About this Project

This blog comments on Canadian (and occasionally comparative) national security law to update my National Security Law textbook and now also my 2015 book, False Security: The Radicalization of Anti-terrorism, co-authored with Kent Roach.

Please also see www.antiterrorlaw.ca for Bill C-51-related analyses by Craig Forcese and Kent Roach.

For narrated lectures on various topics in national security law, please visit my 2017 "national security nutshell" series, available through iTunes.


For a continuing conversation on Canadian national security law and policy, please join Stephanie Carvin and me at A Podcast Called INTREPID.


Please also visit my archive of "secret law" in the security area.

By Craig Forcese

Full Professor
Faculty of Law

Email: cforcese[at]uottawa.ca

Twitter: @cforcese


National Security Law Blog Search
Subscribe to National Security Law Blog

Best Law School/
Law Professor Blog Award


Most Recent Blog Postings

Afghan Detainees and the Charter: Another Perplexing Treatment by the Federal Court

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


Laws may not silent in times of war, but the Charter is (at least if the war is a foreign one).  At core, this is the implication of the Federal Court of Appeal's December 2008 decision in Amnesty International v. Minister of National DefenceThe Federal Court of Appeal upheld a Federal Court decision, a judgment discussed elsewhere in this blog.  It did so based on a reading of the Supreme Court's decision in Hape that seems impossible to defend in light of that Court's further decision in Khadr.


In my view, Hape and Khadr stands for the following proposition: absent consent by the foreign state, the Canadian Charter of Rights and Freedoms does not attach to the overseas actions of Canadian officials except where those actions are contrary to Canada's international obligations (and more specifically, its international human rights obligations).  The threshold question is, therefore, whether the Canadian officials are acting in compliance with Canada's international human rights obligations.  If they are not, then the Charter applies.


The analysis that should have been undertaken by the Federal Court of Appeal  concerns, therefore, the scope of Canada's international human rights obligations in relation to Canadian Forces practices on Afghan detainees.  There, the inquiry must focus on the reach of instruments such as the International Convenant on Civil and Political Rights, the Convention Against Torture -- and there is a jurisprudence on the Covenant in particular that discusses the extraterritorial reach of human rights obligations to state agents acting beyond their borders -- international humanitarian law, and the issues of lex specialis that governs the interaction between this international humanitarian law and the aforementioned human rights treaties.  The question of "effective control" by the CF of Afghan territory, which has appeared in the Federal Court case as a separate ground for asserting the application of the Charter, is not a separate ground.  Instead, "effective control" is a short-hand for the test developed in the international jurisprudence to decide when the Covenant reaches the extraterritorial actions of states bound by it.


So, in sum, here is the analysis: The Charter applies if Canada's international human rights obligations apply.  Canada's international human rights obligations apply either because some norm of international human rights or humanitarian law reaches the conduct of the CF in Afghanistan.  Humanitarian law obviously does -- the Geneva Conventions and their customary equivalents attach to the armed conflict and extend georgraphically to the entire theatre of conflict.  Human rights law may, (a) so long as it is not displaced by the lex specialis of IHL (and, in a case concerning conditions of detention and treatment, there likely isn't any displacement, in my view -- lex specialis requires a contradiction between the two areas of law, not just overlap) and (b) so long as on the face of those treaties (and customary norms) which govern human rights law, there is extraterritorial reach.  On the latter point, the question then terms to the level of control exercised by the CF.


The Federal Court of Appeal, however, launches down a path of its own, denying that Hape and Khadr mean much at all when it comes to tying the Charter to Canada's international obligations.  Its most puzzling statement is this one: "I understand the Supreme Court of Canada to say that deference and comity end where clear violations of international law and fundamental human rights begin. This does not mean that the Charter then applies as a consequence of these violations."  But what else could the following statement from the Supreme Court in Khadr possibly mean:


[18] In Hape, however, the Court stated an important exception to the principle of comity.  While not unanimous on all the principles governing extraterritorial application of the Charter, the Court was united on the principle that comity cannot be used to justify Canadian participation in activities of a foreign state or its agents that are contrary to Canada’s international obligations.  It was held that the deference required by the principle of comity “ends where clear violations of international law and fundamental human rights begin” (Hape, at paras. 51, 52 and 101,  per LeBel J.).  The Court further held that in interpreting the scope and application of the Charter, the courts should seek to ensure compliance with Canada’s binding obligations under international law (para. 56, per LeBel J.). 

[19] If the Guantanamo Bay process under which Mr. Khadr was being held was in conformity with Canada’s international obligations, the Charter has no application and Mr. Khadr’s application for disclosure cannot succeed: Hape.  However, if Canada was participating in a process that was violative of Canada’s binding obligations under international law, the Charter applies to the extent of that participation.


A search and replace of para 19 would read:


If the detention process under which Afghan detainnes are being held was in conformity with Canada’s international obligations, the Charter has no application and the Afghan detainee's application for Charter relief cannot succeed: Hape.  However, if Canada was participating in a process that was violative of Canada’s binding obligations under international law, the Charter applies to the extent of that participation.


It should be as simple as that.  We'll see what the Supreme Court does on appeal.


Special Advocates: The Continued Saga on Continued Communication

Cross-referencing: National Security Law (NSL), Ch. 10, pp. 414 et seq.


As noted in past posts, an issue of contention in Canada's new "special advocate" process for immigration security certificates is the capacity of special advocates to persist in communicating with the named persons once the former have seen the secret information.  See, e.g., December 13, 2007 post.  In essence, such communications must be approved by a judge.  Indeed, on the wording of the Act, all special advocate communications concerning the "proceeding" (and not just those that actually involve secret information) must be approved by a judge.   


Not surprisingly, special advocates and public counsel for the named persons in the current round of security certificate cases view these constraints as a) overbroad and b) greatly impairing of their functions.  (In the interests of full disclosure, I must report that I agree with them, and so attested in an affidavit ultimately filed in the motion discussed below).


In terms of overbreadth, the restrictions on communications lack nuance: communication about the proceeding could include telling one's spouse that one is in court, or a parliamentary committee that one is unhappy with the workings of the process. 


In terms of impairing special advocate functions, first, there is good reason to believe that carefully worded questions to named persons that reveal no secrets can elicit information material to the outcome of proceedings.  That has been the experience of some special advocates, acting in prior capacities as legal agents in proceedings before the Security Intelligence Review Committee (and the view shared by commission counsel, now special advocates, in the Arar inquiry).  For more on this point, see the report on the UK special advocate system Lorne Waldman and I prepared in 2007.


Second, special advocates are obviously concerned that seeking advance clearance from judges (possibly in the presence of government counsel) on what questions they will ask of the named person will be prejudicial -- a special advocate searching out exculpatory detail will presumably have to explain how it is given questions could do that (without revealing secrets to the named person).  If, subsequently, permission is given, and no exculpatory result produced, the judge (and government lawyer if present) will be keenly attuned to the fact that the questions were asked and nothing useful produced.  If anything, this would have the affect of bolstering the government's case (and the judge's acceptance of it).  There's a reason solicitor-client confidentiality in conventional legal relationships includes not just answers given, but questions asked.  For exactly these sorts of reasons, in the UK, special advocates have for the most part declined to seek permission to ask questions on substantive matters because of the need to produce these in writing, vetted by tribunal and government.  (The issue of whether seeking permission from the tribunal can be done on a ex parte basis is currently before the English courts).


Armed with these concerns, public counsel in Almrei mounted a constitutional challenge to the new communications regime, one joined by other named persons and the special advocates as a class.  In early November, Justice Lutfy of the Federal Court declined to address the constitutional issue, determining it to be premature because an insufficient factual record existed -- specifically, no one had actually asked for continued communication and been denied.


Again, readers should be conscious that I am an affiant with a declared view on this matter in reviewing my assessment of this decision.


First, to his credit, Justice Lutfy cut away at least some of the perversity caused by the overbroad communications rules, through an exercise in statutory interpretation.  So special advocates can communicate with their staff, colleagues and family members (and hopefully anyone else!) about their whereabouts.  Special advocates can communicate with the special advocate support unit at Justice concerning their resource and administrative needs.  (But note, since communication is confined to resource and administrative needs, it is unclear whether we can move to a UK style system where the support office is privy to the secret information and can provide analytic support on the secret side of the ledger).  And special advocates on the same case can talk to each other.


On other issues, however, the decision resolves little.  First, the status quo -- communications that concern the proceedings, even if not dealing with secret information is impermissible without advance vetting by judges -- really does put special advocates at a disadvantage: the two special advocates assigned to each case stand in splendid isolation, unable to bounce ideas off other special advocates or lawyers on issues far removed from secret information (such as legal arguments) without pre-clearance from a judge.  Note the Court's sweeping conclusion: "special advocates cannot communicate with another person about the proceeding, absent judicial authorization, even concerning an order or direction made public by the presiding judge."  While special advocates could find themselves isolated in their efforts to respond to this (public) order, government lawyers presumably caucus at will.  This is hardly the equality of arms vital to a functioning adversarial system.


Second, prematurity seems to have no place in this scenario: at issue (at least in part) is whether special advocates should be obliged to go to judges in the first place.  That question is more than ripe.  To ask special advocates to go to a judge first, ask permission, be refused and only then be able to challenge whether they should have to go to a judge in the first place at best delays resolution of this issue (and at worst may arm the government with arguments on mootness and waiver, or some such thing).


Third, Justice Lutfy -- whose experience includes practical involvement with SIRC and therefore should attract substantial respect -- was unpersuaded that SIRC counsel's freedom to communicate was as ample as asserted by the applicants (and by this affiant).  In the Court's view, SIRC counsel is subject to the direction of the presiding SIRC member.  It only stands to reason that special advocates, in the even more emphatic scheme created by Parliament, should be subject to direction by a judge.  I confess, however, to having some doubt that the level of supervision exercised by the SIRC member comes close to the advance permission for questions we seem to be moving towards in the Federal Court -- certainly, no one I have spoken to who has participated in the SIRC process has described such close control.  (And there was nothing on this motion record that suggested that there is such close control).


In any event, what Parliament has done with the special advocate system (that differs from SIRC) is to create a closer relationship -- maybe best described as proto-solicitor-client -- between special advocate and named person.  That relationship -- which does not exist in the SIRC context -- is what creates particular difficulty in advance vetting of special advocate questioning by the (same) judge who serves as the ultimate arbiter of law and finder of fact.


The difficulty appears to lie in reconciling this proto-solicitor-client relationship (and the independence it connotes, quite different from that of SIRC counsel, as representative of SIRC) and wariness about disclosure (however involuntary) of secrets.  Justice Lutfy seems persuaded that this circle can be squared by individualized advance scrutiny of questions by judges, possibly in the absence of government lawyers (that is, on an ex parte basis).  That leaves intact, however, the concerns about chatting about questions with the actual decision-maker you need to persuade if you receive (or not) answers to those questions.


The compromise that takes us out from under this conundrum is to have the question of continued communication decided by a sort of "case management" judge, and not the judge hearing the case.  That resolves the problem of tainting the mind of the actual decision-maker by revealing questions while at the same time allowing the external supervision Justice Lutfy views as desirable.  If the managing judge refuses continued communication, that then becomes the appealable issue (and it would be best to treat it as immediately appealable since what follows could turn on this matter).


The security certificate process has, for a long time, evolved in fits and starts, with compromise piled on compromise and without a tremendous amount of careful advance design.  (On this point, more blame lies with the government in proposing and Parliament for legislating half-measures than with the Court in grappling with them).  But advance avoidance of train wrecks is in the interest of all concerned.  For this reason, the time is more than ripe for the Federal Court to take up the invitation proferred by Parliament in the amended immigration law, s.85.6(1):


"The Chief Justice of the Federal Court of Appeal and the Chief Justice of the Federal Court may each establish a committee to make rules governing the practice and procedure in relation to the participation of special advocates in proceedings before the court over which they preside."


A Tale of Two Disclosure Regimes: the Aftermaths of Charkaoui II and MB

Cross-referencing: National Security Law (NSL), Ch. 10, pp. 404 et seq.


The senior courts of two common law jurisdictions have contemplated the question of disclosure to parties (that is, public disclosure, as opposed to disclosure during the secret proceedings to judge/special advocate) in the adjudication of national security cases in front of administrative tribunals.  In the past, this blog has discussed the House of Lord's 2007 decision in MB.


More recently, the Supreme Court of Canada issued its decision in June 2008 in Charkaoui II.  At issue in this case was whether affirmative disclosure obligations exist for the Canadian Security Intelligence Service (CSIS), when CSIS information is at issue in immigration security certificate proceedings, and if so, what are those obligations.  Cutting to the chase, the Court held that disclosure obligations do exist.  Since section 7 of the Charter is engaged by security certificates, fundamental justice attaches, and that fundamental justice (on the particulars of this situation) requires ample disclosure, albeit disclosure attuned to bona fide secrecy concerns.  (Unfortunately, we don't have a clear sense yet from the Court as to whether there is a constitutional test that governs the actual quantum of disclosure to the named person. I have opined elsewhere that the Canadian security certificate standard of disclosure treats national security as the only consideration, and on paper at least imports no balancing of national security against other interests such as a fair trial, unlike the Canada Evidence Act procedure or my reading of MB.  A frontier question is whether a balancing is a constitutional necessity.  Note that what MB really means is becoming less clear with time -- as discussed below the English Court of Appeal seems to have rejected the idea that MB establishes at least a basement floor of disclosure. )


Even more notably in Charkaoui II, to preserve a meaningful disclosure obligation, the Court's decision imposes an information retention obligation on CSIS -- no more destruction of notes, for example, "when conducting investigations that are not of a general nature” –- which appears to mean “whenever CSIS conducts an investigation that targets a particular individual or group”.  This latter holding is probably the most revolutionary, and will probably change the way CSIS operates in manifold manners.


Charkaoui II orders are now being issued by Federal Court judges in the current round of security certificate cases.  See, for example, the order issued in Harket in September 2008See also Mike Larsen's interesting discussionof this order.  This order obliges disclosure of "all information and intelligence related to Mohamed Harkat including but not limited to drafts, diagrams, recordings and photographs in CSIS’s possession or holdings with the designated proceedings section of the Court."  Once produced, this information will presumably be vetted by the judge (and special advocate) and released (or not) to the named person per the usual manner in which sensitive is assessed by public disclosure in the security certificate process.  The government has said complying with this disclosure order will take six months.


One doesn't know exactly why complying with this order will take so long.  But on its face, this amounts to a shocking turn of events.  Yes, Charkaoui II underscores a rich disclosure obligation on CSIS and the government.  But, the Federal Courts (and for that matter, the Supreme Court) have for a long time obliged disclosure by the government to the judge in ex parte proceedings like these to a standard of "utmost good faith".  Given how long these security certificate cases have been on the books, one would have thought that there is nothing left to disclose, and that this fact has been verified in "utmost good faith".  The fact that CSIS will require another six months to comply with the order suggests that past disclosure in "utmost good faith" was something much less than full disclosure.


All this tends to affirm the concerns about how disclosure works (or doesn't) in these security certificate cases.  (On this point, Lorne Waldman and I make some observations about the problems of full disclosure in the security certificate context in a forthcoming article in the Supreme Court Law Review -- not least the problem for judges and special advocates of figuring out if there has been full disclosure of intelligence, a form of information that doesn't look at lot like the sort of evidence to which most lawyers are attuned).  I strongly suspect that this is not the last time that the question of disclosure will stir controversy.


Across the Atlantic, the British are struggling to iron out the implications of the House of Lord's 2007 MB decision -- a ruling with lots of different speeches from law lords, not all of which add up to a single easily discerned holding.  In October 2008, the English Court of Appeal dealt with a series of consolidated appeals on the question of MB disclosure.  In this decision, the Court of Appeal reduced MB disclosure to the following (rather ambiguous) principles:


"i) ... the question is whether the effect of the process is that the controlee [this case involved disclosure in "control order" cases] is exposed to significant injustice. In what follows 'fair' and 'unfair' are used in this sense.


ii) All proper steps should be made to provide the controlee with as much information as possible, both in terms of allegation and evidence, if necessary by appropriate gisting.


iii) Where the full allegations and evidence are not provided for reasons of national security at the outset, the controlee must be provided with a special advocate or advocates. In such a case the following principles apply.


iv) There is no principle that a hearing will be unfair in the absence of open disclosure to the controlee of an irreducible minimum of allegation or evidence. Alternatively, if there is, the irreducible minimum can, depending on the circumstances, be met by disclosure of ... very little indeed.


v) Whether a hearing will be unfair depends upon all the circumstances, including for example the nature of the case, what steps have been taken to explain the detail of the allegations to the controlled person so that he can anticipate what the material in support might be, what steps have been taken to summarise the closed material in support without revealing names, dates or places, the nature and content of the material withheld, how effectively the special advocate is able to challenge it on behalf of the controlled person and what difference its disclosure would or might make.


vi) In considering whether open disclosure to the controlee would have made a difference to the answer to the question whether there are reasonable grounds for suspicion that the controlee is or has been involved in terrorist related activity, the court must have fully in mind the problems for the controlee and the special advocates and take account of all the circumstances of the case, including the question what if any information was openly disclosed and how effective the special advocates were able to be. The correct approach to and the weight to be given to any particular factor will depend upon the particular circumstances.


vii) There are no rigid principles. What is fair is essentially a matter for the judge, with whose decision this court should very rarely interfere."


All of which is to say that the quantum of disclosure to the named person necessary to satisfy the fair trial rights varies with the chancellor's foot.


This matter is no doubt on its way back to the House of Lords, where additional clarity would be welcomed.