Printer Friendly

Behind Closed Doors: Secret Law and the Special Advocate System in Canada.


I. The Legislative and Constitutional Framework: of the Special Advocate System

A. Secret Hearings and the Right to a Fair Trial

B. The Special Advocate System in Canada

C. The Constitutionality of the Special Advocate System Reviewed: Harkat v Canada

II. Theoretical Context and Methodology

A. Secret Hearings and the Rule of Law

B. Methodology

III. Rules of Practice and Procedure in the Federal Court: Formal Law, Custom, and Culture

A. Rules of Practice and Procedure in SIRC

B. Rules of Practice and Procedure in the Federal Court

C. Summary

IV. The Conditionality of Special Advocate Powers: Material, Epistemic, and Discursive Barriers

A. Selection and Training of Special Advocates

B. Appointments, Tainting, and Knowledge Sharing

C. Disclosure

D. Administrative Support

E. Summary

V. Trust, Control, and the Limits of the Special Advocate System

A. Communication Between Special Advocates and Named Persons

B. Expert Witnesses

VI. Conclusion


Executive secrecy has a long pedigree in Canadian law, (1) and even deeper roots in English common law. (2) Until relatively recently, it was not often invoked in courts, falling for the most part within the realm of the royal prerogative, cabinet confidences, and select administrative and bureaucratic quarters. (3) The role of courts has grown greatly following 9/11, when dynamics between shifting security practices and constitutional litigation generated more intricate juridical frameworks oriented to rights and the rule of law. (4) Judges now administer hearings concerned in one way or another with sensitive information on a daily basis, with large portions of information being heard in the absence of interested parties (ex parte) and/or the public (in camera). The subject matter of hearings relate to such disparate fields as immigration and refugee law, security intelligence warrant applications, passport revocations, listing of terrorist entities, reviews of classified decisions related to government employment and contracts, and the non-disclosure of evidence in civil and criminal proceedings. (5)

Governments typically justify the non-disclosure of sensitive material on grounds of national security, although information may also be protected for reasons related to national defence and international relations. (6) Secrecy is, of course, the bread and butter of intelligence work, to an extent being necessary to protect the integrity of ongoing (and past) operations, the strategic and tactical value of intelligence, and the safety of officers, human sources, and agents of the state. (7) But it is an altogether different matter to protect the secrecy of information that is used, or should be used, as evidence in judicial hearings. (8) Counterbalancing the interest of the state in secrecy are the rights of affected parties (e.g., to a fair trial, privacy, access to information), the public interest in the open court and open justice principles, and the rule of law. (9)

Formalized in Canada, the United Kingdom, the United States, the Netherlands, Australia, New Zealand, and elsewhere, the framework of secret proceedings are composed of public laws, such as statutes and regulations. But these tend to be vague, for the most part being concerned with controlling conditions of disclosure while conferring upon judges a broad discretion over rules of practice and procedure. (10) In some instances, such as those concerned with security intelligence warrant applications, affected parties are unaware the proceeding ever took place. In others, a proceeding is akin to a trial, insofar as an affected party is informed of an allegation against him and has an opportunity to participate in open proceedings that run parallel to closed proceedings.

Secret hearings are, to say the least, hard to square with rights and the rule of law. (11) Legislatures have made efforts to provide substitutes for disclosure and adversarial challenge, albeit in many instances only after appellate courts have required improvements as a matter of constitutional or international law. In Canada and the UK, these substitutes have come in the form of a "Special Advocate system", where Parliaments have allowed for the assignment of security-cleared counsel to represent the interests of affected parties during ex parte, in camera hearings. Some are satisfied that Special Advocate (SA) systems vindicate constitutional law; (12) others are less sanguine, insisting that secrecy of any kind is inherently incompatible with procedural fairness and the rule of law. (13)

As academic and professional debates continue, high courts have upheld the legality of SA systems. The Supreme Court of Canada did so in the 2014 case of Canada (Citizenship and Immigration) v Harkat, (14) while the SA system in the UK has been endorsed or upheld by both UK courts and the European Court of Human Rights. (15) Indeed, these judgments have permitted the UK's Parliament to expand closed material proceedings to new settings. Beginning with immigration proceedings, (16) SAs have since been used in proceedings related to: the listing of terrorist organizations, (17) control orders, (18) terrorism prevention and investigation measures, (19) asset-freezing orders, (20) and civil proceedings where national security information might be disclosed. (21)

But one of the curious features of these cases is that high courts know little more than excluded parties, the public, and academia about the internal workings of secret hearings. In both Canada and the UK, they have generally declined to receive secret evidence, citing adherence to the open court and open justice principles. (22) But another reason relates to formal distinctions between fact and law, of which only the latter is at issue in most appeals. (23) Following the UK Supreme Court's lead in Bank Mellat v HM Treasury, (24) the Court in Harkat said "closed evidence is factual in nature, whereas the points debated before appellate courts are essentially legal" and so "closed hearings before it would rarely, if ever, be necessary for the proper disposition of an appeal". (25) Although the Court did view evidence and hear submissions in secret, it sought to assure critics that the material "did not assist this Court in deciding the issues before it". (26)

While sound from the perspective of the open court and open justice principles, the hard and fast distinction between facts and law raises a methodological question: how can an appellate court pronounce on the soundness of judicial reasoning, much less the general fairness of an SA system, if it does not know what is being decided or even how decisions are made? Recalling the maxim "justice should not only be done, but be seen to be done", we might just as well ask how high courts know justice is being done unless they peer behind closed doors.

The purpose of this paper is to shed some light on two questions: how do secret hearings work, and, to what extent do they abide by the rule of law? Answering either question is obviously difficult--secret hearings are a black box, the content of which can be gleaned only by observing input and output. To maximize our ability to describe and interpret how secret hearings work, we combined legal research (e.g., case law, statutory law, official documents) and interviews with thirty-nine stakeholders, including judges, lawyers, administrators, and institutional representatives involved in secret hearings in Canada and the UK. As this research is ongoing, this paper represents the first of our findings, focusing on the Canadian experience and the Canadian SA system in particular.

Our first and primary finding is that, although Parliament has outlined the essential powers and functions of SAs, the exercise of these powers is conditional on a host of informal factors. Before sketching out some of these factors, we should pause to define "conditionality" as including the material, epistemic, and relational conditions that must be satisfied before SAs can acquire and exercise powers notionally available to them through positive law. (27) Conditionality entails hard work--SAs have to actively find ways to establish and negotiate claims for specific powers on a case-by-case basis, working both with and against a host of dynamic factors. Pertinent factors include: relations of trust and distrust among judges, government lawyers, intelligence communities, and excluded parties; governmental and judicial concerns about inadvertent disclosure; the shifting institutional culture of the Federal Court; and the idiosyncrasies of particular judges. Our second claim is that the confluence of these forces has in some areas produced a body of customary norms or "stabilized interactional expectancies" that reduce but do not eliminate contingency. (28) This sort of stability is highly desirable, providing some semblance of the rule of law. However, SAs remain effectively unable to negotiate some vital powers and, when they are successful, the scope and durability of these powers is subject to change. We identify some areas where greater fairness can be provided without increasing reasonable risks of inadvertent disclosure.

The paper is organized as follows. First, we provide a brief overview of the SA system, including pertinent constitutional norms and legislative provisions. We also survey some of the core criticisms of SA systems, from the perspective of the right to a fair trial and the adversarial tradition. Second, we outline the theoretical context for the study and our methodology. From here we introduce the empirical results of our research. Section III outlines how rules of practice and procedure emerge and apply in the Federal Court of Canada, with specific regard to the interaction between path dependency (29) and organizational change. Section IV explores the material, epistemic, and discursive challenges SAs face, as well as whether and how these obstacles have been overcome. Matters of interest include disclosure, administrative support, and the development of institutional knowledge and expertise. We identify trust and control as critical variables, exploring these themes more fully in Section V, using as case studies struggles for the power to communicate with outside parties and to call expert witnesses. We end by analyzing the implications of our findings with respect to both the fairness of the SA system and the future of secret hearings in Canada.

I. The Legislative and Constitutional Framework of the Special Advocate System

A. Secret Hearings and the Right to a Fair Trial

The right to a fair trial is protected through section 7 of the Canadian Charter of Rights and Freedoms. (30) It requires that one be tried before an independent and impartial adjudicator, that the decision of the adjudicator be based on the facts and the law, and that one know and be able to meet the case against her. (31) A precondition of a fair trial is adequate disclosure and adversarial challenge. In the context of criminal law, the government is obligated to disclose all information in its possession that is relevant to the defence. (32) Exceptions to disclosure include privileged material, such as information that would identify a confidential informant or place the safety of a person at risk (e.g., a spy or undercover agent). Modified disclosure obligations apply in administrative law settings as well as when the government is named as a defendant in a civil trial. An example of the latter is the civil suit launched by Abdullah Almalki, Ahmad Abou-Elmaati, and Muayyed Nureddin, who claimed damages from the Government of Canada for its role in their detention and torture in Syria. (33)

Secret hearings allow for decisions to be based on material that has not been disclosed to the affected party. To be clear, secret hearings are a function of legislation and, hence, occur due to parliamentary intent and executive policies and priorities. The Parliament of Canada has decided that almost all secret hearings occur before a designated Federal Court judge. Security certificates are the most conspicuous example of a secret hearing. Governed through Division 9 of the Immigration and Refugee Protection Act (IRPA), (34) certificate proceedings lead to the detention and deportation of a non-citizen alleged to be inadmissible to Canada on the grounds of security, serious criminality, organized criminality, or the violation of human or international rights. The bulk of evidence used to support the allegations is provided by the Canadian Security Intelligence Service (CSIS). A designated judge reviews this information and decides whether ongoing detention is justified (35) and whether the certificate is reasonable. If it is, the named person is subject to removal from Canada, subject to legal proceedings designed to avert deportation to face the substantial risk of torture or similar abuses. (36) The designated judge is authorized to make any decision on the basis of evidence not disclosed to the named person. (37)

Designated judges also preside over section 38 Canada Evidence Act (CEA) proceedings, (38) in which the government argues for the non-disclosure of sensitive information during a criminal or civil trial. In the absence of the defendant or plaintiff, respectively, the judge reviews the information and decides whether the public interest in non-disclosure outweighs the public interest in disclosure. Unlike in certificate proceedings, section 38 proceedings are concerned with whether sensitive information can be excluded from the evidentiary record. This compromises the fairness of a trial, as decisions cannot be based on all of the facts that are relevant to the issue at hand, nor is an affected party provided with all information relevant to her case. It also means, however, that the information cannot be used to support the government's case.

Courts have decided that both of these legislative frameworks comply with the Charter. In Charkaoui v Canada (Citizenship and Immigration) (Charkaoui I), (39) Charkaoui v Canada (Citizenship and Immigration) (Charkaoui II), (40) and the 2014 case of Canada v Harkat, the Supreme Court of Canada ruled that certificate legislation coheres with the right to a fair trial if and only if persons named in certificates have a substantial substitute for disclosure. This substitute has come in the form of SAs, who attend closed hearings in order to challenge the Minister of Public Safety and Emergency Preparedness' (Minister of Public Safety) claims that certain information cannot be disclosed and to challenge the relevance, reliability, and sufficiency of secret evidence. The efficacy of this system depends on the disclosure of "all information in its possession regarding the person named in a security certificate" to the Federal Court and SAs. (41) Similarly, the Supreme Court of Canada approved of section 38 of the CEA in R v Ahmad, (42) resting this decision on the fact that the Federal Court can provide trial judges with "conditional, partial and restricted disclosure", or even all the sensitive information "for the sole purpose of determining the impact of non-disclosure on the fairness of the trial". (43) Trial judges retain the power to issue remedies for any breaches of section 7, up to and including staying proceedings.

We should note the SA system only applies to select immigration law proceedings, including certificates, the decisions of the Immigration and Refugee Board (IRB) made under Division 9 of the IRPA, and judicial reviews of those decisions. (44) However, the Federal Court often appoints amici curiae to section 38 CEA proceedings. The Court also appoints amici in proceedings concerning CSIS warrant reviews, with the most notable, recent example being an en banc hearing concerning a breach of CSIS' duty of candour with respect to a data collection program (as well as the lawfulness of the past and future retention of data). (45) It can also appoint amici when hearing appeals of no-fly listing designations and passport revocations (46) or conducting judicial reviews of any decision within which a claim of national security confidentiality is made. To emphasize, amici serve altogether different roles and possess different powers than SAs.

B. The Special Advocate System in Canada

Parliament introduced the SA system on October 22, 2008. (47) The IRPA states that the statutory role of the SA is to "protect the interests" of the named person in closed hearings. (48) The SA is not a party to the proceeding and the relationship between the SA and named person is "not that of solicitor and client". (49) Nonetheless, communications between SAs and named persons are protected as if they were subject to solicitor-client privilege. (50) Specific responsibilities of SAs include challenging the Minister of Public Safety's claims that certain information cannot be disclosed, and challenging the relevance, reliability, and sufficiency of secret evidence. Section 85.2 of the IRPA specifies that SAs may make oral and written communication with respect to sensitive information. SAs may also participate in, and cross-examine witnesses who testify during, closed hearings. With the authorization of the designated judge, SAs may employ any other powers necessary to protect the interests of the named person.

This was not the first time security-cleared lawyers participated in certificate proceedings. Prior to the Federal Court, certificates were administered by the Security Intelligence Review Committee (SIRC)--an independent administrative body that oversees and reviews the actions of CSIS. SIRC used security-cleared counsel, empowering them to call and rigorously cross-examine CSIS witnesses, communicate with affected parties and outside counsel, and access government files relevant to the issues raised in a proceeding. As we will detail more fully below, the SA system differs from the SIRC system in a number of key respects, two of which are most relevant. First, whereas SIRC counsel worked at the direction of SIRC, SAs are independent counsel charged with representing the interests of named persons. Second, the SA system is the first time in which the powers of security-cleared lawyers were itemized in enabling legislation; the SIRC system ran more or less according to customary law and at the discretion of SIRC members. (51)

For better or for worse, the SA system was modelled after the UK system, (52) which was structured by the Special Immigration Appeals Commission Act 1997 (53) and the Prevention of Terrorism Act 2005. (54) Both pieces of UK legislation were drafted in response to negative judicial rulings, these being Chahal v The United Kingdom, (55) and A v Secretary of State for the Home Department. (56) The IRPA directs that a judge "shall" appoint an SA in any certificate hearing, after considering the position of the named person and the Minister of Public Safety, but "giving particular consideration and weight to the preferences" of the named person. (57) But section 83 of the IRPA goes on to say that the judge shall appoint a particular person selected by the named person, except if so doing would result in unreasonable delays, produce a conflict of interest, or if the person possesses knowledge of sensitive information and there is a risk of "inadvertent disclosure" to the named person or his counsel. (58)

The latter exception is colloquially referred to in the UK as being "tainted". The most common example of tainting would be when an SA has worked on a file that relates to the same network of people or the same region or country that forms the basis of a prospective file. (59) For example, having represented someone on a file relating to specific terrorist networks in Somalia would disqualify a UK SA from taking any future case relating to Somalia. (60) In the UK, the Secretary of State may block the appointment of a specific SA on the grounds of tainting. By contrast, the IRPA does not give the Canadian government a veto over the appointment of SAs. However, it provides the designated judge the power to decide whether there are sufficient reasons to think there is a risk of inadvertent disclosure of sensitive material.

Canada and the UK have tried to minimize the risk of inadvertent disclosure by prohibiting unauthorized communication between SAs and third parties, with some exceptions. In the UK, SAs may freely communicate with presiding judges, the Secretary of State, and specified law and administrative officers. In the case of proceedings before courts, SAs may communicate with affected parties and their counsel only if they secure approval from the presiding judge; in the case of proceedings before Special Immigration Appeals Commission (SIAC), the decision is made by the Commission. (61) Requests to communicate are made in writing and are sent to the intelligence community as well, which may file an objection; although presiding judges are formally empowered to grant or withhold permission, the views of the intelligence community are in practice decisive. The Parliament of Canada has similarly vested designated judges with the discretion to authorize communications between SAs and named persons "subject to any conditions that the judge considers appropriate". (62) Judges shall give the Minister of Public Safety an opportunity to be heard in such instances. (63) Given the practice in the UK, it was open to question how greatly judicial decisions about the powers of the SA would be subject to the influence of CSIS.

Section 85.4(1) of the IRPA states the Minister of Public Safety shall provide SAs with a copy of all information and other evidence that is provided to the judge but that is not disclosed to the named person. (64) The scope of disclosure under this provision was unclear, since it did not make any reference to whether the government was obligated to submit all relevant information, including that which could be helpful to the named person. Relatedly, it was unclear whether the government would be obligated to disclose all relevant information to the judge or indeed continue to search for information that may challenge its own case once proceedings were underway. The Supreme Court of Canada clarified matters in the 2008 case of Charkaoui II--the third case before the Court regarding the constitutionality of the certificate regime--stating that the government must disclose to the designated judge (and SAs) all information on file relevant to the named person. Parliament mischievously narrowed the scope of disclosure in 2015 through Bill C-51 in an ill-advised attempt to test the boundaries of Charkaoui II. (65) Section 77(2) of the IRPA now states that the Minister of Public Safety is to disclose to the court and SAs only information that is "relevant to the ground of inadmissibility stated in the certificate". (66)

As noted above, named persons are entitled to summaries of information and other evidence that enables them to be "reasonably informed" of the case made by the government. While summaries may serve as a substitute for disclosure, it was at the time open to question whether fairness required a bare minimum amount of actual disclosure to the named person. The UK House of Lords issued a decision on just this matter as the Canadian SA system was winding its way through Parliament in the case of Secretary of State for the Home Department v MB, (67) deciding that a SA system can cohere with the right to a fair trial, but that affected parties are entitled to enough disclosure to convey to a detainee the gist of the government's allegations. It added content to the "gisting" requirement in the 2009 case of Secretary of State for the Home Department v AF (68) (also concerned with control orders) but refused to extend the requirement to immigration hearings before SIAC in RB (Algeria) v Secretary of State for the Home Department, decided in the same year. (69)

C. The Constitutionality of the Special Advocate System Reviewed: Harkat v Canada

The Supreme Court of Canada upheld the constitutionality of the SA system in its 2014 Harkat decision. At issue were, first, restrictions on communication between SAs and named persons. Mr. Harkat and interveners on his behalf drew the Court's attention to the UK experience, where SAs unequivocally stated that communication bans were the "most significant restriction on the ability of SAs to operate effectively". (70) This flaw is well-documented and routinely canvassed in UK courts and Parliament, (71) where judicial authorization procedures require UK SAs to reveal to presiding judges and to the government the proposed content of the communication as well as what was discussed after the fact. (72)

The Court did not expressly note criticisms of the UK system. It was satisfied that the "broad discretion" of designated judges over authorization "averts unfairness" to the extent that judges "take a liberal approach". (73) In its view, the real issue is with the standards used to assess requests to communicate and not the process. In fairness, the Court tangentially noted that the "evolving practices of the Federal Court" with respect to safeguarding solicitor-client privilege are generally to prevent injustices. (74) Although it did not see sufficient evidence of a past or current conflict between the judicial authorization process and solicitor-client privilege, (75) it opined that such a conflict may lead either to an exception to privilege (76) or to a finding of a breach of section 7. (77)

Another issue was whether named persons receive sufficient disclosure. Invoking UK case law on gisting, the Court held that named persons are entitled to an "incompressible minimum amount of disclosure", the absence of which may render a proceeding unfair. (78) It again left this matter to the discretion of the designated judge, who will decide matters of fairness on a case-by-case basis. (79)

And so the Court was satisfied the SA system is fair. While legislative language leaves open the possibility for procedural and substantive injustices, it also provides the framework for what the Court considered to be justifiable limitations on rights. The question is one of discretion: how do judges apply legislation and underlying constitutional principles? (80) In the aftermath of the Harkat decision, the government expressed confidence that broad discretion settles most constitutional questions. In 2015, the Department of Justice published an evaluation of aspects of the SA system, concluding that Harkat sets aside any "ambiguity" about the fairness of the SA system which now "stands on firm ground that can only be shifted through legislative amendments". (81)

We do not share this view--the constitutional ground is anything but firm while ambiguities abound. Apart from arguments about the distinctions to be drawn between what is constitutional and what courts say is constitutional, the fact is courts have only pronounced on a fraction of the rights and rule of law issues that arise in secret hearings. The UK experience highlights a wide range of structural problems that are likely to have arisen in Canada but which have not been fully dealt with, including: inadequate administrative support, late disclosure of sensitive material, the non-disclosure of sensitive material, the inability to effectively identify and then challenge non-disclosure, paucities of formal rules of evidence, the lack of a searchable database of closed judgments accessible by SAs, unduly expansive or aggressive governmental stances on tainting, and the (practical) inability of SAs to call expert witnesses.

II. Theoretical Context and Methodology

A. Secret Hearings and the Rule of Law

Academics and legal professionals have debated the constitutionality of security certificates for many years, with the preponderance of opinion being that certificates are "exceptional". (82) A contested term, exceptionality connotes the suspension of law and the concomitant unfettering of sovereign power from normative constraints, which then becomes subject only to political constraints. (83) When political authority is split from legal authority in liberal democracies, legal institutions (courts, legislatures) do one of two things: they find ways of reasserting the authority of law (thereby contesting the authority to govern), or, they cobble together a passable veneer of legality (thereby normalizing exceptionality). (84) Referring to the expansion of secret trials to the UK civil law context, Lord Hope reflected on the fine line between these two scenarios:
[I]t is a melancholy truth that a procedure or approach which is
sanctioned by the court expressly on the basis that it is applicable
only in exceptional circumstances none the less often becomes common
practice.... That would create a state uncertainty in an area of our
law which would be inimical to the concept of a fundamental right. (85)

Lord Hope was in this instance speaking to the question of whether the common law allowed judges to hold closed-material hearings or whether they could only lawfully administer such proceedings if required to do so by Parliament. Deciding it was the latter, Lord Hope expressed great unease with secret hearings and a reluctant willingness to use them, at present, only when subject to the authority of Parliament. (86) Running counter to the tenets of the adversarial system, it is easy to see how some judges would be uncomfortable administering secret hearings. But it is also easy to see how the SA system's promise of simulating the adversarial tradition might render secrecy more tolerable. The SA system has now reached such a stage of normalcy that even human rights organizations refer to it as a viable means of shoring up rights and the rule of law--albeit with a healthy dose of ambivalence. In its 2016 submission regarding the Government of Canada's National Security Green Paper, the Canadian Bar Association stated that secret hearings
risk undermining public confidence in the courts themselves if judges
are seen as complicit with security agencies rather than transparent,
neutral arbiters safeguarding the rule of law. As a general principle,
courts should function in an open, adversarial system to the furthest
extent possible. To the extent that secrecy is at times required, other
mechanisms such as special advocates can be put into place to ensure a
strong defence of the rights and interests of individual citizens. (87)

We glimpse here a hesitant embrace of the SA system, the value of which is understood not in absolute terms, but in relative terms as the lesser of two evils--that is, as something which is to be preferred to unconditional secrecy, but which is not therefore made morally justifiable. It should be added that accepting the SA system in principle does not determine the bare minimum elements such a system must possess to stand as a meaningful alternative and, indeed, whether providing the bare minimum is sufficient when more is possible.

The lesser of two evils stance is intriguing because it highlights the problematic aspects of accepting, making use of, and in a way endorsing a component of secret trials while at the same time contesting their very legitimacy. The best doctrinal and conceptual resource for making sense of ambivalence is the rule of law. Drawing from Joseph Raz, we adopt a formalist conception of the rule of law that denies necessary connections between the rule of law (and more broadly the content of law) and morality. On this view, the rule of law requires that law be (1) prospective, open, and clear, (2) relatively stable, and (3) general. (88) The rule of law also includes procedural requirements, including the right to a fair trial, access to courts, and the existence of judicial review. (89) In this way, the rule of law constrains the arbitrary exercise of discretionary power, where arbitrariness denotes the "use of criteria of decision that are inappropriate in view of the underlying purposes of the rule", regardless of what the moral content of that rule might be. (90)

How can secret law ever satisfy the rule of law? While we will not attempt to definitively answer this question, we will note that first and foremost, any effort to do so would have to define secret law as a body of general and stable rules that both are known to a group of legal officials (e.g., judges, government lawyers, SAs) and are sourced in public law. Dakota Rudesill suggests that such a form of law can potentially mediate interactions between discretionary power and public law, if "it is not in any way at odds with other law created by or known to other institutors and people who are not aware of the secret law". (91) This requires judges and other authoritative decision makers to "exercise great deference to Public Law and public understanding of it." (92) In theory, discretionary decisions cease to be arbitrary, even though this fact would be known only to a small group of state officials.

It should be reiterated that this model says little about the content of the principles against which discretionary decision making is scrutinized. But even if one were satisfied with the principles outlined in Charkaoui I and II (including all those bundled up in the right to a fair trial), this account of the rule of law displaces the problem. While secret operations are no longer confined to the darkest corners of the executive branch or conceptually unrelated to public law, they are now enrobed in a body of secret decisions and legal opinions, where the judgment of designated judges has to be taken on faith in just the same way as would the goodwill of executive officials. The addition of SAs within a process of oversight or review introduces greater potential for criticism of the government's legal and factual arguments. But outsiders never really know what goes on behind closed doors. All things being equal, we wonder how effective SAs can be as checks on judicial discretion when secrecy by its nature "brings power and status, and is itself inherently a form of regulation that brings influence and status to insiders." (93) Then of course we must confront the inherent limitations of secrecy vis-a-vis the core elements of the rule of law, including generality, openness, and stability. Detailing the deficiencies of the United States Foreign Intelligence Surveillance Court, Orin S Kerr notes how the non-publication of reasons for decisions, the lack of precedent, and the absence of serious appellate review undercut the possibility of relying on feedback mechanisms that would normally reduce the occurrence of mistakes of fact and of law. (94) It is possible that concretizing judgment in the form of law solidifies rather than corrects error. (95)

All of this raises the question of how secret law might be institutionalized--how might such law come to operate as a free-standing body of rules that, while unknown to outsiders, nonetheless impose interpretive and normative constraints on discretionary decision making consistently with autonomous legal values? It is only when decision making is sourced in and abides by the principles and purposes of public law that one may speak of a system that coheres with the rule of law. SAs play a necessary although by no means sufficient role here, insofar as they are uniquely situated to submit the government and judges to criticism by reference, not only to facts and arguments about facts, but by reference to public law, secret law, and the linkages between. Shifting the referent, then, we should be asking about how we might design a system whereby SAs can consistently wield the powers they require to discharge this role.

Given the experiences of UK SAs, we hypothesize that the powers of Canadian SAs are conditional, which is to say SAs must satisfy a host of material, epistemic, and relational conditions before they can acquire and exercise their powers. Conditionality describes a process of constant struggle, where SAs have to actively establish and negotiate claims amidst a host of counter influences that include: judicial and governmental anxieties about inadvertent disclosure, relations of distrust, personal and professional isolation, lack of access to closed judgements, and under-resourcing. However, we also hypothesize that the work of conditionality can over time produce some stable interactional expectancies between SAs, judges, government lawyers, and administrators, which we hereafter would refer to simply as custom or customary law. (96) In some areas, the emergence of customary law might render the work of special advocacy less capricious.

B. Methodology

Our methodology consisted in documentary research and qualitative research. First, we examined case law, judicial orders and directions, statutes and regulations, government and academic reports, and parliamentary records. This research was greatly assisted by the Department of Justice's Special Advocate Program (SAP), which provided us with access to an electronic SA portal containing the totality of Federal Court decisions and orders made in certificate cases, as well as a range of training and professional development material. We also researched materials on the UK's experience.

Second, we conducted semi-structured interviews with thirty-six persons involved in secret hearings, in both Canada and the UK. Interviews lasted between one and two hours. We audio recorded most interviews, with the express permission of the participant, and treated the data collected according to Ryerson Research Ethics Board guidelines. The audio recordings were transcribed verbatim and then analyzed using (1) a descriptive content analysis approach, (2) an interpretative content analysis, and (3) a framework analysis inspired by our previous research studies on these topics. We combined these qualitative methods with ongoing legal and documentary research. In some instances, we conducted second follow-up interviews to explore themes in greater detail.

Participants were informed of the basic objectives of the project, and were assured that we were interested only in the professional, procedural, and administrative realties of working with secret materials, i.e., we were not interested in the specifics of any case. SAs described past experiences with security intelligence and secret hearings, and the nature of their roles. They also shared evaluations of the quality of training and professional development, as well as their views on the fairness of the SA system. SAs were specifically asked about the ways in which they request and deploy statutory rights and powers, including those related to disclosure, calling expert witnesses, and communicating with outside parties, e.g., affected persons, outside counsel, other SAs. Judges were asked similar questions, although we focused on overall philosophies of court administration, how rules of practice and procedure have been developed, how improvements to the SA system could be made, training and professional development, and the nature and extent of interactions with other judges on recurring legal, practical, and administrative matters.

A core task was to trace the historical development of various models of administering secret hearings and the variety of ways customs emerged and affected interactions within these formally distinct frameworks. For reasons that will become clear, we focused on the approach used by SIRC--an independent, external body that reviews CSIS--when it administered security certificates between 1984 and 2001. To this end, we spoke with current and former SIRC counsel and a former Chair of SIRC. We also spoke with representatives of the SAP, which is an independent administrative and professional support body housed within the Department of Justice. We spoke with representatives of the IRB and with Designated Registry Officers (DROs) within the Courts Administration Service (CAS). We were unable to secure interviews with government lawyers or CSIS at the time of this writing.

Canadian representatives included: one judge of the Federal Court, three representatives from the IRB, five Canadian SAs, three representatives from the Department of Justice SAP, three DROs in the Federal CAS, one outside counsel who represented persons subject to a certificate, two current and three past counsel with SIRC. UK participants included: two justices of the High Court of England and Wales and Special Immigration Appeals Commission, ten UK SAs, two representatives of the UK's Special Advocate Support Office (SASO) and the Independent Review of Terrorism Legislation. While some participants were fine with their identity being revealed, others were not. In the interest of maintaining confidentiality, we decided to keep all names confidential, with the exception of Simon Noel J (since no other Federal Court judge participated). We also reveal the name of Ron Atkey, as his experience as first Chair of SIRC was essential to supporting our interpretation of how the SIRC system operated (and no other chairs were interviewed).

We should add that because we were interested in the experiences of SAs and generally the internal operation of the SA system, we have not included any interviews with named persons. We have not included perspectives of outside counsel except as these related to prior experience one may have had as an SA or SIRC special counsel, or to the extent one could speak to the ways in which SIRC proceedings worked.

III. Rules of Practice and Procedure in the Federal Court: Formal Law, Custom, and Culture

This section provides an overview of how rules of practice and procedure are generally produced in certificate proceedings. It begins with the approach used by SIRC, which we compare and contrast with that used by the Federal Court. We observe two dynamics. First, the Federal Court is acculturating to the rule of law demands of the SA system, where traditional philosophies of isolation are giving way to a more collective and self-reflective approach to matters of substantive and procedural law. Second, however, this process is largely informal and has yet to lead to the production of formal rules of practice and procedure that apply across certificate files. For better or for worse, the Federal Court has retained a largely responsive, problem-oriented philosophy where individual judges retain considerable freedom to handle files in their own, distinctive ways. This is attributable in part to path dependency or the continued influence of organizational culture, partly to the wisdom of ensuring that judges can be maximally responsive to complex, novel, and time-sensitive problems, and partly because Parliament has offloaded responsibility for designing fair rules of procedure upon the Federal Court.

A. Rules of Practice and Procedure in SIRC

Literature on how secret hearings operated in SIRC and in the Federal Court prior to 2002 is scant, with Ian Leigh providing the most comprehensive account. (97) Under both systems, the Solicitor General (now the Minister of Public Safety) and the Minister of Citizenship and Immigration (now the Minister of Immigration, Refugees and Citizenship) issued a certificate. Between 1984 and 1992, all certificate proceedings occurred before SIRC. Between 1992 and 2002, certificates issued against permanent residents were heard before SIRC, while certificates issued against foreign nationals were heard before the Federal Court. As noted, all certificates (against permanent residents and foreign nationals) were reviewed by the Federal Court after 2002.

Although no longer involved in certificate proceedings, SIRC wields inquisitorial powers while investigating complaints against CSIS as well as a selection of other matters; in each instance, investigations include hearings which are governed by many of the rules that were used in certificate proceedings. (98) The basic function of the SIRC in this respect is to collect and assess information, sometimes outside of the context of a hearing altogether, in some cases compelling witnesses to testify. (99) In the context of certificates, the decisions of SIRC were advisory and not binding, although these recommendations had some authoritative force and were often assented to. (100)

Perhaps the most significant feature of certificate proceedings was the extent to which they infused elements of adversarial challenge into proceedings that were otherwise nested within an inquisitorial process. Certificate proceedings began when the sole SIRC board member assigned to a file examined information submitted by the government that was relevant to the certificate. The board member was assisted by SIRC counsel, who were either in-house counsel or external counsel hired from the private bar. As SIRC staff, both the member and SIRC counsel would have access to the file submitted by CSIS, and any additional information in the possession of CSIS that was relevant to the investigation, short of cabinet confidences. (101)

The functions of SIRC counsel were threefold: to assist the board member, to liaise with the affected person and her counsel, and to cross-examine government witnesses in closed hearings. (102) There was no functional difference in the powers or responsibilities of in-house or external counsel. (103) The decision of which type of counsel to use typically related to resource considerations or a desire to preserve the appearance of impartiality, since cross-examinations of CSIS witnesses could be aggressive.

Prior to the commencement of hearings, but after the initiation of the investigation, pre-hearing conferences were conducted to establish ground rules. Subsequently, hearings consisted of an open and a closed portion. At the open portion, the named person and his counsel would be able to make submissions based on disclosed government evidence or their own evidence. Closed hearings were attended at by the member, SIRC counsel, and the government. Witnesses would attend at closed hearings only to testify and to answer questions. SIRC counsel would examine and cross-examine witnesses.

Rules of practice and procedure came in two forms. Some were formally encoded in SIRC Rules of Procedure, which entered into force in 1985. (104) Several persons played roles in the enactment of these rules, including Maurice Archdeacon (first Director of SIRC), Ron Atkey (first Chair of SIRC), Lutfy CJ and Noel J (to be clear, of this list of persons, only Ron Atkey and Noel J were interviewed). Formalization had limited effects; as rules they simply reinforced the broad discretion of board members. For example, the rules stated that it was "within the discretion of the assigned members" to provide excluded parties with summaries of secret evidence, to disclose material, and to allow excluded parties to cross-examine witnesses. (105) But in each case, discretion was to be guided by a balancing of the interest in "preventing threats to the security of Canada and providing fairness to the person affected". (106) Board members consistently provided summaries of closed material and allowed outside counsel to ask SIRC counsel to cross-examine secret witnesses on a particular point. (107)

Other rules were nowhere to be seen in legislation or regulations, which is not to say they were bereft of normativity--they simply remained customary in nature. Examples included communications between SIRC counsel and affected parties or outside counsel. During its years administering certificates up until about 2001 or so, board members regularly allowed SIRC counsel tocommunicate with relevant parties before and after viewing secret material. (108) One former SIRC counsel stated that
procedures that are adopted at the discretion of the presiding member,
in consultation with the counsel to the committee, ah, and the
executive director of the committee. And so, it would be at--this is
one of the attractions of the SIRC model is that it is flexible enough,
for example, for a presiding member to say to his or her counsel, on a
complaint inquiry, "I want you to talk to the complainant or the
complainant's counsel and clarify such and such an issue" or "put them
at ease as to the procedure that's going to be followed. Give them a
briefing about what's going to happen; give them an update on what has
happened" et cetera. (109)

It should be noted that one outside SIRC counsel recalls not having permission to communicate with relevant parties. (110) But another stated: "[I]n at least twenty, if not more, cases that I've done, I have never been restricted in that way, and indeed often, ah, that was an important part of the role I was playing." (111)

The role of SIRC counsel was indeed a determining factor. To be clear, SIRC counsel were (and remain) agents of the committee member and did not represent or act on behalf of relevant parties; any effect thereof would be incidental to their role of assisting the committee member in the discovery of truth. While in one sense removed from the premises of the adversarial system, the inquisitorial function of SIRC meant that committee members had more direct control over SIRC counsel and seemed to trust them. We should bear in mind that liberal rules of communication could prejudice the interests of affected persons for this reason, as they could easily incriminate themselves when giving information to SIRC counsel. Despite the inquisitorial setting, SIRC was attuned to principles of fairness and adversarial challenge, as evidenced by the custom of SIRC counsel to advise affected persons that any relevant discussions would have to be reported back to the committee member--even if this was adverse to the named person. (112) For some SIRC counsel, there were occasions when the committee member authorized SIRC counsel ahead of time to withhold any inculpatory statements made by the affected party, although inculpatory information found outside the context of direct communication with the affected party would still be reported. Even so, this could produce a sense of discomfort, as SIRC counsel would have to withhold at least some information they knew was relevant. (113)

The discretion of members was therefore extremely broad, but decisions did not appear to be arbitrary. Discretion was subject to a small body of formal rules of practice and procedure of a general nature, and a much larger pool of customs. These customs emerged over time, and were themselves shaped by the culture of SIRC, including its organizing principles and purposes. Finally, it seems that relationships of trust developed and alignments of interest among a close-knit community of actors were at least as important as conceptions of fairness with respect to the powers of SIRC counsel.

One final set of variables is relevant, and that is the authority of the chair and the executive director of SIRC to impose some measure of consistency and order across files. In the early years of SIRC, the legal expertise and involvement of these actors was high, as one could surmise by reviewing the biographies of Messrs. Atkey and Archdeacon. We were told this collection of experience and hands-on approach was necessary to provide the newly-established committee direction, a common identity, and institutional legitimacy. (114) This corresponds with the fact that SIRC had yet to prove itself as a legitimate review and oversight body and, indeed, CSIS had yet to demonstrate it would conduct itself differently from the Royal Canadian Mounted Police Security Service. (115)

Newly appointed committee members would have needed some additional support and direction at this time, although that too depended on whether they had prior legal experience or expertise. According to some interview participants, there were times when committee members would consult about matters of procedure with SIRC counsel, the chair of the committee, other committee members, and even the executive director and the executive secretary of SIRC. (116) Bearing in mind that SIRC still presides over secret hearings unrelated to certificates, it is relevant to note that its customs have continued to evolve and in some respects resemble that of the Federal Court. With the maturation of the committee, its growing bureaucratic size, heavy case load, and the stabilization of norms, committee members now handle files entirely on their own and do not consult with outside parties. (117)

B. Rules of Practice and Procedure in the Federal Court

According to the Supreme Court of Canada, the Federal Court preferred a pseudo-inquisitorial style when administering certificate hearings right up until Charkaoui I, (118) when elements of the adversarial system were introduced as part of the newly-instituted SA system. But the conditions for organizational change arose shortly following 9/11, which played a role in how the Federal Court internalized the Charkaoui I decision. These conditions included the slow and kaleidoscopic transmission of the traditions of SIRC through the work of two personalities: Allan Lutfy CJ (as he was from 2003 to 2011) and Simon Noel J. Chief Justice Lutfy had extensive experience as counsel on a number of commissions of inquiry into national security matters, including the McDonald Commission. He also worked as SIRC counsel, alongside Noel J, who was appointed to the Federal Court in 2002. During their time at SIRC, each played an instrumental role developing many of the valued practices and procedures of the SIRC model noted above, including those relating to communication with affected parties and the provision of summaries of protected information. By no means were these customs or procedures adopted within the Federal Court at the time, but this considerable experience would prove useful later.

Following the spate of post-9/11 legislative and operational changes in the context of certificates and secrecy in general, Lutfy CJ gave the newly appointed Noel J primary responsibility over security files and related administrative matters; (119) this role was incommensurate with Noel J's level of judicial experience, eliciting some measure of internal discord on the bench. The transitional years from a bifurcated regime to one housed entirely in the Federal Court were rough. According to Noel J, Lutfy CJ's past experience prepared him well to handle the filing of new security certificates in 2002, when the IRPA legislative regime was changed, as he "knew the area and what the concerns were". (120)

Despite a long history with certificate files, warrant applications, and other forms of secret hearings, there was a period of uncertainty, if not confusion, about how to manage the new certificate files. Justice Noel recalls that designated judges "didn't know what was happening" but that they had to "get the job done". (121) He also recalls the pull of existing institutional culture and the resistance of long-standing designated judges used to handling files in their own way. Once appointed as Chief Justice, Lutfy CJ gave Noel J considerable responsibility for directing designated proceedings that was incommensurate with his short time on the bench. He recalled being reticent, as a newly designated judge, to "go in there and try and change everything." (122)

This environment was characterized by a rather polycentric, informal approach to judicial administration. We have noted that the Trial and Appellate Divisions of the Federal Court never produced binding rules uniquely applicable to certificates, as they were authorized to do through section 85.6(1) of the IRPA. (123) Justice Noel informed us that the philosophy of the Court was and remains one of broad individual discretion. The premise of this philosophy is that too many formal rules hinder the ability of designated judges to craft creative and fair solutions to the problems they encounter; judges retain the flexibility they believe is required to adapt to novel and complex problems that require a timely and often singular response. There are also broader, institutional trappings to this philosophy. Justice Noel informed us that draft rules must pass through a rules committee of the Federal Court, which could take up to two years to be enacted. By then, designated judges would have resolved the issue at hand by other, informal means. We were also told that the designated judges were reluctant to go through the rules committee because it is composed of non-designated judges who have limited knowledge of the realities of secret hearings. And so Chief Justices of the Federal Court have, at least since 2003 but likely well before, encouraged designated judges to manage his or her file in their own respective way.

But this philosophy is also an outcrop of path dependency, including the inertia of judicial silos that, prior to 2002, limited the extents to which designated judges shared or had collective access to classified information germane to their respective files. We should pause to explain this latter point. Designated judges are not security cleared. Although they are vetted prior to their appointment to the bench, they are not subject to renewed security clearance, even when appointed as designated judges by the Chief Justice. The authority of any given designated judge to access secret material is an extension of her statutory responsibility to administer proceedings related to the individual certificate assigned to her. Justice Noel informed us that one of his goals as coordinator of closed proceedings was to facilitate awareness and communication of classified judgements among designated judges so that "the right hand knows what the left hand is doing". (124) This involved ensuring that designated judges have regular access to the classified files and judgments of fellow designated judges. This material is securely stored by the CAS, where DROs, paired with a designated judge, retrieve and return case-specific material from a secure site. (125)

Judges do strive to report as much of the reasons for a decision as they can, (126) but there exist bodies of decisions that remain secret. The perforation of silos is essential if the conditions of general rule production are to be met. Access to the judgments of their peers supports the sharing of reasons in written form and, to a degree, supplements the emergence of shared understandings about how recurring substantive or procedural problems unique to secret trials may be resolved. However, we should not exaggerate the implications this has for the rule of law, since SAs do not have access to the same bodies of law that designated judges and, presumably, government counsel do. Instead, SAs are permitted to access only material that is internal to their assigned file. This presents a clear epistemic barrier and, what is more, limits the contributions SAs can make to the evolution and criticism of law; SAs are not well positioned to know (1) if there are customary practices or even written rules of a general nature, or (2) if the designated judge assigned to their file is breaking with custom or written law. Later, we will discuss ways in which SAs have been able to enhance knowledge of what goes on within other files, but for now we should note that, being security-cleared, SAs can and should be given access to all closed judgments to which judges and government counsel have access. It should be noted that this is the practice in the UK, where the High Court of England and Wales recently established a library of closed judgments that judges and SAs can access freely. (127) UK judges expressed to us that they expect the impact of the library to be minimal. (128) But the reason for this is that, according to them, questions of law rarely arise and are in any event seldom complex or multi-faceted. But the SAP, SAs, and Noel J informed us that Charter and other legal issues are frequent. (129)

There are other examples of important and commendable shifts in the organizational culture of the Federal Court towards a more coordinated and orderly approach. Some shifts are informal in nature. As one should expect in any setting, designated judges confer with each other in their chambers or elsewhere about recurring problems. The conditions of these exchanges (when, with whom, how frequently, how detailed, etc.) are unknown, but they may relate to idiosyncratic factors such as shared judicial philosophies or friendship, or they may relate to more objective criteria such as similarities in the facts or legal issues which judges may convey in open judgments and meetings. Presumably, these conversations would relate to the content of judgments whereby the interpretation and application of written secret law is affected by informal interactions.

More concrete evidence of coordination can be found in formal interactions which bear more of the features of community than does ad hoc conversations between several judges. The training of new designated judges and continuing professional development are cases in point--both of which were sporadic and unsystematic up until about 2006 or 2007. Training occurs in two phases. First, the chief justice or a senior designated judge briefs the newly designated judge on such matters as: what secret files look like, formal laws, human sources, and the practices and procedures used in warrant proceedings, certificate proceedings, section 38 CEA proceedings, and section 87 IRPA proceedings. Second, the judge observes secret hearings, and either sits on the bench next to the presiding judge, or at the back. When the judge is ready, the chief justice assigns him to a case.

Designated judges have also been provided two systematic briefings on the nature of security intelligence, provided by CSIS. (130) The Federal Court was concerned that these briefings might (be seen to) compromise judicial independence. As its relationship with the intelligence community was still developing, it was also cautious about the quality of the briefings. It circumvented these problems by creating a panel of eminent jurists, including the Honourable Frank Iacobucci, who vetted the briefing before it was presented to designated judges. Together with senior counsel at the Federal Court, the commissioned jurists would evaluate the briefing along two parameters: (1) to ensure the information was accurate and substantial, and (2) to ensure accessing the material would not impact judicial independence. Once the jurists approved the material, it was presented to the designated judges during two three-hour sessions led by two senior CSIS representatives.

Among other things, designated judges gained a privileged perspective from adjudicating individual applications for warrants, but this view provided only a glimpse and not a global or holistic understanding of the warrant, the underlying operations that may have taken place, and the full implications of the powers when they are executed. Justice Noel recognized that a more comprehensive approach of the entire investigative process would improve the knowledge base of judges and that such knowledge would improve other adjudicative functions. (131) This view is likely animated by a number of recent instances in which CSIS breached its duty of candour in applying for warrants. In the case of X (Re), the Trial and Appellate Divisions of the Federal Court chastised CSIS for misleading a judge in order to secure a warrant for intelligence gathering abroad. (132) The gist of the case was that CSIS attempted to bypass judicial unwillingness to grant warrants for extraterritorial activities by directing foreign intelligence partners to collect and share such foreign information from abroad, omitting this aspect of its operations when requesting a warrant.

More recently, the Federal Court learned in 2016 that CSIS had been illegally retaining extensive volumes of metadata on persons not subject to authorized investigations. (133) Although it did so pursuant to a formal program launched in 2006, CSIS did not inform the Court until 2016. The Court held an en banc hearing with all available designated judges, and assigned two persons listed as SAs (Messrs. Gordon Cameron and Francois Dadour) to act as amici. During this hearing, it was revealed that SIRC recommended that CSIS inform the Court about the program, but CSIS replied it did not need to because the "CSIS Act does not confer any general supervisory authority to Federal Court judges". (134) Justice Noel issued a powerful critique of this position, stating:
The response provided to the SIRC's recommendation by the CSIS shows a
worrisome lack of understanding of, or respect for, the
responsibilities of a party benefiting from the opportunity to appear
ex parte. If the CSIS unduly limits the flow of information the Court
needs to make proper determinations, then the CSIS can be seen as
manipulating the judicial decision-making process. (135)

The use of en banc hearings suggests solidarity and a common identity among designated judges, which contrasts with traditional individualism. The choice to appoint amici at this hearing further evidences a culture shift, since the Court has traditionally preferred not to adopt adversarial methods in warrant applications, a position one can trace at least as far back as the McDonald Commission. (136) As we will discuss below, there are reasons to expect the use of amici will continue. For now, we will note that Noel J and the SAP informed us that designated judges as a group are reflecting on whether and how to restructure the role of amici, including in warrant review proceedings. (137) As part of training and professional development, designated judges were also briefed on Canada's role in global counterterrorism and security as well as international relations. The decision of CSIS to include global issues and international relations in their briefing was sound, as judges should have a firm understanding of how it does and does not relate to national security. One of the more contentious issues in both Canada and the UK is whether information should be kept secret if its disclosure would or may harm international relations. (138) As a net importer of security intelligence, (139) Canada has an interest in sound relations with intelligence partners. But the experiences of the Arar Commission taught us that CSIS over-claims national security confidentiality on this ground. (140) There are arguments to be made that international relations should not stand as a ground of non-disclosure, since this value generally cannot justify the limitation of basic procedural rights. UK proceedings under the Justice and Security Act allow for non-disclosure in civil proceedings only in order to safeguard national security, in part for this reason. (141)

Finally, designated judges meet as a group at least four times a year--sometimes more often. Meetings follow a set agenda where judges discuss, among other things, questions of law, practice, and procedure. Canadian and UK SAs, academics, and other experts sometimes attend these meetings. Reliance on the input of UK perspectives has been steady, although it has waxed and waned in response to changes in our respective laws, policies, and practices. Designated judges have spoken with UK SAs about such matters as administrative support and the nature of closed material proceedings in civil proceedings. They also interact with UK judges and, in one instance, participated in a global conference attended by UK, US, and Canadian judges. (142) Most recently, the National Judicial Institute convened a public conference on secret hearings, which was open to academics, lawyers, sitting and retired judges, and other stakeholders.

C. Summary

The history of certificate proceedings in SIRC highlights that broad discretion and the absence of formal rules of practice and procedure have been the norm in secret hearings in Canada--not the exception. There is some evidence to suggest that discretion was shaped by customary law emerging from the interactions of committee members, SIRC counsel and, in the early years at least, the chair of the committee, the executive director, and the executive secretary of SIRC. Discretion was also shaped by relations of trust, the control that committee members had over SIRC counsel, and the inquisitorial functions and powers of SIRC. Over time, committee members began to decide matters with no outside consultations and, as of 2010, have all but ended ongoing communication between SIRC counsel and outside parties during the secret hearings over which they currently preside.

In some ways, the Federal Court is undergoing a contrapuntal process, whereby solidarity, interaction, and coordination are becoming more valuable commodities. This process is unfolding for reasons that do not relate directly to the express provisions of statutes or, it seems, the four corners of constitutional doctrine. Neither of these sources of law require the abandonment of broad individual discretion; to the contrary, they consistently encourage it and even valorize it as the best means of averting substantive injustice. There is some evidence that the Federal Court is making an effort to magnetize discretion in at least some areas around shared understandings. The conditions for changes to the culture of the Federal Court relate to, among other things, the influence of the traditional SIRC system as interpreted and administered by Lutfy CJ, and Noel J, both of whom were conscious of the customary nature of most SIRC rules. Within the Federal Court, formal interactions such as training, briefings, meetings, judicial conferences, and en banc hearings seem to have contributed to the emergence of interactional expectancies and the codification of customs into written case law--a possibility we will explore in the remaining sections of this paper. There are also informal correspondences among judges about any number of issues relevant to a case. Yet, this change has been subject to path dependency (cultural and doctrinal) and the resiliency of a problem-oriented approach to decision making that prizes supple adaptability to complex, novel, and time-sensitive problems.

As a final note, these practices carry some negative implications that we believe can be addressed through changes to the powers of SAs. The formal and informal interactions noted above are almost entirely confined to the community of designated judges. The sharing of closed judgements occurs only among designated judges, while SAs remain limited to materials internal to their respective files. Government counsel presumably also have total access to closed judgments and can communicate with their security-cleared peers about these materials. This unequal distribution of knowledge inhibits the performance capacity of SAs, but it also results in unequal distributions of law-making power. Unaware of the existence or content of bodies of secret law, SAs cannot share in its interpretation, criticism or conscious change, thereby limiting the responsiveness of such law to the maximum possible range of expectations, interests, and relations. At root, this means that decisions rest on legal materials SAs do not know about and cannot reckon with; this, of course, is precisely the problem the SA system is meant to correct. It should be clarified that exclusion from secret law extends beyond informal conversations among designated judges about (closed) judgments, which is a common practice in any legal field. (143) The core concern is that there are clusters of secret law within secret law, which SAs cannot access, interpret, invoke, challenge, or consciously shape.

IV. The Conditionality of Special Advocate Powers: Material, Epistemic, and Discursive Barriers

We will now turn our attention to the process by which SAs have adapted to the formal and informal components of the SA system. As noted above, the core criticisms of the UK system have been material, discursive, and epistemic in nature, with SAs lacking "the resources of an ordinary legal team for the purpose of conducting a full defence in secret". (144) Central to this problem has been the lack of disclosure or of timely disclosure, the lack of training, inability to correspond with other SAs about shared problems, and limited administrative support. (145) We have also touched upon the negative impact of the government's tough stance on tainting, where SAs with prior experience in matters relating to a prospective file are barred from serving as SAs. Sometimes justifiable from a security perspective, the practice limits the access of named persons to SAs whose knowledge and experiences are best suited to a case, with appointed SAs having a rather steep learning curve and little time to adapt. (146)

Finally, we highlighted early doubts about the independence of SAs, with affected parties suspecting some were closely aligned with the government. (147) This section surveys the extent to which these problems have been an issue in the Canadian SA system.

We would like to be clear that this section aims to contrast the position of designated judges, who have been able to shape the conditions of their adaptation, with that of SAs, who have had to negotiate most of their claims from positions of disempowerment. The subsections that follow will outline the nature of conditionality, where SAs have had to work hard to negotiate and apply essential powers. We will highlight some of the means by which the work of conditionality has been eased, with specific reference to the supporting role of the SAP. In the final analysis, this section will reveal two things. First, there are sub-communities within the community of secret law, each of which has found ways to address matters of concern to them. The move towards coordination found within the Federal Court has yet to transcend the boundaries demarcating these sub-communities, and so SAs have turned to other institutional resources. Second, though, there are indications of the emergence of customary law that responds to interactions between SAs, the Federal Court, and the SAP. Some of this law addresses the material and epistemic barriers faced by SAs, but more progress is needed.

A. Selection and Training of Special Advocates

Section 85(3) of the IRPA vests the Minister of Justice with responsibility for ensuring that SAs receive "adequate administrative support and resources". (148) In 2008, the executive branch established the SAP, which is an independent body housed within the Department of Justice. The SAP is rather small, for most of its history consisting in one senior counsel, one counsel, one administrative assistant, and one half-time financial advisor. (149) Among other things, the SAP: solicits and selects SAs; provides SAs access to sensitive information, case law, and related materials; shares in the provision of legal, professional, logistical, and administrative support (along with the CAS); provides ongoing professional development and training to SAs; and attends all secret hearings involving SAs. (150)

The first step taken by the SAP was to establish a list of persons who may act as SAs. There has only been one appointment process thus far, which began with the dissemination of a formal request for expressions of interest in December 2007. The request outlined eligibility criteria, which included being a member of the bar for at least ten years and having significant litigation experience. There was an expectation that the successful candidate would have experience in immigration law, criminal law, national security law, or human rights law. No mention was made of language skills, ethnicity, or gender. (151)

An assessment committee reviewed the applications and issued recommendations to the Minister of Justice. Former Federal Court Justice William MacKay chaired the committee, which was also composed of representatives from the Canadian Bar Association and the Federation of Law Societies. The committee recommended twenty-eight individuals, all of whom were excellent lawyers with a proven record of independence; all recommended persons were listed as SAs. We have been informed that one SA speaks both Arabic and English, while a good number speak both French and English. There were only three women among the initial group. Two retired for reasons unrelated to the SAP. Of the three, only Ms. Nancy Brooks was appointed to a certificate proceeding (in the Harkat file), although she withdrew several months thereafter. (152)

With some exceptions, SAs had little prior experience in Division 9 or in security matters. Most had backgrounds in criminal or civil litigation and some with immigration law, in either case carrying with them a powerful command of the workings of adversarial systems. But the sui generis nature of certificates meant that initial training was essential. This began shortly after the appointment process in 2008. Initial training consisted of three one-week blocks of meetings, which were divided between open and closed sessions. Briefings covered several core themes. First, there were briefings from persons with experience and expertise in Division 9 proceedings, including a designated judge, government counsel, and outside counsel (including Lorne Waldman). Second, there was discussion of the legal and institutional frameworks of security intelligence agencies, including CSIS and the Canada Border Services Agency (CBSA). Some of these sessions were delivered by academics, such as Professors Craig Forcese and Wesley Wark. Finally, there were closed briefings organized by CSIS and, to a lesser extent, the Royal Canadian Mounted Police. These resembled the briefings designated judges received, including an overview of the intelligence process (how it is gathered, analyzed, stored, shared, etc.) and the global security and threat environment. (153)

Because SAs are dispersed geographically, involved in full-time private practice, and assigned to different files, the SAP endeavoured to build a sense of community and shared knowledge. The primary means of doing so has been an annual, day-long meeting among SAs, including those working on certificate files or as amici as well as persons listed as SAs but who have not yet been called upon to perform either task. Discussions are structured, revolving around research papers solicited by the SAP with recurring issues and problems in mind. Sometimes, SAs present papers. Mr. Norris, for example, shared his knowledge about the differences between the abuse of process doctrine in criminal versus Division 9 proceedings. (154) Outside counsel sometimes participate. Norm Boxall, who represented Mr. Harkat as outside counsel, shared a paper on legal strategies and arguments he used, and how outside counsel contend with communication bans. Sometimes sessions contain a comparative law focus, with UK SAs and academics being among the most frequent participants. (155)

Finally, the SAP maintains an electronic portal of open and closed material that SAs may access. Open material includes both commissioned and refereed research papers, webinars, government reports, and a complete list of open court judgments relevant to certificates. There is also a list of international, regional (European Convention of Human Rights) and foreign case law (principally UK, Australia, and New Zealand). Closed material consists of material, some of which all SAs may access, some of which only persons working certificates may access, and some of which only persons working on a particular file may access. (156)

B. Appointments, Tainting, and Knowledge Sharing

Due to retirement or death, there are now only twenty listed SAs. Of this number, only a handful have been appointed to certificate proceedings. A small cadre of five persons have done the bulk of work, these being Messrs. Gordon Cameron, Anil Kapoor, John Norris, Paul Copeland, and Paul Cavalluzzo. These five have collectively handled the Almrei, Mahjoub, Harkat, and Jaballah files; several other SAs worked on these files for brief periods of time. (157) Messrs. Denis Couture and Francois Dadour were involved in the Charkaoui file, but this ended in December 2009 when Tremblay-Lamer J quashed the certificate. (158)

There is no indication that named persons or their counsel doubted the credibility of the system or the independence of SAs, as was the case in the early years of the UK system. It is unclear why named persons selected the above-mentioned SAs to represent them, although one can surmise from the list that extensive litigation experience was one criterion. The SAP and the Federal Court have established the custom of having two SAs work as partners on a file. (159) Because five SAs have done the balance of work on four certificate files (Charkaoui being the exception), the team approach has contributed to knowledge sharing. For example, Mr. Cameron has worked with Mr. Copeland on the Almrei file and Mr. Kapoor on the Mahjoub file. Mr. Cavalluzzo worked with Mr. Copeland on the Harkat file and Mr. Norris on the Jaballah file. In this way, every file has had at least one SA who has also worked or is working on another file.

The extent to which a small cadre of SAs handle the entirety of certificate files highlights a relatively relaxed position in Canada on the question of tainting. We are aware of only two ministerial objections to the appointment of SAs. The first occurred in 2008 and related to a perceived "conflict of interest" on the part of Messrs. Norris and Copeland. Prior to the introduction of the SA system, both had acted as outside counsel for named persons and were serving as outside counsel in the Iacobucci Inquiry. The government objected to the two maintaining their role as outside council if appointed as SAs, citing the principles of the informed reader and the mosaic effect. (160) The government dropped its objection when Messrs. Norris and Copeland agreed to withdraw as outside counsel in these cases. The other related to the appointment of two SAs (Messrs. Copeland and Cavalluzzo) to the Harkat file. After initially accepting the appointment of two SAs, the government changed its mind and filed a motion before the Federal Court of Appeal to prevent it. The Court of Appeal allowed the original motion to appoint two SAs to one file, (161) which has now become standard practice in all files save for that of Mr. Jaballah.

We infer from this that there is customary law governing the appointments process, which reflects shared understandings among judges across all files, government lawyers, and of course, SAs. This inference is supported by the Federal Court's skepticism of the mosaic effect in the case of Khadr v Canada (Attorney General) (162) and with respect to the Arar Commission in Canada (Attorney General) v Canada (Commission of Inquiry into the Actions of Canadian Officials in Relation to Maher Arar). (163) It is also supported by the fact that the UK is far more vigilant about the risk of tainting, even though our legislative regimes and security concerns are similar. The more important implication for our purposes is the blurring of epistemic barriers between files which resembles the perforation of judicial silos in the Federal Court. The overlapping of SAs across all files means that, on any one file, you might have collective knowledge of three (ongoing) or four (past and ongoing) files. This would include not simply the facts of a case, but knowledge of the styles, decisions, and procedural preferences of different judges. It is likely that acting SAs are able to use this experience to predict and influence the dynamics of the regime, even though they do not officially have access to a library of all closed judgments; to the extent there are general rules of practice and procedure, most SAs would encounter them as such. To the extent there are social or ideological patterns to the individuated decision making of any given judge, high-intensity workloads improve the adaptability of SAs. The pairing of SAs obviously hastens adaptation.

There is evidence of customary law that encourages the acquisition and sharing of knowledge. In 2009, the Federal Court authorized SAs to participate in knowledge-sharing sessions organized and supervised by the SAP. This tradition began when Messrs. Cavalluzzo, and Copeland (advocating on behalf of Mr. Harkat) sought permission from Noel J to correspond with SAs appointed to other proceedings on matters of substantive and procedural law. Justice Noel authorized the request "to communicate with other special advocates (who have obtained the same judicial authorization from their respective designated judge) appointed in other security certificate proceedings to discuss common issues related to questions of jurisdiction, procedure, and substantive law and orders rendered or orders to be sought". (164)

Justice Noel expressly required the SAP to organize and supervise these meetings (165) and, to be clear, participating SAs had to secure permission from the designated judges presiding over whatever files they might be working on. According to our interviews with representatives of the SAP, there have been seven knowledge-sharing sessions thus far, which have included discussion of both substantive and procedural law and, interestingly, strategies for handling particular designated judges. (166) This latter subject highlights the impact of the Federal Court's philosophy of individualized adjudication, with the idiosyncrasies of a judge playing a prominent role in decisions.

There is no formal right or authorization to participate in knowledge-sharing sessions, and certainly no order from an appellate court. The fairly regular occurrence of these sessions signifies agreement among designated judges that appointed SAs can be trusted to speak with each other. We should note that the UK adopted a similar practice for a time, with the SASO supervising sessions. We were told that these sessions have not occurred in several years, owing to a combination of resource constraints and concerns within government about the dissemination of closed material, if even among SAs. (167)

C. Disclosure

We will end by reviewing the role of the SAP in facilitating access to protected information. As noted above, the Supreme Court of Canada in Charkaoui II required CSIS to provide designated judges and SAs with all information on file relevant to the named person. (168) Formally, the Minister of Justice is responsible for ensuring that SAs have adequate administrative support and resources, which includes access to protected information. But since this information is filed with the Federal Court (or the IRB, depending on the venue of a hearing), the SAP had to liaise with the CAS and the IRB to coordinate the relative roles of these institutions. To be clear, material used in hearings before the Federal Court or the IRB is always stored in these locations; it is not kept in the SAP office and certainly is not kept by SAs themselves, in contrast to the practice in the UK. (169)

In April 2008, the SAP and CAS signed a memorandum of understanding, whereby the CAS agreed to assume primary responsibility for providing secure access to secret material, although the SAP was responsible for helping find solutions to problems reported by SAs. (170) In the same month, the SAP and the CAS met with SAs to discuss the nature and extent of support they expected they would need. The SAP and CAS indicated what they were able to provide, and were open to exploring ways of enhancing support to meet "the reasonable needs" of SAs. (171) DROs of the CAS, the SAP, and SAs told us that the CAS was not prepared to handle the demands of the SA system. (172) Before Charkaoui II, litigation proceeded on the basis of a moderate volume of written or hard copies of materials. However, Charkaoui II disclosure included voluminous quantities of information that was composed mostly in digital format. Due to inadequate document management software, SAs had major problems accessing files. First, the separate computers used by a team of two SAs could not be linked together, so that changes or notes made to a file on one team member's database did not appear on the other's database. The SAs would have to print up changes and hand them to the team member, who would manually input the changes on his computer. Second, SAs could not employ searches in the files, as the electronic copies were photographs of originals and not amenable to reliable coding and recognition; sometimes there would be random spaces in script and other times spelling mistakes, e.g., an "I" was coded as "1". For example, the key word "Almrei" would yield only a portion of the times the word was used in files, with SAs having to try different spelling variations at random. (173)

The Federal Court was seized of the matter. Justice Blanchard--then the designated judge in the Mahjoub file--required the SAP to provide two progress reports on how the problems were being addressed. The SAP and CAS eventually solved the problems. The SAs with whom we spoke all said that their designated judges were willing to postpone hearings until they could adequately prepare, suggesting customary practice. (174)

Charkaoui II disclosure also prompted some doctrinal confusion. (175) One issue was whether all information regarding a named person had to be disclosed or, in the alternative, if only information relevant to the certificate or allegations made therein had to be disclosed. In one case, the Federal Court held that the government is only obligated to disclose to SAs such secret material as is "necessary to examine and verify the accuracy of the information submitted" against the named person, i.e., evidence supporting its allegations. (176) In later cases, this resulted in the denial to SAs of access to records about persons or organizations to which a named person was linked, but which did not form part of the investigation into the named person himself. (177) Several SAs with whom we spoke were dissatisfied with this ruling. They thought it is always an open question whether an SA can make constructive use of information (only) the government possesses. (178)

Courts and SAs have had to contend with several instances of non-compliance by the intelligence community. In the Harkat file, CSIS failed to disclose that one of its officers had made changes to the results of a lie detector test in order to strengthen the credibility of a human source. (179) In the Almrei case, the Court discovered that both CSIS, the Minister of Public Safety, and the Minister of Immigration failed to disclose material unfavourable to their case, including information impinging the credibility of human sources. CSIS was also found to have consistently failed to explore, assess, or share new information that challenged the original basis for the certificate. Government lawyers defended this inactivity by arguing there is "no requirement" for the government to "advance a case against a finding of inadmissibility". (180) If accepted, this kind of "tunnel vision" reasoning would allow the government to shirk its responsibility to disclose exculpatory information by failing to collect it in the first place. Justice Mosley rightfully rejected this position and, moreover, found that the Minister of Public Safety, the Minister of Immigration, and CSIS breached their duty of candour to the Court. In the final analysis, the (undisclosed) exculpatory information undermined the government's case that Mr. Almrei posed a current or continuing threat to the security of Canada, and so Mosley J quashed the certificate in December 2009.

Parliament reacted by amending the IRPA through Bill C-51, significantly changing the dynamics of disclosure. (181) Section 77(2) of the IRPA now states that the Minister of Public Safety must only disclose to designated judges information and other evidence "that is relevant to the ground of inadmissibility stated in the certificate and on which the certificate is based". (182) Further, section 85.4(1)(a) states that the Minister of Public Safety only has to share with SAs information that the Minister considers to be relevant to the named person. (183) This means that the government has a wider range of discretion to withhold information from SAs on the basis of perceived irrelevance. Even more concerning is section 85.4(1)(b), which states the Minister of Public Safety does not have to even share "relevant" information that it has not filed with the designated judge, unless he is ordered to do so. (184) Notice that this section expressly states that SAs may be denied access to relevant information in the possession of the government. This includes information that is exculpatory or unfavourable to the government's case. It has been suggested that this provision would have led to a different result in the Almrei file, as SAs relied on information that government counsel would now be statutorily authorized to withhold on the basis of irrelevance. (185)

Under section 83(1)(c.1), judges may even "exempt" the Minister of Public Safety from disclosing relevant information under section 85.4(1)(b) because it "does not enable the permanent resident or foreign national to be reasonably informed of the case made by the Minister". (186) This is a confusing provision, since disclosure (of open material) to the named person is a different matter entirely from disclosure (of secret material) to SAs; it was always the case that the Minister of Public Safety could withhold relevant information from named persons, subject to the minimum core of disclosure required by Harkat. These are quite different matters: whether or not a named person is reasonably informed of a case, the effectiveness of SAs is conditional on access to relevant information. To make matters worse, section 83(1)(c.2) cuts SAs out of the process for deciding whether the Minister of Public Safety should disclose relevant information, stating that a judge "may" allow SAs to make submissions on exemptions under section 83(1)(c.1) if "fairness and natural justice require it". (187)

These amendments are clearly meant to test the ambits of Charkaoui II disclosure. The timing is poor, considering CSIS' multiple breaches of candour. We should be clear we are not suggesting the government will deliberately withhold information. The point is that disclosure is subject to a technical, mechanistic threshold triggered by the evaluations of government counsel and not SAs, who are best positioned to determine if information would be relevant. Secret hearings are already unfair enough without having to restrict the flow of information that appertains both to the named person and his activities and relationships during times covered by government allegations, as narrow as those allegations may be construed. Indeed, there may be an incentive to tailor allegations to reduce the flow of exculpatory information to SAs and designated judges. It is notable that Parliament chose not to address these problems in Bill C-59, which will make broad-sweeping changes to the entire federal security apparatus. (188)

We should add that SIRC counsel with whom we spoke encountered similar problems with accessing relevant information in the recent past. (189) As noted, SIRC is entitled access to the entirety of CSIS records, save for cabinet confidences. Until recently, access to CSIS records was mediated by a liaison officer who lacked familiarity with the nuances of a particular investigation. The nature of the material and the fact that relevance is contingent on the unique nature of a case meant that relevant information was not always produced. SIRC counsel could bypass this problem by renewing more specific requests, but this took extra time and was possible only if they had virtually untrammeled access to CSIS records. SIRC now has enhanced, digital access to records, so it can search the entirety of CSIS records independently of liaison officers. It would be simpler in the certificate context to just return to the practice of providing SAs with all relevant materials, and let SAs decide what is useful. But if the new, restricted disclosure rules are kept, decisions of relevance should be made by an independent and experienced party, such as SIRC.

For obvious reasons, we cannot independently assess whether disclosure of sensitive material has been adequate. However, many of the SAs we interviewed were generally satisfied that the government has complied with its disclosure obligations, the above-noted instances (and probably a few more besides) notwithstanding. (190)

D. Administrative Support

Access to a searchable database and judicial alertness to non-disclosure or over-claiming of national security confidentiality has been helpful, but the sheer volume of disclosed material has at times been overwhelming. The biggest practical problem is that SAs must conduct their own clerical work, such as photocopying, composing factums, conducting outside research, etc. In the first few years of the SA system, the government strongly opposed the appointment of administrative support personnel, stating that SAs did not need the support and, in any event, that widening the scope of persons authorized to access material was too risky. (191) Justice Noel rejected the argument in a 2009 Harkat proceeding, noting that SAs are to be "in the same position as counsel for the Ministers", as far as is possible. (192) He also noted that the absence of administrative support would prolong proceedings to the prejudice of Mr. Harkat. He ordered that an administrative assistant be assigned to assist Messrs. Cavalluzzo and Copeland in organizing and summarizing secret material. (193)

While this ad hoc order is helpful, most SAs expressed a desire to have permanent administrative support that is not conditional on judicial discretion. (194) They criticized the government for reflexively challenging reasonable requests, which both stalls proceedings and creates a disincentive for SAs to seek judicial authorization. Interviews with government lawyers would have shed more light on this issue. Several SAs suggested a solution to this problem: an independent and external body (e.g., SIRC) could replace the CAS in providing SAs with secure and protected access to secret material. (195)

E. Summary

In sum, SAs have had to overcome a wide range of material, epistemic, and discursive barriers in order to access the powers promised to them. Many of their claims have been mediated by the SAP, which has brokered discussions between and among SAs, designated judges, the CAS, CSIS, and, to a lesser extent, government lawyers. After some initial roadblocks, the SAP has ensured meaningful access to sensitive material, and runs an open database that contains Canadian, foreign, and international case law, as well as academic materials. It is engaged in important, non-chirographic work as well, engineering or supervising interpersonal knowledge-sharing sessions among all SAs and court-authorized sessions among appointed SAs. This latter role elevates the position of SAs vis-a-vis governments while giving SAs access to juridical and other material essential to effective advocacy.

Because the SAP is present at all secret certificate hearings, it has total awareness of the factual and legal issues that arise. We can infer that this has helped it guide knowledge sharing in ways that are beneficial to SAs. Being trusted by the Federal Court and the government, we might also infer that the SAP's supervisory role has helped increase judicial trust in SAs or at least somewhat grounded anxieties about inadvertent disclosure.

Lest we be misinterpreted, we should clarify that SAs remain at a considerable disadvantage relative to government lawyers. Our point is that SAs have had to press for many of the resources they now have, and that this effort has for the most part been strongly supported by, and mediated through, the SAP. This is true in logistical terms, but more importantly, the supervisory role of the SAP seems to ground judicial concerns about inadvertent disclosure. While this has some advantages, SAs remain subject to constant supervision and control.

V. Trust, Control, and the Limits of the Special Advocate System

Trust and control are recurring themes. This section explores the role of the two in the process by which SAs claim and negotiate: (1) the power to communicate with named persons or outside counsel, and (2) the power to call expert witnesses. We should make one methodological note: access to judicial orders and directions in this area was initially limited, since the vast majority of them are not reported. Through the SAP, we received access to all 518 open decisions issued across the five certificate files. (196) Interestingly, only 153 are available through a search of the Canadian Legal Information Institute database. (197) This on its own presents a rule of law and open court problem that the Federal Court should consider resolving, although it should also be reiterated that all SAs have access to all 518 of these decisions.

A. Communication Between Special Advocates and Named Persons

We have noted that SAs in the UK generally find communication bans to be the greatest obstacle to effective advocacy; Canadian courts and legislators knew this when constructing the constitutional and statutory architecture of the SA system in Canada. Parliament's decision to include a judicial authorization procedure prompted an immediate Charter challenge in late 2008. (198) Representing Hassan Almrei, Lorne Waldman argued that SAs should be free to determine for themselves whether communication with named persons would occur, and courts should trust that sensitive information would be protected. Mr. Waldman alluded to SIRC as an example of how this process would work effectively. Interveners in Almrei (Re) added that, if an authorization process is used, it should occur ex parte the government and before a designated judge other than the presiding judge.

Chief Justice Lutfy declined to decide these issues on the ground that they were hypothetical--the regime was just starting to operate, so there was no basis upon which to decide one way or the other. (199) He was not convinced that the SIRC model provided enough empirical traction to reformulate the SA system. The interesting feature of the SIRC model is that there actually were no formal rules of practice and procedure governing communication under SIRC. We noted above that norms emerged incrementally and were purely customary in nature, and that SIRC counsel would generally communicate only after being directed to do so by a committee member--and with conditions. Chief Justice Lutfy thought that the legislative framework for the SA system was conducive to the emergence of a similar dynamic, so that the powers of SAs, "like that of SIRC counsel, will evolve based on the rulings of presiding judges". (200)

Comparisons with SIRC are inapt, in our view, since customs there emerged through frequent, sustained, and intense interactions of members of a small community, in which committee members trusted, as well as possessed considerable control over, SIRC counsel. According to the late Ron Atkey, SIRC "trusted counsel to sort of say 'You're under an oath. You can't disclose this information or in any way indicate.' And we never had any difficulty." (201)

The same sort of dynamic was observable in the Arar Commission, where the power of Commission counsel and amici to communicate with outside parties and each other throughout the proceeding rested on trust and a relationship of dependence with the Commissioner. (202) The community dynamics in certificate proceedings do not exhibit the level of trust that seems to have existed in SIRC. In addition, unlike SIRC counsel and Commission counsel or amici, who take instruction from the committee members and the Commissioner, SAs are independent of the designated judge. (203) This institutional setting was bound to produce different customs.

The undue nature of Lutfy CJ's optimism is evident in documentary records pertaining to the judicial authorization process; perusing all court orders in the five certificate files, we found only four authorizations to communicate on substantive matters--ironically, none were filed in the Harkat case. (204) To be clear, the nature and extent of authorization varies, depending on the type of communication sought. Across all certificate proceedings, designated judges have provided SAs with blanket orders to communicate with outside counsel on procedural matters, such as scheduling or timing. (205) But the willingness to authorize communications about more substantive matters "very much depends on the judge. Some are very stringent, imposing lots of restrictions and requiring written or oral reports on what was talk[ed] about." (206) There is a general protocol for every designated judge to "vet all communications between the special advocates and the named person and/or his counsel", on the grounds that even "the smallest risk of inadvertent disclosure must be of concern to the Court". (207)

Neither Lutfy CJ nor the Court in Harkat recognized the structural disadvantages faced by SAs. The Court's misreading of the situation is abundantly clear in its directive that designated judges be liberal in granting communication requests. We have not seen or heard any evidence that worries about being denied is the reason SAs do not make requests. When we asked SAs why they do not make requests, the universal answer was that the procedures for making a request risk prejudicing the named person: if the SA tells a judge what she is looking for, then does not rely on information produced from this interaction, the designated judge may infer something inculpatory was found. (208)

It may take a moment to understand how this is a problem. In a conventional adversarial setting, a trial judge would only hear and decide upon evidence submitted by counsel, who obviously would have thought carefully about all the ways certain evidence and the way it is presented can help or hinder their case. In many instances, he would have to make a judgment call about the value of a piece of evidence or a style of argument, deciding against including information that may be misinterpreted, incomplete, or in some way used against them by opposing counsel. By contrast, an SA never knows in advance what information he will receive from named persons when he requests permission to communicate, although he will hope to find something sufficiently precise to support a request. If he has to tell the designated judge the sort of evidence he is looking for and why, then he has lost the freedom to control his litigation strategy and, practically, to control what information a judge receives--or gathers through inference.

Judges would of course strive to remain objective by not drawing or relying on negative inferences. But the fact that SAs would rather not seek permission to communicate for fear of prejudicing the interests of named persons is proof that this assurance is not enough. In particular, it indicates that SAs distrust judges to a certain degree, and likewise insist on maintaining control over what they consider to be protected information. One SA observed that Harkat did not provide a "doctrinal response" to the issue of communication requests, which remain a function of "familiarity and trust", stating that it was a "purely sort of a club-y kind of response. Like, you know, I'm sort of in the club, so they, they can trust me, because I know the secret handshake." (209)

This observation underscores conditionality, where SAs have to work at cultivating relations of trust before they can wield the powers of an ordinary advocate. One SA reported receiving (along with the other SA working on the file) authorization to communicate in two instances that exemplified a high level of trust on the part of the judge, given the SA had already accessed confidential information. The first was an authorization to communicate with an academic expert about "complex politics behind the terrorism allegations". But the SA was careful to point out that the request was only filed because he and his partner did not have to disclose anything related to their legal strategy, nor did they have to report back to the judge. In the second example, the judge granted the SA's request to speak with the named persons "one-on-one" (in the absence of the outside counsel, with the agreement of the named person and his counsel). (210)

But these requests have been the exception. There are enormous structural impediments to claiming and negotiating particular powers, which include the distorting impact of constant supervisory control over the flow of information. Even if there is enough trust to secure authorization, SAs remain subject to the supervision of the SAP and, depending on the judge, may have to satisfy stringent conditions, including reporting on precisely what a named person has said. The idiosyncrasies of judges, and the SAs' own strategic interest in controlling their own information, profoundly affect the distribution of power in certificate proceedings.

B. Expert Witnesses

The final issue we explore is the role of expert witnesses. Some SAs told us there are occasions when they could use an expert in security intelligence for two reasons. First, expert witnesses would be valuable in challenging claims of national security confidentiality. Across the board, we heard from SAs that their primary responsibility is to have relevant information pushed out, so that outside counsel can participate in proceedings as fully as possible. As one SA put it to us, "the best case scenario is if there's no secret evidence". (211) An expert witness could help debunk claims that certain information cannot be safely disclosed. The second reason is that, short of this event, expert witnesses could help in challenging evidence behind closed doors. We would add that UK SAs have also complained about the inability to call expert witnesses. (212)

As with their UK counterparts, Canadian SAs are reluctant to request authorization for expert witnesses. This is not for want of need, nor was the problem formal inability. Officially, an SA could apply for authorization under section 85.2(c), which states an SA may "exercise, with the judge's authorization, any other powers that are necessary to protect the interests of the permanent resident or foreign national". (213) This provision would work in tandem with authorizations to communicate about a proceeding to a third party, and the authority of the judge to make orders that do not lead to a serious risk to national security or the safety of any person.

As with communication requests, the decision to not make requests is strategic. The general consensus among SAs was that attempting to call expert witnesses in most cases would lead to a protracted argument about whether secret material should be disclosed to a third party. (214) In most cases, the expert would have to be security cleared. Special advocates anticipate that government lawyers would contest requests, much as they contested applications for administrative assistants. Given the limited time and resources SAs have to work on complex files, they have not found this to be a worthwhile strategy. Efforts should be made by courts, government lawyers, the intelligence community, and the SAP to find ways of enhancing accessibility to expert witnesses.

By contrast, there have been some cases where outside counsel have called witnesses. Justice Blanchard permitted counsel for Mr. Mahjoub to call Professor Wesley Wark and former Canadian diplomat Henry Garfield Pardy as expert witnesses, despite vigorous opposition from the government. (215) He also allowed Mr. Mahjoub to subpoena Mr. Ted Flanigan (former Manager and Assistant Director of CSIS) and Mr. Michael Duffy (former Senior General Counsel with CSIS Legal Services). Representatives for Messrs. Richard Fadden (former Director of CSIS) and Stephen Rigby (former President of the CBSA and National Security Advisor to the Prime Minister) agreed to attend the hearing on consent. (216) In Jaballah (Re), Dawson J ordered the testimony of a CSIS witness who testified in an open hearing and in the view of the Court and counsel, but was shielded from the public and Mr. Jaballah. (217)

VI. Conclusion

The deep history and increasing use of secret hearings are a challenge from the standpoint of legal philosophy and everyday constitutional work. For some, they necessarily stand outside the constitution, with courts providing a mere veneer of legality over executive practices that remain effectively unconstrained. (218) For others, courts play a valuable role in reviewing and overseeing security intelligence practices; while this carries significant and perhaps unresolvable implications of legal principle, describing the judiciary as mere instruments in the rationalization of exceptionality is inaccurate, reductive, and unhelpful. The role of courts is made more complex when one recalls that their role is triggered by legislative provisions that reflect the exigencies of the intelligence community. But legislative provisions hardly determine the role of courts, which always possess the authority to submit legislation, executive practice, and the common law to Charter review. And of course, legislative ambiguities about rules of practice and procedure, as well as the inherent authority of courts over their own procedures, provide judges with ample room to fashion fair approaches to the administration of secret hearings. It goes without saying that this helps explain acute differences between the Canadian and the UK SA systems, despite the fact that legislative provisions are similar.

One of the difficulties here is that reasonable people can disagree about what the law allows and thereafter what are the best rules to use. A distinct, less fundamental or ideological problem is that we lack an evidentiary basis for making an informed decision about whether secret hearings might in some contexts be justifiable relative to even a formalist conception of the rule of law. Grand claims about states of exception or, conversely, about the constitutional stability and worth of secret trials are not subject to falsification when practices are, at best, known to a small cadre of judges or lawyers, but not the public. (219) Surely our starting point ought to be one of skepticism about the legality of secret hearings, and our expectation should be that legal institutions do as much as they can to demonstrate how they are shoring up deficiencies in rights, the open court and open justice principles, and the rule of law. The SA system has been the best courts and legislatures have been able to offer, in lieu of full disclosure and adversarial challenge.

Framed by a formalist conception of the rule of law, this paper has searched for evidence that the SA system helps ensure that decisions of procedure and substance cohere with public law. Our hypothesis was that the powers of SAs, which are a critical means to this coherence, are likely to be conditional on a host of informal factors that shape the discretionary power of judges in diverse and sometimes incongruous ways; legislation and high law, in other words, provide for the possibility of fairness, but the operation of the SA system is a function of material, epistemic, and discursive processes that can either reinforce or distort public law. We expected to find a host of variables, including relations of trust, control as a grounding for anxieties about inadvertent disclosure, the idiosyncrasies of individual judges, the institutional culture of the Federal Court, and differing levels and kinds of power as between SAs and government counsel. But we also expected that customary law would play a distinct role, easing the work of conditionality by producing stable interactional expectancies about what powers SAs would receive, as well as when, how, and why. It would be reasonable to suppose that a significant component of this law would be secret, appertaining to the facts of a case.

The first hypothesis was supported, while the latter hypothesis was not as well supported, although this may have much to do with our inability to fully observe what goes on behind closed doors. On the whole, interactions between diverse public laws and judicial discretion with respect to SA powers is byzantine and, in many respects, unmediated by general and stable rules. There is no body of secret case law equally available to all stakeholders, no real system of appellate review of secret decisions, and no formal rules of practice and procedure unique to certificate files. This is not surprising--none of this existed in the SIRC system, nor any other example of a secret hearing in Canada or abroad. But there is evidence of customary law in the Federal Court that has grown from the repeated interactions of all stakeholders, most notably SAs, designated judges, and the SAP; this law has in several instances arisen from and reinforced interactional expectancies about such matters as the sharing of knowledge among SAs, tainting, procedural remedies for delayed disclosure or other practical difficulties faced by SAs, solicitor-client privilege, (220) and communication with named persons or outside counsel about procedural matters. This dynamic was observable in the SIRC system as well, suggesting either that the culture of the Federal Court does not determine the operationalization of law, (221) or that the court has absorbed some of the customs of SIRC. In any event, these customs have for the most part eased the work of conditionality and should be counted as an essential feature of any fair system.

To an extent, stability is also provided by open judgments and redacted reports of closed judgments. In the 2016 en banc hearing in X (Re), for example, judges publicly reproached operational failures and procedural unfairness while advancing legal clarity in the areas of candour, warrant applications, and the retention of metadata. (222) This highlights a genuine and important shift in the culture of the Federal Court characterized by the perforation of judicial silos and the emergence of shared understandings. The hearing also suggests that general, written and unwritten rules have emerged and carry weight in the context of CSIS warrant applications. But the fact remains that only a subset of persons involved in security certificate proceedings have access to the full sweep of written and unwritten law. While there are shared understandings and interactional expectancies among designated judges, the implications this has for the rule of law are delimited by the fact that SAs are not privy to much secret law. They are excluded from the inner circle of what is already a secretive community.

And then we are faced with the fact that public law itself is astoundingly vague and serves more as a permissive framework for broad discretion than a source of normative and interpretive constraint. The IRPA is virtually silent on critical questions of practice and procedure because Parliament chose to offload this responsibility to designated judges; we saw the myriad obstacles to the production of formal rules of practice and procedure that have to this day not been effectively surmounted. In the meantime, constitutional issues related to the powers of SAs have not been squarely addressed and continue to arise with regularity. (223) The tendency has been for appellate courts to accept the legality of broad and/or vague legislative provisions in the hope that designated judges will get it right on a case-by-case basis. What remains resoundingly clear is that appellate courts are in no position to review secret decisions for consistency with public law and, in any event, seldom do. We are certainly not arguing that appellate courts should rely on secret evidence; we claim only that the absence of an adequate system of appellate review and of a simple compendium of secret case law accessible by all stakeholders shakes confidence in a discretionary approach to procedural fairness.

But perhaps the biggest factors affecting conditionality are anxieties about inadvertent disclosure, relations of (dis)trust, and struggles for control over information. There is no question that the anxieties of the intelligence community form the substratum of the SA system by virtue of statutory law as well as the position of government counsel on procedural issues. Lack of trust, and a desire to control information, knowledge, and power filters down from the government and percolates up from within separate files. This is laid bare in the judicial authorization process. We noted already how one SA referred to the negotiation of SA powers as a function of "familiarity and trust", of being "in the club" and knowing "the secret handshake". (224) Another SA questioned why trust should need to be earned anew for each SA and each judge, given that protecting confidential information is
part of a lawyer's training, because, what you do in a cross
examination is try to get people to give you information, without
telling them what you know and that's what lawyers are trained in.
Special advocates have to have twenty years of that experience before
they venture into this task, whereas government employees, such as
people doing immigration briefings and things like that, might have two
weeks of training, and a couple of years experience as a border guard,
and then be handed a classified brief and be told to interview some
immigration applicant or refugee applicant about matters that include
allegations coming from confidential sources or foreign agencies. And
they are trusted with that task. (225)

This sentiment was shared by a third SA, who criticized the government for opposing the assignment of junior lawyers to assist SAs on the grounds that junior lawyers could not be trusted: "[I]t doesn't make sense. I mean half of CSIS lawyers are junior lawyers." (226) A fourth SA agreed that: "[T]o some degree, the concerns about inadvertent disclosure are overblown. But at the same time, I'm happy not to be put in a position of making that mistake, and I, you know, um, you never know whether somebody might actually been trying to mine you for information." (227)

But there is also evidence that the work of conditionality has been eased by the SAP. The institutional supports provided by the SAP has helped steer the Canadian system away from some of the pitfalls of the UK system. Canadian SAs have reliable access to relevant information, including searchable databases to both closed material relevant to their file and a broad range of useful primary and secondary open materials. One SA endorsed the epistemic benefits of the SAP:
[I]n terms of creating the portal of information, creating some young
lawyers working in the system, who will create these portals, get this
research memoranda. You know, and the Harkat decision came out, for
example, within days we got an analysis of staff, not in the justice
department but of special advocates group of the case and the relevance
of what we do. (228)

Conflicts over knowledge sharing are perhaps the best example of how the work of conditionality has been eased, eventually concretizing into custom. Well before they had earned the trust of individual judges, SAs sought permission to communicate with each other about matters relating to the content of their respective files. Opposed by the government, judges approved the requests, inasmuch as the SAP would supervise them; they are now fairly well-established features of the SA system, with there being stable interactional expectancies about their use. One participant reflected on the value of professional development and knowledge-sharing sessions:
SAs aren't allowed to talk to each other about these cases, which is a
big limitation on their effectiveness, because the government lawyers
can get together and talk strategy, talk precedents, talk about what's
worked in one case, what hasn't work in another case, what judges
disposed one way, et cetera. But the special advocates aren't allowed
to coordinate in that way, and keep each other knowledgeable about
cases they've had, precedents that have been made, et cetera. So, a way
to get, to get over that obstruction is to have the special advocate's
representative sort of chair a meeting, of the special advocates where
they can share their purely legal experience, without disclosing any
information from their cases. (229)

Another spoke approvingly of the annual sessions convened by the SAP:
And I think that, you know, one of the, the benefits of the special
advocate program is that there is this group of lawyers and we all know
each other and we all talk, not about the individual cases, but just
about the work that we do. So I think in those kind of friendships and
professional connections, you can learn from one another. You can
support one another. (230)

The SAP can be contrasted with the UK's SASO in this respect, which has become less effective in recent years due to budget cuts and government obstruction of knowledge-sharing sessions. But again, the UK allows SAs to access a library of closed judgments. One wonders why Canadian SAs can be authorized, on an ad hoc basis, to participate in knowledge-sharing sessions but not given blanket access to written secret law.

But the biggest possible role for the SAP in the years to come relates to communication with named persons. While rare, requests are approved on the condition that the SAP supervise interactions. This is suggestive of distrust, but one SA informed us that the practice is sometimes used to protect the SA against governmental allegations of inadvertent disclosure:
[There is] a representative of the special advocate program in
attendance, just as a sort of neutral witness, to make sure everything
goes as it's supposed to, and so that if there's any questions
raised--it's largely for the protection of the special advocates, so if
there's some allegation later on, that the special advocate let
something slip, that oughtn't to have been disclosed, the
representative of the program will be able to say "No, I was there.
That didn't happen." (231)

The work of conditionality would be eased greatly were the SAP to have a more central role, perhaps hearing and making final determinations on communication requests. The UK experience is again instructive here, with its government exploring the option of placing a "Chinese wall" between government counsel and the person(s) clearing communication requests. (232) But resource and logistical barriers are stymieing this proposal:
One difficulty will be to regularly source an official, or cadres of
officials, from within the relevant government department or Agency who
will have sufficient knowledge of the case, the sourcing of the
relevant material, issues around the litigation itself and the context
of the case relative to other similar cases, who will as a result be
able to provide definitive assessments of the risk level of proposed
Special Advocate communication, but who is not in contact with, nor can
have contact with, the litigation team itself and government counsel.

This is less of a problem in Canada, where the SAP already plays a role in supervising communications, is known to be independent from government counsel, and has total awareness of the content of all files relating to Division 9 of the IRPA.

The work of conditionality is also lessened by the growing trust many designated judges have in SAs. One SA said:
I think in part, it's just the personalities of the judges, that we've
had, who have been quite open to the role of special advocates. In
part, I think, even some judges who were skeptical about, you know,
whether we were necessary, have really come around. And I think in
part, it was, um, you know, we were sort of the, we were the new guys,
and they still weren't really sure what to make of us. And, now that we
have, I hope, proven our value, the court is much more willing to put
its confidence in us. And so I think that's made it easier to find
solutions in this situations as well. (234)

This sentiment was echoed by another SA: "Now they trust us. And, not because the Supreme Court of Canada says they should, which they do, in Harkat, but they've gotten to know us. And, I mean, obviously, we were, we were known by reputation only but not, but not personally." (235)

But trust is a double-edged sword, as it applies equally to government counsel and CSIS witnesses:
I don't care who you are, if you're seeing the same guy before you all
the time, the same kind of, all, and you know, there will come a point
where, when the government lawyer gets up and says, "This is a real
serious threat. You just gotta, you just gotta accept it." Just in the
same way that when I get up and I say "I want to talk to the public
counsel." they trust me. Right? They will, they will trust you when you
say "It's a threat." (236)

And it goes without saying that relations of trust relate to the personalities involved. As judges retire and as new SAs and government counsel conduct future work, the process will have to start anew.

It would seem then, that there has been a progressive development in trust, institutional support and knowledge, and the emergence of stable interactional expectancies about certain powers. But customs are likely to ebb and flow as personalities change. What is more, core powers remain effectively unavailable to SAs due either to structural flaws (e.g., judicial authorization process) or the strategic choice of SAs to avoid staunch government resistance (e.g., expert witnesses).

We would like to end by reflecting on what meaning is to be found in struggles over the character of the Canadian SA system. There is a sense in which these struggles may be seen as merely historical. Security certificates proceedings seem to be coming to an end; the Federal Court upheld the reasonableness of certificates against Mohamed Harkat and Mohammad Mahjoub, (237) while finding the certificates against Mahmoud Jaballah, Adil Charkaoui and Hassan Almrei to be unreasonable; (238) earlier, the government chose not to reissue a certificate against Manickavasagam Suresh following the Charkaoui I judgment. (239) Although there remain legal obstacles to the removal of Messrs. Harkat and Mahjoub, (240) the Supreme Court declined to hear Mr. Mahjoub's appeal--presumably because it raised issues already settled in the Harkat j udgment. (241)

But security-based detention and deportation proceedings have not abated, with there being a strong likelihood that the government will start invoking section 86 of the IRPA to conduct secret hearings before the IRB. It successfully argued that Mr. Suresh is inadmissible in 2015 and attempted to remove Mr. Almrei through an inadmissibility hearing before the IRB after the Federal Court quashed the certificate against him. (242) There are likely to be new (and the revisitation of old) Charter issues as the government shifts its energies towards the IRB. Among the issues are the extents to which Charkaoui I and II apply outside of the certificate context (243)--a question framed by the Court's consistent position that section 7 generally does not apply to IRB hearings. (244)

But one of our core findings has been that the SA system has produced high-quality, security-cleared lawyers capable of doing a wide range of legal work. This has corresponded with changes in the traditional roles and responsibilities of amicus curiae in a wide range of proceedings. Of particular note are section 38 CEA and CSIS warrant application proceedings.

One interviewee told us the role of amici:
[Amici have] evolved some distance from the purely neutral, legal
advisor to the court, to, you know... answer a question the court may
have, or to sort of slightly counterbalance the role of the government
in those proceedings. I think that the courts have looked to us to
really perform a more adversarial role, or at least a role that comes
closer to recreating an adversarial hearing in the closed proceedings.

This is interesting since, unlike SAs, the appointment of amici is not constitutionally required, nor is it governed by legislation. It is reasonable to suppose that the need for judges to acculturate to the rule of law demands of the SA system has played a central role in a broader shift to adversarial challenge in secret proceedings generally, although the changing regulatory environment, breaches of candour on the part of the intelligence community, and the Court's growing expertise with the use of intelligence as evidence are also driving the change. This is not to say that the SA system has brought secret hearings into alignment with the rule of law. The value of the SA system is best viewed in relative terms, as being the lesser of two evils. Secret hearings are here to stay, and the SA system is the best means at our disposal to contend with the implications this has for rights and the rule of law. But being the lesser of two evils does not bestow legal or moral virtue to secret trials; there remains much that courts, legislatures, and the executive can do to make the system the best it can possibly be.

Graham Hudson (*) & Daniel Alati (**[dagger])

(*) Dr. Graham Hudson is an Associate Professor in the Department of Criminology at Ryerson University.

(**) Dr. Daniel Alati is an Assistant Professor in the Department of Criminology at Ryerson University.

([dagger]) The authors want to thank the interview participants (and their respective institutions), the late Ron Atkey for his insights, support, and friendship, Dr. Lucia Zedner for her invaluable help facilitating research in the United Kingdom, and Omar Abou El Hassan, Rajpreet Atwal, and Sasan Ansari for their outstanding research assistance. The research in this article was funded by the Social Sciences and Humanities Research Council of Canada as well as by a Foundation for Legal Research Grant.

(1.) For a comprehensive survey of secret proceedings in Canada before 1996, see Ian Leigh, "Secret Proceedings in Canada" (1996) 34:1 Osgoode Hall LJ 113.

(2.) Secret trials were reportedly held in the Star Chamber in the fifteenth to seventeenth centuries, although it is disputed how often this occurred. See Edward P Cheyney, "The Court of Star Chamber" (1913) 18:4 Am Hist Rev 727. For a broader analysis of secret hearings in the UK, see Laurence Lustgarten & Ian Leigh, In from the Cold: National Security and Parliamentary Democracy (Oxford: Clarendon Press, 1994).

(3.) Prior to 2001, there were four classes of proceedings concerned with protected information which were administered by the Federal Court: reviews of Canadian Security Intelligence Service warrant applications (which by this time was already vetted through an internal review mechanism), certain classes of security certificates, reviews of access to information requests, and non-disclosure motions under the Canada Evidence Act, RSC 1985, c C-5 [CEA]. While still in operation, most of these proceedings occurred less frequently than they do now and involved less material and legal issues than today.

(4.) See Craig Forcese & Kent Roach, False Security: The Radicalization of Canadian Anti-Terrorism (Toronto: Irwin Law, 2015); Daniel Alati, Domestic Counter-Terrorism in a Global World: Post-9111 Institutional Structures and Cultures in Canada and the United Kingdom (London: Routledge, 2018); Graham Hudson, "As Good as it Gets?: Security, Asylum, and the Rule of Law After the Certificate Trilogy" (2015) 52:3 Osgoode Hall LJ 905 [Hudson, "As Good as it Gets?"].

(5.) See Graham Hudson, "Secret Hearings and the Right to a Fair Trial: 2015 and Beyond" [2015] Can Hum Rts YB 101.

(6.) The Immigration and Refugee Protection Act, SC 2001, c 27 [IRPA], protects information, the disclosure of which would be injurious to national security or the safety of any person. By contrast, the CEA, supra note 3, protects information relating to international relations, national defence, or national security.

(7.) See Adam DM Svendsen, "Connecting Intelligence and Theory: Intelligence Liaison and International Relations" (2009) 24:5 Intelligence & National Security 700; Martin Rudner, "Hunters and Gatherers: The Intelligence Coalition Against Islamic Terrorism" (2004) 17:2 Intl J Intelligence & Counterintelligence 193; Glen M Segell, "Intelligence Agency Relations Between the European Union and the U.S." (2004) 17:1 Intl J Intelligence & Counterintelligence 81; Stephane Lefebvre, "The Difficulties and Dilemmas of International Intelligence Cooperation" (2003) 16:4 Intl J Intelligence & Counterintelligence 527.

(8.) For a discussion on the use of intelligence as evidence generally, see Canada, Commission of Inquiry into the Investigation of the Bombing of Air India Flight 182, The Unique Challenges of Terrorism Prosecutions: Towards a Workable Relation Between Intelligence and Evidence, vol 4 (Ottawa: Public Works and Government Services Canada, 2010) at 11-40; Craig Forcese, "Staying Left of Bang: Reforming Canada's Approach to Anti-Terrorism Investigations" (2017) University of Ottawa Faculty of Law Working Paper No 2017-23; Kent Roach & Craig Forcese, "Intelligence to Evidence in Civil and Criminal Proceedings: Response to August Consultation Paper" (12 September 2017) [unpublished], online: SSRN <>.

(9.) We define the rule of law as decision making guided by open, general and stable rules that cohere with principles of access to justice and procedural fairness. See Joseph Raz, "The Rule of Law and its Virtue" (1977) 93:2 Law Q Rev 195 [Raz, "Virtue"]; Joseph Raz, Ethics in the Public Domain: Essays in the Morality of Law and Politics (Oxford: Clarendon Press, 1994). For a good overview of some philosophical debates about the meaning of the rule of law, see Paul P Craig, "Formal and Substantive Conceptions of the Rule of Law: An Analytical Framework" [1997] Public L 467; John Finnis, Natural Law and Natural Rights (Oxford: Clarendon Press, 1980); Lon L Fuller, "The Forms and Limits of Adjudication" (1978) 92:2 Harv L Rev 353; Andrei Marmor, "The Rule of Law and Its Limits" (2004) 23:1 Law & Phil 1; Brian Tamanaha, On the Rule of Law: History, Politics, Theory (Cambridge, UK: Cambridge University Press, 2004); Jeremy Waldron, "The Concept and the Rule of Law" (2008) 43:1 Ga L Rev 1; Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-Based Decision-Making in Law and in Life (Oxford: Oxford University Press, 1991).

(10.) Some secret proceedings are not held before an independent and impartial judge, with the Guantanamo Military Commission serving as a notable example. See David Cole, "Military Commissions and the Paradigm of Prevention" in Fionnuala Ni Aolain & Oren Gross, eds, Guantanamo and Beyond: Exceptional Courts and Military Commissions in Comparative Perspective (New York: Cambridge University Press, 2013) 95.

(11.) For critics, they exemplify what David Dyzenhaus terms a legal "grey hole"--a phenomenon where "official lawlessness" is cloaked under a facade of legality. See David Dyzenhaus, "The Compulsion of Legality" in Victor V Ramraj', ed, Emergencies and the Limits of Legality (Cambridge, UK: Cambridge University Press, 2008) 33 [Dyzenhaus, "The Compulsion of Legality"]; David Dyzenhaus, "Schmitt v. Dicey: Are States of Emergency Inside or Outside the Legal Order?" (2006) 27:5 Cardozo L Rev 2005; David Dyzenhaus, "Cycles of Legality in Emergency Times" (2007) 18 Public L Rev 165.

(12.) See David Cole, Federico Fabbrini & Arianna Vedaschi, eds, Secrecy, National Security and the Vindication of Constitutional Law (Cheltenham, UK: Edward Elgar, 2013); David Cole & Stephen I Vladeck, "Navigating the Shoals of Secrecy: A Comparative Analysis of the Use of Secret Evidence and 'Cleared Counsel' in the United States, the United Kingdom and Canada", in Liora Lazarus, Christopher McCrudden & Nigel Bowles, eds, Reasoning Rights: Comparative Judicial Engagement (Oxford: Hart, 2014) 161.

(13.) See Simon Chesterman, "Secrets and Lies: Intelligence Activities and the Rule of Law in Times of Crisis" (2007) 28:3 Mich J Intl L 553; Shirin Sinnar, "Rule of Law Tropes in National Security" (2016) 129:6 Harv L Rev 1566.

(14.) 2014 SCC 37 [Canada v Harkat].

(15.) For cases where the SA system was upheld pursuant to the right to a fair trial, notwithstanding failings of exercises of discretionary power on a case-by-case basis, see A and Others v The United Kingdom [GC], No 3455/05, [2009] II ECHR 137, 49 EHRR 29; Secretary of State for the Home Department v AF, [2009] UKHL 28; Secretary of State for the Home Departmentv MB, [2007] UKHL 46; AZ v Secretary of State for the Home Department, [2017] EWCA Civ 35. For less direct or tacit approvals of the SA system, see Tinnelly & Sons Ltd and Others and McElduff and Others v The United Kingdom (1998), [1998] ECHR 56, 27 EHRR 249 at para 78; Jasper v The United Kingdom [GC], No 27052/95, [2000] ECHR 90, 30 EHRR 441; Fitt v The United Kingdom [GC], No 29777/96, [2000] II ECHR 367, 30 EHRR 480 (dissenting judgments); Edwards and Lewis v The United Kingdom [GC], No 39647/98 & No 40461/98, [2004] X ECHR 560, 40 EHRR 24. See also Eva Nanopoulos, "European Human Rights Law and the Normalisation of the 'Closed Material Procedure': Limit or Source?" (2015) 78:6 Mod L Rev 913; Aileen Kavanagh, "Special Advocates, Control Orders and the Right to a Fair Trial" (2010) 73:5 Mod L Rev 836.

(16.) See Special Immigration Appeals Commission Act, 1997 (UK).

(17.) See Terrorism Act 2000 (UK) at Schedule 3.

(18.) See Prevention of Terrorism Act 2005 (UK) [PTA], as repealed by Terrorism Prevention and Investigation Measures Act 2011 (UK), s 1 [TPIM].

(19.) See TPIM, supra note 18.

(20.) See Counter-Terrorism Act 2008 (UK), s 68; Terrorist Asset-FreezingAct etc 2010 (UK).

(21.) See Justice and Security Act 2013 (UK), s 6 [Justice and Security Act]. The UK Supreme Court strongly criticized the SA system from the perspective of the common law but accepted the authority of Parliament to legislate for closed material civil hearings. See Al Rawi v Security Service, [2011] UKSC 34 [Al Rawi].

(22.) See Canada v Harkat, supra note 14 at para 24. See also Vancouver Sun (Re), 2004 SCC 43; David M Paciocco, "When Open Courts Meet Closed Government" (2005) 29 SCLR (2nd) 385; Canadian Broadcasting Corp v New Brunswick (Attorney General), [1996] 3 SCR 480, 139 DLR (4th) 385 [CBC].

(23.) See Canada v Harkat, supra note 14 at para 24. See also Vancouver Sun (Re), supra note 22; Paciocco, supra note 22; CBC, supra note 22.

(24.) [2013] UKSC 38.

(25.) Canada v Harkat, supra note 14 at para 24.

(26.) Ibid at para 26.

(27.) We derive this definition from Patricia Landolt and Luin Goldring, who apply it in the context of the access which precarious migrants (do not) have to formal rights and public services. See Patricia Landolt & Luin Goldring, "Assembling Noncitizenship Through the Work of Conditionality" (2015) 19:8 Citizenship Studies 853.

(28.) Lon L Fuller, "Human Interaction and the Law" (1969) 14:1 Am J Juris 1 at 10 [Fuller, "Human Interaction"]. See Martha-Marie Kleinhans & Roderick A Macdonald, "What is a Critical Legal Pluralism?" (1997) 12:2 CJLS 25 at 32.

(29.) See Oona A Hathaway, "Path Dependence in the Law: The Course and Pattern of Legal Change in a Common Law System" (2001) 86:2 Iowa L Rev 601.

(30.) s 7, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.

(31.) See Charkaoui v Canada (Citizenship and Immigration), 2007 SCC 9 at para 29 [Charkaoui I].

(32.) See R v Stinchcombe, [1991] 3 SCR 326, 130 NR 277.

(33.) See Nazim Baksh & Terence McKenna, "Federal Government Reaches Settlement with 3 Canadian Men Tortured in Syria and Egypt", CBC News (17 March 2017), online: <>. The claimants and the government of Canada settled the case in 2017. The government agreed to pay $31.3 million and issued an apology. See "Ottawa Pays $31.3M to Canadian Men Tortured in Syria", CBC News (26 October 2017), online <>. For background information, see Canada, Internal Inquiry into the Actions of Canadian Officials in Relation to Abdullah Almalki, Ahmad Abou-Elmaati and Muayyed Nureddin (Ottawa: Public Works and Government Services Canada, 2008).

(34.) Supra note 6.

(35.) See ibid, s 82.

(36.) See Hudson, "As Good as it Gets?", supra note 4 at 907.

(37.) See IRPA, supra note 6, s 83(1)(i).

(38.) Supra note 3.

(39.) Supra note 31.

(40.) 2008 SCC 38 [Charkaoui II].

(41.) Ibid at para 2.

(42.) 2011 SCC 6.

(43.) Ibid at paras 44-45.

(44.) See IRPA, supra note 6, ss 85.1, 86, 87.1.

(45.) See X (Re), 2016 FC 1105.

(46.) See Secure Air Travel Act, SC 2015, c 20, s 11, s 16(6); Prevention of Terrorist Travel Act, SC 2015, c 36, s 42, s 4(4).

(47.) See Craig Forcese & Lorne Waldman, "A Bismarckian Moment: Charkaoui and Bill C-3" (2008) 42 SCLR (2nd) 355; Kent Roach, "Charkaoui and Bill C-3: Some Implications for Anti-Terrorism Policy and Dialogue Between Courts and Legislatures" (2008) 42 SCLR (2nd) 281; Canada, Library of Parliament, Bill C-3: An Act to Amend the Immigration and Refugee Protection Act (Certificate and Special Advocate) and to Make a Consequential Amendment to Another Act (Legislative Summary), by Penny Becklumb, No 39-2-LS-567-E (Ottawa: Parliamentary Information and Research Service, 2008).

(48.) Supra note 6, s 85.1(1).

(49.) Ibid., s 85.1(1)(3).

(50.) See ibid., s 85.1(4).

(51.) We support this claim in Section III.A, below.

(52.) See David Jenkins, "There and Back Again: The Strange Journey of Special Advocates and Comparative Law Methodology" (2011) 42:2 Colum HRLR 279 at 281.

(53.) Supra note 16.

(54.) Supra note 18.

(55.) [1996] ECHR 54, 23 EHRR 413.

(56.) [2004] UKHL 56.

(57.) Supra note 6, s 83(1)(b).

(58.) Ibid, s 83.

(59.) See Craig Forcese & Lorne Waldman, "Seeking Justice in an Unfair Process: Lessons from Canada, the United Kingdom, and New Zealand on the Use of 'Special Advocates' in National Security Proceedings" (1 August 2007) at 28, online: SSRN <> [Forcese & Waldman, "Seeking Justice"].

(60.) Members of the intelligence community refer to this phenomenon as the "mosaic effect", whereby the combination of separate units of information contained in several files produce knowledge of a larger pattern of, among other things, intelligence operations, organizations, tactics, and sources. See Canada (Attorney General) v Almalki, 2010 FC 1106 at paras 115-19; Canada (Attorney General) v Canada (Commission of Inquiry into the Actions of Canadian Officials in Relation to Maher Arar), 2007 FC 76 at para 82; Henrie v Canada (Security Intelligence Review Committee) (1988), [1989] 2 FC 229 at 242-43, 53 DLR (4th) 568. See also David E Pozen, "The Mosaic Theory, National Security, and the Freedom of Information Act" (2005) 115:3 Yale LJ 628 at 633-34. Related to the mosaic effect is the need-to-know principle, which restricts access to sensitive information to those who need to access it in order to perform a legitimate intelligence function. See Harkat (Re), 2009 FC 204 at paras 35-46.

(61.) See Special Immigration Appeals Commission (Procedure) Rules 2003 (UK), SI 2003/1034, rr 36(4)-(5); UK, Secretary of State for Justice, Justice and Security Green Paper (Cm 8194, 2011) at paras 2.30-2.36.

(62.) IRPA, supra note 6, s 85.4(2).

(63.) See ibid, s 83(1)(g). However, one SA informed us that the Minister of Public Safety is not always given an opportunity to be heard and, indeed, may not even be aware that a request to communicate has been filed. Interview of Participant 2 (8 May 2017) [Interview 2].

(64.) See supra note 6, s 85.4(1).

(65.) We will explore this issue in Section IV.C, below. See also Hudson, "As Good as it Gets?", supra note 4.

(66.) Supra note 6, s 77(2).

(67.) Supra note 15.

(68.) Supra note 15.

(69.) [2009] UKHL 10.

(70.) Angus McCullough QC et al, "Justice and Security Green Paper: Response to Consultation from Special Advocates" (16 December 2011) at 11, online (pdf): WordPress <adam1cor.files.>.

(71.) See UK, HL & HC, Joint Committee on Human Rights, Counter-Terrorism Policy and Human Rights (Sixteenth Report): Annual Renewal of Control Orders Legislation 2010, Ninth Report of Session 2009-10, HL Paper 64/HC 395 (26 February 2010) (Chair: Andrew Dismore); UK, HL & HC, Joint Committee on Human Rights, Counter-Terrorism Policy and Human Rights: 28 Days, Intercept and Post-Charge Questioning, Nineteenth Report of Session 2006-07, HL Paper 157/HC 394 (30 July 2007) (Chair: Andrew Dismore); UK, HL & HC, Joint Committee on Human Rights, Counter-Terrorism Policy and Human Rights (Eighth Report): Counter-Terrorism Bill, Ninth Report of Session 2007-08, HL Paper 50/HC 199 (7 February 2008) (Chair: Andrew Dismore); UK, HC, Constitutional Affairs Committee, The Operation of the Special Immigration Appeals Commission (SIAC) and the Use of Special Advocates, Seventh Report of Session 2004-05, HC Paper 323, vol I (3 April 2005) [Constitutional Affairs Committee, Operation of SIAC]; Al Rawi, supra note 21 at paras 36-50, 76, 83, 93; Martin Chamberlain, "Special Advocates and Procedural Fairness in Closed Proceedings" (2009) 28:3 CJQ 314; Martin Chamberlain, "Update on Procedural Fairness in Closed Proceedings" (2009) 28:4 CJQ 448.

(72.) See McCullough et al, supra note 70; Secretary of State for Justice, supra note 61 at paras 2.33-2.34.

(73.) Canada v Harkat, supra note 14 at para 70.

(74.) Ibid at para 72.

(75.) See Canada v Harkat, supra note 14. See also Almrei (Re), 2008 FC 1216 at para 41 [Almrei 2008]; Almrei (Re), 2009 FC 322 at para 24 [Almrei 2009 No 1].

(76.) For cases on exceptions to solicitor-client privilege, see Smith v Jones, [1999] 1 SCR 455 at para 53, 169 DLR (4th) 385, Cory J; Almrei 2008, supra note 75 at paras 60-62.

(77.) See Canada v Harkat, supra note 14 at para 93.

(78.) Ibid at paras 54-56.

(79.) See ibid at para 57.

(80.) This is a familiar line of reasoning. See e.g. Little Sisters Book and Art Emporium v Canada (Minister of Justice), 2000 SCC 69.

(81.) Canada, Department of Justice, Special Advocates Program Evaluation: Final Report (Ottawa: Department of Justice Canada, 2015) at 10 [SAP Report].

(82.) Giorgio Agamben, State of Exception (Chicago: University of Chicago Press, 2005); David Dyzenhaus, "The State of Emergency in Legal Theory" in Victor V Ramraj, Michael Hor & Kent Roach, eds, Global Anti-Terrorism Law and Policy (New York: Cambridge University Press, 2005) 65; Oren Gross, "Chaos and Rules: Should Responses to Violent Crises Always be Constitutional?" (2003) 112:5 Yale LJ 1011; John Ferejohn & Pasquale Pasquino, "The Law of Exception: A Typology of Emergency Powers" (2004) 2:2 Intl J Constitutional L 210; Colin McQuillan, "The Real State of Emergency: Agamben on Benjamin and Schmitt" (2010) 18 Studies in Soc & Political Thought 96; Walter Benjamin, "On the Concept of History", in Howard Eiland & Michael W Jennings, eds, Walter Benjamin: Selected Writings, Volume 4: 1938-1940, translated by Edmund Jephcott et al (Cambridge, MA: Harvard University Press, 2003) 389.

(83.) See Colleen Bell, "Subject to Exception: Security Certificates, National Security and Canada's Role in the 'War on Terror'" (2006) 21:1 CJLS 63; Colleen Bell, The Freedom of Security: Governing Canada in the Age of Counter-Terrorism (Vancouver: UBC Press, 2011). See especially ibid at 55-86 ("The Socio-Legal Paradox of Freedom: Security Certificates and the Politics of Exception"). See Irina Ceric, "The Sovereign Charter: Security, Territory and the Boundaries of Constitutional Rights" (2012) 44:2 Ottawa L Rev 353; Mike Larsen & Justin Piche, "Incarcerating the 'Inadmissible': KIHC as an Exceptional Moment in Canadian Federal Imprisonment" (2007) York Centre for International and Security Studies Working Paper No 45, online (pdf): YorkSpace <>; Kent Roach, "The Law Working Itself Pure?: The Canadian Experience with Exceptional Courts and Guantanamo" in Ni Aolain & Gross, supra note 10, 201.

(84.) See Dyzenhaus, "The Compulsion of Legality", supra note 11.

(85.) Al Rawi, supra note 21 at para 73.

(86.) See ibid at para 71.

(87.) "Our Security, Our Rights: National Security Green Paper, 2016" (2016) at 4, online (pdf): Canadian Bar Association <>.

(88.) See Raz, "Virtue", supra note 9 at 198-200.

(89.) See ibid at 201.

(90.) Duncan Kennedy, "Form and Substance in Private Law Adjudication" (1976) 89:8 Harv L Rev 1685 at 1688. See also Raz, "Virtue", supra note 9 at 202-03.

(91.) Dakota S Rudesill, "Coming to Terms with Secret Law" (2015) 7:1 Harvard National Security J 241 at 318.

(92.) Ibid.

(93.) Ibid at 312. See also David E Pozen, "Deep Secrecy" (2010) 62:2 Stan L Rev 257 at 278 [Pozen, "Deep Secrecy"]; Daniel Patrick Moynihan, Secrecy: The American Experience (New Haven: Yale University Press, 1998); Max Weber, "Bureaucracy", in HH Gerth & C Wright Mills, eds, From Max Weber: Essays in Sociology, translated by HH Gerth & C Wright Mills (New York: Oxford University Press, 1946) 196 at 233-35.

(94.) See Orin S Kerr, "A Rule of Lenity for National Security Surveillance Law" (2014) 100:7 Va L Rev 1513 at 1518.

(95.) See Pozen, "Deep Secrecy", supra note 93 at 278-79.

(96.) See Fuller, "Human Interaction", supra note 28 at 2-3, 9-10. See also Jutta Brunnee & Stephen J Toope, "International Law and Constructivism: Elements of an Interactional Theory of International Law" (2000) 39:1 Colum J Transnat'l L 19 at 28-29.

(97.) See Leigh, supra note 1. See also Forcese & Waldman, "Seeking Justice", supra note 59; James K Hugessen, "Watching the Watchers: Democratic Oversight" in David Daubney et al, eds, Terrorism, Law and Democracy: How Is Canada Changing Following September 11? (Montreal: Editions Themis, 2002) 381; Murray Rankin, "The Security Intelligence Review Committee: Reconciling National Security with Procedural Fairness" (1990) 3 Can J Admin L & Prac 173.

(98.) Particular areas of inquiry include denial of security clearance, discrimination, and denial of citizenship. See Canadian Security Intelligence Service Act, RSC 1985, c C-23, s 38(c) [CSIS Act].

(99.) See ibid, s 50(a).

(100.) See Leigh, supra note 1 at 161; Canada (Attorney General) v Al Telbani, 2012 FC 474 at para 115.

(101.) See CSIS Act, supra note 98, s 39(2).

(102.) See Leigh, supra note 1 at 163.

(103.) See Forcese & Waldman, "Seeking Justice", supra note 59 at 8.

(104.) Rules of Procedure of the Security Intelligence Review Committee in Relation to its Function Under Paragraph 38(c) of the Canadian Security Intelligence Service Act (entered into force 9 March 1985, in force prior to 1 May 2014).

(105.) Ibid, ss 48(2), 48(4)-(5).

(106.) Ibid, s 48(2).

(107.) See Interview 2, supra note 63; Interview of Participant 5 (18 April 2017) [Interview 5]; Interview of Participant 8 (26 June 2017) [Interview 8]; Interview of Participant 9 (26 June 2017) [Interview 9]; Interview of Participant 10 (15 December 2016) [Interview 10]; Interview of Participant 34 (20 April 2017) [Interview 34].

(108.) See Interview 2, supra note 63; Interview 5, supra note 107; Interview 8, supra note 107; Interview 9, supra note 107; Interview 10, supra note 107; Interview 34, supra note 107; Forcese & Waldman, "Seeking Justice", supra note 59.

(109.) Interview 2, supra note 63.

(110.) See Interview of Participant 7 (5 May 2017) [Interview 7].

(111.) Interview 2, supra note 63.

(112.) See Interview 2, supra note 63; Interview of Participant 6 (16 February 2017) [Interview 6]; Interview 8, supra note 107; Interview 9, supra note 107; Interview 10, supra note 107; Interview of Participant 25 (31 May 2016) [Interview 25]; Interview 34, supra note 107.

(113.) See Interview 2, supra note 63.

(114.) See Interview 5, supra note 107; Interview 7, supra note 110.

(115.) See Leigh, supra note 1; Martin Rudner, "Challenge and Response: Canada's Intelligence Community and the War on Terrorism" (2004) 11:2 Can Foreign Policy J 17; Reg Whitaker, "The 'Bristow Affair': A Crisis of Accountability in Canadian Security Intelligence" (1996) 11:2 Intelligence & National Security 279.

(116.) See Interview 5, supra note 107; Interview 7, supra note 110.

(117.) See Interview 8, supra note 107; Interview 9, supra note 107.

(118.) See Charkaoui I, supra note 31 at para 51.

(119.) See Interview 10, supra note 107; Interview 35, supra note 107.

(120.) Interview 10, supra note 107.

(121.) Ibid.

(122.) Ibid.

(123.) See supra note 6, s 85.6(1).

(124.) Interview 10, supra note 107.

(125.) See ibid; Interview of Participant 11 (12 October 2016) [Interview 11]; Interview of Participant 12 (12 October 2016) [Interview 12]; Interview of Participant 13 (12 October 2016) [Interview 13].

(126.) See Interview 10, supra note 107; Interview of Participant 14 (16 December 2016) [Interview 14 No 1]; Interview of Participant 15 (16 December 2016) [Interview 15]; Interview 35, supra note 107.

(127.) Participant 16 informed us that the existence of this library is recent. See Interview of Participant 16 (15 June 2017) [Interview 16].

(128.) See ibid; Interview of Participant 29 (8 June 2017) [Interview 29].

(129.) See Interview of Participant 1 (25 May 2017) [Interview 1]; Interview 2, supra note 63; Interview of Participant 3 (May 2016) [Interview 3]; Interview of Participant 4 (21 March 2016) [Interview 4]; Interview 6, supra note 112; Interview 10, supra note 107; Interview 11, supra note 125; Interview 14 No 1, supra note 126; Interview 15, supra note 126.

(130.) See Interview 10, supra note 107; Interview 35, supra note 107.

(131.) See Interview 10, supra note 107.

(132.) 2014 FCA 249, aff'g 2009 FC 1058. See also Canadian Security Intelligence Service Act (Re), 2008 FC 301.

(133.) See X (Re), supra note 45.

(134.) Ibid at para 99 [emphasis in original omitted].

(135.) Ibid at para 100.

(136.) See Canada, Commission of Inquiry Concerning Certain Activities of the Royal Canadian Mounted Police, Freedom and Security Under the Law, vol I (Ottawa: Supply and Services Canada, 1981) at 558.

(137.) See Interview 10, supra note 107; Interview 14 No 1, supra note 126; Interview 15, supra note 126. See also an SA quote, below at 67.

(138.) See R (Mohamed) v Secretary of State for Foreign and Commonwealth Affairs, [2010] EWCA Civ 65. For an excellent academic commentary on the case, see CRG Murray, "Out of the Shadows: The Courts and the United Kingdom's Malfunctioning International Counter-Terrorism Partnerships" (2013) 18:2 J Conflict & Security L 193.

(139.) See Craig Forcese, "Canada's Security & Intelligence Community after 9/11: Key Challenges and Conundrums" (2016) University of Ottawa Faculty of Law Working Paper No 2016-35. This is likely to change to some degree, following Parliament's decision to authorize CSIS to engage in foreign operations. See CSIS Act, supra note 98, ss 12(2), 12.1, 21(3.1), 21.1(4).

(140.) See Canada, Commission of Inquiry into the Actions of Canadian Officials in Relation to Maher Arar, Report of the Events Relating to Maher Arar: Analysis and Recommendations (Ottawa: Public Works and Government Services Canada, 2006) [Arar Report].

(141.) See Justice and Security Act, supra note 21; Al Rawi, supra note 21.

(142.) See Interview 10, supra note 107.

(143.) The practice of judicial side-bars is well known, as are the implications this has on power and exclusion. See Constance Backhouse, "Gender and Race in the Construction of 'Legal Professionalism': Historical Perspectives" in Adam Dodek and Alice Woolley, eds, In Search of the Ethical Lawyer: Stories from the Canadian Legal Profession (Vancouver: UBC Press, 2016) 126; Vitalius Tumonis, "Legal Realism & Judicial Decision-Making" (2012) 19:4 Jurisprudence 1361; Neil MacCormick, Legal Reasoning and Legal Theory (Oxford: Clarendon Press, 1978) at 15-16; Brian Leiter, "Legal Formalism and Legal Realism: What Is the Issue?" (2010) 16:2 Leg Theory 111.

(144.) Constitutional Affairs Committee, Operation of SIAC, supra note 71 at para 52.

(145.) See also UK, Independent Reviewer of Terrorism Legislation, Terrorism Prevention and Investigation Measures in 2012: First Report of the Independent Reviewer on the Operation of the Terrorism Prevention and Investigation Measures Act 2011, by David Anderson QC (London: The Stationery Office, 2013) at 79-81; UK, Independent Reviewer of Terrorism Legislation, Control Orders in 2011: Final Report of the Independent Reviewer on the Prevention of Terrorism Act 2005, by David Anderson QC (London: The Stationery Office, 2012) at 82.

(146.) One UK SA also criticized this rule as contravening the "cab rank rule". Interview of Participant 19 (9 June 2017) [Interview 19]. See also Bar Standards Board, Bar Standards Board Handbook, 2nd ed, London, UK: BSB, 2015, rr rC29, rC30.

(147.) See Forcese & Waldman, "Seeking Justice", supra note 59; Jenkins, supra note 52.

(148.) Supra note 6, s 85(3).

(149.) See SAP Report, supra note 81 at 5.

(150.) See Interview 14 No 1, supra note 126; Interview 15, supra note 126; SAP Report, supra note 81.

(151.) See Interview of Participant 14 (27 June 2017) [Interview 14 No 2]; Human Resources and Professional Development Directorate, Request for Expression of Interest (EOI): Special Advocates for Bill C-3 (Security Certificates under the Immigration and Refugee Protection Act), (Ottawa: Justice Canada, 18 December 2007).

(152.) See Interview 14 No 2, supra note 151.

(153.) See Interview 14 No 1, supra note 126.

(154.) The SAP provided us access to this and other papers.

(155.) Recent examples include Martin Chamberlain (a leading UK SA) and Professor John Jackson, Nottingham Trent University.

(156.) See Interview 14 No 1, supra note 126.

(157.) See ibid.

(158.) See Charkaoui (Re), 2009 FC 1030.

(159.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 4, supra note 129; Interview 6, supra note 112; Interview 14 No 1, supra note 126; Interview 15, supra note 126; Interview 10, supra note 107; Interview 35, supra note 107; Harkat (Re), 2009 FC 204.

(160.) See Mahjoub (Re), 2008 CanLII 90747 at Schedule A, para 5 (FCTD).

(161.) Canada (Minister of Public Safety and Emergency Preparedness) v Harkat (30 May 2011), Ottawa, A-76-11 (FCA).

(162.) 2008 FC 549 at para 77.

(163.) Supra note 60 at para 84.

(164.) Harkat (Re), 2009 FC 59, [2009] 4 FCR 528 at 542.

(165.) See Harkat (Re), supra note 164; Interview 14 No 1, supra note 126.

(166.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 14 No 1, supra note 126; Interview 15, supra note 126.

(167.) See Interview of Participant 17 (31 May 2016) [Interview 17]; Interview 19, supra note 146; Interview of Participant 27 (1 June 2016) [Interview 27]; Interview of Participant 28 (1 June 2016) [Interview 28].

(168.) See supra note 40 at para 2.

(169.) See Interview 16, supra note 127; Interview 17, supra note 167; Interview of Participant 18 (5 June 2016) [Interview 18]; Interview 19, supra note 146; Interview of Participant 20 (24 May 2016) [Interview 20]; Interview of Participant 21 (19 May 2016); Interview of Participant 22 (26 May 2016); Interview of Participant 23 (25 May 2016); Interview of Participant 24 (24 May 2016) [Interview 24]; Interview 25, supra note 112; Interview of Participant 26 (1 June 2016) [Interview 26]; Interview 27, supra note 167; Interview 28, supra note 167.

(170.) See Interview 14 No 1, supra note 126; Interview 15, supra note 126.

(171.) Mahjoub (Re) (22 June 2009), Ottawa DES-07-08 (FC) (Report to Justice Blanchard about the Minister of Justice's support under section 85(3) of the Immigration and Refugee Protection Act (IRPA), 26 June 2009) [First Progress Report]; Mahjoub (Re) (22 June 2009), Ottawa DES-07-08 (FC) (Second Progress Report to Justice Blanchard about the Minister of Justice's support under section 85(3) of the Immigration and Refugee Protection Act (IRPA), 10 July 2009).

(172.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 4, supra note 129; Interview 6, supra note 112; Interview 11, supra note 125; Interview 12, supra note 125; Interview 13, supra note 125; Interview 14 No 1, supra note 126; Interview 15, supra note 126.

(173.) See Interview 2, supra note 63; First Progress Report, supra note 171.

(174.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 3, supra note 129; Interview 4, supra note 129; Interview 6, supra note 112.

(175.) See Graham Hudson, "A Delicate Balance: Re Charkaoui and the Constitutional Dimensions of Disclosure" (2010) 18:3 Const Forum Const 129 at 130.

(176.) Harkat (Re), 2009 FC 203 at para 12.

(177.) See Harkat (Re), 2009 FC 340.

(178.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 6, supra note 112.

(179.) See Harkat (Re), 2009 FC 553 at paras 1-3; Harkat (Re), 2009 FC 1050 at para 27.

(180.) Almrei (Re), 2009 FC 1263 at para 501 [Almrei 2009 No 2].

(181.) An Act to enact the Security of Canada Information Sharing Act and the Secure Air Travel Act, to amend the Criminal Code, the Canadian Security Intelligence Service Act and the Immigration and Refugee Protection Act and to make related and consequential amendments to Other Acts, 2nd Sess, 41st Parl, 2015 (assented to 18 June 2015), SC 2015, c 20.

(182.) Supra note 6, s 77(2).

(183.) See ibid, s 85.4(1)(a).

(184.) Ibid, s 85.4(1)(b).

(185.) See Craig Forcese & Kent Roach, False Security: The Radicalization of Canadian Anti-Terrorism (Toronto: Irwin Law, 2015) at 68-69.

(186.) IRPA, supra note 6, ss 83(1)(c.1), 85.4(1)(b).

(187.) Ibid, ss 83(1), 85.4(1)(b).

(188.) An Act Respecting National Security Matters, 1st Sess, 42nd Parl, 2017 (second reading by the Senate 8 November 2018).

(189.) See Interview 8, supra note 107; Interview 9, supra note 107.

(190.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 4, supra note 129; Interview 6, supra note 112.

(191.) See Harkat (Re), 2009 FC 173.

(192.) Ibid at para 16.

(193.) See ibid at para 1.

(194.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 3, supra note 129; Interview 5, supra note 107; Interview 6, supra note 112.

(195.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 5, supra note 107; Interview 6, supra note 112.

(196.) The number of decisions per file is as follows: Almrei (37); Charkaoui (39); Harkat (69); Jaballah (159); Mahjoub (214).

(197.) Almrei (22 overall, 16 from the Federal Court); Charkaoui (51 overall, 30 from the Federal Court) Harkat (43 overall, 35 from the Federal Court); Jaballah (39 overall, 32 from the Federal Court); Mahjoub (47 overall, 40 from the Federal Court).

(198.) See Almrei 2008, supra note 75.

(199.) For examples of cases where this exact decision was made, see Harkat (Re), 2010 FC 1242; Harkat (Re), 2010 FC 1243; Almrei 2009 No 1, supra note 75.

(200.) Almrei 2008, supra note 75 at para 52.

(201.) Interview 5, supra note 107.

(202.) See Interview 2, supra note 63; Interview 5, supra note 107; Interview 6, supra note 112. Arar Report, supra note 140 at 291-95.

(203.) Chief Justice Lutfy recognized this latter point in Almrei. See Almrei 2008, supra note 75 at para 49.

(204.) For Almrei, only 1 authorization to communicate (June 1, 2009); 0 authorizations for Charkaoui; 0 authorizations for Harkat; for Jaballah, 2 authorizations to communicate (September 26, 2012, and November 28, 2013); for Mahjoub, 1 authorization to communicate (December 7, 2010).

(205.) See Interview 10, supra note 107; Interview 14 No 1, supra note 126; Interview 15, supra note 126.

(206.) Interview 6, supra note 112.

(207.) Almrei 2009 No 1, supra note 75 at para 31.

(208.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 3, supra note 129; Interview 6, supra note 112.

(209.) Interview 1, supra note 129.

(210.) Interview 2, supra note 63.

(211.) Ibid.

(212.) See Interview 17, supra note 167; Interview 18, supra note 169; Interview 19, supra note 146; Interview 26, supra note 169; Interview 24, supra note 169; Interview 20, supra note 169. McCullough et al, supra note 71 at paras 17, 30.

(213.) IRPA, supra note 6, s 85.2(c).

(214.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 6, supra note 112.

(215.) See Mahjoub (Re), 2010 FC 380; Mahjoub (Re), 2010 FC 379.

(216.) See Mahjoub (Re), 2010 FC 1193 at paras 5, 32.

(217.) (24 November 2009), DES-6-08 (FCTD). For a similar example in the UK, but in the context of a criminal proceeding, see Guardian News and Media Ltd v AB CD, [2014] EWCA Crim 1861.

(218.) See Secretary of State for the Home Department v MB, [2006] EWCA Civ 1140; Amnesty International, "UK: Justice Perverted under the Anti-Terrorism, Crime and Security Act 2001" (11 December 2003) at 12, online (pdf): Amnesty International <>.

(219.) See Chesterman, supra note 13.

(220.) At least with respect to the practice of excluding government counsel from requests for authorization to communicate with named persons or outside counsel.

(221.) This contrasts somewhat with some recent work on the distorting impact organizational culture has on legal values. See Lauren B Edelman, "The Legal Lives of Private Organizations" in Austin Sarat, ed, The Blackwell Companion to Law and Society (Malden, Mass: Blackwell, 2004) 231; Lauren B Edelman, Sally Riggs Fuller & Iona Mara-Drita, "Diversity Rhetoric and the Managerialization of Law" (2001) 106:6 Am Jour Soc 1589; Dagmar Soennecken, "The Managerialization of Refugee Determinations in Canada" (2013) 2:84 Dr et soc 291.

(222.) See supra note 45.

(223.) See Interview 1, supra note 129; Interview 2, supra note 63; Interview 4, supra note 129; Interview 6, supra note 112; Interview 14 No 1, supra note 126; Interview 15, supra note 126; Interview 10, supra note 107; Interview 35, supra note 107.

(224.) Interview 1, supra note 129.

(225.) Interview 2, supra note 63.

(226.) Interview 6, supra note 112.

(227.) Interview 4, supra note 129.

(228.) Interview 5, supra note 107.

(229.) Interview 2, supra note 63.

(230.) Interview 4, supra note 129.

(231.) Interview 2, supra note 63.

(232.) Secretary of State for Justice, supra note 61 at para 2.33.

(233.) Ibid, s 2.34.

(234.) Interview 4, supra note 129.

(235.) Interview 1, supra note 129.

(236.) Ibid.

(237.) See Mahjoub (Re), 2013 FC 1092; Mahjoub v Canada (Citizenship and Immigration), 2017 FCA 157 [Mahjoub 2017]; Harkat (Re), 2010 FC 1241.

(238.) See Charkaoui (Re), supra note 158; Almrei 2009 No 2, supra note 180; Jaballah (Re), 2016 FC 586.

(239.) See Suresh v Canada (Public Safety and Emergency Preparedness), 2017 FC 28 at para 6 [Suresh].

(240.) One of the issues is whether the men can be deported to face the substantial risk of torture or, in the alternative, how long they may be detained until they can be safely removed. See Hudson, "As Good as it Gets?", supra note 4.

(241.) See Mahjoub 2017, supra note 237, leave to appeal to SCC refused, 37793 (17 May 2018). However, the issues raised by Mr. Mahjoub differ from those raised in Canada v Harkat, supra note 14, and were addressed in detail by the Federal Court of Appeal three years after the Court's decision (ibid).

(242.) See Suresh, supra note 239 at para 24; Almrei v Canada (Citizenship and Immigration), 2014 FC 1002.

(243.) See Suresh, supra note 239; B095 v Canada (Citizenship and Immigration), 2016 FC 962; Torres Victoria v Canada (Public Safety and Emergency Preparedness), 2011 FC 1392.

(244.) See B010 v Canada (Citizenship and Immigration), 2015 SCC 58; Febles v Canada (Citizenship and Immigration), 2014 SCC 68.

(245.) Interview 4, supra note 129.
COPYRIGHT 2018 Queen's University, Faculty of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Hudson, Graham; Alati, Daniel
Publication:Queen's Law Journal
Date:Sep 22, 2018
Previous Article:When Efficiency Calls: Rethinking the Obligation to Provide Reasons for Administrative Decisions.
Next Article:MAID in Canada? Debating the Constitutionality of Canada's New Medical Assistance in Dying Law.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters