The Rise of Facts in Public Law II: Factual Assessments in Judicial Review of Legislation

In a draft book chapter I am working on with co-author Kseniya Kudischeva, we discuss the increased importance of factual assessments in public law. This is the second of four substantive parts (see Part I here). Comments and thoughts welcome.

In federal systems, factual assessments can never be entirely off limits in judicial review of legislation. Where the powers of legislative bodies in different levels of government are limited to particular fields, a factual assessment will often be necessary to ensure that the level of government has remained in one of its fields. As has been observed on the High Court of Australia, “it is the duty of the Court in every constitutional case to be satisfied of every fact the existence of which is necessary in law to provide a constitutional basis for the legislation”.[1]

A straightforward example is provided by the decision of the Supreme Court of Canada about the validity of emergency legislation to combat the inflation crisis of the 1970s: extensive evidence was filed to demonstrate (or, from the opponents’ perspective, undermine) the factual predicates for federal legislation under Parliament’s authority to legislate on matters of national emergency.[2]

There, as in Canadian federalism generally, the effects of a law are relevant to determining its constitutionality, which permits the courts to make factual assessments.[3] This type of factual assessment, however, does not typically present great conceptual or practical difficulties. Much as reference to government policy papers and Hansard has become commonplace in discerning the meaning of statutory provisions, the availability of background material about the passage of legislation and its intended and actual effects can assist judges in determining whether a given law falls within one of the law-maker’s constitutionally accorded fields of responsibility. The judge does not have to evaluate the law-maker’s factual assessment of the situation, testing it for rationality, reasonableness or proportionality but at most only to determine whether the factual assessment was accurate. Put another way, federalism questions are black and white, in the sense that a particular matter either will, or will not be, within the constitutional competence of a given level of government, and factual material is helpful in resolving these binary questions.

Judicial review of legislation for compliance with fundamental rights raises very different issues, at least as this practice has developed in common-law jurisdictions. The proportionality test typically used to determine whether a legislative provision interferes unduly with a protected right requires a court to ask about importance, rationality, necessity and fair balance: (1) whether the legislature pursued an important objective; (2) whether there is a rational connection between the means chosen by the legislation and the achievement of the objective; (3) whether the means chosen were necessary to achieve the objective; and (4) whether the detrimental effects on individual interests caused by the means used outweighed the public interest in achieving the objective.

I acknowledge that it is “hazardous to talk of proportionality analysis simpliciter, rather than the commitments of particular proportionality theorists, or the doctrines and practices of particular courts”.[4] Nonetheless, it is tolerably clear that the factual assessments required by the various prongs of the proportionality test – especially prongs (3) and (4) – are not comparable to the factual assessments required in federalism cases. The questions are not binary, or black and white, but invariably involve complex value judgements about “importance”, “rationality”, “necessity” and “balance”. They are by definition evaluative. Consider the following passage from the work of Professor David Beatty, the leading exponent of the view that judicial review of legislation for proportionality is an exercise in technical reasoning rather than moral reasoning:

Turning conflicts about people’s most important interests and ideas into matters of fact, rather than matters of interpretation or matters of moral principle, allows the judiciary to supervise a discourse in which each person’s perception of a state’s course of action is valued equally and for which there is a correct resolution that can be verified empirically.[5]

The premise of Professor Beatty’s view is that proportionality involves consideration of “matters of fact”, where there is in principle a “correct resolution” capable of empirical verification. Even here, then, proportionality review requires courts to conduct detailed factual assessments.

The requirement to engage in such factual assessments raises principled and practical difficulties. On the practical side, “it will sometimes be difficult, if not impossible, for the state to provide reliable and direct evidence of the benefit its measures will achieve”,[6] a difficulty which arises especially at prongs (3) and (4). Particularly in areas of social policy where legislation is designed to protect vulnerable groups from inchoate harms – varying from the psychological dangers of exposure to harmful material[7] to pressure felt to avail of assisted suicide[8] – hard evidence which “can be verified empirically” may be hard to come by. Even if extensive social science evidence is put before the court, and tested by way of cross-examination of expert witnesses, it may be difficult to determine that the legislation must founder on the necessity or fair balance prongs of the proportionality test. Notice that the disadvantages here are not all on the state side: individuals or organizations challenging the constitutionality of legislation must also expend significant resources in order to mount plausible challenges in the first place; and it is reasonable to ask as a general proposition (and sometimes urgent in respect of specific challenges) whether this is the best use of the scarce resources of the individuals or organizations concerned.[9]

On the principled side, the factual assessments judges must conduct in proportionality review require them to consider the quality of the legislature’s work product and, by extension, the quality of a legislature’s deliberation. Bar-Siman-Tov Ittai has argued that there is an emergent species of “Semiprocedural Judicial Review”, based on a series of cases in which “the quality of the legislature’s decision-making process seemed to influence the court’s decision about the constitutionality of the law” but “the quality of that process was not the sole consideration determining the validity of legislation”.[10] This “rationality check of the policy process” might be attributable to “the increasing evidence based nature of the policy process”[11] but there can be little doubt that the factual assessments required by proportionality review must involve at least indirect scrutiny of the legislative process.

There have been several responses to these difficulties: a preference for testing social science evidence by the tendering of expert evidence which can be subjected to cross-examination; marking off areas of legislative activity from judicial oversight; and the development of mechanisms for judicial restraint, modulating the application of prongs (3) and (4) in cases involving legislative value judgements.

Some facts relevant to determining the proportionality of legislation will be in the public domain or uncontroversial and, as such, judges may take judicial notice of them.[12] But the social science-type evidence which often underpins legislative interventions (or legislative abstentions) is not generally suitable for judicial notice, as the interpretation of the evidence is often contestable and may be subject to multiple, equally reasonable interpretations. Accordingly, the Supreme Court of Canada has cautioned against the use of “Brandeis briefs” in constitutional litigation, expressing a preference for the tendering instead of expert evidence which can be subject to cross-examination.[13] The hope here is that controversial ‘legislative’ facts can become settled ‘adjudicative’ facts by the time the court reaches the stage of conducting a proportionality test.[14]

In addition, courts have made clear that there are some areas of legislative activity which are non-justiciable. Direct judicial consideration of the legislative process is, in the common law tradition, strictly forbidden as it would cause judges to trespass into areas protected by parliamentary privilege:[15] “assessing the quality of [legislative] debate and examining the extent to which members participated in it…are not matters which are apt for judicial supervision”.[16]  Parliamentary privilege, traceable at least to the Bill of Rights 1688, means that judges may not opine on or grant remedies in respect of areas of legislative activity which are central to the work of the legislature.[17] Semiprocedural review is as far as the judges may go.

Most importantly, courts have developed a variety of mechanisms of restraint. On the one hand, it is clear that where legislation can be characterized as addressing complex social, cultural, economic or other issues on which reasonable people hold a range of differing views judges will shy away from the conclusion that the legislation unjustifiably infringes fundamental rights. Whether the term “deference”, “margin of appreciation” or some other equivalent is used, the message is the same: on sensitive matters, legislatures will get more leeway on judicial review; put another way, judges will be reluctant to second-guess legislatures’ factual assessments where legislators are better-placed than courts to make such assessments.[18] On the other hand, when judicially reviewing legislation, courts focus less on whether a legislature’s factual assessment was right or wrong than they do on whether the government defending the legislation can establish, on a reasoned basis and by reference to evidence in the record a sufficiently convincing justification for an interference with a fundamental right. There is no need, in this mode of analysis, to directly impugn a legislature’s factual assessment or indeed its internal processes.[19] Once more, however, it is clear that contemporary courts find themselves of necessity making factual assessments and having to develop responses to the resulting difficulties.


[1] Australian Communist Party v Commonwealth (1951) 83 CLR 1, 222, per Williams J.

[2] Re: Anti-Inflation Act [1976] 2 SCR 373, 386-391.

[3] Reference re Firearms Act (Can.) 2000 SCC 31, [2000] 1 SCR 783, [16]-[18].

[4] Bradley Miller, “Proportionality’s Blind Spot: “Neutrality” and Political Philosophy” in Grant Huscroft, Bradley W Miller and Grégoire N Webber eds., Proportionality and the Rule of Law: Rights, Justification, Reasoning (Cambridge University Press, Cambridge, 2014), p. 370 at pp. 371-372.

[5] David Beatty, The Ultimate Rule of Law (Oxford University Press, Oxford, 2004), at p. 171.

[6] R. v K.R.J., 2016 SCC 31, [2016] 1 SCR 906, at para 144, per Brown J (dissenting).

[7] See e.g. Irwin Toy Ltd. v Quebec (Attorney General) [1989] 1 SCR 927.

[8] See e.g. R (Nicklinson) v Ministry of Justice [2014] UKSC 38, [2015] AC 657, [81], [201], [228].

[9] The classic text in this regard remains Gerald N Rosenberg, The Hollow Hope: Can Courts Bring About Social Change? 2nd ed.(University of Chicago Press, Chicago, 2008).

[10] (2012) 6 Legisprudence 271.

[11] Alberto Alemanno, “The Emergence of the Evidence-Based Judicial Reflex: A Response to Bar-Siman-Tov’s Semiprocedural Review” (2013) 1 Theory and Practice of Legislation 327.

[12] Edmund M Morgan, “Judicial Notice” (1944) 57 Harvard Law Review 269.

[13] R. v Spence 2005 SCC 71, [2005] 3 SCR 458, [68]. See also M. v H. [1999] 2 SCR 3, [296], per Bastarache J; Public School Boards’ Assn. of Alberta v Alberta (Attorney General) 2000 SCC 2, [2000] 1 SCR 44, [4]-[5]per Binnie J; R. v K.R.J. 2016 SCC 31, [2016] 1 SCR 906, [144], per Brown J.

[14] For the classic distinction, see Kenneth Culp Davis, “An Approach to Problems of Evidence in the Administrative Process” (1942) 55 Harvard Law Review 364, 402-403.

[15] R. (HS2 Action Alliance) v Secretary of State for Transport, [2014] UKSC 3, [110].

[16] Ibid. at para. 109.

[17] See e.g. Mikisew Cree First Nation v Canada (Governor General in Council) 2018 SCC 40, [2018] 2 SCR 765.

[18] See e.g. See e.g. Irwin Toy Ltd v Quebec (Attorney General) [1989] 1 SCR 927, 993-1000.

[19] See e.g. Moshe Cohen‑Eliya and Iddo Porat, “Proportionality and Justification” (2014) 64 University of Toronto Law Journal 458.

This content has been updated on May 2, 2022 at 14:18.