by C. L. Ostberg and Matthew E. Wetstein. Vancouver: UBC Press, 2007. 288pp. Cloth. CDN$85.00/ US$93.95. ISBN: 9780774813112. Paper. CDN$32.95/US$36.95. ISBN: 9780774813129.
Reviewed by Roy B. Flemming, Department of Political Science, Texas A&M University. Email: roy [at] politics.tamu.edu.
The attitudinal model in the study of judicial decision making draws its influence from its parsimony and explanatory power. Judges single-mindedly pursue their personal policy preferences which can be located on a single ideological dimension along with the issue raised in a legal dispute. The interaction of preferences and issue determines how judges ultimately vote (Segal and Spaeth 2002). It is simple, though some complain it is too simple. But it works, most of the time, and perhaps in many places, though that remains an open question. One reason this remains an open question is that the institutional features of courts vary considerably.
Next year will be the fortieth anniversary of the publication of a book, edited by Glendon Schubert and David J. Danelski, devoted exclusively to studies of comparative judicial behavior. This path breaking book included two chapters on the Supreme Court of Canada that found the justices’ voting patterns during the 1950s and 1960s were structured along various policy dimensions (Fouts 1969; Peck 1969). This led Fouts (at 284) to conclude “the Supreme Court of Canada resembles its U.S. counterpart in significant respects.” Interest in the behavior of the justices unfortunately waned after this burst of enthusiasm until various internal and external changes occurred that expanded the Supreme Court’s authority, particularly the adoption of the Charter of Rights and Freedoms in 1982. The Charter piqued scholarly curiosity about how the justices behaved in this new legal regime and raised expectations that the resemblances between Canadian and American judges would become stronger.
For several years now, C. L. Ostberg and Matthew E. Wetstein have probed the limits of the attitudinal model in Canada’s Supreme Court; in their estimation “one of the easiest test cases” around to see if this warhorse of American judicial behavioralism can go the distance in another venue. Ostberg’s dissertation compared the American and Canadian high courts following Canada’s adoption of its Charter of Rights and Freedoms (Ostberg 1995). In subsequent work, among other things, Ostberg and Wetstein extended Segal’s (1984) search and seizure model to Canada and assayed how case facts and attitudes influenced judicial votes regarding this issue (Wetstein and Ostberg 1999; Ostberg and Wetstein 1998). A recent article more broadly explores attitudinal decision making on Canada’s high court during the mid-1990s (Ostberg, Wetstein and Ducat 2002). [*501]
Ostberg and Wetstein’s new book builds on their prior research but expands on it in ambitious ways, making it a stand-alone and worthy piece of work. Their study surveys the voting decisions of twenty-three justices who sat on the bench from 1984 to 2003. The authors’ central concern is determining if the justices’ policy preferences significantly affected whether they voted liberally in three different legal domains (criminal, civil rights and civil liberties, and economic) with cases in two specific issues drawn from each issue domain – for example, cases involving unions and taxes in the economic domain – the foundations for the analyses .
Measuring judicial preferences, of course, is an immediate and critical concern. Taking a cue from Segal and Cover (1989), Ostberg and Wetstein conducted content analyses of commentaries in nine Canadian newspapers regarding the justices at the time they were nominated. Their measure is operationally different, however, from the original one. Their coding procedure is more broadly framed to avoid the civil liberties/civil rights bias of the Segal/Cover scores, and they varied the weights that make up the scores to take into account the particular features of Canada’s newspaper industry. Ostberg and Wetstein are sensitive to this measure’s underlying assumption that judicial preferences are stable over time; tests of this assumption persuaded them that sufficient stability existed that the measure was appropriate. In addition to these scores, the authors employ indices of personal attributes of the justices and the party of the prime minister who appointed them as other ideological indicators. Chapter 3 fully and carefully covers the construction of these variables, assesses their external validity, and how they match up with measures used in previous research.
A common template structures Chapters 4-6, the book’s central chapters that correspond to the three legal domains. First, the domain’s docket is described. The domain’s annual share of Canada’s docket over the twenty year period is presented along with how often the domain’s cases are decided non-unanimously and the average size of the corams, or panels, that heard the cases; Canada’s court does not always sit en banc but often in panels of five or seven justices. (The chief justice decides both the size of panels and which justices will sit on them if the court does not sit en banc.) The average size of panels, typically around seven, varied across the three legal domains and over time. Next, the voting patterns of the individual justices are described: how often they dissented, how often they wrote majority, concurring, or dissenting opinions (to flag those justices who were leaders or followers on the bench in a particular domain), how often they voted liberally and to overturn a law as an indictor of “activism,” and whether these voting patterns were stable over time. The chief concern of these descriptions deals with what appears to be a strong “consensus norm” on the Canadian court and the leadership of the chief justice.
The final piece tests the efficacy of the attitudinal model in the specific issue areas of the particular domain. In addition to the ideology measure and the party of the prime minister or other personal attributes of the justices, the logistic models include the facts related to the cases, what litigants were [*502] involved, and who the chief justice was at the time of the votes. (There were three chief justices during the twenty year period.) Ostberg and Wetstein weave into their statistical analyses illustrations from pertinent court decisions to help explain the results of the models. The book, nonetheless, is driven by its findings both expected and unexpected (of which there are many). Broader theoretical considerations tend to be left in the back seat as ad hoc explanations for anomalous results steer the discussion.
This becomes especially evident with Ostberg and Wetstein’s strategy of testing their models with data for both unanimous and non-unanimous decisions and then running the models separately for only the non-unanimous cases. In many instances, this switch dramatically affects the results; statistical significance rises or falls; signs for variables flip from what they were before; variables get dropped because the smaller number of non-unanimous cases creates multicollinearity problems. As a consequence, it is often difficult to interpret the competing results. For example, the ideology measure in search and seizure and in right to counsel cases jumps in importance when only non-unanimous decisions are used as compared to the “all cases” results. But would this variable be influential if only unanimously decided cases were used? This strategy also makes it difficult to know whether case facts are more strongly related to voting decisions in unanimous cases than in non-unanimous cases and thus evidence for the influence of the “legal model” when the justices agree on cases.
Setting aside these concerns, does the attitudinal model work in Canada? Chapters 6-7 wrap up the book’s findings and place them in perspective. Ostberg and Wetstein ultimately conclude “the impact of ideology is not as crystal-clear or as systematic as that found in the US context” (p.226). Their biggest surprise, given the success of the Segal/Cover scores in the US, is that the newspaper-based measure of judicial ideology flunked the test in cases involving equality and free speech issues drawn from the civil rights and liberties domain. The same thing occurred for the two issue areas in the economic domain. Only in the criminal law domain did the measure meet expectations. Indicators of the justices’ personal attributes as alternative indicators of ideology did much better in civil rights and liberties issues; female justices were significantly more likely than their male counterparts to vote liberally, and the party of the appointing prime minister had a marginal effect. In the economic domain, justices who spent most of their careers in private practice were more likely to vote against unions, but no other personal attribute variable came into play in tax cases. (For a recent reconsideration of the personal attribute model in Canada, see Songer and Johnson 2007.) While these are suggestions that policy views matter some of the time in Canada, it nonetheless means that case characteristics, the type of litigant, and who the chief justice was at the time of the cases did much of the empirical heavy lifting in the statistical models.
It is a mixed picture to be sure and one that Ostberg and Wetstein anticipate in Chapter 2 when they consider multiple political, institutional, and [*503] Charter-related reasons for why Canada’s justices might be variously encouraged or discouraged to follow their ideological leanings. Most of these factors, however, remain on the shelf until Ostberg and Wetstein need them to explain findings; they are generally not incorporated or not capable of being incorporated into the statistical models. For instance, the chief justice’s discretion to set the size and composition of panels, an institutional factor, and how it relates to consensus on the bench frequently comes up in the book’s discussions, but it is not systematically integrated into the analysis. Nor is it immediately obvious how the chief justices’ decisions are linked to whether the justices on the panels vote liberally or not.
So, what’s next? Perhaps a new, improved measure of Canadian judicial ideologies? Alarie and Green (2007) adapted Martin and Quinn’s (2002) approach to generate estimates of the ideal points of Canadian justices and compared them to those for the US Supreme Court. The ideal points for the Canadian justices were less dispersed and more clustered than those for the American justices, which is consistent with a stronger norm of consensus on Canada’s court. However, further analysis suggested that the justices’ votes “are not driven by an underlying attitude that is distributed unidimensionally” (at 211), which may be why Ostberg and Wetstein’s newspaper-based ideology measure proved disappointing and which may also limit the utility of the Martin and Quinn approach in Canada.
Maybe the way forward is to acknowledge the intricacies of judicial behavior in Canada while continuing direct comparisons with the US. Ostberg and Wetstein, joined by Songer and Johnson, in a forthcoming article take exactly this approach to question whether a unidimensional voting model makes sense in Canada’s Supreme Court compared to the US Supreme Court. They conclude it does not, after factor analyses of non-unanimous decisions in the two courts reveal not only that the primary dimensions underlying voting on the two courts differ but also that Canadian jurists are less consistent and more unpredictable than American justices. Another way may be to take a long term view of Canada’s Supreme Court with the hope of capturing the effects of the institutional changes it has undergone that simple applications of the attitudinal model cannot reveal. Songer’s forthcoming book, THE TRANSFORMATION OF THE SUPREME COURT OF CANADA: AN EMPIRICAL EXAMINATION holds out the promise that it will begin to unravel the complexity of the Canadian case. In the meantime, it seems that Fouts may have misjudged how closely Canada’s Supreme Court would come to resembling its American cousin.
Alarie, Benjamin R.D., and Andrew Green. 2007. “The Reasonable Justice: An Empirical Analysis of Frank Iacobucci’s Career on the Supreme Court of Canada.” 57 UNIVERSITY OF TORONTO LAW JOURNAL 195-226.
Fouts, Donald E. 1969. “Policy-Making in the Supreme Court of Canada, 1950-1960.” In COMPARATIVE JUDICIAL BEHAVIOR: CROSS-CULTURAL STUDIES OF POLITICAL DECISION-MAKING IN THE EAST AND WEST, eds. Glendon Schubert and David J. Danelski. New York: Oxford University Press. [*504]
Ostberg, C.L. 1995. “A Comparison of US and Canadian Supreme Court Decisions after the Addition of the Charter of Rights and Freedoms to the Canadian Constitution.” Ph.D. dissertation, Northern Illinois University, DeKalb, Illinois
Ostberg, C.L., and Matthew E. Wetstein. 1998. “Dimensions of Attitudes Underlying Search and Seizure Decisions of the Supreme Court of Canada.” 31 CANADIAN JOURNAL OF POLITICAL SCIENCE 767-87.
Ostberg, C. L.,and Matthew E. Wetstein. 2002. “Attitudinal Dimensions of Supreme Court Decision Making in Canada: The Lamer Court, 1991-1995.” 55 POLITICAL RESEARCH QUARTERLY 237-58.
Ostberg, C.L., Matthew E. Wetstein, Donald R. Songer, and Susan W. Johnson. Forthcoming. “Ideological Consistency and Attitudinal Conflict: A Comparative Analysis of the U.S. and Canadian Supreme Courts.” COMPARATIVE POLITICAL STUDIES.
Peck, Sidney R. 1969. “A Scalogram Analysis of the Supreme Court of Canada, 1958-1967.” In COMPARATIVE JUDICIAL BEHAVIOR: CROSS-CULTURAL STUDIES OF POLITICAL DECISION-MAKING IN THE EAST AND WEST, eds. Glendon Schubert and David J. Danelski. New York: Oxford University Press.
Segal, Jeffrey A. 1984. “Predicting Supreme Court Cases Probabilistically: The Search and Seizure Cases, 1962-1981.” 78 AMERICAN POLITICAL SCIENCE REVIEW 891-900.
Segal, Jeffrey A., and Albert D. Cover. 1989. “Ideological Values and Votes of US Supreme Court Justices.” 83 AMERICAN POLITICAL SCIENCE REVIEW 557-65.
Segal, Jeffrey A., and Harold J. Spaeth. 2002. THE SUPREME COURT AND THE ATTITUDINAL MODEL REVISITED. Cambridge: Cambridge University Press.
Songer, Donald R. Forthcoming. THE TRANSFORMATION OF THE SUPREME COURT OF CANADA: AN EMPIRICAL EXAMINATION. Toronto: University of Toronto Press.
Songer, Donald R., and Susan W. Johnson. 2007. “Judicial Decision Making in the Supreme Court of Canada: Updating the Personal Attribute Model.” 40 CANADIAN JOURNAL OF POLITICAL SCIENCE 911-34.
Wetstein, Matthew E., and C.L. Ostberg. 1999. “Search and Seizure Cases in the Supreme Court of Canada: Extending the American Model of Judicial Decision Making across Countries.” 80 SOCIAL SCIENCE QUARTERLY 757-74.
© Copyright 2008 by the author, Roy B. Flemming.