Thursday Thinkpiece: Roach on Terrorism Prosecutions in Post-9/11 Canada

Each Thursday we present a significant excerpt, usually from a recently published book or journal article. In every case the proper permissions have been obtained. If you are a publisher who would like to participate in this feature, please let us know via the site’s contact form.

Be Careful What You Wish For? Terrorism Prosecutions in Post-9/11 Canada

Kent Roach, Professor and Prichard Wilson Chair in Law and Public Policy at the University of Toronto | antiterrorlaw.ca
(2014) 40:1 Queen’s LJ 99

Excerpts: Introduction and Part I

[Footnotes omitted. They can be found in the original via the link above]

Introduction

Since 9/11, there has been international movement away from criminal law toward less restrained alternatives, including immigration detention. As an antidote to current approaches, Professor Conor Gearty has recently revived the traditional call for “charge or release”. He argues that political violence must be treated as criminal in order to protect the liberty of the least powerful and to ensure that basic rules of justice are met, such as “open justice; a presumption of innocence; careful rules of evidence to prevent abuse; an independent system of sentencing, and much else besides”.

I agree with Professor Gearty that the criminal law is preferable to its less restrained alternatives. Nevertheless, the criminal law that Gearty wishes us to return to is not the same criminal law that eventually acknowledged the wrongful convictions of the Guildford Four and the Birmingham Six. In those cases, the convictions rested on whether the accused actually bombed civilians sitting in pubs. Today, the criminal law has moved so far in a preventive direction that charges are almost always laid long before any such event. Issues of guilt or innocence have become far more complex and dependent upon legislative definitions. Indeed,there is a danger that legislative overbreadth can define innocence out of existence.

It is particularly difficult for Canadians to understand that the criminal law may not be the perfect solution, as Canada has been exposed to the worst effects of using immigration law for counter-terrorism, while Canadian criminal law on terrorism is more restrained than its American or British counterparts. Of course, the criminal law is preferable to immigration law, which risks indeterminate detention, judicially sanctioned use of secret evidence, and Canada’s shameful “loaded weapon” of the threat of deportation to torture under the Suresh v Canada (Minister of Citizenship and Immigration) exception. Unrestrained celebration of the criminal law would, however, be a mistake. In this article I will argue that Professor Gearty and other defenders of the criminal law have overestimated and romanticized criminal law’s contemporary restraints.

Canadian criminal law is comparatively restrained. The Supreme Court of Canada has stressed the importance of requiring proof of terrorist purposes and has read broadly defined terrorism offences to exclude conduct that is not harmful or creates only a “negligible risk of harm”. The Court has reminded judges of the importance of sentencing discretion, including the potential relevance of rehabilitation when sentencing terrorists. It has also warned Canadian trial judges that they should stay terrorism trials if they are concerned that the accused cannot obtain a fair trial because of non-disclosure of information on national security confidentiality grounds. These are all important and praiseworthy safeguards.

Nevertheless, Canadians should not be complacent about how criminal law is actually used against suspected terrorists. The criminal law promises to respect the presumption of innocence, and the need to prove guilt beyond a reasonable doubt with public evidence was affirmed when two men were acquitted of the 1985 Air India bombings. Part I of this article will demonstrate, however, that those charged with terrorism offences face a reverse onus that requires them to establish that they should not be denied bail. Moreover, the preventive detention that characterizes security certificates under immigration law is not foreign to the criminal law, as the grounds for denying bail are almost as expansive as the grounds for determining whether a non-citizen is a threat to national security under security certificates. While Canada has not yet used preventive arrests and peace bonds for suspected terrorists, these legal instruments remind us that the criminal law, like immigration law, has preventive aspirations.

The criminal law also promises the accused effective remedies for unfair treatment. Part II of this article will demonstrate, however, that the accused in criminal terrorism cases in Canada have had no more success persuading judges to grant strong remedies like stays of proceedings than detainees in security certificate cases. Accused in three Canadian terrorism prosecutions have asked for stays of proceedings on the basis of entrapment, but all have had their claims rejected. Entrapment has similarly failed as a defence in terrorism prosecutions in the United States. Although the Canadian experience is less intense than the American experience, it still opens up the possibility of random virtue testing of those who associate and share extreme religious and political views with suspected terrorists. This Part will also examine the promise that the Supreme Court made in R v Ahmad that trial judges would stay proceedings before allowing a trial that was unfair because of non-disclosure due to national security confidentiality rulings by the Federal Court. It will be suggested that there is no reliable means to ensure that trial judges have the information they need to enforce fair trial rights through strong remedies.

Part III of this article will examine why it is unlikely that wrongful convictions will emerge from the terrorism cases of the 2000s and will relate this to problems of legislative overbreadth in defining terrorism offences. Although Canadian law makes some exemptions for freedom fighters and does not base terrorism convictions on objective fault, some Canadian terrorist offences are as broad as the American material support offences which allow convictions in the absence of a terrorist purpose.

The final promise that the criminal law makes is of proportionate punishment. This is an important restraint that is not made in immigration law, where long-term indeterminate detention is not considered a form of punishment. Sentencing has the ability to act as a final check on overbroad offences. This was demonstrated by the British Columbia Court of Appeal’s decision upholding a controversial six-month sentence for a man who collected three thousand dollars in support of the Tamil Tigers and was the first person convicted of financing terrorism. It will be seen in Part IV of this article, however, that this promise of proportionate punishment is fragile. The 2001 Anti-terrorism Act challenges proportionate punishment by providing for mandatory consecutive sentences for overlapping terrorism crimes. Moreover, little would stop the government from introducing mandatory minimum penalties for terrorism offences or simply extraditing people to countries with higher sentencing tariffs.

Some of the concerns raised in this article about the criminal law are based on abuses that may occur in the future but cannot be said with certainty to have yet occurred. The speculative nature of some of my concerns largely reflects Canada’s limited experience with terrorism prosecutions compared with the United States, the United Kingdom or even Australia. But that could change. The argument made in this article is nuanced because it does not dispute that the criminal law is more restrained than immigration law in its approach to terrorism. Indeed, the criminal law is preferable to immigration law because of its emphasis on fair trial rights and the presumption of innocence. The virtues of the criminal law revolve around fair trials, but the criminal law is less restrained when it comes to offence definition as well as the pre- and post-trial stages of the criminal process. This article attempts to redirect scholarship on terrorism toward these often-neglected parts of the criminal process. It also warns against complacency in preferring the criminal law solution to terrorism. In particular, it is dangerous to ignore Parliament’s ability to broadly define all crimes and, in doing so, erase the importance of innocence that Professor Gearty and other defenders of the criminal law rightly celebrate.

I. Criminal Law and the Promise of Being Punished for What You Have Done

Criminal law is often defended on a retributive basis: It punishes people for what they have done and for the choices they have made. Such accounts, however, neglect the criminal law’s ability to detain people for preventive reasons in the pretrial stage of the criminal process. Pretrial custody is particularly relevant to those charged with terrorism offences because they are presumed to be ineligible for bail. They can also face the additional sanctions of preventive arrest and peace bonds. These are non-trivial features of the criminal law that are too often ignored.

A. Bail

Bail or pretrial release can be denied on any one of three grounds. The primary ground is that pretrial detention is necessary to ensure the attendance of the accused at trial. The secondary ground relates to public danger, and the third ground relates to the need to maintain confidence in the administration of justice. Those accused of terrorism may be vulnerable to detention on all three grounds. Moreover, in the immediate aftermath of 9/11, Parliament enacted a reverse onus that requires anyone charged with a terrorism offence to justify why they should be granted bail. As Gary Trotter, Canada’s leading authority on bail, observed at the time of the enactment of these provisions: “If a reverse-onus provision is constitutional with respect to drug trafficking, a provision focused on terrorism is surely to be upheld. Whether a reverse-onus provision is necessary in this context (or any context, for that matter) is another question.” As McLachlin J (as she then was) observed in her dissent in R v Pearson, the reverse onus and presumption against bail will not be necessary for large-scale organized drug traffickers, but could be both determinative and disproportionate if applied to small-scale offenders. Similarly, while a reverse onus would hardly be necessary for an accused with links to a well-resourced international terrorist group, it could be decisive where an accused is alleged to have acted with only a few individuals and has flight and danger risks that can be controlled by appropriate release conditions.

In R v Morales, the Supreme Court unanimously upheld the practice of denying bail and pretrial release for public safety reasons. The Court focused on the wording of section 515(10)(b) of the Criminal Code and concluded that it was consistent with the Canadian Charter of Rights and Freedoms because:

Bail is not denied for all individuals who pose a risk of committing an offence or interfering with the administration of justice while on bail. Bail is denied only for those who pose a “substantial likelihood” of committing an offence or interfering with the administration of justice, and only where this “substantial likelihood” endangers “the protection or safety of the public”. Moreover, detention is justified only when it is “necessary” for public safety. It is not justified where detention would merely be convenient or advantageous.

The Court in Morales accepted that it was appropriate to deny bail for reasons of preventive detention. Chief Justice Lamer reached this conclusion simply by asserting that a bail system that released a person who committed a crime while on bail would “not function properly”. The legitimacy of preventive detention was assumed because “one objective of the entire system of criminal justice is to stop criminal behaviour”.

The Court also candidly recognized that the preventive detention provision would result in a number of false positives where accused persons are kept in pretrial detention even though they would not commit an offence if released. Chief Justice Lamer accepted “that the art of predicting recidivism and future dangerousness is, at the very least, a somewhat inexact process”, but responded to this problem by asserting that “the bail system does not aim to make exact predictions about future dangerousness because such predictions are impossible to make”. In other words, the impossibility of exact predictions excused the Court from attempting to measure or improve the accuracy of the predictions that are made.

In the post-9/11, but non-terrorism, case of R v Hall, the Supreme Court upheld denying bail to maintain confidence in the administration of justice, but struck down an open-ended provision that authorized denying bail “on any just cause”. Chief Justice McLachlin concluded that the ground for denying bail on the basis of public confidence was not vague or overbroad because it instructed judges to focus on the gravity and circumstances of the offence, the potential lengthy term of imprisonment faced by the accused and the apparent strength of the prosecution’s case. Justice Iacobucci and three others dissented, arguing that independent courts should resist “irrational public fears” and “outrage” about “a highly publicized serious crime”.

Although Parliament enacted the public confidence provision in 1997, it almost seems to have been written with terrorism in mind, and it has played an important role in post-9/11 terrorism cases. For example, it was used to detain Saad Gaya of the Toronto 18 plotters, even though the judge was convinced that he had discharged his onus on the two other grounds of bail because his passport had expired and he could be subject to electronic monitoring. Justice Hill concluded that pretrial release of Gaya, who had been “demonstrably immersed in a viable plan capable of wreaking catastrophic consequences within Canadian society. . . would significantly diminish the public’s confidence in the administration of justice”.

Charter challenges, such as those in Morales and Hall, often focus on the legal wording of provisions that authorize preventive detention and overlook how these provisions are actually used and against whom they are used. Concerns about systemic discrimination are relevant at the bail stage in post-9/11 terrorism cases as the overwhelming majority of the accused have been Muslim men, and by definition, all those accused of terrorism will have controversial political or religious motives. In both Morales and Hall, the Court failed to examine the 1991 findings of the Manitoba Aboriginal Justice Inquiry that Aboriginal people in Manitoba were more likely to be denied bail and held for longer periods of time. The commission concluded that “Aboriginal detainees had a 21% chance of being granted bail while non-Aboriginal detainees had a 56% chance”. The Court in Hall did not examine the findings of a subsequent inquiry that found African-Canadian accused were disproportionately denied bail in Ontario. In addition, bail decisions and reasoning will often not be available for public scrutiny until after the trial owing to extensive pretrial publication bans. Bail decisions may also be influenced by increasing risk aversion of judges making bail decisions. This risk aversion is demonstrated by the fact that over half of those imprisoned in provincial correctional centres are people who are formally presumed innocent and awaiting trial.

It is noteworthy that not all those accused of terrorism in Canada have been denied bail. Although many of these bail decisions are not publicly available, they do suggest that some judges are capable of resisting public pressure for pretrial detention. However, there are some published examples of decisions denying bail where the accused went on to receive an acquittal or a stay of proceedings. This false positive issue is particularly pressing in the terrorism context given the complexity of terrorism trials and the lengthy period between arrest and trial.

(i) The Stanikzy Case

Matin Stanikzy was arrested and taken into custody on November 17, 2010. Along with unrelated domestic violence charges, Stanikzy was charged with a number of terrorism offences: counselling the commission of an offence under section 464, uttering threats under section 264.1 and attempting to possess explosives under section 81(1)(d). He was denied bail, and bail reviews were also twice rejected. He was held in detention for almost a year before being acquitted on November 4, 2011. It is worth noting that Stanikzy’s pretrial delay was relatively short for terrorism trials because the evidence against him did not originate from intelligence agencies.

At a bail review, Stanikzy’s detention was denied on all three grounds. There were concerns that he would not attend trial because he had recently arrived in Canada from Afghanistan. There were also concerns about public safety concerning both the threats he allegedly made to kill thirty to one hundred people at a Canadian Forces Base and the threats he allegedly made against his wife. There were also concerns about the public confidence in the administration of justice, with the reviewing judge stressing that the

Crown’s evidence is strong. It manifests from the defendant’s own words. The gravity of the offence is very serious. It falls into the category of the most serious since it includes the threat of indiscriminate wanton killing of innocent Canadians with the intent to undermine the very fabric of our society, rule of law, and our constitutional democracy. . . . Because of the seriousness of the offence, upon conviction, the defendant would potentially face a very long prison term.

Justice Ray rejected the accused’s request that he be allowed to live away from his estranged wife, under house arrest with sureties. This suggests that the primary reason for denying bail was the terrorism charges as opposed to those of assaulting and uttering threats against his estranged wife.

(ii) The Khadr Case

The Abdullah Khadr case is another example of someone held in lengthy pretrial custody without ever being convicted. He was indicted for material support of terrorism in the US and was subject to four and a half years of detention before proceedings to extradite him were stayed. The judge on the first bail review held that the reverse onus that applied to those charged with terrorism offences in Canada would also apply to bail sought in relation to extradition proceedings. The judges at both bail reviews found Khadr had not discharged the onus of demonstrating that he was not a flight risk given his family’s notorious history of involvement in terrorism and the fact that he was apprehended in Pakistan. Justice Molloy also agreed with the prosecutor’s submission that “this is a tertiary ground case, if ever there was one”. She elaborated:

[A] reasonable person apprised of all of the circumstances in this case would be disturbed to learn that Mr. Khadr had been released into the community under the supervision of his grandmother pending his extradition hearing. This case has attracted considerable public attention because of the nature of the allegations against Mr. Khadr and the strength of the evidence connecting him to Al Qaeda terrorists. This is a rare and extraordinary case. If Mr. Khadr were released in these circumstances and then disappeared from the jurisdiction before his extradition hearing, the consequences could be horrific and the Canadian justice system would decidedly be brought into disrepute.

Justice Molloy cited evidence that during a television interview in February 2004, Abdullah Khadr had said “he dreams himself of becoming a martyr for Islam, expressed his admiration for the terrorists who crashed into the World Trade Buildings on September 11, 2001 and referred to Osama Bin Laden as a ‘saint’”.

In 2008, Khadr sought bail review again and relied in part on the Supreme Court’s 2007 decision in Charkaoui. The decision expressed approval for regular bail reviews for security certificate detainees and accepted that the detainee’s threat to national security should decrease with time, while the government’s need to produce evidence of such danger should increase with time. However, Trotter J held that the Court’s decision in Charkaoui was specific to immigration security certificates and found that Khadr’s risk of flight was still present, stating: “[T]he question of whether delay has severed ties is highly fact-specific”. Justice Trotter also concluded that bail should be denied in order to maintain public confidence in the administration of justice. In 2010, Khadr was released when extradition proceedings against him were stayed because American officials abused process when they offered a bounty for his capture and detention in Pakistan.

My point is not to argue that either Stanikzy or Khadr should have been granted bail or that the decisions were unreasonable or wrong. It is only to suggest that the bail issue may be as problematic under criminal law as it is under immigration law. A person charged with a terrorism offence has to establish cause for being granted bail and such a reverse onus is likely consistent with the Charter. The grounds for denying bail are broad and the concept of denying bail to maintain public confidence is vague and somewhat subjective. Information that does not satisfy the ordinary rules of evidence can be used in bail hearings. In the terrorism context, the accused’s expressive and associational activities may, as they did in the Abdullah Khadr case, provide a basis for denying bail. Accused who come from or have been in foreign countries may likely be denied bail. There may be a presumption of danger and the judge may also conclude that public confidence in highly publicized terrorism cases will be adversely affected by granting bail.

Bail practices should not be ignored when realistically assessing use of the criminal law as a response to terrorism. Those denied bail in the Toronto 18 terrorism prosecutions were subject to prolonged detention in administrative segregation, and for the most part, had little success challenging their conditions of confinement. The difference between bail in immigration and criminal terrorism cases is more a matter of degree than a difference of kind. Although the criminal law is rhetorically associated with the presumption of innocence, those accused of terrorism bear the onus of justifying their release before trial. While presumed to be innocent, they may spend years in prison awaiting trial.

B. Preventive Arrests and Peace Bonds

Unlike in the United Kingdom, where preventive arrests have been used extensively, Canada’s preventive arrest provisions have had no reported use. However, there has been use of preventative peace bonds in Canada. Some of the Toronto 18 agreed to preventive peace bonds as a condition of having charges dropped. Ali Mohamed Dirie, of the Toronto 18, was subject to a peace bond after his sentence expired in October 2011. The peace bond prohibited his possession of a passport, but Dirie was able to leave Canada and died fighting in Syria.

It is difficult to know why Canadian authorities have not used the preventive arrest powers and have used peace bonds sparingly. Authorities may have concluded that broad new terrorism offences, as well as existing inchoate offences, are sufficient. Furthermore, a terrorism suspect can be detained much longer if he is charged with an offence and is unable to show cause for bail than if he is subject to a preventive arrest or peace bond. Bail denial will last much longer than the seventy-two hours maximum for preventive arrest or even the one-year period for a peace bond. These provisions will briefly be examined in order to advance the theme that it is a mistake to underestimate the restraints of the criminal law at the pretrial stage and think that it only responds to past acts.

Since 2001, the Criminal Code has allowed judges to impose peace bonds or recognizances to keep the peace and comply with reasonable conditions when there are reasonable grounds to fear that a person would commit a terrorism offence without such orders. The Ontario Court of Appeal has upheld similar non-terrorism provisions from Charter challenge, stressing that peace bonds were included in the original 1892 Criminal Code and had their origins in a 1361 British statute.

The language of the peace bond provisions is vague and archaic in its reference to the oxymoronic concept of “reasonable fears”. Nevertheless, the Ontario Court of Appeal salvaged the provision by stressing that a judge must be satisfied on the basis of evidence that there were reasonable grounds for the fear. The Court of Appeal concluded that peace bonds were a proportionate restriction on liberty. As is common in proportionality analyses, the existence of a potentially more draconian response helped justify the existing response. Here, the court stressed that a peace bond was less drastic than imprisoning the accused.

Peace bond restrictions on terrorist suspects could, however, be as intrusive as some of those imposed in the immigration security certificate cases. Restrictive conditions caused one security certificate detainee to seek full imprisonment rather than impose the hardships caused by his conditions on his family. Similar to some security certificate cases, recent amendments to the Criminal Code’s peace bond provisions provide that reasonable conditions could include prohibitions on the use of the internet, use of electronic monitoring, curfews and geographic restrictions. The difference between preventive detention in the criminal law and immigration law is a matter of degree.

Preventive arrests were also introduced in the 2001 Anti-terrorism Act and have been reintroduced as of 2013, after they were allowed to sunset. They generally require approval from both the Attorney General and a judge, as well as reasonable grounds to believe that a terrorist activity will be carried out. However, when it comes to the individuals subject to the preventative arrest, police are only required to show a reasonable suspicion that their recognizance is necessary to prevent the terrorist activity from being carried out. Although the actual preventive arrest period is capped at seventy-two hours (as opposed to fourteen days in the UK), a person subject to a preventive arrest can be required to enter into a recognizance for a one-year period.

If the past is any indication, the preventive arrest provisions are unlikely to be used in the future. Although the difference between the reasonable grounds and reasonable suspicion standards looms large in constitutional jurisprudence, it is likely smaller in practice. In most cases it would be far more advantageous to the state to lay a criminal charge under one of the broad terrorism offences and argue that the accused has not shown cause for bail, as opposed to employing the novel preventive arrest powers. The “ordinary criminal law”, in the form of bail denial, can result in lengthy preventive detention of the accused without having to resort to the novel and controversial devices of peace bonds and preventive arrests. It is, therefore, a mistake to think that the criminal law will only detain people in response to proven wrongdoing. As we have seen, an accused may be held in pretrial detention for years, only to be acquitted or have proceedings stayed.