Law Commission of Canada, ed., Law and Risk (Vancouver: UBC Press, 2005).
Pp. viii, 208.
Risk permeates virtually all human activity, including law. Whether in devising
sentencing guidelines, influencing legal reasoning, or assigning the burden of proof,
concerns about an individual, a group, or a societys exposure to risk are often at the
fore. Demystifying the meaning of risk and how we measure it is a complicated and
layered
task. It requires not only a disaggregation of numerous analytic
methodologies, but also an intermingling of the worlds of science, mathematics, and
social values.
The six essays comprising the book Law and Risk, edited by the Law
Commission of Canada, take specific aim at what is meant by risk, how we
understand it, and how it is applied in a legal context. The common tenor of the
essays is that risk analysis is a function more of social values than of the cold logic of
precise calculations. The different authors engage this thesis from varied and
provocative vantage points. Some of the essays analyze specific legislation (see
Mariana Valverde, Ron Levi & Dawn Moore, Legal Knowledge of Risk 86, which
discusses Megans Lawa term used in the United States for statutes that mandate
community advisement upon the release of a sex offender who has a moderate to high
risk of recidivism). Some deconstruct the reasoning in Supreme Court cases (see
Danielle Pinard, Evidentiary Principles with Respect to Judicial Review of
Constitutionality: A Risk Management Perspective 121, which discusses Gosselin v.
Quebec (A.G.)1 and R. v. Sharpe,2) while others look at the role of values in creating
risk assessments (see e.g. Duff R. Waring & Trudo Lemmens, Integrating Values in
Risk Analysis of Biomedical Research: The Case for Regulatory Law Reform 156).
The book, in focusing on the nexus between social values and risk analysis,
nearly completely neglects to consider risk from a quantitative basis. At one point,
when mathematical calculations are mentioned, it is done for the purpose of
undermining the very reliability of scientific data.3 The authors of the opening essay
anecdotally discuss the likelihood of the guilt of a person selected from a population
of 500 000 and whose DNA matches that found at a crime scene. While the error rate
for the DNA analysis is one in a million, the authors explain that the probability that
this person is the culprit is surprisingly only sixty-seven per centfar below the
standard of beyond a reasonable doubt (William Leiss & Steve E. Hrudey, On Proof
and Probability: Introduction to Law and Risk 1 at 8).4 The example serves as
1 [2002] 4 S.C.R. 429, 221 D.L.R. (4th) 257, 2002 SCC 84 [Gosselin cited to S.C.R.].
2 [2001] 1 S.C.R. 45, 194 D.L.R. (4th) 1, 2001 SCC 2.
3 While there are parts of the book that are numerate, such analysis does not form a significant part
4 Although the result may seem counterintuitive, it is easily arrived at by a Bayesian formulation of
of the discussion.
conditional probabilities.
[Vol. 51
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
614
useful rhetoric in fostering wariness towards scientific absolutism (in this case, the
persuasiveness of DNA test results).
The discussion and analysis in the book continually draws the readers attention
back to the value-laden underpinnings of the risk discourse. For instance, the authors
of Legal Knowledges of Risk (86), in discussing Megans Law, point to the
significant discretion held by judges in weighing and balancing the private and public
concerns inherent in risk assessment. It is described as a process of translationthe
alchemical change that risk and its evaluation undergo when taken from the scientific
to the legal field. They argue that, in applying Megans Law, the courts have broad
discretion to control or even reject expert testimony relating to actuarial or clinical
data in favour of the knowledge gained by the prosecutor through his or her work
experience (103-104).
Dayna Nadine Scott, in her essay about the precautionary principle, discusses the
political and unscientific nature of choosing amongst statistical tests and standards
that are relied upon to verify the existence of a given harm (i.e., environmental
degradation) (Shifting the Burden of Proof: The Precautionary Principle and Its
Potential for the Democratization of Risk 50 at 63). Scott uses as an example the
discussion of genetically engineered foods; the decision as to whether the statistical
analysis should be organized around ruling out the possibility of false positives or
ruling out false negatives is, as she says, a value-based judgment (62-63).5 Do we
want to protect the interests of businesses or avoid all possible consumer exposure to
genetically engineered foods? The answer to this question will influence, if not
determine, how tests for such foods are carried out.
In the final essay of the book, entitled Integrating Values in Risk Analysis of
Biomedical Research: The Case for Regulatory Law Reform (156), Duff R. Waring
and Trudo Lemmens candidly discuss the influential role of social values in the
approval of medical trials by research ethics boards. The authors point out that risk
assessments may have more to do with the values that inform the definitions of
harm and benefit than they do with the quantitative methods being put forward
(160).
The hegemony of social values in risk analysis is used throughout the book to
underwrite various critiques of the adjudicative process. One such critique is how
conclusions drawn from risk analysis steward the administration of justice. David
MacAlisters essay (Use of Risk Assessments by Canadian Judges in the
Determination of Dangerous and Long-Term Offender Status, 1997-2002 20), which
compares the dangerous offender with the long-term offender designations in
Canadian criminal law, is a good example. MacAlister identifies how, through the use
of various risk measures, courts create the impression of bestowing well-supported,
actuarial justice. The essay explains that one of the most popular scientific tests for
5 A false positive would show genetic modification where there is none; a false negative would not
show genetic modification in every instance where it has occurred.
615
BOOK NOTE / RECENSION
2006]
these designations is the Hare Psychopathy Checklist. A test taker who scores at least
thirty on this test will be deemed to be a prototypical psychopath (28). The test result
will then be used to assist in determining recidivism and, ultimately, will aid in
determining the test takers appropriate offender designation (i.e., long-term or
dangerous).
MacAlister, in evaluating the utility of these actuarial results, is able to highlight
the inherent problem of not being able to refute probability results (37). If someone is
ruled to have only a five per cent chance of reoffending if released from jail, and so is
released and then actually does reoffend, the analysis is not wrong per se. It did, after
all, foresee a five per cent possibility. A court is in an obvious bind, because in
deciding whether the person would be classified as a dangerous offender or a long-
term offender, it does not know if it is dealing with someone who is in the five per
cent or the ninety-five per cent range. MacAlister additionally warns of the danger of
using results from group behaviour to make decisions about individuals (38).
In the essay Legal Knowledges of Risk (86), the authors argue, among other
things, that risk analyses are used to revive needs-based solutions to drug-related
crime. The authors explain that characterizing drug crimes within a risk to society
construct has led to the development of special drug courts in Toronto, Vancouver,
and elsewhere. These special courts operate in a unique fashion. They are designed to
enable a dialogue between the drug offender and the judge, whereby a discussion of
the formers drug violations are undertaken with suggestions by the respective
lawyers of appropriate sentencing (e.g., community service, counselling, etc.). These
courts also highlight how framing analysis in a risk context leads different actors to
draw from different kinds of expertise. For instance, judges using pharmacological
knowledge to try to convince drug offenders of the likelihood of further impeding
their health, or lawyers using psychologically based argumentation to convince a
judge of the benefit of giving a particular drug offender a second chance (111, 114).
This fluidity of knowledge is termed knowledge swapping and is described as a
dynamic and interactive development (116).
Another critique that emerges from these essays is the bias underlying the
assignment of the burden of proof. As Danielle Pinard writes in her essay
Evidentiary Principles with Respect to Judicial Review of Constitutionality: A Risk
Management Perspective (121), the party with the burden of proof will bear the cost
of scientific uncertainty. Who bears the evidentiary burden is a normative decision
and a key determination with respect to risk management (127). Pinard is writing
about claims based on the Canadian Charter of Rights and Freedoms,6 where the
burden of proving a rights infringement rests with the claimant, and the burden of
proving the reasonableness of such infringement rests with the Crown.
6 Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c. 11.
[Vol. 51
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
616
Pinard dissects the reasoning in the 2001 Supreme Court of Canada decision of
Gosselin,7 demonstrating how the uncertainty of the evidence and of the risks being
claimed worked in favour of the Crown. Uncertainty ultimately contributed to the
Crowns victory. The decision of who bears the burden of proof is not unlike Scotts
discussion of how the choice of statistical methods influences the outcome.
The numerous authorsten in allwho contributed to this book, discuss risk by
conveying their ideas through advocacy, research, and rhetoric-based writing. The
book nicely catalogues various instantiations of risk analysis, its relative usefulness,
and its underlying arbitrariness. Though some of the writing verges on the polemical,
and perhaps in some cases overstates the politicization of scientific testing, the cross-
section and currency of legal topics covered provides a compelling case for the
importance of risk in even the most generalized consideration of law.
Hugh Sandler
7 Supra note 1.
