An Empirical Comparison of the Exclusion of Evidence Pre/Post 9/11: A Model of Judicial Discourse (January, 2015 issue of CJCCJ, vol. 57, no.1)

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Canadian Exclusion of Evidence Under Section 24(2) of the Charter: An Empirical Model of Judicial Discourse Melanie Janelle Murchison Queen’s University Belfast Richard Jochelson* Department of Criminal Justice, University of Winnipeg En 2005, Margit Cohn et Mordechai Kremnitzer ont créé un modèle multidimensionnel à 17 paramètres pour mesurer le discours judiciaire inhérent à la prise de décisions par les tribunaux constitutionnels. Un tribunal rendait une décision jugée « activiste » lorsqu’elle se trouvait à l’extérieur des limites constitutionnelles traditionnelles ou « empreinte de retenue » lorsqu’elle adhérait aux principes traditionnels. En 2012, Jochelson et coll. avaient opérationnalisé ce modèle pour analyser les changements importants dans l’interprétation, par la Cour suprême du Canada, des dispositions de la Chartre canadienne des droits et libertés sur les fouilles, les perquisitions et les saisies, avant et après les événements du 11 septembre 2011. Cet article utilise maintenant le modèle pour ajouter à cette analyse l’article 24(2) de la Chartre canadienne des droits et libertés sur l’exclusion des preuves. En utilisant l’échelle de Likert (de 1 à 10) pour chaque indicateur de Cohn-Kremnitzer, une valeur a été assignée à chaque variable de tous les dossiers, puis les dossiers précédant le 11 septembre 2011 ont été comparés aux dossiers suivant cette date. Notre analyse des données démontre une retenue accrue de la part de la Cour suprême dans 7 des 13 variables de discours judiciaires mesurés après le 11 septembre 2011, même après avoir tenu compte de la décision historique de R c. Grant (2009). Ces changements correspondent à la littérature sur la sécurisation suivant le 11 septembre 2011 : ces événements ont donné à l’État la permission de contrôler et d’utiliser la sécurité comme justification pour des actions préventives et prudentes. Bien que nous ne puissions pas affirmer l’existence d’un lien causal entre ces changements et les événements du 11 septembre 2011, les avertissements permettent tout de même des découvertes importantes, la plus intrigante voulant que la Cour ait modifié son discours sur l’exclusion de la preuve et fasse preuve de plus de retenue. Ces découvertes sont en lien avec d’autres études qui ont eu des résultats similaires dans le domaine des fouilles, des perquisitions et des saisies. * Please direct correspondence to Professor Richard Jochelson, Dept. of Criminal Justice, University of Winnipeg, 515 Portage Avenue, Winnipeg, MB R3B 2E9; e-mail [email protected]

© 2015 CJCCJ / RCCJP doi:10.3138/cjccj.2914.E08

116

Revue canadienne de criminologie et de justice pénale

janvier 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Mots clés : activisme judiciaire, analyse de discours, exclusion d’une preuve lors d’un procès pénal

Margit Cohn and Mordechai Kremnitzer developed a multidimensional 17parameter model, in 2005, to measure the judicial discourse in the decisions of constitutional courts. A court rendered a decision that was activist when it made a decision outside the traditional scope of judicial constraints on government action and that was restrained when they adhered to the principles of traditional adjudication roles. Previously, this model was successfully operationalized, by Jochelson et al. (2012), to analyse significant changes in the interpretation of search and seizure law in the judicial discourse of the Supreme Court of Canada, before and after 9/11. We now use the model to expand that analysis to section 24(2) exclusion of evidence cases under the Charter. By using a 1–10 Likert scale for each Cohn/Kremnitzer indicium of analysis, a value was assigned to each variable of every case and then the pre-9/11 case group were compared to the post-9/11 one. Our data analysis shows increased restraint on the part of the Supreme Court in 7 of the 13 variables of judicial discourse measured after 9/11, even when factoring in the landmark decision in R v Grant (2009). These changes are consistent with the post-9/11 literature on securitization: 9/11 was a moment when the state was given excuses to control, using security as justification for precautionary and risk-averse actions. While we cannot assert any causal relationships between these changes and 9/11, the caveats still permit significant findings, the most intriguing being that the Court has shifted, in its discourse on the exclusion of evidence, toward an ethic of more restraint. These findings stand alongside other studies that have found similarly in the area of search and seizure law. Keywords: judicial activism, discourse analysis, exclusion of evidence in criminal trial

Introduction Since the terrorist attacks of 9/11, scholars have hypothesized that judicial decision making might change, with the judiciary putting more emphasis on security and the expansion of police powers and less emphasis on the protection of civil liberties, particularly in criminal cases (Lindquist and Cross 2009; Kelly and Murphy 2001; Jochelson, Weinrath, and Murchison 2012; Bloss 2008: 208; Romero 2003). The present article seeks to answer whether, in the wake of 9/11, the Supreme Court has changed the way it discusses constitutionally prescribed exclusion of evidence under section 24(2) of the Canadian Charter of Rights and Freedoms that would otherwise have been excluded.

An Empirical Model of Judicial Discourse

117

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

The Charter is, of course, supreme Canadian constitutional law, by virtue of section 52(1) of the Constitution Act, 1982, and it came into effect on 17 April 1982. Section 24(2) states that Where, in proceedings under subsection (1), a court concludes that evidence was obtained in a manner that infringed or denied any rights or freedoms guaranteed by this Charter, the evidence shall be excluded if it is established that, having regard to all the circumstances, the admission of it in the proceedings would bring the administration of justice into disrepute. (Canadian Charter of Rights and Freedoms 1982)

This means that Canadian courts have the power to exclude evidence if it has been obtained in a manner that is unconstitutional. The original importance of the section was that it provided a constitutional exclusionary remedy, even in situations where evidence was probative, important to the Crown’s case, and reliable. In order for a case to be analysed under section 24(2), there must first be an identified violation of a Charter protected individual right (such as the rights contained in ss 7, 8, 9, 10b, etc.). Justice McLachlin (as she then was) discusses the new role of the Court in R v Hebert (at para 178), stating, The Charter introduced a marked change in philosophy with respect to the reception of improperly or illegally obtained evidence. Section 24(2) stipulates that evidence obtained in violation of rights may be excluded if it would tend to bring the administration of justice into disrepute, regardless of how probative it may be. No longer is reliability determinative. The Charter has made the rights of the individual and the fairness and integrity of the judicial system paramount. The logic upon which Wray was based, and which led the majority in Rothman to conclude that a confession obtained by a police trick could not be excluded, finds no place in the Charter. To say there is no discretion to exclude a statement on the grounds of unfairness to the suspect and the integrity of the judicial system, as did the majority in Rothman, runs counter to the fundamental philosophy of the Charter.

According to Morton and Knopff (2000: 39), by 2000 (before this article’s benchmark of 9/11) the Supreme Court of Canada had used the Charter as a tool for judicial activism – a means of asserting judicial discretion to undo the legislative prerogative of the democratically elected government. The notion of judicial activism has often been used to critique the

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

118

Canadian Journal of Criminology and Criminal Justice

January 2015

perceived political machinations of a Court in contravention of Parliamentary will. As we describe below, we have refined a type of activism analysis as a discourse analysis – an assessment of how the Court speaks about itself in decision making. This allows us to analyse any case before the Supreme Court for discourse effects, regardless of whether the Court is responding to legislation or not. In the case of section 24(2), the Court has interpreted the Charter and is thus making determinations about the scope and application of the Canadian Charter – the supreme law of the land. Using the model we have developed, our data has revealed more restrained behaviour of the Court in the years following 9/11. September 9, 2011 is used as a pivot point of analysis partly in order to compare our results to previous findings, using the same model as for search and seizure jurisprudence (Jochelson et al. 2012), and partly because the general discussion of security and deployment of law in the security and policing literature routinely pivots around 9/11 as a watershed moment (Bloss 2008; Haggerty 2009). Thus, our principle inquiry is whether measured discourse effects align with the conceptions of more secure state actions advanced by socio-legal scholars. In this regard, 9/11 is a convenient temporal pivot point and, as we explain later, the examination of other temporal pivot points appears to support the use of 9/11 as an analytic pivot point in complicated ways.

Development of section 24(2) tests The first significant judicial test post-Charter for whether or not evidence should be excluded under section 24(2) was developed by the Supreme Court in R v Collins, a case which established three factors to be considered before deciding whether or not evidence should be excluded in response to a Charter violation by the state in the course of criminal investigation. The test was an objective, reasonable person test: would the admission of the evidence bring the administration of justice into disrepute in the “eyes of the reasonable person, dispassionate and fully apprised of the circumstances of the case” (R v Collins at paras 33–36).1 The first factor that would be considered under the new Collins test is trial fairness: whether the admission of the evidence in question would affect the overall fairness of the trial (R v Collins at para 55). The analysis was further developed in R v Stillman in 1997 and subsequently became known as the Stillman test (R v Stillman at para 74). In Stillman, Justice Cory for the majority proposed a two-step test: First, is the evidence of a conscriptive nature (i.e., was the accused required to participate in the creation or the discovery of this evidence?) (R v Stillman at para 77). If so, the second step of the test requires identifying whether or not the

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

119

evidence was otherwise discoverable by non-conscriptive means (R v Stillman at paras 102–5). The second factor goes to the overall seriousness of the Charter violation in the case at bar (was it a serious violation or merely a minor one?), and the third and final prong of the test is whether or not the inclusion of the evidence would bring the administration of justice into disrepute (R v Collins at paras 29–33). Thus the state of the law before 9/11 was largely informed by three major questions that an accused was required to answer. First, was the fairness of the trial compromised, a matter which was usually answered in the affirmative if conscriptive evidence was used as a means of conviction. If the trial was not rendered unfair by the use of the evidence, then the court could address the final two issues – how serious the breach was and whether the evidence’s admission would bring the administration of justice into disrepute. The test as developed was notable because it possessed near exclusionary rules in the case of conscripted evidence – an aspect of the test that was favourable to accused persons. This test remained in use until it was completely overhauled by the Supreme Court in 2009 in R v Grant, and a new three-part test was instituted instead. The movement toward the Grant overhaul was a slow and steady one, in which seeds were planted across multiple cases, but a new legal interpretation did not manifest itself until 2009. The new test did away with near automatic exclusionary rules and asked that all evidence be assessed using three broad questions. The exclusion analysis was to be undertaken in light of (1) the seriousness of the Charter breach (including a fulsome assessment of the actions of police and whether the actions were in good faith); (2) the impact upon the accused due to the severity of the breach (including whether the accused suffered serious or multiple violations of interests considered significant by court decisions); and (3) whether the administration of justice would be brought into disrepute by inclusion of the evidence. Factors to be weighed in the inclusion/exclusion of evidence decisions also included the reliability of the evidence and the societal interest in truth seeking and the administration of justice, long-term and prospective, as seen by the reasonable society member (R v Grant at paras 71–79). Each branch of the test would need to be assessed to determine exclusion or inclusion of evidence.

Understanding judicial activism The term judicial activism is a contested one. For example, the term is often used as a means of critiquing a particular decision and suggests, when lobbed at the judiciary, that the judiciary has somehow abused its

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

120

Revue canadienne de criminologie et de justice pénale

janvier 2015

constitutional or legislative role (Kelly and Manfredi 2009: 295). Charges of judicial activism are usually meant to suggest that the court has adopted a political stance, and worse, that this stance has resulted in judicial decision making beyond the “proper limits” of the judiciary (Cohn and Kremnitzer 2005; Jochelson et al. 2012). The literature on judicial activism is, of course, vast. The general conception is that judicial activism occurs at the boundary of judicial decision making, when a court moves beyond its adjudicative function and into the legislative realm (Cohn and Kremnitzer 2005; Jochelson et al. 2012). A Court may appear to make this shift (from restrained to activist) when it makes decisions that trouble the supremacy of Parliament, when it defies the intent of the law’s drafters, when it skirts the plain meaning of legislation, when it makes decisions that affect social policy, when it makes large-scale, high-impact, and significant decisions on the basis of relatively poor and narrow judicial resources (such as facts and social science evidence), when it overturns settled law (statutory, constitutional, or common law principles), when it makes decisions on the basis of materials not in force within the jurisdiction (e.g., American jurisprudential materials or international instruments), when it creates new law or invents legal tests, or when it alters legislative language (Cohn and Kremnitzer 2005) (this list is not meant to be exhaustive). Yet some scholars note that even decisions that seem activist may be emanating from a restrained ethic if a court is implementing its decision on the basis of core constitutional philosophies; for example, using the conception of equality to strike down a law could be considered restrained (rather than activist, despite a possible deviation from Parliamentary intent) if one defined the ethic of equality as a core and settled constitutional foundation (Cohn and Kremnitzer 2005). The advantage of the Cohn/Kremnitzer model is that it provides possibilities for understanding multiple dimensions of activism and moving beyond the binary of activist/restrained to see the Court’s logic as complex and multivalent. Discussions of activism and restraint have often been motivated by political interests; it is not unusual for detractors of the Court to accuse it of unbridled activism (Jochelson et al. 2012). The politicization of the term activism has not been useful to the development of comprehensive empiricism in studying Courts (Jochelson et al. 2012; Muttart 2011).

Cohn/Kremnitzer judicial activism scale In 2005, Margit Cohn and Mordechai Kremnitzer (2005: 333) created what they termed a “multidimensional model of judicial activism and analysis” to measure the levels of judicial discourse that are inherent in

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

121

the decision making of constitutional courts. To do this, they created a model that encompasses 17 different parameters contained within three separate categories (Cohn and Kremnitzer 2005: 333). In a previous article, this model was used to examine the changes in judicial decision making in search and seizure cases, before and after 9/11, demonstrating that the Cohn/Kremnitzer model of analysis can be used to examine changes in judicial discourse over discrete time periods (Jochelson et al. 2012; Jochelson 2009a, 2009b). The present article seeks to operationalize the same model, using it to evaluate changes in the judicial interpretation of section 24(2) of the Canadian Charter of Rights and Freedoms during the same time period both because of its relevance to the supposed state securitization that followed 9/11 and because there is enough case law at the Supreme Court level in the area to generate relatively statistically significant results in a quantitative analysis. Below, the benefits of a judicial discourse model are discussed, and the methodology of the study is explained. Finally, the article discusses the findings and concludes that, ultimately, the Court has, in the area of section 24(2) of the Charter, been using reasoning rooted in a more judicially restrained ethic since the events of 9/11. The deployment of this model is somewhat different than the traditional approach in Canada. Previous studies have looked at a court’s entire pantheon of cases across multiple areas of law to assess activism (Muttart 2007, 2011). Such studies have the potential to obscure results that may emerge in the sub-disciplines of law. A court may be behaving very differently in its criminal law jurisprudence than in its tort law jurisprudence, for example. Further, a court may also be behaving very differently within the sub-discipline. A court’s approach to search and seizure law may be very different than its approach to constitutional exclusion of evidence. Measuring judicial output according to Charter rights in the criminal procedure context yields enough cases to generate data and also allows cross comparison outputs with the necessary precision to see how qualitatively differently a court behaves across different constitutional protections and areas of law. This model has been previously operationalized using search and seizure jurisprudence, and it was found that, since 9/11, the Court has used more restrained language in its decision making (Jochelson et al. 2012). With that data already available, it is now possible to ask how the Supreme Court used language in those cases where it assessed exclusion of evidence in the same time frame. This article operationalizes the adapted Cohn/Kremnitzer model in the context of constitutional exclusion of evidence law. As noted in the

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

122

Canadian Journal of Criminology and Criminal Justice

January 2015

introduction, the legal “tests” for exclusion of evidence have changed several times over the lifespan of the Charter, most dramatically in 2009, in R v Grant. The studies of the legal effects of this change are well worn (Currie 2011: 213; Coughlan 2011; Stewart 2011; Madden 2011), but the Court’s language, apart from precedential effects, can be measured across two broader epochs – pre-9/11 and post-9/11. This will allow the assessment of Court discourse outside the framework of precedent and the identification of any broad changes that occur in the way the Court makes its decisions. The use of 9/11 as the measurement point is somewhat a practical decision. The events of 9/11 have a particular relevance to the study of state security responses that could yield interesting findings about Court language before and after this time point. Even more pragmatically, using 9/11 as the pivot point provides us with enough case law at the Supreme Court level to complete a rigorous analysis and provides ample populations of cases both before and after the 9/11 date.

From asserting judicial activism to measuring judicial discourse While empirical studies of judicial decision making are available in the literature, the Canadian context of legal empiricism remains relatively undeveloped, and certainly, there have been relatively few attempts to utilize the Cohn/Kremnitzer model (Muttart 2011; Muttart 2007; Jochelson 2009a; Ostberg and Wetstein 2007). The application of the model has principally been qualitative and has largely been limited to single case contexts, as opposed to being operationalized across larger Court eras (Muttart 2007; Khosla 2009: 59). Indeed, some scholars have called for more expansive quantitative explorations of activism research to respond to a greater range of activism or restraint (Muttart 2011). The most significant critique of the Cohn/Kremnitzer model stems from the argument that it does not measure judicial activism at all, but instead some new species of judicial analytic (Muttart 2007; Khosla 2009). This concern has been answered, in the present article, by reorienting the Cohn/Kremnitzer model to a discourse-based analysis of court decision making. When we use the term discourse analysis, in this context, we are not employing the conception of discourse developed by Foucault, which refers to institutional knowledge patterns that inhere in disciplinary structures connecting knowledge and power (Kendall and Wickham 1999). We are, however, interested in the broader conception of discourse studies that understand language as social interaction concerned with specific social contexts in which discourse is mostly embedded

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

123

(Keller 2013; Bhatia 1993). Here, we examine the legal text as the professional setting in which contact with the socio-political world occurs. We use the text of cases to uncover discourse in terms of how the Court sees itself in relation to the constitutional role of the Court established by Parliament through constitutional law. Thus the discourse analysis we undertake is more limited to an exploration of how the Court sees itself in relation to Parliament, the constitution, and the governance of society through its constitutional interpretations. Rather than asking whether the Court has behaved in an activist or restrained fashion – which is, in itself, a political call – the focus of the analysis becomes the content of what a Court has said about its own decision making. The effect of this change is to alter the judgement call of the researcher in the coding of judicial text, eliminating individual inclinations to lob accusations of activism at a court. This reorientation, then, assures us that the text of the decision is the empirical source of primary research. The reorientation of the project to textual empirical method means that the Cohn/Kremnitzer factors that attempt to measure social outcomes of cases (and, in turn, to judge a Court as activist or restrained) must be the work of a separate, differently empiricized project (e.g., social effects could be recorded through media reactions to a court case, etc.). This reformulation of the Cohn/Kremnitzer method is essentially an “in their own words” methodology in relation to a Court and prohibits, therefore, examining the words of other entities, such as media, politicians, or other courts.

Previous limits overcome? One of the limitations of measuring discourse might be that it precludes making claims that a Court has made an activist or restrained decision. The discourse method asks, instead, that we analyse whether a court uses language of restraint or of activism and to quantify its usage. The recording of language as an indication of restraint or activism is grounded in measurable deliverables and, thus, is more likely to provide data to support or refute political claims of a Court as being activist or restrained. This adapted model is, in essence, an attempt to resolve the limitations of typical studies of judicial activism, which represent a qualitative response to a qualitative question. Discourse analysis provides a means of assessing larger quantities of cases and providing an empirical basis for qualitative conclusions about activism or restraint, using the Court’s own language. This means that new discussion can begin to measure a Court’s output, rather than beginning with the political question of whether a Court has behaved in a legitimate or

124

Revue canadienne de criminologie et de justice pénale

janvier 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

illegitimate manner, a question that may, sometimes, relate in the pundit’s mind to whether a decision can be described as restrained or activist. Some scholars have openly questioned whether a study of a single case could ever reveal whether a Court is activist or restrained – the case may be an outlier, for example (Muttart 2007: 66). Previous assessments have required that the scholar uncover a Court as activist or restrained, and others have attempted to measure the political factors that form part of judicial reasoning (Muttart 2007; Jochelson 2009a: 246). For instance, Ostberg and Wetstein (2007) recorded the pre-appointment political affiliations of judges and then assessed whether the attitudes implicit in these affiliations affected judicial decision making. The present article will contribute a new empiricism to the current discussion, using a model that seeks to foster debate about court analytics and to identify the statistical trends that emerge. The model is an assessment of judicial discourse that takes the Court’s own words seriously. The present article suggests that there is potential in assessing judicial analytics apart from measuring precedential effects (Jochelson et al. 2012). Below, the methodology of the study is explained and the findings are reported. Finally, there is a discussion of the results, followed by a conclusion about the results.

Methodology Sampling The sample for this study consisted of all Canadian Supreme Court criminal cases dealing with section 24(2) that discuss the exclusion of evidence, since the Charter’s inception in 1982. All of the cases were found on the legal databases Westlaw, Quicklaw, CanLII, and LexisNexis, and the databases were cross-referenced with each other to ensure that all cases that contained the search terms “section 24, s 24(2), Charter, Evidence, Exclusion” were included. To ensure reliability, this was the same process used in the previous operationalization of this model (Jochelson et al. 2012). The search was done in October 2013 and created an original sample size of 414 cases, spanning 31 years. Each case was then individually analysed for both content and relevance. For example, cases that only made a passing reference to section 24(2) (such as R v Duhamel) or that only explicated another section of the Charter as a comparison were excluded, due to their lack of detailed analysis. Cases that were less than 10 paragraphs long were also excluded from the

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

125

sample, since they would provide too little material for discourse analysis. A total of 98 cases remained for analysis; the first case for section 24(2) was in 1985, and the end point was in 2013 (these cases are listed in Appendix A). To determine whether or not a change had taken place in judicial discourse, the sample was divided into pre-9/11 cases and post-9/11 cases. There were 72 cases in the first group (pre-9/11) and 26 cases in the second group (post-9/11). Each of the relevant individual variables Cohn and Kremnitzer created were created as ordinal variables so that differences in cases could be ranked and meaningful comparisons could be created. A 1–10 (10-point) Likert scale was created to have a scale rank of each individual variable, with 1 indicating the lowest level of judicial activism, while a 10 represented the most activist type of discourse. During initial data collection, it became apparent that, in this study, only two of the three dimensions should be measured, as the second dimension, socio-legal deviation (legislative reaction, administrative reaction, or public reaction) is not relevant to this project for reasons outlined earlier. This left only 13 of the 17 variables to be analysed, in large part because the goal of this specific project was to measure judicial discourse, not sociopolitical reaction to judicial decision making.

Variables: Traditional visions of activism2 Judicial stability. Here, we measure whether the Court was ready to depart from its own or former decisions. Departure from precedent is a fairly standard indicator of judicial activism (Cohn and Kremnitzer 2005). When the Court affirmed the decisions of all lower courts, we scored the case as closer to 1 and as more restrained. When the Court overturned a previous decision and overturned legislation and created new law, we score the case as closer to 10 and more activist. For all the variables we discuss, it is worth noting that we used the scales contained in Appendix B. We scored cases across the entire Likert range in each of the variables. Interpretation. Did a Court interpret a legal text in possible contradiction with the assumed original intent of the constitution or its plain linguistic meaning? Where the Court interpreted section 24(2) in light of an original intent or plain meaning approach, we scored closer to 1, and therefore restrained. Where the Court interpreted the section in a way that was unremittingly interpretive, we scored the case as closer to 10 and as more activist.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

126

Canadian Journal of Criminology and Criminal Justice

January 2015

Majoritarianism and autonomy. Here, we measure whether the Court interfered with policies set by democratic processes and whether the Court was willing to supply its own solution and/or policy and was thus activist. When the Court did not interfere with policies set by democratic policies or left all legislation unimpeached, we scored the case as closer to 1, as more restrained. Where the Court struck down legislation and applied its own policy or solution, we scored the case as closer to 10, as more activist. Judicial reasoning: Process/substance. With these indicia, we measure how heavily the Court relied, in its decision, on strict legal and procedural grounds, and thus behaved in a restrained fashion. Where the Court entirely relied on strict legal or procedural grounds in making a decision, we scored a case as closer to 1, as more restrained. Where the Court relied on open-ended legal tests, such as reasonableness-based assessments, we scored a case as closer to 10, and thus more activist. Threshold activism. Here, we measure the extent to which the Court was willing to forgive threshold hurdles. In this context, a rigorous application of threshold issues, such as denying that the state caused the breach, would score as closer to 1, and be more restrained. Where the Court found new reasons to exclude or include evidence where previous cases had not, we score the case as closer to 10, as more activist. Judicial remit. These indicia measure whether the decision expands or redefines the jurisdiction of the Court. When the decision did not expand or redefine the jurisdiction of the Court we scored the case as closer to 1 and as more restrained. A decision that expanded the judiciary’s remit into areas previously immune from intervention was scored as closer to 10, as more activist. Rhetoric. These indicia measure whether a judicial decision was used as a platform for expression of broader positions and values or whether the use of extra-legal rhetoric was restricted in the Court’s explication of legal principles. We scored the absence of rhetoric (usually correlating with shorter decisions) as closer to 1, and more restrained. High levels of extra-legal rhetoric, combined with long discussions of political implications, were scored as closer to 10, as more activist. Obiter dicta. These indicia measure how far the Court expanded its opinion beyond the legal requirements of the specific case. When the Court did not delineate any obiter, we scored the case as closer to 1 and more restrained. When the Court used extensive amounts of obiter and

An Empirical Model of Judicial Discourse

127

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

discussed issues not relevant to the case we recorded the score as closer to 10, and as more activist. Reliance on comparative sources. Here, we examined how extensively the Court relied on foreign sources that are not legally binding in the domestic sphere. Where the Court used domestic law exclusively we scored the case as closer to 1, as more restrained. When the Court used comparative sources to create new legal conceptions with extensive comparative referencing we scored the case as closer to 10 and as more activist since extra jurisdictional materials would not have been intended by Parliament to be part of the interpretive context of courts. Judicial voices. Here, we examined the extent of other judicial decisions, besides the majority decision. A unanimous decision was scored as closer to 1, and as restrained. On occasions where we saw two concurring and two dissenting judgments (the most judicial voices we saw), we scored as closer to 10 and more activist. This indicium is based on Cohn and Kremnitzer’s assertion that a unanimous Court is more likely to be reined in and bound by traditional judicial constraints and less likely to contain any “maverick” opinions (2005). Extent of decision. Here, we examined whether the Court’s ruling expressly applied to a single or specified set of circumstances or whether the law that resulted had broad implications for larger sections of society. If the Court simply applied the legal rules, we typically scored the case as closer to 1 and as restrained. Where the Court created a new standard that effected broader populations, we scored a case as closer to 10, as more activist. These indicia were based on Cohn and Kremnitzer’s assertion that decisions with high “extent” act more like legislation and are, therefore, a departure from the traditional judicial role (2005). Legal background. Here, we examined whether the legal framework on the basis of which the Court made its decision was inclusive and clear or whether the rules concerned were vague, complex, self-contradictory, or incomplete. Where the Court applied clear rules that did not extend beyond the prior case law, we scored the case as closer to 1 and as restrained. Where the framework was murky and where the Court generated a new framework for analysis, we scored the case as a 10 and as more activist. These indicia are based on Cohn and Kremnitzer’s assessment that decisions with new and abstract frameworks tend to apply to larger populations and are thus more legislative in nature than adjudicative (2005).

128

Revue canadienne de criminologie et de justice pénale

janvier 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Variables: Core values activism Intervention and value content. Here, we examined whether the subject matter under examination was highly value-laden, in that it had bearing on democratic principles and human liberties accepted domestically. Where the case dealt with important human rights issues, and the Court appeared to assert its guardianship of the Constitution, we scored the case as closer to a 1 and as restrained. Where the Court declined to discuss the constitutional values at stake, we scored toward a 10 and closer to activist. This indicium is based on Cohn and Kremnitzer’s assertion that a focus on core values may offer a vision of restraint different than the traditional one; that is, core values can explain the restraint in a decision that otherwise might appear to be unabashedly activist (2005). For example, the creation of a new constitutional test may, in light of traditional visions, seem activist; but if the new test asserts a core value, the restraint implicit in the core value being asserted offers a compelling counter-description for explaining what would otherwise be seen to be an activist decision. This reasoning is the one asserted in traditional core value arguments, but could be understood by thinking of core values of the Charter as the supreme will of Parliament.

Sample cases To better understand the process by which the Cohn/Kremnitzer model can be operationalized – namely, how some cases were scored as activist, others as restrained, it is important to understand how the coding decisions were made in each case. This section will explain how each of the variables was coded using two sample cases to illustrate the substantial differences between activism and restraint in judicial discourse. These cases occurred within one year of each other: R v Grant is one of the most activist cases included in the data set, while R v Wittwer, decided just one year earlier, is one of the most restrained (using the discourse model). In Grant, two plain-clothed police officers stopped the accused, a younger black male, after noticing his “suspicious movements.” After a brief conversation with these officers, Grant disclosed that he was carrying marijuana and a firearm. He was subsequently arrested. At trial, Grant submitted that his section 8, 9, and 10(b) rights had been violated, and that the evidence should be excluded under section 24(2). He was convicted at trial, and the Court of Appeal upheld his conviction. In Wittwer, the police attempted to obtain a confession from the accused using his own previous admissions, which the police knew had been obtained while violating his right to counsel. After

An Empirical Model of Judicial Discourse

129

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

repeated questioning, Wittwer confessed, but later submitted that this confession was still a violation of his 10(b) right, as the interrogations were all connected, and sought to have it excluded under section 24(2). As set out in Appendix B, each Cohn/Kremnitzer variable has a possible score of 1–10, and each of those scores corresponds to a particular behaviour of the Court or result of the proceedings. For example, the variable judicial stability tells us whether or not the Court overturned previous decisions, while the number tells us to what degree they varied from those previous decisions. A 1 would indicate that they did not overturn any previous decision, while a 10 would tell us that they overturned not only a previous Supreme Court decision but also legislation, by making a new test. These cases will illustrate the differences in each variable specifically. The decision in Wittwer is only 27 paragraphs long, while, at 230 paragraphs, Grant is more than 10 times that length. The length of the Grant decision is strongly correlated with the high score for both rhetoric and obiter, as, in longer decisions, there is more space for engaging in such behaviours, and these behaviours tend to make the decisions longer. Wittwer scored 1 on both rhetoric and obiter, while Grant scored 10 on both. Wittwer is also a unanimous judgment, with no mention of extra-jurisdictional case law or social science evidence, while Grant uses extensive American case law. This means that Wittwer scored 1 on both comparative sources and judicial voices, but Grant scored 8 and 4 on the same variables, respectively. The only variables that received the same score in both cases were judicial remit and majoritarianism, and both scored 1, as the Court neither expanded its jurisdiction in these cases, nor did it overturn democratically enacted legislation. In Wittwer, the Supreme Court overturned both the appeal court and the trial court’s decisions and held that evidence obtained by violating Wittwer’s section 10(b) right to counsel should have been excluded and ordered a new trial. Because the Supreme Court overturned both lower court decisions, this case scored 5 on judicial stability. Meanwhile, in Grant, the Supreme Court agreed with both lower court decisions on four out of the five charges but overturned both lower courts on the fifth charge. This behaviour alone would have led it to also score 5; however, the Court went further and completely overhauled the test for section 24(2), which then met the criteria for a score of 8 on judicial stability. It is because of this substantial change in the exclusion-of-evidence test that Grant also was given the highest available score – 10, for the variables legal background, judicial reasoning, and extent of decision. In Wittwer, the Court

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

130

Canadian Journal of Criminology and Criminal Justice

January 2015

merely applied previously created tests, and the decision did not extend beyond any previous case law, so it scored the complete opposite and the lowest possible value – 1, on the aforementioned variables. Threshold activism measures the degree to which the Court is willing to forgive or overlook the threshold for the Charter protection (for s 24(2), that threshold is whether or not there is a temporal connection between the breach and the remedy). Wittwer scores 1 on this measurement, as the Court explicitly discussed the temporal connection and its importance, while Grant scored 3, as the Court was willing to impose limits on this threshold, in this case. The variable interpretation measures the level of original intent, and Wittwer scored 1 on this variable, as the case held true to what is set out in section 24(2). Grant scored 3, however, as the majority in the decision did not discuss the original intent of the Charter, though the concurring justices did. Finally, in core values, the last variable measured, Wittwer scored 1, as the justices in the case discussed the importance of the adherence to Charter values and protections and excluded the evidence because those values were violated by police conduct in that case. In Grant, there was no discussion of the inherent values at stake but the justices did interpret those values purposively, so it scored 5.

Data analysis All of the coding (including that which was done in Jochelson et al. 2012) was done by the same researcher to ensure reliability. The cases were read in chronological order, and each case was read through in its entirety twice and was coded according to the created scale (see Appendix B). After the data was collected and coded, it was loaded into the IBM program “Statistical Package for the Social Sciences” (SPSS) for analysis. It was at this time that statistics were run for the relevant time periods. Prior to this, the only information that the coder possessed was the instruction to code. All 13 variables for the period from 1984 to 2001 (pre-9/11) and from 2002 to 2013 (post-9/11) were compared by recording the means, and calculating mean differences and t-statistics to ascertain if judicial decision making had been impacted in the aftermath of the 9/11 attacks. The sample size for section 24(2) is slightly bigger (98 cases) than the initial operationalization of the model using section 8 (90 cases), but is still not large, making it more difficult to achieve reliable outcomes and high levels of statistical significance. Thus, the data are reported using one relaxed standard of statistical significance (alpha = .10) and the two typical standards of statistical significance (alpha = .05; and alpha = .01).

An Empirical Model of Judicial Discourse

131

Results and discussion of Section 24(2) Jurisprudence

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Descriptives Overall, the mean scores were relatively low, with only one of the Cohn/ Kremnitzer variables attaining a mean greater than the median score of 5, or the middle measurement in the scale (see Table 1). This indicates overall restraint on the part of the Court. The variable rhetoric received the highest overall mean score, at 5.29, with judicial voices (4.96) being the only variable approaching the median measurement. These were the only variables to achieve a mean value greater than 4, as core values had a mean at exactly 4.00. Obiter had the next highest mean score, 3.86, while interpretation, judicial stability, and activism threshold, all clustered closely together, scored approximately 3.5 (3.59, 3.54, and 3.53, respectively). Extent of decision had a mean of 3.10, while the means for judicial reasoning and legal background were 2.71 and 2.55. Three variables had means under 2, which indicate extreme restraint by the judiciary. Comparative sources (1.92), majoritarianism (1.26), and judicial remittance (1.24) all scored in the bottom 19% of the scale and demonstrate the Court’s reluctance to overrule Parliament, go beyond its instructed scope, or rely on foreign precedent when making decisions. The standard deviation and range of scores show considerable variability in the level of involvement by the Court in all of the judicial discourse variables. It can be observed that the standard deviations for this sample are relatively high, as are the ranges between the minimum and maximum scores reported for each indicator. This is logical, as the variables Table 1: Section 24(2) descriptives of multidimensional indicia Variables

Mean

Std. Deviation

Range

Minimum

Maximum

N

Judicial stability Interpretation Majoritarian/Autonomy Judicial reasoning Activism threshold Judicial remit Rhetoric Obiter Comparative sources Judicial voices Extent of decision Legal background Core values

3.54 3.59 1.26 2.71 3.53 1.44 5.29 3.86 1.92 4.96 3.10 2.55 4.00

2.140 2.238 0.803 1.795 2.197 1.075 2.964 2.528 1.679 3.025 2.602 2.429 2.411

9 9 5 9 9 5 9 9 9 9 9 9 9

1 1 1 1 1 1 1 1 1 1 1 1 1

10 10 6 10 10 6 10 10 10 10 10 10 10

97 97 97 97 97 97 97 97 97 97 97 97 97

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

132

Revue canadienne de criminologie et de justice pénale

janvier 2015

that the Court considers in a given case are highly contextual and vary from case to case to a larger degree than for more simplistic events. Several of the indicia included rankings as high as 10 and as low as 1, and most ranged over the full Likert scale, meaning that every number from 1 to 10 was assigned to most variables.

Pre- and post-9/11 mean scores It is important to recognize that, while it uses pre- and post-9/11 as the timeframe for measurement, this article is, in no way, attributing the changes found to the events of 9/11 directly. While there may be some strong correlations pre- and post-9/11, these do not prove a causal relationship; it is not being suggested that 9/11 caused these shifts to occur. It must be noted that the changes measured may be, in part, due to Court composition issues, Court leadership issues, as well as to sociopolitical factors that discourse analysis cannot uncover. The result, when read along with the results of the previous section 8 findings in Jochelson et al. (2012), indicate a Court that, on the whole, is more restrained in terms of its use of language since 9/11 – a Court which is speaking in less activist terms than the during the pre-9/11 period (see Table 2). Judicial activism decreased in a narrow majority (8/13) of the variables from pre-9/11 to post-9/11. This trend is in keeping with our previous analysis of section 8 jurisprudence, albeit a less drastic one, as, in our section 8 analysis, all but one variable decreased (Jochelson et al. 2012: 200). In the wake of 9/11, it is possible to posit that, initially, Courts would respond to security concerns using more restrained language as to the scope of constitutional rights. However, the lack of similar discourse-based restraint by the Court after 9/11 may simply demonstrate that the Court had begun its slow march toward the exclusion-ofevidence revolution that occurred in R v Grant. A less restrained Court would be evidenced by a Court that was slowly beginning to dabble in discourses dealing with the constraints in applying section 24(2). Unfortunately, at the time of data collection not enough cases discussing Grant had occurred since 2009 to make any statistical findings regarding the impact of Grant in post-Grant cases as compared to pre-Grant cases. The most significant shift can be seen in the comparative sources variable, which decreased by 49.3% after 9/11, and the decrease was statistically significant at the highest measurement (.001). This could be attributed to the fact that, in 2001, the Charter had already been operational for the past 19 years, and it may be that the Court felt more

An Empirical Model of Judicial Discourse

133

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Table 2: Comparison of means pre-9/11 and post-9/11 in section 24(2) Supreme Court decisions Variables

Pre-9/11

Post-9/11

N Judicial stability

72 3.61 (2.13) 3.46 (2.16) 1.33 (.92) 2.88 (1.84) 3.50 (2.13) 1.39 (.93) 4.93 (2.90) 3.58 (2.42) 2.21 (1.86) 5.03 (3.08) 3.33 (2.71) 2.78 (2.52) 3.82 (2.30)

26 3.35 (2.21) 3.96 (2.46) 1.04 (.20) 2.23 (1.61) 3.63 (2.42) 1.58 (1.42) 6.27 (2.97) 4.62 (2.71) 1.12 (.43) 4.77 (2.91) 2.46 (2.20) 1.92 (2.08) 4.50 (2.67)

Interpretation Majoritarian/Autonomy Judicial reasoning Activism threshold Judicial remit Rhetoric Obiter Comparative sources Judicial voices Extent of decision Legal background Core values

Difference of Means‡

t statistic

−7.20%

−0.539

14.45%

0.983

−21.80%

−1.617*

−22.57%

−1.581**

3.71%

0.228

13.67%

0.763

21.37%

2.005*

29.05%

1.805†

−49.32%

−2.957***

−5.16%

−0.372

−26.13%

−1.473

−30.94%

−1.549†

17.80%

1.237

‡

The Mean Difference was calculated as follows: [(Mean1−Mean2)/Mean 1] ∗ 100 p < .10 * p < .05 ** p < .01 *** p < .001. All tests one-tailed. Standard deviations are in parentheses. †

confident relying on a new wealth of previous Canadian jurisprudence, no longer feeling the need to draw from other jurisdictions. The Court itself stated that it was less likely to cite or be influenced by other nation’s legal systems, so this result is no surprise. The Court’s restraint in relying on extra-jurisdictional sources to justify its legal prose can also be explained by understanding the evolution toward the Grant case as a Canadian response to a Canadian test, one that was originally created through the Collins/Stillman line of cases. It would be unsurprising that a Court seeking to repair defects in seminal cases would work within the Canadian context.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

134

Canadian Journal of Criminology and Criminal Justice

January 2015

Also decreasing by a substantial and statistically significant amount after 9/11, legal background and extent of decision (30.1% and 26.1%, respectively) demonstrate a substantial shift in the Court’s behaviour (see Table 2). At first glance, this may appear incongruous with our understandings of Grant’s implications, but it can be explained by understanding the development of section 24(2) jurisprudence as occurring at three distinct points – Collins in 1987, Stillman in 1997, and Grant in 2009. Unsurprisingly, the pre-9/11 measurements show more discourse in line with affecting large populations, since two major jurisprudential changes occurred before 9/11. Noteworthy is that, in neither epoch (pre- or post-9/11), did the variable of extent of decision exceed 4, indicating that, despite three relevant section 24(2) legal overhauls, the Court expressed itself as relatively restrained in making large-scale legal change. The same reasoning could explain the restraint it showed in its discourse by choosing to establish clearer and more concise guidelines for the legal framework it set (legal background). Collins and Stillman were cases that were seminal in the development of section 24(2) jurisprudence, and their opaqueness was largely what had to be redressed in the Court’s decision in Grant. This is not to say that the Grant test provides a perfect or sound metric for section 24(2) exclusionof-evidence analysis. Rather, it demonstrates that, on the slow march to the Grant decision, the Court began to use language aimed at tightening up the jurisprudence for exclusion of evidence, in an attempt to solve some of the problems the Court perceived as resulting from the Collins/ Stillman factors. The variables obiter and rhetoric showed large increases post-9/11 (29% and 21%, respectively), also with statistical significance. Post-9/11, the Court also began to use more activist language in terms of reaching outside the four corners of the case in front of it and relying on principles not directly at issue in the case at hand (obiter) as well as in its use of discussion that seemed to enter the political realm (rhetoric). Again, these changes post-9/11 can be seen as part of the evolution process that would eventually lead to the Grant decision. Grant represented a significant retreat from the Collins/Stillman test (Currie 2011: 213; Coughlan 2011; Stewart 2011; Madden 2011). It is reasonable to assume that the Court would be well aware of such concerns and would begin to lay the philosophical foundation for making a change to the section 24(2) analysis, across several years. In particular, during this time period, the Court began to become concerned about the stringency of the pre-Grant near automatic exclusionary rules (see R v Orbanski; R v Elias [Orbanski/ Elias], for examples). For example, Justices Fish and Lebel, in dissent, in

An Empirical Model of Judicial Discourse

135

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Orbanski/Elias, noted that the majority had interestingly avoided exclusion-of-evidence analysis: It is likely that few Charter provisions have generated so much academic comment, conflicting jurisprudential developments, media rhetoric or just plain uneasiness as section 24(2). Since the Charter came into force, our Court has returned on many occasions to the interpretation and application of this provision. It has developed and refined methods of analysis and application. Despite all these efforts, doubts and misunderstandings remain. They arise mostly from views, which attempt to read into the jurisprudence of our Court the creation of an exclusionary rule in the case of conscriptive evidence. (Orbanski/Elias at para 87)

Justices Fish and Lebel (Orbanski/Elias at para 93) also note that cracks in the Collins and Stillman approach were evident in the Court’s reasons in Buhay (2003), Law (2002), and Fliss (2002). The dissenting decision provided a written record for the slow change in law that would ultimately result in Grant (2009). Judicial discontent with the state of established law would likely show up in measurements of deviation from past decisions and in the Court’s use of stricter limitations in application of law. This is demonstrated by “Judicial Reasoning” decreasing by 22.6%. The variables of “Core Values,” “Judicial Remit,” and Interpretation” increased by 17.8%, 13.7%, and 14.6%, respectively, while “Judicial Stability” and “Judicial Voices” decreased by 7.2% and 5.16% in that order. These were not statistically significant shifts but can still be informative about the changes taking place in the Court. The expansion of a Court’s jurisdiction by measuring judicial remit also indicates results consistent with the development of the Grant test. The Grant case was a capstone on the Court’s tendency to reify the findings of the trial Court in exclusion of evidence matters. This was a matter explained in the Collins/ Stillman line of cases, but one that became significantly more prominent in the evolution of the Grant case. Indeed this type of prose is evident in cases such as Buhay or Orbanski/Elias, as examples. The Court’s discussion of its jurisdiction would trigger higher remit findings, even when, ultimately, the Court delineated the primacy of trial findings. Again, our analysis measures the amount the Court speaks of jurisdiction, not the resultant jurisdiction the Court purports to hold. The less markedly significant increase in restraint in the Court’s willingness to retreat from past cases (judicial stability) and the Court’s

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

136

Revue canadienne de criminologie et de justice pénale

janvier 2015

tendency to foster more unanimous decisions (judicial voices) represents interesting findings that should not be overstated. The small increase in fidelity to prior cases is a measure of the extent to which the Court spoke about fidelity to prior cases. Though the Court deviated from past precedent in Grant, in substance, it spoke of those deviations in a manner that made Grant seem to be a logical extension of previous section 24(2) jurisprudence. This is a typical style of judicial prose, which allows for common law development, and could be expected in situations where a Court is looking to tread lightly before making dramatic legal change. Similarly, the increasing tendency for the Court to issue unanimous decisions could be more related to issues such as Court composition and leadership. It has now been established that the Court under Chief Justice McLachlin is more likely to render unanimous decisions than previous Courts did (Jochelson et al., 2012). The Court’s shift toward higher measurements in core values can also be read in light of the changing exclusion-of-evidence landscape. The higher measurement indicates less language used by the Court referring to its stewardship of the constitution. Since the evolution of the Grant test would lead to the absence of automatic exclusionary rules and a more explicit focus on the reliability of the evidence, it is unsurprising that the Court would speak less of its need to stand firm between the individual and the state – such language would indicate a Court that was taking its stewardship less seriously, and undoubtedly a Court would speak less of that core value before making an important decision like R v Grant (Stuart 2010: 313).

Explaining confounding effects: What is the meaning of post-9/11? When choosing 9/11 as our pivot for comparison, we wanted to make sure that we could be confident in the results of our statistical analysis. In particular, we wanted to determine whether the fact that cases were post-Grant confounded our findings for cases post-9/11, although as we explain below, this is a complicated question. We also wanted to explore whether using a different point of bisection of the data might yield results similar to our primary analysis, in order to determine whether 9/11 was a useful pivot point for analysis. We assert that 9/11 is a useful point to explore, given that the extant security literature uses 9/11 as a jumping off point (Lindquist and Cross 2009; Kelly and Murphy 2001; Jochelson et al. 2012; Bloss 2008). Nonetheless, we wanted to see if our findings applied to other temporal pivot

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

137

points. To do this, we ran a pre/post comparison of the median (or the exact mid-point case in our data set), which occurred in 1994. In doing this, we found that, after that pivot point, judicial restraint only increased in 5 of the 13 variables, and less significantly, which demonstrates that the restraint does, indeed, show an increase when 9/11 is used as the pivot point (in our analysis 7 of 13 variables show restraint after 9/11). The increased restraint scores align with conceptions that using 9/11 as the pivot point demonstrates unique effects when compared with the median-point analysis (See Table 3). We were interested in seeing whether measurements after 9/11 aligned with claims made in the security literature; we are not claiming that 9/11 was causative of restraint findings. Nonetheless, our analysis of the median point bolsters claims about the salience of 9/11 as a security inspiring event, or at least using 9/11 as a useful launching point for discussing judicial restraint. In Grant, the test for exclusion of evidence under section 24(2) was completely overhauled. In the five years since Grant was decided, 12 section 24(2) cases have been decided that meet the criteria for analysis, which means that approximately half of the post-9/11 cases were decided before Grant and the other half after. Because Grant had such substantial effects on the resulting jurisprudence, we ran tests to see whether or not Grant had the same effect on our data set. In particular, we were interested in measuring whether post-Grant cases were so restrained that they would have confounded our results showing overall restraint after 9/11. First, great caution must be taken in looking at these results, as the sample size of cases is quite small in a pre/post-Grant analysis. Second of all, it is important to note that restraint post-Grant could be seen as a 9/11 effect as much as a post-Grant effect, since, as some scholars have argued, Grant can be read as a securitizing jurisprudential moment post-9/11 (Jochelson and Kramar 2014). What we found was that when we controlled for Grant, and ran all of our data before Grant and after Grant, we still had judicial restraint increasing on the same 7 of the 13 variables as in the post-9/11 analysis. Grant’s effect on Table 3: Number of variables classified as activist or restrained Era Comparisons Pre-9/11–Post-9/11 Pre-Median–Post-Median Pre-Grant–Post-Grant Pre-9/11–Post-9/11 (with post-Grant cases removed)

Variables Demonstrating Restraint

Variables Demonstrating Activism

7 5 7 9

6 8 6 4

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

138

Canadian Journal of Criminology and Criminal Justice

January 2015

measured restraint is undoubtedly important then, even given the small sample size (see Table 3). Yet, if we look for restraint post-9/11 and remove cases following Grant, we, in fact, see restraint in 9 of 13 variables, indicating that temporally post-9/11 cases had a greater effect on restraint than when we include the post-Grant data (9 variables of restraint as opposed to 7). This strengthens the suggestion that the temporal lens of 9/11 correlated with greater measured restraint, and that post-Grant results do not confound the analysis and indeed might be supportive of the analysis. The insertion of the post-Grant cases into the equation lowers restraint in two variables. This does not mean that restraint is on the decrease in perpetuity but shows that the immediate reaction after Grant was modestly less restrained (perhaps since interpreting Grant requires some creativity by the Court), and supported the overall trend of post-9/11 restraint measured within the first 8 years after 9/11. Our results using 9/11 as the pivot point thus owe some of their import to the Grant decision, but since Grant can be read as a response to securitization and state action after 9/11 it would be difficult to divine whether Grant or 9/11 was causative of our findings. Given the academic literature on Grant as a security-based response to 9/11, the results for pre/post-9/11 and pre/post-Grant can both be read as showing a securitizing of jurisprudential language and as intertwined (Jochelson and Kramar 2014). In short, if Grant gave effect to securitizing logic (arguably a response to 9/11), then comparing restraint pre/post-Grant and pre/post-9/11 may, indeed, be an attempt to measure the same phenomenon through slightly different temporal and interpretive lenses. The restraint measured could be affected by a post-9/11 effect, a post-Grant effect, a Charter maturation effect, or a Court composition effect. The central point is that restraint is measured at a time when critical scholarship has argued that security-based state responses were on the rise. Regardless of how our results are interpreted, our findings align with these literatures. When we control for post-Grant by removing cases after the pivotal decision, temporal responses post-9/11 show restraint in 9 variables (compared to 7 when post-Grant is included in the analysis).

Conclusion The present article continued to adapt the Cohn/Kremnitzer method and has found significant changes to the way the Supreme Court has used discourse in exclusion of evidence cases since 9/11 (but not

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

139

necessarily because of 9/11): Canadian judicial discourse has shifted substantially toward restraint over activism in 7 of the 13 variables measured, a phenomenon that mirrors the authors’ earlier analysis of search and seizure cases (Jochelson et al. 2012). The events of 9/11 have been described as a discursive moment, where states were provided with excuses for governing more coercively, as the need for security seemed to justify precautionary and risk averse actions; the findings of the present study appear to support those claims (Bloss, 2008: 208; Romero 2003; Haggerty 2009; Ericson and Doyle 2003). We do not assert that 9/11 caused these changes, merely that these findings are correlative in nature. They could possibly be explained by other situational factors, such as Court composition, leadership, administrative shifts, and jurisprudential (i.e., post-Grant) changes (although the latter is less likely, given our finding that removing post-Grant cases shows restraint in even more variables in the post-9/11 era). The point, here, is that in the era of scholarship that asserts governmental responses to security are on the rise, we are seeing apposite changes in the way the Court talks about itself (i.e., in a more restrained fashion). Socio-legal scholars and criminologists often look at the Court’s output in terms of its disposition effects – that is, how does this legal interpretation affect persons, policies, and society? Revealing the layers of how a Court packages its reasons through discourse analysis allows us to see the way the Court forms its reasons and opens up the possibility of exploring the social links that inform judicial reasoning. This approach considers the formation of law in process rather than only considering the output of law, and it permits us to park our political judgements (at least, until the coding exercise has ceased). Notes 1 This test replaced the former Rothman Test (which was intended to deal with confessions) but developed from that case, in 1981 (R v Rothman). Because R v Rothman was a pre-Charter decision, the test was subsequently adapted (1981). 2 For a detailed explication please see Appendix A. These variables are adapted from Jochelson et al. 2012.

140

Revue canadienne de criminologie et de justice pénale

janvier 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

References Bhatia, V. J. 1993 Analysing Genre: Language in Professional Settings. England: Longman. Bloss, William 2008 Escalating U.S. police surveillance after 9/11: An examination of causes and effects. Surveillance and Society 4 (3): 208–28. Cohn, Margit and Mordechai Kremnitzer 2005 Judicial activism: A multidimensional model. Canadian Journal of Law and Jurisprudence 18: 333–56. Coughlan, Steve 2011 Good faith, bad faith and the gulf between: A proposal for consistent terminology. Canadian Criminal Law Review (London, England) 15: 197–213. Currie, Robert C. 2011 The evolution of the law of evidence: Plus, ça change : : : ? Canadian Criminal Law Review (London, England) 15: 213–38. Ericson, Richard and Aaron Doyle (eds.) 2003 Risk and Morality. Toronto: University of Toronto Press. Haggerty, Kevin D. 2009 Forward: Surveillance and political problems. In Surveillance: Power, Problems, and Politics, ed. Sean P. Hier and Josh Greenberg. Vancouver: UBC Press. Jochelson, Richard 2009a Multidimensional analysis as a window into activism scholarship: Searching for meaning with sniffer dogs. Canadian Journal of Law and Society 242: 231–49. Jochelson, Richard 2009b Trashcans and constitutional custodians: The liminal spaces of privacy in the wake of Patrick. Saskatchewan Law Review 72(2): 199–222. Jochelson, Richard and Kirsten Kramar 2014 Situating exclusion of evidence analysis in its socio-legal place: A tale of judicial populism. Crime, Law and Social Change 61(5): 541–61. http:// dx.doi.org/10.1007/s10611‑014‑9515‑9

An Empirical Model of Judicial Discourse

141

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Jochelson, Richard, Michael Weinrath, and Melanie Janelle Murchison 2012 Searching and seizing after 9/11: Developing and applying empirical methodology to measure judicial output in the Court’s section 8 jurisprudence. Dalhousie Law Review 35(1) 179–213. Keller, R. 2013 Doing Discourse Research. An Introduction for Social Scientists. London: Sage. Kelly, James B. and Christopher P. Manfredi 2009 Contested Constitutionalism. Vancouver: UBC Press. Kelly, James B. and Michael Murphy 2001 Confronting judicial supremacy: A defence of judicial activism and the Supreme Court of Canada’s legal rights jurisprudence. Canadian Journal of Law and Society 16(1): 3–27. Kendall, Gavin and Gary Wickham 1999 Using Foucault’s Methods. London: Sage. http://dx.doi.org/10.4135/ 9780857020239 Khosla, Madhav 2009 Addressing judicial activism in the Indian Supreme Court: Towards an evolved debate. Hastings International and Comparative Law Review 32: 55–100. Lindquist, Stefanie A. and Frank Cross 2009 Measuring Judicial Activism. Oxford: Oxford University Press. http:// dx.doi.org/10.1093/acprof:oso/9780195370850.001.0001. Madden, Mike 2011 Marshalling the data: An empirical analysis of Canada’s s. case law in the wake of R v Grant. Canadian Criminal Law Review 15(2): 229–51. Morton, Frederick Lee and Rainer Knopff 2000 The Charter Revolution and the Court Party. Peterborough: Broadview. Muttart, Daved 2007 The Empirical Gap in Jurisprudence: A Comprehensive Study of the Supreme Court of Canada. Toronto: University of Toronto Press. Muttart, Daved M. 2011 One step forward, One step back: Measuring activism in the Supreme Court of Canada. SSRN working paper. http://dx.doi.org/10.2139/ ssrn.1470709

142

Canadian Journal of Criminology and Criminal Justice

January 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Ostberg, C. L. and Matthew Wetstein 2007 Attitudinal Decision Making in the Supreme Court of Canada. Vancouver: UBC Press. Romero, Anthony D. 2003 Living in fear: How the U.S. government’s War on Terror impacts American lives. In Lost Liberties: Ashcroft and the Assault on Personal Freedom, ed. Cynthia Brown. New York: New Press. Stewart, Hamish 2011 Section 24(2): Before and after Grant. Canadian Criminal Law Review (London, England) 15: 253–65. Stuart, Don 2010 Welcome flexibility and better criteria from the Supreme Court of Canada for exclusion of evidence obtained in violation of the Canadian Charter of Rights and Freedoms. Southwestern Law School Journal of International Law 16: 313–32.

Legislation cited Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, being Schedule B to the Canada Act 1982 (UK), 1982, c 11. Constitution Act, 1982, being Schedule B to the Canada Act 1982 (UK), 1982, c 11.

Appendix A: Case List Section 24(2) * Cases cited in text that were part of the analysis; ** cases cited in text that were not part of the analysis Total number of cases = 100 Total number of listed cases disqualified = 2: Duhamel (length) and Rothman (Pre-Charter) Total Number of Cases Analysed in the Model = 98 Note: Orbanski and Elias are cited as the same case but were treated as two cases because they pertained to different accused persons in different causes of action and thus reflect different decisions; the two cases from 2001 were counted as pre-9/11.

An Empirical Model of Judicial Discourse

Clarkson v The Queen, [1986] 1 SCR 383. Mooring v Canada (National Parole Board), [1996] 1 SCR 75.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Proulx v Quebec (Attorney General), [2001] 3 SCR 9. R v AM, [2008] 1 SCR 569. R v Arp, [1998] 3 SCR 339. R v Aucoin, 2012 SCC 66. R v Bartle, [1994] 3 SCR 173. R v Beaulieu, [2010] SCC 7. R v Belnavis, [1997] 3 SCR 341. R v Black, [1989] 2 SCR 138. R v Borden, [1994] 3 SCR 145. R v Brown, [2002] 2 SCR 185. R v Broyles, [1991] 3 SCR 595. R v Brydges, [1990] 1 SCR 190. *R v Buhay, [2003] 1 SCR 631. R v Burlingham, [1995] 2 SCR 206. R v Calder, [1996] 1 SCR 660. R v Caslake, [1998] 1 SCR 51. R v Clayton, 2007 SCC 32. R v Cobham, [1994] 3 SCR 360. R v Colarusso, [1994] 1 SCR 20. R v Cole, 2012 SCC 53. *R v Collins, [1987] 1 SCR 265. R v Cook, [1998] 2 SCR 597. R v Cornell, 2010 SCC 31. R v Côté, 2011 SCC 46. R v Debot, [1989] 2 SCR 1140. R v Dersch, [1993] 3 SCR 768. R v Duarte, [1990] 1 SCR 30. R v Duguay, [1989] 1 SCR 93. **R v Duhamel [1984] 2 SCR 555. R v Dyment, [1988] 2 SCR 417.

143

144

Revue canadienne de criminologie et de justice pénale

R v Edwards, [1996] 1 SCR 128. R v Elshaw, [1991] 3 SCR 24.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

R v Evans, [1991] 1 SCR 869. R v Evans, [1993] 3 SCR 653. R v Feeney, [1997] 2 SCR 13. *R v Fliss, [2002] 1 SCR 535. R v Garofoli, [1990] 2 SCR 1421. R v Généreux, [1992] 1 SCR 259. R v Genest, [1989] 1 SCR 59. R v Goldhart, [1996] 2 SCR 463. R v Grant, [1993] 3 SCR 223. *R v Grant, 2009 SCC 32. R v Greffe, [1990] 1 SCR 755. R v Hamill, [1987] 1 SCR 301. R v Hape, [2007] 2 SCR 292. R v Harper, [2004] 1 SCR 827. R v Harrison, 2009 SCC 34. *R v Hebert, [1990] 2 SCR 151. R v Hynes, [2001] 3 SCR 623. R v I (LR) and T(E), [1993] 4 SCR 504. R v Jacoy, [1988] 2 SCR 548. R v Jacques, [1996] 3 SCR 312. R v Jarvis, [2002] 3 SCR 757. R v Kang-Brown, [2008] 1 SCR 456. R v Kokesch, [1990] 3 SCR 3. R v Ladouceur, [1990] 1 SCR 1257. R v Latimer, [1997] 1 SCR 217. *R v Law, 2002 SCC 10. R v Ling, [2002] 3 SCR 814. R v MRM, [1998] 3 SCR 393. R v MacKenzie, 2013 SCC 50. R v Mann, [2004] 3 SCR 59.

janvier 2015

An Empirical Model of Judicial Discourse

R v Manninen, [1987] 1 SCR 1233. R v Matheson, [1994] 3 SCR 328.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

R v McCrimmon, [2010] 2 SCR 402. R v Mellenthin, [1992] 3 SCR 615. R v Monney, [1999] 1 SCR 652. R v Morelli 2010 SCC 8. R v Nolet 2010 SCC 24. *R v Orbanski [2005] 2 SCR 3. *R v Elias, [2005] 2 SCR 3. R v Plant, [1993] 3 SCR 281. R v Pohoretsky, [1987] 1 SCR 945. R v Pozniak, [1994] 3 SCR 310. R v Prosper, [1994] 3 SCR 236. R v RJS, [1995] 1 SCR 451. R v Ross, [1989] 1 SCR 3. **R v Rothman, [1981] 1 SCR 640. R v Schmautz, [1990] 1 SCR 398. R v Silveira, [1995] 2 SCR 297. R v Simmons, [1988] 2 SCR 495. R v Sinclair 2010 SCC 35. R v Singh 2007 SCC 48. R v Smith, [1987] 1 SCR 1045. R v Smith, [1992] 2 SCR 915. *R v Stillman, [1997] 1 SCR 607. R v Strachan, [1988] 2 SCR 980. R v Suberu, 2009 SCC 33. R v Terry, [1996] 2 SCR 207. R v Therens, [1985] 1 SCR 613. R v Thompson, [1990] 2 SCR 1111. R v Tremblay, [1993] 2 SCR 932. R v Wijesinha, [1995] 3 SCR 422. R v Wiley, [1993] 3 SCR 263.

145

146

Canadian Journal of Criminology and Criminal Justice

January 2015

R v Wise, [1992] 1 SCR 527. *R v Wittwer, [2008] 2 SCR 235.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

R v Wong, [1990] 3 SCR 36. Thomson Newspapers Ltd v Canada (Director of Investigation and Research, Restrictive Trade Practices Commission) [1990] 1 SCR 425.

Appendix B: Section 24(2) Variable List Category 1: Activism under the traditional vision Judicial stability. Here, whether the Court was ready to retract from its own or former decisions was measured. When the Court affirmed the decisions of all lower courts, this was scored as being the lowest level of activism. When the Court overturned a previous decision, overturned legislation, and created new law, the case was scored as a 10, or the highest level of activism. Court affirms the decisions of both lower courts = 1 Court affirms the decisions of the lower courts but make any changes to the previous decisions = 2 Court overturns a decision made by lower court but affirms a decision by the appeal court = 3 Court affirms a decision made by the lower court but overturns a decision made by the appeal court = 4 Court overturns both lower court decisions = 5 Court changes a law = 6 Court goes beyond overturning lower court decisions and overturns legislation as unconstitutional = 7 Court implements new tests or increases or changes previously existing tests = 8 Court overturns a previous Supreme Court decision but does not create any new law = 9 Court overturns a previous Supreme Court decision and overturns legislation and creates new law = 10 Interpretation. Does a Court interpret a legal text in possible contradiction with the assumed original intent of the constitution or its plain

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

An Empirical Model of Judicial Discourse

147

linguistic meaning? The decisions that interpret the law for its original meaning and avoid straying from the protections of the Charter will render lower numbers on the activism scale. Courts that use their position and power to interpret legal texts to suit their own intentions, straying from its original meaning, will be seen as highly activist. Supreme Court interprets section 24(2) as the safeguard against illegally obtained evidence and an important Charter protection = 1 Majority interprets section 24(2) as a valued Charter protection, but not one that is without limits as prescribed by law = 2 Concurring justices discuss the high valued Charter protections; however, the majority does not = 3 Majority interprets the Charter as it was intended, but the dissent does not = 4 Majority of the Court interpret the Charter to their own ends but still discuss original intent of Charter = 5 Charter is interpreted by all Court members to their own ends = 6 Majority of the Court interpret the Charter to their own ends but a dissent holds true to the original interpretation = 7 Supreme Court is open to overriding section 24(2) Charter protection, in favour of other interests = 8 Supreme Court interprets section 24(2) as being easily overridden = 9 Supreme Court interprets section 24(2) in a way that contradicts being a high protection from illegally obtained evidence = 10 Majoritarianism and autonomy. Here, whether the Court interfered with policies set by democratic processes and whether the Court was willing to supply its own solution and/or policy was measured Court does not interfere with policies set by democratic processes and leaves all legislation alone = 1 Court discusses the constitutionality of legislation but ultimately upholds it = 2 Court splits on the constitutionality of legislation but ultimately upholds it = 3 Concurring decision would read down (interpret the law narrowly by removing words or meanings) the legislation to amend the law, but the majority upholds the law = 4

148

Revue canadienne de criminologie et de justice pénale

janvier 2015

Dissent of the Court reads down the legislation to amend the law = 5 Majority of the Court reads down the legislation to amend the law = 6

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Dissent would strike down legislation, but the majority would not = 7 Majority strikes down legislation; the dissent would not have done so = 8 Entire Court strikes down legislation = 9 Court strikes down legislation and applies its own solution = 10 Judicial reasoning – Process/Substance. These indicia measured how heavily the Court relied on strict legal and procedural grounds. Where the Court relied entirely on strict legal or procedural grounds in making a decision, the case scored 1. Where the Court relied on open-ended legal tests, such as reasonableness, the case scored 10. Court relies entirely on strict legal and procedural grounds = 1 Court relies entirely on both strict legal and procedural grounds and adopted case law tests = 2 Court mostly relies on legal and procedural grounds but discusses an open-ended test = 3 Court relies more on legal and procedural grounds but also uses an openended test = 4 Court relies equally on legal and procedural grounds and open-ended tests = 5 Dissent uses more open-ended tests than legal and procedural grounds, but the majority does not = 6 Majority uses more open-ended tests than legal and procedural grounds, but the dissent does not = 7 Court uses legal and procedural grounds only to inform its reasons for using open-ended tests = 8 Court uses legal and procedural grounds to create a new open-ended test = 9 Court relies entirely on open-ended tests = 10 Threshold activism. Here, the extent to which the Court was willing to forgive threshold hurdles was measured. In this context, a rigorous application of threshold issue (in this section, whether or not there is a temporal connection between the Charter violation and the remedy as outlined in section 24(2) of the Charter) would score 1. Where the Court

An Empirical Model of Judicial Discourse

149

found reasons to allow for bypassing the temporal connection, where previous cases had not, the case scored 10.

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Court entirely defers to the threshold issue – temporal connection = 1 Court interprets the temporal connection as important, but not always the most important factor = 2 Temporal connection threshold is interpreted as being of substantial importance, but the Court imposes limits on it = 3 Majority of the Court interprets the temporal connection as exceedingly important, however the dissent varies from this view = 4 No discussion of any thresholds = 5 Dissent interprets the temporal connection as incredibly important; however, the majority feels it is not above being overridden = 6 Temporal connection is seen by the Court as being a consideration, but not an important one = 7 Temporal connection is deemed to be unimportant in the case at bar = 8 The temporal connection is almost entirely overlooked in favour of other doctrines = 9 The Court entirely ignores whether or not there is a threshold or temporal connection = 10 Judicial remit: These indicia measured whether the Court’s decision expanded or redefined the jurisdiction of the Court. When the decision did not expand or redefine the jurisdiction of the Court, the case scored 1. A decision that expanded the judiciary’s remit into areas previously immune from intervention was scored 10. Decision does not expand or redefine the jurisdiction of the Court = 1 Dissent would slightly redefine the jurisdiction of the Court = 2 Dissent would slightly expand the jurisdiction of the Court = 3 Decision does not expand the jurisdiction of the Court but slightly redefines it = 4 Decision does not expand the Court’s jurisdiction but somewhat redefines it = 5 Decision does not expand the Court’s jurisdiction but redefines it entirely = 6

150

Canadian Journal of Criminology and Criminal Justice

January 2015

Decision slightly expands the jurisdiction of the Court and slightly redefines it = 7

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Decision slightly expands jurisdiction and entirely redefines it = 8 Majority entirely expand and redefine the jurisdiction, but the dissent would not = 9 Decision entirely redefines and expands the jurisdiction of the Court = 10 Rhetoric. These indicia measured whether judicial decisions were used as platforms for expression of broader positions and values or whether the use of rhetoric was restricted to the explication of legal principles. The absence of legal rhetoric (usually correlating with shorter decisions) scored 1. High levels of extra-legal rhetoric, combined with long discussions of political implications, were scored 10. No rhetoric by the Courts and a very short decision = 1 Minimal rhetoric, still a fairly short decision = 2 Some expression of broad positions and values, but still a relatively short decision = 3 Substantial expression of values and positions = 4 The Court engages in brief discussion of non-legal principles, as well as potential implications = 5 The decision begins to become more about positions than law = 6 A lengthy decision; substantial discussion on non-legal issues = 7 A long decision; broad positions as well as political ideology discussed = 8 A very long decision, with substantial expression of non-legal principles and political values = 9 Extremely long decision and copious amounts of rhetoric used = 10 Obiter dicta. These indicia ask how far the Court expands its opinion beyond the legal requirements of the specific case. When the Court did not delineate any obiter, the case scored 1. When the Court used extensive obiter and discussed issues not relevant to case, a 10 was recorded. Court does not discuss its opinion at all; no obiter = 1 Court does not expand its opinion beyond the case at bar = 2

An Empirical Model of Judicial Discourse

151

Court briefly discusses potential ramifications of decision but stays neutral = 3

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Court discusses potential ramifications of decision at length but stays neutral = 4 Court briefly discusses ramifications and appears to be expressing an opinion = 5 Opinion clearly being expressed; ramifications are expressed = 6 Ramifications of decision are discussed in detail; opinions extend beyond the case at bar = 7 Court uses substantial amount of obiter = 8 Court uses extensive obiter and discusses issues not relevant to the case at bar = 9 Court uses extensive obiter and discusses issues not relevant to case at bar = 10 Reliance on comparative sources. Here, how extensively the Court relied on foreign sources that are not legally binding in the domestic sphere was examined. Where the Court used domestic law exclusively, a score of 1 was given. When the Court used comparative sources to create new legal conceptions, with extensive comparative discussion, a score of 10 was given. Court uses domestic law exclusively = 1 Court makes minimal references to England’s common law or texts written by non-judges to understand law, not to change it = 2 Use of American law to understand law but not to change it, with minimal reference = 3 Use of comparative law to justify existing domestic laws, still with fairly minimal reference = 4 Use of comparative law justifying existing domestic laws, with extensive reference = 5 Use of comparative law to justify changing existing laws, with minimal reference = 6 Use of comparative law to justify changing existing laws, with extensive reference = 7 Use of comparative law to create new legal tests = 8 Use of comparative law to create new laws, with minimal reference = 9

152

Revue canadienne de criminologie et de justice pénale

janvier 2015

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Use of comparative law to create new laws, with extensive reference = 10 Judicial voices. Here, the extent of other judicial decisions besides the majority decision was examined. A unanimous decision scored 1. On occasions where there were two concurring and two dissenting judgments (the most judicial voices seen), the case scored 10. A unanimous decision = 1 One concurring decision with no additions to the overall law = 2 One concurring decision adding other law = 3 Two concurring decisions = 4 Two concurring decisions adding other law = 5 One dissent or three concurring decisions = 6 One concurring decision and one dissenting opinion = 7 One concurring decision adding other law and one dissent = 8 Two concurring judgments and one dissenting judgment or two dissenting judgments = 9 Two concurring judgments and two dissenting opinions = 10 Extent of decision. Here, the question of whether the Court’s ruling expressly applied to a single or specified set of circumstances or whether the law that resulted had broad implications for larger sections of society was examined. If the Court simply applied the legal rules, the case scored 1. Where the Court created a new standard that affected broader populations, the case scored 10. Court simply applies the rules set forth in the criminal code = 1 Court narrows its application to the case at bar = 2 Application to case at bar; discussion on potential future standing = 3 Application to one or two other cases = 4 Application to a few cases as well as discussion on future standing = 5 Creation of a new standard which applies only to case at bar = 6 Creation of a new standard which applies to a few cases = 7 Creation of a new standard with a broad scope = 8 Creation of new powers then used in later cases = 9

An Empirical Model of Judicial Discourse

153

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

Creation of a new standard that is then adopted into the common law = 10 Legal background. Here, the question of whether the legal framework on the basis of which the Court made its decision were inclusive and clear or whether the rules concerned were vague, complex, self-contradictory or incomplete was examined. The absence of a clear rule essentially requires the judiciary to extend its decision beyond earlier case law. Once the Court does not or cannot abstain from deciding, decision making in later cases will almost necessarily involve some creative judicial expertise and will thus be activist. Legal framework is clear; rules are clear; decision does not extend beyond prior case law = 1 Legal framework is clear; rules are clear; decision slightly extends beyond prior case law = 2 Legal framework is clear; rules are clear; decision extends well beyond prior case law = 3 Legal framework is clear; rules are not clear; but decision does not extend beyond prior case law = 4 Legal framework is clear; rules are not clear; decision extends beyond prior case law = 5 No clear framework or rules, but decision does not extend beyond case at bar = 6 No clear framework or rules; Court adopts rules from other areas of law = 7 No clear framework or rules; Court makes new rules = 8 No clear framework or rules; Court makes new rules that override old rules = 9 No clear framework or rules; Court makes new framework for analysis = 10

Category 3: Activism and the protection of core values Variables: Core Values Activism Intervention and value content. Here, the question of whether the subject matter under examination was highly value laden in its bearing

http://www.utpjournals.press/doi/pdf/10.3138/CJCCJ.2914.E08 - Friday, June 30, 2017 8:26:17 AM - IP Address:50.71.76.72

154

Canadian Journal of Criminology and Criminal Justice

January 2015

on democratic principles and human liberties accepted domestically was examined. Where the case dealt with important human rights issues and the Court appeared to assert its guardianship of the constitution the case scored 1. Where the Court declined to discuss the constitutional values at stake, the case scored 10. Unanimously high value context; Charter protections paramount = 1 Majority discusses high value context; Charter protections paramount = 2 Majority discusses high value context; Charter extremely important dissent does not = 3 In a concurring statement, a justice discusses high value context = 4 No value context discussion, but Charter is interpreted purposively = 5 Majority discusses value context, but determines that it is not above being overridden = 6 Dissent discusses high value context, but majority does not = 7 Little mention of values, Charter discussion easily overridden = 8 No discussion of value context, brief discussion of liberties, mostly technical decision = 9 No one discusses value context at all; technical decision, not about liberties = 10