One Law's Unprivileged Belligerent is another Law's Terrorist: Potential Legal Wrinkles in the Edmonton Arrests

NSL, Ch. 7, pp. 273 et seq.

 

The Globe and Mail is reporting the arrest today of an Edmonton man for "helping organize a pipeline for foreign suicide bombers who were intent on blowing themselves up in Iraq."  Specifically, the arrest was made in response to a US criminal complaint, alleging that the man conspired to kill US military personnel in Iraq over the last several years, culminating in a serious and lethal bombing attack in Iraq.  The man is now subject to extradition to the United States.  Canadian officials are quoted as saying that the extradition will be swift, so long as uncontested.

 

This blog focuses on whether this extradition might plausibly be contested (based on the reported facts). 

 

Double Criminality

 

The core precept of extradition law is double criminality.  "Double criminality requires that the conduct upon which the extradition request is made be criminal in both the requesting and requested countries," ( United States of America v. Barbu, 2010 ONCA 891 at para. 16) and is codified in section 3 of the Extradition Act.  In Canadian law, "it is not necessary that the Canadian offence described in the ... committal order “match” the foreign offence for which the person is sought or surrendered in name or in terms of its constituent elements; it is 'the essence of the offence' that is important on the conduct-based approach" (Canada (Justice) v. Fischbacher, 2009 SCC 46 at para. 29.

 

The US charges are: "conspiracy to murder US nationals" overseas; and, "material support for terrorists".  At issue, therefore, is what Canadian offences reflect "the essence" of these offences. 

 

Conventional Crimes

 

Most Canadian Criminal Code offences are territorial in scope -- that is, they relate to crimes committed within the terroritory of Canada.  Only in exceptional cases does Canadian criminal law extend to extraterritorial conduct.  A number of terrorism offences are extraterritorial in scope, a matter I return to below.  Still, even if the offence is a coventional crime that is not extraterritorial in scope, if enough of its constituent elements arise in Canada (even if the crime in ultimately culminated oversees) it is amenable to prosecution in Canada.  In the Supreme Court's words in Libman v. The Queen, [1985] 2 S.C.R. 178 at para. 74: "all that is necessary to make an offence subject to the jurisdiction of our courts is that a significant portion of the activities constituting that offence took place in Canada. As it is put by modern academics, it is sufficient that there be a  'real and substantial link' between an offence and this country, a test well‑known in public and private international law".  In the result, a conspiracy to murder a person, where the conspiracy is partially pursued in Canada, although the murder is to take place overseas, should prosecutable as plain vanilla conspiracy to murder under the Criminal Code.

 

In these circumstances, the double criminality element is satisfied -- we have a close analogue to the US crime -- and the Edmonton arrestee is extraditable.  The problem is that the US complaint is a bit ambiguous on the geography of the conspiracy -- it is not clear if the man was conspiring in Canada or abroad.  The description of the facts supporting the material support charge do seem to have more detail, and do suggest that the accused was communicating from Canada.  If these same facts undergird the conspiracy charge, then this implies that the Libman test is satisfied.  If not, then life becomes more complicated.

 

Terrorism Crimes

 

Moving into the complicated part: Given the nature of the alleged offence, terrorism crimes seem the more natural fit to the charge.  The Criminal Code is now replete with terrorism offences, the most relevant of which are extraterritorial in scope.  That is, they are not confined to terrorist actions in Canada.  On their face, therefore, these are offences that cover the "essence" of the US charges, meeting the double criminality requirements.

 

However, terrorism offences are tied, ultimately in almost all instances, to the Criminal Code's definition of "terrorist activity".  And that definition excludes from its ambit "an act or omission that is committed during an armed conflict and that, at the time and in the place of its commission, is in accordance with customary international law or conventional international law applicable to the conflict".  I have written about the so-called "armed conflict exemption" elsewhere on this blog as well as in National Security Law.  A couple of points are worth making about it.

 

"Armed Conflict"

 

The exemption is clearly intended to echo equivalent (but not identical) concepts found in international terrorism treaties, a point the Canadian courts have made.  (See, e.g., R. v. N.Y., 2008 CanLII 24543 (ON S.C.) at para. 12, approved R. v. Khawaja, 2010 ONCA 862 at para. 159)  Its language implicitly invokes principles of international law, and specifically the laws of war, where the concept of "armed conflict" is a term of art.  (It is not a term of art in Canadian law).  In international law, there are two sorts of armed conflict: international and non-international.  These are discussed in National Security Law.  Of note for the purposes of this blog, an non-international armed conflict arises in a limited series of circumstances, perhaps best described as a sufficiently protracted armed struggle between a government and a nongovernmental group, the latter with a degree of organization and control over territory.  Put another way, the classic expression of a noninternational armed conflict is a civil war.

 

In Iraq's recent history, it has indisputably been the theatre of a violent noninternational armed conflict between the Iraqi government (supported by US and other foreign troops) and a sometimes quick sophisticated (in terms of organization) insurgency.  Whether the periodic unrest that persists in Iraq still amounts to an armed conflict is less certain.  A noninternational armed conflict does not exist, to use the language of Additional Protocol II of the Geneva Coventions, "in situations of internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature".

 

"During an armed conflict"

 

The armed conflict exemption applies "during an armed conflict".  One read of "during" might be "while participating in".   Put another way, "during" is limited to circumstances where the acts took part as part of the armed conflict itself.  Another read of "during" is "at a time when an armed conflict was occurring".  Unfortunately (I think), the Ontario Court of Appeal offered up the latter reading of "during" in its recent Khawaja decision, at para. 165: "The availability of the exception does not require proof of an accused’s physical presence in an area of armed conflict.  Nor does it contemplate that an accused’s impugned acts or omissions must be carried out within the territorial limits of an area of armed conflict.  As we have already said, all that is required to trigger the exception is some evidence that: (1) an accused’s acts or omissions were committed “during” an armed conflict; and (2) those acts or omissions, at the time and at the place of their commission, accorded with international law applicable to the armed conflict at issue."

 

It is not clear why the Court of Appeal preferred this construction (or that the Crown apparently conceded this point).  I think it is wrong.  The international analogues to this provision are found in, e.g., the International Convention for the Suppression of Terrorist Bombing, G.A. Res. 164, U.N. GAOR, 52nd Sess., Supp. No. 49, at 389, U.N. Doc. A/52/49 (1998), entered into force May 23, 2001.  Article 19(2) of that instrument reads: "The activities of armed forces during an armed conflict, as those terms are understood under international humanitarian law, which are governed by that law, are not governed by this Convention".  There are several important differences between this language and that of the Criminal Code-- I shall return to some of these.  But of immediate note is the implication of speaking of "armed forces during an armed conflict".  An activity of an armed force during an armed conflict that is governed by international humanitarian law (IHL) is one that, by definition, takes place in or at least as part of the armed conflict.  That is, the actions exempted are actions that are covered by IHL, making this exemption a choice of law provision of sorts (to wit, you are to apply IHL and not the terrorist bombing convention).  This is a much narrower read than that given to the "during armed conflict" concept by the Ontario Court of Appeal. 

 

The Canadian exemption also differs from its international analogue in failing to restrict its application to an "armed force", a reasonably well understood concept in international humanitarian law.  So, while in international law, to benefit from the exemption, the act that occurs during the armed conflict must be by an armed force -- in essence, a state military -- in Canada, anyone and anything may lay claim to the exemption. 

 

Combined Effect

 

If you combine the broad availability of the Canadian exemption with the broad read of "during" offered by the Ontario Court of Appeal, you have a ready made defence for every aspiring guerrilla, insurgent, freebooter or terrorist inclined to do something in Canada in favour of their preferred violent cause overseas: they needn't be an "armed force" and all they have to do is do their wrongful act at the same time (and presumably be motivated by) the foreign armed conflict.  This is a truly concerning possibility given the historical reality that a lot of  the activity that is now penalized by Canadian terrorism provisions has been fuelled by grievances stemming from foreign disputes.

 

The Only Backstop to Open Season

 

The only thing that reigns in the scope of the Canadian exemption (as construed by the Ontario Court of Appeal) is the second prong of the exemption test: the impugned act, to benefit from the exemption must "at the time and in the place of its commission, [be] in accordance with customary international law or conventional international law applicable to the conflict".  That is, it must comply with the Geneva Convention, their additional protocols and/or the customary equivalents and those other aspects of international law (such as human rights law) that may persist in applying "during the armed conflict". 

 

Note that here the Canadian exemption is more rigorous than its international analogue: the acts must be in accordance with this international law, and not merely governed by it.  So a failure to comply, e.g., with the laws of war precludes the application of the exemption.  It is exactly on this basis that the Ontario Court of Appeal rejected the exemption's application to Mr. Khawaja: "We agree with the Crown’s submission that the appellant’s own emails belie any suggestion that he viewed the violent Jihad that he espoused as lawful.  Contrary to the central tenets of international humanitarian law, he made no distinction between civilians and soldiers or lawful combatants.  Indeed, his emails reveal that he supported the indiscriminate and random murder of civilians, for instance, by means of a suicide attack in a crowded public venue in Israel or by terrorist attacks similar to those in the United States on September 11, 2001."

 

The Armed Conflict Situation and the Edmonton Case

 

All of this means, of course, is that an accused who directed violence against an armed force in an armed conflict in a manner that didcomply with international humanitarian law (for instance, targeting only combatants and not non-combatants, among other things) might be entitled to invoke the exemption.  This may be an extremely rare occurrence -- indiscriminate violence, for instance, and violence directed as much against civilians as soldiers was commonplace in Iraq.  So the facts in the Edmonton case may be uncertain -- but the possibility that the violence was more focused on military targets than, say, in the Khawaja case, is raised by the nature of the allegations against the accused in the US complaint.

 

A simple way out of this conundrum might be simply to say that even if the combatants in the armed conflict otherwise comported fully with IHL, the mere fact that the accused was an unprivileged belligerent -- essentially a civilian that had participated in the conflict outside of the circumstances where such things are permissible in IHL -- is enough to render the accused in non-"accordance" with IHL.  Personally, I think this is entirely a plausible construction of the law: to be in accordance with IHL requires more than "don't commit a war crime".  It also means honouring those aspects of IHL which give the monopoly of lawful use of armed violence to a handful of actors (and almost never to civilians).

 

Summary of Implications

 

So in sum, here are some possible issues that may arise in the Edmonton case (another of my predictions bound to founder, I suspect):

 

1. Can the government satisfy the dual criminality requirements through the application of the Libman test to a conventional Criminal Code conspiracy charge? If yes, life is simplified.  If no, it is not.

 

2. If the dual criminality requirement is to be satisfied by reference to a terrorism offence, can the government argue that the "essence" of the Canadian equivalent to the US crime does not require consideration of the armed conflict exemption -- that is, the exemption is a particular defence that is not related to the "essence" of the offence and does not need to be considered in deciding whether the Edmonton man is extraditable.  (I note that the US terrorism definition does not appear to have an equivalent to the armed conflict exception).

 

3. If instead the armed conflict exemption describes the limit of the actual overarching offence -- that is, it confines the reach of the offence in a manner that goes to its "essence"-- then there will be a question about whether the US charge really does relate to actions in an armed conflict and whether the conduct in question was, in fact, consistent with how force can be used in that armed conflict.  If the answer is yes, at the time of the incidents there was an "armed conflict" (not a certain thing for 2009 and 2010 incidents in Iraq, it seems to me) and yes, force was used in a manner that was in accordance with IHL, then the exemption could apply to annul the dual criminality requirement.  By this point, we have entered the murky and uncomfortable world of deciding whether being one law's (IHL) combatant in an armed conflict against a close friend and ally precludes being a terrorist under Canadian law. 

 

At the end of this day, and as suggested above, I think the best solution is to hold that being an unprivileged belligerent under the laws of war is enough to render the exemption non-applicable.  Any other construction makes the Canadian armed conflict exemption sweeping.  But by the time we reach this decision point, we may be several months -- if not years -- into a complicated extradition case.