New Zealand Yearbook of New Zealand Jurisprudence
Last Updated: 16 April 2015
Charter-Proofing in Canada
Governments in liberal democratic societies may justifiably limit the rights and freedoms of citizens to protect them from terrorist threats. The question is only how extensive they can make those limitations. Politicians may, to appear strong on security issues, feel considerable pressure to limit rights as much as possible within certain 'absolute' boundaries (whether these boundaries are imposed by written constitutional provisions or unwritten norms). In these difficult political times, we may wonder if there is any basis for criticizing governments, not for limiting rights at all (which reasonable people agree is necessary to an extent), but for taking inadequate care when fashioning those limits to ensure they infringe rights as little as possible. This paper, drawing on the Canadian context, makes the case that saying why states ought to exercise more caution when limiting rights requires us to re-conceptualize rights themselves.
CHARTER-PROOFING: PRELIMINARY THOUGHTS
Shortly after the attacks upon the World Trade Centre and the Pentagon, the Liberal government in Canada introduced Bill C-36 — later the Anti-Terrorism Act.' The legislation, which introduced new police powers2 and created new criminal offences (and expanded some existing ones),3 attracted
Dr. Michael Plaxton, School of Law, University of Aberdeen, UK. I am grateful to Carissima Mathen, Fiona Leverick, James Chalmers and David Jenkins for their comments and suggestions on an earlier draft. Any errors or omissions are mine alone.
Anti-Terrorism Act, S.C. 2001, c. 41.
2 See Criminal Code, ss. 83.28, 83.3 (recognizing, respectively, investigative hearings and preventive detention).
considerable criticism from civil libertarians.' In response, the Minister of Justice remarked that the soon-to-be Act was `Charter-proof, meaning that the courts would not strike it down as contrary to the Charter of Rights and Freedoms.' The comment has provoked much discussion, in part owing to the difficulties we have in figuring out just how the fact that a piece of legislation is 'Charter-proof makes it defensible to critics outside the courtroom. Some of the controversy stems from the remark's possible implication that constitutional legislation is necessarily good legislation. Kent Roach, on the topic of 'Charter-proofing', has observed:
Charter-proofing is now an entrenched part of the legislative process in Canada, but it presents dangers especially if governments become more concerned about avoiding invalidation of legislation under the Charter than living up to its broader purposes and spirit. Charter-proofing can be a matter of shrewdly predicting what the courts will be prepared to do. Concerns exist, however, that courts, especially on sensitive matters, such as security, will be reluctant to strike legislation down. In short, a conclusion that Bill C-36, or for that matter any other piece of legislation, should survive Charter review does not mean it is good law.6
Roach is, of course, quite right: legislation may pass constitutional muster yet be 'bad law'. But tad law' in what sense? We might speak of law as 'bad' (or, more diplomatically, 'sub-optimal') in the sense it is inefficient or
See, for criticism of Bill C-36, many of the articles in Ronald J. Daniels, Patrick Macklem, and Kent Roach, eds., The Security of Freedom: Essays On Canada's Anti-Terrorism Bill (University of Toronto Press, 2001). See also Don Stuart, "The Anti-Terrorism Bill C-36: An Unnecessary Law and Order Quick Fix that Permanently Stains the Canadian Criminal Justice System" (2002) 14 N.J.C.L. 153; Kent Roach, September 11: Consequences for Canada (McGill
Queen's University Press, 2003) at ch. 3 [September 11].
Part I of the Constitution Act, 1982, being Schedule B to the Canada
Act 1982 (U.K.), 1982, c. 11 [Charter].
Kent Roach, "The Dangers of a Charter-Proof and Crime-Based Response to Terrorism" in Daniels, Macklem, and Roach, supra note 4 at 132.
costly. Alternatively, we might refer to a law as 'bad' because we think it immoral.' Depending on the moral theory we adopt, the first judgment can lead inexorably to the second, but it need not: only consequentialists — and only some consequentialists at that — would take the efficiency or cost-ineffectiveness of a policy as a prima facie reason to proclaim it 'immoral'. We usually have little trouble conceptually distinguishing the morality of a law from its efficiency. (There are few thoroughgoing utilitarians.)
When the Minister of Justice defended Bill C-36 as `Charter-proof , she surely meant to respond to accusations that the legislation was immoral — i.e., that it failed to respect the political morality of liberal democratic states. Charter review, we will suppose, seeks to identify (and invalidate) legislation that is incompatible with that political morality. To say, then, that the Act is 'Charter-proof' is only to say that it satisfies the strict requirements of morality — that, however else one might characterise the Act as `sub-optimal',8 no one can complain that it is immoral.
In at least one respect, that is a coherent argument. There is nothing wrong with 'Charter-proofing' per se. The government is supposed to pay attention, when drafting statutes, to what the constitution requires9— this is why courts interpret legislation with the presumption that the government intended to make it fit with those requirements.10 That
presumption anticipates that lawmakers will take the utmost care when crafting legislation, so that it does not conflict with Canadians' constitutional rights; or, if it must, conflicts with them as little as possible. The court, however, cannot know for sure that the legislature, in a given case, has undertaken such an exhaustive analysis of the issues and the possible solutions. Legislatures have far more institutional competence than the judiciary when it comes to evaluating and discovering policy alternatives, meaning that the judiciary might well conclude (by mistake) that a provision limiting rights and freedoms does so to the minimum extent possible, though there are in fact far less intrusive measures available. We must, then, distinguish limiting measures that can be justified, as proportionate,' to a court during the s.-1 stage of constitutional litigation,' from limiting measures that a lawmaking assembly can justify, as proportionate, to itself when crafting the legislation in the first place. When commentators deride the idea of `Charter-proofing', they have in mind a situation where the legislature cannot justify, to itself, the legislation at hand — either because it knows that the legislation goes further than necessary or because it has not taken the steps necessary to hold a good-faith belief — but enacts it in the hope (and, perhaps, the expectation)" that the courts will wave it past the s.-1 roadblock. We (hopefully) can imagine, however, situations where the government crafts limiting legislation having first formulated a good-faith belief that the measures to be deployed are genuinely proportionate to the objective. When that happens, the government has
counterparts in other countries. See, e.g., R. v. Secretary of State for the Home Department, Ex Parte Simms,  UKHL 33;  2 A.C. 115 at para. 44; United States v. Morrison,  USSC 42; 529 U.S. 598 at 607 USSC 42; , 120 S. Ct. 1740 at 1748 (2000).
11 See R. v. Oakes,  1 S.C.R. 103, where the Supreme Court laid down the proportionality test for the s.-1 analysis.
13 See David Dyzenhaus, "The Permanence of the Temporary: Can Emergency Powers Be Normalized?" in Daniels, Macklem, and Roach, eds., supra note 4 at 25.
`Charter-proofed' the statute in question, and acted rightly in doing so.
Defenders of Bill C-36 argued that the legislation was `Charter-proofed' in this latter sense:4 Since, again, there is a legal presumption to that effect, it is an easy argument to make. We have positive reasons, though, to think that Parliament crafted the Anti-Terrorism Act with some care. Stanley Cohen, of the Department of Justice, observed on at least two occasions that the Act did not go as far as antiterrorism legislation devised in other Western democracies.15 That is not, by itself, an answer to civil libertarians: every Western democracy, post-9/11, became panic-stricken over terrorist threats. We cannot merely look to the anti-terrorist legislation of other countries, when we want to decide what sort of legislation would respect rights and freedoms to an extent necessary in a 'free and democratic society', because that foreign legislation may not respect rights and freedoms to that necessary (minimum) degree.16 Even if we accept on faith that Canadian anti-terrorist legislation respects rights and freedoms more than that of other countries, we have not yet established it respects them as much as a 'free and democratic society' demands. In a sense, though, this line of thinking misses the mark: the point is not that the state can justify legislation to a court, scrutinising it under s.-1, just by comparing it to foreign statutes and policies devised to achieve similar aims. The point that Cohen makes is that we have reason to think that the government has 'Charter-proofed' legislation in the responsible sense when it chooses not to blindly follow the lead of countries that have imposed
14 See, e.g., Stanley Cohen, "Safeguards in and
Justifications for Canada's New Anti-Terrorism Act" (2002) 14 N.J.C.L. 99
at 101; "Concluding Comments from the Department of Justice" in Daniels,
Macklem, and Roach, eds., ibid.
15 Cohen, "Safeguards", ibid.; "Concluding Comments", ibid. at 440-1.
Thus, in Sauve v. Canada (Chief Electoral Officer),  3 S.C.R. 519, 218 D.L.R. (4th) 577, a majority of the Supreme Court struck down legislation disenfranchising serious criminal offenders for the duration of their imprisonment, though (as the dissent pointed out) many Western democracies disenfranchise prison inmates.
harsher limits on rights and freedoms (whether or not those limits are justifiable). The government may miscalculate, but we have no prima facie reason to think it has.
The public scrutiny, to which Bill C-36 was subjected, also gives us reason to think that alternatives were at least considered." Given the pressure on opposition parties to accept the legislation without serious challenge for the first two readings, and given the Liberal government's invocation of 'closure' after only a week of debate on the third reading, Bill C-36 was not debated in the House of Commons as much as one would like.' Nevertheless, the Bill was debated vigorously in the media and the academic community' before it became law, and the government participated in that discussion.' The government, at the committee stage, invited a great deal of input from public interest groups — input that impacted upon the content of the Act, albeit not as much as some civil libertarians would have wished.'
Parliament's 'Charter-proofing' of the Anti-Terrorism Act need not have been of the illegitimate variety. When the Minister of Justice used the term, she may only have meant that the legislation was neither drafted recklessly, nor intended to run contrary to the Charter.' We might regard that as quite a statement in itself, if we define a state of
See David Jenkins, "In Support of Canada's Anti-Terrorism Act: A Comparison
of Canadian, British, and American Anti-Terrorism Law"
(2003) 66 Sask. L. Rev.
419 at 422. Roach provides a fine account of the debate: September 11, supra
note 4 at 58-68.
Roach, ibid. at 65-8. Indeed, some opposition-party members claimed
that, had they known the government would invoke closure on the third reading,
they would have given Bill C-36 a harder look during the first two
See Daniels, Macklem, and Roach, supra note 4.
See, e.g., Irwin Cotler, "Does the anti-terrorism bill go too far? No:
we need powerful new legal tools to fight the new global terrorist threat"
and Mail (20 November 2001). Irwin Coder, a human-rights lawyer, served as a
Liberal Member of Parliament when Bill C36 was
introduced. He is the present
Minister of Justice.
See September 11, supra note 4 at ch. 3.
See Charter, s. 33 (`the notwithstanding clause').
emergency (at least in part) as a situation in which it becomes politically feasible to say, in explicit language, that certain measures should be implemented 'notwithstanding' Charter rights." In that case, it would appear significant that Parliament refused to make use of s. 33, despite the government's clear interest in presenting itself as 'tough on terrorism' .24 By opting to 'Charter-proor the Anti-Terrorism Act, rather than bypass the issue of constitutional compliance altogether by using the notwithstanding clause, the government effectively hinted that, whatever else 9/11 changed, it did not change Canadians' collective attitude towards their fundamental rights and freedoms." At a time when human-rights activists frequently call for a more robust institutional role for judges, on the basis that judges are less susceptible to political winds, it is worth recalling that politicians' vulnerability can (if we take seriously the idea that democracy is something that ordinary people want and not something foisted upon them by mandarins who know better) give them an incentive to protect fundamental rights —an incentive that unelected judges might lack.
We commonly think there is such a thing as good-faith `Charter-proofing'. Suppose, though, that we believe that the government, in devising the Anti-Terrorism Act, did not `Charter-proof it in good faith; that we believe the political
The Supreme Court of Canada, in rejecting the Crown's claim that the objective of the Anti-Terrorism Act concerned 'national security' rather than ordinary law enforcement, observed that the Crown, in drafting the Act, did not resort to the notwithstanding clause: Re Application under s. 83.28 of the Criminal Code,  2 S.C.R. 248 at para. 39.
25 In anticipation of an upcoming Parliamentary review of the Anti-Terrorism Act, a consulting firm has recently convened focus groups, asking participants what they thought of the Act. Many expressed concerns for its impact on rights. See Jim Bronskill, "Canadians leery of Anti-Terrorism Act" Globe and Mail (10 February 2004), online: Globe and Mail <www.globeandmail.com>.
pressure on the government and opposition parties to appear `tough on terror' motivated them to create measures that would limit rights and freedoms as much as possible, in the expectation that those measures would not be struck down by the courts. It is worth considering why we think that problematic. After all, there need not be anything inherently wrong with limiting fundamental rights and freedoms as much as the impugned legislation permits (or requires) —given the proper circumstances and the proper objective, a court might have no difficulty accepting that rights and freedoms should be limited to that degree. If the legislature does not cross any sacred lines — if it does not infringe rights to a degree that is per se unacceptable — we can only (if we insist that this bad-faith 'Charter-proofing' remains illegitimate) claim that it is illegitimate because the legislature does not have the proper objective yet. Why, though, should that matter? It is no good arguing that the legislature must have the 'right sort of objective' because, under the Oakes test, it must justify the extent of the infringement, and not merely the bare fact of it: that answer only pushes the problem back a step. We might wonder why it should make a moral difference that the Crown has limited rights and freedoms more than necessary, so long as it chooses not to cross certain 'absolute' boundaries.
There is another dimension to this problem. Even if the government acts in good faith, when 'Charter-proofing' legislation, it may fail to limit rights to the minimum extent possible (given its objectives). If the moral significance of `Charter-proofing' depends on the government's intention, it would appear that civil libertarians and others concerned with human rights necessarily have, in that case, no grounds for complaint. Yet we might think it peculiar that citizens could be conceptually unable to claim, on moral grounds, greater protection from the state for their constitutional rights and freedoms just because the incursion happened for innocent reasons. We surely think that the mere fact that legislation infringes rights and freedoms more than necessary is a reason to criticize it.
SUPEREROGATION AND DEONTOLOGY
Both these problems arise as a result of a presupposition we often make that rights are grounded in deontological arguments. A deontological account takes a rights-violation as per se morally abhorrent "whatever the consequences"." The account is agent-relative in the sense that it places top priority on the moral virtue of each agent (who, in the context of rights embodied in the Constitution, can be an institution or state agent)." An agent's moral virtue can be, but need not be, affected by what she does.2s It is affected only if the things she does have a particular causal relationship to certain outcomes or risks,' making them attributable to her. So long as she does not bear that relationship, the outcome or risk is not attributable to the moral actor, and her virtue is unaffected. It does not matter that others cause horrific things to happen; those consequences do not affect the actor's moral status as a 'good person' so long as she does not bear the requisite causal relationship to them. A deontological theory seeks to ensure only that agents do not have that causal relationship. It, therefore, forbids agents from acting in certain ways, but does not require them to positively act in other ways (since an obligation to help one person may require the moral agent to simultaneously act to the detriment of someone else).30 Absolutist conceptions of morality are, by definition, non-consequentialist because on that kind of view consequences per se have no moral significance. Only the causal relationship counts. So, a deontological analysis would ask whether an executive actor (like a police officer) treated
26 See Bernard Williams, "Consequentialism and Integrity" in Samuel Scheffler, ed., Consequentialism and Its Critics (Oxford, 1988) at 27 [edited volume cited as CAIC].
28 See Thomas Nagel, "War and Massacre" in CAIC, supra 26 at 52.
29 Moore describes the various theories of culpability in TBE, supra note 27 at 298-310.
30 See Isaiah Berlin, "Two Concepts of Liberty" in Four Essays On Liberty (Oxford, 1969). The actor, in that case, would bear the requisite causal relationship to those adverse consequences, and her virtue would accordingly be compromised.
someone in an impermissible way, or if rules devised by the legislature require that sort of impermissible treatment. If not, the Crown's moral virtue remains intact, even if other entities interfere with person in ways one recognizes as 'out of bounds'.
The absolutist model suffers from well-known defects. Because deontological approaches conceptually divorce rights from consequences, they seem unable to adequately explain why anyone should care about rights in the first place. Recall that a deontological theory is concerned only with the individual agent's moral virtue and, so, does not concern itself with consequences that have not been caused by the agent (remembering that, on this theory, an agent is capable of 'causing' things to happen according to a relatively narrow set of criteria). This means that, so long as the agent does nothing to cause (in that narrow sense) prohibited outcomes, her virtue remains intact even if her failure to act in a certain way causes (in a broader, more metaphysical sense) outcomes that one recognizes as far worse. Thus, the agent who indirectly (though intentionally) prevents one hundred (or one million) rights-violations by directly violating the rights of a single person is morally worse, not better. A deontological theory cannot say that rights aim to prevent the realization of bad outcomes, since the rights themselves may require agents to allow those bad outcomes to occur. In particular, it cannot say that rights advance the well-being of persons. But, if rights do not aim at human well-being, one wonders why rights are worth having. Rights, in that case, appear to be just an arbitrarily selected list of rules that serve no discernible human purpose.'
31 See T.M. Scanlon, "Contractualism and utilitarianism" in Amartya Sen and Bernard Williams, eds., Utilitarianism and beyond
(Cambridge, 1982) at 108-9 [edited volume cited as "Claims
about individual well-being are one class of valid starting points for moral argument. But many people find it much harder to see how there could be any other, independent starting points. Substantive moral requirements independent of individual well-being strike people as intuitionist in an objectionable sense." See also Scanlon, "Rights, Goals, and Fairness" in LAIC, supra note 26 at 74.
We can, right away, see why this view of rights could permit the Minister of Justice to use the idea of 'Charter-proofing' as a way of shutting down debate. If morality is not pegged to human well-being, then morality requires the agent only to avoid certain causal relationships to specified outcomes; that certain outcomes occur makes no moral difference to the agent. Institutional respect for rights may happen to make people well-off, but that respect is not right because it makes people well-off. Making people well-off, on a deontological account, is morally irrelevant. But if it is morally irrelevant that people are made well-off, institutions and other agents cannot improve their moral status by making people even better off. An absolutist framework simply cannot make sense of any moral distinction between doing what one must and doing more than that.' Rights — and their corresponding duties — exhaust the moral universe. And, if rights exhaust the moral universe, there is but one moral question for us to ask: `has the government violated rights?' If it has not, the government must walk with the angels, and we (its would-be critics) can have nothing to say. Even bad-faith 'Charter-proofing' is not, in that case, morally reproachable.
What human-rights critics need, when they attack legislation that respects rights but not as much as they want, is a conception of rights that allows for supererogative moral duties (or 'permissive oughts'). To say that someone (or, in this case, the state) has a supererogative moral duty is to say that she (or it) is not morally required to do X, but that she earns moral 'bonus points' if she does X; that she is morally satisfactory if she fails to do X, but that she is morally better if she does it. For our purposes, recognition of permissive oughts would allow civil libertarians to adopt a more nuanced position than the Minister of Justice, armed only with her `Charter-proofing' claim, can answer. It has that result because it allows critics to acknowledge that the Canadian
32 See Joseph Raz, "Right-Based Moralities" in R.G. Frey, ed., Utility and
Rights (University of Minnesota Press, 1984) [edited volume cited as UR];
Joseph Raz, The Morality of Freedom (Clarendon, 1985) at ch.
228 Yearbook of New Zealand Jurisprudence Vol 8.1
government, in devising the Anti-Terrorism Act, has not cast morality to the wind; that it has not, so far as this particular piece of legislation is concerned, transformed itself into a `wicked legal system'." This, in itself, would be a salutary development since it depolarizes politically charged debates. To be sure, a depolarized debate lowers the stakes for the government, meaning that critics themselves might prefer a more antagonistic approach. As we have seen, though, the government may have little difficulty replying to criticisms that call into question its status as a rights-respecting regime. All-or-nothing attacks either get quickly rebuffed in a public fashion that ultimately discredits civil libertarians and human-rights activists; or bogged down in debates concerning the actual content of rights — debates that activists may be unable to win convincingly enough to turn public opinion against the proposed legislation. Critics can bypass that debate by conceding that the Crown's legislation is not wicked — that it is only uncommitted to rights. Critics, in that case, can agree that the Crown, relative to other governments in other countries, is morally pretty good, while pointing out that it could be morally better.
But, as we have already seen, deontological theories of rights have trouble accommodating this idea of supererogation. The Crown's implicit reliance on a deontological model of rights would not be so threatening if that conception were not so popular. When, in our day-to-day lives as moral beings and armchair philosophers, we think about rights, we are inclined to think of them as, by definition, inviolable 'whatever the consequences'. This way of thinking about rights may also be reflected in the increasing domination, within the political sphere, of 'rights-talk' at the expense of less 'loaded' language — the implication being that, if rights are not engaged, an issue ceases to matter."
33 I have in mind David Dyzenhaus' Hard Cases In Wicked Legal Systems (Oxford, 1990).
There are, however, other ways of grounding rights — ways that, to an extent, are both more consistent with our moral intuitions and more able to recognize the concept of permissive oughts. Consider some ways in which deontological theories of rights — despite their superficial appeal — often strike us as implausible. First, deontological moral theories chiefly aim to preserve the virtue of individual agents at all costs, but there seems no good reason to believe that any moral theory should prioritize an individual person's moral virtue over the well-being of others,' particularly when there are catastrophic costs attached to respecting a person's rights.' That objection acquires greater force when the agents in question are state agents. People have a strong intuition that, when a person enters into public service, her obligations to citizens can overwhelm her ordinary moral duty, as a private person, to respect others' rights 'whatever the consequences' .37 As Moore puts it, "the ruler should take the guilt upon himself rather than allow his people to perish [or suffer-If' A deontological theory of morality, and therefore rights, fails to account for the sense that public actors must use their position to prevent certain consequences, even when that means 'dirtying their hands'.
These objections, in turn, relate to the most fundamental issue that people have with absolutist theories — that any plausible moral theory must take consequences per se as morally significant enough that they themselves can provide a moral reason for acting. For many people, it simply does not ring
21; Nigel Simmonds, "The Analytical Foundations of Justice" (1995) 54 C.L.J.
306 at 308.
The actor herself would not necessarily think of the choice in these terms:
if she thinks she ought to do whatever is necessary to
well-being under the circumstances, she will think that her moral virtue is
preserved by doing that, not that she
prefers their wellbeing over her own
virtue. See Nagel, supra note 28 at 61-2. A deontological theory,
however, seems to expect moral actors to think this way.
36 See TBE, supra note 27 at 313-4.
38 See TBE, supra note 27 at 329.
230 Yearbook of New Zealand Jurisprudence Vol 8.1
true that morality forbids state agents from torturing a terrorist so they may learn the location of a ticking time bomb; or kill an innocent man to avert a nuclear war.39 Even many die-hard non-consequentialists cannot accept that there could be moral rules that hold 'whatever the consequences'; at the very least, there must be exceptions built into the rules.' Regardless of whether the rules have built-in exceptions, though, some avowed non-consequentialists ( `threshold deontologists') appear prepared to say that an absolutist moral rule can 'give way' in the face of catastrophic consequences.' That view, while attractive, ultimately strikes one as untenable.' Still, it suggests a basic
39 See TBE, ibid. at 328.
40 See TBE, ibid. at 297-8.
Threshold deontology is untenable because it must rest on the premise that rights and consequences are somehow commensurable; that they are composed of the same moral 'stuff'. Moore denies that premise — he must, or else he cannot explain why threshold deontology does not 'collapse into' consequentialism. He wants to say that a moral agent can coherently think that it is bad for others to produce certain consequences, but that it is worse for her to produce those consequences herself, and therefore the consequences produced by others must be extraordinarily bad in order to justify doing ordinarily bad acts oneself. TBE, ibid. at 330n. That is a mysterious claim. If it is based on the idea that, as a rule, the best consequences will ultimately flow if everyone privileges her own projects and commitments, 'threshold deontology' cashes out as just a form of two-tier consequentialism. In that case, the agent might unreflectively apply a rule, or recognize a right, based on consequentialist reasoning, without considering the consequences in most instances where the rule applies. (This looks like the 'intuitive reasoning' described in R.M. Hare, Moral Thinking: Its Levels, Method and Point (Clarendon Press, 1981) [MT].) If the right action is ultimately
intuitive problem with deontological theories with which even deontologists have trouble coping: they simply cannot square the idea that moral rules could be indefeasible and agent-relative with the intuition that morality must first and foremost advance human well-being.
Utilitarianism trades on this very intuition. For the rule-utilitarian, "the only fundamental moral facts are facts about individual well-being.' Rights, as moral rules, deserve respect if and only if they maximize well-being in the long term." They may accomplish that because the long-term and consistent recognition of, and respect for, rights actually makes the world a markedly better place in which to live (notwithstanding the short-term costs in well-being accrued each time the rights in question are respected)." Alternatively, rights may deserve respect because individual moral actors lack the perspective, information, and impartiality to decide which courses of conduct will actually maximize well-being in the long run. Moral rules, on that view, limit the disutility that would result if actors, with all their human shortcomings, regularly tried to maximize overall well-being at the expense of particular persons.' A rule-utilitarian will claim that, when the state violates a person's rights, it diminishes the overall well-being of
determined by looking to consequences, though, then it seems wrong to say
that a moral person could (still less, should) regularly
act without giving any
thought to consequences. If she regularly considers consequences, instead of
rights, then rights cease to
do any moral work and one loses the deontological
thread. So Moore's `threshold deontology' seems to collapse into
after all, precisely because of the commensurability of rights
and consequences. See also Amartya Sen, "Rights and Agency" in CAIC, supra
43 Scanlon, "Contractualism and utilitarianism" supra note 31 at 108.
44 Act-utilitarianism is utterly incapable of accounting for any moral rules
at all (beyond 'thou shalt, on every possible occasion,
well-being') and will not be the object of attention in this paper.
See David Lyons, Forms and Limits of Utilitarianism (Oxford University
See Sumner, supra note 34 at 190-1.
232 Yearbook of New Zealand Jurisprudence Vol 8.1
persons (however one defines 'well-being'). This kind of justification for rights has a 'pragmatic' appeal to those who find deontological justifications ungrounded.
Rule-utilitarianism, however, is a notoriously unstable foundation for rights. Most glaringly, it rests on an empirical assumption that is, at best, untestable — there seems little reason to accept, on faith, the rule-utilitarian' s claim that discrete 'goods' are maximized is improved because agents adhere to the moral rules embodied in `rights'.' That point exposes two weaknesses in the utilitarian thesis. First, it suggests that rule-utilitarianism cannot really explain rights at all. Second, and more importantly, even if utilitarianism can explain rights, the rights it posits are not as robust as one might expect them to be. On some occasions, it will strike the actor as highly unlikely, perhaps impossible, that the costs of adhering to the moral rule will be one day offset by the longterm benefits. This concern is especially present when one considers minority rights, or the legal rights of criminal defendants, because utilitarianism, unlike absolutism, is agent-neutral: when calculating overall well-being, each person's interests count and no person's interests count for more than anyone else's. One may think this just another intuitively appealing feature of utilitarianism, since it is hard to see how this method of tallying well-being could be inequitable.' By calculating utility in this fashion, though, utilitarianism leads to the troubling conclusion that the rights of an individual or minority group exist only so long as they improve the well-being of the majority. That kind of thinking does not square with the accepted wisdom that constitutional rights are designed precisely to insulate individuals and minority groups from the whims of the majority.
See Ronald Dworkin, A Matter of Principle (Harvard, 1985) at 82-4 [AMOP].
48 See Hare, "Rights, Utility and Universalization: a Reply to John Mackie" in UR, supra note 32 at 106-7; Charles Taylor, "The Diversity of Goods" in UB, supra note 31 at 130.
WELL-BEING AND ASPIRATIONS
If utilitarianism satisfactorily explains neither rights nor suprerogation, one can nonetheless learn something about both by considering utilitarianism's hold on our moral thinking. Utilitarianism is attractive inasmuch as it reflects the common intuition that rights guarantee some measure of well-being for persons. It has intuitive appeal also because of its respect for the equality of persons: each person's interests count in the utilitarian calculus, and count as much as those of everyone else. That seems like a forthright and simple method of treating everyone as equally worthy of respect. The method only appears problematic because one has a further intuition — one that utilitarianism does not accommodate — that equal treatment cannot only mean treating everyone's interests as equivalent. Some interests, like the interest a community might have in the conviction of an innocent person, or the interest the majority might have in the segregation, disenfranchisement, or disempowerment of a minority group, seem like interests that should not count in the moral calculus. Yet there seems no way, on a utilitarian theory, to exclude them without undermining utilitarianism's claim to be a theory that pragmatically ties rights to wellbeing.' There might, after all, be more well-being in the world if those interests were counted and allowed to trump individual or minority rights. The problem, put bluntly, is that this net increase in well-being would be bought at the price of a just world. Because utilitarianism equates a just world with a world in which a given good is maximized, it cannot make sense of that problem. Ordinary people can, and this is why few people are thoroughgoing utilitarians.
Any credible moral theory must recognize that well-being matters, but it must also recognize that people aspire to relate
49 Some theorists have attempted to exclude certain interests or preferences from the notion of 'well-being': see, e.g., Mirrlees, "The Economic Uses of Utilitarianism" in UB, .supra note 31; Harsanyi, "Morality and the Theory of Rational Behaviour" in UB, supra note 31. But see "Introduction" in UB, supra note 31 at 10.
234 Yearbook of New Zealand Jurisprudence Vol 8.1
to each other as equals even if some quantity of 'goodness' is lost as a result.' To put the matter another way, a credible moral theory must reflect the manner in which living human beings measure well-being. We do not, unlike traditional utilitarians, measure our well-being according to the amount of happiness or money (or whatever) we collectively have. Rather, we measure it by the extent to which we, as individuals and as communities, have succeeded or failed in the worthwhile projects we undertake.' Thus we can, and often do, think ourselves better off when we sacrifice some good (e.g., happiness or money) for the sake of the successful completion of a project we have devised for ourselves.' Liberal democratic states, as servants of citizens, are designed to help us achieve our projects.
That line of thinking would not assist us in this thought-experiment if individuals' fundamental projects were radically at odds with each other.53 (In that case, the state could not advance one person's projects without setting back those of another — the state would either fail as a servant of the people, or would fail to create an equal society.) We may, however, reasonably choose to reject that possibility. Thinking that people lack common moral aspirations seems to require one to ignore the way people make political arguments in democratic societies. Sumner is correct to say that people disagree about the content and scope of their rights,' but virtually no one thinks it is acceptable if a person is treated unequally or unfairly (though people may disagree about whether that person actually has been treated unequally or unfairly). Even officials in notoriously undemocratic
See Scanlon, "Rights, Goals, and Fairness", supra note 31 at
See John Gardner, "On the General Part of the Criminal Law" in Antony Duff, ed., Philosophy and the Criminal Law: Principle and Critique (Cambridge, 1998).
See Sumner, supra note 34 at 158-9.
54 Ibid. at 1-17.
societies feel the need to pretend that citizens are treated equally, as a means of sustaining (however precariously) the illusion of legitimacy.55
One could object that if people aspire to live in a certain kind of society (we will suppose, arguendo, an equal society) they must surely think they would be better off in that society. In that case, people surely would not have the sort of interests or preferences that make rights untenable in a traditional utilitarian analysis. Since people do have those interests, it might seem implausible to think that people have the moral aspirations attributed to them. The objection comes down to this: if people share certain aspirations in common, we cannot explain how and why they disagree about the content and scope of rights. Here, though, one needs to distinguish between the abstract moral aspirations people have, and their intuitions concerning what a society that realizes those aspirations would look like.' A person who fails to carefully consider what a fair society looks like may think that such a society would permit certain interests to function as reasons for action when, in fact, it could not. Many people can have this impression. It would be wrong to think, however, that they do not have aspirations, just because they have fallen into error with respect to what their satisfaction entails.
Notice that the addition of moral aspirations makes it possible to say something that traditional utilitarians arc unable to say; namely, that a person may have 'superficial preferences' and `real preferences' and that only the latter should 'count'. Traditional utilitarianism's agent-neutrality prevents it from privileging certain interests over others — it has no resources for choosing a person's 'deep' preferences over her 'shallow' ones. If it tries to do so, it loses the two features that make utilitarianism intuitively appealing: it ceases to look like a theory that treats everyone's interests equally, and therefore
55 See Hard Cases, supra note 33.
236 Yearbook of New Zealand Jurisprudence Vol 8.1
ceases to look like a theory that treats people equally; and it becomes less readily identifiable as a theory that ties moral duties to people's actual well-being. Once we say that wellbeing is not to be equated with the accumulation of 'goods' —whether they be wealth or happiness — but with the successful completion of a common project, the problem disappears. We can draw a distinction between superficial and deep preferences, since a person (or community) can by pursuing a given course of conduct fail to achieve her goal even if she superficially believes that this course will advance it.
Because people share moral goals, mistakes about the scope and content of rights are amenable to correction through moral argument. A person who wants to criticize a community's understanding of rights may do so by pointing out the ways in which her peers have allowed personal interests and biases to lead them into error (i.e., with respect to what a society looks like when its people govern themselves according to fair terms of interactions). Critics can do this by showing what kind of world their peers would choose to inhabit if they were temporarily stripped of their prejudices and defining personal characteristics.' And it is here that contractarian moral theory springs into action. By showing people who want to live as equals what policies they would reject if stripped of their immediate preferences and personal commitments, contractarian theories make claims about what rights such people are committed to having by virtue of their commitment to a just society.58 Likewise, it reveals which rights a state, committed to the well-being of its citizens, must be committed to protecting.
As citizens in democratic societies (and, perhaps, as human beings in any society), we have a common project: that of achieving a just world. We expect liberal democratic states to respect rights because (a) we think that the just society we aspire to create must look a certain way; and (b) we think
See, e.g., John Rawls, A Theory of Justice, Rev. Ed. (Harvard, 1999).
58 See Scanlon, "Contractualism and utilitarianism", supra note 31 at 116.
that, by honouring rights, the state helps us create the society we want. And now we have a sense of what is wrong with a state that respects rights only as much as it absolutely must: when it acts as though rights are nothing more than an obstacle to be overcome, rather than 'moral furniture' that, just by existing, advance the projects (and therefore wellbeing) of citizens, the state either acts as though its citizens have no aspirations for a particular kind of world or is forced to say that it is not committed to the well-being of its citizens. Neither claim is acceptable. We think that the Crown should achieve its discrete objectives in a fashion that infringes rights as little as possible, because in doing so it re-affirms that we are committed to a particular end-state and that it is committed to helping us realize that goal. If the state merely respects rights — i.e., if it recognizes there are lines it should not cross under any circumstances — it behaves morally well, but if the state commits itself to the defense of rights, it behaves morally better. The Minister of Justice's 'Charter-proofing' claim cannot go far enough to satisfy those of us who think that the Crown should not only respect rights, but commit itself to them.
In times of crisis, liberal democratic governments may feel the strong temptation to 'sit on the fence' so far as rights are concerned — to refuse to abandon them, yet (in the name of appearing 'tough on terror') refuse to give them as much protection as they could. When we consider the intuitions we have about rights, we see that this approach is unacceptable. Once we tie rights to well-being — and, specifically, the wellbeing people have when they succeed in worthwhile activities — we find that a state committed to rights must do more than honour an ethereal line in the sand. It must have the character to protect rights as much as it can, even when doing so is inconvenient. When engaging the state in debates about rights, critics should insist on treating it as an entity devoted
238 Yearbook of New Zealand Jurisprudence Vol 8.1
to (rather than tolerant of) our rights and freedoms. Otherwise, the fight against terrorism, however much it preserves our lives, will — by preventing us from achieving the world we want — strip them of meaning.