Extraterritorial Application of the Charter to Canadian Forces

Cross-referencing: National Security Law (NSL), Ch. 14, pp.541 et seq.; Ch. 2, p.32.

 

It would appear that the Supreme Court’s decision in Hape is creating substantial confusion concerning the extraterritorial reach of the Charter.  In Amnesty International v. Canada, the Federal Court concluded in March 2008 that the Charter did not reach the conduct of Canadian Forces in Afghanistan in their transfer of Afghan detainees to Afghan authorities.  In construing Hape to arrive at this conclusion, the Court’s reasoning reflects some of the uncertainty in that decision and produces confusion about exactly what Hape stands for. 

 

 

Put simply, in Hape, the Supreme Court concluded that the Charter typically does not follow the flag – that is, that it does not generally attach to the extraterritorial conduct of Canadian government actors.  The Court did, however, raise caveats to that conclusion.  Consent of the foreign state to the application of the law is an obvious exception.  But so too is what the Court called “some other basis under international law” (at para. 65).  The difficulty in deciding what those other bases are stems from the Supreme Court’s rather unpersuasive approach to prescriptive and enforcement jurisdiction in international law. 

 

 

Put simply, extraterritorial enforcement jurisdiction is impermissible, without consent of the territorial state.  Extraterritorial prescriptive jurisdiction is entirely appropriate, so long as justified on one or more of several recognized grounds.  These latter include, for example, the nationality principle – states can and do legislate what it is their nationals can do internationally.

 

The Supreme Court acknowledges this distinction (see paras 60 et seq.).  The Court then muddies the issue by implying, however, that the application of the Charter must necessarily engage extraterritorial enforcement jurisdiction.  Hape concerned the issue of search and seizures conducted outside Canada and whether they were subject to the Charter’s section 8 protections.  Because Canada cannot dictate how other states conduct their police investigations, it cannot dictate how Canadian police officers cooperating (while overseas) in these investigations act, including in relation to Charter rights (para. 105).

 

But it can.  There is absolutely no bar on interpreting the Charter as a prescriptive tool that precludes Canadian police officers participating in overseas investigations that violate Charter norms.  Here, Canada would not be applying its norms to foreign actors, just asking its own nationals to abstain from overseas behaviour inconsistent with the Charter.  There is no difference, in other words, between this approach and that the Court invokes as proper in its own hypothetical: a law barring extraterritorial smoking by Canadian citizens.  Nor is there any difference between this approach and, e.g., our Criminal Code provision that bars Canadians abroad engaging in sexual tourism.  There is no issue here of inappropriate extraterritorial enforcement jurisdiction, unless Canadian authorities arrive on foreign streets to physically apprehend offenders.  (Nor, for that matter, are there any serious comity concerns – only if the foreign law demanded that people smoke or engage in sexual tourism would there be a conflict of law giving rise to comity preoccupations.)

 

It is also notable that the Court did say that “deference [to foreign law and international comity] ends where clear violations of international law and fundamental human rights begin” (at para. 52) and that Canadian officers may not participate “in activities that, though authorized by the laws of another state, would cause Canada to be in violation of its international obligations in respect of human rights” (at para. 90).  It also reaffirmed that  “[i]n interpreting the scope of application of the Charter, the courts should seek to ensure compliance with Canada’s binding obligations under international law where the express words are capable of supporting such a construction” (at para. 56).

 

From this discussion, three points may be drawn.  First, Hape suggests that the Charter generally will not apply extraterritorially.  Second, a straightforward reading of prescriptive jurisdiction rules means that the Charter could be construed as requiring Canadian government actors to desist from extraterritorial actions violating Charter norms.  Third, at the very least, they have to desist if in participating, they would violate Canada’s human rights obligations.  So at minimum, the Charter reaches as far as Canada’s international human rights obligations reach.  It follows that if Canada’s international human rights obligations extend extraterritorially, so too does the Charter.

 

As noted, Hape concerned search and seizures conducted outside Canada and whether they were subject to the Charter’s section 8 protections.  Here, no real questions of compliance with Canada’s human rights obligations were truly engaged.  The section 8 protections against search and seizure are more robust than anything found in the International Covenant on Civil and Political Rights (ICCPR), for example.

 The Amnesty International case is entirely different, and rests on section 7: the protection against deprivation of life, liberty and security of the person.  Particularly of note is the right to be free from torture and not removed in circumstances where there are reasonable grounds to believe one will be tortured.  These rights are amply guarded in international treaties of which Canada is a party (the ICCPR, Art. 7 and the UN Torture Convention, Art.3) and at customary international law.

 

The Amnesty International case should, therefore, hinge on whether these international obligations extend to Canadian soldiers operating in Afghanistan.  As discussed in NSL, Ch. 2, p. 28 and Ch. 14, pp. 542-43, there is sound reason to believe that they do, although the matter is not without some doubt.  The test (at least under the ICCPR) hinges on the degree of control exercised by the state’s forces.  In the words of the UN Human Rights Committee: “a State party must respect and ensure the rights laid down in the Covenant to anyone within the power or effective control of that State Party, even if not situated within the territory of the State Party.”  U.N. Hum. Rts. Comm., General Comment 31, para. 10, U.N. GAOR, 59th Sess., Supp. No. 40, Vol. 1, at 175, 177, U.N. Doc. A/59/40 (2004).

 

The Federal Court had a very different view of the standards to be applied.  Unfortunately, the Court focused on the Supreme Court’s admittedly murky discussion about whether extraterritorial application of the Charter is a matter of prescriptive jurisdiction or whether it also necessarily involves enforcement jurisdiction.  Since extraterritorial enforcement jurisdiction is impermissible without consent, the Court focuses its inquiry on whether this consent is forthcoming from Afghanistan.  Responding in the negative, the Court then considers whether an exception to this consent rule should be applied because the actor here was the Canadian military (often deployed without consent).  Apparently, the applicants urged that because the military was in a position to assert effective control over persons in their custody (unlike a situation where Canadian police are simply assisting foreign authorities), the Charter should apply.

 

This is the right test in the wrong place.  As noted above, the effective control test is the measure of whether Canada’s obligations under the ICCPR extend extraterritorially.  The logical sequence should be: Charter applies where international law applies; international law applies where there is effective control.  This approach would eliminate the confusion that ultimately flows from the Court’s reasoning.

 

In this last respect, the Court casts doubt on whether “effective control” is really an exception to Hape (correctly, because its importance is not as a direct exception, but rather than the metre stick of whether international law applies or not).  More importantly, it later asserts in a review of international law disconnected from the effective control discussion that “[i]t cannot be that the Charter will not apply where the breach of a detainee’s purported Charter rights is of a minor or technical nature, but will apply where the breach puts the detainee’s fundamental human rights at risk … That is, it cannot be that it is the nature or quality of the Charter breach that creates extraterritorial jurisdiction, where it does not otherwise exist. That would be a completely unprincipled approach to the exercise of extraterritorial jurisdiction” (at para. 310-11).

 

One understands this concern.  But it is exactly this variable standard that international law and Hape (in my view) dictate: the Charter will follow the flag to the extent that Canada’s international human rights obligations also follow that flag.  And because those human rights obligations hinge on the effective control test, the result may be a patchy reach of the Charter to the overseas activities of Canadian officials.

 

To compound the difficulty, I have argued elsewhere that the extraterritorial reach of the ICCPR and the Torture Convention are not necessarily identical.  See Craig Forcese, “A New Geography of Abuse? The Contested Scope of US Cruel, Inhuman and Degrading Treatment Obligations,” (2006) 24 Berkeley Journal of International Law 908-938.

 

But at core, this case (on appeal) stands to be won or lost on the effective control test.  Setting aside the confusion about where that test should be situated, with respect, the Federal Court gives rather short shrift to the international authorities endorsing this test as the measure for the extraterritorial reach of the ICCPR and its closest analogues. 

 

Also left unaddressed in this case (presumably because it was not argued) is the customary international argument.  If, as Hape affirms, customary international law is the common law of Canada, is there not a simple common law argument to be made that Canada cannot transfer detainees where there are reasonable grounds to fear torture?

 

 

 

 

 

 

Subsequently, if the extraterritoriality hurdle is ever overcome, other legal questions will press.  For example, what exactly are the substantive obligations not to transfer to torture under ICCPR Art. 7 and Torture Convention Art 3 (or under customary international law)?  And have they been met by, inter alia, the Canadian-Afghan transfer arrangement?