. The operational reality may not have changed, but the informational ecosystem within which the operations are understood has been fundamentally restructured. This restructuring has consequences for how the campaign is perceived by potential allies, potential critics, and potential targets.

For potential targets, the post-Guardian environment provides an explicit warning that the campaign is systematic rather than opportunistic. Before the investigation, an individual on India’s most-wanted list living in Pakistan might have calculated that the killings were isolated events, random violence, or internal Pakistani score-settling. After the investigation, the same individual knows, or should know, that intelligence operatives from both countries have described a targeted program with institutional backing and government authorization. This knowledge may change behavior: targets may relocate, increase security, alter routines, or seek protection from Pakistani intelligence services. Whether these behavioral changes have occurred and whether they have affected the campaign’s operational effectiveness is an empirical question that subsequent articles analyzing the post-2024 acceleration address.

For Pakistan’s security establishment, the post-Guardian environment creates a paradoxical pressure. The investigation vindicated Pakistan’s allegations against India but simultaneously exposed Pakistan’s failure to protect individuals on its soil, many of whom it had denied sheltering. Every subsequent killing after the Guardian investigation is a demonstration that Pakistan’s security agencies cannot prevent Indian operations within Pakistani borders, even after those operations have been publicly documented and the operational model has been exposed. The investigation that was supposed to build international pressure on India also built domestic pressure on Pakistan’s security establishment, a pressure that the continued killings intensify with each successive incident.

The investigation also created a new analytical obligation. Every subsequent article in this series, every profile of an eliminated target, every analysis of an operational pattern, every assessment of the campaign’s strategic implications, must now reckon with the investigation’s claims. The question is no longer simply who is killing these people but who is killing these people, given that intelligence operatives from both countries have told a major Western newspaper that India’s intelligence service is responsible, and given that India has denied this categorically, and given that the evidence falls in the gap between credible allegation and proven fact. That gap is where the shadow war lives. The Guardian investigation did not close it. It illuminated it.

The campaign that followed the investigation’s publication is, by any measure, the most aggressive phase of the shadow war to date. Over thirty militants with ties to organizations like LeT and Hizbul Mujahideen have been killed by unknown gunmen across Pakistan’s cities, from Karachi and Lahore to Rawalpindi and Peshawar, representing an unprecedented acceleration in operational tempo that defies the expectation that exposure would produce restraint. Whether this acceleration was caused by the investigation, whether the exposure emboldened the campaign by demonstrating that international consequences would not materialize, or whether the acceleration had been planned independently of the investigation, remains a question that demands rigorous temporal analysis rather than casual correlation. But the temporal fact is striking: the moment the shadow war entered the Western public record was also the moment it shifted into its highest gear. If the purpose of exposure is deterrence, the Guardian investigation is a case study in deterrence failure. If the purpose of exposure is documentation, the investigation succeeded. The shadow war is now documented. It continues anyway. And the continuity itself is the strongest single piece of evidence that the campaign’s architects had already calculated the cost of exposure and judged it acceptable, a judgment that every subsequent killing validates.

Frequently Asked Questions

Q: What did The Guardian’s investigation claim about India’s killings in Pakistan?

The Guardian investigation, published on April 5, 2024, claimed that the Indian government had ordered assassinations of individuals in Pakistan as part of a wider strategy to eliminate terrorists on foreign soil. The report cited unnamed Indian and Pakistani intelligence operatives and documents shared by Pakistani investigators. It described an operational architecture involving RAW sleeper cells based in the UAE, the recruitment of local criminals and jihadists to carry out the actual killings, and payments of millions of rupees through Dubai-based financial channels. The most significant claim was from an unnamed Indian intelligence officer who stated that such operations required approval from the highest level of government and that the program was triggered by the Pulwama attack of February 2019.

Q: How did India respond to the Guardian investigation?

India’s Ministry of External Affairs dismissed the entire investigation as false and malicious anti-India propaganda. The ministry reiterated a previous statement by External Affairs Minister S. Jaishankar that targeted killings in other countries were not the government of India’s policy. India did not address the investigation’s specific claims, including the named Indian agents, the financial evidence, or the apprehended shooters in Pakistan. The categorical denial, without specific rebuttal, is consistent with standard covert-operations protocol but leaves the investigation’s detailed allegations standing without point-by-point refutation.

Q: Who were the sources for the Guardian investigation?

The investigation cited at least two unnamed Indian intelligence officers, unnamed senior officials from two separate Pakistani intelligence agencies, and documents shared by Pakistani investigators, including witness testimonies, arrest records, financial statements, WhatsApp messages, and passport details. The Guardian acknowledged that the Pakistani documents could not be independently verified. The use of multiple source categories from both countries is a methodological strength, but the anonymity of all sources creates an irreducible information asymmetry between the journalist and the reader.

Q: What specific evidence did Pakistan present linking India to targeted killings?

Pakistan’s Foreign Secretary Muhammad Syrus Sajjad Qazi, at a January 25, 2024, press conference, presented evidence linking Indian agents to two specific killings: Shahid Latif in Sialkot (October 2023) and Muhammad Riaz in Rawalakot (September 2023). Pakistan named two Indian agents, Yogesh Kumar and Ashok Kumar Anand, released their passport details, described the recruitment of local shooters through Telegram, and cited documentary, financial, and forensic evidence. The apprehended shooters, Muhammad Umair and Muhammad Abdullah Ali, provided confessional statements described by Pakistan as corroborating Indian involvement.

Q: Was the Guardian investigation timed to influence India’s 2024 election?

The investigation was published on April 5, 2024, weeks before the first phase of India’s general election. Critics, including the Indian government, alleged that the timing was designed to damage Prime Minister Modi’s reelection campaign. The Guardian did not address the timing question directly. From an analytical perspective, the timing is notable but does not determine the investigation’s accuracy. Major intelligence stories are frequently published during politically sensitive periods, and editorial timing decisions are distinct from the investigative validity of the reporting. The investigation’s sourcing and evidence should be evaluated independently of when it was published.

Q: What role did the UAE play in the alleged assassination operations?

Pakistani investigators described the UAE, particularly Dubai, as the primary logistical hub for the alleged killing program. RAW sleeper cells allegedly operated from the UAE, organizing operations, recruiting intermediaries, and channeling payments. The UAE’s large Pakistani and Indian diaspora populations, its position as a financial hub, and the ease of travel between the UAE and Pakistan make it a plausible operational base. RAW handlers reportedly met in multiple locations including the UAE, Nepal, the Maldives, and Mauritius. Payments for the killings were allegedly facilitated through Dubai-based financial transactions.

Q: How reliable is The Guardian as a source on intelligence matters?

The Guardian has a strong institutional track record on intelligence reporting. The newspaper published the Edward Snowden disclosures on NSA mass surveillance (2013), participated in the Pegasus Project consortium investigating NSO Group spyware, and has broken multiple intelligence-related stories that governments initially denied but subsequently acknowledged. This track record supports the inference that the newspaper’s editorial standards for intelligence reporting are rigorous. The institutional credibility does not guarantee the accuracy of every specific claim in any given investigation, but it establishes that the newspaper’s gatekeeping processes are above average for intelligence-related journalism.

Q: Did the Guardian investigation stop or slow down the targeted killings?

No. The campaign continued and by multiple accounts accelerated significantly after the investigation’s publication. Over thirty militants with ties to organizations like Lashkar-e-Taiba and Hizbul Mujahideen have been killed in the months following the report. The post-investigation acceleration suggests that public exposure had no deterrent effect and may have demonstrated to the campaign’s operators that international consequences for the killings would not materialize, effectively reinforcing rather than constraining the operational tempo.

Q: What is the Pulwama attack and why is it relevant to the Guardian investigation?

The Pulwama attack occurred on February 14, 2019, when a Jaish-e-Mohammed suicide bomber rammed a vehicle laden with explosives into a Central Reserve Police Force convoy on the Jammu-Srinagar National Highway, killing forty paramilitary personnel. The Guardian investigation’s Indian sources cited Pulwama as the catalytic event that shifted RAW’s approach to extraterritorial operations: after Pulwama, the approach changed to target elements outside the country before they could launch attacks. The temporal pattern of the killings, which began approximately two years after Pulwama, is consistent with this claimed trigger, allowing for a build-up period during which the operational infrastructure was established.

Q: Why was Pakistan reluctant to publicly acknowledge the targeted killings?

Pakistan’s reluctance stemmed from a strategic dilemma. Acknowledging the targeted killings required acknowledging that the individuals killed, most of whom were designated terrorists wanted by India, were living in Pakistan. This acknowledgment contradicted years of Pakistani denials about sheltering designated terrorists. For months after the killings began, Pakistan said nothing publicly. The silence broke only when the pattern became too visible to ignore and when the Guardian investigation made continued silence unsustainable. When Pakistan did speak, it framed the killings as sovereignty violations by India rather than addressing the targets’ terrorist designations.

Q: How does the Guardian investigation relate to the Canada and US allegations against India?

The Guardian investigation appeared in a context where Canada and the United States had both accused India of involvement in extraterritorial killings. Canada accused India of involvement in the killing of Hardeep Singh Nijjar in British Columbia in June 2023. The United States indicted an alleged Indian intelligence operative for the attempted assassination of Gurpatwant Singh Pannun on American soil. The Guardian investigation added a third theater, Pakistan, to the picture of alleged Indian extraterritorial operations, and Pakistani Foreign Secretary Qazi explicitly noted that the method of assassination was similar across all three jurisdictions. Together, the allegations from Canada, the United States, and the Guardian investigation suggest a broader pattern of Indian extraterritorial operations, though each case rests on different evidence and different levels of specificity.

Q: What did the Guardian investigation say about the Mossad and KGB connections?

An unnamed Indian intelligence officer told The Guardian that RAW drew inspiration from intelligence agencies such as Israel’s Mossad and Russia’s KGB, which have been linked to extrajudicial killings on foreign soil. The officer also stated that the killing of Saudi journalist Jamal Khashoggi had been directly cited by RAW officials. These references position India’s alleged program within a specific tradition of state-sponsored targeted killing. The Mossad reference is the most analytically significant because Israel’s targeted killing program is the most extensively documented and debated precedent for a democracy conducting extraterritorial assassinations as a counter-terrorism tool.

Q: What evidence could not be independently verified in the Guardian investigation?

The Guardian explicitly acknowledged that documents shared by Pakistani investigators could not be independently verified. These documents included witness testimonies, arrest records, financial statements, WhatsApp messages, and passport details. The newspaper stated that it had seen the documents but that independent verification was not possible. This is a standard limitation in intelligence reporting, but it means the documentary foundation of the investigation rests on the credibility of the Pakistani investigators who provided the materials, whose institutional interests in establishing Indian culpability should be factored into the assessment.

Q: How did the Guardian investigation affect India-Pakistan relations?

The investigation intensified an already deteriorating relationship. Pakistan used the investigation as additional evidence for its formal allegations against India. India dismissed it as propaganda. The investigation did not produce any new diplomatic channel, mediation effort, or conflict-resolution mechanism. It reinforced existing positions: Pakistan’s narrative that India was conducting extraterritorial assassinations and India’s narrative that Pakistan was peddling anti-India propaganda to deflect from its own role in sheltering terrorists. The relationship between the two countries reached its lowest point during this period, compounded by the subsequent Pahalgam attack and Operation Sindoor.

Q: Why did the international community largely ignore the Guardian investigation?

The muted international response suggests several factors. Western intelligence agencies likely already suspected Indian involvement and had decided that the killings of designated terrorists in Pakistan did not warrant diplomatic confrontation with India, a major strategic partner. The targets’ status as designated terrorists created a different moral and diplomatic category from the Nijjar and Pannun cases, where the targets were political activists on allied soil. India’s economic and strategic importance to Western governments likely shielded it from consequences that might have followed similar allegations against a less strategically important state.

Q: Is the Guardian investigation sufficient to prove India ordered the targeted killings?

No. The investigation provides strong circumstantial evidence and compelling testimony from intelligence operatives on both sides, but it does not establish proof to a legal or definitive standard. The unnamed sources cannot be cross-examined, the documentary evidence has not been independently verified, and India has not been afforded or compelled to provide a specific rebuttal. The investigation is best understood as a significant escalation from hypothesis to informed assessment: it moves the attribution question from speculation supported by pattern evidence to credible allegation supported by pattern evidence, insider testimony, and documentary claims.

Q: What did the investigation reveal about the recruitment methods for the killings?

Pakistani investigators described a layered recruitment model. RAW sleeper cells in the UAE allegedly recruited intermediaries who then recruited local Pakistanis, including criminals and impoverished individuals, to carry out the actual killings. In one case, a twenty-year-old illiterate Pakistani working in an Amazon packing warehouse in the UAE was allegedly recruited, paid 1.5 million Pakistani rupees to track a target, and promised fifteen million rupees and his own catering company upon completing the killing. Indian agents also allegedly recruited jihadists by telling them they were killing infidels, exploiting religious ideology to create deniable triggermen who would leave an investigative trail pointing toward jihadi violence rather than state-sponsored assassination.

Q: How does the Guardian investigation change the analytical framework for understanding the shadow war?

The investigation changes the analytical framework by moving the attribution question from inference to testimony. Before the investigation, attributing the targeted killings to India required assembling pattern evidence: the consistent modus operandi, the exclusive targeting of India-wanted individuals, the absence of any Pakistani group claiming credit. After the investigation, the attribution assessment can also draw on the testimony of intelligence operatives from both countries, who stated directly that India’s intelligence service was responsible. The investigation does not close the evidentiary gap between belief and proof, but it narrows it significantly and makes the gap itself visible as a deliberate strategic feature of the campaign rather than an accidental limitation of the evidence.

Q: What is the significance of Pakistan releasing passport details of alleged Indian agents?

Pakistan’s release of passport details for Yogesh Kumar and Ashok Kumar Anand represents an unusual step in diplomatic accusation. By providing specific biographical information, including birthdates and birthplaces, Pakistan invited verification of the named individuals’ identities and implicitly challenged India to deny their existence. If the named agents are genuine RAW operatives, the passport release constitutes a significant intelligence exposure. If the names are cover identities or fabrications, the release is a diplomatic maneuver that adds false specificity to the allegations. India’s refusal to address the named individuals leaves the question unresolved and the analytical space open.

Q: What lessons does the Guardian investigation offer for understanding state-sponsored covert operations?

The investigation demonstrates several enduring principles of covert operations. First, deniability is maintained not by preventing suspicion but by preventing provable attribution. Second, the evidentiary gap between belief and proof is a deliberate operational feature, not an analytical limitation. Third, public exposure of covert operations does not necessarily produce deterrence; it may instead demonstrate that the international community lacks the will to impose consequences. Fourth, intelligence agencies on opposing sides of a conflict may both have incentives to share information with journalists, for different strategic reasons, creating a journalistic environment where multiple sources converge on a narrative without that convergence constituting independent corroboration.

Q: How does the Guardian investigation fit within the broader history of Western media investigations into Indian intelligence operations?

The Guardian investigation is the most comprehensive Western media investigation into alleged Indian intelligence operations on Pakistani soil, but it is not the first Western media engagement with India’s extraterritorial activities. The Washington Post reported on the US allegations regarding the Pannun assassination plot. Canadian media extensively covered the Nijjar killing and its alleged links to Indian intelligence. The Financial Times and BBC have published investigative pieces on aspects of India’s intelligence activities abroad. The Guardian investigation is distinct because it combines sources from both India and Pakistan, covers the largest number of alleged killings, and presents the most detailed operational reconstruction of how the killings were allegedly organized. Its place in the broader investigative landscape is as the most ambitious and comprehensive single investigation, even as its limitations regarding unverifiable documents and unnamed sources apply equally to other investigations in this space.

Q: Could the apprehended shooters’ confessions in Pakistan be considered reliable evidence?

The reliability of the confessional statements obtained from apprehended shooters Muhammad Umair and Muhammad Abdullah Ali depends on the conditions under which the confessions were obtained. Pakistani law enforcement agencies have been criticized by international human rights organizations for coercive interrogation practices, and confessions obtained under duress do not meet the evidentiary standard required by most legal systems. The investigation does not address the conditions under which the shooters were interrogated. If the confessions were freely given and corroborated by independent physical evidence (financial records, communication logs, travel documentation), they constitute significant evidence. If they were obtained through coercion and corroborated only by documents provided by the same investigators who obtained the confessions, the evidentiary chain becomes circular. The investigation’s value would be substantially strengthened if the apprehended individuals were made available for independent interview by an international body or if the documentary evidence were subjected to third-party forensic verification.

Q: What does the Guardian investigation reveal about the future trajectory of the shadow war?

The investigation’s most consequential revelation about the shadow war’s future is contained not in what it reported but in what happened afterward. The campaign’s continuation and acceleration following the investigation demonstrate that public exposure does not function as a constraint on the program’s operations. This finding has directional implications: if the shadow war has survived the most significant investigative exposure it has faced and emerged more active rather than less, the analytical expectation should be that the campaign will continue until one of three conditions is met: the target list is exhausted, Pakistan develops the security capacity to prevent the operations, or a geopolitical shift makes the diplomatic cost of continuation unacceptable. None of these conditions appears imminent. The investigation thus functions as documentation of a program that has demonstrated resilience against the primary accountability mechanism that democratic societies employ against covert violence: journalistic exposure. The shadow war’s architects appear to have solved the exposure problem not by preventing exposure but by rendering it strategically inconsequential.