Standing Apart: Separate Concurrence
and the Modern Supreme Court
of Canada, 19842006
Peter McCormick*
Separate concurrences, like dissents, are a common
feature of many common law court decisions. At first
glance though, it is difficult to understand their purpose, as
they have no effect on the outcome of a case and leave in
doubt legal questions that would otherwise be perceived as
resolved. In addition, valuable amounts of time and effort
are in some instances invested in their preparation.
Seeking rationales for separate concurrences, the
author provides an empirical analysis of their use by the
Supreme Court of Canada under Chief Justices Dickson,
Lamer, and McLachlin. From the resulting data, the author
generates a typology of separate concurrences and contends
that most express significant doctrinal disagreements with
the reasoning of the majority rather than minor quibbles.
Despite those that are merely intended to bridge reasons
found in other opinions or confirm a judges break from a
previous dissent, many concurrences argue in favour of
more sweeping statements of law while others advocate for
narrower legal grounds. A significant proportion of
concurrences are even more disapproving of the majority
opinion, either asserting the incorrectness of the majoritys
reasons or disagreeing with the majoritys very approach.
Most striking are separate concurrences written in the form
of complete decisions, which the author suggests are
intended to sway other judges of the Court who have
initially committed to signing other opinions.
that writing separate
concurrences is a common and often rational strategy. The
author also suggests that their use frequently reflects
broader understandings of the judicial function in evolving
contexts. This is exemplified by the declining use of
separate concurrences at the Supreme Court of Canada as
Charter matters have become more settled.
typology
indicates
This
les
rsultats obtenus,
Les dcisions conjointes, tout comme les dcisions
dissidentes, sont frquentes dans le systme de common
law. premire vue, il est difficile de comprendre leur
raison dtre, nayant aucun effet sur le rsultat dune
affaire et semant le doute sur des questions de droit
considres comme rsolues. De plus, leur prparation
requiert souvent un investissement de temps considrable.
Cherchant comprendre leur fonction, lauteur
analyse de facon empirique lutilisation de dcisions
conjointes au sein de la Cour suprme du Canada durant les
annes de service des Juges en chef Dickson, Lamer et
McLachlin. Utilisant
lauteur
develope une typologie et suggre que la plupart des
dcisions conjointes refltent dimportants diffrends
doctrinaux entre les juges et non pas des dsaccords
mineurs. Un grand nombre de dcisions conjointes
proposent des dclarations plus gnrales sur le droit alors
que dautres promeuvent des motifs plus restreints. Une
grande proportion de ces dcisions exprime des dsaccords
encore plus grands, rejetant les motifs de la majorit ou
allant mme jusqu questionner lapproche de la majorit.
Les plus surprenantes de ces dcisions conjointes sont
celles crites comme des dcisions compltes; selon
lauteur, celles-ci ont t redig dans cette forme pour
tenter de persuader les juges qui ont initiallement decid de
se rallier une opinion diffrente.
La typologie indique que la rdaction de ces
dcisions conjointes est une stratgie commune et souvent
rationelle de la part des juges. Lauteur argumente aussi
que leur utilisation reflte des perspectives gnrales sur la
fonction des tribunaux judiciaires dans des contextes
juridiques qui voluent. Par exemple, le dclin du nombre
de dcisions conjointes ces dernires annes est
possiblement le rsultat de lexistence dune jurisprudence
plus tablie ayant trait la Charte canadienne des droits et
liberts.
* Professor, Department of Political Science, University of Lethbridge. I would like to acknowledge
the assistance of Ben Thibault, who was an active and aggressive collaborator both for the collection
and coding of data for the Lamer Court, and for the development of my databases coding categories.
Peter McCormick 2008
To be cited as: (2008) 53 McGill L.J. 137
Mode de rfrence : (2008) 53 R.D. McGill 137
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
138
Introduction
I. The Frequency of Separate Concurrence
II. Special Concurrences: What Kind of Law?
III. The Length of Separate Concurrences
IV. Toward a Typology of Separate Concurrences
A. Bridging
B. Let me add …
C. Narrower Grounds
D. Except for …
E. Different Route
F. Ditto
G. Seriatim Style
H. Other
Conclusion
[Vol. 53
139
142
143
145
146
149
149
152
155
156
157
159
162
163
2008]
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
139
Introduction
At first glance, writing a separate concurrence is the most curious of available
choices for a judge of a panel appellate court. To be part of the majority (even more
so, the lead writer for the majority) is clearly to be preferred, as it carries with it the
capacity both to declare the winner in the immediate case and to declare the law in a
way that will direct decisions in future cases. Since it is widely understood that the
named judge1 is the lead rather than the sole author, to be part of this winning
coalition also implies an opportunity to persuade or bargain with other judges about
the details of reasons, which by all accounts usually involves a considerable degree of
give and take.2
To write (or sign) a dissent is clearly less preferable, but on its face still makes
perfectly good sense. If one thinks that the majority has got the outcome wronghas
allowed an appeal that it should have dismissed or vice versathen it is a widely
accepted part of the judicial role in common law countries3 to speak up, to say that the
majority is wrong and to explain where and how they went wrong.4 To dissent all the
1 Or judges, given recent co-authorship practices on the Supreme Court of Canada. There have been
86 coauthored decisions since 1984. Perhaps the most famous example is the decision in Irwin Toy
Ltd. v. Quebec (one of the first statements about the nature and scope of freedom of expression under
the 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]), authored jointly by Dickson, Lamer and
Wilson JJ. ([1989] 1 S.C.R. 927, 58 D.L.R. (4th) 577). For a recent significant example, see Health
Services and Support-Facilities Subsector Bargaining Assn. v. British Columbia (repudiating a 20-
year old precedent to include collective bargaining within the Charter right to freedom of association),
authored jointly by McLachlin and LeBel JJ. (2007 SCC 27, [2007] S.C.R. 391, 283 D.L.R. (4th) 40).
2 All descriptions of the Supreme Court of Canadas decision-making process emphasize a practice
of drafting and circulating that typically draws serious rather than perfunctory response, and that goes
through more than one iteration. See e.g. Bertha Wilson, Decision-Making in the Supreme Court
(1986) 36 U.T.L.J. 227; Ian Green et al., Final Appeal: Decision-Making in Canadian Courts of
Appeal (Toronto: James Lorimer & Company, 1998); Donald Songer, The Transformation of the
Supreme Court of Canada: An Empirical Examination (Toronto: University of Toronto Press)
[forthcoming]. For the Supreme Court of the United States, for which judicial papers become widely
available to the public after a period of 20 years, norms of interaction clearly include explicit
negotiation as well as collegial persuasion. See e.g. Forrest Maltzmann, James F. Spriggs & Paul J.
Wahlback, Crafting Law on the Supreme Court: The Collegial Game (Cambridge: Cambridge
University Press, 2000).
3 This assertion is not universally shared though. Some commentators, such as William D. Popkin,
would reject this flat and static statement, pointing out that the practice varies from one country to
another, from one court to another even within the same country, and from one time to another
(Evolution of the Judicial Opinion: Institutional and Individual Styles (New York: New York
University Press, 2007)).
4 Of course, it is never quite that simple: it is unlikely that judges are, and arguably undesirable that
they should be, completely candid all the time about every aspect of every question. See e.g. Scott
Altman, Beyond Candor (1990) 89 Mich. L. Rev. 296; Scott C. Idleman, A Prudential Theory of
Judicial Candor (1995) 73 Tex. L. Rev. 1307.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
140
time would raise questions, but never to dissent at all over an extended period of time
would be equally remarkable.5
To write a separate concurrenceto agree with the outcome but disagree (or at
least not agree completely) with the majority reasons for that outcomeseems the
much less obvious choice. It does not change the outcome; indeed, it does not even
suggest that the outcome should be changed. A judge who writes a concurring
opinion gives up on (and thereby admits the failure of) the attempt to persuade or
negotiate with the majority to accept at least part of his diverging ideas. It undermines
the impact of the decision itself and the lessons it provides for other courts in the
future by reducing the size of the majority and concomitantly leaving in doubt
questions that would otherwise be understood to have been resolved. And in the
process it absorbs the time and energysometimes to quite a considerable extentof
the separately concurring judge or judges.
Nonetheless, judges of the Supreme Court of Canada continue to write separate
concurrences. By 25 October 2007, the McLachlin Court, which began in January
2000, had already written eighty-seven separate concurrencesabout eleven every
yearand these have totalled just over 300,000 words. To be sure, this is only about
half as often as the judges wrote dissents (175 over the same time period), and
dissents tended to be twice as long as concurrences. But this practice still represents a
considerable investment of time and energy, limited resources on any national high
court. And since every member of the Court has taken part in one or more
concurrences,6 we must be observing something that is rooted in the practices of the
Court and not simply an attribute of a subset of judgesexplanations based on
personality cannot carry all the weight in accounting for such a pervasive
phenomenon.
This paper will examine the practice of separate concurrence in the modern
Supreme Court of Canada, which I will take as including the Chief Justiceships of
Dickson, Lamer, and McLachlin.7 My intention is to develop a typology of separate
concurrences, in the hope that such a classification and frequency count will suggest
5 See e.g. Paul H. Edelman & Jim Chen, Duel Diligence: Second Thoughts about the Supremes as
the Sultans of Swing (1996) 70 S. Cal. L. Rev. 219 at 221 (writing disparagingly of a Justice
Milquetoast who never fails to follow the majority, and wondering if his participation in judicial
decisions amounts to anything).
6 Rothstein J., the most recent appointee, was the coauthor of a separate concurrence in R. v. Bryan,
2007 SCC 12, [2007] S.C.R. 527, 276 D.L.R. (4th) 513.
7 This selection is not altogether arbitrary, because by the end of the Laskin Court, the Court was
exhibiting the most unified decision-delivery practices in its history, with more unanimous decisions,
fewer dissents and separate concurrences, fewer cases with multiple minority decisions, and a total
absence of plurality decisions. The amount of disagreement rose until the mid-1990s; it fell after this
point, but is still higher than during the Laskin Court. The years selected are also very convenient. For
one thing, a complete set of Court decisions starting in 1985 is available online (Judgments of the
Supreme Court of Canada, online: LexUM
For another, it limits my inquiry to the post-Charter period.
141
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
something about the purpose and function of separate concurrence in common law
appellate courts. An earlier product of this project, looking only at the concurrence
behaviour of the McLachlin Court, has already been published.8 The present
endeavour is a refined study that takes the analysis further and applies it to a more
extended time period.
It might be thought that concurrences do not need to be investigated very closely
because they ultimately do not matter. I admit that I once belonged to this school of
thought myself, blithely assuming that dissents were big disagreements deserving
further exploration, while separate concurrences were little disagreements that
could be treated more casually. Curiously, the Supreme Court of Canada itself seems
to have adopted this same evaluation. The statistics that the Court collects on its own
performance keep track of unanimous/split decisions, where unanimous refers
only to unanimity of outcome,9 suggesting that only dissents are indications of
disagreements worth noting.
However, with respect, this practice is simply wrong, as illustrated by the trilogy
of cases led by R. v. Van der Peet.10 In these cases, Chief Justice Lamer for the
majority set out a new test for establishing a claim to an aboriginal right, using this
test twice to dismiss the claim11 and once to allow it.12 Justices LHeureux-Dub and
McLachlin vigorously disagreed with the new test, each writing a lengthy dissent in
the first two cases13 and a comparably lengthy concurrence in the third.14 Despite the
different designations used to describe their separate opinions, the depth of the
disagreement expressed, and its future doctrinal implications, are identical in all three
of these cases.15
At least some of the timeindeed, as I will suggest, most of the timeseparate
concurrences express differences of opinion that are just as significant (and
sometimes more so) as those expressed in dissents. If we wish to study what judges
8 Peter McCormick, The Choral Court: Separate Concurrence and the McLachlin Court, 2000
2004 (2005) 3 Ottawa L. Rev. 1.
9 Supreme Court of Canada, Bulletin of Proceedings: Statistics, 19972007, special ed. at 8, online:
Supreme Court of Canada
10 [1996] 2 S.C.R. 507, 137 D.L.R. (4th) 289 [Van der Peet cited to S.C.R.]; R. v. N.T.C.
Smokehouse Ltd., [1996] 2 S.C.R. 672, 137 D.L.R. (4th) 528 [Smokehouse cited to S.C.R.]; R. v.
Gladstone, [1996] 2 S.C.R. 723, 137 D.L.R. (4th) 648 [Gladstone cited to S.C.R.].
11 Smokehouse, ibid.; Gladstone, ibid.
12 Van der Peet, supra note 10.
13 Ibid. at paras. 95-223, LHeureux-Dub J.; ibid. at paras. 224-322, McLachlin J.; Smokehouse,
supra note 10 at paras. 30-82, LHeureux-Dub J.; ibid. at paras. 83-99, McLachlin J.
14 Gladstone, supra note 10 at paras. 126-54, LHeureux-Dub J.; ibid. at paras. 155-175,
McLachlin J.
15 One could make the same point about the companion cases Pecore v. Pecore (2007 SCC 17,
[2007] 1 S.C.R 795, 279 D.L.R. (4th) 513) and Madsen Estate v. Saylor (2007 SCC 18, [2007] 1
S.C.R. 838, 279 D.L.R. (4th) 547), Abella J. separately concurring in the former and dissenting in the
latter.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
142
do, we cannot overlook the fact that one of the things they do reasonably often is
write separate concurrences. My purpose in generating a typology of this type of
disagreement is to support the argument that a concurrence is worthy of examination,
as it is more often used to indicate significant doctrinal disagreement with, rather than
preference for minor adjustments to, the majority opinion.
The database I have used to obtain the numbers presented in this paper was
initially created in the mid-1990s to cover Supreme Court of Canada decisions since
1949.16 Since then, the information has been kept up to date on a personal and
voluntary basis and the data points have been expanded. Information inputted into the
database now includes the style of cause, the application and hearing and decision
dates, the result, the type of law involved, whether the decisions are unanimous or
include dissents or concurrences, which judges served on the panel and what reasons
they wrote or signed, which court the case is being appealed from, whether the action
was successful in that court and whether the decisions set of reasons were
unanimous, and a word count for every set of reasons. The full data-point expansion
applies only to the last three Chief Justiceships, and is under way for the Laskin
Court. For more recent years, the LeXUM online database is the source that has been
used to find cases, although, as the website long acknowledged, there are some gaps
for cases in the late 1980s.17 I have therefore also used the CanLII online database,18
which has a better (but still not complete) coverage. The source which was used in
the initial data compilation in the 1990s was the print version of the Supreme Court
Reports. It is only since 1970 that the Supreme Court Reports have attempted to
report every decision; data collection for years before that date included all major law
reports (e.g., Dominion Law Reports, Canadian Criminal Cases, Criminal Reports,
Canadian Cases on the Law of Torts, Motor Vehicle Reports) but there may still be
some omissions. The database has supported the publication of 18 articles as well as
one book.19
I. The Frequency of Separate Concurrence
How frequent is the practice of writing a separate concurrence? From the
beginning of Dicksons Chief Justiceship to the end of December 2006, there were
432 cases with 610 separate concurrences bearing 906 judicial signatures. For present
purposes, I will concern myself with the middle figurethe existence of a separate
concurring opinionnot with the question of how often there were multiple separate
concurrences in a single case, and only slightly with the collegiality of separate
16 The project was initially funded by the Manitoba Legal Research Institute (my thanks to Alvin
Esau) and the Alberta Law Foundation (my thanks to Owen Snider).
17 Supra note 8.
18 Online: CanLII
19 Peter McCormick, Supreme at Last: The Evolution of the Supreme Court of Canada (Toronto:
James Lorimer & Co., 2000).
143
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
concurrencesthe extent to which separate concurrences are more than the action of
a solitary judge.
To put these figures in the broader decisional context, there were about 2,200
decisions handed down by the Court over this same period, which means that about
one case in five involved a separate concurrence (or, to put it differently, decisions
outnumbered separate concurrences by about three and a half to one). However, this
description slightly misstates the pattern because it ignores the clear ebb and flow that
marked this practice. If one were to chart the frequency of separate concurrences (on
any of the three measures suggested in the above paragraph), the graph would start
low (twenty-four signatures on sixteen separate concurrences in twelve different cases
in the 198485 Court term), rise steadily and peak in the 199495 term (which had
111 signatures on 75 separate concurrences in 43 different cases), then begin to slide
back down at almost exactly the same rate as the steady rise (ending with eleven
signatures in seven separate concurrences in six different cases in the 200506 term).
The two average terms (forty-three signatures on twenty-seven concurrences in
nineteen cases) are 199192 and 199899, nicely framing the peak year as well as
marking something close to the beginning and end of Lamers Chief Justiceship.
However, the fact that the peak comes in the middle and not at the end of the Lamer
Court suggests that whatever causal factors may have driven this pattern, they are
nothing as simple as a change of chief justices.
II. Special Concurrences: What Kind of Law?
It would hardly be expected that all types of law would be equally likely to draw
separate concurrences; presumably we are drawing on those aspects and issues that
are still to some extent unsettled, and it is this uncertainty that leads even judges who
agree on the outcome to disagree on the best way to explain why it is appropriate. For
present purposes, the caseload has been divided into four blocks. The firstobvious
since the beginning of the Dickson Courtconsists of cases involving the Canadian
Charter of Rights and Freedoms.20 Although there is not a perfectly clean line
dividing Charter cases from other cases, and researchers sometimes generate slightly
different lists,21 this group is generally easy to identify. I have also allowed myself to
be guided by the headnotes in the Supreme Court Reports (which are drawn up by the
20 Supra note 1.
21 This is a hard point to document now that there are so many Charter cases that it is no longer
possible (as the first generation of Charter articles did) to include an appendix listing all the cases that
were included as Charter decisions. I will therefore illustrate the problem with a not-so-simple
question: what was the Courts first Charter decision? Most accounts would accord this honour to
Skapinker v. Law Society of Upper Canada ([1984] 1 S.C.R. 357, 9 D.L.R. (4th) 161); but LexUM
(supra note 7) includes as a Charter decision the other contender, R. v. Westendorp ([1983] 1 S.C.R.
43, 144 D.L.R. (3d) 259 [cited to S.C.R.]), despite the fact that Laskin C.J.C. found in a single
paragraph that the Charter issue does not even arise unless the bylaw in question survives a division-
of-powers test (ibid. at 46-47), which it did not. My point is that both answers are defensible.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
144
judges themselves, not by staff lawyers or law clerks). The second block is public
law, which involves noncriminal cases to which the government (provincial, federal
or municipal) is a party. The third block is criminal law, where the Crown is a party to
the case. This block also includes appeals arising from provincial offences, although
in practice these instances make up a very small part of the Courts caseload. The last
block is private law, which generally involves individuals and corporations in actions
dealing with such matters as liability and contract. To be sure, the categories are not
quite exclusive of one another (because every Charter case has some other element
usually criminal, sometimes public). However, they seem preferable to the blunt
civil/criminal distinction of the Supreme Court Reports themselves, which create an
enormous and unwieldy residual civil category.
Table 1: Separate Concurrences and Total Number of Decisions
by Type of Law
Supreme Court of Canada, 19842006
Type of law
Charter
Public
Criminal
Private
Total:
Number of
Concurrences
260
Percentage of
Concurrences
42.6%
Number of
Decisions
334
134
127
89
610
22.0%
20.8%
14.6%
100%
495
494
393
1716
The separate-concurrence proportions do not parallel those of the overall
caseloadthere are almost three times as many Charter cases involving concurrences
as one would expect from their share of the total caseload, and only about half as
many criminal cases. The criminal-law case figures admit of an easy answer: about
one-half of the criminal-law caseload was made up of appeals by right,22 which are
disproportionately likely to be dismissed unanimously by the legal minimum panel
22 When the Courts leave jurisdiction was expanded in 1975 (An Act to Amend the Supreme Court
Act, S.C. 1974-75-76, c. 18), the proportion of appeals by right promptly dropped to about 15 per cent
of the caseload, but their numbers moved steadily upward to the point where they accounted for more
than 30 per cent of the caseload during the Lamer Court. During the first seven years of the
McLachlin Court, they were down again to 17.4 per cent, or a little higher if the Courts own coding
category of mixed issues of right and leave is recognized. See e.g. R. v. Rmillard, 2004 SCC 41,
[2004] 2 S.C.R. 246, [2005] J.E. 311 (as an example of this category).
145
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
with the briefest of reasons23 or, less often, allowed unanimously by the minimum
panel with comparably brief reasons).24 These summary cases aside, the pool of real
criminal cases with substantive issues that are comparable to the rest of the leave-
generated caseload was therefore only about half as large as one would deduce from
the raw numbers. And this in turn takes us to a mildly surprising conclusion: the hotly
contentious Charter issues aside, private-law, public-law, and (substantive) criminal-
law appeals all seemed to draw separate concurrences with roughly comparable
frequency. Separate concurrences, in other words, are routine, and thus not triggered
only by extraordinary circumstances.
III. The Length of Separate Concurrences
The judges of the Court invest a considerable amount of time and effort in writing
separate concurrences. Under Lamers Chief Justiceship, the Court wrote fully half as
many words in minority reasons as it did in decisions of the Court; looking more
closely at these minority reasons, the Court wrote fully half as many words in its
separate concurrences as it did in its dissents.25 Assuming that word length is a useful
proxy for the investment of the scarce resource of judicial time, this suggests that
judges on the Lamer Court spent on average one-ninth of their working time on
separate concurrences. Under McLachlins Chief Justiceship, the Court wrote just
over a quarter as many words in minority reasons as it did in decisions, and just under
a quarter as many words in separate concurrences as dissents; judges spent on average
one-twentieth of their working time on separate concurrences. This in turn suggests
that separate concurrence is a practice of declining significance in the operation of the
Court, the implications of which I shall discuss in the Conclusion.
The average number of words in reasons for judgment moved up from just over
5,000 for the Dickson Court to just over 6,500 for the McLachlin Court; the average
dissent was very close to the same length, and increased similarly; the average
separate concurrence was just under half as long, also tracking upward. The median
separate concurrence was about 700 words longbetween two and three pages of
text. About one-fifth of all separate concurrences were less than one hundred words
long, not enough for anything more than an enthusiastic paragraph. At the other
extreme, fully one hundred separate concurrences were more than 5,000 words long,
forty of them more than 10,000 words and half a dozen more than 20,000 words. The
23 The standard formula was The appeal is dismissed for the reasons given in the Court below, a
bare dozen words. See e.g. R. v. Huang, [1990] 2 S.C.R. 1105, 11 W.C.B. (2d) 216; United States of
America v. Whitley, [1996] 1 S.C.R. 467, 132 D.L.R. (4th) 575; R. v. Howell, [1996] 3 S.C.R. 604, 110
C.C.C. (3d) 192.
24 The standard formula was The appeal is allowed for the reasons given in the dissent in the Court
below. See e.g. R. v. Grover, [1991] 3 S.C.R. 387, 67 C.C.C. (3d) 576; R. v. Fleurant, [1995] 4 S.C.R.
47, 28 W.C.B. (2d) 513; United States of America v. Jamieson, [1996] 1 S.C.R. 465, 197 N.R. 1.
25 The numbers really are that neat: 66.8 per cent of all its words were written in decisions, 22.3 per
cent in dissents, and 10.9 per cent in concurrences.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
146
differences in length correlate only mildly with type of law. The average concurrence
was 2,500 words long; concurrences in Charter and private-law cases were on
average about 20 per cent longer, and in criminal law cases about 25 per cent shorter.
But there is far more mileage to be made by working word length into the typology of
concurrences that I will be presenting below. There are very distinct types of
concurrence, and each type implies a certain scope of argument that directs it toward
a standard word length, something that does not correlate very strongly with the type
of law involved.
IV. Toward a Typology of Separate Concurrences
There is no single style, length, tone, or content to separate concurrences; they are
quite a diverse collection of arguments and ideas. Nor do they cluster in particular
areas of law (except for a tendency to be somewhat more frequent in Charter cases
than in non-Charter cases). It might be, however, that there are substantially different
types of separate concurrence, each of which has distinctive characteristics and a core
logic. This Part will attempt to generate a typology of separate concurrence on the
Supreme Court of Canada by identifying these types.
I suggest that there exist seven different types of separate concurrences (with an
inevitable but very small residual group). In order to generate these categories, and to
assign specific minority opinions to one category or another, I have relied on my own
judgment, based on a close reading of the minority reasons and an analysis of their
relation to the decision of the Court. However, as I worked through this process, I was
intrigued to discover that judges of the Court have a distinctive language that
coincides with the categories I have identified. To a surprising extent, these judges
seem to have developed a set of stock phrases used in the opening paragraph of the
concurrence (occasionally but rarely the second paragraph) that serve to identify the
nature and extent of the disagreement that will direct their separate concurrence;
rather than imposing my categories on the words of the judges, I had a real feeling
that I was discovering a typology of which they were already aware. This is the
phenomenon that I shall refer to as self-labellingthe judges themselves have
provided the unambiguous language that enables their reasons to be assigned to an
appropriate category. The categories I have identified are listed, along with their
frequency and their average word count, in the table below:
2008]
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
147
Table 2: Types of Separate Concurring Opinions,
Supreme Court of Canada, 19842006
Type
Bridging
Let me add …
Narrower Grounds
Except for …
Different Route
Ditto
Seriatim Style
Other
Number
51
106
79
141
126
77
15
14
Average Length
(words)
244
1447
658
2403
4057
144
6872
367
It is intriguing that the judges of the Court have invented language for separate
concurrences that clearly indicates from the very start the nature and extent of their
disagreement. I believe that the most useful frame for understanding minority opinions
is to speak of concerned publics and signals. By concerned publics I refer to
organizations, groups and individuals within civil society who wish to push legal
doctrine on certain issues in a specific direction. Applications for intervener status,
resulting in the filing of briefs to press these arguments, are an obvious example of
such activities; publishing academic articles or establishing journals as vehicles for a
steady flow of certain kinds of articles are others. But since the evolution of law is
usually incremental rather than total, and since the judges of the Court are discrete
individuals who do not respond identically to the arguments and interpretations that are
suggested to them, minority opinions become the vehicle for sending signals to these
publicswhat is important is not just that a judge is claiming one of his colleagues
made a mistake about the law (which could be done in a personal note, or over coffee),
but that the judge is stating this publicly in an official pronouncement of the Court.
Part of the motive in writing separately may be intellectual (This is wrong, and I want
the record to show that I know it is wrong), but part of it may also be tactical, with the
opinion serving as a signal to those concerned elements in civil society (I agree with
you, Im doing what I can, give me more to work with). A standard language for
different types of disagreement serves this signalling function well.26
26 I take the notion of concerned publics from Charles R. Epps argument that courts do not develop
doctrine on their own, but rather in collaboration with a network of supporting institutions and groups
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
148
There is one procedural complication that needs to be resolved: it concerns what
might be called concurrence swing decisions, where a judge who was initially
writing the majority reasons loses signatures to a judge who was initially writing a
concurrence. From an external perspective, what matters is that there is a set of
reasons written by a judge who agrees with the outcome but not with the reasons
given by a majority of the Court, even if at the time of writing them the judge in
question thought they would constitute the decision of the Court and thus followed
the appropriate formatfor mere observers, that is a separate concurrence. However,
from an internal perspective, which is what I will adopt, it does matter that this judge
thought she was writing a judgment, while another colleague was writing what started
out as concurring reasons but became the judgment when the votes shifted. It is in
this second set of reasons that we need to look for the acknowledgment, the terms of
respect, and the self-labelling that characterize judicial disagreement in Canada, and
that permits us to categorize them appropriately. For the same reason, it is also this
second set of reasons that gives us a better idea of the conventions on word length (a
proxy for both thoroughness of analysis and allocation of time) that accompany the
various types of disagreement.
To expand this argument, the Court has over the last twenty years developed a
standard format for delivering decisions (which it follows in over 90 per cent of
reserved decisions) that includes the facts of the case, the actions of lower courts, the
issues, an analysis, and a conclusion. Concomitantly, it has developed a minority-
reason style that includes an acknowledgment (I have read the reasons), respect
terms (but with respectdeference), and a self-location (I cannot agree with regard
to [specific issue]). But on a number of occasions, the set of reasons designated by
the Supreme Court Reports as the judgment of the court began with the
acknowledgmentrespectlocation language of minority reasons; and the (concurring
or dissenting) minority reasons contained the labelled elements of the standard
majority reasons. My argument is that these occasions clearly identify a judgment
swing as deliberations after conference moved enough votes to allow the initial
minority reasons to become a judgment and vice versa; this has happened sixty times
for separate concurrences (and ninety times for dissents) since 1984. When this
happens, I think it is more appropriate to seize on the initial expectations (when the
minority judge thought she was writing minority reasons) rather than the final
outcome (when this became the judgment of the Court because signatures moved).
The ordering of the items in the table (except of course for the residual other
category) reflects an increasing level of disagreement, a point that I will demonstrate
by describing typical examples of each type and discussing their implications. To
anticipate my conclusion, the frequency of the various types is heavily skewed toward
those types reflecting significant disagreement: although some separate concurrences
(The Rights Revolution: Lawyers, Activists, and Supreme Courts in Comparative Perspective
(Chicago: University of Chicago Press, 1998)). I take the notion of signalling from Tonja Jacobi (The
Judicial Signaling Game: How Judges Shape Their Dockets (2007) [forthcoming in Sup. Ct. Econ.
Rev.]).
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
represent rather modest levels of disagreement within the panel, concurrences
typically organize around much more substantive and deep-seated disagreements. The
typical concurrence is not motivated by a small matter.
149
A. Bridging
Concurrences in this category are never the second set of reasons for a judicial
decision, but arise when a decision of the Court is in place and a set of minority
reasons has been written (separate concurrences are more likely, but one-fifth of the
time it is a dissent). A judge may in such circumstances write a separate opinion to
agree with some of the reasons of the majority and some of the reasons of the
minority, while of course accepting the outcome proposed by the majority. As an
example, consider Chief Justice Lamers reasons in R. v. Osolin:27 I agree with my
colleague that the trial judge erred in refusing to permit cross-examination on the
medical records. … With respect to the defence of mistaken belief, I am in agreement
with my colleague, Justice Sopinka28 (who wrote his own, more substantial,
concurrence). Similarly, in R. v. Marquard,29 Justice Gonthier wrote that he agreed
with Justice McLachlins reasons for the majority, subject however to the comments
of Justice LHeureux-Dub pertaining to s. 16 of the Canada Evidence Act, … which I
adopt.30 These reasons tend to be short (about a single page of argument) for the
obvious reason that they do not need to make the full argument themselves, but
simply refer the reader to the relevant section or sections of the majority and minority
opinions.
If a single judge is to be associated with this style of disagreement, it is Justice
Gonthier: alone, he accounted for more than one-quarter of the fifty-one examples of
this category of disagreement, and conversely, this category of disagreement
accounted for roughly half of his own separate concurrences (Justices Cory, La Forest
and Lamer were in an approximate three-way tie for second place, each with about
half as many bridging opinions as Justice Gonthier). Bridging concurrences are also
the most solitary type of disagreement: 88 per cent of bridging concurrences, in
comparison with 66 per cent of all concurrences, were signed by a single judge.
B. Let me add …
A more substantive type of separate concurrence expresses agreement with the
majority disposition and its supporting reasons, usually without any significant
qualification, but then goes on to add to those majority reasons. There were 106
examples of this type. Sometimes the stated purpose is to clarify the majority reasons,
or to emphasize some element within them. For example, in Mitchell v. Peguis Indian
27 [1993] 4 S.C.R. 595, 109 D.L.R. (4th) 478 [cited to S.C.R.].
28 Ibid. at 607-08.
29 [1993] 4 S.C.R. 223, 108 D.L.R. (4th) 47 [cited to S.C.R.].
30 Ibid. at 253. LHeureux-Dub J. dissented in this case.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
150
Band, 31 Justice Wilson wrote, I agree with my colleague that the Garnishment Act
does not apply on the facts of this case. But it seems to me important to explain why
this is so.32 And more recently, in Blank v. Canada,33 Justice Bastarache wrote, I
have read the reasons of Fish J. and concur in the result. I think it is necessary to
provide a more definitive and comprehensive interpretation of s. 23 of the Access to
Information Act … 34 Nothing in the language suggests any serious substantive
disagreement; rather, what is intended is an expansion or a closer reading of one
particular aspect of the majority reasons.
Of course, there is always something more to this type of concurrence than an
initial analysis may reveal. The reasons might read as if the justices were saying,
This just occurred to me while you were reading your decision, but that is never
actually the case. Judicial decisions are collegial and negotiated products, released
only after they have been crafted and polished.35 The reasons pronounced in public
and recorded in the Supreme Court Reports are the product of a considerable and
often extended interaction, as drafts are circulated, comments received and responded
to, and compromises worked out between different points of view within the majority.
The final product only appears in public when the judges have determined that it is
indeed the final product and not some evolution toward a broader consensus. We can
feel confident that every one of the expansions or clarifications expressed within this
category were conveyed to the writing judge, indicating general agreement, but
suggesting that the majority reasons might be amended to address these points more
specifically. And, because they appear in a separate concurrence, it is clear that the
writing judge (and the other judges in the majority), upon reflection, expressly and
deliberately rejected such suggestions. The disagreement between the separately
concurring judge and the judges in the majority is strong enough that the majority is
willing to reject the suggestion at the cost of losing one or more signatures, and
strong enough that the disagreeing judge is willing to break from the majority to write
separately. However, this type of disagreement is still a relatively innocuous one that
does not significantly compromise any core doctrine.
More commonly, this type of separate concurrence not only clarifies the majority
reasons, but explicitly takes the logic behind these reasons one step further. To qualify
for inclusion in this set of concurrences, the general tone must always be one of
31 [1990] 2 S.C.R. 85, 71 D.L.R. (4th) 193 [cited to S.C.R.].
32 Ibid. at 116.
33 2006 SCC 39, [2006] 2 S.C.R. 319, 270 D.L.R. (4th) 257.
34 Ibid. at para. 66. Actually, this wording is subtly subversive, and the message may be more
negative than I have suggested. A judicial decision consists of an outcome and supporting reasons,
both of which are important; to write I have read the reasons and support the outcome carries with it
a subtle hint of but I dont support the reasons. My contention, however, is that the justices have
verbal formulae that they routinely use to express varying levels of disagreement, and therefore I do
not think it necessary to try to read between the lines.
35 See supra note 2.
151
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
addition, never of subtraction. In Peel (Regional Municipality) v. Canada,36 for
example, Chief Justice Lamer agreed with the disposition and the analysis of the
majority, but went on to say that he agreed with the outcome for these additional
reasons.37 Similarly, in Bhinder v. Canadian National Railway Co.,38 Justices Wilson
and Beetz agreed with the outcome [f]or these [reasons] and the reasons given by …
McIntyre J. for the majority.39 In R. v. Melnichuk,40 Justice Iacobucci agreed with the
majority, but would go further.41 And in Chaoulli v. Quebec (A.G.),42 the concurring
judges agreed with Justice Deschamps that the Quebec ban on private health-care
insurance violated the Quebec Charter of Human Rights and Freedoms,43 but unlike
Justice Deschamps, also wanted to take the logical step of considering whether it
violated the Canadian Charter as well.
Again, these comments are carefully contemplated additions rather than
spontaneous outbursts: the further steps have obviously been presented to (and
rejected by) the majority, yet the minority still wishes to break from what would
otherwise be a stronger majority (or even unanimous) decision to make the
observations public. It sends a message to the public that an element of the Court
wishes to push certain ideas further in order to develop the Courts precedence on a
particular issue, and hence that there is some value in considering litigation that
would bring similar issues back in front of the Court in the hope that opinions would
move in this direction. In doctrinal terms, it still constitutes a relatively modest and
unsubversive disagreement, which is why I have placed it fairly low on the
continuum.
However, it would be a mistake to write such concurrences off altogether. In
Mitchell v. M.N.R.,44 for example, Justice Binnie blandly introduced his separate
concurrence with the words There are, however, some additional considerations.45
But those additional considerations required more than 12,000 words to be discussed,
longer than the average Court decision. Also, in Toronto (City of) v. C.U.P.E., Local
79,46 Justice LeBel agreed with Justice Arbours disposition, but saw a need for
further discussion of the administrative law aspects47 of the question, and spent
11,500 words elaborating. The basic message of this type of disagreement is that a
minority of the Court wishes to go further in a particular direction than the majority.
36 [1992] 3 S.C.R. 762, 98 D.L.R. (4th) 140 [cited to S.C.R.].
37 Ibid. at 773.
38 [1985] 2 S.C.R. 561, 23 D.L.R. (4th) 481 [cited to S.C.R.].
39 Ibid. at 580.
40 [1997] 1 S.C.R. 602, 146 D.L.R. (4th) 686 [cited to S.C.R.].
41 Ibid. at para. 6.
42 2005 SCC 35, [2005] 1 S.C.R. 791, 254 D.L.R. (4th) 577.
43 R.S.Q. c. C-12.
44 2001 SCC 33, [2001] 1 S.C.R. 911, 199 D.L.R. (4th) 385.
45 Ibid. at para. 66.
46 2003 SCC 63, [2003] 3 S.C.R. 77, 68 O.R. (3d) 799, 232 D.L.R. (4th) 385.
47 Ibid. at para. 61.
152
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
[Vol. 53
C. Narrower Grounds
If let me add suggests a judge who wants to push the Court to go further, the
narrower-grounds category of separate concurrence comes from a judge who thinks
the majority may already have gone too far. There are three slightly different forms
that this category can take. One is the not at this time argument. As a good example
of the tone of this argument, consider Chief Justice McLachlins reasons in B. v.
Ontario (Human Rights Commission):48 We do not disagree in the result, … [but] we
would reserve for another day the more general question … 49 (note the double
negative). Similarly, Justice La Forest in R. v. Felawka50 agreed with the majority but
preferred not to take a definitive position regarding the interaction of certain
sections of the Criminal Code at this time.51
Slightly different, but still conveying the same general meaning, is the narrower
grounds wording. Of all types of disagreement, this is intriguingly the one most
likely to reveal a certain testiness in the language, although the general tone of the
Court, even in disagreement, is so formal and polite that one must look quite closely
to find instances of it. For example, in Canadian Broadcasting Corp. v. Canada
(Labour Relations Board),52 Justice La Forest rather pointedly asserted, As I see it,
this case raises a very narrow point and I prefer to confine myself to that point,53 a
statement that comes across as an unusually curt dismissal of his colleagues finding
of a less narrow point that deserved close consideration. Citing Justice LaForest
again, his concurrence in R. v. Papalia54 is a delightfully pithy I have had the benefit
of reading the judgment of Justice McIntyre and would dismiss the appeal solely for
the reason set forth in the last paragraph of his judgment.55
Slightly different again is the third variant, which suggests that it is not necessary
to consider some aspects or elements of the majority decision. Justice McIntyre, in
Hunter Engineering v. Syncrude Canada,56 agreed with the disposition but found it
unnecessary to deal with the further concept of fundamental breach in this case.57 In
R. v. Park,58 Justice Sopinka noted the comments of a colleague on honest but
mistaken belief in consent but claimed that it was unnecessary and undesirable to
deal with this in the immediate case.59 And in Sport Maska v. Zittrer,60 Justice Beetz
48 2002 SCC 66, [2002] 3 S.C.R. 403, 219 D.L.R. (4th) 701.
49 Ibid. at para. 1.
50 [1993] 4 S.C.R. 199, 85 C.C.C. (3d) 248 [cited to S.C.R.].
51 Ibid. at 209.
52 [1995] 1 S.C.R. 157, 127 D.L.R. (4th) 385 [cited to S.C.R.].
53 Ibid. at para. 88.
54 [1988] 2 S.C.R. 137, 43 C.C.C. (3d) 129 [cited to S.C.R.].
55 Ibid. at 197.
56 [1989] 1 S.C.R. 426, 57 D.L.R. (4th) 321 [cited to S.C.R.].
57 Ibid. at 481.
58 [1995] 2 S.C.R. 836, 99 C.C.C. (3d) 1 [cited to S.C.R.].
59 Ibid. at 873.
60 [1988] 1 S.C.R. 564, 83 N.R. 322 [cited to S.C.R.].
153
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
concurred in the result but added, I do not consider it necessary to rule on the
possible similarity between common law and Quebec law in this respect.61 More
bluntly, Justice LeBel in Entreprises Sibeca v. Frelighsburg (Municipality of)62 wrote,
I do not believe that it is necessary to state an opinion regarding the application of
those principles to the facts of this appeal … ,63 which seems to be a rather pointed
comment.
The general content of this style of disagreement is straightforward and
consistent: an expression of general agreement with the outcome, but a bluntly
expressed reluctance to sign on to some major element of those reasons. Initially, they
were coded as three different types of disagreement, but this seems to go too far.
Instead, think of a single type of disagreement, taking the pure form We prefer to
decide on narrower grounds because it is not necessary at this time to resolve the
broader issues, with different elements being emphasized in different exemplars. The
cross-section of not at this time, narrower grounds, and not necessary
demarcates this set very nicely.
There have been seventy-nine examples of this type of disagreement. Narrower
grounds concurrences tend to be shortonly a handful are over 1,000 words long
and the average, at just over 500 words, is barely a page and a half of text. These
concurrences also tend to be individual efforts66 per cent of all concurrences, but
75 per cent of narrower-grounds concurrences, are signed by a single judge.
This type of disagreement is significant because it offers a glimpse of the Court
as a strategizing institution that does not simply answer whatever questions are
dropped before it, but plans collegially and collectively over the longer term to
develop legal doctrine. The most obvious arena for judicial discretion in the service
of strategic calculation is the process of granting or denying leave to appeal (which
does not even involve giving reasons); most of the time, the Court has the power to
decide whether a specific legal issue should be addressed immediately, or whether the
Court should deny leave and wait for the matter to come up when the lower-court
debate is more developed, or when Canadian public opinion is more prepared for a
resolution.
But the other, less visible face of strategic judicial discretion is the panels
decision about how broadly or narrowly it wishes to respond to the question before it.
In the American literature, the term for this latter dimension of discretion is issue
fluidity, which deals with the capacity of the Court to decide for itself how broadly
or how narrowly to treat the issue before it.64 The battles that have been fought out on
61 Ibid. at 570.
62 2004 SCC 61, [2004] 3 S.C.R. 304 at para. 41, 243 D.L.R. (4th) 513.
63 Ibid. at para. 41.
64 See e.g. Kevin T. McGuire & Barbara Palmer, Issue Fluidity on the U.S. Supreme Court (1995)
89 American Political Science Review 691; Kevin T. McGuire & Barbara Palmer, Issues, Agendas,
and Decision Making on the Supreme Court (1996) 90 American Political Science Review 853; Lee
Epstein, Jeffrey A. Segal & Timothy Johnson, The Claim of Issue Creation on the U.S. Supreme
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
154
the pages of the American Political Science Review and the Political Research
Quarterly revolve around the codification of hundreds of decisions of the Supreme
Court of the United States, comparing the issues identified by the appellants
application with the issues actually addressed by the court. But in narrower-grounds
concurrences, we actually have the Canadian judges saying, out loud and in public,
that the members of the panel specifically consider and sometimes disagree on the
appropriate timing for the consideration of the broader issues that always and
inevitably lurk beneath specific cases.
Narrower-grounds disagreements serve as a measure of increasing reluctance
within the Court to pursue a particular doctrinal development. Think of a string of
cases that are filling out the implications of a particular legal issue. At a certain point
in time, the solid Court majority shreds as some judges evince reluctance to extend
the core principle beyond a certain point; narrower grounds is the signal to the
concerned publics that the boundaries of doctrinal extension have been reached and
further litigation on the question may not be fruitful. The set of decisions on
constructive murder, hinging on the nature of foreseeability, does not quite make
my point, but it comes close. This set includes six different decisions delivered on
13 September 1990, all dealing with the (now unconstitutional) crime of
constructive murder, which addresses situations in which a person dies as a result
of a felony.65 These cases work through an intriguing variety of factual circumstances
to display a range of subissues working off the basic concept. The five-judge majority
on the seven-judge panel held firm through all six cases. However, Justice
LHeureux-Dub separately concurred in three of the decisions,66 and dissented in the
other three. Justice Sopinka separately concurred with all six, but in different terms.
The right to counsel set, consisting of six decisions handed down on 29 September
1994, is even more useful.67 In these cases, a seven-judge majority in R. v. Matheson
eroded to a four-judge plurality in R. v. Prosper, and a number of judges (Justices
LaForest, McLachlin, LHeureux-Dub, Gonthier and Major) moved from the
majority to separate concurrences to dissents as the factual context shifted.
But, given the lack of any overt indication of a desire to roll back a legal doctrine,
this style of concurrence is still a comparatively mild form of disagreement. There is
an air of ambiguity to it: the judge who signs such a concurrent opinion is merely
communicating an unwillingness, at this time, to commit to a position on the broader
Court (1996) 90 American Political Science Review 845; Barbara Palmer, Issue Fluidity and
Agenda Setting on the Warren Court (1999) 52 Political Research Quarterly 39.
65 R. v. Martineau, [1990] 2 S.C.R. 633, 109 A.R. 321; R. v. Rodney, [1990] 2 S.C.R. 687, 112 N.R.
167; R. v. Arkell, [1990] 2 S.C.R. 695, [1990] 6 W.W.R. 180 [Arkell]; R. v. Luxton, [1990] 2 S.C.R.
711, 111 A.R. 161 [Luxton]; R. v. Logan, [1990] 2 S.C.R. 731, 73 D.L.R. (4th) 40 [Logan]; R. v.
J.(J.T.), [1990] 2 S.C.R. 755, [1990] 6 W.W.R. 152;
66 Arkell, ibid.; Luxton, ibid.; Logan, ibid.
67 R. v. Bartle, [1994] 3 S.C.R. 173, 118 D.L.R. (4th) 83; R. v. Prosper, [1994] 3 S.C.R. 236; R. v.
Pozniak, [1994] 3 S.C.R. 310, 118 D.L.R. (4th) 154; R. v. Matheson, [1994] 3 S.C.R. 328, 118 D.L.R.
(4th) 323; R. v. Harper, [1994] 3 S.C.R. 343, 118 D.L.R. (4th) 312; R. v. Cobham, [1994] 3 S.C.R.
360, 118 D.L.R. (4th) 301.
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
question, and is not necessarily suggesting that that she will reject a majority position
at an appropriate time in the future.
155
D. Except for …
A more serious form of disagreement occurs when the writer of a separate
concurrence specifically rejects elements of the majority reasons because they are not
only unnecessary to reach the outcome (or are being raised prematurely), but also
simply incorrect. A pure example of the wording used in such concurrences (and the
one that provides this sections title) is found in Quebec (A.G.) v. Greater Hull School
Board68 in the reasons of Justice LeDain, who agreed with the writer of the majority
reasons, except for his conclusion concerning the validity of the requirement of
referendum … 69 Another example is R. v. Creighton,70 in which Justice McLachlin
bluntly started off with I respectfully disagree with the Chief Justice on two
points.71 And in R. v. Demers,72 Justice LeBel stated, I agree with my colleagues
conclusion regarding the breach of s. 7 of the Charter, but I disagree with respect to
the division of powers issue … 73 The general form of the concurrence is
straightforward: concurring judges agree with the disposition, but disagree or
cannot agree with one or more specifically identified points in the majority reasons
supporting that disposition. Alternative wording indicates general agreement but then
qualifies that agreement or expresses significant reservations about specifically
indicated aspects of the majority reasons, as Justice Lamer did in R. v. Bulmer74 and
R. v. Thatcher.75
This type of concurrence expresses serious and major disagreement, comparable
in many ways to the significance and impact of a dissent. The minority is not just
adding a gloss to the majority reasons or differing on tactics or timing, but flatly
stating that the majority is wrong in the way that it has resolved some of the legal
issues involved. Indeed, in one case in this category, R. v. Potvin,76 Justice McLachlin
described herself as respectfully dissent[ing]77 from the majority even though she
agreed with the outcome. There were 141 such separate concurrences, including 40
per cent of the concurrence swing decisions. And they tended to be longer than any of
68 [1984] 2 S.C.R. 575, 15 D.L.R. (4th) 651 [cited to S.C.R.].
69 Ibid. at 598.
70 [1993] 3 S.C.R. 3, 105 D.L.R. (4th) 632 [cited to S.C.R.].
71 Ibid. at 40.
72 2004 SCC 46, [2004] 2 S.C.R. 489, 240 D.L.R. (4th) 629.
73 Ibid. at para. 68.
74 [1987] 1 S.C.R. 782 at 796, 4 W.W.R. 577, Lamer J. (I wish to add, however, the following
qualifications to certain statements made in his reasons).
75 [1987] 1 S.C.R. 652 at 703, 4 W.W.R. 193, Lamer J. (I should, however, qualify my concurrence
in those of his reasons addressing the ground he refers to as Jury Unanimity as to the Material
Facts).
76 [1993] 2 S.C.R. 880 at 886, 105 D.L.R. (4th) 214 [cited to S.C.R.].
77 Ibid. at 886.
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
156
the other forms of disagreement that have been mentioned so far, averaging 2,550
words; the longest, at 11,000 words, was Justice LeBels concurrence in Figueroa v.
Canada (A.G.),78 where he agreed with much of the majority opinion but expressed
reservations about the methodology for identifying a Charter violation.79 Except
for is the modal type of separate concurrence, the largest single category, and it
accounts for roughly one-quarter of the total.
[Vol. 53
E. Different Route
Another type of concurrence that expresses even deeper disagreement with the
majority begins with a deceptively simple self-description. To use an example from
the very beginning of the time period, in Re B.C. Motor Vehicle Act80 (from which I
take the section title), Justice Wilson agreed with the result reached by Justice Lamer
but added, I reach that result … by a somewhat different route.81 Alternate phrases
that clearly locate themselves in the same zone were on a different basis,82 for
somewhat different reasons,83 for different considerations,84 by a different
approach,85 by a somewhat different analysis and characterization,86 or in a
somewhat different manner.87 Though this variation brings to light the fact that the
judges of the Court do not have a laminated, official style sheet pinned to their
bulletin boards, it still suggests that they deliberately and consistently use similar
language to highlight the nature and extent of their differences.
I suggest that this concurrence is the most serious form of disagreement. If
narrower grounds suggests that the majority answered questions that it did not need
to ask, and if except for suggests it got part of the answer to a question wrong, then
different grounds suggests that the majority got the question itself wrong. The
language here is often very forthright about this opinion. In Norberg v. Wynrib,88
Justice McLachlin somewhat testily wrote, I do not find that the doctrines of tort or
contract capture the essential nature of the wrong done to the plaintiff.89 In Reference
78 2003 SCC 37, [2003] 1 S.C.R. 912, 67 O.R. (3d) 440.
79 Ibid. at para. 95.
80 [1985] 2 S.C.R. 486 at 522, 24 D.L.R. (4th) 536 [cited to S.C.R.].
81 Ibid. at para. 92.
82 R. v. Shoker, 2006 SCC 44, [2006] 2 S.C.R. 399 at para. 27, 271 D.L.R. (4th) 385, LeBel J.
83 National Corn Growers Assn. v. Canada (Import Tribunal), [1990] 2 S.C.R. 1324 at 1331, 74
D.L.R. (4th) 449.
94.
84 Douglas/Kwantlen Faculty Assn. v. Douglas College, [1990] 3 S.C.R. 570 at 607, 77 D.L.R. (4th)
85 R. v. Chaulk, [1990] 3 S.C.R. 1303 at 1366-67, 119 N.R. 161.
86 Canadian National Railway v. Norsk Pacific Steamship, [1992] 1 S.C.R. 1021 at 1166, 91 D.L.R.
(4th) 289, Stevenson J.
87 R. v. Bain, [1992] 1 S.C.R. 91 at 101, 87 D.L.R. (4th) 449, Cory J.
88 [1992] 2 S.C.R. 226, 92 D.L.R. (4th) 449 [cited to S.C.R.].
89 Ibid. at 268-69.
157
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
Re Young Offenders Act (P.E.I.),90 Justice La Forest wrote, I do not believe that the
test enunciated in Re Residential Tenancies Act, 1979, [1981] 1 S.C.R. 714, is fully
applicable to the case at the bar.91 In Multani v. Commission Scolaire Marguerite-
Bourgeoys,92 Justice Deschamps objected to the majoritys choice of treating the case
as a Charter issue, and firmly stated that it should have been handled purely as an
issue.93 Justice LHeureux-Dub similarly argued for an
administrative-law
administrative-law approach in her massive separate concurrence in 2747-3174
Quebec Inc. v. Quebec (Rgie des permis dalcool).94 Under slightly different
circumstances, these are the kinds of disagreements that lead to major dissents. In
Trinity Western University v. College of Teachers95 for example, the majority chose to
resolve the issue in terms of municipal and administrative law, while Justice
LHeureux-Dub wrote a vigorous dissent arguing that this was the appropriate time
and case to clarify the Courts position on the scope of freedom of religion under the
Charter.96
I would make three further observations about the different route style of
disagreement. First, the concurrences in this category were twice as long as the
average separate concurrence, at just over 4,000 words. This was not quite as long as
the average dissent, but it is sufficient to make an argument of some scope and depth.
Second, 27 per cent of all concurrences in Charter cases, but only 18 per cent of
concurrences in non-Charter cases, were on different grounds. Third, almost 40 per
cent of the concurrence swing decisions fell within this category. This is the second
most common type of concurrence, accounting for just over one-fifth of the total.
F. Ditto
About one-seventh of separate concurrences are very short comments directing
attention to the reasons presented in another Court decision. In Kourtessis v. M.N.R.,97
for example, Justice Sopinka wrote, Although I was part of the minority in Knox
90 [1991] 1 S.C.R. 252, 77 D.L.R. (4th) 492 [cited to S.C.R.].
91 Ibid. at 281.
92 2006 SCC 6, [2006] 1 S.C.R. 256, 264 D.L.R. (4th) 577.
93 Ibid.
94 [1996] 3 S.C.R. 919, 140 D.L.R. (4th) 577.
95 2001 SCC 31, [2001] 1 S.C.R. 772, 199 D.L.R. (4th) 1.
96 I am peering over a very high cliff in legal theory by raising this question. If we are persuaded by
the formalist model, then of course all judicial decisions are determinate; there is a single objectively
correct answer reached by mechanical processes, and it makes no difference in what order the
questions come or when the Court decides to decide a matter. But formalism is no longer persuasive,
and once we admit any degree of judicial discretion, we are committed to a certain amount of path
dependency: it does matter in what order the questions come, and this contingency can direct the
course of the law. For an introduction to this (very extensive) argument, see Oona A. Hathaway, Path
Dependency in the Law: The Course and Pattern of Legal Change in a Common Law System (2001)
86 Iowa L. Rev. 601.
97 [1993] 2 S.C.R. 53 at 93, 102 D.L.R. (4th) 456 [Kourtessis cited to S.C.R.].
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
158
Contracting Ltd. v. Canada, [1990] 2 S.C.R. 338, I feel bound by the majority
decision in that case and, accordingly, join Justice La Forest J.s reasons in the present
case.98 This particular example is a nice demonstration of the doctrine of precedent.
Knox Contracting Ltd. v. Canada99 raised an issue that had not yet been definitively
resolved, and the Court was divided on the best way to resolve it; three years later,
when the issue resurfaced in connection with Kourtessis, the former minority
declared that the matter had been definitively resolved and distanced itself from its
earlier disagreement with the majority.100 To be sure, it is mildly curious that the
former minority found it necessary to tell us so. One would expect that this
surrendering of earlier views in the light of previous majority decisions is a fairly
frequent occurrence that usually goes unremarkedclearly, not every nonunanimous
decision that is subsequently cited as authority calls for such an explicit act of self-
abnegation. Nor is there anything in the tone or content of the earlier decision referred
to that seems to call for particular comment. Generally speaking, disagreement on the
Supreme Court of Canada is expressed in terms that are mild, temperate, and polite;
indeed, there does not seem to be anything pointedly critical or offensive in the rather
bland wording of the minority reasons in Knox.101 Apart from wondering why
anything was said at all, however, the content seems unproblematic. Eleven
concurrences took this form.
The reason that I have put this category higher on the disagreement continuum is
the subsequent evolution of the use of the ditto citation. The most explicit
renunciations of former minority reasons in the name of precedent are found in the
first decade (198494), and there have been no explicit renunciations since 1994.
More recently, the terminology of the ditto concurrence has hardened so as to make it
considerably more ambiguous. These concurrences now tend to be extremely terse
(typically around fifty words) and uncommunicative, beginning with the phrase
subject to my comments [in another specific case], and concluding simply with I
concur in the disposition of the appeal. Tracking down the reference, and going to
the earlier case for a closer look, one invariably finds a dissent that is being referred
to. The effect on the critical observer is to see a reinforcement of, not a retreat from,
the earlier disagreement.
At first glance, the effects of both types of ditto concurrence seem to be very
similar: an earlier dissent is indicated, but in the immediate case the judge concurs.
What is missing in this modern expression, however, is any declaration of distance,
98 Ibid. at 93.
99 [1990] 2 S.C.R. 338, 106 N.B.R. (2d) 408 [Knox cited to S.C.R.].
100 Supra note 98.
101 In his dissent, Justice Sopinka begins very neutrally with I have had the advantage of reading
the reasons for judgment herein of my colleague, Cory J., but I am unable to agree with either his
reasons or his disposition of this appeal (supra note 100 at 356). And the most critical comments in
the pages that follow are While I agree with the statement of Cory J. … it does not follow … (ibid. at
360), and then a concluding I am concerned that, contrary to the views expressed by my colleague,
the appellants and others in the same position will find themselves without a remedy (ibid. at 365).
159
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
any indication that the issue has been decisively settled. Justice LHeureux-Dubs
acknowledgment of authority in the 1993 Kourtessis decision indicated no desire to
raise the issue again;102 the more modern terminology (used by a broad cross-section
of current judges) suggests a more reluctant acquiescence, which keeps the issue, at
least theoretically, alive. To my ears, the message seems to be subject to the
reservations expressed in an earlier case, reservations I still have. And this in turn
leads me to suggest that this subtle shift in wording hints at a softening in the modern
doctrine of stare decisis, such that previous decisions of the Courteven recent
oneshave a less privileged status in directing the deliberations and declarations of
the Court than they might have had a few decades ago. The message is less I was
wrong than I am still waiting for my chance.
For obvious reasons, these concurrences tend to be individual efforts66 per
cent of all concurrences, but 73 per cent of all ditto concurrences, are signed by a
single judge. Curiously though, there were three occasions on which reasons that
started off as ditto concurrences swung enough votes to become the judgment of the
Court,103 creating something that seems almost a caricature of a Court decisionwith
a terse judgment of the Court immediately followed by an extended separate
concurrence. In W.(V.) v. S.(D.),104 the most extreme example, the judgment of the
Court (so identified in the Supreme Court Reports) is a single twenty-five-word
sentence by Justice McLachlin, signed by three other judges; but Justice LHeureux-
Dubs separate concurrence is over 15,000 words long.
G. Seriatim Style
The normal appearance of a decision of the Supreme Court of Canada is a
judgment of the Court, containing the facts of the case, the actions of the lower
courts, the relevant legislation, the issues, the analysis, and the disposition.105 This
may be accompanied by one or more sets of minority reasons (dissents or separate
concurrences), but typically these reasons are derivative rather than free-standing
they refer specifically to the judgment of the Court, they position themselves in
relation to the reasons in that judgment, they self-label the focus and extent of their
disagreement, and they omit the more routine elements such as reporting the facts or
102 Supra note 103 at 93.
103 And another three occasions when it came within a single vote of being the majority, these being
R. v. Wilson, [1990] 1 S.C.R. 1291, 5 W.W.R. 188; R. v. Forster, [1992] 1 S.C.R. 339, 88 D.L.R. (4th)
169; R. v. Pearson, [1998] 3 S.C.R. 620, 233 N.R. 367.
104 [1996] 2 S.C.R. 108, 134 D.L.R. (4th) 481.
105 The evolution of such formats is interesting in itself. The Court of Appeal for England and Wales
just marches in with straight text, unbroken by headings, while the Supreme Court of the United States
divides its opinions into roman-numbered and lettered parts in a unique and sometimes very
confusing way. See B. Rudolph Delson, Typography in the U.S. Reports and Supreme Court Voting
Protocols (2001) 76 N.Y.U.L. Rev. 1203. The closest analogue to the Canadian style that I have come
across is that of the United States Court of Appeals for the Ninth Circuit; other circuits have different
practices.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
160
the actions of the lower courts. The Canadian Supreme Court Reports explicitly
reinforce this normal appearance: almost (but not quite) invariably, one set of reasons
is identified as the judgment of the Court,106 with a list of the judges signing on to it
and an identification of the lead author or authors; the other sets of reasons are
labelled as reasons of a given judge or group of judges.
Sixty-three cases depart from this pattern, though for a very simple reason: the
separate concurrence looks like a judgment of the Court because the majority
decision looks like a concurrence. At some point well into the collegial decision-
making process, a judge who was writing a separate concurrence persuaded enough
of her colleagues to sign her reasons rather than the initially designated judges
reasons, such that the concurrence and the decision changed places. Since it usually
does not make sense for everyone to start again by redrafting the decision in the more
conventional way, the new judgment of the Court typically accepts what is now the
concurrences statement of the facts, actions of the lower courts, and legislative
summary, and it may adopt some (even many) of the reasons as well. In other words,
the departure from the normal appearance is easily explained and largely illusory
(although it does generate some enormously long separate concurrences).
This explanation simply makes the other exceptions all the more surprising.
There is a style of minority opinion, usually as long as or longer than the majority
reasons, that is written as if the majority reasons simply were not there. A full
decision format (facts, actions of the lower courts, relevant legislation, issues,
analysis and disposition) is provided even though the first four elements simply
parallel and duplicate the material in the majority reasons. But at no time is the
majority opinion referred to or confronted on any specific point. The two sets of
reasons display a complete indifference to one anotherthis is a dual monologue, not
a conversation. I call this type of concurrence seriatim style because it recalls the
concurrence format of the Court from decades ago, a format that is still widely
followed by the High Court of Australia107; all or most of the judges on the panel
write their own reasons, often running parallel and even citing the same authorities.
106 [A]lmost because the exercise can fail in both directions. Sometimes, as in Committee for the
Commonwealth of Canada v. Canada, there are so many concurring reasons playing off each other in
such complex ways that no set of reasons is identified as the judgment ([1991] 1 S.C.R. 139, 77
D.L.R. (4th) 385). And sometimes, as in Comit paritaire de lindustrie de la chemise v. Potash, there
will be two different sets of reasons, both free-standing, both signed by a majority of the judges, and
both identified in the Supreme Court Reports as a judgment ([1994] 2 S.C.R. 406, 115 D.L.R. (4th)
702 [Comit paritaire]). In this case, Sopinka and McLachlin JJ. signed on to the reasons of La Forest
J. as well as those of LHeureux-Dub J. However, both scenarios together comprise less than 1 per
cent of the full caseload.
107 See Matthew Groves and Russell Smyth, A Century of Judicial Style: Changing Patterns in
Judgment Writing on the High Court, 19032001 (2004) 32 Federal Law Review 255. See also
Matthew Lynch, Report from Australia (24 April 2007), online: The Court
judgments on the Supreme Court of Canada, see Claire LHeureux-Dub, The Length and Plurality
of Supreme Court of Canada Decisions (1990) 28 Alta. L. Rev. 581.
161
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
The reasons are then totalled as votes to direct the outcome. No indication is made of
any exchange between the writers, or any attempt to engage points on which they
differ. The Court has long since abandoned this styletheir last true seriatim decision
was in 1964108and I do not wish to suggest that they are returning to it; my use of
the label is intended only to catch the notion of multiple decisions that are written as
if each of them were intended to constitute a judgment of the Court.109
Fifteen cases effectively had two sets of reasons written like judgments of the
Court, one with enough signatures to constitute the decision of the Court.110 These
cases do not seem to have any centre of gravity: they are scattered across the three
Chief Justiceships (four for Dickson, seven for Lamer, four for McLachlin) and
across the four areas of law indicated above. The closest to a clustering is the fact that
four such cases occurred in the two calendar years of 1990 and 1991, but even this
correlation is weakened if one refers to Court terms instead of calendar years. The
seriatim style is normally not an individual effort; almost two-thirds of all
concurrences, but less than two-fifths of seriatim concurrences, are signed by a single
judge.
What could explain this phenomenon? Should we be thinking in terms of a
failure of communication at the post-hearing conference, such that two different
judges think they have been assigned the judgment? That understanding is simply not
credible, the more so because there are another sixty-six examples of this practice that
involve dissents. It seems more likely that the conference revealed two strongly held
views with a number of judges expressing either uncertainty between the two, or
choosing between them with some obvious hesitation. None of the accounts of what
happens at conferencenot Justice Wilsons 1986 article about decision making at
the Court,111 Ian Greenes account drawn from interviews with seven of the sitting
judges of the Court,112 nor Donald R. Songers more recent description113suggest
that it is acceptable for a judge to abstain while waiting to be persuaded. Therefore,
108 Hossack Estate v. Hertz Drive Yourself Stations of Ontario Ltd., [1966] S.C.R. 28, 54 D.L.R. (2d)
109 At first glance, this seems to be a massively wasteful duplication of effort; but I think the practice
of writing in full seriatim mode can be understood in terms of the formalist approach to decision
making, which assumes that there is a single correct answer to any legal question that can be
rigorously and objectively derived from purely legal materials. If this is the case, then having each
judge run their own separate analysis is the judicial equivalent of seeking a second opinion in a
medical diagnosis, only with the additional opinions being generated simultaneously rather than
sequentially. The perfect result would be to have all the judges on the panel not only reach the same
outcome, but do so by drawing on the same set of authorities in strikingly parallel ways. It is only if
the panel of judges is seen as negotiating within a zone of discretion that multiple and repetitive
reasons are counterproductive.
110 Sixteen if we include Comit paritaire, which has two free-standing sets of reasons both signed
148.
by a majority of the panel (supra note 106).
111 Supra note 2.
112 Supra note 2.
113 Supra note 2.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
162
perhaps the first of my scenarios is too fanciful to survive. But the second is enough
to explain why both the nominal judgment writer and a strong challenger, sensing real
uncertainty among some potential swing votes, would invest time and effort in
writing a set of reasons in full judgment format.
It is difficult to tell how many of these challenges succeeded. It could be that
none did, that the original bloc of judges held firm and that the challenger wound up
simply writing an unusual separate concurrence with extraneous material included;
but I find this unlikely, the more so because the sixty-three swing concurrences
mentioned above show that judges do change their minds after conference. Or it
could be that every single challenge succeeded, and on all fifteen occasions the
original majority writer wound up in minority; but I find this equally unlikely.
This is a significant category of disagreement because it shows that delivering
reasons for the Court is a contested territory even at the time of writing these reasons.
Combined with the number of swing concurrences, it also suggests a reason why
judges are willing to invest time and effort in writing separate reasons: they do so
because they have a not-insignificant chance of prevailing, of persuading enough of
their colleagues that their reasons should become the judgment of the Court.
H. Other
In any categorization, there are always some leftover cases. In this case, I am
relieved that this category is needed for barely 2 per cent of the total concurrences.
For almost all of these concurrences, I am unable to detect any reason for separate
reasons having been written at allthe writer seems to do nothing more than echo
some or all of the major elements of the majority reasons. On two occasions,114 a
judge wrote to indicate agreement with the ideas of another judge who was writing a
separate concurrence; again, a signature on those reasons would have made the same
point.
Only once, in R. v. Burlingham,115 was a separate concurrence written simply to
take stronger exception to the ideas in a dissent than that expressed in the majority
reasons. In Burlingham, the only example from the twenty-five concurrences to have
a clear substantive content, Justice LHeureux-Dub wrote an individual dissent
severely criticizing the majority for reverting to a formalist approach to decision
making (reminiscent of … R. v. Wray, [1971] S.C.R. 272),116 rather than the
contextual approach that has characterized the style of the modern Court. This
assertion provoked every judge in the majority except one to sign separate reasons
114 Air Canada v. British Columbia, [1989] 1 S.C.R. 1161, 59 D.L.R. (4th) 161; R. v. Tutton, [1989]
1 S.C.R. 1392, 35 O.A.C. 1.
115 [1995] 2 S.C.R. 206, 124 D.L.R. (4th) 7 [cited to S.C.R.].
116 Ibid. at para. 107.
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
vigorously and categorically refuting this accusation.117 The case is significant for
acknowledging and reinforcing the shift from formalist to contextual decision
making, and also demonstrates the frequent marginality of Justice LHeureux-Dub.
Clearly, these separate concurrences do not amount to very much. If all
concurrences looked like these, there would be no reason to study them; only
Burlingham is even mildly interesting.
163
Conclusion
I have tried to show that separate concurrences are a regular and ongoing aspect
of the work of the Supreme Court of Canada, whose judges invest a considerable
amount of time and effort in preparing them. I have also sought to show that separate
concurrences are a normal and routine part of the output of the Court, such that all
judges participate to a greater or lesser degree, and such that all the major
components of the Courts caseload provide the opportunity for this type of
disagreement. I have tried to show that writing separate concurrences is a rational
strategy, given that about one-tenth of separate concurrences can sway the signatures
of enough members of the panel to become the judgment of the Court (and another
tenth of separate concurrences come within a single signature of accomplishing this).
And I have tried to show that there are just over a half-dozen distinctive types of
separate concurrences of escalating importance, these types being routinely self-
labelled by the judges of the Court in a consistent and transparent manner. What
remains is to put judicial disagreement in general, and the separate-concurrence mode
of disagreement in particular, in a broader frame.
Mitchel Lasser has suggested that courts in the modern world are pulled between
two equally important yet mutually exclusive expectations.118 On the one hand, courts
are expected to follow the rules that have been laid down, to be the mouthpiece of the
law rather than a source of the law, and to impartially apply rules, standards, and
demands that are not of their own making. The formalist viewthat there is a correct
solution to all legal issues that can be derived by legal professionals through formal
techniques to generate an objective and replicable resultis no longer fully
convincing, even to the judges themselves. But the desire lives on and is reflected in
the wording of many judicial decisions. On the other hand, courts are expected to pay
attention to the policy consequences of their decisions, to consider the significance of
social conditions that change faster than the laws themselves, and to take special
circumstances into appropriate consideration even when these events cut in a different
117 Ibid. at 287-97, Sopinka J. The curious implication is that sometimes judges will sign on to more
than one set of reasons; in fact, this has occurred less than a dozen times in twenty years, usually
involving only a single judge, so I see no reason to break it out for separate discussion.
118 Mitchel de S.-O.-LE. Lasser, Judicial Deliberations: A Comparative Analysis of Judicial
Transparency and Legitimacy (Oxford: Oxford University Press, 2004) at 63.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
164
direction than the text of the rules. Modern courts balance these conflicting demands
in a variety of ways.
Lassers work explores this idea by examining three appellate courts that
exemplify distinct systems: the Supreme Court of the United States, the French Cour
de cassation, and the European Court of Justice. Lasser argues that the American
solution (which is also in large part the Canadian solution) is to have judges take
responsibility for their decisions through extensive, authored reasons for their
decisions, and to have these reasons include a dynamic consideration of both sets of
demands and a finding that strikes an appropriate balance between these forces in the
immediate case.119 I will take the argument a step further: the written expression of
disagreement, in the form of both dissent and separate concurrence, is a further part of
this process of assuring the concerned public that there has been serious consideration
of a wide range of materials, ideas, and approaches. The reasons why the winner must
win are often accompanied, in the official text, by suggestions that there was much to
be said for the losers concerns as well. Thus, the formal certainty of the statement of
legal doctrine by the majority is subtly shadowed by minority statements of
alternative tracks and options, and the archival embodiment of these subordinated
concerns in the same document as the winners vindication keeps these ideas in
play.120
Because the two aforementioned expectations are in unavoidable tension, they
constantly conflict with one another; the appearance of following the rules laid down,
of observing binding precedent, of staying within judicial tests created in earlier
decisions, suggests rigidity, uniformity, and inflexibility. But the often-heated give
and take of judicial disagreement and the deployment of on the other hand
arguments backed by judicial and academic citations constantly threaten this stability
with the potential for judicial discretion and creativity, risking the appearance of a
court that is simply making it up as it goes along.121 Both extremes are constantly in
play, as both, if isolated from one another, are intolerable. While neither can take over
the judicial function, neither can be conclusively defeated without serious
consequences.
But, as noted above, it would seem that I am plotting the cartography of a rapidly
shrinking landscape. The numberssix hundred concurrences in twenty-three
yearssuggest vigorous ongoing activity, but in fact this is overwhelmed by the ebb
and flow that saw a steady rise in the use of separate concurrence to a dramatically
high peak in 19951996, followed by steady decline. The Supreme Court of Canada,
119 Ibid. at 302-03.
120 About one-tenth of all Court citations to its own prior decisions are to minority opinions. See
Peter McCormick, Second Thoughts: Supreme Court Citation of Dissents & Separate Concurrences,
1949[1999] (2002) 81 Can. Bar Rev. 369 at 375.
121 Paul Weiler, In the Last Resort: A Critical Study of the Supreme Court of Canada (Toronto:
Carswell, 1974) at 207.
165
P. MCCORMICK THE MODERN SUPREME COURT OF CANADA
2008]
which once delivered twelve separate concurrences on a single day,122 and eighteen
separate concurrences on two consecutive days,123 has over the last three calendar
years averaged only eight concurrences every year. Part of this is explained by a
caseload that has fallen from a high of 136 cases in one calendar year124 to its current
level of less than sixty every calendar year125 (a phenomenon curiously paralleled by
the Supreme Court of the United States),126 but even on a relative rather than absolute
basis, separate concurrence is declining. Taking word count as a reasonable surrogate
measure, the Supreme Court of Canada under Chief Justice Lamer devoted more than
one-ninth of its efforts to writing concurrences; under Chief Justice McLachlin the
figure is just over one-twentieth. In what context should we place this decline?
M.T. Henderson suggests that we should not think of the expression of judicial
disagreement as something that is idiosyncratic, accidental, personality driven,127 or
even explainable purely in terms of internal variables.128 Instead, he sees this activity
as representing a strategy on the part of courts to respond to threats or opportunities in
the external environment in order to better position these institutions for survival and
influence. His first North American example (U.S. Chief Justice John Marshalls
displacement of seriatim practices in favour of unanimous and often anonymous
expressions of doctrine) is presented as a strategy for building an institutional space
for the Supreme Court of the United States in a new democratic society, where clarity
and consistency in articulating a generally settled, if still developing, vision were
critical.129 His second North American example, the fragmentation of the same courts
122 This occurred on 25 January 1991. See R. v. Ratti, [1991] 1 S.C.R. 68, 44 O.A.C. 161; R. v.
Romeo, [1991] 1 S.C.R. 86, 110 N.B.R. (2d) 57; Committee for the Commonwealth of Canada v.
Canada, [1991] 1 S.C.R. 139, 77 D.L.R. (4th) 385; R. v. Landry, [1991] 1 S.C.R. 99, 119 N.R. 293.
123 This occurred on 2930 September 1994. See R. v. Bartle, [1994] 3 S.C.R. 173, 118 D.L.R. (4th)
83; R. v. Cobham, [1994] 3 S.C.R. 360, 118 D.L.R. (4th) 301; R. v. Matheson, [1994] 3 S.C.R. 328,
118 D.L.R. (4th) 323; R. v. Harper, [1994] 3 S.C.R. 343, 118 D.L.R. (4th) 312; R. v. Prosper, [1994] 3
S.C.R. 236, 118 D.L.R. (4th) 154; R. v. Pozwak, [1994] 3 S.C.R. 310, 118 D.L.R. (4th) 205; R. v.
Swolinski, [1994] 3 S.C.R. 481, 119 D.L.R. (4th) 309; R. v. Bordon, [1994] 3 S.C.R. 145, 119 D.L.R.
(4th) 74; Hodgkinson v. Simms, [1994] 3 S.C.R. 377, 117 D.L.R. (4th) 161; R. v. Daviault, [1994] 3
S.C.R. 63, 118 D.L.R. (4th) 469.
124 This occurred in 1993.
125 Based on neutral citation, 59 cases were decided in 2006 while 55 were decided in 2007.
126 See e.g. David M. OBrien, A Diminished Plenary Docket Judicature 89:3 (November-
December 2005) 134.
127 See e.g. Philip Cooper, Battles on the Bench: Conflict Inside the Supreme Court (Lexington, Ky.:
University of Kentucky Press, 1995) (describing disagreement on the Supreme Court of the United
States in terms of personality clashes between pairs of justices).
128 From Seriatim to Consensus and Back Again: A Theory of Dissent (2007) [unpublished],
online: Social Science Research Network,
How Many Judges Does it Take to Make a Supreme Court? And Other Essays on Law and the
Constitution (Lawrence, Kan.: University Press of Kansas, 2006) at 42, 35, 38 (emphasizing
Marshalls institution-building accomplishments, and the important role played by single-opinion
decisions).
129 Henderson, ibid. at 8ff.
[Vol. 53
MCGILL LAW JOURNAL / REVUE DE DROIT DE MCGILL
166
decision writing after the mid-1930s, is explained in terms of the juxtaposition of the
new legal realismunconvinced by a formalism that sought to validate itself by its
own unanimitywith the new and controversial challenges faced by the Court before
and after World War II.130 He argues that divided decisions demonstrate a court that is
both open to a variety of arguments (and that therefore mollifies the losing side) and
willing to change its mind over time (which keeps the losing side in the game).131
Both enhance long-term legitimacy in this new landscape.
Perhaps we should make a parallel case for the Supreme Court of Canada. During
the first twenty years of the Charter, as the Court sought to flesh out a new
constitutional document and as Canadian politicians and citizens received a crash
course in judicial policy involvement, the Court (like the Supreme Court of the
United States during the New Deal and the rights revolution) needed to present a
variety of faces, an openness to new ideas, and a willingness at times to reconsider its
initial positions. This need was served by an extensive number of multiple-opinion
decisions, and in particular by a separate-concurrence rate that grew to exceed that of
dissents. Separate concurrences are particularly useful in this context because they
resist a simple winnerloser dichotomy and suggest a variety of directions in which
the prevailing point of view could be nudged in future. But the declining caseload of
the Supreme Court of Canada in this century, as well as its declining rates of
disagreement in general and separate concurrence in particular, suggests that this
dynamic period of flux and change has come to an end. Although the Charter is not
yet completely explored ground, it is nonetheless true that most of the major
questions have been answered; as a result, fewer big questions are coming before
the Court, and fewer policy-divergent responses need to be generated to prepare the
field within which these can be managed. To be sure, this is a big conclusion to hang
on two decades of caseload numbers; the point is to suggest that we need to look
beyond specific personalities or the impact of a chief justice, or even the inner
workings of the Court, to broader notions of the judicial function in evolving
contexts.
130 Ibid. at 28ff.
131 Ibid. at 32ff.
