Relations between India and Canada have deteriorated dramatically this week after Prime Minister Justin Trudeau told Canada’s Parliament that Canadian security agencies were investigating what he described as “credible allegations” of a potential link between agents of the Indian government and the killing of Hardeep Singh Nijjar, a Canadian Sikh separatist activist shot outside a gurdwara in British Columbia in June. Canada subsequently expelled a senior Indian diplomat. India rejected Trudeau’s allegation as “absurd and motivated”, expelled a Canadian diplomat in retaliation and accused Ottawa of tolerating Khalistani extremists who threaten Indian interests. India has since suspended visa services for Canadian citizens.
The first requirement in analysing this dispute is separating allegation from established fact. Trudeau’s statement is unusually serious, and a Canadian prime minister would presumably understand the diplomatic consequences of making it publicly. That gives the allegation political weight. It does not by itself establish responsibility. India has denied involvement, and the evidence underlying Canada’s assessment has not yet been presented publicly in enough detail for outsiders to adjudicate the claim independently. The appropriate position at this stage is therefore not neutrality about the seriousness of the allegation but precision about what is known and what remains unproven.
If Canadian investigators eventually establish that agents acting for the Indian state participated in an assassination on Canadian territory, the issue would extend far beyond India-Canada relations. It would constitute an extraordinary violation of Canadian sovereignty and of the principle that governments cannot simply enforce their security preferences inside other countries without those countries’ consent. Democracies cannot accept a system in which one state unilaterally decides that someone living abroad is dangerous and then exercises lethal force inside another jurisdiction.
The principle should remain consistent regardless of which government is involved. India’s democratic system, strategic importance and relationships with Western states would not make such conduct more acceptable if established. Likewise, Canada’s political disagreements with India do not reduce the evidentiary burden required before attributing responsibility. Sovereignty and evidence both matter precisely when the accusation is politically inconvenient.
India’s underlying security concern should not be dismissed either. The Khalistan movement, which advocates an independent Sikh state centred principally on Punjab, has a long and violent history. During the 1980s and early 1990s, Punjab experienced a severe insurgency involving militant organisations, assassinations, bombings and extensive violence by both militants and state security forces. The conflict included the Indian Army’s Operation Blue Star in 1984, the subsequent assassination of Prime Minister Indira Gandhi by two Sikh bodyguards and the horrific anti-Sikh violence that followed her assassination.
For many Indians, particularly those old enough to remember that period, Khalistani militancy is therefore not an abstract debate about self-determination. It is associated with terrorism, separatist violence and one of the most traumatic periods in modern Indian domestic politics. India’s concern about organisations abroad glorifying militants or advocating renewed separatism consequently has a genuine historical and national-security context.
Yet historical trauma cannot determine another country’s speech laws. Canada contains one of the world’s largest Sikh diaspora communities, and Canadian citizens possess extensive constitutional protections for political expression. Advocating a political objective that India considers unconstitutional does not automatically make the advocacy illegal in Canada. A person can support Quebec independence, Scottish independence, Catalan independence or Khalistan without necessarily committing a criminal offence.
This is where the two countries’ legal perspectives diverge. India sees some overseas separatist activism through the lens of territorial integrity and national security. Canada sees political advocacy principally through Canadian law, under which the threshold between radical political speech and criminal conduct may be different.
Neither state gets to erase the other’s jurisdiction.
Canada cannot reasonably tell India that violent threats against Indian diplomats are irrelevant merely because they originate abroad. India cannot reasonably demand that Canada criminalise peaceful political advocacy simply because the objective advocated would violate India’s territorial integrity.
The relevant distinction is between separatism and violence.
Democracies need to be capable of tolerating the first while prosecuting the second.
That sounds straightforward in principle and becomes difficult in practice because political movements often contain several organisations and individuals operating across a spectrum. One activist organises a peaceful referendum campaign. Another glorifies people who committed violence. Another makes a direct threat. Another raises money that authorities believe reaches militants.
Governments then face the temptation to treat the entire movement according to its most extreme members.
That creates both security and civil-liberties problems.
If Canada underreacts to genuine violent activity because officials fear appearing hostile towards Sikh political expression, India has legitimate grounds for frustration. If India treats broad support for Khalistan as equivalent to terrorism, it risks collapsing political dissent and violent extremism into one category.
Good security policy requires behavioural specificity.
What did the individual actually do?
Was there incitement?
Financing?
Planning?
Threats?
Violence?
Political identity should not substitute for evidence of conduct.
The Nijjar case makes this distinction especially important. India had accused Nijjar of involvement in terrorism and sought action against him, while supporters described him as a political activist. Those competing descriptions should not determine the legal analysis by themselves. If India possessed evidence of criminal activity, the appropriate international mechanisms include intelligence-sharing, law-enforcement cooperation and extradition requests subject to Canadian legal procedures.
Those procedures can be frustrating, particularly if Indian officials believe Canadian authorities are insufficiently responsive.
But frustration with another country’s courts does not create jurisdiction.
This is the fundamental principle underneath the dispute: states do not acquire extraterritorial enforcement authority merely because they consider another state’s legal response inadequate.
If they did, sovereignty would become conditional on whether powerful governments approved of weaker governments’ law enforcement.
International politics already contains uncomfortable exceptions and contested practices around extraterritorial counterterrorism. The United States itself has used lethal force against suspected terrorists in other countries, sometimes with host-government consent and sometimes under controversial legal theories. Russia has repeatedly been accused of targeting opponents abroad. Saudi agents murdered journalist Jamal Khashoggi inside the Saudi consulate in Istanbul.
Western governments therefore cannot discuss the principle as though extraterritorial state violence is something only adversaries practise.
Consistency matters.
If Canada demands that India respect Canadian sovereignty, it should defend the same principle when allies are accused of violating other states’ sovereignty. International rules become more credible when governments apply them according to conduct rather than geopolitical alignment.
India should recognise the same problem from the opposite direction. New Delhi strongly defends sovereignty and non-interference as principles of international relations. India frequently criticises Western governments when they appear to interfere in domestic political questions and has historically emphasised territorial sovereignty within the UN system.
If Canadian allegations were substantiated, involvement in a killing inside Canada would therefore conflict directly with principles India itself invokes.
A state cannot reasonably demand strict sovereignty when others comment on its internal affairs while adopting a looser interpretation when pursuing perceived enemies abroad.
Again, that conclusion depends on the evidence.
The geopolitical context makes the dispute especially awkward because India has become increasingly important to Western strategy. Washington wants deeper cooperation with New Delhi on China, technology, defence and supply-chain diversification. Canada is a NATO member and one of America’s closest allies. The United States therefore has strong incentives to avoid choosing between two important partners.
This creates a test of the US-India partnership discussed in June. Strategic relationships are easiest when interests align. Their maturity becomes visible when a partner is accused of conduct the other considers unacceptable.
If Washington minimises credible evidence because India is strategically useful against China, its claims about a rules-based international order become less convincing. If it treats an unresolved allegation as established because Canada is an ally, India will reasonably question whether strategic partnership provides any insulation from premature judgement.
The appropriate response is institutionally boring but politically difficult: cooperate with investigation, examine intelligence and allow evidence to determine attribution.
Rules-based order requires exactly this kind of discipline.
India’s growing power makes consistency increasingly important. New Delhi has spent years arguing that the international system should give India greater influence because existing institutions do not reflect contemporary distributions of power. That argument is persuasive.
Greater power brings greater scrutiny.
A country seeking permanent membership of the UN Security Council and a larger role in global governance will increasingly be judged not only by whether international rules protect Indian interests but by how India behaves when those rules constrain Indian options.
This is part of becoming a great power.
Strategic autonomy cannot mean freedom from consequences. It means possessing enough capability to make independent choices while accepting responsibility for those choices.
Canada faces its own institutional questions. Trudeau’s government needs eventually to provide enough evidence that the public allegation can be evaluated. Intelligence cannot always be disclosed fully without revealing sources and methods, but accusations of state-sponsored assassination are too serious to remain indefinitely at the level of governmental assertion.
Trust cannot replace evidence permanently.
The stronger the accusation, the stronger the obligation to demonstrate its basis through appropriate legal or investigative processes.
Otherwise the dispute risks becoming entirely political. Canadians predisposed to trust Trudeau believe Canada. Indians predisposed to distrust him believe India. Evidence becomes secondary to national identity.
That would be the worst informational outcome because each government could mobilise domestic support while the underlying factual question remained unresolved.
Nationalism creates powerful incentives in disputes like this. Trudeau cannot appear indifferent to the possibility that a foreign government killed a Canadian citizen on Canadian territory. Modi’s government cannot appear to accept an accusation that India conducts assassinations abroad. Both therefore face domestic political costs for backing down.
The bilateral dispute can consequently develop its own escalation ladder.
Canada expels a diplomat.
India expels one.
India suspends visas.
Canada considers further responses.
Each action creates pressure for another action demonstrating resolve.
This resembles military signalling in a much less dangerous domain. Governments want to impose enough cost to defend credibility without damaging the relationship beyond what the underlying dispute requires.
The danger is that retaliation becomes detached from the original objective.
Diplomatic punishment should create incentives for cooperation or signal unacceptable conduct. If each side begins retaliating principally because the other retaliated, escalation becomes self-sustaining.
That would be particularly costly because the social relationship between India and Canada is much larger than the diplomatic dispute. Canada has a substantial Indian diaspora and large numbers of Indian students, migrants and families maintain relationships across the two countries, and bilateral economic ties affect people with no involvement in national-security policy.
Government conflict therefore produces civilian externalities.
Visa restrictions affect students and families.
Diplomatic reductions slow consular services.
Political rhetoric can increase suspicion towards diaspora communities.
This is why leaders need to distinguish governments from populations carefully.
Canadian Sikhs are not collectively responsible for Khalistani extremism.
Indian Canadians are not representatives of the Indian government.
Sikhs in India are not responsible for the politics of diaspora activists abroad.
Collective attribution would transform a diplomatic dispute into a social one.
Diaspora politics itself deserves more attention because migration changes the geography of political conflict. People can leave a country while remaining deeply involved in its politics through money, media, activism and family networks. Political movements defeated or marginalised domestically can survive abroad where legal protections and funding environments differ.
Host governments then face a difficult sovereignty question. Their citizens possess rights under domestic law even when their political activities affect another country.
Diasporas therefore create transnational political space that does not fit neatly into a world organised around territorial states.
India sees activists in Canada influencing Indian security.
Canada sees Canadian citizens exercising Canadian rights.
Both perceptions can be simultaneously genuine.
The solution cannot be one country simply imposing its interpretation on the other. It requires legal cooperation precise enough to distinguish protected expression from conduct that crosses into criminality.
That means evidence-sharing, extradition processes where appropriate, financial investigations and protection of diplomats from genuine threats.
These mechanisms are slower than unilateral action.
Their slowness is intentional.
Legal procedures exist partly because governments are not supposed to be final judges of accusations involving their own security interests.
The current dispute therefore returns to a principle running through many of these posts: institutions matter most when governments believe bypassing them would be easier.
India may believe Canadian institutions have responded inadequately to extremist activity.
Canada may believe Indian institutions are unwilling to address its allegation.
The answer cannot be abandoning institutions entirely.
It has to be creating enough credible cooperation that each state believes the other’s legal processes can produce meaningful outcomes.
That may be difficult given how quickly trust has deteriorated this week.
But the alternative is a world in which states increasingly police political opponents across borders according to their own definitions of threat.
For dissidents, activists and diaspora communities everywhere, that would be an extraordinarily dangerous precedent.
Sovereignty protects governments.
It also protects individuals from foreign governments exercising power where they have no legal authority.
That is why the Nijjar allegation matters beyond India and Canada.
If Canada proves that Indian agents were involved, India should face serious diplomatic and legal consequences because states cannot treat another democracy’s territory as an extension of their own security jurisdiction.
If Canada cannot substantiate the allegation, the political consequences of publicly accusing a major partner will also be serious.
The evidence therefore has to carry more weight than either country’s nationalism.
India’s security concerns deserve investigation.
Canada’s sovereignty deserves respect.
Political advocacy deserves protection.
Violence deserves prosecution.
Those principles are not contradictory if institutions distinguish them carefully.
The dispute becomes dangerous when governments collapse them into one another.
For now, neither Canadians nor Indians should be asked simply to choose which government they believe.
The allegation is serious enough that belief is not enough.
Evidence has to decide.
