Taking Deterrence Seriously: Excluding
Unconstitutionally Obtained Evidence Under
Section 24(2) of the Charter
Steven Penney*
Section 24(2) of the Canadian Chatter of Rights and Freedoms
grants courts the authority to exclude evidence obtained in a manner that
infringes or denies any Charter rights or freedoms, provided the adtnission
of such evidence would bring the administration of justice into disrepute.
While the problem of deciding in what circumstances to exclude illegally
obtained evidence has occupied both the Supreme Court of Canada and
scholarly commentators, their efforts to dats have been largely devoid of
theory. This article seeks to remedy this situation by developing a single,
coherent exclusionary theory. To this end, Professor Penney argues that the
only worthwhile reason to exclude evidence under section 24(2) is to deter
constitutional violations. A deterrence rationale allows for balance of the
conflicting purposes of section 24(2): encouraging constitutional
compliance and convicting the factually guilty.
rationale,
the “corrective
justice” rationale, and
In Part I, Professor Penney examines
the most common
justifications for exclusion of illegally obtained evidence-what he calls the
“condonation”
the
“deterrence” rationale-and conclades that deterrence is the only one that is
normatively plausible. He further argues that the objective of the
exclsionary rule should not be “maximum” deterrence, but “optimal”
deterrence. In other words, evidence should be excluded only when the
benefit of increased deterrence outweighs the cost of lost convictions. Part
ll of the article considers the available empirical evidence on the costs and
benefits of the exclusionary rule and concludes that while an exclusionary
rule can have a significant deterrent effect with few “lost convictions”, its
ability to deter is also limited by a number of phenomena, including most
importantly, the complexity of constitutional rules governing investigative
behaviouc. In Part ll, Professor Penney critiques Supreme Court section
24(2) jurisprudence in light of his exclusionary theory. Specifically, he
argues that the Court should develop ways to deter violations of the Charter
rights of third parties; that it should maintain
its liberal approach to
causation; that it should abandon both its “tral fairness” approach to self-
incriminating evidence and its “balancing approach” to other evidence; and
finally, that it should adopt a bright-line rule that encourages police to
become reasonably well-informed about their constitutional obligations and
signals to them that intentional and negligent violations will always result
in exclusion. This approach would be consistent with both the wording of
section 24(2) and much of the Court’s jurisprudence. It would also go some
way toward achieving a better balance between the rights-protection and
truth seeking finctions of section 24(2).
l6ments de preuve devraient dtre rejetdes.
L’article 24(2) de la Charte canadienre des droirs et libertds
confdre aux tribunaux le pouvoir d’exclure des 6lments de preuve obtenus
de mamire contraire aux droits garamis par Ia Charte, si leur admission
causerait prmudice A la bone administration de la justice. La Cow supree
do Canada et Ins commentatcurs ont examind la question de savoir dana
quelles circonmtances de tels
Toutefois, ces efforts ont Wt jusqu’l mainteant peu supportds par une
rdflexion th6orique. Cet article se propose de resudier A cetie situation en
proposant une thdorie de l’exclusion unifide et cohrente. Le professeur
Penney s’emploie A ddmontrer que la seule raison valable d’exclure des
4l6mnents de preuve en vertu de l’article 24(2) de Ia Charte est de
ddcourager Ins violations constitutionnelles. Un principe fondo sr Ia
dissuasion permet de rdconcilier les objectifs opposes do l’aticle 24(2), A
savoir d’encoumager l’obdssance A la constitution de trouver coupables
ceux qul le sont dana les faits.
la complexird des
Dana la premidre partie, l’auteur examine les motifs usuels pour
l’exclusion de la prmuve ildgalement obtenue, qu’il regroupe en vi raisons
fonddes sur l’excuse o,
i raisons de justice correctives) et i masons de
dissuasion s. nl conclut que seule Ia demiire est plausible d’un point de vrue
normatif. f soutient dgalement que le but de la tgle d’exclusion n’est pas
la dissuasion os maximale >), mais plutt la dissuasion v optimale o. Cela
signifie qu’un 6ldment de preuve ne devrait ftre exclu que lorsque le
bendfice d’one plus grande dissuasion ssepnsse
t cofit de condamnations
perdues. La deuxidme partie de I’article examine Ins donndes empiriques
disponibles smr les coots et bdndfices de In tidgle d’exclusion. L’auteur
conclut que si celle-ci permet un effet dissuasif substantiel tout en
occasiomant pea de os condamnations perdues o, sa capacitd A dissuader est
lirnitde par un certain nombre de facters, parmi lesquels on comptero
surtout
le
comportement des enqudeorus. Dana In troisine partie, l’auteur se sert de
sa thorie de I’exclusion poor critiquer la jurisprudence de la Cour supr&m
sur l’article 24(2). 1l soutient en particulier que la Cour devrait dvelopper
des manirmes de ddcourager les violations des droits des tiemces parties
garamnis par ln Charte, qu’elle devrait maintemr son approche hbrale se
les questions de causalit6; qu’efle devrait abandooner A la fois son approche
dite do v l’dquit6 du procds s quant A la preuve auto-incmnante et son
approche dite de vs l’6quilibre )s quant anx autres cattgories de preuve; et
elicter one rdgle stricte encoutageant Its corps
enfin qu’elle devrait
policiem A tre aisonnablement bien informrds de
leurs obligations
toute preuve obtenue en
constitotionelles
contravention intentionnelle ou ndgligente sern toujours rejetbe. Cette
approche serait A la fois conforme an texte de l’article 24(2) et A la plus part
de la jurisprudence de In Cowr. Elle contibuenan 6galement A mieax
6luilibrer les fonctions do protection des droits et d’enquhte de I’article 24(2).
rigles constitutionnelles gouvernant
indiquant que
et
leur
. Associate Professor, Faculty of Law, University of New Brumswick. I wish to thank Don Stuart,
Randal Graham, and two anonymous reviewers for helpful comments on earlier drafts. I am also
grateful to Bonny Miller and Emily Head for excellent research assistance.
McGill Law Journal 2003
Revue de droit de McGill 2003
To be cited as: (2003) 49 McGill L.J. 105
Mode de r~fdrence : (2003) 49 R.D. McGill 105
MCGILL LAw JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
Introduction
I. Justifications for Excluding Illegally Obtained Evidence
II. The Deterrent Effect of Exclusion and Its Alternatives
A. The Benefits and Costs of the American Exclusionary
Rule
B. Alternatives to Exclusion
Ill. Deterrence and Section 24(2) of the Charter
A. Standing
B.
C. Bringing the Administration of Justice into Disrepute
“Obtained in a Manner”
1. Evidence Affecting Trial Fairness: Automatic Exclusion
2. Evidence Not Affecting Trial Fairness: The Balancing
Approach
Conclusion
107
110
113
114
120
124
125
126
128
129
133
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
107
Introduction
Whether and in what circumstances courts should exclude illegally obtained
evidence is one of the most hotly contested questions in criminal procedure and
evidence law. It is not difficult to understand why. On the one hand, excluding
reliable evidence on the basis that it was obtained improperly detracts from the truth-
seeking function of criminal trials and makes it more likely that factually guilty
defendants will evade conviction and punishment. Reflecting this concern, the
common law traditionally denied the existence of any such power.’ On the other
hand, jurists have long argued that the absence of a power to exclude renders courts
impotent to control or dissociate themselves from police misconduct.’ Recent decades
have thus witnessed a loosening of the common law position, and many jurisdictions
now recognize a limited discretion to exclude improperly obtained evidence.3
But until the enactment of section 24(2) of the Canadian Charter of Rights and
Freedoms,4 only United States courts recognized a broad power to exclude evidence
obtained in violation of constitutional norms. Section 24(2) was conceived against the
is often
backdrop of the American “exclusionary
(inaccurately) perceived to mandate exclusion in every instance of constitutional
infringement.6 The language of section 24(2), which authorizes the exclusion of
unconstitutionally obtained evidence when the court determines that admission could
“bring the administration of justice into disrepute,” reflects a desire to avoid this
result.7 Beyond that, the meaning of the provision is unclear. The Supreme Court of
rule”,5 which
in Canada
1 See Quebec (A.G.) v. Begin, [1955] S.C.R. 593, 5 D.L.R. 394; R. v. Wray, [1971] S.C.R. 272, 11
D.L.R. (3d) 673. See generally Don Stuart, Charter Justice in Canadian Criminal Law, 3d ed.
(Scarborough, Ont.: Carswell, 2001) at 466-69 [Stuart, Charter Justice].
2 See Law Reform Commission of Canada, Report on Evidence (Ottawa: Information Canada,
1975) at 61-62.
3 See Police and Criminal Evidence Act 1984 (U.K.), 1984, c. 60, s. 78 (evidence may be excluded
“if it appears to the court that, having regard to all the circumstances, including the circumstances in
which the evidence was obtained, the admission of the evidence would have such an adverse effect on
the fairness of the proceedings that the court ought not to admit it.”); Lawrie v. Muir (1949), [1950]
J.C.A., [1949] S.L.T. 58 (H.C.J. Scot.); Bunning v. Cross (1978), 19 A.L.R. 641 (H.C.A.). See
generally Steven Penney, “Unreal Distinctions: The Exclusion of Unfairly Obtained Evidence Under
s. 24(2) of the Charter” (1994) 32 Alta. L. Rev. 782 at 784-95.
4 Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B
to the Canada Act 1982 (U.K.), 1982, c. 11 [Charter].
5 See A. Anne McClellan & Bruce P. Elman, “The Enforcement of the Canadian Charter of Rights
and Freedoms: An Analysis of Section 24” (1983) 21 Alta. L. Rev. 205 at 234; Kenneth Jull,
“Remedies for Non-Compliance with Investigative Procedures: A Theoretical Overview” (1985) 17
Ottawa L. Rev. 525 at 527-28.
6 See e.g. R. v. Collins, [1987] 1 S.C.R. 265 at 280, 38 D.L.R. (4th) 508 [Collins cited to S.C.R.]; R.
v. Simmons, [1988] 2 S.C.R. 495 at 532, 55 D.L.R. (4th) 673 [Simmons cited to S.C.R.].
7 Charter, supra note 4. For discussions of the legislative history of the provision see McClellan &
Elman, supra note 5 at 206-208; Stuart, Charter Justice, supra note 1 at 454-55, 476; David M.
Paciocco, “The Judicial Repeal of s. 24(2) and the Development of the Canadian Exclusionary Rule”
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
Canada has consequently devoted considerable energy to developing principles to
guide trial judges in deciding when to exclude evidence on the basis of section 24(2).
Commentators have subjected the complicated jurisprudence emerging from this
effort to frequent and trenchant criticism.8 This criticism, however, has been largely
void of theory.9 Commentators have pointed out ambiguities and contradictions in the
doctrine and have argued that exclusion should be less or more frequent on the basis
of general preferences for truth seeking or rights protection in the criminal justice
process.1″ They have also canvassed the various rationales for exclusion and pointed
out aspects of the Court’s jurisprudence that are consistent or inconsistent with those
rationales.” But few have attempted to prescribe an exclusionary regime that is tied to
and justified by a single, coherent exclusionary theory.
This article represents such an attempt. I argue that the only worthwhile reason to
exclude evidence under section 24(2) is to deter constitutional violations. Taking
deterrence seriously allows us to penetrate the ambiguity and confusion surrounding
section 24(2) and pursue an optimal accommodation between the competing interests
implicated by the provision: encouraging constitutional compliance and convicting
this accommodation, courts should exclude
the factually guilty. To achieve
unconstitutionally obtained evidence unless doing so would be unlikely to deter; that
is, when all state actors responsible for the violation honestly and reasonably believed
that they were complying with the Charter. This position is close to the contemporary
American exclusionary rule, 2 which at least some of the Charter’s architects sought
(1990) 32 Crim. L.Q. 326 at 354 [Paciocco, “Judicial Repeal”]; Martin L. Friedland, “Controlling the
Administrators of Criminal Justice” (1989) 31 Crim. L.Q. 280 at 292; James A. Stribopoulos,
“Lessons From the Pupil: A Canadian Solution to the American Exclusionary Rule Debate” (1999) 22
B.C. Int’l. & Comp. L. Rev. 77 at 117-19.
8 See e.g. R.J. Delisle, “Collins: An Unjustified Distinction” (1987) 56 C.R. (3d) 216 [Delisle,
ibid.; Richard Mahoney, “Problems with the Current
“Collins”]; Paciocco, “Judicial Repeal”,
Approach to s. 24(2) of the Charter: An Inevitable Discovery” (1999) 42 Crim. L.Q. 443; Carol A.
Brewer, “Stillman and Section 24(2): Much To-Do About Nothing” (1997) 2 Can. Crim. L. Rev. 239;
Peter Mirfield, “The Early Jurisprudence of Judicial Disrepute” (1988) 30 Crim. L.Q. 434; Stuart,
in Criminal
Charter Justice, supra note 1 at 482-520; Stephen G. Coughlan, “Developments
Procedure: The 1996-97 Term” (1998) 9 Sup. Ct. L. Rev. 273; John A.E. Pottow, “Constitutional
Remedies in the Criminal Context: A Unified Approach to Section 24 (Part II)” (2000) 44 Crim. L.Q.
34 [Pottow, “Unified Approach”]; Paul L. Moreau, “R. v. Stillman”, Case Comment (1997) 40 Crim.
L.Q. 148 at 153.
9 For an exception, see Jull, supra note 5. Richard Mahoney has gone so far as to argue that theory
is irrelevant to the proper interpretation of section 24(2). It is improper, in his view, to substitute any
theoretical framework for the plain language of the provision, which commands that exclusion be
based on the repute of the justice system. See Mahoney, ibid. at 446-50. It seems clear to me,
however, that the phraseology of section 24(2) is largely devoid of intrinsic meaning, and that
inevitably some effort must be made to ground the provision in theory and policy.
10 See e.g. Stuart, Charter Justice, supra note 1 at 480-520; Paciocco, “Judicial Repeal”, supra note 7.
11 See e.g. Kent Roach, Constitutional Remedies in Canada, looseleaf (Aurora, Ont.: Canada Law
Book, 1994) c. 10 [Roach, Constitutional Remedies].
12 See United States v. Leon, 468 U.S. 897 (1984) [Leon]; Illinois v. Krull, 480 U.S. 340 (1987)
[Krull]; Arizona v. Evans, 514 U.S. 1 (1995) [Evans].
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
to avoid. 3 But having started down the exclusionary road, it is the only reasonable
destination.
My argument proceeds as follows. In Part I, I evaluate the most common
justifications for exclusion and conclude, as the United States Supreme Court has
long maintained, 14 that deterrence is the only one that is normatively plausible. The
other main exclusionary rationales-avoiding
judicial condonation of police
misconduct and compensating victims of overreaching-fail straightforward cost-
benefit tests.
Concluding that deterrence is a normatively plausible justification for exclusion,
however, gets us only part of the way. We must also demonstrate that in practice
evidentiary exclusion is better able to achieve its objective than are its alternatives.
That objective is not “maximum” deterrence (awarding remedies to deter as much
misconduct as possible), but rather “optimal” deterrence (awarding remedies only
when the benefit of increased deterrence outweighs the cost of lost convictions). We
can assess the utility of the exclusionary remedy, then, by estimating its impact in
inducing constitutional compliance as discounted by its effect in causing factually
guilty defendants who would otherwise have been convicted to go free. Part II
therefore considers the available empirical evidence on the benefits and costs of the
American exclusionary rule. Contrary to the views of many, this evidence shows
fairly definitively that exclusion has a substantial deterrent effect and causes few “lost
convictions”. It also demonstrates, however, that this deterrent effect is limited by a
number of phenomena, including most importantly, the fact that some constitutional
rules governing investigative behaviour are too complex for police to understand. This
part also examines alternative methods of enforcing constitutional compliance,
concluding that they would be less likely to achieve optimal deterrence than exclusion.
In Part III, I critique the Supreme Court of Canada’s section 24(2) jurisprudence
with the conclusions of Parts I and H1 in mind. I argue that the Court should develop
ways to deter violations of the Charter rights of third parties and that it should
maintain its liberal approach to causation. I also recommend that the Court abandon
both its “trial fairness” approach to self-incriminating evidence and its “balancing”
approach to other evidence. The trial fairness approach, which mandates the virtually
automatic exclusion of self-incriminating evidence,
relies on the discredited
“corrective justice” exclusionary rationale and cannot be justified on deterrence
grounds. The balancing approach, which decides the admissibility of evidence that is
not self-incriminating by weighing the seriousness of the Charter violation against the
seriousness of the offence and the importance of the evidence to conviction, is
normatively plausible. But it does not produce enough ex ante certainty for optimal
13 See Paciocco, “Judicial Repeal”, supra note 7 at 342.
14 See Elkins v. United States, 364 U.S. 206 at 217 (1960); United States v. Calandra, 414 U.S. 338
at 347 (1974). See also Stone v. Powell, 428 U.S. 465 (1976) [Powell]; United States v. Janis, 428 U.S.
433 (1976); Leon, supra note 12; Evans, supra note 12; Pennsylvania Board of Probation and Parole
v. Scott, 524 U.S. 357 (1998).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
deterrence; that is, it fails to deter many violations ex ante that a court applying an ex
post balancing test would conclude warrant exclusion. Optimal deterrence instead
requires a bright-line exclusionary rule that encourages police to become reasonably
well-informed about their constitutional obligations and signals to them that
intentional and negligent violations will always result in exclusion. It does not insist,
however, that the Court adopt an automatic exclusionary rule. Optimal deterrence can
be achieved by admitting evidence when (and only when) the violations that
produced them were inadvertent and reasonable. This standard is compatible with the
wording of section 24(2) as well as much (but not all) of the Court’s jurisprudence
relating to the “seriousness” of the constitutional violation.
I. Justifications for Excluding Illegally Obtained Evidence
justice”
rationale),
and deterring constitutional violations
Jurists have put forward three main justifications for excluding illegally obtained
evidence: avoiding judicial condonation of police misconduct (the “condonation”
rationale), compensating victims of violations to restore the status quo ante (the
“corrective
(the
“deterrence” rationale). The Supreme Court of Canada has directly or indirectly
endorsed all three as reasons to exclude evidence under section 24(2) of the Charter.
In Collins, it stated that while police misconduct “often has some effect on the repute
of the administration ofjustice, … s. 24(2) is not a remedy for police misconduct… “”
Rather, the provision’s purpose is to prevent the “further disrepute” occasioned by the
admission of evidence that would “deprive the accused of a fair hearing, or from
judicial condonation of unacceptable conduct by the investigatory and prosecutorial
agencies.”16 This statement explicitly refers to the condonation rationale. And as will
be discussed in more detail later, the Court’s approach to ensuring the “fairness” of
the hearing, which requires the automatic exclusion of non-discoverable, self-
incriminating evidence, is grounded in the corrective justice rationale. 7 Lastly, while
the Court in Collins was skeptical of the use of exclusion to directly discipline police,
in later cases it has stated that exclusion is designed in part to “oblige law
enforcement authorities to respect the exigencies of the Charter” and to “promot[e]
the decency of investigatory techniques …8
15 Collins, supra note 6 at 280-8 1.
16 Ibid. at 281. The idea that admitting illegally obtained evidence stains the integrity of the
judiciary first gained prominence in the dissents of Justices Holmes and Brandeis in Olmstead v.
United States, 277 U.S. 438 (1928). See also Robert M. Bloom, “Judicial Integrity: A Call for its Re-
emergence in the Adjudication of Criminal Cases” (1993) 84 J. Crim. L. & Criminology 462.
17 See discussion infra notes 116-17 and accompanying text.
18 R. v. Burlingham, [1995] 2 S.C.R. 206 at paras. 25 and 50, 124 D.L.R. (4th) 7 [Burlingham cited
to S.C.R.]. See also R. v. Kokesch, [1990] 3 S.C.R. 3 at 35, [1991] 1 W.W.R. 193 [Kokesch cited to
S.C.R.]; R. v. Stillman, [1997] 1 S.C.R. 607 at para. 126, 185 N.B.R. (2d) 1 [Stillrnan cited to S.C.R.];
R. v. Feeney, [1997] 2 S.C.R. 13 at para. 82, 146 D.L.R. (4th) 609 [Feeney cited to S.C.R.]; R. v.
Buhay (2003), 225 D.L.R. (4th) 624 at para. 70,2003 SCC 30 [Buhay cited to D.L.R.].
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
We can dismiss the condonation rationale fairly easily. In the eyes of most
citizens, excluding illegally obtained evidence harms the reputation of the justice
system much more often than admitting it does.19 There is no indication, moreover,
that in the years before the Charter the absence of an exclusionary remedy fomented
widespread disrespect for the system.2′ But even if we assume that the system’s
reputation is occasionally harmed by the admission of the evidentiary fruits of police
misconduct (for example, where police have acted reprehensibly and the offence is
not terribly serious), it is not because the use of such evidence in the proceedings is
inherently unseemly. Rather, it is because admission represents a lost opportunity for
deterrence. If there were no possibility that excluding evidence might deter future
misconduct, then the costs of exclusion would clearly outweigh whatever marginal,
ephemeral benefit that might be gained from avoiding judicial “complicity” in police
impropriety. Factually guilty defendants would be more likely to be acquitted and
future criminal suspects would be no less likely to endure abusive investigative
(or possibly
practices.
compensation, as I discuss in more detail below), the reputation of the justice system
would likely suffer from any instance of exclusion, no matter how egregious the
in these (hypothetical)
misconduct or trivial the offence. Admitting evidence
circumstances does not condone police misconduct. It simply recognizes that the
violation cannot be undone, that exclusion would not prevent future misconduct, and
therefore no purpose would be served in suppressing reliable evidence of guilt. To the
extent that condonation rhetoric has any normative appeal, it is as a proxy for the
deterrence rationale. It does not in itself justify the exclusion of illegally obtained
evidence.
from considerations of deterrence
Indeed, divorced
The corrective justice rationale is initially more appealing. It justifies the
exclusion in order to compensate victims of overreaching and reinstate the status quo
ante between the individual and the state.2′ The state should not be permitted, on this
view, to benefit from violating criminal suspects’ rights by using unconstitutionally
obtained evidence to help convict them. The problem with this argument, however, is
that evidentiary exclusion provides entirely too much compensation to criminal
defendants. In the vast majority of cases, the exclusion of inculpatory evidence will
be worth much more to defendants than avoiding the harm induced by constitutional
violations. The exclusion of unconstitutionally obtained evidence makes it more
likely that defendants will be acquitted. The social stigma, economic repercussions,
and penal sanctions accompanying criminal conviction are generally extremely
severe. If they could afford to do so, most defendants would be willing to pay
19 See Alan W. Byrant et al., “Public Attitudes Toward the Exclusion of Evidence: Section 24(2) of
the Canadian Charter of Rights and Freedoms” (1990) 69 Can. Bar Rev. 1 at 33.
20 See R. v. Duguay (1985), 50 O.R. (2d) 375 at 392, 18 C.C.C. (3d) 289 (Ont. C.A.), Zuber J.,
dissenting, aff’d [1989] 1 S.C.R. 93.
21 See Jull, supra note 5 at 530-3 1; Kent Roach, “The Evolving Fair Trial Test Under Section 24(2)
of the Charter” (1996) 1 Can. Crim. L. Rev. 117 [Roach, “Evolving”]; Roach, Constitutional
Remedies, supra note 11 at paras. 10.30-10.60.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
considerable sums to reduce the likelihood of suffering these consequences. These
payments would in most cases far exceed the amount that defendants would accept,
ex ante, as compensation for permitting the police to violate their Charter rights.22
in other words, gives most
Excluding unconstitutionally obtained evidence,
to the wrong.23
defendants a windfall-the remedy is grossly disproportionate
Corrective principles only justify exclusion in those rare cases where the violation
causes severe harm and the consequences of conviction are unusually benign.
If the foregoing analysis appears too crassly economic,
then consider the
corrective justice rationale through a moral lens. The immediate benefits of exclusion
accrue to the factually guilty.24 In most cases, the state’s misconduct in violating the
Charter pales in comparison to the accused’s misconduct in committing a criminal
offence.25 Criminal procedural rights protect both the innocent and the guilty against
investigative abuses, but this does not mean that the latter are as deserving of the
state’s concern and respect as the former. The Charter protects guilty suspects chiefly
because police cannot be certain, ex ante, who is guilty and who is not.26 If ex ante
certainty were possible, then we would surely grant criminals less constitutional
protection than law-abiding persons. When guilt has been established the state may
legitimately restrict the liberty and privacy of persons convicted of criminal offences
22 See Donald Dripps, “The Case for the Contingent Exclusionary Rule” (2001) 38 Am. Crim. L.
Rev. 1 at 16-17 [Dripps, “The Case”].
23 Exclusion also does nothing to compensate persons whose Charter rights are violated without
producing inculpatory evidence, though such persons can theoretically seek compensation by other
means. See Paciocco, “Judicial Repeal”, supra note 7 at 335.
24 While the exclusion of evidence under section 24(2) may benefit the innocent accused, it does so
incidentally and arbitrarily. As I discuss at infra note 116, courts can exclude unreliable evidence that
might lead to wrongful convictions without resort to section 24(2).
25 Exceptions might arise when the accused’s conduct is controversially criminal, such as the
possession of “soft” drugs, and the police employ heavy-handed, privacy-invasive tactics.
26 See R. v. Edwards, [1996] 1 S.C.R. 128 at para. 64, 26 OR. (3d) 736, La Forest J., concurring
[Edwards cited to S.C.R.] (“We exercise discretion to exclude evidence obtained by unconstitutional
searches from being used against an accused, even when it would clearly establish guilt, not to protect
criminals but because the only really effective safeguard for the protection of the constitutional right
we all share is not to allow use of evidence obtained in violation of this public right when doing so
would bring the administration of justice into disrepute.”); Henry J. Friendly, “The Bill of Rights as a
Code of Criminal Procedure” (1965) 53 Cal. L. Rev. 929 at 951 (“A defendant is allowed to prevent
the reception of evidence proving his guilt not primarily to vindicate his right of privacy, since the
benefit received is wholly disproportionate to the wrong suffered, but so that citizens generally … may
be ‘secure … against unreasonable searches and seizures.”‘); Brinegar v. United States, 338 U.S. 160
at 181 (1949), Jackson J., dissenting (“Courts can protect the innocent against such invasions only
indirectly and through the medium of excluding evidence obtained against those who frequently are
guilty.”) See also Steven Penney, “A Concern for Innocence: Justice La Forest and the Criminal Law”
in Rebecca Johnson et al., eds., G~rard V La Forest at the Supreme Court of Canada: 1985-1997
(Winnipeg: The Supreme Court of Canada Historical Society / Canadian Legal History Project, 2000)
217 at 235 [Penney, “A Concern”].
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
in ways that would clearly violate the Charter if applied to non-offenders. 27 Criminal
defendants have no right not to be convicted on the basis of illegally obtained
evidence.28 Excluding evidence to restore the status quo ante between criminals and
the law abiding public, therefore, is morally unjustified. There is nothing wrong with
compensating victims of police overreaching for the injuries they have suffered, even if
those victims have committed crimes. The compensation should, however, be
proportional to the harm caused by the violation. In the vast majority of cases,
evidentiary exclusion would not be.
To summarize, whether measured in terms of ensuring judicial integrity or
providing appropriate compensation to victims of overreaching, the benefits of
exclusion are clearly outweighed by its social costs. Divorced from deterrence
considerations, the reputation of the justice system would be better served by
admitting illegally obtained evidence than by excluding it. And if constitutional
remedies are supposed to compensate victims of investigative abuses, then monetary
damages or other remedies would be more appropriate than evidentiary exclusion.
II. The Deterrent Effect of Exclusion and Its Alternatives
The merits of excluding evidence to deter constitutional violations are more
difficult to assess. Deterring police from violating the rights of innocent suspects is
undoubtedly a laudable objective. Excluding unconstitutionally obtained evidence
may give some factually guilty defendants a compensatory windfall, but the
deterrence rationale’s proponents assert that this cost must be incurred in order to
ensure constitutional compliance. 9
We do not, however, want as much deterrence as we can get. As with criminal
punishments, courts should aim for “optimal” and not “maximum” deterrence.3″
Sanctions for constitutional violations, in other words, should only be awarded when
their social benefits
their social costs (lost
convictions). Many jurists have accordingly argued that deterrence does not justify
exclusion.3 They maintain that exclusion is not an especially effective deterrent; that
its social costs are unacceptably high; and that there are other methods of deterring
(increased deterrence) outweigh
27 See e.g. Weatherall v. Canada (A.G.), [1993] 2 S.C.R. 872, 105 D.L.R. (4th) 210 (prison inmates
subject to substantially reduced expectation of privacy).
28 See Richard A. Posner, “Rethinking the Fourth Amendment” [1981] Sup. Ct. Rev. 49 at 51-53
[Posner, “Rethinking”].
29 See Potter Stewart, “The Road to Mapp v. Ohio and Beyond: The Origins, Development and
Future of the Exclusionary Rule in Search-and-Seizure Cases” (1983) 83 Colum. L. Rev. 1365 at
1396.
30 See Posner, “Rethinking”, supra note 28 at 54-55. See generally Gary S. Becker, “Crime and
Punishment: An Economic Approach” (1968) 76 Journal of Political Economy 169; Robert Cooter &
Thomas Ulen, Law and Economics, 3d ed. (Reading, Mass.: Addison Wesley Longman, 2000) at 444-
46.
31 See infra notes 33-37.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
constitutional violations that are likely to be at least as effective and do not entail the
loss of reliable evidence and the acquittal of factually guilty defendants. Determining
who is correct requires detailed empirical and theoretical analyses of the benefits and
costs of the exclusionary remedy and its alternatives.
A. The Benefits and Costs of the American Exclusionary Rule
There is a long-standing debate over the degree to which evidentiary suppression
influences police conduct. Deterrence theory dictates that legal sanctions can only
influence behaviour when people understand the law intended to influence them,
bring that understanding to bear on their conduct, and perceive that the costs of non-
compliance outweigh its benefits.3″ Critics of the American exclusionary rule have
accordingly argued that police are frequently unaware of the complex and ever-
changing constitutional law regulating their investigative efforts;33 are not informed of
the suppression of evidence that they have collected;34 fail to understand the reason for
exclusion when informed of it;3″ and/or perceive that their interests are better served by
non-compliance than by compliance, even when they know that non-compliance may
result in evidentiary exclusion.36 Critics also point to empirical studies purporting to
show that the deterrent impact of the exclusionary rule is minimal.37 Even many jurists
32 See Paul H. Robinson & John M. Darley, “Criminal Law as an Instrument of Behavioral Control:
Should Deterrence Have a Role in the Formulation of Criminal Law Rules?” Social Science Research
Network Working Paper Series (October 2002) at 24, online: Social Science Research Network
Working Paper Series
33 See e.g. Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388 at
417 (1971) [Bivens], Burger C.J., dissenting; Harry M. Mitchell Caldwell & Carol A. Chase, “The
Unruly Exclusionary Rule: Heeding Justice Blackmun’s Call to Examine the Rule in Light of
Changing Judicial Understanding About its Effects Outside the Courtroom” (1994) 78 Marq. L. Rev.
45 at 54-56; L. Timothy Perrin, H. Mitchell Caldwell & Carol A. Chase, “An Invitation to Dialogue:
Exploring the Pepperdine Proposal to Move Beyond the Exclusionary Rule” (1999) 26 Pepp. L. Rev.
789 at 794; L. Timothy Perrin et al., “If It’s Broken, Fix It: Moving Beyond the Exclusionary Rule: A
New and Extensive Empirical Study of the Exclusionary Rule and a Call for a Civil Administrative
Remedy to Partially Replace the Rule” (1998) 83 Iowa L. Rev. 669 at 676; Dallin Oaks, “Studying the
Exclusionary Rule in Search and Seizure” (1970) 37 U. Chi. L. Rev. 665 at 730-31; James E. Spiotto,
“Search and Seizure: An Empirical Study of the Exclusionary Rule and Its Altematives” (1973) 2 J.
Legal Stud. 243 at 274,276.
34 See Caldwell & Chase, ibid. at 55; Perrin et al., ibid. at 675.
35 See Caldwell & Chase, ibid.; Perrin et al., ibid.
36 See Caldwell & Chase, ibid. at 56; Perrin et al., ibid.; Oaks, supra note 33 at 720-24; Jerome H.
Skolnick, Justice Without Trial: Law Enforcement in Democratic Society (New York: John Wiley &
Sons, 1966) at 219-25.
37 See Oaks, supra note 33; Spiotto, supra note 33; Perrin et al., ibid.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
who support exclusion have conceded that its deterrent effect has yet to be proven
definitively.38
At first glance this skepticism seems justified. The empirical evidence is both
conflicting and vulnerable to methodological critique. But if we strip away the
partisan rhetoric and examine the data carefully, we can conclude with reasonable
confidence that the American exclusionary rule has had a significant but limited
deterrent effect on police overreaching. Perhaps the best evidence of the rule’s impact
is the dramatic increase in the use of warrants after Mapp v. Ohio,39 the case that
imposed the exclusionary rule on states that had not already adopted it as a matter of
state law.4″ Since warrant-based searches are more likely to be constitutional than
warrantless ones, we can infer greater compliance from an increase in the proportion
of the former to the latter.4″ We can also infer a significant deterrent effect from the
fact that successful suppression motions are exceptionally rare.42 Further evidence of
38 See e.g. Wayne R. LaFave, Jerold H. Israel & Nancy J. King, Criminal Procedure, 3d ed. (St.
Paul, Minn.: West, 2000) at 115; Stuart, Charter Justice, supra note 1 at 474.
‘9 367 U.S. 643 (1961).
40 See Michael J. Murphy, “Judicial Review of Police Methods in Law Enforcement: The Problem
of Compliance by Police Departments” (1966) 44 Tex. L. Rev. 939 at 941-42; Bradley C. Canon, “Is
the Exclusionary Rule in Failing Health? Some New Data and a Plea Against a Precipitous
Conclusion” (1974) 62 Ky. L.J. 681 at 708-11; Sam J. Ervin, Jr., “The Exclusionary Rule: An
Essential Ingredient of the Fourth Amendment” [1983] Sup. Ct. Rev. 283 at 293. Researchers have
also attempted to gauge the rule’s deterrent effect by tracking changes to post-Mapp arrest and
property seizure rates. See Oaks, supra note 33 at 707 (finding that adoption of the exclusionary rule
initially had little effect on the number of arrests or on the amount of stolen property recovered by the
police in Cincinnati, but may “have had a slight long range effect of inducing greater conformity in
searches for stolen property.”); Canon, ibid. at 703-707 (replicating Oaks’ Cincinnati study in 13 cities
and finding in many cases a significant decline in arrests after adoption of the exclusionary rule). It is
unclear, however, whether these changes were caused by the adoption of the exclusionary rule. Other
variables may have been responsible. See generally Oaks, supra note 33 at 713-15; Canon, ibid at
698-702; Perrin et al., supra note 33 at 711.
41 The data, it should be noted, discloses an absolute increase in warrant usage. It does not reveal
whether the proportion of warrant-based searches has increased. Some scholars have therefore
attributed the spike in warrants to an increase in all types of searches induced by rising drug-related
crime rates. See Steven R. Schlesinger, “The Exclusionary Rule: Have Proponents Proven That It Is a
Deterrent to Police?” (1979) 62 Judicature 404 at 407. This conclusion is belied, however, by both the
intensity and immediacy of the post-Mapp warrant explosion. See Dripps, “The Case”, supra note 22
at 15. There is little reason to think, moreover, that in the absence of the exclusionary rule an increase
in the frequency of searches would lead to a significant increase in warrant usage. Without the
prospect of exclusion, police have little incentive to obtain a warrant. See Canon, supra note 40 at
714.
42 See Canon, supra note 40 at 708-709; Peter F. Nardulli, “The Societal Cost of the Exclusionary
Rule: An Empirical Assessment” [1983] Am. Bar Found. Res. J. 585 [Nardulli, “Empirical
Assessment”]; Thomas Y Davies, “A Hard Look at What We Know (and Still Need to Learn) About
the ‘Costs’ of the Exclusionary Rule: The NIJ Study and Other Studies of ‘Lost’ Arrests” [1983] Am.
Bar Found. Res. J. 611; General Accounting Office, Impact of the Exclusionary Rule on Federal
Criminal Prosecutions: Report of the Comptroller General of the United States (Washington, D.C.:
U.S. General Accounting Office, 1979); Richard Van Duizend, L. Paul Sutton & Charlotte A. Carter,
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
the exclusionary rule’s deterrent impact can be garnered from surveys questioning
police and other justice system officials about their perceptions and practices. Almost
all of these studies show that these officials believe that the rule exerts significant
deterrent force and that they behave accordingly.43
This research also reveals, however, that the rule’s deterrent effect is far from
perfect. As deterrence
is often hindered by
informational deficits and unfavourable cost-benefit calculations. The greatest barrier
theory would predict, compliance
The Search Warrant Process: Preconceptions, Perceptions, Practices (Williamsburg, Va.: National
Center for State Courts, 1985). But see National Institute of Justice, The Effects of the Exclusionary
Rule: A Study in California (Washington, D.C.: The National Institute, 1983). It must be pointed out,
however, that this measure does not account for violations that do not result in the laying of charges.
See William C. Heffernan & Richard W. Lovely, “Evaluating the Fourth Amendment Exclusionary
Rule: The Problem of Police Compliance with the Law” (1991) 24 Mich. J.L. Reform 311 at 320-21.
43 See Michael Katz, “The Supreme Court and the States: An Inquiry into Mapp v. Ohio in North
Carolina: The Model, the Study, and the Implications” (1966) 45 N.C. L. Rev. 119 at 134 (substantial
majority of prosecutors, defence lawyers, and judges surveyed believed the exclusionary rule is an
effective deterrent); Myron W. Orfield, Jr., “The Exclusionary Rule and Deterrence: An Empirical
Study of Chicago Narcotics Officers”, Comment (1987) 54 U. Chicago L. Rev. 1016 [Orfield,
“Exclusionary Rule and Deterrence”] (survey of police finding significant compliance due to
exclusionary rule); Van Duizend, Sutton & Carter, ibid. at 120 (“[T]he exclusionary rule, though
seldom invoked, serves as an incentive for many police officers to follow the limits imposed by the
Fourth Amendment as defined in their jurisdiction.”); Craig D. Uchida & Timothy S. Bynum, “Search
Warrants, Motions to Suppress and ‘Lost Cases’: The Effects of the Exclusionary Rule in Seven
Jurisdictions” (1991) 81 J. Crim. L. & Criminology 1034 at 1065 (“Our interviews indicated that
police were willing to follow guidelines established by the Constitution, the district attorney’s office,
and the courts when writing search warrant applications.”); Myron W. Orfield, Jr., “Deterrence,
Perjury, and the Heater Factor: An Exclusionary Rule in The Chicago Criminal Courts” (1992) 63 U.
Colo. L. Rev. 75 at 91 [Orfield, “Deterrence, Perjury”] (survey of defence lawyers, prosecutors, and
judges finding significant deterrence); Milton Loewenthal, “Evaluating the Exclusionary Rule in
Search and Seizure” (1981), 49 UMKC L. Rev. 24 at 29-30 (participant-observer study finding that
police accept exclusionary rule as necessary to deter violations); Perrin et al., supra note 33 at 720-21,
tbl. 1 (80 per cent of police surveyed consider possibility of exclusion to be either the “primary” or an
“important” concern in conducting investigations) and 732 (majority of police believed that “the
interests of the criminal justice system are well served by excluding unlawfully seized evidence”);
Heffernan & Lovely, ibid. at 347-48 (85 per cent of police surveyed indicate willingness to comply
with their understanding of search and seizure law). See also Commission of Inquiry Concerning
Certain Activities of the Royal Canadian Mounted Police, Freedom and Security Under the Law:
Second Report, vol. 2 (Ottawa: The Commission, 1981) at 1041 (“It can now be said, at least in this
country and in regard to the R.C.M.P., that the attitude of members of that Force … is to regard the
absence of critical comment by the judiciary as tacit approval of forms of conduct that might be
unlawful.”) But see Skolnick, supra note 36 at 215-29 (participant-observer study of one department
finding that police compliance with constitution not greatly influenced by exclusionary rule).
There has been no empirical study to date of the deterrent effect of exclusion under section 24(2)
of the Charter. A study of the implementation of two Supreme Court of Canada criminal procedural
decisions by one police force concluded, however, that “police have managed to cope with new
requirements imposed by the Charter and have effectively
implemented changes to standard
procedures” (Kathryn Moore, “Police Implementation of Supreme Court of Canada Charter
Decisions: An Empirical Study” (1992) 30 Osgoode Hall L.J. 547 at 577).
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
to deterrence is legal complexity. The more complex the rules governing criminal
investigation, the less likely it is that police will understand them and apply them
correctly on the ground.’ Police find it particularly difficult to apply rules associated
with warrantless searches, especially where the impact on privacy interests is only
moderate.45 They have much less difficulty applying straightforward regulations, such
as the Miranda rules dictating the content of pre-interrogation warnings.’
It is also
apparent that in a significant minority of cases police are unwilling to obey the rules
even when they know what is expected of them.47 They may fail to comply because
they believe that some other objective (such as restoring order, preserving safety,
removing contraband, “clearing” cases, or harassing undesirables) is more important
than obtaining evidence for prosecution.48 Or they may believe that any evidence that
they obtained illegally will be admitted as a result of police perjury or judicial
reluctance to exclude.49
None of this should be particularly surprising. Logic and intuition suggest that the
prospect of evidentiary suppression and the concomitant loss of convictions must
have some impact on police behaviour.50 Exclusion may not affect police as acutely as
monetary sanctions or imprisonment, but it does matter. It influences police conduct
directly by serving as a basis for performance evaluations or spurring training
44 See Heffeman & Lovely, supra note 42 at 322-24. See also Moore, ibid. at 552, 572.
45 See Heffernan & Lovely, supra note 42 at 322-23. Heffeman & Lovely presented three groups of
individuals (police, criminal lawyers, and laypersons), with six search and seizure scenarios derived
from Supreme Court cases and asked them to decide whether the intrusion by the police was legal.
The problems were “difficult” in the sense that they involved warrantless search powers and relatively
innocuous privacy intrusions. In the researcher’s estimation the legality of the intrusion in each
scenario was “not clear cut” (ibid. at 328-29). Individuals in each group were also required to respond
to a series of multiple-choice questions on search and seizure law. The average score (out of 16) for
police officers was 8.3, compared to 10.3 for lawyers and 7.0 for laypersons. Ibid. at 334, tbl. 3. See
also Perrin et al., supra note 33 at 727-28 (administering similar tests to police and finding that police
answered 65 per cent of questions correctly; the control group of law students averaged 70 per cent).
Of course, these studies do not tell us the extent to which police responses to hypothetical scenarios
mirror their actual behaviour in the field. See Gregory D. Totten, Peter D. Kossoris and Ebbe B.
Ebbesen, “The Exclusionary Rule: Fix It, But Fix It Right-A Critique of ‘If It’s Broken, Fix It:
Moving Beyond the Exclusionary Rule”‘ (1999), 26 Pepp. L. Rev. 887 at 897-98.
46 Ibid. at 339, tbl. 7. See also Perrin et al., supra note 33 at 728, tbl. 6.
47 Heffernan & Lovely, ibid. at 348 (indicating that in 15 per cent of cases police indicated that they
would proceed with search in presented scenario despite their determination that it was prohibited).
48 See Skolnick, supra note 36 at 220; “The Effect of Mapp v. Ohio on Police Search-and-Seizure
Practices in Narcotics Cases”, Comment (1968) 4 Colum. J.L. & Soc. Probs. 87 at 99-101.
49 See Orfield, “Deterrence, Perjury”, supra note 43 at 82-83; Christopher Slobogin, “Testifying:
Police Perjury and What To Do About It” (1996), 67 U. Colo. L. Rev. 1037 at 1057-59; Gabriel J.
Chin & Scott C. Wells, “The ‘Blue Wall of Silence’ as Evidence of Bias and Motive to Lie: A New
Approach to Policy Perjury” (1998), 59 U. Pitt. L. Rev. 233 at 248-50; Teny v. Ohio, 392 U.S. 1 at 14
(1968); Oaks, supra note 33 at 683-84.
50 See Dripps, “The Case”, supra note 22 at 15 (“[B]asic assumptions about rational behavior
suggest that the rule ought to deter, just as the admission of evidence of subsequent remedial measures
in tort suits is thought-without empirical support of any kind-to deter needed repairs.”)
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
efforts.” It also helps to influence behaviour indirectly by shaping and reinforcing
norms of legality and propriety. Legal prohibitions influence behaviour not simply by
ensuring that the costs of illegality outweigh its benefits, 2 but also by influencing the
development of internal moral codes, shaping preferences, and providing a basis for
informal community and peer group sanctions. 3 As Heffernan and Lovely explain,
“[e]xclusion provides officers with a day-to-day reminder of the importance of
adherence to the law.”54
In summary, we can confidently conclude that as a general matter the exclusion
of unconstitutionally obtained evidence has an appreciable impact on constitutional
compliance.5” We can also say with assurance, however, that the exclusionary remedy
does not deter all violations. It is particularly unlikely to deter when police believe
that their conduct is legal, give less priority to obtaining a conviction than to some
other law enforcement objective, or when they believe that evidence may be admitted
despite the violation.
The social costs of exclusion are easier to calculate. Almost all studies of the
American exclusionary rule have found that the proportion of convictions lost due to
51 The empirical literature supports the intuition that evidentiary exclusion induces police to increase
training efforts. See Orfield, “Exclusionary Rule and Deterrence”, supra note 43 at 1036-48; Craig M.
Bradley, “The ‘Good Faith Exception’ Cases: Reasonable Exercises in Futility” (1985) 60 Ind. L.J.
287 at 291-92; Murphy, supra note 40 at 941; Heffeman & Lovely, supra note 42 at 337-38; Wayne
R. LaFave, “Controlling Discretion by Administrative Regulations: The Use, Misuse, and Nonuse of
Police Rules and Policies in Fourth Amendment Adjudication” (1990) 89 Mich. L. Rev. 442 at 448.
Empirical studies also confirm the intuition that greater training leads to greater compliance. See
Heffernan & Lovely, supra note 42 at 336-37.
52 There is skepticism among some criminal law scholars regarding the deterrent effect of criminal
sanctions. See e.g. David M. Paciocco, Getting Away with Murder: The Canadian Criminal Justice
System (Toronto: Irwin Law, 1999) at 28-35 [Paciocco, Getting Away with Murder]. But among
empirical researchers there is a strong consensus that, on the whole, punishments exert a substantial
deterrent effect. The real question is whether any particular policy innovation (typically increased
enforcement or penalties) will add measurably to the existing preventive effect. See David J. Pyle, The
Economics of Crime and Law Enforcement (New York: St. Martin’s, 1983) at 55, 57-58; Daniel
Nagin, “Criminal Deterrence Research at the Outset of the Twenty-First Century” (1998) 23 Crime &
Just. 1 at 1-3.
53 See generally Johannes Andenaes, Punishment and Deterrence (Ann Arbor: University of
Michigan Press, 1974) at 110-26; Kenneth G. Dau-Schmidt, “An Economic Analysis of the Criminal
Law as a Preference-Shaping Policy” [1990] Duke L.J. 1; Paul H. Robinson & John M. Darley, “The
Utility of Desert” (1997) 91 Nw. U. L. Rev. 453 at 468-78; Louis Michael Seidman, “Soldiers,
Martyrs, and Criminals: Utilitarian Theory and the Problem of Crime Control” (1984) 94 Yale L.J.
315 at 336-38; Neal Kumar Katyal, “Deterrence’s Difficulty” (1997) 95 Mich. L. Rev. 2385 at 2442-
49; Dan M. Kahan, “Social Meaning and the Economic Analysis of Crime” (1998) 27 J. Legal Stud.
609.54 Heffernan & Lovely, supra note 42 at 351.
55 See Powell, supra note 14 at 492 (“[W]e have assumed that the immediate effect of exclusion will
be to discourage law enforcement officials from violating the Fourth Amendment by removing the
incentive to disregard it.”)
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
evidentiary exclusion is very low.56 Most researchers estimate the rate of lost
convictions to be between one and two per cent.57 Given the large number of criminal
cases, the number of lost convictions in absolute terms is far from insignificant.
Society bears a considerable cost each time a factually guilty defendant is set free
is excluded. 8 That cost, however,
because unconstitutionally obtained evidence
should not be exaggerated. The criminal is rarely set free because the constable
blundered.59
It seems, then, that unlike the condonation and corrective justice rationales, the
deterrence rationale is a good justification for excluding unconstitutionally obtained
evidence. The American experience teaches us that exclusion exerts a significant
deterrent effect and generates few lost convictions. The next question is whether there
are any alternatives to exclusion that are likely to achieve an even better balance
between deterring constitutional violations and convicting the factually guilty. If there
are, then despite the advantages of the exclusionary remedy we would be better off
interpreting section 24(2) of the Charter in a manner that would rarely result in
exclusion.
56 A General Accounting Office study found that successful suppression motions were made in only
1.3 per cent of federal cases in Chicago. Dismissals or acquittals were granted in approximately half
of those cases. This does not account, of course, for instances where cases are dropped before formal
charges are laid because prosecutors believe that evidence necessary for conviction would very likely
be excluded. The GAO study found, however, that only 0.4 per cent of cases were declined for this
reason. The researchers thus concluded that the costs of the exclusionary rule in terms of lost
convictions were minimal. See General Accounting Office, supra note 42 at 10-14. A similar study
conducted by the National Institute of Justice (“NIU) found, however, that 4.8 per cent of all felony
arrests were dropped due to tainted evidence. National Institute of Justice, supra note 42 at 1. The NIJ
study was criticized by Davies, who re-analyzed the data and concluded that only 0.8 per cent of
felony arrests were rejected on the basis of illegal searches. He found that the proportion of lost cases
was between 0.3 and 2.35 per cent. See Davies, supra note 42. More recent studies have confirmed
these findings. See Nardulli, “Empirical Assessment”, supra note 42 at 606-607 (study of 7,500 felony
cases in three states finding that less than 0.69 percent of cases were lost due to exclusion of physical
evidence); Van Duizend, Sutton & Carter, supra note 42 at 119 (“a properly administered and
supervised search warrant review process can protect privacy and property rights without significantly
interfering with the ability of police officers to conduct thorough and effective investigations of
criminal activity.”); Uchida & Bynum, supra note 43 at 1064-66 (only 1.5 per cent of search warrant
cases lost due to exclusion); Peter F. Nardulli, “The Societal Costs of the Exclusionary Rule
Revisited” [1987] U. Ill. L. Rev. 223 at 234 (study of 2,759 Chicago cases found that approximately
1.77 per cent were lost due to exclusion).
57 See sources cited ibid.
58 The exclusionary rule may also produce other social costs, such as the wasted resources generated
by meritless suppression motions, the promotion of police perjury, and the narrowing of constitutional
rights to avoid exclusion. See Caldwell & Chase, supra note 33 at 49-54; Perrin et al., supra note 33 at
677-78. I discuss the latter two “costs” at infra note 78 and accompanying text and infra notes 139-40
and accompanying text, respectively.
59 I allude to Justice Cardozo’s famous statement in People v. Defore, 150 N.E. 585 at 587 (N.Y),
cert. denied 270 U.S. 657 (1926) (“The criminal is to go free because the constable has blundered.”)
MCGILL LAw JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
B. Alternatives to Exclusion
Few critics of the exclusionary rule argue that the constitutional rules regulating
police investigative practices should not be enforced. Instead, they maintain that other
mechanisms are as capable of deterring police overreaching as exclusion.6′ Most
recognize, however, that conventional criminal, civil, and administrative remedies
have not been effective in deterring misconduct.61 They have therefore proposed new
schemes (or reforms to existing ones) that would provide stronger incentives for
police to refrain from overreaching.62 These alternative schemes, their proponents
claim, would achieve at least as much deterrence as the exclusionary rule without
producing lost convictions.63
There is little evidence, however, that the legislative and administrative will that
would be required to implement these alternatives is presently strong.’ Efforts to
restrict police power rarely generate political or popular support.65 Enforcing the
rights of criminal suspects is an intrinsically countermajoritarian exercise,66 primarily
because the victims of police overreaching are disproportionately members of
disfavoured minorities.67 Evidentiary exclusion does not require political innovation
and is effected by officials who are relatively insulated from majoritarian political
pressures.68
60 See e.g. Caldwell & Chase, supra note 33 at 67-68; Posner, “Rethinking”, supra note 28.
61 See Oaks, supra note 33 at 673-74; Silas J. Wasserstrom & Louis Michael Seidman, “The Fourth
Amendment as Constitutional Theory” (1988) 77 Geo. L.J. 19 at 83; Caleb Foote, “Tort Remedies for
Police Violations of Individual Rights” (1955) 39 Minn. L. Rev. 493; Stewart, supra note 29 at 1387-
88. But see Posner, “Rethinking”, supra note 28 at 58-62 (arguing that with modest reforms,
conventional tort law remedies are sufficient to optimally deter fourth amendment violations).
62 See e.g. Perrin et al., supra note 33 at 744-53; Akhil Reed Amar, “Fourth Amendment First
Principles” (1994) 107 Harv. L. Rev. 757 at 811-19; Bivens, supra note 33 at 421-24, Burger C.J.,
dissenting.
63 See sources cited ibid.
64 See Carol S. Steiker, “Second Thoughts About First Principles” (1994) 107 Harv. L. Rev. 820 at
848-50; Laurie L. Levenson, “Administrative Replacements: How Much Can They Do?” (1999) 26
Pepp. L. Rev. 879 at 882; Dripps, “The Case”, supra note 22 at 2.
65 See Steiker, ibid. at 850-51.
66 See Antonio Lamer, “Protecting the Administration of Justice From Disrepute: The Admissibility
of Unconstitutionally Obtained Evidence in Canada” (1998) 42 St. Louis U. L.J. 345 at 354-55.
67 See Steiker, supra note 64 at 850; Morgan Cloud, “Pragmatism, Positivism, and Principles in
Fourth Amendment Theory” (1993) 41 UCLA L. Rev. 199 at 284; Wasserstrom & Seidman, supra
note 61 at 94; Richard V. Ericson, Reproducing Order: A Study of Police Patrol Work (Toronto:
University of Toronto Press, 1982) at 200-201; Commission on Systemic Racism in the Ontario
Criminal Justice System, Report (Toronto: Queen’s Printer, 1995); Julian V. Roberts & Anthony N.
Doob, “Race, Ethnicity and Criminal Justice in Canada” in Michael Tonry, ed., Ethnicity, Crime and
Immigration: Comparative and Cross-National Perspectives, vol. 21 (Chicago: University of Chicago
Press, 1997) 469 at 519; David M. Tanovich, “Using the Charter to Stop Racial Profiling: The
Development of an Equality-Based Conception of Arbitrary Detention” (2002) 40 Osgoode Hall L.J.
145; R. v. Brown (2003), 64 O.R. (3d) 161 at para. 9, 173 C.C.C. (3d) 23.
68 See generally Leon, supra note 12 at 930, Brennan J., dissenting.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
Even if alternatives to exclusion could be implemented without political initiative
(for example, by substituting non-exclusionary section 24(1) remedies for exclusion
in
under section 24(2)),69 no alternative is likely to be superior to exclusion
optimizing the balance between deterrence and truth-seeking. Non-exclusionary
remedies are very likely to generate either too little or too much deterrence. To avoid
underdeterrence, alternative remedies must impact police interests severely enough to
influence their future conduct. Most Charter violations would warrant only modest
compensatory damages.” Few victims would find it worthwhile to incur the costs
required to obtain these awards.7 As a result, police would likely consider damage
awards a minor cost of doing business.72 In theory, this problem can be overcome by
the use of such mechanisms as class actions, administrative hearings, and non-
compensatory remedies (such as punitive damages, statutory liquidated damages, and
injunctions).7 But such initiatives would require significant legislative or judicial
innovation, and it is not clear that they would be financially, administratively, or
politically feasible.74
Any scheme attempting to avoid the underdeterrence problem, moreover, would
likely overdeter by chilling legitimate investigative methods.75 As discussed, the
complexity of the law governing investigative practices generates high rates of
inadvertent error. Police make two types of unintentional mistakes. The first occurs
when they incorrectly believe that an intrusion is legal and proceed to collect
evidence illegally. The second occurs when police incorrectly believe that an
intrusion is illegal and therefore forego an opportunity to lawfully obtain evidence.
The first type of error is likely to be common under both exclusionary and non-
exclusionary schemes.76 The second type of error, in contrast, is much more likely to
69 See J.A.E. Pottow, “Constitutional Remedies in the Criminal Context: A Unified Approach to
Section 24 (Part LI)” (2000) 44 Crim. L.Q. 223. See generally Stuart, Charter Justice, supra note 1 at
458-61.
70 See Daniel J. Meltzer, “Deterring Constitutional Violations by Law Enforcement Officials:
Plaintiffs and Defendants as Private Attorneys General” (1988) 88 Colum. L. Rev. 247 at 284.
71 Other barriers to this type of litigation stem from the fact that victims of police overreaching are
often poor, ignorant of their rights, fearful of police reprisals, and incarcerated. Ibid.
72 See Heffeman & Lovely, supra note 42 at 363.
73 See Perrin et al., supra note 33 at 744-5 1; Amar, supra note 62 at 814-16; Posner, “Rethinking”,
supra note 28 at 62-64.
74 See Levenson, supra note 64.
75 See Stewart, supra note 29 at 1399 (“Unless the administrative sanctions imposed on officers for
violating the fourth amendment are harsh, officers may reasonably decide to resolve close questions-
and perhaps less close questions as well-against compliance with constitutional requirements. And,
if administrative sanctions are too harsh, officers may refrain from beneficial law enforcement
activities.”)
76 It is possible that a sufficiently vigorous non-exclusionary remedial scheme could prompt police
to significantly improve their training efforts. But it is very likely that the error rate would still be
substantial. See Heffernan & Lovely, supra note 42 at 362 (“Even with substantial training, our results
suggest that police officers are likely to be mistaken about a quarter of the time about the legality of
intrusions where specific rules of search and seizure are involved.”)
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
occur under a non-exclusionary regime. Consider what is likely to happen when
police believe that an intrusion is probably legal, but are not certain. If the penalty for
error is evidentiary exclusion, they are likely to proceed with the intrusion, especially
if they believe that there is no other way to obtain the evidence that is more likely to
be found constitutional. The worst that can happen is that a court will decide that they
were wrong and exclude the evidence.77 If the evidence would not likely have been
available in any event, then police have lost nothing. But if the consequences of error
include fines, damage awards, or impediments to career advancement, police may not
be prepared to risk even a modest risk of violation. They may therefore forego legal
intrusions and fail to obtain evidence necessary for conviction.78 Fear of non-
exclusionary sanctions, in other words, may cause police to steer far clear of the
sometimes blurry boundary between legal and illegal conduct, prompting them to use
only those methods that they believe are patently constitutional. Some might applaud
this result, believing that police should not act unless they are absolutely certain of
the legality of their conduct. But given criminal procedure’s inevitable complexity,
such a mindset would likely deter too many legal intrusions.
The overdeterrence problem could be lessened to some degree by granting police
immunity for “good faith” errors (i.e., violations committed when police were
reasonably confident of the legality of their conduct). But this would severely
undercut the capacity of sanctions to promote training, which we know significantly
increases constitutional compliance.79 If police know that an honest belief in the
legality of an intrusion immunizes them from potential sanctions, then they will be
inclined to remain ignorant of their constitutional obligations. Exclusion, in contrast,
can be used to sanction inadvertent errors (thereby increasing compliance) without
fear of chilling legitimate tactics. The exclusionary rule gives police an incentive to
learn as much law as possible; but it does not dissuade them from taking reasonable,
calculated risks in the face of legal uncertainty. What is often characterized as a
weakness of the exclusionary rule-its inability to deter marginal violations-turns
out to be one of its strengths. Given the unfortunate inevitability of complexity-
induced error, we are better off with a remedy that leaves a few marginal violations
undeterred than with one that is likely to either grossly under or overdeter.
Alternative sanctions, then, face daunting obstacles of implementation and
optimization. But let us imagine, if only for the sake of argument, that we could
implement an alternative regime that deters constitutional violations at least as well as
exclusion without deterring an unacceptable proportion of legal intrusions. Would it
really represent much of an advance over the exclusionary remedy? Alternative
77 Empirical evidence confirms that police are very unlikely to face civil or criminal liability or
internal discipline for committing non-obvious and non-egregious constitutional violations. See
Heffernan & Lovely, supra note 42 at 328-29, 350.
78 See Loewenthal, supra note 43 at 31-32; Totten, Kossoris & Ebbesen, supra note 45 at 908;
Heffernan & Lovely, supra note 42 at 362-63; Dripps, “The Case”, supra note 22 at 17; Jull, supra
note 5 at 539.
79 See supra note 53 and accompanying text.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
sanctions promise to deter overreaching without diminishing the probability of
convicting the factually guilty. But as many jurists have noted, if alternative sanctions
deterred effectively, then police would never discover evidence that could only have
been obtained illegally.8″ Just as with the exclusionary rule, the deterrent impact of the
sanctions would cause factually guilty suspects
to escape conviction. A non-
exclusionary regime would generate fewer lost convictions than exclusion, of course,
because even strong sanctions cannot deter all violations. But the quantum of this
benefit is only marginal. As discussed, the empirical record indicates that very few
convictions of the factually guilty are lost because of evidentiary exclusion.” Given
the weaknesses of alternative sanctions, attempting to reap this meagre benefit is not
worth the effort. The unfortunate reality is that protecting constitutional rights entails
significant social costs.82 Roughly speaking, the relationship between constitutional
compliance and the acquisition of evidence to convict the factually guilty is zero-sum.
Non-exclusionary remedies may foster less popular discontent with the criminal
justice system by better “hiding” the costs of constitutional compliance, but they do
not eliminate them.83 More compliance equals less evidence, no matter what the
remedy for violations.
that provided
the
regime
Alternative sanctions are not the answer. While we could in theory imagine a
non-exclusionary
for optimal
deterrence,84 in practice the obstacles to achieving such a result are far too great.
Despite its flaws, the exclusionary remedy gets us closer to optimal deterrence than
any alternative is likely to do. Admittedly, exclusion cannot deter when the purpose
of the intrusion is not to obtain admissible evidence. Non-exclusionary sanctions
should be available to deter these intrusions and compensate victims, but as an
adjunct to-not a replacement for-evidentiary exclusion.85 Exclusion also cannot
incentives necessary
80 See Richard A. Posner, “An Economic Approach to the Law of Evidence” (1999) 51 Stan. L.
Rev. 1477 at 1533 (If alternative sanctions deterred violations, “there would not be any fruits, and so
there would be no net gain from the standpoint of accuracy in adjudication.”); Yale Kamisar,
“‘Comparative Reprehensibility’ and the Fourth Amendment Exclusionary Rule” (1987) 86 Mich. L.
Rev. 1 at 47-48 (“A society whose officials obey the fourth amendment in the first place (because of
an effective tort or other ‘direct alternative’ remedy) ‘pays the same “price’ as the society whose
officials cannot use the evidence they acquired because they obtained it in violation of the fourth
amendment. Both societies convict fewer criminals.”); Stewart, supra note 29 at 1392 (noting that
while the exclusionary rule often deprives courts of relevant evidence, that evidence frequently
“would not have been obtained had the police officer complied with the demands of the fourth
amendment in the first place.”)
81 See supra notes 58-61 and accompanying text.
82 See Steiker, supra note 67 at 820; Stewart, supra note 29 at 1393; Leon, supra note 12 at 941,
Brennan J., dissenting; Jul], supra note 5 at 526.
83 See Caldwell & Chase, supra note 33 at 51-52; Kamisar, supra note 80 at 47.
84 See e.g. Posner, “Rethinking”, supra note 28 at 58-62.
85 See Stewart, supra note 29 at 1396 (though the exclusionary rule is “powerless to deter invasions
of constitutionally guaranteed rights where the police either have no interest in prosecuting or are
willing to forego successful prosecution in the interest of serving some other goal” this “does not
suggest that the rule is not a necessary remedy, only that it is not a sufficient one.”)
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
deter when police are prepared to lie in an effort to avoid exclusion. But neither can
alternative remedies. Police facing the prospect of monetary or professional sanctions
are at least as likely to falsely deny misconduct than those facing the possibility of
evidentiary exclusion. 6 Lastly, exclusion cannot deter when police incorrectly believe
that their conduct is legal. But again, neither can alternative remedies. Consequently,
in developing an interpretive framework for section 24(2) of the Charter, courts
should not rely on the capacity of non-exclusionary remedies to deter violations. If
we want police to respect the Charter, we must be prepared to live with the regular
exclusion of unconstitutionally obtained evidence and the occasional acquittal of the
factually guilty.
Ill. Deterrence and Section 24(2) of the Charter
I have argued to this point that deterrence is the only justification for excluding
unconstitutionally obtained evidence; that the empirical record shows that the
deterrent effect of exclusion is significant but limited; and that alternative methods of
inducing police to respect constitutional guarantees are neither feasible nor desirable.
What would section 24(2) doctrine look like if it respected these conclusions? Put
simply, it would require judges to: (i) exclude only when doing so is likely to have a
significant deterrent effect; and (ii) exclude in every such case. In other words,
evidence should never be excluded pursuant to section 24(2) for any reason other
than deterrence. Conversely, it should be an error for a judge to admit evidence on the
basis that deterrence could be achieved by other means or that deterrence should be
sacrificed to achieve some other objective.
As I discuss in the remainder of this article, the Supreme Court of Canada’s
section 24(2) jurisprudence strays considerably from
these dictates. But the
foundations of a deterrence-based approach are there. It is possible, without doing
intolerable violence to either constitutional text or stare decisis, to strip away that
which is incompatible with deterrence and begin building an exclusionary doctrine
that pursues an optimal accommodation between our need to prevent constitutional
violations and our need to convict the guilty. I elaborate this argument below in the
course of examining
the jurisprudence associated with section 24(2)’s key
components, which require applicants to establish that: (i) they have standing to bring
the application; (ii) the evidence that they are seeking to exclude was “obtained in a
manner” that violated a Charter right; and (iii) the admission of the evidence could
“bring the administration of justice into disrepute.”
86 See Levenson, supra note 64 at 881 (“There is absolutely no evidence that a police officer will be
any less motivated to lie in an administrative hearing, where their reputation and job position are at
risk, than in a criminal proceeding where the court threatens to exclude evidence.”)
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
A. Standing
Section 24(2) applications occur in the context of proceedings under section
24(1) of the Charter. Section 24(1) states that persons whose Charter rights have been
infringed may apply for a remedy. The Supreme Court of Canada has interpreted this
to mean that defendants may only seek exclusion if their rights were denied, and not
if the evidence was obtained by violating the rights of third parties.87 As a matter of
plain language interpretation this conclusion may be inevitable. But the Court’s
holding limits the capacity of the exclusionary remedy to deter investigative abuses,
especially in search and seizure cases.8″ For example, police who believe that
evidence against a suspect might be obtained at a location where that suspect has no
reasonable expectation of privacy may with relative impunity commit egregious
violations of the privacy of persons who do have such an expectation.89
There may be ways to surmount this problem. One is to interpret Charter rights
expansively to protect the privacy and security of persons without a direct proprietary
interests in the location searched.9 Another is to find that admitting evidence
obtained by violating a non-defendant’s rights would be “unfair” or constitute an
abuse of process.91 Whatever method is used, courts should have some means of
excluding evidence to deter police from violating the rights of innocent third parties.
As Justice La Forest has noted, to focus exclusively on the rights of the accused is “to
accord greater protection to the right of privacy to the accused or other wrongdoer
than to a person against whom there may be no reasonable suspicion of
wrongdoing.”92 Though section 24’s wording may limit its remedial scope to
violations of applicants’ rights, the purpose of exclusion is not to achieve corrective
87 See R. v. Wijesinha, [1995] 3 S.C.R. 422 at para. 66, 127 D.L.R. (4th) 242; Edwards, supra note
26 at paras. 45, 51-56.
88 See Roach, Constitutional Remedies, supra note 11 at paras. 10.450-10.570. See also Jonathan
Dawe, “Standing to Challenge Searches and Seizures Under the Charter. The Lessons of the
American Experience and Their Application to Canadian Law” (1993) 52 U.T. Fac. L. Rev. 39 at 68-71.
89 See generally Edwards, supra note 26, La Forest J., concurring.
90 See R. v. Belnavis, [1997] 3 S.C.R. 341, 34 O.R. (3d) 806 [Belnavis cited to S.C.R.], La Forest J.,
dissenting; Dawe, supra note 88 at 53-56, 60-61.
91 See Ursula Hendel & Peter Sankoff, “R. v. Edwards: When Two Wrongs Just Might Make a
Right” (1996) 45 C.R. (4th) 330. The Supreme Court has repeatedly held that evidence that is not
obtained in a manner that violated a Charter right may be excluded where admitting it would be
“unfair”. The Court has variously described the source of this exclusionary discretion as arising from
the common law and sections 7, 1 l(d), and 24(1) of the Charter. See R. v. Harrer, [1995] 3 S.C.R. 562
at paras. 21-24, 128 D.L.R. (4th) 98 (recognizing common law discretion-constitutionalized by
section 11 (d) of the Charter-to exclude evidence where admission would undermine right to a fair
trial); R. v. Terry, [1996] 2 S.C.R. 207 at para. 25, 135 D.L.R. (4th) 214 (abusively-obtained evidence
may be excluded under sections 7 or 11(d) of the Charter); R. v. White, [1999] 2 S.C.R. 417 at paras.
86-89, 174 D.L.R. (4th) 111 (evidence excluded under section 24(1) where admission would violate
self-incrimination principle protected by section 7); Buhay, supra note 18 at para. 40 (“[E]ven in the
absence of a Charter breach, judges have a discretion at common law to exclude evidence obtained in
circumstances such that it would result in unfairness if the evidence was admitted at trial.”)
92 Edwards, supra note 26 at para. 64, La Forest J., concurring.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
justice for criminals. Rather, it is to deter violations of rights generally. If this cannot
be accomplished under the aegis of section 24(2), then courts should employ other
means.
B. “Obtained in a Manner”
In addition to establishing standing, defendants seeking exclusion under section
24(2) must also demonstrate that the evidence was “obtained in a manner” that
infringed one of their Charter rights.” Applicants do not need to establish a strict
causal connection between the violation and the discovery of the evidence; it is
sufficient if the violation and discovery are part of “a single transaction”.94 There
need only be, in other words, a non-remote, temporal connection between the two
events.95 The mere presence of a causal connection, however, may not be sufficient.96
Courts must consider the presence and strength of both temporal and causal
connections in determining “on a case-by-case basis” whether the discovery of the
evidence was linked closely enough to the violation to justify exclusion.97
If the purpose of exclusion is to deter constitutional infringements, then the Court
has been wise to permit exclusion in the absence of a causal connection between
violations and evidentiary fruits. Insisting on causation would blunt section 24(2)’s
deterrent force. The Court noted in Strachan, for example, that a causation
requirement would in most cases prevent the exclusion of physical evidence obtained
after a violation of the section 10(b) right to counsel.98 Consider the following
scenarios, assuming that applicants must prove causation to exclude evidence under
section 24(2). Police conducting a search of a suspect’s residence for physical
evidence believe that there is also a reasonable prospect of obtaining a confession.
Police in this scenario have a strong incentive to comply with section 10(b). If they
did not, then a court would likely find that any confession obtained was causally
related to the violation. The suspect would be able to show, in other words, that she
might not have confessed had the police properly cautioned her.99 Now suppose that
93 R. v. Therens, [1985] 1 S.C.R. 613 at 648, 18 D.L.R. (4th) 655 [Therens cited to S.C.R.].
94 R. v. Strachan, [1988] 2 S.C.R. 980 at 1005, 56 D.L.R. (4th) 673 [Strachan cited to S.C.R.].
95 Ibid. at 1005-1006. See also Therens, supra note 93 at 649, Le Dain J., dissenting; Kokesch, supra
note 18 at 19; R. v. Grant, [1993] 3 S.C.R. 223 at 254-55, 84 C.C.C. (3d) 173 [Grant cited to S.C.R.];
R. v. Wiley, [1993] 3 S.C.R. 263 at 278, 84 C.C.C. (3d) 161 [Wiley cited to S.C.R.]; R. v. Plant, [1993]
3 S.C.R. 281 at 299, 145 A.R. 104 [Plant cited to S.C.R.]; R. v. Bartle, [1994] 3 S.C.R. 173 at 208,
118 D.L.R. (4th) 83 [Bartle cited to S.C.R.]; R. v. Goldhart, [1996] 2 S.C.R. 463 at paras. 33-36, 136
D.L.R. (4th) 502 [Goldhart cited to S.C.R.].
96 See Goldhart, ibid.
97 Ibid. at para. 40.
98 Strachan, supra note 94 at 1003-1005.
99 The Court has been very reluctant to find that self-incriminating evidence would have been
obtained even if police had complied with section 10(b), casting the burden to establish this on the
prosecution. See Bartle, supra note 95 at 211-13; R. v. Harper, [1994] 3 S.C.R. 343 at 354, 118 D.L.R.
(4th) 312.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
police either have no need for a confession or do not believe they can obtain one.
They would consequently be free to deny the suspect access to counsel in order to
isolate, intimidate, or harass her in the course of conducting the search.’ The abuse
would not have in any way contributed to the discovery of the evidence.
Requiring defendants to prove causation would also encourage police to violate
the Charter to obtain evidence in support of warrant applications. In Grant, Wiley, and
Plant, police used information obtained from unconstitutional perimeter searches to
obtain warrants to search residences for illegal drugs.’ ‘ In each case, however, the
Court determined that the warrants could have been issued absent the tainted
evidence.” 2 As police could have obtained the drugs without violating the Charter, a
“but for” causation requirement would have likely precluded exclusion. Police
collecting evidence in support of warrant applications would consequently have little
reason
to forego constitutionally dubious techniques. If an issuing justice or
reviewing court determines that the technique violated the Charter, then police would
be in no worse a position than they would have been had they not used it.
This analysis also suggests that courts should be reluctant to conclude that
temporal or causal connections between violations and the obtaining of evidence are
too “remote” to engage section 24(2). A broad interpretation of the “obtained in a
manner” requirement allows courts to sanction and deter misconduct that does not
directly or immediately lead to the discovery of evidence. There may, however, be a
point at which the connection is so tenuous that exclusion may have little deterrent
effect. If violations are only remotely related to the obtaining of evidence, then police
may not perceive that they are being sanctioned for specific transgressions. They may
consequently view exclusion as arbitrary and fail to adjust their investigative
practices. Excluding evidence unconnected to overreaching may also lead to a failure
of marginal deterrence by diminishing the incentive of police to comply with the
Charter after they become aware that they have committed a violation. In R. v.
Upston, 3 for example, police failed to inform a suspect of his section 10(b) rights
immediately after detaining him. They did so later, however, and he subsequently
confessed. The Court determined that the statement was not “obtained in a manner”
that infringed the Charter because there was no causal connection between the
violation and the confession.”4 A better justification for the decision is that exclusion
would have diminished section 24(2)’s capacity to deter subsequent constitutional
100 A causal connection requirement would also do little to dissuade police from conducting
intrusive, unconstitutional searches in circumstances where there is little expectation of discovering
evidence, such as routine strip searches of suspects arrested for impaired driving. See generally R. v.
Flintoff(1998), 16 C.R. (5th) 248 at paras. 28-34, 126 C.C.C. (3d) 321 (Ont. C.A.).
101 Grant, supra note 95; Wiley, supra note 92; Plant, supra note 95.
102 Grant, ibid. at 253-54; Wiley, ibid. at 277; Plant, ibid. at 299.
103 11988] 1 S.C.R. 1083, 63 C.R. (3d) 299.
104 This holding is inconsistent with the Court’s later decision in Strachan, supra note 94, which
held that exclusion does not always require defendants to establish a causal connection between the
violation and the discovery of the evidence.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
violations. If police had believed that their initial violation would have resulted in the
exclusion of any subsequently obtained evidence, they would have had little reason to
belatedly comply with section 10(b).
It is difficult to state precisely when a temporal or causal connection between a
violation and the obtaining of evidence is so tenuous that exclusion will not likely
deter future misconduct. But focusing on the likely impact of exclusion on police
behaviour both before and after trial promises to provide more concrete guidance to
courts than relying solely on the evasive concept of remoteness.
C. Bringing the Administration of Justice into Disrepute
Once defendants establish that evidence was obtained in manner that infringed
one of their Charter rights, they must then prove that its admission could “bring the
administration of justice into disrepute.”‘0 5 As is well known, the Supreme Court has
instructed trial judges to consider three sets of factors in making this determination:
those relating to the fairness of the trial, the seriousness of the violation, and the effect
of exclusion on the repute of the administration of justice.0 6 If admitting the evidence
would threaten trial fairness, then the evidence will generally be excluded.0 7 If it
would not, then admissibility turns on a balancing of factors in the latter two
categories. Courts must weigh the seriousness of the violation against the seriousness
of the crime and the importance of the evidence.” 8 The outcome of this balancing
process is difficult to predict, but in many cases results in admission.0 9
105 Though the English version states “would bring the administration of justice into disrepute,” in
Collins (supra note 6 at 287) the Court determined that the correct translation of the French version of
the provision [“est susceptible de diconsidgrer l’administration de la justice”] is “could bring the
administration of justice into disrepute” [emphasis added].
106 See Collins, supra note 6; R. v. Law, [2002] 1 S.C.R. 227 at para. 33, 245 N.B.R. (2d) 270 [Law
cited to S.C.R.].
107 See Collins, supra note 6 at 284; R. v. Broyles, [1991] 3 S.C.R. 595 at 619, 120 A.R. 189
[Broyles]; Stillman, supra note 18 at paras. 72, 110, 118-19. See also Stuart, Charter Justice, supra
note 1 at 493, 507. In many cases the Court has gone on to consider the seriousness of the violation
and the effect of exclusion even after concluding that admitting the evidence would compromise trial
fairness. In only a few of these cases has the Court admitted the evidence.
108 See Buhay, supra note 18 at paa. 51; David M. Paciocco, “Stillman, Disproportion and the Fair
Trial Dichotomy under Section 24(2)” (1997) 2 Can. Crim. L. Rev. 163 at 172-73 [Paciocco, “Fair
Trial Dichotomy”]; Roach, “Evolving”, supra note 21 at 119; David M. Paciocco & Lee Stuesser, The
Law of Evidence, 3d ed. (Toronto: Irwin Law, 2002) at 300-309.
109 See Don Stuart, “Eight Plus Twenty-Four Two Equals Zero” (1998) 13 C.R. (5th) 50 [Stuart,
“Eight”]; Stuart, Charter Justice, supra note 1 at 504, 513-16. A recent study found that courts
exclude non-conscriptive evidence obtained in violation of sections 8 or 9 of the Charter in
approximately 50 per cent of cases. See Nathan J.S. Gorham, “Eight Plus Twenty-Four Two Equals
Zero-Point-Five” (2003) 6 C.R. (6th) 257.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
1. Evidence Affecting Trial Fairness: Automatic Exclusion
The Court’s
trial
reflects
fairness doctrine
its especial concern
for
unconstitutional self-incrimination. Admitting self-incriminating evidence obtained
from Charter violations is “unfair”, according to the Court, because it violates the
“case-to-meet” principle, which requires that the prosecution establish its case
without
the compelled participation of the accused.”‘ The definition of the
evidentiary category triggering this concern for unfair self-incrimination has shifted
over the years.”‘ It currently consists of evidence that is both “conscriptive” and
“non-discoverable”.” 2 Conscriptive evidence arises only when the accused “is
compelled to incriminate himself at the behest of the state by means of a statement,
the use of the body or the production of bodily samples.””‘ 3 Evidence is “non-
discoverable” if it could not have been obtained by legal, non-conscriptive means.” 4
There are two ways to interpret this doctrine. The conventional approach views
the case-to-meet principle as justifying the exclusion of evidence apart from any
concern for deterring police misconduct. The wrong remedied by exclusion, in other
words, is the admission at trial of unconstitutionally obtained, self-incriminating
evidence.” 5 Deterring misconduct involved in obtaining evidence is the job of the
second group of Collins factors-those relating to the seriousness of the violation.
110 See Burlingham, supra note 18 at para. 145, Sopinka J. (“The participation of the accused in
providing incriminating evidence involving a breach of Charter rights is the ingredient that tends to
render the trial unfair as he or she is not under any obligation to assist the Crown to secure a
conviction.”); R. v. Ross, [1989] 1 S.C.R. 3 at 16,46 C.C.C. (3d) 129 (“Any evidence obtained, after a
violation of the Charter, by conscripting the accused against himself through a confession or other
evidence emanating from him would tend to render the trial process unfair.”)
’11 See Roach, Constitutional Remedies, supra note 11 at paras. 10.430-10.440.
112 Stillman, supra note 18.
113 Ibid. at para. 80. It is fairly clear that the Stillman Court’s reference to the “use of the body”
refers to either the use of a suspect’s body for identification (e.g., for fingerprinting or an identification
lineup) or the extraction of bodily substances, and not to the mere participation of the accused in the
discovery of pre-existing “real” evidence. See especially ibid. at paras. 77, 94, 98, 113. See also
Feeney, supra note 18 at para. 64; R. v. Fliss, [2002] 1 S.C.R. 535 at para. 77; Law, supra note at para.
34; Buhay, supra note 18 at para. 49. See also Peter Cory, “General Principles of Charter Exclusion”
(National Criminal Law Program, July 1997, Halifax) [unpublished] at 12-13. But see R. v. M(MR.),
[1998] 3 S.C.R. 393 at paras. 87-89, 166 D.L.R. (4th) 261, Major J., dissenting. Courts of appeal have
almost unanimously favoured this “narrow” interpretation of Stillman. See e.g. R. v. Davies (1998),
127 C.C.C. (3d) 197 (Y.T.C.A.), leave to appeal to S.C.C. denied, [1998] S.C.C.A. No. 460; R. v.
Lewis (1998), 38 O.R. (3d) 540 at 551-52, 13 C.R. (5th) 34 (Ont. C.A.); R. v. Ellrodt (1998), 130
C.C.C. (3d) 197 (B.C.C.A.); R. v. Richardson (2001), 153 C.C.C. (3d) 449 at paras. 19-22, 43 C.R.
(5th) 371 (B.C.C.A.). These decisions are contra R. v. Young (1997), 34 O.R. (3d) 177, 8 C.R. (5th)
343 (Ont. C.A.). It should also be noted that “conscriptive” evidence includes “derivative” evidence;
that is, evidence that is discovered as a result of conscription: Stillman, supra note 18 at paras. 99-101;
Feeney, supra note 18 at paras. 67, 69-70.
114 See Stillman, ibid. at paras. 108-10; Feeney, ibid. at para. 65.
115 See Law, supra note 106 at para. 34 (“The concept of trial fairness is ultimately concerned with
the continued effects of unfair self-incrimination on the accused…” [emphasis added]).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
The Court has accordingly stated that section 24(2)’s purpose is not only to prevent
constitutional violations but also to “preclude improperly obtained evidence from
being admitted to the trial process when it impinges upon the fairness of the trial.””‘ 6
This conception of the Court’s fair trial test is grounded on the corrective justice
rationale.” 7 It attempts to prevent the harms associated with self-incrimination by
restoring the status quo ante between the suspect and the state.” 8 But as I have
discussed, corrective justifications for exclusion are instrumentally and morally
dubious. When police infringe the Charter to obtain self-incriminating evidence, they
violate the privacy and dignity of both innocent and guilty suspects.”‘ Exclusion is
justified to prevent this. But only factually guilty suspects suffer from any additional
harm generated by the admission of self-incriminating evidence at trial. 2 It may be
psychologically painful to witness the admission of self-incriminating evidence, but
how is it is any more painful than witnessing the admission of any other kind of
inculpatory evidence? Even if we assume that the admission of illegally obtained,
self-incriminating evidence causes factually guilty defendants some measure of
discrete psychological distress, that distress is almost always outweighed by the
state’s interest in securing their conviction. Like other applications of the corrective
justice rationale, this interpretation of the Court’s trial fairness doctrine perversely
values the interests of guilty defendants in avoiding conviction over society’s interest
in securing it.
The second way of reading the Court’s trial fairness doctrine is to view it as
treating constitutional violations
in self-incriminating evidence as
particularly serious and therefore warranting automatic exclusion in order to further
resulting
116 Burlingham, supra note 18 at para. 25.
17 See Roach, “Evolving”, supra note 21 at 123; Roach, Constitutional Remedies, supra note 11 at
para. 10.280.
18 This is why the court has limited the trial fairness category to evidence that is both conscriptive
and non-discoverable. The problem is not simply the unconstitutional conscription of the suspect, it is
the use at trial of the fruits of that conscription that the state could not have obtained by other means.
against false confessions and wrongful convictions, they especially benefit the innocent.
119 To the extent that some Charter rights, such as section 10(b), serve as prophylactic protections
120 It is true that some evidence obtained by unconstitutional conscription, such as that resulting
from coerced confessions, is inherently unreliable. But not all such evidence is unreliable. Some
conscriptive evidence, such as DNA identification evidence derived from bodily samples, may be
especially reliable. Even unconstitutionally obtained statements can be reliable, for example, those
taken after violations of section 10(b) in non-coercive circumstances. So the near automatic
exclusionary rule for evidence affecting trial fairness cannot be justified by reliability concerns. It is
questionable, moreover, whether reliability ought to play any role in section 24(2) determinations. In
most cases, judges can exclude unreliable self-incriminating evidence under the voluntary confession
rule. See Mahoney, supra note 8 at 456-57. Where that rule does not apply, evidence may be excluded
on the basis that its prejudicial effect outweighs its probative value. See R. v Sweitzer, [1982] 1 S.C.R.
949 at 953, 37 A.R. 294; R. v. Corbett, [1988] 1 S.C.R. 670, 41 C.C.C. (3d) 385, La Forest J.,
dissenting on other grounds; R. v. Potvin, [1989] 1 S.C.R. 525, 47 C.C.C. (3d) 289, La Forest J.,
concurring; R. v. Seaboyer, [1991] 2 S.C.R. 577, 83 D.L.R. (4th) 193; R. v. Arp, [1998] 3 S.C.R. 339,
166 D.L.R. (4th) 296.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
deterrence.121 Charter violations that create self-incriminating evidence are not
categorically more serious, however, than those that do not. 22 Some methods of
obtaining self-incriminating evidence are highly invasive of privacy and dignity, such
as interrogations involving physical or psychological coercion and takings of bodily
samples without authorization or cause. But others are not. For example, police may
obtain a confession after infringing a suspect’s right to counsel without employing
any coercive interrogation techniques; indeed, without even questioning the suspect.
This is not to say that section 10(b)’s requirements are not warranted as prophylactics
against abuse and false confessions. Nor does it mean that excluding uncoerced
confessions cannot be justified on deterrence grounds. But the “uncoerced
confession” example does demonstrate that some Charter violations producing self-
incriminating evidence do not cause serious harm. 123 Have police who obtain a breath
sample after neglecting to inform a suspect of duty counsel’s telephone number
committed a more serious wrong than those who burst into a suspect’s residence with
their guns drawn to secure the premises
in anticipation of obtaining a search
warrant? 124 Is a warrantless search involving the plucking of a single hair from the
head of a suspect for DNA analysis more abusive than one involving a rectal search
121 This reading is incompatible, it should be noted, with the Court’s discoverability doctrine. If the
aim of exclusion is to deter constitutional violations, then the fact that the evidence could have been
obtained otherwise is at best irrelevant. See Stuart, Charter Justice, supra note I at 513.
122 See Steven Penney, “What’s Wrong With Self-Incrimination? The Wayward Path of Self-
Incrimination Law in the Post-Charter Era (Part ID” 48(3) Crim. L.Q. [forthcoming in 2004] [Penney,
“Self-Incrimination
(Part I)”]. The Court regularly characterizes violations producing self-
incriminating evidence as serious without any evidence that police acted in an abusive manner. See
Roach, Constitutional Remedies, supra note 11 at paras. 10.1700-1710, 10.1740.
123 Some would argue that violations of the right to counsel are inherently serious, not only because
the right helps to prevent abuse and false confessions, but also because it protects against unwitting
self-incrimination. See e.g. Therens, supra note 93 at 652-53, Le Dain J., dissenting (“[T]he right to
counsel is of such fundamental importance that its denial in a criminal law context must prima facie
discredit the administration of justice.”) This is simply a reiteration, however, of the corrective justice
rationale for exclusion. The violation is serious, on this view, not because the means used to obtain the
evidence are cruel or abusive, but because suspects are faced with the prospect of being convicted by
their “own” evidence. As discussed, this is not a compelling rationale for exclusion. The law permits
unwitting self-incrimination, moreover, in a variety of circumstances, including undercover operations
and electronic surveillance, as well as in permitting suspects to be questioned even if they have only a
“limited cognitive capacity” to understand the right to counsel or are incapable of making a best-
interests decision as to whether to speak (R. v. Whittle, [1994] 2 S.C.R. 914 at 941, 116 D.L.R. (4th)
416) and in permitting police to use considerable psychological pressure, manipulation, and trickery in
interrogations (see R. v. Oickle, [2000] 2 S.C.R. 3, 187 N.S.R. (2d) 201). See generally Penney, “Self-
Incrimination (Part 11)”, ibid.
124 Compare Bartle, supra note 95 (breath sample, taken after police failed to inform suspect of toll-
free telephone number for duty counsel, excluded on basis that its admission would affect trial fairness
and despite the fact that counsel’s advice was of marginal utility in the circumstances) with Silveira,
supra note 18 (police raided suspect’s residence without a warrant and placed his family members
under “house arrest” for over an hour while awaiting the issuance of a warrant; drugs obtained did not
affect trial fairness and the violation was not sufficiently serious to warrant exclusion). See also Stuart,
Charter Justice, supra note 1 at 517.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
for concealed narcotics? Again, deterrence may justify exclusion in each of these
cases. As will be discussed further on, deterrence dictates that certain types of Charter
violations tending to generate self-incriminating evidence should almost always
trigger exclusion, but this result flows solely from the capacity of exclusion to deter
this type of violation, and not from the self-incriminating character of the evidence
produced.
The idea that constitutional violations producing self-incriminating evidence are
intrinsically egregious is belied, moreover, by the existence and constitutionality of
laws compelling self-incriminating information from criminal suspects.’25 Courts
have upheld numerous provisions requiring suspects to submit to takings of bodily
samples and impressions. 26 In these cases, the only questions have been whether the
laws require police to establish adequate grounds for suspicion and conduct the
procedures
in a minimally intrusive manner. If compelled self-incrimination is
acceptable when subject to regulatory safeguards, then how can we conclude that
unconstitutional conscription is so inherently abusive that its evidentiary products
must always be suppressed? The fact that these same judicially upheld legislative
provisions do not provide any evidentiary immunity proves, moreover, that the
admission at trial of compelled, self-incriminating evidence
is not intrinsically
“unfair”.
To summarize, neither deterrence nor any other rationale supports the automatic
exclusion of self-incriminating evidence under section 24(2) of the Charter. Further,
there is no reason for courts to even consider whether evidence is “self-incriminating”,
“conscriptive”, or “discoverable”.’27 The Court’s trial fairness analysis is unjustifiable
125 See Paciocco, “Fair Trial Dichotomy”, supra note 108 at 169; Pottow, “Unified Approach”,
supra note 8 at 54; Steven Penney, “What’s Wrong With Self-Incrimination? The Wayward Path of
Self-Incrimination Law in the Post-Charter Era (Part LI)” 48(4) Crim. L.Q. [forthcoming in 2004].
126 The Criminal Code’s impaired driving provisions give police power to demand breath and in
some circumstances blood samples for alcohol analysis. It is an offence to fail to comply with any of
these demands. See Criminal Code, R.S.C. 1985, c. C-96, s. 254. The Supreme Court of Canada has
rejected attacks on these provisions based on Charter sections 8, 9, and 10(b). See R. v. Thomsen,
[1988] 1 S.C.R. 640, 40 C.C.C. (3d) 411; R. v. Hufsky, [1988] 1 S.C.R. 621, 40 C.C.C. (3d) 398. The
Ontario Court of Appeal has recently rejected the argument that roadside alcohol screening infringes
section 7’s protection against self-incrimination: R. v. Thompson (2001), 52 O.R. (3d) 779, 151 C.CC.
(3d) 339. The Supreme Court has also upheld the mandatory fingerprint and photograph identification
provisions in section 2 of the Identification of Criminals Act, R.S.C. 1985, c. I-1. See R. v Beare,
[1988] 2 S.C.R. 387, 55 D.L.R. (4th) 481. It has also recently been decided that the Criminal Code’s
DNA warrant provisions, which authorize police to compel suspects to provide bodily samples for
identification purposes do not violate the Charter. See R. v. S.A.B., 2003 SCC 60.
127 As discussed, the Court’s discoverability doctrine hinges on the corrective justice rationale and is
therefore unsustainable. As I discuss at infra note 164 and accompanying text, the fact that evidence
could have been obtained legally should in some circumstances militate in favour of exclusion. The
scrapping of the trial fairness category also eliminates the need to determine whether facially non-
conscriptive evidence is causally “derived” from conscriptive evidence such that it should be
considered to affect trial fairness. In the absence of trial fairness considerations, the question of
whether evidence is “derivative” is subsumed into the “obtained in a manner” inquiry.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
and should therefore be scrapped. Self-incriminating evidence should be treated the
same as all evidence and excluded only to further deterrence.
2. Evidence Not Affecting Trial Fairness: The Balancing Approach
As mentioned, when trial fairness is not implicated, the Supreme Court of Canada
has instructed trial courts to decide section 24(2) applications by weighing the
seriousness of the violation against the negative impact of exclusion on the reputation
of the justice system. The more serious the violation, the more likely it is that its
evidentiary fruits will be excluded. Conversely, the more serious the crime and the
more important the evidence to securing a conviction, the more likely it is that the
evidence will be admitted. Among these factors, the seriousness of the violation is the
most weighty. 28 The assessment of the violation’s seriousness turns primarily on
whether it “was committed in good faith, or was inadvertent or of a merely technical
nature, or whether it was deliberate, wilful or flagrant.” ‘129 Other considerations
include whether there was an urgent need to preserve evidence,13 whether the
evidence could have been obtained legally, 131 and the extent to which the violation
unjustifiably invaded the suspect’s privacy.13 Factors such as offence seriousness and
evidentiary importance have received less jurisprudential attention and likely play a
secondary role in deciding admissibility. Some commentators have even suggested that
they do no real work in section 24(2) determinations. 3 3 Nonetheless, courts regularly
consider these factors and they likely make a difference in at least some cases.134
128 See Stuart, Charter Justice, supra note 1 at 499.
129 Therens, supra note 93 at 652.
130 See R. v. Silveira, [1995] 2 S.C.R. 297 at para. 150, 23 O.R. (3d) 256; R. v. Wong, [1990] 3
S.C.R. 36 at 59,60 C.C.C. (3d) 460 [Wong cited to S.C.R.].
131 The Supreme Court’s jurisprudence on this point has been somewhat contradictory. In Collins,
(supra note 6 at 285) the Court indicated that the availability of constitutional means heightens the
seriousness of the offence by demonstrating a “blatant disregard for the Charter.” See also Grant,
supra note 95 at 260; Law, supra note 106 at para. 38; Buhay, supra note 18 at paras. 52, 56, 63.
Conversely and consistently, it has also held that the fact that there was no other way to obtain the
evidence diminishes the seriousness of the offence. See R. v. Thompson, [1990] 2 S.C.R. 1111 at 1155,
73 D.L.R. (4th) 596 [Thompson]; Wong, supra note 130 at 59. But in one case the Court held that a
violation is less serious where police could have obtained the evidence constitutionally. See R. v.
Duarte, [1990] 1 S.C.R. 30 at 59-60, 71 O.R. (2d) 575 [Duarte cited to S.C.R.].
132 R. v. Dyment, [1988] 2 S.C.R. 417, 73 Nfld. & P.E.I.R. 13 [Dyment cited to S.C.R.]; Buhay,
supra note 18 at par. 52; R. v. Caslake, [1998] 1 S.C.R. 51 at para. 34, 123 Man. R. (2d) 208
[Caslake]; Belnavis, supra note 90 at para. 40.
133 See R.J. Delisle, “Mellenthin: Changing the Collins Test” (1992) 16 C.R. (4th) 286 at 290;
Robert Harvie & Hamar Foster, “Different Drummers, Different Drums: The Supreme Court of
Canada, American Jurisprudence and the Continuing Revision of the Criminal Law Under the
Charter” (1992) 24 Ottawa L. Rev. 39 at 46, n. 23; Stuart, Charter Justice, supra note 1 at 506;
Brewer, supra note 8 at 250.
134 See e.g. Grant, supra note 95 at 261 (“the negative effect of the exclusion of the evidence and
the good faith of the officers outweighed the seriousness of the violations … “). See also R. v.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
The Court’s balancing approach is superficially compelling. It purports to give
courts the capacity to make fine-tuned valuations of competing social interests,
excluding when necessary to deter egregious violations and admitting when the need
for evidence is most pressing.’35 As David Paciocco puts it, the balancing approach
(or as he calls it, the “principle of proportionality”) “enables the complex mix of
competing interests to be measured on a case by case basis.”’36 “Sometimes the costs
[of exclusion],” Paciocco continues, “simply outweigh the benefits.”’37 Balancing
appears, in other words, to be the best method of achieving optimal deterrence.
Closer examination reveals, however, that it is not. Like many other expostfacto,
“all of the circumstances” inquiries, the balancing approach fails to give police the ex
ante certainty necessary for optimal deterrence.’38 As discussed, exclusion cannot
deter unless police understand the limits of their investigative powers. The more
complex and uncertain the rules, the less likely it is that police will obey them. The
same logic applies to the sanctions that the law imposes for non-compliance. Police
are less likely to be deterred from violating the Charter when the test for exclusion is
complex and uncertain rather than clear and predictable. Under the Court’s balancing
approach, police know that deliberate, flagrant violations are likely to result in
exclusion. But where the unconstitutionality of an intrusion is not obvious, the
balancing approach gives them little reason to refrain from intruding. At the moment
when they are contemplating intruding, they will often have little idea as to whether a
court will characterize the violation (if it is a violation) as serious.’39 They may also
be unable to assess whether the crime they are investigating is serious, both because
the nature of the crime (if it is a crime) may not be fully known at the time and
because they may not be able to accurately predict whether a court will agree with
their characterization. 4 Similarly, it will often be very difficult for police to predict
whether a court will eventually find that the evidence they are seeking is important to
obtaining a conviction. With all of this uncertainty, police will often decide to commit
the intrusion, rationally calculating that there is a decent chance that a court will
Colarusso, [1994] 1 S.C.R. 20 at 78, 110 D.L.R. (4th) 297 [Colarusso] (emphasizing the “appalling
circumstances in which the underlying offence … was committed.”). To reiterate, the Supreme Court
has been clear that the seriousness of the charge and the importance of the evidence cannot be used to
justify the admission of evidence found to threaten the fairness of the trial. See Collins, supra note 6 at
286; Burlingham, supra note 18 at 242.
135 See Paciocco, “Fair Trial Dichotomy”, supra note 108 at 172-75, 181; Don Stuart, “Stillman:
Limiting Search Incident to Arrest, Consent Searches and Refining the Section 24(2) Test” (1997) 5
C.R. (5th) 99 at 108; Don Stuart, “Questioning the Discoverability Doctrine in Section 24(2) Rulings”
(1996) 48 C.R. (4th) 351 at 356; Paciocco, Getting Away with Murder, supra note 49 at 172-73.
136 Paciocco, “Fair Trial Dichotomy”, ibid. at 172.
137 ibid.
138 See generally Penney, “A Concern”, supra note 26 at 235-36; Antonin Scalia, “The Rule of Law
as a Law of Rules” (1989) 56 U. Chicago L. Rev. 1175.
139 See Levenson, supra note 64 at 884.
140 The Supreme Court has yet to develop coherent standards for determining offence seriousness
under section 24(2) of the Charter. See Roach, Constitutional Remedies, supra note 11 at para.
10.1910.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
decide either that the intrusion did not violate the Charter or that, on balance, the
section 24(2) factors weigh in favour of admission. Such calculations are particularly
likely when police believe that
to obtain evidence by
constitutional means. In such cases police have little to lose by overreaching, even if
there is a good chance that evidence will be excluded. They are only likely to be
deterred if they believe that exclusion is virtually certain. 14 1
it would be difficult
We can see, then, that the ex ante uncertainty inhering in the Court’s balancing
approach undercuts the ability of exclusion to deter all but the most flagrant
violations. From an ex post facto perspective, it seems reasonable to weigh the costs
and benefits of exclusion through an assessment of the severity of the violation, the
seriousness of the offence, and the importance of the evidence. But from an ex ante
point of view it does not. The balancing approach fails to deter violations in many
cases where a court would have determined, ex post, that the violation warranted
exclusion.
Optimal deterrence is much more likely to be achieved by a bright-line rule that
makes it clear to police that violations will almost always result in exclusion. Such a
rule would substantially increase constitutional compliance in criminal investigations.
And as the American experience demonstrates, it would not cause many guilty
defendants to be set free.142 Some commentators have argued, however, that robust
exclusionary regimes simply encourage courts to narrow the scope of constitutional
rights to avoid lost convictions. 143 Rights may indeed be interpreted more broadly
when unconstitutionally obtained evidence is likely to be admitted than when it is
likely to be excluded. The true level of protection afforded by constitutional rights,
however, depends not only on how broadly they are defined, but also on how likely
they are to be respected. A regime that generously interprets rights but rarely excludes
evidence obtained in violation of those rights does not provide much protection
against overreaching. Disparities between rights definition and enforcement also
obscure the trade-offs that constitutional criminal procedure must inevitably make
between competing interests. If courts determine, for example, that for a particular
type of search, crime control should prevail over privacy, then they should declare
141 For some, this may beg the question as to whether deterrence is possible in any case where
police believe that sought-after evidence cannot be found by constitutional means. It is. Where
exclusion is substantially certain to flow from an intrusion, police will generally (but of course not
always) forego the evidence, either to conserve resources (Why engage in a costly intrusion when its
evidentiary fruits will almost inevitably be suppressed?) or to avoid formal or informal non-
exclusionary sanctions for non-compliance (Why risk such consequences, even if they are unlikely to
be imposed, when there is no advantage to be gained by committing the violation?).
142 See discussion supra notes 56-59 and accompanying text.
143 See Paciocco, “Fair Trial Dichotomy”, supra note 108 at 174; Pottow, “Unified Approach”,
supra note 8 at 64-67; George C. Thomas III & Barry S. Pollack, “Saving Rights from a Remedy: A
Societal View of the Fourth Amendment” (1993) 73 B.U.L. Rev. 147 at 147; Perrin et al., supra note
33 at 677; Stribopoulos, supra note 7 at 131-38.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
such searches constitutional.'” They should not be saying, “there is a right to privacy
in these circumstances, but we are not willing to stop police from violating it.”
The only question, then, is whether to adopt a rule mandating the exclusion of all
unconstitutionally obtained evidence (the “absolute rule”) or one that permits
exclusion when prosecutors are able to demonstrate that exclusion would fail to deter
because the violation was inadvertent and reasonable (the “good faith rule”). The
latter is obviously more consistent with section 24(2)’s text and legislative history. It
is also more compatible with existing section 24(2) doctrine. And properly conceived,
it is at least as likely to achieve optimal deterrence as the absolute rule. Recall that
deterrence requires the subjects of regulation to know and understand the law. If
police are not aware that a particular investigative tactic might violate the Charter,
then they have no reason to refrain from using it.’45 The logic of deterrence also
dictates, however, that misconduct should only be excused when ignorance of the law
is reasonable.’46 Otherwise police will have an incentive to remain oblivious to their
constitutional obligations.’ 47 Police should not be held to have acted in good faith,
therefore, unless at the time of violation they honestly and reasonably believed that
they were complying with the Charter.’48 It is important, moreover, to limit the good
faith exception to cases where all government actors
in a criminal
investigation have made reasonable, good faith efforts at constitutional compliance,
including prosecutors giving advice to police and judicial officers issuing search and
arrest warrants. Constitutional criminal procedure can be very complex and police
often reasonably rely on authorities with greater expertise to define the boundaries of
their power. If intentional or negligent mistakes by those authorities do not result in
exclusion, then they will have little reason to resist pressure to grant police maximum
investigative liberty.’49 Courts should deny section 24(2) applications, therefore, only
involved
‘A This is in essence the “reasonable expectation of privacy” test that defines the bounds of privacy
protection under both section 8 of the Charter and the fourth amendment. See Hunter v. Southam Inc.,
[1984] 2 S.C.R. 145, 55 A.R. 291; Katz v. United States, 389 U.S. 347 (1967).
145 See Powell, supra note 14 at 540, White J., dissenting (“When law enforcement personnel have
acted mistakenly, but in good faith and on reasonable grounds, and yet the evidence they have seized
is later excluded, the exclusion can have no deterrent effect.”)
146 See Stuart, “Eight”, supra note 109 at 62-63.
147 See Jull, supra note 5 at 549.
148 The reasonableness requirement also eliminates the difficulty in proving that police were aware
that their behaviour violated the Charter. It is much easier for police officers to lie about their states of
mind than their conduct.
149 My proposal, it should be noted, is in this sense broader than the specific good faith exceptions to
the exclusionary rule that the United States Supreme Court has developed in its recent fourth
amendment jurisprudence. See Leon, supra note 12 (no exclusion where police reasonably rely on
warrant subsequently determined to have been issued in the absence of probable cause); Krull, supra
note 12 (no exclusion where police reasonably rely on statute authorizing search where statute
subsequently determined to violate fourth amendment); Evans, supra note 12 (no exclusion where
officer reasonably relied on police record containing clerical error in effecting an unconstitutional
arrest). In particular, the Leon warrant exception does not apply to judges and magistrates issuing
warrants. The Court reasoned that there was no basis “for believing that exclusion of evidence seized
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
when prosecutors establish that all those responsible for the violation honestly and
reasonably believed that they were complying with the Charter. Excluding evidence
in these circumstances would not further deterrence. 5 It would, on the other hand,
potentially allow factually guilty defendants who would otherwise have been
convicted to go free. Even if this were to occur only rarely, it is a substantial social
cost that should not be incurred in the absence of any countervailing benefit.
American critics of the good faith exception to the exclusionary rule have argued
that suppression can deter even when police have acted on a reasonable but mistaken
belief in the constitutionality of their actions.’
in these
circumstances, the argument runs, has a long-term, general deterrent effect. It gives
law enforcement officials an incentive to devote greater care to ensuring that their
policies, procedures, and practices minimize
the possibility of constitutional
infringement to the greatest extent possible.’52
Excluding evidence
But as long as courts assess the reasonableness of all facets of the state’s
investigative apparatus-including both the actions of police directly responsible for
committing the violation as well as any policies, practices, training, legal advice, or
judicial authorization that may have contributed to it-then excluding the fruits of
reasonable mistakes would have little deterrent impact.’53 Moreover, there is very
likely an “uneliminable” margin of error among “even well-trained officers” in
pursuant to a warrant will have a significant deterrent effect on the issuing judge or magistrate” (Leon,
supra note 12 at 916). The argument in favour of requiring prosecutorial reasonableness is admittedly
stronger than that requiring magisterial reasonableness. Unlike prosecutors, justices issuing warrants
are required to be independent and impartial. Empirical evidence suggests, however, that the ex parte
nature of warrant applications may result in systemic bias in favour of police interests. See Casey Hill,
Scott Hutchison & Leslie Pringle, “Search Warrants: Protection or Illusion?” (2000) 28 C.R. (5th) 89;
Wayne R. LaFave, “‘The Seductive Call of Expediency’: United States v. Leon, Its Rationale and
Ramifications” [1984] U. Ill. L. Rev. 895 at 906-909 [aFave, “Expediency”]. Subjecting issuing
justices to the possibility of the exclusionary remedy seems likely to attenuate this bias. It also
dissuades police from judge-shopping and from uncritically relying on justices’ determinations of
probable grounds. See generally Leon, supra note 12 at 955-56, Brennan J., dissenting, and 974-76,
Stevens J., dissenting; Stewart, supra note 29 at 1403. But see Donald Dripps, “Living with Leon”
(1986) 95 Yale L.J. 906 at 929-33.
150 See Cloud, supra note 67 at 267.
151 See Leon, supra note 12 at 953-55, Brennan J., dissenting.
152 See LaFave, “Expediency”, supra note 148 at 910.
153 As Dickson J. (as he then was) put it in discussing the virtues of strict versus absolute liability for
regulatory offences:
If a person is already taking every reasonable precautionary measure, is he likely to
take additional measures, knowing that however much care he takes, it will not serve as
a defence in the event of breach? If he has exercised care and skill, will conviction have
a deterrent effect upon him or others? Will the injustice of conviction lead to cynicism
and disrespect for the law, on his part and on the part of others? (R. v. Sault Ste. Marie
(City), [197812 S.C.R. 1299 at 1311, 85 D.L.R. (3d) 161).
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
applying the most complex criminal procedural rules. 54 Ideally, police would have
perfect knowledge of existing Charter requirements and accurately predict the
resolution of outstanding constitutional questions. But in reality we cannot reasonably
expect them to. Excluding evidence obtained by unintentional, non-negligent error not
only fails to deter, it is also likely to increase police frustration with courts and the
Charter, exacerbate the police perjury problem, and damage the reputation of the
criminal justice system.
Admittedly, the good faith approach cannot prevent judges who resist exclusion
from applying the reasonableness standard too meekly. Adopting the absolute rule,
however, would not prevent this. Recalcitrant judges would simply fail to find a
constitutional violation. 5′ The good faith rule does not have to be applied perfectly,
moreover, in order to deter effectively. Because many judges would hold them to a
rigorous standard, police would have ample incentive to maximize training and avoid
negligent errors.
It could also be argued that by employing a reasonableness standard, the good
faith approach is as incapable of generating ex ante certainty as the Supreme Court’s
balancing approach. But unlike the balancing test, the good faith approach does not
permit courts to weigh competing interests. The only circumstances considered are
those relating to the intentionality and reasonableness of police conduct. The good
faith approach sends a clear signal to police that all evidence derived from intrusions
that they knew or should have known were unconstitutional will be excluded, even
where they have not behaved egregiously and the evidence is needed to obtain a
conviction for a serious crime. A rule mandating exclusion in every case is unlikely to
achieve much more deterrence than this.156
How, then, should courts go about determining whether violations were
inadvertent and reasonable? Many of the Supreme Court’s “seriousness of the
154 Heffeman & Lovely, supra note 42 at 345 (suggesting this margin of error lies between 20 and
30 per cent). See also discussion supra notes 47-48 and accompanying text.
155 See Orfield, “Deterrence, Perjury”, supra note 43 at 118.
156 Some commentators have suggested that the good faith exception eliminates defence lawyers’
incentive to bring forward novel constitutional claims. See Stephen G. Coughlan, “Good Faith and
Exclusion of Evidence under the Charter” (1992) 11 C.R. (4th) 304. Defence lawyers’ ethical
obligations, however, oblige them to advance any argument that has a reasonable prospect of success.
In most cases where police have violated the Charter, there will at least be a plausible argument that
their mistake was unreasonable. In cases where it is obvious that exclusion will be denied on good
faith grounds, courts can award remedies under section 24(1) of the Charter (such as damage or cost
awards or modest reductions in sentence) to provide the necessary incentive. See generally Stuart,
Charter Justice, supra note I at 459-60 (discussing cases endorsing non-exclusionary remedies for
Charter breaches); John Pottow, “Constitutional Remedies in the Criminal Context: A Unified
Approach to Section 24” (2000) 43 Crim. L.Q. 459 at 487 (“[T]here is no reason why … an applicant
seeking only exclusion of a given piece of evidence should not be taken as presumptively asking for a
lesser, monetary award if the judge concludes that outright exclusion is excessive.”); Caldwell &
Chase, supra note 33 (proposing sentence reductions as incentive for fourth amendment claims where
exclusion not available).
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
violation” factors are relevant. “Deliberate” violations should obviously result in
exclusion. The more “flagrant” the intrusion into a clearly protected interest (such as
bodily integrity or residential privacy), the more likely it is that a court will conclude
that police either knew or should have known that it infringed the Charter.
Conversely, “technical” violations are more
to be characterized as
unintentional, non-negligent errors. It would be a mistake, however, to consider these
factors in order to determine the intrinsic “seriousness” of the violation, as measured,
for example, by the degree of harm it caused to the defendant. Courts should only
consider them to help decide whether the violation was inadvertent and reasonable
(and hence undeterrable). When police
intentionally or
negligently, deterrence requires that evidentiary fruits be suppressed no matter how
trivial the violation may appear. Conversely, unintentional, reasonable violations
should not result in exclusion even when they cause serious harm.
infringe the Charter
likely
Much of what the Court has said about “good faith” is compatible with this
approach. It has upheld the admission of non-conscriptive evidence obtained by
police conducting searches pursuant to legislation,’57 policy directives,’58 widely used
practices, 59 legal advice,160 and lower court decisions that were later held to violate
the Charter.’61 It has also excused violations by police who relied in good faith on the
validity of search warrants.’62 In many cases, moreover, it has insisted that good faith
reliance on these authorities be reasonable.’63 But it has not always done so, holding
in numerous cases that “inadvertent” or “careless” (but arguably negligent) mistakes
did not warrant exclusion.”6 The Court should clarify, therefore, that the “good faith”
exception applies only to non-negligent violations.
157 See R. v. Sieben, [1987] 1 S.C.R. 295, 38 D.L.R. (4th) 427; R. v. Hamill, [1987] 1 S.C.R. 301, 38
D.L.R. (4th) 611; Grant, supra note 95; Wiley, supra note 95; Plant, supra note 95.
158 See R. v. Simmons, [1988] 2 S.C.R. 495, 55 D.L.R. (4th) 673; R. v. Jacoy, [1988] 2 S.C.R. 548,
[1989] 1 W.W.R. 354; Caslake, supra note 132.
159 See Colarusso, supra note 134; ibid.
160 See Wong, supra note 130.
161 See Thompson, supra note 131.
162 R. v. Goncalves, [1993] 2 S.C.R. 3, 135 A.R. 397, rev’g (1992), 81 C.C.C. (3d) 240 (Alta. C.A.);
R. v. Erickson, [1993] 2 S.C.R. 649, 81 C.C.C. (3d) 447, aff’g (1992), 125 A.R. 68, 72 C.C.C. (3d) 75
(Alta. C.A.).
163 See e.g. Duarte, supra note 131 at 60 (police misunderstanding of law was “entirely
reasonable”); Wong, supra note 130 at 59; R. v. Genest, [1989] 1 S.C.R. 59 at 87, 45 C.C.C. (3d) 385
(no evidence of bad faith but “defects in the search warrant were serious and the police officers should
have noticed them”); Dyment, supra note 132 at 440 (“no evidence that the respondent’s rights were
knowingly breached,” but “lax police procedures cannot be condoned”); Kokesch, supra note 18 at 32
(“Either the police knew they were trespassing, or they ought to have known. Whichever is the case,
they cannot be said to have proceeded in ‘good faith’, as that term is understood in s. 24(2)
jurisprudence.”); Buhay, supra note 18 at para. 59 (“[T]he officer’s subjective belief that the
appellant’s rights were not affected does not make the violation less serious, unless his belief was
reasonable.”)
164 See e.g. R. v. Wise, [1992] 1 S.C.R. 527 at 545, 70 C.C.C. (3d) 193 (use of expired search
indicated “carelessness” but not bad faith); Grant, supra note 95 at 253 (police
warrant
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
The Court should also direct trial judges to apply the good faith exception to self-
incriminating evidence. As discussed, the fact that unconstitutionally obtained
evidence is self-incriminating does not justify its exclusion. Deterrence principles
dictate, however, that certain types of Charter violations that tend to produce self-
incriminating evidence should generally trigger exclusion. It should be easy for
police, for example, to comply with their informational obligations under section 10
of the Charter. Apart from cases where courts retrospectively amend
those
obligations,’65 violations should rarely be excused as honest, non-negligent errors.
But other types of section 10 violations may sometimes be characterized as
reasonable, inadvertent mistakes, for example where police incorrectly conclude,
after considering all of the circumstances, that a suspect they are questioning is not
detained and is therefore not entitled to be cautioned.’66
Of course, the same principles should apply to violations of Charter rights, such
as section 8, that typically produce non-conscriptive evidence. Exclusion should
result when police violate clear search and arrest rules, for example when they arrest
suspects in their residences without warrants and in the absence of exigent
circumstances.’67 But where police violate less precise rules, such as those governing
whether, considering all of the circumstances, there is a “reasonable expectation of
privacy” in a context that has not yet been considered by the courts, admission may
well be justified.’68 If liberty or privacy appears unjustifiably compromised by the
admission of evidence in these circumstances, courts should be faulted for crafting
indeterminate substantive rules and not for excusing police for blamelessly failing to
follow them. As I have repeatedly stressed, the prospect of evidentiary exclusion can
do little to deter police misconduct unless it is reasonably clear, ex ante, what
constitutes misconduct. This should prod courts to craft substantive constitutional
criminal procedure rules precisely, avoiding whenever possible the indeterminacy
endemic to ex post facto, “all of the circumstances” inquiries.
As mentioned, the Court has also held that a violation is more serious if its
evidentiary fruits could have been obtained by legal means. Under the deterrence-
based approach, the availability of legal alternatives should only be considered
insofar as it is relevant to good faith. When the legality of an intrusion was uncertain
ex ante, the existence of an obviously lawful alternative strongly indicates that the
decision to proceed was unreasonable. Police should be expected to forego
constitutionally questionable methods in favour of clearly legal ones. Conversely,
“inadvertently” failed to infonn issuing justice of previous wanantless perimeter search; no evidence
of “bad faith”); Plant, supra note 95 at 298 (misstatement to issuing justice exaggerating specificity of
informant’s tip was “good faith, albeit erroneous, attempt to draft the information concisely,” not
“deliberate attempt to mislead”).
165 See e.g. Broyles, supra note 107; Bartle, supra note 95.
166 See R. v. Moran (1987), 36 C.C.C. (3d) 225, 21 O.A.C. 257; R. v. Hicks, [1990] 1 S.C.R. 120, 54
C.C.C. (3d) 575 aff’g (1988) 64 C.R. (3d) 68, 42 C.C.C. (3d) 394 (Ont. C.A.).
167 See e.g. Feeney, supra note 18.
168 See generally Duarte, supra note 131.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
when there are no apparent legal alternatives, a court is more likely to determine that
the violation was inadvertent and reasonable. This does not mean, of course, that
courts should justify obvious improprieties when there is no other means of obtaining
evidence.169 Neither does it excuse police agencies from making reasonable efforts to
ensure adequate officer training. It simply recognizes the fact that even well-trained
officers will encounter situations in which the bounds of legality are not clearly
marked, and that when alternative legal means are unavailable the prospect of
exclusion is unlikely to dissuade them from intruding.
interests protected by a Charter right
The presence of exigent circumstances, on the other hand, which the Court has
held mitigates a violation’s seriousness, should rarely be a relevant factor in section
24(2) determinations. As I have argued, if in certain circumstances an intrusion into
the liberty or privacy
is justified by
countervailing interests, then courts should interpret that right to permit the intrusion.
Allowances for exigency should therefore be built into the definition of rights, as the
Court has done in permitting warrantless searches and arrests to prevent physical
harm or the imminent loss of evidence.17 But if circumstances are not urgent enough
to justify what would otherwise be a violation, then perceived exigency should not
weigh in favour of admission. Otherwise, exclusion loses its capacity to deter
intrusions ex ante that courts would find were unjustified ex post. It is precisely when
police feel the need to act urgently that exclusion is most needed to deter
overreaching. Again, the only exceptions to this should arise when the constitutional
rules governing police conduct in the circumstances are unclear. Police who are
reasonably unsure as to whether an intrusion is legal are especially unlikely to be
deterred by the prospect of exclusion if they also believe that there is an urgent need
to act.
Conclusion
Interpreting section 24(2) of the Charter is no easy task. Like many other
constitutional provisions, it is expressed in broad and indeterminate language. That
language, moreover, was the product of an ambiguous compromise between the
traditional common
the American constitutional
exclusionary rule.171 About all that can be discerned from the provision’s phraseology
inclusionary
rule and
law
169 See Kokesch, supra note 18 at 29 (“Where the police have nothing but suspicion and no legal
way to obtain other evidence, it follows that they must leave the suspect alone, not charge ahead and
obtain evidence illegally and unconstitutionally.”)
170 See Grant, supra note 95 (warrantless search power read down
to require exigent
circumstances); Feeney, supra note 18 (exigent circumstances exception to warrant requirement for
residential arrests); R. v. Godoy, [1999] 1 S.C.R. 311, (1998), 41 O.R. (3d) 95 (warrantless entry of
residence to ensure safety is permitted in response to emergency calls). See also Strachan, supra note
94 at 999 (immediate section 10(b) warning requirement may be deferred until police achieve control
over “potentially volatile situation”).
171 See Simmons, supra note 6 at 532 (“[T]he Charter enshrines a position with respect to evidence
obtained in violation of Charter rights that falls between two extremes. Section 24(2) rejects the
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
and legislative history is that they support neither an absolute inclusionary nor
absolute exclusionary regime.
Working out a sensible framework for section 24(2) determinations consequently
requires a thorough immersion in theory. An examination of the various rationales for
excluding unconstitutionally obtained evidence leads to an ineluctable conclusion:
deterring constitutional violations is the only defensible reason to exclude. The
United States Supreme Court has recognized this for many decades and few
contemporary American jurists question it.’72 While initially reluctant to do so, our
own Supreme Court seems increasingly prepared to concede that deterrence is an
important aim of exclusion. It is now time for it to take the final step and admit that it
is the only one.
Once we recognize that deterrence
is the only compelling justification for
exclusion, we can then go on to determine the circumstances in which exclusion
constitutes a net social benefit. The empirical record demonstrates that exclusion is an
effective (though far from perfect) deterrent and that it produces very few lost
convictions. As many critics of the exclusionary rule have shown, it is possible to
imagine a non-exclusionary regime that would come closer to optimal deterrence than
the exclusionary remedy. But alternative schemes are unlikely to be implemented, and
if implemented, would probably either underdeter (if sanctions are weak) or overdeter
(if sanctions are strong) in comparison to exclusion. The great advantage of the
exclusionary rule is that it provides a substantial deterrent against intentional and
negligent constitutional violations without chilling legitimate investigative methods.
What it cannot do, however, is prevent the honest and reasonable mistakes that are
inevitably produced by the complexity and uncertainty of the constitutional law
governing criminal investigations.
With these insights in mind, constructing a sensible framework for section 24(2)
determinations becomes fairly straightforward. While the plain language of section
24(2) may not support the exclusion of evidence to deter violations of the Charter
rights of third parties, other avenues are and should be made available to do so.
Deterrence also requires that courts be permitted to exclude even in the absence of a
causal connection between the violation and the discovery of the evidence. The
deterrence
treatment of
“conscriptive” and “non-conscriptive” evidence. Constitutional violations producing
self-incriminating evidence are not inherently more egregious than those that do not.
It is time for the Supreme Court to finally abandon its trial fairness analysis.
Deterrence theory suggests, furthermore, that the Court stop attempting to balance the
seriousness of the violation against the seriousness of the offence and importance of
rationale does not, however, support the differential
American rule that automatically excludes evidence obtained in violation of the Bill of Rights … It
also shuns the position at common law that all relevant evidence is admissible no matter how it was
obtained… “).
172 This is not to say, of course, that many do not question how the court has applied the deterrence
rationale.
2004]
S. PENNEY- TAKING DETERRENCE SERIOUSLY
the evidence. The balancing approach is intuitively appealing, because it appears to
strike a reasonable compromise between irreconcilable interests. But it is simply too
indeterminate to provide police with enough ex ante certainty to prevent unjustifiable
violations. Balancing appears moderate, but in practice it fails to deter all but the most
obvious and egregious abuses. Optimal deterrence requires a bright-line rule
mandating exclusion for all but reasonable, inadvertent infringements.
Adopting a deterrence-based approach would not necessarily require a radical
overhaul of existing section 24(2) jurisprudence. Trial fairness is undoubtedly a
cornerstone of the Court’s approach to section 24(2). But while its conceptual
foundations are flawed, many of its outcomes can be supported on deterrence
grounds. Violations of the informational requirements of section 10(b), for example,
will rarely be justified as reasonable, inadvertent errors. There is consequently no
reason to fear that the deterrence approach will permit trial courts to sacrifice the right
to counsel to crime control interests, even in cases where the violation appears trivial
in relation to the seriousness of the crime and importance of the evidence. The
deterrence approach is also consistent with many of the Court’s decisions involving
non-conscriptive evidence. Most of the “seriousness of violation” factors can be
viewed as proxies for inadvertence or reasonableness. And the relative lack of
attention that courts have payed to the “effect of exclusion” factors may in part be
explained by an unexpressed recognition that applying them robustly would severely
blunt section 24(2)’s deterrent impact.
Taking deterrence seriously does not mean, moreover, that courts must abandon
the notion that section 24(2) represents a compromise between competing values or
admit unconstitutionally obtained evidence only rarely.’73 The good faith exception
recognizes that police often perform their difficult work in an atmosphere of legal
uncertainty, and that even well-intentioned and well-trained officers will make
mistakes in the heat of the moment. Exclusion can have little deterrent force in these
circumstances. But when the” rules are clear, and police intentionally or negligently
violate them, exclusion is the only practical means we have to persuade them to
improve their behaviour. In these circumstances, sacrificing deterrence to obtain a
conviction–even in cases when the sacrifice appears worthwhile-dramatically
undercuts the capacity of exclusion to deter when we would want it to.
Perhaps it is naive to think that the Supreme Court will adopt a deterrence-based
approach to section 24(2) anytime soon. It has invested considerable energy and
institutional prestige into the current regime, and only a few years have passed since
173 Nor does it require that we straightforwardly import American jurisprudence into our own. As I
have mentioned, in requiring all state actors involved in the criminal investigative process to act
reasonably, the “good faith” exception to exclusion that I have proposed is in one sense narrower than
the American rule. But in another (perhaps more important) sense, it is broader. Unlike the American
doctrine, which to date only excuses violations generated by reasonable reliance on facially valid
warrants, statutes, and official records, my proposal would excuse all non-negligent infringements.
See supra note 149.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 49
it authoritatively reconfirmed the status quo in the face of widespread calls to
reconsider it.’7 4 It is possible, however, that the Court’s reluctance to depart from
section 24(2) orthodoxy has stemmed in part from a dearth of compelling alternatives.
Commentators have proposed eliminating the automatic exclusionary rule for evidence
affecting trial fairness’75 or scrapping the trial fairness category altogether and relying
exclusively on the “seriousness of the violation” and “effect of exclusion” factors.’76
None of these proposals, however,
the murky,
underdeterring “all of the circumstances” balancing test. The proposal outlined here
promises to achieve something closer to an optimal compromise between rights-
protection and truth-seeking than any of these alternatives. Hopefully, the Court will
see fit to adopt it.
represents an advance over
174 After hearing oral argument in Stillman, supra note 18, the Court ordered a re-hearing to “invite
a re-consideration of established principles as regards the application of s. 24(2) … “. Stuart, Charter
Justice, supra note 1 at 508, n. 408.
175 See e.g. Delisle, “Collins”, supra note 8; Paciocco, “Fair Trial Dichotomy”, supra note 108.
176 See e.g. Mahoney, supra note 8; Stuart, Charter Justice, supra note 1 at 519-20.
