NOTES
The Proposed Canada Evidence Act and the “Wray Formula”:
Perpetuating an Inadequate Discretion
David M. Paciocco*
A fundamental policy underlies our rules of
evidence: they should serve to ensure a fair
trial. But the mere operation of these rules
does not always achieve this goal, and a dis-
cretion to exclude otherwise admissible evi-
dence is necessary. This discretionary power
was recognized by the Supreme Court in R.
v. Wray, but was defined there extremely nar-
rowly. The Court articulated a rigid tripartite
test, requiring the evidence to be gravely prej-
udicial, only tenuously admissible, and of tri-
fling probative force. The proposed Canada
Evidence Act, 1982 contains in section 22(2)
a virtually unchanged version of the “Wray
formula”. Convinced that the present draft-
ing is dangerously inadequate, the author ar-
gues for a more workable statutory formula.
To underscore the general importance of this
power to exclude, he illustrates the numerous
cases in which a judge could require the dis-
cretion in order to prevent an unfair trial. In
light of this importance, the shortcomings of
section 22(2) become manifest. The present
formula suggests the trial judge must find ac-
tual prejudice, while the only reasonable rule
would be to require a real risk of prejudice.
The terms of section 22(2) are generally un-
duly restrictive. They prevent the free oper-
ation of a sliding scale by which to weigh
relevance against prejudice. They add con-
fusion to the rule of relevance itself, suggest-
ing that the minimum standard of probative
value may now be lowered to some point be-
neath “trifling”. More particulary, the crite-
rion of “tenuous admissibility” is triply flawed.
It is inappropriate, as it defeats the very pur-
pose of the discretion by chiselling it down
to an effectively unexercisable power. It is
often redundant, overlapping in most cases
with the rule of relevance. It is unworkable,
requiring a fine line to be drawn where none
is possible. The author concludes that a better
formula is clearly desirable, and he proposes
an elementary balancing test. A discretion to
exclude should always exist where the real
risk of a prejudicial effect outweighs proba-
tive value.
Un principe fondamental sous-tend nos rgles
de preuve: elles doivent assurer un procs
6quitable. Leur application m~canique n’at-
teint toutefois pas toujours cette fin et la dis-
cr~tion d’6carter une preuve, mime recevable,
est parfois n~cessaire. Ce pouvoir discretion-
naire a et6 reconnu par la Cour supreme dans
l’arr~t La Reine c. Wray, quoique dans un
sens tr~s restreint. La Cour supreme y a alors
pos6 trois crit6res: il doit s’agir d’une preuve
fortement pr~judiciable A l’accus6, dont la re-
cevabilHt6 tient A une subtilit6 et dont la force
probante est insignifante. Le texte propos6
de rarticle 22(2) de la Loi fedrale de 1982
sur lapreuve reprend en des termes A peu pros
inchang6s la formule Wray. Convaincu de la
dangereuse insuffisance de ce projet de loi,
l’auteur propose une disposition statutaire
d’application plus facile. Pour souligner lim-
portance de ce pouvoir discr~tionnaire, il
numire plusieurs cas oi lejuge pourrait choisir
d’6earter une preuve dans le but d’6viter un
proc s inequitable. Les lacunes de rarticle 22(2)
deviennent alors 6videntes. Cet article semble
indiquer que le juge du proc~s pourra 6carter
une preuve, autrement recevable, seulement
s’il est convaincu que son admission cause-
rait un prejudice reel. La seule rgle raison-
nable, selon l’auteur, devrait plutft permettre
aujuge d’exercer sa discretion lorsqu’il existe
un risque rel deprejudice. De faron g~n~rale,
le libell de l’article 22(2) s’avre inutilement
restrictif. I empeche de comparer la perti-
nence d’une preuve avec le prejudice que son
admission pourrait causer. II rend obscure la
r~gle de pertinence mime, laissant entendre
qu’il faudra d~sormais exiger une force pro-
bante encore moins que “insignifiante” avant
d’exclure une preuve. Plus particulirement,
le crit&e de “recevabilit6 tenant i une sub-
tilit6” est triplement vici: il est impropre,
souvent redondant et irr6alisable. L’auteur
conclut qu’il faudrait recourir A une formule
selon laquelle le risque de prejudice s’va-
luerait en regard de la force probante. La dis-
cration d’carter une preuve devrait toujours
exister lorsque le juge estime que le risque
r6el de prejudice d~passe la force probante de
cette preuve.
*Assistant Professor of Law, University of Ottawa.
Mc GILL LAW JOURNAL
[Vol. 29
Synopsis
Introduction
I.
II.
Inflammatory Evidence
Inadequate Evidence
The Necessary Scope of the Exclusionary Discretion
4.
B.
C. Non-assessable Evidence
The Technical Failure of Section 22(2)
A.
B.
C. An Inappropriate Criterion: “Tenuous Admissibility”
“Actual” Prejudice versus a “Real Risk” of Prejudice
The Unduly Restrictive Terms of Section 22(2)
1.
2.
3.
Its Inappropriateness
Its Redundance
Its Unworkability
Conclusion
Introduction
A primary expectation that we have of our rules of evidence is that
they will help to ensure a fair trial. While based upon the principles and
policies which reflect our generally accepted standards of fairness, these rules
too often do not fulfill their purposes exhaustively or exclusively. The con-
sequence of this technical shortcoming can be that a rule operates unfairly.
It follows that the mere application of rules will not necessarily ensure
a fair trial. The Supreme Court of Canada recognized this fact in the case
of R. v. Wray,’ where, for the majority, Martland J. indicated that a trial
judge possesses a discretion to exclude technically admissible evidence which
would operate unfairly, based upon his duty to ensure that the accused
receives a fair trial. 2 Justice Martland defined that discretion narrowly:
It is only the allowance of evidence gravely prejudicial to the accused, the admissibility
of which is tenuous, and whose probative force in relation to the main issue before the
Court is trifling, which can be said to operate unfairly.3
Borrowing largely from this formulation, which can be referred to as
the “Wray formula”, the framers of the proposed Canada Evidence Act,
1[1971] S.C.R. 272, [1970] 4 C.C.C. 1, (1970) 11 C.R.N.S. 235, (1970) 11 D.L.R. (3d) 673
[hereinafter cited to S.C.R.].
2Ibid., 287.
3Ibid., 293.
1983]
NOTES
19824 have included a provision which, if enacted, will statutize this dis-
cretionary power. According to s. 22(2):
The court may exclude evidence the admissibility of which is tenuous, the probative force
of which is trifling in relation to the main issue and the admission of which would be
gravely prejudicial to a party.5
Martland J. adopted the strict formula now found in section 22(2) from
a passage stated obiter in Noor Mohammed v. The King.6 While the passage
does employ the same terminology contained in the Wray formula, there
is no compelling indication that Du Parcq L.J. intended these words to
serve as legal terms of art, delimiting a general formula for the exclusionary
discretion.7 Yet if, as appears to be the case, Martland J. was searching for
strict controls upon the use of an exclusionary discretion, then it is little
wonder that the passage commended itself to him and that he was quick
to adopt it as the exclusionary formula. That decision seems to have been
more a matter of choice than of authority. Ironically, this apparent desire
to preserve the integrity of rules by so narrowly confining the discretion
appears to prevent the discretion from vindicating its very rationale. The
tripartite criteria and the restrictive language employed seriously impair the
discretion’s ability to ensure a fair operation of rules. The Wray formula
thus falls prey to the very malady which gave rise to its existence; it, and
hence section 22(2), are both technically flawed.
I.
The Necessary Scope of the Exclusionary Discretion
Since the focus of the discretion is the unfair operation of evidence at
trial, it must encompass every identifiable case where technically admissible
information could cause errors in reasoning and judgment. An examination
4Bill S-33, given first reading 18 November 1982, since withdrawn and returned to committee.
It is the product of the Report of the Federal/Provincial Task Force on Uniform Evidence,
initially prepared for the Uniform Law Conference of Canada, and since published by Carswell
(1982).
5A major change made by section 22(2) is the extension of the exclusionary discretion to
evidence tendered by the accused. This appears to undermine the discretion’s original purpose
of ensuring that the accused has a fair trial because the section allows exclusion of evidence
which has the possibility of raising a reasonable doubt on behalf of the accused. However, the
primary goal of guaranteeing the fair operation of evidentiary rules justifies the change. Where
evidence tendered by the accused could mislead the trier of fact, it should also be subject to
exclusion. See the discussion of the potential scope of the discretion infra, Part I.
6[1949] A.C. 182, 192, [1949] 1 All E.R. 365 (P.C.).
7For criticisms of the legal analysis of Martland J., see Weinberg, The Judicial Discretion to
Exclude Relevant Evidence (1975) 21 McGill L.J. 1, 4-5. Other criticisms are offered by Shep-
pard, Restricting the Discretion to Exclude Admissible Evidence [:] An Examination of Regina
v. Wray (1972) 14 Crim. L.Q. 335, 342-7.
REVUE DE DROIT DE McGILL
[Vol. 29
of these cases demonstrates the critical importance of the exclusionary dis-
cretion and of a properly drafted statutory formula. It also provides a useful
benchmark against which to measure section 22(2). For convenience, po-
tentially dangerous evidence can be divided into three categories: inflam-
matory evidence, inadequate evidence, and non-assessable evidence.
A.
Inflammatory Evidence
Evidence within this category can have an emotive effect, causing a
trier of fact to exaggerate the importance of the information provided. The
largest class of inflammatory information relates to the bad character of an
accused. Evidence which invites the conclusion that the accused is a bad
person can cause the trier of fact to accept guilt where such a finding is not
warranted. The evidence suggests that the accused is a contemptible indi-
vidual. The sober judgment of the trier can be clouded by an emotional
reaction. In the words of Lord Hailsham, the evidence may add more “heat
than light”.8
Evidence which is technically admissible can often have the incidental
effect of showing that the accused is of bad character. Such is almost always
the case with similar fact evidence. Information adduced to prove particular
acts of misconduct on occasions other than the one giving rise to the charge
in question will be technically admissible, according to the conventional
articulation, where it gives rise to a relevant inference other than the mere
general conclusion that the accused is the type of person who could do such
a thing.9 Although its specific probative value is, as a general rule, sufficient
to justify its admissiblity, such evidence will still give rise to the prohibited
inference of bad character. If the trier of fact were to place emphasis upon
that inference the evidence would operate unfairly. Hence, in NoorMohammed
v. The King Du Parcq L.J. speaks in terms of an auxiliary discretion to
exclude. o
The same undesirable incidental effect can occur where evidence of an
accused’s bad reputation is called. Where an accused calls one or more
witnesses to testify as to his good reputation with respect to a particular
8D.P.P. v. Boardman [1975] A.C. 421, 454, [1974] 3 All E.R. 887 (H.L.).
9D. Piragoff, Similar Fact Evidence [:] Probative Value and Prejudice (1981), refers to this
as “Propensity Reasoning”. This traditional formulation is derived from Makin v. The Allor-
ney-Generalfor New South Wales [1894] A.C. 57, [1891-1894] All E.R. Rep. 24 (P.C.). The
author, like many before him, points out deficiencies in the workability of this formulation.
See, for example, Sklar, Catchwords and Cartwheels (1977) 23 McGill L.J. 60.
‘Supra, note 6, 192. The exercise of an exclusionary discretion was accepted by the Supreme
Court of Canada in Boulet v. The Queen [1978] S.C.R. 332, (1977) 75 D.L.R. (3d) 223.
1983]
NOTES
and relevant character trait,” the Crown may call witnesses to testify as to
the accused’s bad reputation. The problem is that while such evidence is
called ostensibly to rebut the inference of good character, the Crown may
in fact lead evidence of bad reputation with. respect to any character trait
of the accused. 12 Given the potential lack of relevance and the diverse as-
pects of one’s character, such rebuttal evidence can be extremely inflammatory.
Similarly, where an accused puts his good character into issue by calling
evidence of good reputation, or by implying generally that he is not the type
of person to commit the crime in question, the Crown may adduce evidence
of the accused’s criminal record.’ 3 Even where an accused does not put his
good character into issue, his criminal record may be presented if he becomes
a witness at the trial. 14 Clearly proof of a prior criminal record has inflam-
matory potential.
In R. v. Stratton15 the Ontario Court of Appeal held that a trial judge
does not have a discretion to refuse statutorily sanctioned questions con-
cerning, or proof of, an accused’s criminal record. 16 The Court seemed rather
concerned by the inflammatory potential of such information but refused
the discretion as an invasion of a statutory right conferred upon the Crown. 17
If the exclusionary discretion is statutized, however, it should follow as a
matter of course that it can be applied to exclude evidence otherwise ad-
missible under statutory rules.’ 8
Occasionally evidence entirely unrelated to the character of an accused
can lead to an emotional rather than rational examination of probative
value. Real evidence can inflame the trier of fact, as has been recognized
Ca. 520, (1865) 169 E.R. 1497, [1861-73] All E.R. Rep. 549.
ICharacter evidence may be initiated by the accused alone. See R. v. Rowton (1865) Le. &
12R. v. Winfield(1939) 27Cr. App. R. 139, 141, per Humphreys J.: “There is no such thing…as
putting haifa prisoner’s character in issue and leaving out the other half.” This has been heavily
criticized by Sir Rupert Cross, Evidence, 5th ed. (1979) 428-9, although the rule was apparently
approved by the House of Lords in Stirland v. D.P.P. [1944] A.C. 315, [1944] 2 All E.R. 13.
‘ 3Criminal Code, R.S.C. 1970, c. C-34 as am., s. 593.
14Canada Evidence Act, R.S.C. 1970, c. E-10, s. 12.
15(1978) 21 O.R. (2d) 258, (1978) 90 D.L.R. (3d) 420 (Ont. C.A.) [hereinafter cited to O.R.].
‘6The Court in Stratton relied upon R. v. Leforte [1962] S.C.R. viii, (1962) 31 D.L.R. (2d)
1. Although the case deals with s. 12 of the Canada Evidence Act, its reasoning applies equally
to s. 593 of the Criminal Code.
17R. v. Stratton, supra, note 15, 278. Curiously, in R. v. Tretter [1974] 3 0.R. (2d) 708, 716,
(1974) 26 C.R.N.S. 153 (Ont. C.A.), it was conceded that the common law discretion could
exclude evidence rendered statutorily admissible by s. 643 of the Criminal Code.
I lmprovements in this area by the proposed Canada Evidence Act, 1982 would reduce the
need for a discretion under s. 12 of the presentAct but not with respect to s. 593 of the Criminal
Code. See s. 123 and s. 25(2) of the proposed Act and cf the recommendations of the Task
Force in this regard in the Report, supra, note 4, 93.
Mc GILL LAW JOURNAL
[Vol. 29
in cases involving exposure of injuries and tendering of explicit photo-
graphs.19 Such evidence can create a sense of horror and disgust. Where
information causes these reactions, the likelihood of a sober decision can
be diminished and the evidence can have a tendency to induce a judgment
based upon unfair considerations.
On other occasions technically admissible evidence tendered pursuant
to rules which typically have nothing to do with inflammatory evidence can
produce exaggerated inferences. For example, the Crown may tender state-
ments “by an accused charged with [sexual assault] which contain […an]
admission with respect to the charge in question, but also contain clear
admissions of two previous [sexual assaults]. 2
The doctrine of res gestae is particularly susceptible of permitting the
coincidental inclusion of dangerous evidence. The offence in question could
be committed under circumstances which, if revealed, would demonstrate
another act of misconduct. For example, an act leading to a charge of dan-
gerous driving could occur where an accused was being pursued by police
in connection with some other discreditable offence. If the nature of that
other discreditable offence was revealed to the trier of fact, the very real
prospect of an unfair inference arises.
B.
Inadequate Evidence
To be admissible, evidence must be relevant. To be relevant it need
not be conclusive as to the existence or non-existence of a fact in issue. It
must merely have a tendency to render the existence or non-existence of
that fact more probable.21 Evidence can be adduced, then, which no rea-
sonable person could consider determinative of the fact in issue. Clearly it
would be unfair to an accused if such inconclusive evidence was inculpatory
and the court relied upon it alone in drawing a damaging inference.
The case of R. v. Bengert (No. 7) provides an example. 22 The Crown
attempted to prove that the accused was involved in a conspiracy to traffic
narcotics. It hoped to show that the accused had recently travelled to Montreal
to visit a known trafficker of drugs. Though not a material issue in itself,
it might have given rise to the material inference that the purpose of the
19For a review of the authority in this regard see MacFarlane, Photographic Evidence: Its
Probative Value at Trial and Judicial Discretion to Exclude it from Evidence (1974) 16 Crim.
L.Q. 149.
20Ratushny, R v. Deleo and Commisso [,] Annotation [:] Discretion, confessions and counsel
21Report of the Federal/Provincial Task Force, supra, note 4, 61.
22(1979) 15 C.R. (3d) 33 (B.C.S.C.).
(1972) 18 C.R.N.S. 268, 271.
1983]
NOTES
meeting involved the distribution of narcotics. The only evidence offered
to prove the “known drug trafficker’s” status was a conviction relating to
a major American drug deal thirteen years earlier. Berger J. conceded that
if the Crown had successfully proved that the Montreal contact was a major
drug dealer, evidence as to the rendez-vous would have been admissible.
Since it had not, Berger J. excluded this evidence as prejudicial, the prejudice
being the danger of an improper inference. Had the Court held otherwise,
evidence would have been left before the trier of fact which was insufficient
to support a legitimate inference.23
A closely related situation can occur where the trial judge is required
to rule on a preliminary issue of admissibility. For example, where an ac-
cused is confronted with an allegation of his guilt before he has been arrested
it is possible that he will be taken to have “adopted” the allegation as true.
Before attributing the adoption to the accused, a trier of fact must be satisfied
that the accused heard and understood the statement and failed to deny it
effectively where he could reasonably have been expected to do so. Ac-
cording to one line of authority no “foundation” is required before the
scenario is left to the trier of fact.24 Clearly it would be unfair if the trier
concluded that an allegation had been accepted by the accused if the pre-
liminary evidence could not support such a conclusion. This was recognized
by MacKeighan C.J.N.S. in R. v. Thompson,25 in which he indicated that
a trial judge could use an exclusionary discretion to keep such evidence
from the jury.
C. Non-assessable Evidence
Evidence can only be relied upon “fairly” where the trier of fact is in
a position to assess its credibility and weight. If for any reason the trier of
fact is unable to gain a rational appreciation of the true value of the evidence,
it is unfair to an accused if the evidence is relied upon to his detriment.
Reliance upon such evidence could only be considered arbitrary.
This concern led to an application of the Wray formula in two cases.
In R. v. Dingham26 the Crown attempted to tender statements made by an
alleged murder victim shortly before her death. The victim had stated that
23Ratushny has suggested that “gravely prejudicial” cannot refer to the probative force of
evidence: “Otherwise it would be in contradiction with the third criterion [trifling probative
value]”. See supra, note 20, 271. Ratushny’s observation highlights the problems created by
the use of terms such as “gravely” and “tenuous”.
24R. v. Christie [1914] A.C 545, (1914) 83 L.J.K.B. 1097 (H.L.).
25(1974) 8 N.S.R. (2d) 417, 425-6, (1974) 26 C.R.N.S. 144 (N.S.S.C., App. Div.).
26(1978) 4 C.R. (3d) 193, 195 (B.C.S.C.).
REVUE DE DROIT DE McGILL
[Vol. 29
the accused “tried to kill” her and “did it on purpose”. Murray J. of the
British Columbia Supreme Court ultimately relied upon the Wray formula
to exclude the evidence. While he did not specifically articulate as much,
the earlier portion of the judgement reveals his concern about the court’s
inability to assess the accuracy of the deceased’s observations.
The decision in R. v. Moore27 resulted in exclusion of evidence for
similar reasons. Purporting to apply the Wray principles, Van Camp J.
refused to admit the preliminary hearing transcript of a deceased rape vic-
tim’s testimony at the rape trial of the accused. The decision to refuse was
based upon the inability of the accused to cross-examine fully the deceased
victim. In the absence of such an investigation the evidence could not be
duly assessed. Thus the accused was seen to be seriously prejudiced by the
prospect of its admission.
Based upon similar reasoning there can be cases where the hearsay
exceptions do not provide a sufficient guarantee of trustworthiness to out-
weigh the risk of unreliability.28 Proper reasoning is not ensured merely
because a rule of evidence renders information admissible. Indeed, such a
realization is the very basis for the discretion.
H.
A.
The Technical Failure of Section 22(2)
“Actual” Prejudice versus a “Real Risk” of Prejudice
To identify cases of prejudice, section 22(2) uses the phrase “the ad-
mission of which would be gravely prejudicial to a party”. The operative
word “prejudice” is a good choice given that it describes “a judgement found
before due examination and consideration”. 29 It is, therefore, particularly
well suited to identifying errors in reasoning. However, apart altogether from
the use of the restrictive adjective “gravely”, the phrase is itself inadequate.
It requires that the admission of the evidence would be prejudicial. Clearly
it is impossible for anyone to know in advance whether a trier of fact will
fall victim to the misleading potential of such evidence. Indeed, this con-
clusion can rarely be made even at the end of a trial. Consequently, the
27(1973) 17 C.C.C. (2d) 348 (Ont. H.C.). Van Camp J. erroneously held that s. 643(1) of the
Criminal Code conferred a discretion on the trial judge to exclude. In R. v. Tretter, supra, note
17, it was held that this was not so. However, the Court of Appeal conceded that the common
law discretion applied, and since Van Camp J. applied the Wray principle, R. v. Moore is still
instructive.
2This sphere of judicial discretion was recognized by Rand J. in Finestone v. The Queen
[1953] 2 S.C.R. 107, 109-10, (1953) 107 C.C.C. 93, (1953) 17 C.R. 211. See Sheppard, Restricting
the Discretion to Exclude Admissable Evidence (1972) 14 Crim. L.Q. 334, 339.
29The Oxford English Dictionary, vol. VIII (1970).
1983]
NOTES
discretion must assume a role often played by our rules of evidence, namely
the prevention of possible mistakes in reasoning. To do so, the discretion
must focus on the real risk of prejudice rather than on actual prejudice.
This preventative function seems to have been recognized as a matter
of necessity at common law, notwithstanding the wording that has been
used. In Noor Mohammed, 30 for example, if the Court had used the dis-
cretion rather than the similar fact evidence rule to exclude the evidence
there could have been no hint beforehand that the improper inference would
be drawn. And in R. v. Bengert (No. 7) 31 there was no indication that the
trier of fact would treat the Montreal rendez-vous as part of the conspiracy.
Thus, the phrase is clearly understood to encompass evidence which has
the very real potential to cause errors in reasoning, not just evidence that
is certain to cause such errors. If the discretion is to be statutized, now is
the time to correct the shortcomings in its wording.
B.
The Unduly Restrictive Terms of Section 22(2)
The discretion to exclude is based upon the relationship between the
likely prejudicial effect of the tendered information and its importance to
the court in the proper assessment of the factual issues. In essence, section
22(2) poses this question: given the importance of the information, is it
justifiable to require the party against whom it has been tendered to bear
the risk that it will be improperly considered? Such a determination must
be made upon a sliding scale. The more important the evidence, the greater
the risk of prejudice we can fairly expect that party to bear. But section
22(2) prevents the free operation of a sliding scale. According to its formula,
the extent of prejudice can be contrasted with the importance of the infor-
mation only when that prejudice can be characterized as “grave”, 32 and the
sliding scale can only be used where the probative value of the evidence
can be characterized as “trifling”. 33 Thus, if evidence is moderately useful,
even the gravest prejudice cannot be grounds for exclusion. If the prejudice
is moderate but not “grave”, even trifling information must be received. It
is therefore suggested that the formula is too restrictive, as it excludes too
many cases which should be subject to this critical balancing.
30Supra, note 6.
31Supra, note 22.
32″Grave” –
“weighty, important, requiring serious thought, serious”; The Oxford English
Dictionary, vol. XI (1970).
33″Trifling” – “of little moment or value, paltry, trumpery, insignificant, petty”; The Oxford
English Dictionary, vol. IV (1970).
McGILL LAW JOURNAL
[Vol. 29
The other side of the problem concerns those situations where section
22(2) does apply. The three-part formula permits a judge to exclude unduly
prejudicial evidence; it does not require him to do so. This raises the ques-
tion of the relationship between the exclusionary discretion and the rule of
relevance itself. It is not only evidence that is “completely bereft of proof
quality” which is excluded by the rule of relevance.3 4 In order to be admitted,
evidence must obtain that degree of significance which tends to change the
probability of a desired inference. Yet, under section 22(2) a judge may
accept information of “trifling” probative value. The standard of relevance
would appear to be lowered to some point beneath “trifling”. The exact
point is not clear, nor is it clear that a change in standard was intended.
What is clear is that the restrictive language of section 22(2) would bring
unwelcome confusion to the very rule of relevance.
Section 22(2) also casts doubt upon a widely held view of the role of
the trial judge in assessing relevance. According to this view, a mere logical
connection between the tendered evidence and the probability of the fact
sought to be proved may not be sufficient to justify admissibility. In exer-
cising a discretion, 35 the trial judge will weigh the probative value of the
evidence against competing factors. In particular he considers the risk that
the evidence might raise distracting side issues, unduly prolong the trial,
unduly surprise the opposing party, or inflame emotions and provoke feel-
ings of prejudice, hostility or sympathy. When some or all of these factors
are present, the judge must decide whether the probative value makes the
evidence worth hearing. Like the Wray formula before it, section 22(2)
clearly calls into question the propriety of this investigation.
The decision in Wray should not be considered to affect the common
law power of the judge to exclude evidence where its receipt would unduly
prolong or complicate the proceedings. An important distinction exists. This
discretionary power permits the court to control its own process; the Wray
formula provides protection for the accused against prejudice. The Supreme
Court decision therefore should only be permitted to affect the judge’s as-
sessment of evidence in light of its prejudicial effect. 36 If the formula is
statutized, however, it could be read as supplanting all of the above dis-
cretionary powers. A judge would have the discretion to consider only prej-
udice as a counter-balancing factor. Since the statutory formula provides
34Report of the Federal/Provincial Task Force on Uniform Evidence, supra, note 4, 64.
35See R. v. Wray (No. 2) [1971] 3 O.R. 843, 848, (1971) 4 C.C.C. (2d) 378, aff’d [1974] S.C.R.
565, (1973) I0 C.C.C. (2d) 215, (1973) 33 D.L.R. (3d) 750. In the Ontario Court of Appeal,
Arnup J.A. considered confusion and complication as important factors in determining whether
to exercise a discretion. Hoffman, Similar Facts After Boardman (1975) 91 L.R.Q. 193, 205,
sees this analysis as part of the rule of relevance itself.
36See S. Schiff, Evidence in the Litigation Process vol. 1 (1978) 69-70, who speculates that
the effect of the Wray decision on this discretion may be so confined.
19831
NOTES
for a discretion based specifically and exclusively upon prejudice, the maxim
expressio unius est exclusio alterius could preclude the exercise of a common
law discretion based upon other factors.
There are other exclusionary discretions that could be endangered by
the statutization of the Wray formula. At present one could argue that a
trial judge may exclude evidence in order to protect a third party.37 He may
also exclude information which, though material to the issue of credibility,
is of little probative value and a source of great embarrassment to a wit-
ness. 38 As before, the Wray case should not be interpreted as having any
effect upon these discretions. Unfortunately, section 22(2) could convey that
impression.39
C.
An Inappropriate Criterion: “Tenuous Admissibility”
The final problem with section 22(2) is its requirement that evidence,
to be excluded, must be only “tenuously admissible” in the first place. Given
the purpose of the discretion this prerequisite is inappropriate. In the ma-
jority of cases it is also redundant, and, where it is not redundant, it is
unworkable.
1.
Its Inappropriateness
The purpose of the discretion is to prevent an unfair operation of rules
which are often technically deficient. To defer, then, to these unsatisfactory
rules, by restricting exclusion to cases where evidence is barely admissible
anyway, is inappropriate. Technical deficiencies do not exist only at the
edges of the rules. All evidence which tends to render the existence of a
material fact more probable can be admitted. But its admission will still be
dangerous if the information admitted cannot satisfactorily establish the
material inference and no other evidence is adduced to support it. Even the
traditional rule of similar fact evidence could be complied with, yet the
37R. v. St-Jean [1976] C.A. 513, (1976) 32 C.C.C. (2d) 438, (1976) 34 C.R.N.S. 378, per
Kaufman J.A. But see also R. v. Hawke (1975) 7 O.R. (2d) 145, 181-3, (1975) 22 C.C.C. (2d)
19, (1975) 29 C.R.N.S. 1 (Ont. C.A.) and Reference Re Legislative Privilege (1978) 39 C.C.C.
(2d) 226, (1978) 83 D.L.R. (3d) 161 (Ont. C.A.). It is submitted that due to the nature of the
enquiry in R. v. Wray the Ontario Court of Appeal position is not the inevitable or even
correct corollary of that case.
38R. v. Sweet-Escott (1971) 55 Cr. App. R. 316, per Lawton J.
39The crucial question is whether we wish to confirm the tendency of lower courts to treat
the Wray formula as outstripping its original context and devouring these separate discretions.
Such a prospect requires careful consideration.
REVUE DE DROIT DE McGILL
[Vol. 29
evidence admitted might still be dangerous. It could give rise to a material
inference apart from bad character. If so, then although the standard of
tenuous admissibility is met, the permitted inference may be trite and the
potential prejudice significant. The criterion of “tenuous admissibility” therefore
does not seek to identify dangerous information: its sole function is to
confine the scope of the discretion. While some control is admittedly nec-
essary, this objective should not be achieved in an arbitrary way. If applied,
the criterion could render the discretion largely impotent.
2.
Its Redundance
The statutory formula is often redundant since it requires evidence to
have an extremely low probative value and to be tenuously admissible. In
many cases the very issue of admissibility requires an examination of pro-
bative value. For example, it is generally agreed that even the admissibility
of similar fact evidence must be resolved by assessing the degree of probative
value.40 This is clearly true for the category described above as “Inadequate
Evidence”. 4’ Thus, in most situations where the section could apply, the
question of tenuous admissibility is already answered by an investigation
of the “probative force” of the evidence “in relation to the main issue”.
3.
Its Unworkability
In cases where the admissibility of the evidence involves something
more than a mere assessment of probative value the criterion of tenuous
admissibility could be given independent meaning. Unfortunately issues of
admissibility cannot always be analyzed against a precisely graduated scale.
Considering the example of a confession which coincidentally contains an
inflammatory admission, Ratushny speculates that there could be “a mar-
ginal point at which proof beyond a reasonable doubt has been established
but, if the evidence of voluntariness were a trifle weaker the statement would
be excluded.” 42 While no doubt the correct interpretation of the criterion
in the context of this example, such a determination is impressionistic at
best and hardly capable of precise dissection. Evidence can only rarely be
weighed with the precision of troy ounces of gold. The criterion of “tenuous
admissibility” calls for a clarity of analysis which simply does not exist.
4See the discussion of “Degree of Probative Force” in Piragoff, supra, note 9, 114, where a
collection of those commentators is presented.
41See supra, Part I(B).
42Ratushny, supra, note 20, 270-1.
19831
Conclusion
NOTES
Problems with the formula can be traced to a distrust of discretion and
a desire to temper its incursion into the integrity of rules. The preference
for rules and the sceptical attitude toward discretion seem to be natural
corollaries of our system of criminal justice. A judge is supposed to be a
technician who implements justice by cutting along the dotted lines of ready-
made rules. There are concerns that a discretion that purports to liberate a
judge from the mandatory direction of rules could undermine this system
and invite the implementation of a personal brand of justice. But these
concerns are misplaced. From the numerous examples above of potentially
dangerous evidence, it is apparent that the evidentiary rules are flawed and
therefore undeserving of zealous respect. In any event, the discretion itself
refrains from deferring to a trial judge’s personal sense of justice. Instead,
the process required is a technical one which involves a weighing of vari-
ables: the probative value of the evidence against the risk and extent of
prejudice. These are variables which do not invite a political response; there
is no danger of personal brands of justice.
These unfounded concerns have resulted in an extreme irony. Section
22(2) attempts to ensure a proper operation of rules by liberating fairness
from the technical straitjacket the rules sometimes impose. Despite this
purpose, the formula is itself placed in a technical straitjacket which in turn
impairs its ability to fulfill its function. The problem is caused by an ex-
ceedingly deferential respect for our imperfect rules of admissibility –
the
very rules which require the discretion in the first place. A formula which
is unbridled by restrictive language is easier to accept once it is recognized
that, although the discretion is a power to side-step rules, it is one that will
in fact reinforce those rules. A less restrictive formula will further the pur-
poses which the rules alone cannot.
After a detailed investigation of the proper scope of the judicial dis-
cretion to exclude, the House of Lords concluded in R. v. Sang that “a trial
judge in a criminal case has always a discretion to refuse to admit evidence
if in his opinion its prejudicial effect outweighs its probative value. ’43 With
slight modification the Sang formula would constitute a significant im-
provement over section 22(2). The Sang formula avoids unduly restrictive
terms, thereby maintaining the sliding scale necessary to ensure the fair
43R. v. Sang [1980] A.C. 402, 437, [1979] 2 All E.R. 1222. The Sang formula also contained
a second branch: “Save with regard to admissions and confessions and generally with regard
to evidence obtained from the accused after commission of the offence, he has no discretion
to refuse to admit relevant admissible evidence on the ground that it was obtained by improper
or unfair means.” This aspect does not touch upon the question at hand.
154
McGILL LAW JOURNAL
[Vol. 29
operation of evidentiary rules. It also avoids the irrelevant criterion of ten-
uous admissibility. Its only downfall is a failure to replace the requirement
of actual prejudice with the requirement of a real risk of prejudice. A trial
judge therefore should always have a discretion to exclude where the real
risk of prejudice outweighs probative worth. This is the formula that should
be statutized in place of the Wray test, lest section 22(2) become just one
more rule incapable of fulfilling its purpose.
