On a Sunday evening in June 2023, two masked men waited in the parking lot of the Guru Nanak Sikh Gurdwara in Surrey, British Columbia. Hardeep Singh Nijjar, the temple’s president and a man New Delhi had branded a terrorist three years earlier, was shot in his pickup truck as he left after the day’s congregation. Roughly eight thousand kilometres away, three days before, Avtar Singh Khanda had died in a Birmingham hospital. Khanda was identified by Indian agencies as the chief of the Khalistan Liberation Force, operating under an assumed name, and he had pulled the Indian flag down from the High Commission in London only weeks earlier. Two bodies, two of the world’s oldest democracies, one separatist project, and a single uncomfortable question that neither Ottawa nor Westminster had ever wanted to answer directly: how had Canada and the United Kingdom become the safest places on earth to be a Khalistan militant?

That question sits at the centre of one of the strangest features of India’s long campaign against the men it accuses of plotting terror. Most of the targeted killings examined across this series occurred inside Pakistan, in cities such as Lahore, Karachi, and Rawalpindi, where a wanted operative could be reached because the state sheltering him was an adversary and the operating environment was permissive. The Western cases break that pattern. They occur in countries that are not adversaries of India but partners, countries with advanced law enforcement, functioning courts, and the legal architecture of liberal democracy. To call Canada and Britain safe havens at all therefore demands a careful definition, because the shelter they provide looks nothing like the shelter Pakistan provides. Pakistan’s terror sanctuary network is built on state policy, military protection, and an intelligence apparatus that treats militants as strategic assets. Canada and the United Kingdom shelter the same movement through an entirely different mechanism. They shelter it through the law itself, through free expression protections, asylum frameworks, and electoral politics that make a hard line against Sikh separatism politically costly. One model is the ungoverned space. The other is the over-governed space, where governance becomes the cover. Understanding the difference is the whole point of this analysis, and getting it wrong in either direction produces a distorted picture of what the West has actually done.

This is not a marginal debate. The Khalistan question has become the single most damaging issue in India’s relationship with two of its most important Western partners, and the way it is analyzed carries consequences for hundreds of thousands of ordinary Sikh citizens in both countries who have nothing to do with separatism and resent being spoken of as if they did. A careful account therefore has to accomplish several things at once. It has to take seriously the documented history of lethal violence, above all the Air India bombing, the deadliest terrorist attack ever planned on Canadian soil. It has to take equally seriously the principle that advocating for a political idea, however much a foreign government dislikes it, is not terrorism in a democracy. The analysis has to distinguish the armed groups with a record of killing from the advocacy bodies whose record is one of litigation and protest. And it has to resist the gravitational pull of both propaganda systems, the separatist one that recasts every militant as a martyr and the official Indian one that recasts every activist as a terrorist. This series has examined the Pakistani theatre of the campaign at length, and the four-decade arc of the movement shows how the violence began and where it travelled. The Western chapter is harder to write than the Pakistani one precisely because the moral and legal categories are genuinely contested rather than settled, and a reader who finishes this analysis certain of a simple verdict has probably been reading a different and less honest piece.

Geography and Strategic Position

The geography that matters for the Western chapter of this story is not physical terrain. It is demographic. Sikh migration to Canada and Britain stretches back more than a century, long predating the separatist project, and the communities that grew from it are now large, prosperous, and politically organized. Canada’s 2021 census counted roughly 770,000 Sikhs, about two percent of the national population, the largest such community of any country outside India. The United Kingdom’s 2021 census recorded a little over half a million people identifying as Sikh. Those numbers alone make both nations central to any diaspora-driven movement, because a movement that has lost most of its support inside Punjab must draw its money, its membership, and its political oxygen from abroad.

Where those populations cluster determines where the movement concentrates. In Canada, the densest Sikh settlement is the Lower Mainland of British Columbia, and within it the city of Surrey, where entire neighbourhoods are majority South Asian and where the gurdwaras function as community anchors. Brampton, in Ontario’s Peel Region, is the second great hub, and Winnipeg, Calgary, and Edmonton each host substantial communities of their own. In Britain the pattern is older and just as concentrated. Southall in west London has been a Punjabi heartland since the 1950s. Birmingham and the wider West Midlands, including Smethwick, Handsworth, and Wolverhampton, hold one of the largest Sikh populations in Europe. Leicester, Slough, and parts of east London round out the map. These are not marginal places. They are economically successful, electorally significant, and woven into the civic life of both countries.

That civic weight is the strategic position. A militant hiding in a Karachi suburb has the protection of distance and a hostile state. A separatist organizer in Surrey or Southall has something Pakistan can never offer: a vote that politicians want, a community whose grievances make headlines, and a set of constitutional protections that apply to citizen and asylum seeker alike. Canada’s electoral system magnifies the effect. In a parliamentary democracy where many seats are decided by a few thousand ballots, a concentrated and mobilized community becomes a constituency that no party can afford to alienate. The federal New Democratic Party was led for years by Jagmeet Singh, a practising Sikh, and successive Liberal and Conservative governments have counted Sikh-Canadian voters as a bloc worth courting. India’s complaint, voiced for two decades through its diplomats, has been that this courtship slides into tolerance, that Canadian politicians attend events where separatist iconography is displayed and decline to condemn what New Delhi regards as glorification of terrorism.

Britain’s version of the same dynamic runs through its constituencies and its long postcolonial relationship with the Punjab. Sikh organizations in the UK are sophisticated lobbyists, and the gurdwara management committees that control temple finances and platforms are themselves contested political spaces. The strategic value of both countries to the movement, then, is not that they are lawless. It is precisely that they are lawful, and that the law in a democracy protects speech, assembly, association, and asylum without first asking whether a foreign government approves. A separatist who would be jailed in India for sedition can in Surrey or Southall hold a rally, run a referendum stall, and address a crowd, all of it legal. The geography of the Western safe haven is the geography of rights.

The history behind that geography matters, because it explains why the diaspora communities are both old enough to be deeply rooted and large enough to carry political weight. Sikh migration to Canada began in the first years of the twentieth century, when men from Punjab travelled to British Columbia to work in lumber mills and on the railways. The 1914 voyage of the Komagata Maru, a ship carrying hundreds of would-be Punjabi immigrants who were turned away from Vancouver harbour, became a foundational episode of grievance and identity for the community. Migration to Britain followed a parallel arc, accelerating after the Second World War when labour shortages drew Punjabi workers to the foundries of the West Midlands and the factories around west London. Southall grew in part because of its proximity to Heathrow and to industrial employers willing to hire migrant labour. By the time the Punjab insurgency erupted in the 1980s, both nations already hosted established, second-generation communities into which a new and more politically charged wave of arrivals would settle.

That later wave is the demographic origin of the diaspora militancy. The violence of the 1980s and early 1990s, including the events around the Golden Temple and the long counter-insurgency that followed, produced a stream of asylum seekers, and a significant share of them settled in Canada and Britain. Some were fleeing genuine persecution. Some carried the politics of the insurgency with them. The separatist networks that India now describes as a Western safe haven were, in large part, seeded in those years, by people for whom the cause was not an abstraction but a recent and personal history. The asylum systems of both countries, designed to protect people at risk, inevitably also admitted some who would continue the political project that had made them refugees in the first place. That is not a flaw unique to the Khalistan story. It is a structural feature of asylum in every conflict, and it is one reason the safe haven question resists simple answers.

The electoral arithmetic deserves a closer look, because it is the mechanism through which demography becomes political influence. In Canada, ridings such as Surrey-Newton and several seats across Brampton contain Sikh-majority or Sikh-plurality populations, and a community that votes as an organized bloc can decide the outcome in a tight three-way contest. Successive Canadian Parliaments have seated more than a dozen Sikh members, and at the height of the community’s federal profile the country had more Sikh ministers in cabinet than the Indian union government. Jagmeet Singh led the federal New Democratic Party from 2017 until 2025, and during the years when the NDP held the balance of power in a minority Parliament, his party’s support was essential to the survival of the Liberal government. India’s diplomats have long argued that this configuration made Canadian politicians structurally reluctant to confront separatist activity. Canadian politicians counter that they were representing constituents, which is what elected representatives are supposed to do. Both statements describe the same fact from opposite ends.

Britain’s version is quieter but real. The United Kingdom Parliament has included Sikh members, an all-party parliamentary group has long advocated on community issues, and gurdwara federations function as organized interlocutors with government. The British state’s posture has historically been shaped by a competing pressure as well, because the United Kingdom maintains a close strategic and trade relationship with India and has at times been visibly anxious not to damage it. That cross-pressure, pulling between a domestic community and a foreign partner, is the reason British policy on the movement has often looked hesitant and inconsistent, lurching between gestures toward New Delhi and reassurances to Sikh constituents.

It is worth stating plainly what the geography is not. It is not a network of training camps. There are no separatist firing ranges in Surrey or Southall, no equivalents of the facilities that the Pakistani safe haven network maintains on physical ground, no zones from which the host state’s writ is absent. The Western geography is entirely a geography of institutions and rights, of temples and committees and constituencies and courts. A separatist in Pakistan is protected by territory and a sponsoring state. A separatist in the West is protected by a constitution and a vote. That distinction recurs throughout this analysis, because almost every confusion in the public debate about Western safe havens comes from importing the imagery of the Pakistani model into a setting where it simply does not apply.

There is a strategic dimension that cuts the other way, and an honest account must include it. The same democratic machinery that protects advocacy also constrains the movement. Canada and Britain both possess capable security services, independent courts, and the legal tools to prosecute genuine criminality. The shelter is conditional. It holds for protected political activity and collapses the moment a prosecutor can prove a crime. That conditional quality is exactly what separates the Western model from Pakistan’s, where protection has historically been near-absolute because it flowed from the state’s own strategic calculation rather than from a neutral legal principle. The diaspora geography gives the movement reach. The democratic geography gives it limits. Both halves of that sentence are true, and the rest of this analysis is an attempt to hold them together.

One further feature of the Western geography deserves emphasis before the organizational map, because it shapes everything that follows. The diaspora communities of Canada and Britain are not closed enclaves. They are deeply integrated into the economic, professional, and political mainstream of both countries, and that integration is itself a kind of constraint on the movement. A separatist faction operating inside a marginalized, isolated community can radicalize it with relative ease. A separatist faction operating inside a prosperous, integrated community competes for attention with every other demand on people’s time and loyalty, with careers, mortgages, children’s schooling, and the ordinary business of belonging to a successful diaspora. This is one reason the movement’s real-world support has remained a minority position even as its online presence has grown. The geography that gives the movement a platform also surrounds it with a community that has, for the most part, chosen a different life. Any analysis that pictures Surrey or Southall as a sea of separatist sentiment has mistaken the loudest voices for the whole population, and that mistake, repeated often enough, becomes a slander against communities that are overwhelmingly law-abiding and civically engaged.

The Organizations and Their Western Footprint

Khalistan as a political cause has never been carried by a single organization. It is a family of groups, some of them genuine armed formations with a documented history of violence, others closer to advocacy fronts, and a few that blur the categories deliberately. Mapping that family across Canada and Britain is the first step toward judging what kind of haven the two countries actually provide. The mapping has to be done carefully, because the names matter and the names are slippery. An Indian charge sheet, a Canadian indictment, a British sanctions notice, and a movement supporter’s own description of the same network will use different vocabulary, attach different boundaries, and imply different things about who is a member and who is merely a sympathizer. The reader who wants to think clearly about the safe haven question has to hold all of those vocabularies at once and trust none of them completely. What follows is an attempt to describe each formation as precisely as the public record allows, noting where the record is solid and where it rests on contested intelligence assessment.

Oldest of the armed groups is the Babbar Khalsa, later the Babbar Khalsa International, founded in the late 1970s and responsible for some of the deadliest violence of the Punjab insurgency. Its Canadian connection is not incidental to the movement’s Western history. It is the foundation of it. Talwinder Singh Parmar, a Babbar Khalsa figure based in British Columbia, was identified by Canadian investigators as the principal organizer of the 1985 bombing of Air India Flight 182, the worst act of aviation terrorism before September 2001. That single fact shaped everything that followed, because it established beyond argument that a Canada-based Khalistan network was capable of mass murder, and it created a permanent reference point against which every later claim and counter-claim would be measured. The Air India connection is the reason the word safe haven attaches to Canada at all.

The Babbar Khalsa’s reconstitution as the Babbar Khalsa International gave the group a structure that survived the death of its founders and the collapse of the insurgency at home. The international suffix was not decorative. It described an organization whose centre of gravity had migrated outward, with cells and sympathizers in Canada, Britain, Germany, and elsewhere, held together by ideology, kinship, and money rather than by territory. For decades the group existed in a strange condition, proscribed in India, listed by some Western governments and not others, and sustained by a diaspora that remembered the 1980s very differently from the way New Delhi remembered them. That asymmetry of memory is the soil in which every later dispute grew. When Britain in December 2025 imposed its first counter-terrorism sanctions on a pro-Khalistan target, naming Gurpreet Singh Rehal and an associated entity in connection with Babbar activity, it marked the first time the United Kingdom had treated any part of the Khalistan ecosystem as a sanctionable terrorism concern rather than a policing matter. The narrowness of that step, a single designation after decades of Indian complaint, is itself evidence for both sides of the safe haven argument. India read it as belated confirmation that the threat was real. British officials read it as proof that the system responds to evidence rather than to pressure.

A second insurgency-era formation, the Khalistan Commando Force, belongs to the same generation. Founded in the mid-1980s, it carried out assassinations and bombings through the peak years of the Punjab violence before Indian counter-insurgency broke it. Its long arc, traced in the organizational history of the KCF, ended not in formal dissolution but in the exile and eventual killing of its chief, Paramjit Singh Panjwar, who was shot during a morning walk in Lahore in 2023. The KCF’s Western footprint was always thinner than its Pakistani one, but its diaspora fundraising and propaganda networks in Britain and Canada helped sustain it through the long dormant decades. The group’s leadership history is a study in fragmentation. Multiple men claimed the title of chief at various points, factions split and reformed, and the organization that existed on paper in Indian charge sheets was often more coherent than the loose reality of scattered exiles and intermittent funding. This gap between the named entity and the lived network recurs throughout the Western story, and it is one of the central difficulties in adjudicating the safe haven question. A government cannot easily proscribe a brand that has no fixed membership, no headquarters, and no bank account in its own name.

The Khalistan Liberation Force occupies a particular place in the British half of this story. Proscribed in India, the KLF was a violent group during the insurgency, and after a long period of near-collapse it was reconstituted under new leadership. Indian agencies identified Avtar Singh Khanda, a UK resident, as its chief, operating under the alias Ranjodh Singh. Khanda’s father had himself been a KLF figure killed by Indian security forces, which gave the lineage a hereditary quality that the movement’s storytellers found useful. The KLF’s relevance to the safe haven question is direct: India says a proscribed terrorist organization was being run from British soil, and Britain neither proscribed the group nor, in New Delhi’s account, acted on the warnings it was given. The reconstitution itself is contested. Indian charge sheets describe a deliberate revival, a chain of command, and operational tasking flowing from Britain to handlers and shooters in Punjab. British and Canadian observers who follow the movement closely describe something looser, a recycling of an old and resonant name by a small number of committed individuals, with the organizational scaffolding existing more firmly in intelligence assessments than in the field. Both descriptions can be partly true at once, and the resulting ambiguity is precisely what makes the proscription decision so fraught. To ban a group, a Western government must be able to define it, and a name that floats free of fixed structure resists definition.

Newest of the armed-labelled groups, and the most internationally consequential, is the Khalistan Tiger Force, because its chief was the man killed in Surrey. The KTF, examined in detail in the dedicated organizational profile, represents the movement’s generational shift. It was not born in Punjab’s fields but assembled in the gurdwaras and social media channels of the diaspora, and Indian intelligence treats it as an organized force with a command structure. Some Western analysts are more sceptical, arguing that the KTF functions more as a brand than as a disciplined organization, a label applied to loosely connected activists. That disagreement is not a footnote. It is the heart of the adjudication this article must perform, because if the KTF is a real terrorist organization then sheltering its members is sheltering terrorism, and if it is mostly a label then the same shelter looks like ordinary protection of dissidents.

Of all these formations, the one that has caused the most friction in recent years is Sikhs for Justice, an organization founded by the lawyer Gurpatwant Singh Pannun, who holds United States and Canadian citizenship. Sikhs for Justice presents itself as a non-violent advocacy body and is best known for the Khalistan Referendum, a series of unofficial, non-binding votes staged among diaspora communities in Canada, Britain, Australia, Italy, and elsewhere. India banned the organization in 2019 and designated Pannun an individual terrorist, pointing to what it calls incitement and threats. Neither Canada nor Britain has proscribed it. In 2025 the group went a provocative step further, declaring the opening of a so-called Embassy of the Republic of Khalistan in Surrey, at the very gurdwara Nijjar had once led. To India the announcement was an affront to its sovereignty staged on the soil of a friendly state. To the group it was political theatre protected by free expression. The gap between those two readings is the gap this article exists to examine.

Two further formations belong on the map. The International Sikh Youth Federation was once a significant diaspora body and was at various points proscribed in the United Kingdom before the ban lapsed. Babbar Akali Lehar, a UK-registered entity, was sanctioned by the British Treasury in December 2025 for what the government described as support to the Babbar Khalsa International. That sanctions action matters enormously to the argument, and it is examined later, because it is the clearest evidence that the British safe haven is neither permanent nor unconditional.

The pattern across all of these groups is consistent. The armed organizations are diaspora-funded and diaspora-amplified, dependent on Western communities for money, recruits, and a platform, and India treats every one of them as a terrorist body. Canada and the United Kingdom proscribe none of the currently active Khalistan groups by name as terrorist organizations. That single fact, more than any other, is what New Delhi means when it uses the phrase safe haven, and it is the fact around which the rest of this analysis turns.

It is worth examining why the proscription gap exists, because it is too easily read as simple negligence or, worse, as deliberate indulgence. Proscribing an organization in a Western democracy is a high-threshold legal act with serious consequences for anyone associated with the named group, and it must withstand challenge in independent courts. The state that proscribes must be able to define the organization, demonstrate that it exists as a coherent entity, and show that it is concerned in terrorism as the relevant statute defines that term. For a fixed, hierarchical group with a clear membership and a record of attacks, that test is straightforward. For the modern Khalistan formations, with their floating brands, their contested leadership, their fluid overlap with diaspora advocacy and with organized crime, the test is genuinely hard to meet on evidence a court would accept. India experiences the non-proscription as a political choice, and sometimes it partly is. But it is also a function of evidentiary standards that apply to every proscription decision a democracy makes, standards that exist precisely to prevent governments from outlawing political movements by decree. The December 2025 British sanctions are instructive here as well. Britain did not proscribe a group. It sanctioned a named individual and a named entity, a narrower and more evidentially manageable step. The choice of instrument reveals the constraint.

There is one more structural point that the organizational map makes clear. The Khalistan project in its Western form is not a pyramid with a single commander at the top. It is a network, and networks of this kind are resilient precisely because they have no centre to decapitate. The killing of Nijjar removed one prominent figure, but it did not remove the Khalistan Tiger Force as an idea, because the idea never depended on him alone. This is the same property that makes the movement so frustrating for India to confront and so difficult for the West to police. A democracy can prosecute an individual for an individual crime. It cannot easily prosecute a diffuse transnational sentiment, and the movement, whatever else it is, is partly a sentiment. That is why the safe haven debate never resolves cleanly, and why each side can point to real evidence for its preferred conclusion.

The Men India Wanted and the West Sheltered

The safe haven debate becomes concrete only when it attaches to individuals, because a country does not shelter an abstraction. It shelters people. Five men illustrate the spectrum, from documented mass-casualty terrorism at one end to contested political activism at the other, and the honest reader will notice that they do not all belong in the same category. The temptation, in a debate this charged, is to reach for a single word that covers all of them, and both sides yield to it. India’s word is terrorist, applied uniformly. The movement’s word is activist, or martyr, applied just as uniformly. Neither word survives contact with the actual biographies. The value of looking at individuals rather than categories is precisely that it forces the abstraction to dissolve into specifics, and the specifics refuse to line up neatly behind either label.

Hardeep Singh Nijjar emigrated from Punjab to Canada in 1997 and built an ordinary suburban life as a plumber in Surrey before becoming president of the Guru Nanak Sikh Gurdwara. To his Canadian neighbours and congregation he was a community leader. To the Indian state he was the operational chief of the Khalistan Tiger Force, the subject of an Interpol notice and an Indian terrorism designation in 2020, with a cash reward attached by the National Investigation Agency for information leading to his arrest. India accused him of running a training module for the group and of involvement in conspiracies to commit violence in Punjab. Canada had received those accusations and had not acted on them, treating Nijjar as a citizen against whom no Canadian charge had been laid. The framing contest around his name is total. There is no neutral noun that both governments would accept. His profile and the diplomatic crisis his killing triggered are examined in full elsewhere in this series, but the essential point for the safe haven question is that India’s most-wanted designation and Canada’s settled citizenship pointed at the same man and could not be reconciled. It is worth noting what Canada did and did not say after his death. Canadian officials did not endorse the Indian terrorism designation, and they did not retract it either. Their position was procedural rather than substantive, namely that Nijjar had been a Canadian citizen, that no Canadian court had convicted him of anything, and that his killing on Canadian soil was therefore a homicide to be investigated like any other. India found that posture maddening, because from New Delhi’s vantage point it amounted to a democracy shrugging at the violent record of a designated terrorist. From Ottawa’s vantage point it was simply the rule of law, which does not allow a foreign designation to substitute for a domestic conviction. The unresolved tension between those two positions is why Nijjar, more than any other figure, became the symbol of the entire dispute.

Ripudaman Singh Malik sits at the opposite, and far darker, end of the documentary record. A Vancouver businessman with historic ties to the Babbar Khalsa, Malik was tried in British Columbia for the Air India bombing that killed 329 people. In 2005 a Canadian court acquitted him, ruling that the Crown had not proven its case beyond reasonable doubt. He lived on as a free man, and in his later years he reportedly distanced himself from the separatist cause, even praising the Indian government, a turn that bewildered the community. The Malik case is the strongest illustration of how the Western system actually works and how its results unsettle every simple narrative. A man whom India and many Canadians believed was connected to a terrorist atrocity walked free because a court applied the ordinary standard of criminal proof. India saw a failure of justice. Canadian law saw the system functioning as designed. Both descriptions are accurate, and that is precisely why the case is so difficult.

Avtar Singh Khanda anchors the British half of the story. Indian agencies identified him as the reconstituted KLF’s chief, operating under an alias, and he had arrived in the United Kingdom on a study visa in 2007 before being granted asylum in 2012. He was named by India as a close associate of the radical preacher Amritpal Singh and was accused of helping him evade Punjab police for more than a month in 2023. Khanda was among the men identified in connection with the March 2023 incident at the Indian High Commission in London, when protesters pulled the Indian flag from the building. The UK dimension of the diaspora network runs through his case, and his story is also where the safe haven argument becomes most contested, because Khanda had been granted protection by the British state precisely as a political asylum seeker, a person the United Kingdom had formally judged to be at risk if returned to India. The asylum grant is the crux of the British argument and the British vulnerability at once. To India it looks like the formal sheltering of a wanted militant. To the British Home Office it was the routine operation of refugee law, a process that assessed Khanda as someone facing a real risk of persecution given his family history and his political profile. The two readings cannot both be the headline, and the choice between them is, in the end, a choice about whether one trusts the asylum process or regards it as a loophole. Khanda’s father had been killed by Indian security forces during the insurgency, and the British system weighed that lineage as a risk factor. India weighed the same lineage as evidence of inherited militancy. The same biography, read by two states, produced opposite conclusions, and that is the safe haven dispute in miniature.

Sukhdool Singh, known as Sukha Duneke, is the case that complicates the terrorism framing in the other direction. Duneke fled India for Canada in 2017, reportedly on a fraudulently obtained passport, and was associated with the organized-crime network linked to the gangster Arsh Dalla. He was wanted in India in connection with a string of killings, and Indian agencies described him as a gangster-terrorist, a hybrid figure in whom criminal enterprise and separatist labelling overlapped. The Winnipeg killing that ended his life looked, to many observers, as much like an episode of inter-gang violence as an episode of the shadow war. Duneke is the reminder that the Khalistan label in the modern era is frequently entangled with the Punjabi organized-crime underworld, and that entanglement, examined in the parallel case of Harvinder Singh Rinda, makes attribution genuinely hard rather than merely politically inconvenient. The gangster-terrorist category is itself worth interrogating, because it serves a purpose for more than one actor. For India it usefully merges two threats into one, so that a man who is primarily a criminal can be folded into the terrorism file and a man who is primarily a militant can be discredited by association with crime. For the gangsters themselves, the separatist label can be a kind of branding, a way of dressing extortion and contract violence in the language of a cause. Duneke sits exactly on that fault line, and his death, like his life, refuses to resolve cleanly into either category. The honest reader should be suspicious of any account that presents him as a pure martyr or a pure criminal, because the evidence supports neither caricature.

Gurpatwant Singh Pannun is the fifth figure and the one furthest from the armed wing. A practising lawyer and the public face of Sikhs for Justice, Pannun has never been credibly accused of personally carrying out violence. He is an organizer, a propagandist, and a litigant, and his referendum campaign is a political project. India designates him a terrorist and points to videos in which he issues what New Delhi calls threats. His defenders point out that issuing inflammatory speech, however offensive, is not the same as committing terrorism, and that a designation built on rhetoric is a designation built on contested ground. Pannun’s importance to this analysis is that the United States Department of Justice, in an indictment unsealed in 2023, alleged that an Indian official had directed a foiled plot to murder him on American soil. He is therefore both an accused man and an alleged intended victim, and his case is the single clearest demonstration that the shadow war did not stop at Pakistan’s borders.

Lay those five lives side by side and the spectrum is obvious. Parmar and the Air India network represent proven mass-casualty terrorism. Khanda and the reconstituted KLF represent a contested armed-group claim resting heavily on Indian intelligence assertion. Duneke represents the criminal-separatist hybrid. Pannun represents political advocacy that India has reclassified as terrorism. Nijjar sits somewhere in the middle, an Indian terrorism designation against a Canadian citizen never charged at home. Any analysis that flattens this spectrum into a single word, whether that word is terrorist or activist, has stopped describing reality and started taking a side.

A sixth name belongs in this section, and it points in the opposite direction from the other five. Jagtar Singh Johal is a British citizen from Dumbarton in Scotland who travelled to Punjab in 2017 for his wedding and was detained by Indian police shortly afterward. He has been held in Indian custody ever since, accused of involvement in a series of targeted killings, charges he and his family reject. His supporters, and a significant body of British parliamentary opinion, describe his detention as arbitrary, allege that a confession was extracted under torture, and point to a United Nations working group that found his detention to lack legal basis. India maintains that he faces ordinary criminal process. Johal matters to the safe haven argument because he is the mirror image of every other figure in this story. Where India accuses the West of sheltering men it considers terrorists, the Johal case allows Britain’s Sikh community and human-rights campaigners to accuse India of reaching across borders to detain a man for his diaspora politics. The phrase transnational repression, which Western governments increasingly apply to the actions of authoritarian states against exiles, is wielded by India against the West and by the West’s Khalistan activists against India. Both sides have learned to speak the language of victimhood, and the Johal case is where that symmetry becomes impossible to ignore.

The Killings, the Deaths, and the Disputed Cases

If Canada and Britain are safe havens, then the events that ended these men’s lives are the moment the haven failed, and the manner of each death tells us a great deal about who was responsible and how confident we can be in saying so. The discipline this section demands is to treat each death as a separate evidentiary problem rather than as an episode in a single pre-written story. Four deaths in roughly a year, all of men with separatist associations, invite the mind to assemble them into a campaign. Sometimes that assembly is warranted and sometimes it is not, and the only way to tell the difference is to look at what each investigation actually produced. Some of these cases carry formal state attribution. Some carry none. Some are officially ruled natural deaths. The honest account keeps those distinctions sharp rather than letting the strongest case lend its certainty to the weakest.

The first of the modern Western killings was Ripudaman Singh Malik’s. In July 2022 Malik was shot dead in his vehicle outside his business in Surrey. Two men were later charged by Canadian authorities, and the investigation produced evidence that the killing may have had roots in the local underworld and in business disputes rather than in any campaign directed from abroad. No state attribution was ever made. The Malik case, taken on its own, is not strong evidence of an Indian operation, and a careful analyst should resist the temptation to fold it into a pattern simply because the victim carried a Khalistan history. Malik had enemies of many kinds.

Malik’s killing also illustrates a methodological trap that recurs throughout this subject. Once a narrative of state-directed assassination takes hold, every subsequent death of a person with a separatist past is read backward through that lens, and the reading can be wrong. Malik, in the years before his death, had publicly thanked the Indian government and distanced himself from the separatist cause, a posture that made him an unlikely target for an Indian operation and a more plausible target for those within the movement who regarded his conversion as betrayal, or simply for the commercial rivals any businessman accumulates. The disciplined approach is to treat each death as its own evidentiary problem, to ask what the investigation actually found, and to resist the gravitational pull of the larger story. By that standard the Malik case sits apart from the others. It is a Canadian homicide with Canadian suspects and a plausible local motive, and folding it into the shadow war would be an act of pattern-making rather than analysis.

The procedural texture of the Canadian response deserves its own attention, because it is what most sharply distinguishes the Western theatre from the Pakistani one. The Integrated Homicide Investigation Team in British Columbia ran the Nijjar file as a conventional murder inquiry, and the arrests in May 2024 followed the ordinary path of surveillance, forensic work, and the gathering of admissible evidence. The four men charged were Indian nationals living in Canada, and they entered the criminal justice system as defendants with the full protections any accused person receives. When the RCMP held its press conference in October 2024, the Commissioner and senior officers did something almost unheard of, describing publicly an alleged campaign of homicides, extortion, and coercion that they linked to agents of a foreign government, and announcing that they had felt compelled to go public because of an imminent threat to public safety. The Canadian government simultaneously expelled six Indian diplomats, including the High Commissioner, and India responded in kind, expelling Canadian envoys and withdrawing its own senior staff. None of this resembles the silence that follows a killing in Pakistan. Whatever else can be said about Canada, its institutions reacted to the violence on their soil as an attack on their sovereignty and processed it through courts, police, and Parliament rather than burying it.

Nijjar’s killing eleven months later is a different matter. On 18 June 2023 he was shot multiple times in his truck in the gurdwara parking lot by two masked gunmen who escaped in a waiting vehicle. The Royal Canadian Mounted Police treated it from the outset as a targeted homicide. In May 2024 Canadian police arrested and charged four Indian nationals with the murder, and in October 2024 the RCMP took the extraordinary step of publicly stating that it had gathered evidence linking agents of the Government of India not only to Nijjar’s death but to a broader pattern of violent activity on Canadian soil, including extortion and intimidation. That public statement, made by a national police force against a friendly foreign government, has no recent parallel in Canadian history. Three months earlier, in September 2023, Prime Minister Justin Trudeau had told the House of Commons that Canadian agencies were pursuing credible allegations of a potential link between Indian agents and the killing. India rejected the accusation as absurd and politically motivated and has consistently denied any role. The Canadian shadow war analysis elsewhere in this series examines the diplomatic cascade in detail, but for the purposes of the safe haven question the relevant fact is narrow and important: Nijjar’s case is the only one of the Western deaths that carries formal state attribution from the host country’s police and prime minister.

Khanda’s death is the most genuinely contested case in the entire Western file, and the responsible thing is to lay out both readings rather than choose one. Khanda died in a Birmingham hospital on 15 June 2023, three days before Nijjar was shot. His supporters immediately alleged that he had been poisoned, and the proximity of the two deaths fed a narrative of a coordinated transnational operation. The Indian press and some Indian-aligned commentators reported, citing medical records, that Khanda had been suffering from acute leukaemia. West Midlands Police reviewed the death and stated that it was not being treated as suspicious. A subsequent investigation by a British newspaper raised questions about the thoroughness of that review and reported that Khanda’s family in Punjab had been subjected to intimidation by Indian police in the months before he died. No British coroner or court has ruled the death a homicide. Khanda’s case, in other words, is one where the medical evidence points one way, the political suspicion points another, and the official verdict is natural causes. Anyone who presents it as a confirmed assassination has gone beyond what the record supports, and anyone who dismisses the suspicion entirely has ignored the documented harassment of his family and the unsettling timing.

Sukhdool Singh Duneke was shot dead in Winnipeg in September 2023, in the same season as Nijjar’s killing, a coincidence that invited the same inference of a campaign. But Duneke’s deep entanglement with the Punjabi organized-crime world means his death has at least two plausible explanations, a shadow-war hit or a gangland settling of scores, and the available evidence does not decisively separate them. The honest verdict on Duneke is uncertainty. It is worth being precise about why uncertainty is the correct verdict rather than a failure of analysis. A man who is both a wanted separatist and an active participant in a violent criminal network has, by definition, two distinct populations of enemies, and a killing that fits the modus operandi of one population fits the modus operandi of the other equally well. There were no arrests immediately publicized, no host-state attribution, and no claim of responsibility that survived scrutiny. In the absence of those things, an analyst can responsibly say that the killing is consistent with a shadow-war operation and equally consistent with an underworld feud, and can refuse to choose. That refusal is not fence-sitting. It is the discipline the evidence requires, and it is precisely the discipline that the louder participants in the public debate, on both sides, decline to exercise.

A wider international fact frames all four cases. The United States indictment in the Pannun plot named a specific Indian official and alleged a directed murder-for-hire conspiracy on American soil, foiled before it could be executed. The indictment described how an alleged intermediary, Nikhil Gupta, was said to have been tasked with arranging the killing and to have contacted a purported hitman who was in fact an undercover American agent. Gupta was arrested in the Czech Republic and, after a contested extradition fight, transferred to United States custody to face trial, while an Indian inquiry committee separately examined the conduct of the official named. That indictment matters here because it establishes, in a court filing by a Five Eyes government, that the campaign’s reach extended to Western democracies, and it lends corroborating weight to the Canadian allegations by pattern. A British documentary later reported that the United Kingdom’s signals intelligence agency had passed Canada intercepted communications discussing three potential targets, naming Nijjar, Khanda, and Pannun. That reporting is unconfirmed by the governments involved and should be treated as allegation rather than established fact, but it is part of the public record and cannot simply be omitted.

Set the four deaths against the dozens of killings inside Pakistan and the contrast is the analytical payoff of this section. In Pakistan the targeted killings are typically clean in the operational sense and untraceable in the evidentiary sense, carried out by motorcycle-borne gunmen and followed by no arrests, no charges, and no host-state attribution, because the host state has neither the capability nor the will to investigate. In Canada the very opposite happened. The killing of Nijjar produced arrests, criminal charges, a public police accusation against a foreign state, and a prime ministerial statement in Parliament. The Western environment did not prevent the violence. It did something Pakistan never does. It generated consequences. That difference is the strongest single piece of evidence that the word safe haven, applied to Canada and Britain, means something fundamentally different from what it means in Lahore or Karachi.

The point cuts in a direction that neither side of the debate finds entirely comfortable. For India, the Canadian response is unwelcome because it turned an alleged operation into a public catastrophe, but it also vindicates, in a backhanded way, India’s long-standing claim that the Western states are serious about the rule of law. For the separatist networks, the Canadian response is superficially welcome, because it cast India as the aggressor, but it also demonstrates that the Western haven is a place where violence is investigated and prosecuted rather than tolerated, which is not the unconditional sanctuary the movement’s rhetoric sometimes implies. The deaths examined in this section do not deliver a clean verdict for anyone. What they deliver is a distinction, the distinction between a haven that is the by-product of a functioning legal order and a haven that is the deliberate product of a state’s strategic policy. Everything else in the safe haven debate flows from getting that distinction right, and most of the confusion in the public conversation flows from getting it wrong.

The Infrastructure of Shelter

Behind the individual cases lies a system, and the safe haven argument ultimately stands or falls on what that system actually is. India’s case is that Canada and Britain host an infrastructure that allows the separatist project to recruit, raise money, propagandize, and intimidate, and that the host states permit it. The counter-case is that most of what India calls infrastructure is the ordinary apparatus of democratic life, and that mislabelling it as a terror network is itself a form of pressure on legitimate communities. Both descriptions capture part of the truth, and the task here is to sort which part belongs to which.

The most important institution is the gurdwara. The Sikh temple is a place of worship, a community kitchen, a welfare hub, and a civic forum, and in the diaspora it is also something more, because the management committees that control gurdwara finances and platforms are elected bodies, and their elections can be fiercely contested between factions. Where a pro-Khalistan faction controls a major gurdwara, it controls a podium, a donor base, and a venue. India points to gurdwaras where separatist iconography is displayed, where images of men it regards as terrorists are honoured as martyrs, and where political messaging is woven into religious gatherings. That is real, and it is a genuine source of the movement’s resilience. But the gurdwara is also, irreducibly, a religious institution protected by freedom of worship, and the fact that some temples are run by separatist sympathizers does not make the institution itself a terror front. The infrastructure here is dual-use, and that is exactly what makes it so hard to police.

The contest for control of gurdwara committees is itself a window into the movement’s true size. These elections are often bitterly fought, and the separatist faction does not always win. In many temples, moderate committees focused on worship, langar, and community welfare hold power, and the pro-Khalistan groups are a vocal minority rather than a dominant bloc. The temples that become flashpoints, the ones that host the most provocative events and display the most contentious imagery, are a subset, not the whole. This matters because the safe haven narrative, in its cruder form, treats the gurdwara as such as a node of terror infrastructure, when the honest picture is of a religious network within which a political faction competes, sometimes successfully and sometimes not, for influence. A government that treated every gurdwara as a security concern would be both factually wrong and guilty of exactly the conflation that damages ordinary Sikh communities. A government that ignored the genuine cases, the temples where the honouring of violence crosses from memory into incitement, would be failing at its actual job. The line runs through the institution, not around it, and that is the hardest kind of line for any state to police.

Money is the second pillar. The movement’s finances run through diaspora donations, some collected openly for political campaigns and some, India alleges, raised under humanitarian or religious cover and diverted. The harder edge of the financial story is the organized-crime nexus, which is most pronounced in Canada. Investigative work, including the reporting compiled by the journalist Sam Cooper in his book on the subject, has documented how Punjabi gangs involved in drug trafficking and extortion intersect with separatist networks, with criminal proceeds and separatist politics flowing through overlapping circles of people. The Duneke and Rinda cases are the human face of that overlap. This is the part of the infrastructure where the safe haven label is most defensible, because criminal financing is a crime in any democracy, and a system that fails to follow the money is failing at ordinary law enforcement, not protecting free speech.

The criminal dimension has its own cast of recurring names, and they matter because they are where the romantic story of national liberation collides with the ugly reality of the extortion racket. Arshdeep Singh Gill, the gangster known as Arsh Dalla, is described by Indian agencies as a Canada-based operative who inherited a slice of Nijjar’s network and combined separatist tasking with conventional organized crime, and he was reportedly detained in Canada in late 2024 in connection with a shooting before becoming the subject of an Indian extradition request. The rival ecosystem associated with the imprisoned gangster Lawrence Bishnoi, whose network has been blamed by Canadian police for a wave of violence, runs partly against and partly alongside these separatist-tinged gangs, and the result is a Punjabi underworld stretching from Surrey to Brampton to Punjab itself in which it is genuinely difficult to say where a political killing ends and a turf war begins. For the safe haven argument this matters in a specific way. The criminal nexus is the strongest evidence that something real and dangerous is being sheltered, but it is also the part of the problem that is least about ideology and most about ordinary cross-border crime, the kind every Western country struggles to police. India is right that the nexus exists. It is on weaker ground when it presents gangland Canada as proof of a state-tolerated terror project rather than as proof of a law-enforcement gap.

Propaganda and recruitment form the third pillar, and they have been transformed by social media. The modern movement is amplified far beyond its organic support base by online channels that reach young people in Punjab, in the diaspora, and across the wider Sikh world. The reconstituted KLF’s press statements, according to Indian and British reporting, were issued from the United Kingdom. The Khalistan Referendum is run as a slick, internet-native campaign. India argues that this machinery radicalizes and incites. Scholars of the British movement, including academics such as Gurharpal Singh who have studied its organizational structure for decades, tend to describe it as a smaller and more fragmented phenomenon than Indian official rhetoric suggests, a movement whose online volume far exceeds its real-world membership.

The referendum deserves a closer look, because it is the single activity that has generated the most diplomatic friction while being, on its face, the most clearly lawful. The votes are organized by Sikhs for Justice, staged in rented halls and gurdwara grounds, and counted by the organization itself, which announces large turnout figures that no independent body verifies. The exercise has no legal standing, changes no borders, and binds no government. Its purpose is theatre and momentum, a way of manufacturing the appearance of a mandate and of keeping the idea of Khalistan in the news. India treats each polling event as a hostile act and lobbies host governments to ban them, and host governments, citing freedom of assembly, generally decline. The events have occasionally been marred by violence, including a stabbing outside one Canadian vote and confrontations between rival groups, and that disorder gives host-state police a public-order reason to monitor them. But monitoring a protest is not the same as dismantling a terror cell, and the referendum, stripped of the inflated rhetoric on both sides, is best understood as exactly the kind of provocative, non-violent political performance that democracies are built to tolerate even when an allied government finds it infuriating.

The fourth pillar is the one India finds hardest to accept and the one that most clearly is not a terror infrastructure at all. It is the law. Canada’s Charter of Rights and Freedoms and Britain’s human rights framework protect speech, assembly, and association, and they protect them for asylum seekers as well as citizens. Khanda held asylum in Britain because the British state had formally assessed that he faced a risk of persecution in India, a judgement made under refugee law, not a favour to a militant. Duneke’s entry to Canada on a fraudulent passport was an immigration-control failure, a different kind of problem from deliberate shelter. The referendum events, however offensive New Delhi finds them, are non-binding political expression. To call the Charter or the Refugee Convention part of a terror infrastructure is to call the rule of law itself a terror infrastructure, and that is a claim no democracy can accept and remain a democracy.

This is where the named disagreement at the centre of the article has to be adjudicated rather than dodged. Are Canada and the United Kingdom genuinely safe havens for terrorism, or are they functioning democracies whose free-speech protections correctly separate political advocacy from criminal violence? The defensible answer is that they are partly the first and substantially the second, and that the proportions are knowable case by case. The historic violence is real: the Air India bombing was mass murder organized from Canadian soil, and that fact is not erased by time. The criminal-financial nexus is real and is a genuine law-enforcement failure when it goes unaddressed. Those are the parts of the picture where the safe haven charge has force. But the referendum campaigns, the protests, the asylum grants, and the ordinary advocacy for an independent Sikh state are protected political activity, and India’s tendency to fold all of it into the single category of terrorism is the reason the dispute has been so intractable. There is a counter-narrative that British and Canadian Sikh organizations advance with increasing confidence, the narrative of transnational repression, which holds that India uses the terrorism label to chill lawful dissent abroad, and which points to cases such as the prolonged detention in India of the British national Jagtar Singh Johal as evidence. That narrative is not the whole truth either, because it tends to minimize the genuine violent fringe and the criminal money. The truthful position lives between the two framings, and a serious analysis says so plainly.

The decisive evidence that the Western haven is conditional rather than absolute is what the host states have started to do. In December 2025 the British Treasury, for the first time, used its domestic counter-terrorism sanctions regime against a pro-Khalistan target, imposing an asset freeze on a UK national accused of supporting the Babbar Khalsa International and on a UK-registered entity accused of aiding the same group. Sikh organizations across a dozen countries condemned the move as an overreach that punished a man never charged in a court, and that objection is itself a serious one about due process. But the action demonstrates the structural point: the British state retains and will use the legal tools to act against what it judges to be genuine terrorism support, and the same is true of Canada. Pakistan’s sanctuary system, built and protected by the intelligence apparatus that treats militants as instruments of state policy, has no equivalent self-correcting mechanism, because the shelter there is the policy. In the West the shelter is a by-product of rights, and rights coexist with enforcement. That is the difference between an over-governed space and an ungoverned one, and it is why the two kinds of safe haven, examined together in the global comparison of sanctuary models, should never be described in the same words.

How the Shadow War Changed These Safe Havens

The Western chapter of the campaign did not unfold in a vacuum, and the years since 2022 have transformed the political environment in both Canada and Britain in ways that neither the movement nor New Delhi fully anticipated. Tracing that change is the way to answer the largest question of all, which is whether reaching into Western democracies achieved anything for India or simply cost it.

Before 2023, the separatist networks in Canada and the United Kingdom operated with what New Delhi regarded as near-total impunity. India’s diplomatic protests about gurdwara platforms, referendum events, and the honouring of insurgency-era figures were lodged year after year and produced little visible response. The Western governments treated the matter, for the most part, as a question of domestic free expression and community relations rather than national security. The movement’s diaspora infrastructure, the gurdwara committees and the online campaigns and the fundraising channels, ran with low friction. That was the baseline, and from India’s perspective it was an intolerable one.

The killing of Nijjar shattered the equilibrium. Trudeau’s statement in the House of Commons in September 2023 converted a long-running irritant into a first-order diplomatic crisis. Both countries expelled diplomats. India suspended visa services for Canadians, recalled its High Commissioner, and demanded a reduction in Canada’s diplomatic presence. Trade negotiations were frozen. For more than a year the relationship between two Commonwealth democracies and natural economic partners was effectively broken, and the rupture deepened in October 2024 when the RCMP went public with its allegation of a broader pattern of Indian-linked violence on Canadian soil. The United States case over the alleged plot against Pannun ran in parallel, ensuring that the controversy was not Canada’s alone but a shared concern across the Five Eyes intelligence partnership. By any conventional measure, the Western operations, if that is what they were, had produced a diplomatic catastrophe for India.

Then the picture turned again, and the turn is the most instructive part of the story. Trudeau, weakened politically, announced his resignation, and in March 2025 Mark Carney became Prime Minister of Canada. Carney’s government signalled an interest in repairing the relationship, and the repair moved quickly. Modi and Carney met on the sidelines of the G7 summit in June 2025. High commissioners were reinstated. A Trade and Investment Ministerial Dialogue followed in November 2025. India’s National Security Adviser visited Ottawa in early 2026 for talks on security cooperation, and Carney made a multi-day visit to India. Most striking of all, in its 2025 public report Canada’s intelligence service, CSIS, described Canada-based Khalistani extremism as a national security threat and acknowledged that some individuals were using Canadian institutions to advance a violent extremist agenda. That assessment, published by Ottawa’s own security establishment, was the closest the Canadian state had ever come to validating a concern India had voiced for decades. In Britain the trajectory ran the same direction, with the December 2025 sanctions action representing the first use of domestic counter-terrorism powers against a pro-Khalistan target and a leaked government extremism review devoting substantial attention to the movement.

So did the shadow war work in the West? The honest answer requires holding two findings together. The first finding is that the operations, on the Western evidence, were extraordinarily costly. They produced criminal charges against Indian nationals in Canada, a murder-conspiracy indictment in the United States naming an Indian official, a year of diplomatic wreckage, and a lasting reputational injury, because a democracy accused by a friendly democracy’s police force of directing violence on its soil does not easily repair that image. The second finding is that the security environment for the separatist networks in both countries has nonetheless tightened, with formal threat designations, sanctions, and a new willingness by Western governments to treat the movement’s violent fringe as a security matter rather than a community-relations one. The temptation is to credit Indian pressure for the second finding, but that reading is too simple. The networks did much of the damage to their own cause. The opening of a self-declared Khalistan embassy in Surrey, attacks on Hindu temples, aggressive referendum confrontations, and the steady drift of parts of the movement into the orbit of organized crime alienated Western publics and policymakers independent of anything New Delhi did. The recalibration in Ottawa and London owes as much to the movement’s own conduct as to India’s diplomacy.

It is worth dwelling on the strangeness of the reset, because it tells us something about how the international system actually weighs these matters. In the autumn of 2023 the relationship between India and Canada was as poisoned as the relationship between two democracies can be without an outright rupture of recognition. A little over two years later, prime ministers were meeting, trade dialogue had resumed, and national security advisers were exchanging visits. Nothing about the underlying dispute had been settled. No Canadian court had delivered a verdict in the Nijjar prosecution. India had neither admitted nor been cleared of the central allegation. What changed was the calculation of cost. Both governments concluded that the strategic and economic value of the relationship, the trade, the energy and education ties, the shared interest in a multipolar Indo-Pacific, outweighed the value of continuing to litigate the quarrel in public. The Khalistan question did not disappear. It was subordinated. That is how states usually handle disputes they cannot resolve, and it is a useful corrective to the assumption that moral clarity drives foreign policy. The diaspora networks, watching their cause become a bargaining chip rather than a cause, learned something uncomfortable about their leverage.

The British trajectory is quieter but follows the same logic. The United Kingdom never had a Nijjar moment, no single killing on its soil with a prime ministerial accusation attached, and so its recalibration was incremental rather than dramatic. The death of Khanda, officially attributed to natural causes, did not force the issue. What moved British policy was the slow accumulation of pressure, the leaked extremism review, the December 2025 sanctions, and a broader hardening of the British state’s posture toward transnational extremism of several kinds. Britain, like Canada, also had a powerful incentive to protect its relationship with India, a major trade partner and a strategic counterweight in Asia, and that incentive pulled in the direction of taking Indian concerns more seriously. The result in both countries is a posture that is harder on the movement’s violent and criminal fringe and unchanged, or even more protective, toward its lawful political expression. That split posture is not a contradiction. It is the democratic answer to the safe haven question, and it is the answer this article has argued is the correct one.

The deepest lesson of the Western chapter is the one stated at the outset and now earned by the evidence. The safe haven that Canada and Britain provided was never the safe haven Pakistan provides. It was never ungoverned, never the product of a deliberate state policy of shelter, and never immune to evidence. It was a by-product of democratic rights, and democratic rights, unlike strategic sponsorship, come bundled with courts, police, and the capacity for self-correction. The campaign’s most consequential effect on these countries was not that it eliminated anyone. It was that it forced two democracies to do in public what they had long avoided, which was to draw an explicit line between the political advocacy they would continue to protect and the criminal violence they would now prosecute. The four-decade arc of the movement, set out in the complete Khalistan terror timeline, shows that the line had always been blurred by both sides, by a movement that hid violence inside advocacy and by an Indian state that labelled advocacy as violence. The shadow war’s Western chapter, for all its recklessness and cost, ended by forcing the line into the open. As the geography of the safe-haven system continues to shift, the Western democracies are likely to keep narrowing the space for genuine violence while continuing to protect genuine speech, because that is what their own laws require of them. That is not the collapse of the safe haven. It is the haven behaving, at last, like a democracy.

Frequently Asked Questions

How did Canada become a safe haven for Khalistan terrorism?

Canada became central to the movement through migration rather than policy. A large, prosperous Sikh community settled there over more than a century, and from the 1980s a separatist faction within it built fundraising and propaganda networks. The defining event was the 1985 Air India bombing, organized by a British Columbia-based Babbar Khalsa network, which proved the diaspora’s capacity for mass-casualty violence. India’s complaint is not that Canada deliberately shelters terrorists but that it has treated separatist activity as protected expression and has been slow to act on intelligence. Canada’s position has shifted recently, with its intelligence service in 2025 describing Canada-based Khalistani extremism as a national security threat. It is important to be precise about the word safe haven here, because it carries an implication of intent that the Canadian record does not support. Canada did not set out to be a sanctuary. Its electoral system, its constitutional protections, and its asylum law together produced an environment in which separatist organizing could occur without much friction, and political incentives discouraged successive governments from confronting it. That is a story of incentives and institutions, not of a state policy of shelter, and the distinction is essential to understanding why Canada’s behaviour differs so sharply from Pakistan’s.

Why does the UK shelter Khalistan separatists?

Britain does not shelter them as a matter of state policy. The United Kingdom hosts a large Sikh population dating to the postwar decades, and its laws protect speech, assembly, and asylum. Avtar Singh Khanda, identified by India as a KLF chief, held British asylum because the UK had assessed he faced persecution risk in India under refugee law. India argues that Britain has been too permissive toward separatist organizing and slow to proscribe groups. The British government has begun to act more firmly, using counter-terrorism sanctions against a pro-Khalistan figure in December 2025 and devoting attention to the movement in a government extremism review. The British case is complicated by a cross-pressure that Canada feels less acutely. The United Kingdom maintains a close trade and strategic relationship with India and has at times been visibly anxious not to damage it, while also representing a sizeable and politically organized Sikh constituency at home. That double bind has made British policy on the movement look hesitant and inconsistent over the years, lurching between gestures toward New Delhi and reassurances to domestic communities. The recent firmer steps suggest the balance has shifted, but the underlying tension between a domestic community and a foreign partner has not disappeared.

How do Western safe havens differ from Pakistani ones?

The difference is structural and total. Pakistan’s sanctuary is a deliberate state policy, built and protected by its military and intelligence services, which treat militants as strategic assets. Protection there is near-absolute because the state itself is the protector. Canada and the United Kingdom shelter the movement only as an unintended by-product of democratic rights, the protections for speech, association, and asylum that apply to everyone. That shelter is conditional. It collapses the moment a prosecutor can prove a crime, which is why the Nijjar killing produced arrests and charges while killings in Pakistan produce neither. One model is the ungoverned space, the other the over-governed space.

Does free speech protection cover Khalistan advocacy?

In Canada and the United Kingdom, advocating for an independent Sikh state as a political idea is generally protected speech, including through protests, rallies, and non-binding referendum events. Free expression protections do not cover incitement to violence, criminal financing, intimidation, or terrorism, and the line between protected advocacy and prosecutable conduct is drawn case by case by courts and prosecutors. India disputes where that line falls, arguing that figures and organizations it designates as terrorist should not enjoy any platform. Western legal systems require an actual crime to be proven before speech-adjacent activity can be punished, which is the core of the disagreement.

Where is the line between advocacy and terrorism in Canada?

The line is legal, not rhetorical. Holding a referendum, displaying separatist symbols, or campaigning for Khalistan as a political project falls on the protected side. Organizing or financing violence, conspiring to commit murder, extortion, and intimidation fall on the criminal side. The Nijjar case showed the system applying that line: Canada had not charged him with any crime, but his alleged killers were charged with murder. The Malik acquittal in the Air India case showed the same standard from the other direction, with a court declining to convict on evidence it found insufficient. The line is real, but it can only be enforced through proof.

How do Khalistan networks operate in both countries?

They operate through three main channels. The first is the gurdwara, where management committees controlled by separatist factions provide platforms, donors, and venues. The second is finance, including diaspora donations and, particularly in Canada, an overlap with Punjabi organized-crime networks involved in trafficking and extortion. The third is propaganda, increasingly run through social media campaigns that reach far beyond the movement’s organic support base. The armed-labelled groups, including the KTF and the reconstituted KLF, depend on this diaspora machinery for money and visibility, while advocacy bodies such as Sikhs for Justice run internet-native political campaigns. It is worth stressing that these channels are not all of one kind. The gurdwara platform and the social media campaign are, in themselves, lawful uses of lawful institutions, and most of the people involved in them are engaged in protected political activity. The criminal-financial channel is genuinely different, because trafficking and extortion are crimes in any jurisdiction. Treating all three channels as a single terror network is the analytical error that drives much of the public confusion. The honest map separates the lawful infrastructure of advocacy from the unlawful infrastructure of crime, and recognizes that the movement’s resilience comes partly from the fact that the two are tangled together.

Why do Western governments tolerate Khalistan activities?

Tolerance is partly a matter of law and partly a matter of politics. Legally, democracies cannot punish political speech or association without proving a crime, so much separatist activity is simply protected. Politically, Sikh communities in both countries are electorally significant, and parties have been reluctant to alienate them. India argues this slides into appeasement. The recent shift suggests the tolerance was always conditional: once the movement’s violent fringe and its criminal financing became impossible to ignore, and once the diplomatic stakes rose, both Ottawa and London moved toward firmer security designations and enforcement. The word tolerate is itself slightly misleading, because it implies a single deliberate choice. What actually happened was the cumulative result of many smaller decisions, each defensible on its own terms, made by courts applying evidentiary standards, by immigration officials applying refugee law, and by politicians weighing constituencies. No one designed the haven. It emerged.

Has India pressured Canada and the UK to crack down?

India has pressed both governments for years, through diplomatic protests, intelligence sharing, requests to proscribe specific organizations, and public statements by its leaders. The pressure intensified after 2023. Whether the recent hardening of Western policy is the result of that pressure is debatable. The movement’s own conduct, including the declaration of a Khalistan embassy in Surrey, attacks on Hindu temples, and the drift toward organized crime, alienated Western publics independently. The honest assessment is that Indian diplomacy and the movement’s self-inflicted damage both contributed to the shift. There is also a paradox worth noting. The most aggressive form of Indian pressure, if the allegations of state-directed violence on Western soil are accurate, was almost certainly counterproductive in the short term, because it cast India rather than the separatists as the aggressor and triggered a year of diplomatic rupture. The quieter forms of pressure, the patient intelligence sharing and the steady diplomatic case-making, are far more plausibly connected to the later hardening of Western policy. Pressure works in this arena when it persuades, and tends to backfire when it shocks.

Who was Hardeep Singh Nijjar?

Nijjar emigrated to Canada from Punjab in 1997 and worked as a plumber in Surrey, British Columbia, before becoming president of the Guru Nanak Sikh Gurdwara. India designated him a terrorist in 2020, identifying him as the operational chief of the Khalistan Tiger Force and accusing him of involvement in violent conspiracies in Punjab. The National Investigation Agency had announced a reward for information leading to his arrest. To his Canadian community he was a temple leader and citizen against whom Canada had laid no charge. He was shot dead outside his gurdwara on 18 June 2023, and his killing triggered the worst crisis in modern India-Canada relations.

How did Avtar Singh Khanda die?

Khanda died in a Birmingham hospital on 15 June 2023. His supporters alleged he had been poisoned, citing the timing relative to Nijjar’s killing three days later. Reporting based on medical records indicated he had been suffering from acute leukaemia, and West Midlands Police stated the death was not being treated as suspicious. A later British newspaper investigation questioned the thoroughness of the review and reported that Khanda’s family in Punjab had faced intimidation from Indian police beforehand. No coroner or court has ruled the death a homicide. It remains the most genuinely contested case in the Western file, with the medical evidence and the political suspicion pointing in different directions.

What is the Khalistan Tiger Force?

The KTF is one of the newer separatist groups and the most internationally consequential, because its chief, Nijjar, was the man killed in Surrey. India treats it as an organized terrorist body with a command structure and has banned it. Some Western analysts argue it functions more as a brand or label applied to loosely connected activists than as a disciplined organization. That disagreement is central to the safe haven debate, because the answer determines whether sheltering its members amounts to sheltering terrorism. The KTF is a diaspora-era formation, assembled through gurdwaras and social media rather than born in Punjab’s insurgency. The brand-versus-organization question is not merely academic. If the KTF is a structured group that recruits, trains, and tasks, then a Western state that fails to act against its members is failing at counter-terrorism. If it is closer to a label that Indian intelligence applies to a scattering of separatist activists, then acting against those members on the strength of the label alone would mean punishing political association rather than crime. The public evidence does not fully settle the question, and that unsettled quality is itself one of the central reasons the safe haven dispute has never reached a clean resolution.

Is Sikhs for Justice a terrorist organization?

It depends entirely on which government is asked. India banned Sikhs for Justice in 2019 and designated its founder, the lawyer Gurpatwant Singh Pannun, an individual terrorist, citing what it calls incitement and threats. Neither Canada nor the United Kingdom nor the United States has proscribed the organization. The group presents itself as a non-violent advocacy body and is best known for the Khalistan Referendum. Its defenders argue that inflammatory speech, however offensive, is not the same as terrorism. Its case illustrates how far the Indian and Western definitions of terrorism diverge. The divergence is not a matter of one side being careless. India’s legal framework allows a body to be designated on the basis of incitement and association, while Western counter-terrorism law generally requires a closer connection to actual violence before a comparable label can attach. Sikhs for Justice sits in the gap between those two thresholds, lawful in one legal order and banned in another, and that gap is the single clearest illustration of why the safe haven argument is, at bottom, an argument about definitions as much as about facts.

What is the Khalistan Referendum?

The Khalistan Referendum is a series of unofficial, non-binding votes organized by Sikhs for Justice among diaspora communities in Canada, the United Kingdom, Australia, Italy, and other countries. It has no legal standing and does not bind any government. Its organizers present it as an exercise in self-determination and a way to demonstrate diaspora support for an independent Sikh state. India regards it as a provocation and a propaganda tool and has protested the events strongly. In Western legal terms the referendum is generally treated as protected political expression rather than criminal activity. The votes are counted by the organizers themselves, who announce large turnout figures that no independent body verifies, so the exercise is best understood as a way of manufacturing the appearance of a mandate and sustaining media attention rather than as a genuine measure of opinion. Some polling events have been marred by violence between rival groups, which gives host-state police a public-order reason to monitor them, but monitoring a contentious protest is a different thing from dismantling a terror operation, and that distinction is exactly what the safe haven debate so often blurs.

Did India operate on Canadian and British soil?

In the Canadian case, Canadian police and the prime minister have publicly alleged that agents of the Indian government were linked to Nijjar’s killing and to a broader pattern of violence, and four Indian nationals have been charged with the murder. India denies all of it. In the United States, a Justice Department indictment alleged an Indian official directed a foiled plot to kill Pannun. In the British case, Khanda’s death has not been ruled a homicide by any court, and allegations of an operation rest on disputed reporting. The Canadian and American allegations are the strongest; the British case remains unproven.

What is the Air India connection to Canada’s Khalistan networks?

The 1985 bombing of Air India Flight 182 killed 329 people and was the worst act of aviation terrorism before September 2001. Canadian investigators identified a British Columbia-based Babbar Khalsa network, led by Talwinder Singh Parmar, as responsible for organizing it. Ripudaman Singh Malik, a Vancouver businessman with Babbar Khalsa ties, was tried for the bombing and acquitted in 2005 when a Canadian court found the evidence insufficient. The atrocity is the historical foundation of the phrase safe haven as applied to Canada, because it proved beyond argument that a Canada-based separatist network was capable of mass murder.

Has the India-Canada relationship recovered?

It has substantially recovered from its 2023 low point. After Justin Trudeau’s resignation, Mark Carney became Prime Minister in March 2025 and pursued a reset. Modi and Carney met at the G7 summit in June 2025, high commissioners were reinstated, and a trade dialogue resumed in November 2025. India’s National Security Adviser visited Ottawa in early 2026, and Carney visited India. The strategic and trade architecture is being rebuilt. The underlying disagreement over Khalistan-linked networks, however, remains unresolved, and incidents such as the declaration of a Khalistan embassy in Surrey continue to create friction.

What did Canada’s CSIS say about Khalistani extremists?

In its 2025 public report, the Canadian Security Intelligence Service described Canada-based Khalistani extremism as a national security threat and acknowledged that some individuals connected to the movement were using Canadian institutions to advance a violent extremist agenda. This was one of the clearest public recognitions by Canada’s own security establishment of a concern India had raised for decades. It marked a significant shift from the earlier Canadian tendency to treat separatist activity primarily as a domestic free-expression and community-relations matter rather than a security one.

Who was Talwinder Singh Parmar and why does he matter?

Talwinder Singh Parmar was a Babbar Khalsa figure based in British Columbia whom Canadian investigators identified as the principal organizer of the 1985 Air India bombing. He was never convicted in a Canadian court and was later killed in India in 1992 in circumstances his supporters dispute. Parmar matters because his case established the founding fact of the Western safe haven debate, the demonstrated capacity of a Canada-based network to organize mass-casualty terrorism. Every later argument about Canada, whether it stresses shelter or stresses the rule of law, is measured against the Air India precedent, which is why his name recurs throughout any serious account of the movement’s diaspora history.

What were the UK’s December 2025 sanctions against Khalistan figures?

In December 2025 the British Treasury imposed counter-terrorism sanctions, including an asset freeze, on a United Kingdom national and an associated UK-registered entity accused of supporting the Babbar Khalsa International. It was the first time Britain had used its domestic counter-terrorism sanctions regime against a pro-Khalistan target. India read the action as belated confirmation of a threat it had described for decades. Sikh advocacy organizations condemned it as an overreach against a man never charged in a court. The measure is significant chiefly because it showed that the British state retains and will use legal tools against what it judges to be genuine terrorism support.

Did India try to assassinate Gurpatwant Singh Pannun?

A United States Department of Justice indictment unsealed in 2023 alleged that an Indian official directed a foiled murder-for-hire plot against Pannun, a Sikhs for Justice leader and dual United States and Canadian citizen, on American soil. An alleged intermediary, Nikhil Gupta, was arrested in the Czech Republic and extradited to face trial. India denied directing any plot and convened an inquiry committee to examine the named official’s conduct. The case has not been resolved in court, so the allegation remains an allegation, but it is significant because it came from a Five Eyes government in a formal legal filing rather than from rhetoric.

Is the Khalistan dispute an example of transnational repression?

It depends on which direction the question is asked. India accuses Canada and Britain of sheltering people it considers terrorists. Sikh advocacy groups and human-rights campaigners accuse India of transnational repression, meaning the use of intimidation, surveillance, and alleged violence to silence diaspora dissent, and they cite cases such as the prolonged detention in India of the British national Jagtar Singh Johal. Both framings contain truth and both are also used tactically. The honest position recognizes that a genuine violent fringe exists within the movement and that lawful political advocacy is also being chilled, and it refuses to collapse those two realities into one.

Are most Sikhs in Canada and the UK supporters of Khalistan?

No. The Sikh communities in both countries are large, diverse, and overwhelmingly engaged in ordinary civic, professional, and religious life, and surveys and community leaders consistently indicate that active support for a separatist state is a minority position. The visibility of the movement online and at organized events can create a misleading impression of its real size. Conflating the broad Sikh population with the separatist project is both factually wrong and a source of genuine harm to communities, and it is a conflation that careful analysis, and responsible governments, are obliged to avoid.