Canadian Electoral Boundaries and the Courts: Practices,
Principles and Problems
David Johnson*
This case comment offers a critical assess-
ment of Reference Re Provincial Electoral
Boundaries (Sask.), decided by the Supreme
Court of Canada in the summer of 1991. This
significant and controversial case marks the
first Charter appeal dealing with the constitu-
tionality of a Canadian electoral distribution
system. The case is significant in that it deals
with the validity of the ways and means by
which electoral districts are created in this
country, the meaning of the right to vote found
in the Charter, the constitutional weight to be
accorded to the principle of “one person –
one vote”, and the extent to which factors
other than population can be used to structure
such districts. The case is controversial, how-
ever, in that the majority decision was anti-
quated and problematic, leading the Court to
uphold a pluralist approach to distribution
which was narrow, questionable, and in certain
respects, unprincipled. The majority was will-
ing to interpret the right to vote in light of a
number of factors, only one of which was
rooted to the principle of “one person –
one
vote”. By giving constitutional weight to other
factors such as geography, access to communi-
cations, minority representation and commtini-
ties of interest, the Court provided legitimacy
to factors which are questionable at best and
unprincipled at worst. This decision, far from
offering persuasive guidance in addressing the
issue in dispute, will likely result in more dis-
tribution cases flowing to the courts.
Ce commentaire d’arrt propose une ana-
lyse critique de ]a d6cision rendue en 1991 par
la Cour suprme du Canada dans l’affaire Ren-
voi relatif aux circonscriptions electorales
provinciales (Sask.). Cette decision, ” ]a fois
importante et controversde, est la premiere at
aborder la constitutionnalit6 du syst~me de
r6partition des districts 6lectoraux au Canada
en vertu de la Charte. C’est une d6cision
importante dans ]a mesure oit elle discute de la
validitd des moyens par lesquels les districts
6lectoraux sont cr 6s au pays, de ]a significa-
tion du droit de vote 6nonc6 dans la Charte, du
poids constitutionnel devant 8tre accord6 au
principe une personne –
un vote et de l’uti-
lit6 de certains facteurs, autres que la popula-
tion, pour organiser les divers districts 6lecto-
raux. I1 s’agit cependant d’une d6cision con-
trovers~e. En effet, le test de ]a majorit6, au
demeurant d6pass6 et probl6matique, mani-
feste une approche pluraliste de ]a rdpartition,
approche qui est restrictive, discutable et, At
certains 6gards, mal fond~e. Une majoritd de
juges a interprdt6 le droit de vote A la lumi~re
d’un certain nombre de facteurs dont un seul
d6coule du principe
aux autres facteurs tels que ]a g6ographie, l’ac-
c~s aux communications, la repr6sentation des
minorit6s et les communautds d’int&rts, la
Cour supreme accorde une 16gitimitd At des fac-
teurs qui sont, au mieux, discutables, sinon
sans fondement. Cette d~cision, loin de pr6-
senter une solution persuasive pour l’analyse
des questions de r6partition, aura probable-
ment pour effet d’engendrer une recrudescence
de cas de ce type devant les tribunaux.
* Assistant Professor, Department of Political Science, University College of Cape Breton.
McGill Law Journal 1994
Revue de droit de McGill
To be cited as: (1994) 39 McGill L.J. 224
Mode de r6f6rence: (1994) 39 R.D. McGill 224
1994]
CASE COMMENTS
Synopsis
Introduction
I.
Political and Legal Origins
A. Sister Province; Related Challenge
B. Judgement in Regina
C. Judgement in Ottawa
II. The Judgements in Retrospect
A. Territory and Communications
B. Community of Interest
C. Minority Representation
D. Deviation Quotients
Final Reflections
Introduction
The issue of representation within legislative assemblies has long been a
matter of concern in this country, and in recent years questions respecting the
quality of such representation have, in fact, spawned a number of constitutional
challenges culminating in the decision of the Supreme Court of Canada in Ref-
erence Re Provincial Electoral Boundaries.1 When this case was heard in the
spring of 1991, the Court was confronted with its first Charter appeal dealing
with the constitutionality of a Canadian electoral distribution system. While this
subject matter has, unfortunately, not been one to excite the passions of most
political scientists or scholars of constitutional law, the issues at stake in this
case were, and still are, highly important; so much so that this decision can be
viewed as the most significant yet to be rendered by the Supreme Court of Can-
ada on the nature and functioning of the parliamentary process in this country.
The significance of the decision stems from its scope. In Reference Re Pro-
vincial Electoral Boundaries, the Supreme Court of Canada was called to rule
upon the validity of the ways and means by which electoral districts are created
in this country, the meaning of the right to vote found in the Charter,3 the con-
‘Reference Re Provincial Electoral Boundaries (Sask.), [1991] 2 S.C.R. 158, (sub nom. Refer-
ence Re Electoral Boundaries Commission Act, ss. 14, 20 (Sask.)) 81 D.L.R. (4th) 16 [hereinafter
Reference Re Provincial Electoral Boundaries cited to S.C.R.].
2Canadian 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 [hereinafter Charter].
31bid., s. 3.
REVUE DE DROIT DE McGILL
(Vol. 39
stitutional weight to be accorded to the principle of “one person –
one vote”,
and the extent to which factors other than population can be used to structure
the size of such districts. The Court was thus given the task of reviewing and
assessing certain of the fundamental aspects of the system of parliamentary rep-
resentation found in Canada.
The controversy with this decision, in turn, stems from the substantive con-
clusions reached in regard to the two competing approaches to distribution pres-
ent in the case. While problems appear in the decisions of both the Saskatche-
wan Court of Appeal and the Supreme Court of Canada, the majority decision
of the Supreme Court has serious weaknesses, as will be illustrated in this paper.
Its ruling respecting the meaning of the right to vote was antiquated and highly
problematic, leading the Court to uphold a traditional, pluralist approach to dis-
tribution which was narrow, questionable and, in certain respects, unprincipled.
The majority was willing to interpret the right to vote in light of a number of
factors, only one of which was rooted to the equality principle of “one person
– one vote”. The majority was supportive of the contention that, in the drafting
of electoral boundaries, governments may take into consideration certain factors
other than population, such as geography, access to communications, minority
representation and communities of interest. It was the acceptance of these fac-
tors as valid guides to determining distribution which elicits problems.
Such factors in relation to distribution are questionable at best and unprin-
cipled at worst. The bestowal of constitutional validity upon the factor of com-
munity of interest was acceptance of a factor so hopelessly indeterminate as to
be worthless as a guiding principle in distribution. Similarly, the acceptance of
minority representation as a factor which may need to be incorporated into
decision-making respecting distribution is a development which raises a host of
practical and theoretical problems related directly to the nature of the system of
representation in this country. Through reflection on the issues enveloping the
process of distribution it will be asserted that the egalitarian approach taken by
the Saskatchewan Court of Appeal, in upholding the primacy of population in
devising electoral boundaries, offers a far more principled and practical
approach to distribution than that found in either of the majority or minority
decisions of the Supreme Court of Canada. From this conclusion it will be
argued that Reference Re Provincial Electoral Boundaries marks not the end of
constitutional litigation on this subject, but only the beginning of a process of
litigation that will continue until the Court sets a principled and practical
approach to the drawing of electoral boundaries.
I.
Political and Legal Origins
The dispute in Reference Re Provincial Electoral Boundaries centred upon
a challenge to the constitutional validity of Saskatchewan’s Representation Act,
1989 and Electoral Boundaries Commission Act.4 Under the latter statute the
provincial legislature established a boundaries commission and instructed it to
4See The Representation Act, 1989, S.S. 1989-90, c. R-20.2; The Electoral Boundaries Conmis-
sion Act, S.S. 1986-87-88, c. E-6.1.
1994]
CHRONIQUES DE JURISPRUDENCE
create 66 electoral ridings, of which 29 were to represent urban centres and 35
were to represent rural areas, with two ridings designed to cover the far north
of the province. As is common with Canadian electoral boundaries legislation,
the Commission was entitled to vary the population of the southern ridings by
plus or minus 25% of the average population size for southern ridings so as to
take account of such factors as geography, mode of communications and com-
munity of interest. Under the Representation Act, 1989, it was recommended
that 66 ridings be varied. This proposed set of electoral districts, though,
quickly came under attack as did the Commission itself.
With regard to the Commission, a number of critics5 impugned its legiti-
macy on the grounds that its enabling legislation dictated that it must establish
a set number of rural and urban ridings regardless of its own judgement as to
how the number of seats should be apportioned between rural and urban areas.
As such, the argument was raised that the discretion of the Commission had
been fettered. While these concerns became significant, most critical attention
was devoted to the substantive outcome of the Commission’s work. The critics
attacked what they considered to be the debasement of the right to vote through
the proposed creation of constituencies with quite widely varying populations.6
Through the use of the 25% variance rule, of the 64 southern ridings, 34 were
within plus or minus 10% of the southern average, 10 ridings exceeded the aver-
age by between 10% to 15%, 12 ridings exceeded the average by between 15%
to 20%, and a total of 8 ridings exceeded the average by a figure of more than
20%, but less than 25%. Fully 20 ridings, 31% of all southern ridings, exceeded
the southern average by more than 15%, which was the stipulated maximum
allowable deviation from the most recent electoral distribution.7 The effect of
these variances, so critics argued, was the devaluation and debasement of the
worth of votes held by citizens living in constituencies with relatively large
electorates. Roger Carter noted that the southern riding with the smallest elec-
torate – Saskatoon Sutherland University, with a population of 7684 – was
only 61% of the size of the riding with the largest electorate – Saskatoon Grey-
stone, with a population of 12,567. As such, it would take at most only 3843
voters to elect a representative in the former riding, whereas the corresponding
figure for the latter riding would be 6284. The effect of this population imbal-
ance on individual votes would be striking: 100 votes in Saskatoon Sutherland
University would carry the same electoral weight as 164 votes in Saskatoon
Greystone. In other words, a single vote in the smaller riding carried 63.5%
more electoral weight than a single vote in the larger riding.’ This dynamic of
5Certain of the leading critics who eventually received intervener standing before the Supreme
Court of Canada included such groups as Equal Justice for All and the British Columbia Civil Lib-
erties Association. Individual interveners included Professor John F. Conway of the University of
Regina. Given that this case developed as a reference, the court-appointed counsel for the critics
was Mr. Roger Carter, Q.C.
6See Respondent’s Factum in Reference Re Provincial Electoral Boundaries at 3-7 [hereinafter
Respondent’s Factum]; Factum of Intervener John F. Conway in Reference Re Provincial Electoral
Boundaries at 20-24 [hereinafter Factum of John Conway]; Factum of Intervener Equal Justice for
All in Reference Re Provincial Electoral Boundaries at 2-5.
7Saskatchewan, Final Report (Regina: Electoral Boundaries Commission, 1988) at 9-10.
8Respondent’s Factum, supra note 6 at 13.
McGILL LAW JOURNAL
[Vol. 39
vote weight imbalance on account of variances in population was found
throughout the province.
Beyond such examples of the discrepancies in constituency electorate size
and vote weight, the critics argued that the distribution created a rural represen-
tational bias. The Boundaries Commission recommended the creation of 29
urban ridings with an average elector population of 10,998, while the average
elector population of the 35 rural ridings was 9,637. The value of the urban vote,
it was then argued, was thereby less equal, thereby debased, in relation to the
rural vote. The critics made the point that, on average, the rural vote carried
14% more electoral weight than an urban vote.9 In his factum to the Supreme
Court of Canada, John Conway went so far as to contend that the proposed dis-
tribution constituted a gerrymander, designed by the then ruling Progressive
Conservative government to maximize its electoral support amongst a rural
population which was historically more predisposed than the urban electorate to
support the Progressive Conservative party.'”
All of these findings led to the criticism that the proposed distribution was
unjust and unconstitutional. Carter, Conway and Equal Justice for All, amongst
others, all challenged the mandate of the Commission and its substantive recom-
mendations as being in violation of the Charter. It was alleged that the set rules
of distribution found in the enabling legislation, as well as the given distribution
itself, violated the right to vote provision of the Charter in that the riding system
would be created in such a manner that the voting power of citizens of Saskatch-
ewan would vary depending on where they lived, resulting in the effective
worth of votes not being equal throughout the province. This, it was argued, vio-
lated the principle of vote equality which the challengers held to be implicit in
the concept of “one person –
one vote”, a concept which they also held to be
fundamental to the right to vote entrenched in the Charter.”
A. Sister Province; Related Challenge
This critique of the merits of the proposed electoral boundary system and
the aforementioned Charter challenge were not unprecedented. For example, in
1989, the constitutionality of British Columbia’s existing electoral boundary
system was challenged in Dixon v. British Columbia (A.G.)”2 on similar grounds
to those raised in Reference Re Provincial Electoral Boundaries. In the decision
of the British Columbia Supreme Court on this matter, the Court found the
impugned distribution system to be unconstitutional as it was held to be a vio-
lation of the right to vote provision within the Charter. In a judgement written
by Chief Justice Beverly McLachlin (as she then was), the Court recognized that
9Factum of John Conway, supra note 6 at 1-2. It is to be noted, however, that this 14% discrep-
ancy did not fully carry over into the distribution of seats between rural and urban Saskatchewan.
Through the proposed distribution, rural Saskatchewan was granted 53% of legislative seats rep-
resenting 50% of the electorate, while urban Saskatchewan was granted 44% of legislative seats
representing 47% of the electorate.
‘Factum of John Conway, ibid. at 16-19.
“Respondent’s Factum, supra note 6 at 3-7; Factum of John Conway, ibid. at 20-25.
12(1989), 59 D.L.R. (4th) 247, [1989] 3 W.W.R. 393 (B.C.S.C.) [hereinafter Dixon cited to
D.L.R.I.
19941
CASE COMMENTS
while the principle of vote equality was central to the right to vote as found in
the Charter, this right did not guarantee all citizens perfect mathematical equal-
ity of vote strength in relation to the votes of all other citizens. The Court
accepted that the elector population size of ridings could deviate from the pro-
vincial average so as to take account of such factors as population shifts over
time, geographic considerations, the quality of communications in territorially
large ridings, and the concept of community of interest. In addressing the issue
of what would constitute an acceptable range of deviation from the principle of
equality, McLachlin C.J. accepted the proposition that deviations up to plus or
minus 25% from the provincial average would be tolerable if, through these
deviations, other important concerns relating to the factors listed above were
incorporated into the distribution system.
Despite this acceptance of a generous deviation rule, McLachlin C.J. nev-
ertheless found the British Columbian boundary system unconstitutional due to
its excessive and unwarranted violation of this rule. The Court found that only
7 of 52 ridings were within 10% of the provincial average. Of those in excess
of this figure, fully 9 ridings had electorates more than 25% below the provin-
cial average, while 10 ridings had electorates more than 25% above the average.
As McLachlin C.J. noted, “[Sluch anomalies cannot but suggest a gross viola-
tion of the fundamental concept of representation by population which is the
foundation of our political system.””3 The Court proceeded to rule this entire
distribution of seats unconstitutional, though providing the provincial govern-
ment time to redistribute its legislative seats in accordance with the 25% devi-
ation rule.
B. Judgement in Regina
Because of the criticisms raised against the Saskatchewan distribution
legislation, the provincial government, in 1990, referred questions to the Sas-
katchewan Court of Appeal respecting the constitutionality of the mandate of
the Boundaries Commission and of the distribution of seats recommended by
the Commission. In a decision rendered during March of 1991, the Court, in a
unanimous decision, accepted the validity of the challenges and ruled unconsti-
tutional the Commission’s general mandate, as well as the bulk of its substantive
conclusions.
4
In reaching this decision, the Court of Appeal stressed that in assessing dis-
tribution systems in light of the right to vote provision in the Charter, “the con-
trolling and dominant consideration in drawing electoral constituency bounda-
ries must be voter population in the province.”‘ 5 The guiding ideal giving mean-
ing to the right to vote and its operationalization through distribution systems,
one vote”. 6 In theo-
according to the Court, was the principle of “one person –
131bid. at 267-68.
“‘Reference Re Electoral Boundaries Commission Act (Sask), ss. 14, 20 (1991), 78 D.L.R. (4th)
449, (sub nom. Reference Re Provincial Electoral Boundaries) [19911 3 W.W.R. 593 (Sask. C.A.)
[hereinafter Reference Re Electoral Boundaries Commission Act cited to D.L.R.].
5Ibid. at 463.
161bid. at 460.
REVUE DE DROIT DE McGILL
[Vol. 39
ry, all electors were to be treated equally, all were to possess a vote of compa-
rable worth. 7 The Court of Appeal was quick to assert, however, that the notion
of absolute equality in distribution was undesirable due to the practical impos-
sibility of devising and maintaining electoral districts with absolutely equal
electorates. Some deviations from the provincial average were then counte-
nanced as a practical necessity, but the Court stipulated that these deviations
must be as limited as practically possible. In creating electoral boundaries, the
legislature was ordered to apportion seats “to constituencies of substantially
equal voter population.”‘8 With respect to the Commission and its work, the
Court of Appeal was condemnatory. Though it held that northern Saskatchewan
required special attention in distribution due to the vastness of its territory and
the difficulties this posed to elected representatives in effectively communicat-
ing and interacting with their constituents, the Court was unwilling to view rural
Saskatchewan from a similar perspective. The Court found that due to those
provisions of its mandate requiring it to create a set number of urban and rural
constituencies, the Commission was fettered in its ability to implement the prin-
ciple of substantial elector equality amongst all such ridings. Socio-geographic
considerations regarding the uniqueness of rural areas, their special communi-
ties of interest, and the difficulties faced by elected representatives in providing
services to territorially large ridings thus came to interfere with the controlling
and dominant consideration of population, with such interference being viewed
as unconstitutional. 9
Despite the arguments of the Attorney General of Saskatchewan respecting
the importance of these socio-geographic considerations, the Court rejected the
contention that rural Saskatchewan deserved any special treatment in distribu-
tion on account of its socio-geographic nature. The Court held that the interests
and concerns of urban areas were as valid and broad as those of rural Saskatch-
ewan, and that it was fallacious to assert that elected representatives of one type
of area are ignorant and insensitive to the concerns affecting other types of
areas.2” The Court stressed, furthermore, the importance which modem technol-
ogies of communications have had in facilitating the ability of persons living in
rural areas to interact with their elected representatives, in particular, and with
government offices, in general. Indeed, the Court argued that when there is seri-
ous concern respecting the ability of a representative to effectively serve a ter-
ritorially large area, policy instruments other than over-representation present
themselves, such as “additional travel allowances, support staff, and ‘up-to-
date’ communication services.”‘”
With regard to the actual distribution, the Court observed the wide devia-
tions from the general average as found in numerous ridings and ruled that these
“egregious” deviations violated the principle of substantial equality.22 The Court
’71bid. at 460-62.
181bid. at 463.
’91bid. at 477-81.
2″Ibid. at 479-80.
2 ‘Ibid. at 480.
221bid. at 478.
1994]
CHRONIQUES DE JURISPRUDENCE
held, moreover, that these violations could not be justified as a reasonable limi-
tation to voting rights under section 1 of the Charter. The Crown had not been
able to demonstrate to the satisfaction of the Court that the interests of rural Sas-
katchewan were so unique from those of urban Saskatchewan, and so endan-
gered by a distribution undertaken in accordance with the principle of substan-
tial equality, as to be “pressing and substantial”, thus deserving special
consideration under the Charter.n Rather, the Court reasserted that any special
communications problems faced by rural residents and their representatives
could be resolved “by other, non-infringing and equally effective methods” of
communication.24
C. Judgement in Ottawa
Following the release of the Court of Appeal’s decision, the government of
Saskatchewan, as expected, moved to appeal this judgement to the Supreme
Court of Canada. On account of the substantive nature of the decision, the fed-
eral government, as well as a host of provincial and territorial governments,
moved to intervene on behalf of the Saskatchewan government.’
If the Sas-
katchewan Court of Appeal was correct, and the province’s boundary system
was unconstitutional, then many other jurisdictions in the country would be in
legal peril since most distribution systems allowed for significant deviations
from average quotients so as to accommodate interests unrelated to the principle
of substantial elector equality.
The Supreme Court of Canada heard the appeal in late April of 1991 and
rendered its judgement in June. In a split decision the Court overturned the
lower court decision and upheld the constitutionality of the Saskatchewan com-
mission and its work. In writing for the majority, Madame Justice McLachlin (as
she now is) rejected the Court of Appeal’s proposition that, with regard to elec-
toral distribution, the right to vote was to be understood, in principle, as being
a right to substantial equality among votes, thus a right to substantial elector
equality. Rather, McLachlin J. contended that the key principle governing the
right to vote in relation to distribution was a guarantee of “effective represen-
tation”.26 Such representation, according to the majority, would be contingent
upon a number of factors. One such factor was the “relative parity of voting
power”. The relative equality of voting power and consequent relative equality
of electorate size per riding were held to be important matters to be taken into
consideration by distribution commissions.27 But they were not the only factors
defining the right to effective representation.
McLachlin J. was quick to proclaim that “[factors like geography, commu-
nity history, community interests and minority representation may need to be
taken into account to ensure that our legislative assemblies effectively represent
23Ibid. at 478-81.
24Ibid. at 480.
25Intervener governments included the federal government and those of Newfoundland, Prince
26Reference Re Provincial Electoral Boundaries, supra note 1 at 183.
27Ibid.
Edward Island, Quebec, Alberta, British Columbia, the Northwest Territories and the Yukon.
McGILL LAW JOURNAL
[Vol. 39
the diversity of our social mosaic.”2 In order to ensure that the various interests
associated with these factors are reflected in the composition of legislatures, she
held that electoral boundary systems may depart from the principle of substan-
tial elector equality so as to give representation to these other valid interests.
. asserted, moreover, that the factors which she enumerated were
McLachlin
not exhaustive: “These are but examples of considerations which may justify
departure from absolute voter parity in the pursuit of more effective represen-
tation; the list is not closed.”29 Deviations from the equality principle, though,
were only to be upheld on the basis that they provided more effective represen-
tation. As Madame Justice McLachlin stated:
I adhere to the proposition asserted in Dixon, … that only those deviations should
be admitted which can be justified on the ground that they contribute to better gov-
ernment of the populace as a whole, giving due weight to regional issues within
the populace and geographic factors within the territory governed.30
Furthermore, McLachlin J. stressed, “[B]eyond this the dilution of one citizen’s
vote as compared with another should not be countenanced. ‘3’
Through this interpretative approach the majority upheld the work of the
Saskatchewan commission as promoting the effective representation of rural,
urban and northern areas of the province. McLachlin J. found that the special
treatment accorded to northern Saskatchewan was necessary to deal with the
transportation and communication difficulties experienced by persons living in
this huge area.32 As for the variations found between rural and urban ridings, she
contended that the distribution was fair, as was the mandate of the Commission
itself. The mandate, according to the majority, simply reflected the historic pat-
tern of electoral representation which had always been used to distribute seats
between rural and urban areas, coupled with an appreciation of demographic
trends in the province. As such, McLachlin J. stressed that the mandate was not
arbitrary and did not threaten the interests of urban electors.33 The majority
noted, in fact, that under the Electoral Boundaries Commission Act, the Com-
mission was instructed to add two additional urban seats to the legislature to
take account of population growth in Saskatoon and Regina.’
The majority furthermore stressed the fact that, given the proposed distri-
bution, rural Saskatchewan was accorded 53% of the legislature’s seats with
50% of the province’s electorate, while urban Saskatchewan was granted 44%
of the seats with 47% of the electorate. This variation was viewed by McLachlin
. to be “relatively small” and ultimately justifiable on the basis of “factors such
as geography, community interest and population growth patterns. 35 With
regard to population variations between specific ridings, McLachlin J. was
again unperturbed. She accepted the legitimacy of the 25% variance rule as
28Ibid. at 184.
291bid.
30Ibid. at 185.
31Ibid.
32Ibid. at 190, 196.
331bid. at 191-92.
34Ibid. at 192.
351bid. at 192, 197.
1994]
CASE COMMENTS
upheld in Dixon and proceeded to justify given deviations on the grounds of the
poorer levels of transportation and communication services in rural areas, as
well as on the basis “that rural voters make greater demands on their elected rep-
resentatives, whether because of the absence of alternative resources to be found
in urban centres or for other reasons.”36 Given this understanding of rural Sas-
katchewan, the majority consequently asserted that “the goal of effective repre-
sentation may justify somewhat lower voter populations in rural areas.”37 As
such, no violation of the Charter right to vote was found. 8
In dissent were Chief Justice Lamer and Justices L’Heureux-Dub6 and
Cory. Their judgement, authored by Cory J., affirmed many of the conclusions
reached by the Saskatchewan Court of Appeal, though it offered some interest-
ing variations. The dissenters accepted that “representation of community inter-
ests” and “geographic considerations” may need to be given attention in distri-
bution, but held that the key principle in interpreting the right to vote and in
drafting electoral boundaries is necessarily “that each vote must be relatively
equal to every other vote.”39 In assessing the nature and work of the Commis-
sion, Cory J. attacked its mandate for dictating that it establish a set number of
rural and urban seats regardless of the Commission’s independent appraisal of
what would constitute a just distribution.’ In addition, the minority attacked the
25% variance rule permitted in the legislation and used by the Commission.
According to Cory J., a 15% deviation should have been the maximum accept-
able figure to provide for the representation of special factors while ensuring the
relative equality of riding electorates and vote strength. The minority noted that
the most recent Saskatchewan distribution of 1981 had resulted in all ridings
existing within a 15% deviation from the general provincial average. On
account of this, Cory J. asserted that he was presented with no pressing and sub-
stantial reasons from the government of Saskatchewan as to why the current dis-
tribution had to allow for significant exceptions from this 15% standard.4′
Accordingly, the minority found that the impugned legislation did not constitute
a reasonable limit on the right to vote enshrined in the Charter.
II. The Judgements in Retrospect
A review of these decisions reveals that the courts were divided between
two significantly different approaches to electoral distribution. The egalitarian
approach, as advocated by the challengers to the Saskatchewan legislation and
as upheld by the Court of Appeal, and further supported in modified form by
36Ibid. at 195.
371bid.
38Mr. Justice Sopinka wrote a short judgement concurring with McLachlin J. on most points,
while elaborating on the right of a legislative assembly to establish electoral boundaries commis-
sions with strict guidelines delineating the powers to be held, and the nature of decisions to be ren-
dered, by such commissions. According to Sopinka J., this was in accordance with the sovereign
power of the delegating legislature to construct a subordinate institution as it deems fit (ibid. at
197-99).
391bid. at 165.
40Ibid. at 169-70.
41Ibid. at 172-74.
REVUE DE DRO1T DE McGILL
[Vol. 39
the dissent in the Supreme Court of Canada, is rooted to the principle of indi-
vidual elector equality of opportunity. As Carter and Conway argued in their
factums, the Charter recognizes the equality of all citizens, with all citizens
having the right to vote. Therefore, all such votes should carry a substantially
equal weight, thereby allowing all citizens to possess an equal ability to influ-
ence electoral outcomes and an equal claim to representation from legislative
assemblies.42 This approach is infused with the ideas surrounding the classical
understanding of representation by population: “One Man – One Vote, One
Vote – One Value”.43
In contrast to this, the pluralist approach to distribution, as supported by
the government of Saskatchewan and all other intervener governments, and as
upheld by the majority of the Supreme Court of Canada, is based on the under-
standing that a variety of social factors must be taken into consideration in the
drafting of electoral boundaries. Population is clearly such a factor and advo-
cates of this approach stress that there should not be a gross variance in the pop-
ulation sizes of various constituencies. But, population is only one factor. As the
factum for the Attorney General for Saskatchewan asserted, the key principle to
be promoted in distribution is not the strict equality of voting power and con-
sequent equal riding populations, but the guarantee of “effective representa-
tion”. The achievement of such representation necessitates the incorporation of
a number of social matters into the thinking about distribution.” As the govern-
ments of Saskatchewan and Canada suggested, these matters may include con-
cern over geography and the ability of elected representatives to service geo-
graphically large areas and concern for the representation of distinct commu-
nities with special interests rooted to their historical-geographical-cultural
identity.
An analysis of the reasoning in Reference Re Provincial Electoral Bound-
aries must then devote attention to the merits of these contending approaches.
As will be argued below, the arguments raised in favour of the pluralist
approach are unconvincing, while also raising a host of problems respecting the
theoretical and practical workings of the electoral process within this country.
In contrast, the egalitarian approach offers a relatively uncomplicated, practical
yet principled means of organizing a democratic electoral process.
A. Territory and Communications
Much concern was expressed regarding the degree of importance which
geography should play in distribution. The majority of the Supreme Court of
4 2Respondent’s Factum, supra note 6 at 4-6; Factum of John Conway, supra note 6 at 20-24.
43F. Schindeler, “One Man One Vote: One Vote One Value” (1968) 3 J. Can. Stud. 13. This
approach, moreover, is very much informed by, and related to, congressional distribution doctrine
as found in the United States and upheld by the United States Supreme Court in such landmark
cases as Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691 (1962); Reynolds v. Sims, 377 U.S. 533, 84
S. Ct. 1362 (1964). The doctrine of substantial elector equality can thus be viewed as one having
its ideological origins in the American political tradition.
“tThis is also the posiion held by various academic commentators. See e.g. R. Knopff & F.L.
Morton, Charter Politics (Scarborough, Ont.: Nelson Canada, 1992) c. 12; K. Roach, “Reappor-
tionment in B.C.” (1990) 24 U.B.C. L. Rev. 79 at 80.
19941
CHRONIQUES DE JURISPRUDENCE
Canada accepted the contention raised by the government of Saskatchewan and
all other intervener governments that rural and remote ridings should be
accorded smaller average electorates than urban ridings on account of the fac-
tors of territorial size and attendant problems in transportation and communica-
tion. It has long been recognized that the operationalization of the principle of
substantial elector equality would result in the creation of territorially large elec-
toral districts in rural and remote areas of the country given their sparseness of
population. According to the critics of the substantial equality principle, the
practical effect of this would be the imposition of a hardship on residents of
such areas in that the transportation and communication facilities of these areas
are neither as developed, nor as numerous, nor as easy to access as those in
urban areas. Hence the acceptance by McLachlin J. of the legitimacy of estab-
lishing rural and remote constituencies with often substantially smaller elector-
ates than urban ridings.
This conclusion is reached, however, in disregard to the arguments made
by the Saskatchewan Court of Appeal respecting both the quality of the trans-
portation and communication services actually existing throughout most parts of
rural Saskatchewan and the variety of alternative instruments which can be used
to enhance effective communication between the represented and the represent-
ative. As the Court of Appeal argued, the transportation and communication
system existing in the rural Saskatchewan of the 1990s is far different from that
which existed in the province in its early years of development. The current sys-
tem of transportation is impressive, based upon a network of highways, second-
ary roads and local air services. Similarly, a system of high technology telecom-
munications is as readily accessible in rural as in urban Saskatchewan.4 5 The
implication should be clear. A rural bias in distribution designed to compensate
for weaknesses in the transportation and communication infrastructure is situa-
tionally specific. The practice of rural over-representation emerged in the nine-
teenth century when the quantity and quality of transportation and communica-
tion services were rudimentary in this country. Under these circumstances the
policy of rural over-representation was understandable and can be viewed as
reasonable
4 6
Such a policy, however, is only justified for so long as the infrastructure
in question is weak and underdeveloped. As the infrastructure is improved, the
need to maintain such a policy should be re-evaluated accordingly. To do oth-
erwise is to perpetuate as a valid policy instrument a legal construct designed
for a differing era and differing social conditions. Yet this is what the majority
did; at no time did McLachlin J. address the specific arguments raised by the
Court of Appeal, by Carter or by other interveners, respecting the quality of the
modern transportation and communications system found in Saskatchewan and
the capabilities this system offers to representatives of rural and remote ridings
desiring to enhance their interactions with constituents.
This narrowness of vision is especially evident in McLachlin J.’s treatment
of the alternative means of enhancing communications among elected represent-
45Reference Re Electoral Boundaries Commission Act, supra note 14 at 479.
46T.H. Qualter, The Election Process in Canada (Toronto: McGraw-Hill, 1970) at 94-95.
McGILL LAW JOURNAL
(Vol. 39
atives and their constituents should existing modes of communication be viewed
as problematic.47 The Court of Appeal stressed that in these instances other pol-
icy instruments exist, such as “additional travel allowances, support staff, and
‘up-to-date’ communication services.”’48 The two former instruments can be
viewed as traditional means to enhance the communications capacities of
elected representatives. Additional travel grants would provide members in ter-
ritorially large ridings with the financial power to engage in more movement
within their constituencies than hitherto, thus enabling them to meet with more
of their constituents. The provision of additional support staff, similarly, would
enable these representatives to increase the size of their constituency offices, if
not to establish a number of such offices throughout their ridings, thereby cre-
ating more linkage points at which constituents could meet with the member
and officials working on behalf of the member.
While these instruments should be well understood reform options, the lat-
ter instrument represents a range of technological developments which are new,
innovative and worthy of close analysis. With the development of fax machines,
electronic mail, teleconferencing and inter-active video technology, the modem
communications services available to legislatures desiring to enhance the com-
munications capacities of members in rural and remote areas are impressive,
ever improving and ever more economical. As Bercuson and Cooper have
argued, through such developments the inhibition which geography has posed
to effective communication is being increasingly eliminated, thereby calling
into question the need to provide for rural and remote over-representation as a
means to promote communication between representatives and the repre-
sented.49 Yet McLachlin J. ignored this information and by so doing failed even
to consider whether there are, as the Court of Appeal proclaimed, effective prac-
tical alternative means to improve the communications capacities of represent-
atives without needing to have recourse to a policy which does damage to the
principle of substantial elector equality. The majority of the Supreme Court of
Canada obviously disagreed with the Court of Appeal on this point, yet good
judicial reasoning should have, at the very least, necessitated that the majority
address the matter on its merits.
B. Community of Interest
The majority’s use of the concept of community of interest as a factor inte-
gral to distribution and capable of justifying deviations from the principle of
substantial elector equality is similarly controversial. Madame Justice McLach-
lin accepted that “community interests” may need to be taken into account in
devising electoral boundaries and she proceeded to justify the work of the Com-
mission, in part, on the basis of the special communities of interest found in
rural ridings. Nowhere in the judgement, however, did the majority address the
tough theoretical and practical issues regarding the definition of the concept.
4 7Reference Re Provincial Electoral Boundaries, supra note I at 195.
48Reference Re Electoral Boundaries Commission Act, supra note 14 at 480.
4 9D.J. Bercuson & B. Cooper, “Electoral Boundaries: An Obstacle to Democracy in Alberta” in
J.C. Courtney, P. MacKinnon & D.E. Smith, eds., Drawing Boundaries: Legislatures, Courts and
Electoral Values (Saskatoon: Fifth House, 1992) at 123-24.
19941
CASE COMMENTS
In the most definitive and supportive work to date on the use of the concept
of community of interest in distribution, even Alan Stewart”0 is forced to admit
that there can be no rigorous definition given to the term. Stewart proceeds to
justify the use of the concept on the basis of practical reasoning. Specific com-
munities and their distinct interests will be identifiable when officials vested
with the authority to devise electoral boundaries closely probe the socio-
economic nature of particular areas and observe the patterns of social behaviour
which link or divide certain groups of people from other groups within those
areas.5′ It is just this lack of definition, coupled with rough reasoning, which
makes the concept vague, subject to great debate as to its meaning, and open to
great inconsistency in application.
Stewart contends that the concept must be operationalized in reference to
geographic space, with such space being partitioned so as to encompass distinct
socio-economic communities.” It was this approach, of course, which was
adopted by the majority in Reference Re Provincial Electoral Boundaries.
According to Stewart’s reasoning, electoral ridings in the British parliamentary
system have always been territorially based with elected members being called
upon to represent the interests of the inhabitants of these territories. 53 Patterns
of settlement, moreover, are not considered random. Particular types of people
sharing various interests tend to congregate together for a variety of socio-
economic reasons, leading to certain territories possessing distinctive commu-
nity traits. As distinctive communities can be identified, they may be accorded
special treatment in the drafting of electoral boundaries so as to facilitate the
correspondence of electoral constituencies to these communities. In theory, this
process of distribution is seen as promoting the representation of these commu-
nity interests within the legislative process.’
Though the logic here has a prima facie attractiveness given its symmetry,
it nevertheless falters by failing to define the key concept. The question simply
and starkly becomes, What constitutes a community? Stewart is correct to assert
that there are a host of answers to this question, but is mistaken in arguing that
no useful purpose can be served in seeking to provide a definition of the con-
cept. A review of the sociological literature on community indicates that the
concept is open to a variety of interpretations. MacIver, for example, stressed
that communities are intrinsically bound to territory, with communities being
identifiable through the study of sentiments, values, loyalties and affections held
by people living in particular areas as distinct from those held by others in dif-
fering areas.55 This territorial sense of community, however, has been vigorously
challenged in the past three decades. Broom and Selznick have asserted that
while communities are usually to be understood in relation to territory, a distinct
5 0 ‘Community of Interest in Redistricting” in D. Small, ed., Drawing the Map: Equality and
Efficacy of the Vote in Canadian Electoral Boundary Reform (Toronto: Dundum Press, 1991) at
117-19.
5 11bid. at 128-35.
52 Ibid. at 124-28.
531bid. at 117-18.
54Ibid. at 119-26.
55R.M. MacIver, Society: A Textbook of Sociology (New York: Rinehart, 1948) at 146-53.
REVUE DE DRO1T DE McGILL
[Vol. 39
contiguous territory is not a necessary condition of community. According to
these authors, a community is “an inclusive group” which is “bound together by
a shared sense of belonging and by the feeling among its members that the
group defines, for them, their distinctive identity.”’56 As such, one can speak of
a Catholic, or a Jewish, or an academic community within a state.” Lenski and
Lenski support this line of reasoning in affirming the existence of “geographic”
and “cultural” communities, with the latter being created through shared cul-
tural traditiofns, values and experiences held by particular groups, such as racial
or ethnic groups.58 Wellman, Carrington and Hall, in turn, stress the importance
of social networks over territory. Community ties, they posit, are the result of
“informal links of companionship and aid between individuals –
and the pat-
terns formed by these linls.” 9 These networks also are established by individ-.
uals sharing common interests and values.’
The development of an understanding regarding the meaning of community
must then take account of the breadth of the approaches to community. Commu-
nities should be understood not only in territorial terms, but also in their social-
cultural terms. In the most general sense, then, a community can be understood
as a collectivity of individuals, of varying numbers, sharing certain interests, attri-
butes and characteristics with one another. On this basis, communities can be
defined by gender, language, ethnicity, race, religion, age, class, profession, occu-
pation or geography; and this list is not exhaustive.6′ The problem then faced in
thinking of community of interest in relation to distribution is not that there is no
such thing as community, but rather that any society is composed of an array of
varied yet inter-related communities. Which communities are then deserving of
special attention in distribution through the concept of community of interest?
The concept itself provides no answer to this question.
An appeal to geographically defined communities, moreover, will not
resolve this problem for the very reason that, given the plurality of communities
in any society, numerous and varied communities will co-exist in all geograph-
ically defined areas of society. Certain areas may have a population composed
of a number of differing ethnic, racial, religious and age groups, with each view-
ing itself as a distinct community. Similarly, certain areas may possess a pop-
ulation blending lower and middle income households of varied ethnic and lin-
guistic backgrounds; certain other areas may possess a population blending
urban and rural households of varied ethnic and religious backgrounds. Which
of these groups become the key groups defining the nature of the community in
a given area? Or do we look to the inter-relationship of such groups leading to
56L. Broom & P. Selznick, Sociology: A Text with Adapted Readings, 3d ed. (New York: Harper
& Row, 1963) at 31-32.
571bid.
58G. Lenski & J. Lenski, Human Societies: An Introduction to Macro Sociology, 5th ed. (New
York: McGraw-Hill, 1987) at 45-46.
59B. Wellman, P. Carrington & A. Hall, “Networks as Personal Communities” in B. Wellman
& S.D. Berkowitz, eds., Social Structures: A Network Approach (Cambridge, Eng.: Cambridge
University Press, 1988) at 131-34.
6Olbid.
61Qualter, supra note 46 at 96.
1994]
CHRONIQUES DE JURISPRUDENCE
the creation of distinct inter-group communities within the areas where group
mingling occurs?
As various communities exist everywhere in society, they necessarily blend
into one another geographically. Differing ethnic and racial groups, for example,
spatially merge into one another as do groups defined by income and language.
It is difficult, therefore, to determine where one community ends and another
begins. Wherever a dividing line is drawn, reasonable arguments can be raised
that the people living on the margins in fact share community interests among
themselves with such interests rooted to the nature of the mix of identities occur-
ring on the margins. Recourse to geographically defined communities of interest
thus cannot resolve the question of what is a community of interest since the
delineation of any geographic area will cut across some community ties.
The logic of the foregoing line of reasoning suggests that the concept of
community of interest, as a general factor to be taken into consideration in elec-
toral distribution, is questionable and ultimately unhelpful. This is not because
the concept of community is invalid, but because the concept is too broad. The
concept itself, in short, cannot be used as a precise directive for instructing and
guiding distribution commissions in the delineation of communities worthy of
special electoral representation. Since all societies are formed of a plethora of
communities, all possessing distinct interests, electoral boundaries can be drawn
in a plethora of manners, with all of them being subject to justification under
the rubric of community of interest. Such over-inclusiveness renders the general
concept worthless as an organizing principle of distribution; it simply promotes
and justifies arbitrary decision-making by electoral boundaries commissions as
certain communities of interest are delineated and given preference, while oth-
ers are ignored.
C. Minority Representation
Given that the theoretical and practical worth of the general concept of
community of interest falters on the plurality of communities in society, an
intriguing variation on the theme of community representation is that of “minor-
ity representation”, a concept to which McLachlin J. gave brief mention.62 The
initial difficulty with McLachlin J.’s use of this concept is that it is undefined
as to meaning and scope, and unrefined as to its practical operationalization.
The concept of “minority representation” is open to a variety of interpretations.
In the most permissive sense, the concept can mean that all minorities are
to be entitled to special representational rights in legislatures. Through this
approach, Canadians of Scottish or German descent, for example, would pos-
sess such an entitlement. In contrast to this open approach, a more restrictive
interpretation of “minority representation” would hold that only those groups
such as native Canadians, certain ethnic and racial groups, and the aged and
handicapped, for example, which can make a claim to having been historically
discriminated against within the life of this society, should be entitled to special
representational rights as a means of redress for past wrongs, as well as a means
62Reference Re Provincial Electoral Boundaries, supra note 1 at 188.
McGILL LAW JOURNAL
[Vol. 39
of integrating historically disadvantaged groups into the mainstream of political
and social life. As McLachlin J. asserted, a system of minority representation
could enable legislative bodies to more accurately reflect the “social mosaic” of
society.6″ This approach would be in keeping with current approaches to com-
batting discrimination within our society via policies of affirmative action.’
If this latter form of reasoning best captures the logic of concern for minor-
ity interests, certain questions of definitional scope immediately arise. Which
minorities are to be accorded special representational status? What is the thresh-
old level of discrimination which a group must have experienced in order to be
considered disadvantaged? And over what time period? As Karl Peter has
argued, all non-Anglo-Saxon ethnic groups in this country have faced severe
discriminatory behaviour at the hands of the Anglo-Saxon ethnic elite and still
suffer from such discrimination in’terms of social status and income. 6 In a more
recent work, though, Robert Brym has concluded that while Canadians of Euro-
pean but non-Anglo-Celtic backgrounds have faced discrimination in this coun-
try in the past, and still confront certain inequities attributable to ethnicity, eth-
nicity now generally “does not strongly influence status or income attain-
ment.”66 For non-European racial minorities, however, such is not the case, with
these groups continuing to bear the burden of racist discriminatory behaviour.67
The question then becomes whether European ethnic groups merit special rep-
resentational rights or not. And if so, should these rights be prorated to the
degree of discrimination which the group has suffered and is still suffering in
comparison to that of other groups?
If minorities are to be accorded such special status on account of having
faced past discrimination, what of certain groups which, although not demo-
graphic minorities, can make analogous claims of discrimination? Representa-
tives of women’s organizations, labour organizations and poverty groups have
all made such claims on behalf of their members. Are these groups to be entitled
to special representation? And would such special representation accorded to
minority groups be contingent upon the geographic concentration of such group
members? In certain instances, historically disadvantaged minority groups have
concentrated in certain distinct geographic areas, as has been the case with var-
ious ethnic groups forming ethnic neighbourhoods, for example, but this is not
universally the case. Numerous members of ethnic minority groups do not live
in ethnic enclaves. Questions then arise as to how these citizens should be
accorded special representation if they constitute but a small fraction of the pop-
63Ibid. at 189.
64See e.g. J. Edwards, Positive Discrimination: Social Justice and Social Policy (London: Tavis-
tock, 1987) c. 1; A. Bayefsky & M. Eberts, eds., Equality Rights and the Canadian Charter of
Rights and Freedoms (Toronto: Carswell, 1985) at 16-21, 116-24; G. Brodsky & S. Day, Canadian
Charter Equality Rights for Women: One Step Forward or Two Steps Back? (Ottawa: Canadian
Advisory Council on the Status of Women, 1989) at 190-98.
Ethnicity, Power and Politics in Canada (Toronto: Methuen, 1981) at 60-64.
65″The Myth of Multi-Culturalism and Other Political Fables” in J. Dahlie & T. Fernando, eds.,
66From Culture to Power: The Sociology of English Canada (Toronto: Oxford University Press,
671bid. at 29-30.
1989) at 109-13.
19941
CASE COMMENTS
ulation of the area in which they live. In certain other instances, minority group
enclaves, while existing as distinct enclaves, may have small populations far
below average riding quotients even taking into consideration a 25% deviation
rule. This is the case with rural and remote areas possessing distinct clusters of
native populations and with urban areas possessing unique ethnic communities,
such as Toronto’s downtown Chinatown. At the other extreme are those groups
which are very diffuse, such as women, the aged and the handicapped, for
example, which are found throughout all areas. It is extremely difficult, there-
fore, to determine how these various minority interests should be represented.
The issue to be confronted, ultimately, is whether there should be a para-
digm shift away from the concept of territorially based constituencies to interest
group based constituencies, thereby providing for the possibility of distinct
native ridings, women’s ridings, elderly ridings, ethnic ridings, class and occu-
pational ridings, and so forth. If this is the direction which political leaders in
this country adopt respecting distribution, and if it is upheld –
if not encour-
aged or even mandated –
by the courts, we will be witnessing a profound and
complex change in our system of representation.
The immediate problems associated with a shift to group representation
would be organizaitional. Precisely which groups would be accorded represen-
tational rights and which would not? The decision-making called for by this
simple question would be contentious and controversial as the foregoing ana-
lysis has illustrated. Other questions arise. Would such representation be estab-
lished to the exclusion of traditional representation by population or would it be
an adjunct to traditionally composed legislatures? If the latter, what proportion
of seats would be allocated to such group representation? And what of voting
rights? Would individuals have a singular or plural entitlement? Would individ-
uals have to register an affinity with only one group, thus claiming a vote for
only one category of constituency, or could individuals claim a variety of group
affinities, thereby claiming a plurality of votes? If the latter, the principle of
individual elector equality could be compromised as certain individuals may be
able to claim more group affinities than others, thereby entitling themselves to
more votes than others.
A move to group based constituencies would also raise difficult questions
regarding the distribution of seats among the represented groups. A move
toward group representation on the basis of general proportional representation
would be fraught with complexity. The initial problem would be ordering the
groups to be accorded such representation. Should legislative apportionment
rules begin, for example, in terms of ethnicity or gender or religion or income?
Depending on the first group chosen, certain tensions can be foreseen as con-
cern is then directed to the proportional representation of other groups into the
first grouping. If racial and ethnic groups constituted the base groups of repre-
sentation, would 52% of all racial and ethnic representatives have to be women,
for example? Would such a rule of gender equality be enforced on certain ethnic
groups notwithstanding that gender equality may be foreign to the social culture
of particular groups? Similarly, if the base groups are the gender groups, would
these groups have to be perfectly subdivided into ethnic and religious sub-
REVUE DE DRO1T DE McGILL
[Vol. 39
groups even if certain of these sub-groups reject female representation? If the
base group was defined, in turn, by income, this could have the beneficial result
for women of them possessing greater than 52% representation given the larger
proportion of women than men comprising the poorer yet more populous lower
echelons of society. An initial income based grouping could also result in some
ethnic groups being accorded greater, and some lesser, representation in propor-
tion to their complement of the general population given the relative wealth
experienced by certain ethnic groups.
Beyond these organizational issues, the development of a system of group
representation would also raise certain theoretical issues of importance.
Whereas the current system of representation is grounded upon the representa-
tion of citizens qua citizens, albeit organized along geographic lines, a system
of minority group representation is geared to the representation of members of
particular groups, with group affiliation alone being the defining factor. Such a
development would not only challenge the principle of “one person –
one
vote”, but it would also challenge the concept of citizenship in this country, as
well as the traditional role of elected representatives being guardians of the
common interest of the community as a whole.
The most controversial form of- minority representation would be one
where group representation was the sole basis of electoral representation, with
individuals casting a single ballot for one designated group constituency. This
development would constitute a clear signal to individuals that they are most
important in political life as members of one particular group, that they should
be primarily concerned with this particular group’s interests, and that they need
not be concerned, as individual citizens, with questions regarding the common
welfare of other groups or of the greater society in which they exist. This
approach, in short, would deny the pluralism which can and does exist in any
society.
As Katharine Swinton has argued,6″ individuals do not possess a unidimen-
sional identity rooted to but one aspect of their character. Rather, we are “com-
plex individualswith multiple identities.”69 An individual identity will be
founded upon a host of factors such as one’s gender, ethnicity, race, religion,
language, class, education, occupation and age, to name just a few. Any one
individual will thus be a member of all of these component groups with which
the individual shares an affinity, and may have strong opinions and beliefs
regarding political and socio-economic matters affecting any and all of these
groups. Individuals, in short, relate to the world through a plurality of group
identities, with these pluralist inter-relationships serving to link together, into a
social whole, the numerous and various groups within society.7″ A move to pro-
mote specific group representation would then act in defiance of this multidi-
mensional understanding of human identity, while serving to divide rather than
unite the groups composing society.
supra note 49 at 31.
68″Federalism, Representation and Rights” in J.C. Courtney, P. MacKinnon & D. Smith, eds.,
69Ibid.
70Ibid.
19941
CHRONIQUES DE JURISPRUDENCE
An initiative to promote minority representation, while avoiding these
problems of exclusivity and division, would then have to be founded upon the
establishment of a multiplicity of group constituencies going far beyond the
mere representation of historically discriminated against groups. Rather, to ade-
quately represent the social mosaic which creates human personality, a system
of representation would have to be founded upon the inclusion of all those
groups within society which contribute to the social mosaic. The list, of course,
would be long. Such a system, as well, would also have to be founded upon a
pluralist voting process with individuals casting ballots for the various and
numerous groups with which they are affiliated. This initiative, in turn, Would
then run into the practical and theoretical problems associated with concern for
elector equality and apportionment of group representation as outlined above.
This initiative would still represent a fundamental rejection of the concept
of general representation within a legislative assembly. Whereas the current sys-
tem posits that elected members can and should strive to represent the diversity
of the individual and group interests within their constituencies, a system of
group representation posits that only members of a specific group can represent
that group’s interests. Unlike the existing system which recognizes that individ-
uals, as representatives, party members and common citizens, can and should
seek to understand and sympathize with people from differing social groups and
even promote and represent their interests, group representation, in theory, chal-
lenges these ideas by stressing that only members of a particular group can truly
understand and represent the interests of the group. The danger implicit in this
line of thinking is that it rejects the human capacity for sympathetic reasoning
and action and, again, serves as a dynamic of division rather than union within
society.
Any initiative to promote group representation, in short, must take account
of the practical and theoretical issues raised by the concept. All of the questions,
concerns and ideas mentioned above, however seemingly far-fetched or idealis-
tic, flow directly from consideration of the meaning and operationalization of
the concept of minority representation in distribution. These are matters which
will have to be addressed should political leaders or the courts move to accept
the concept of minority representation as a requirement of distribution. It is
indeed unfortunate, therefore, that the majority on the Supreme Court of Canada
did not devote serious attention to the breadth of matters related to this concept
or the host of issues emanating from it. On matters of such importance, partic-
ularly with the Court suggesting that this concept may provide a criterion for
making decisions in the future respecting electoral distribution, judicial reason-
ing should be sound and enlightening, not weak or ambiguous.
All of the foregoing is not to belittle the importance of the concept of com-
munity in political or social life. As Charles Taylor and Michael Walzer have
long argued, communities are crucially important to the construction of socie-
ties and the development of human understanding respecting self and others,
and the distributional relationships between the individual and the collectivity.7′
7 1See e.g. C. Taylor, Sources of the Sef. The Making of the Modern Identity (Cambridge, Mass.:
Harvard University Press, 1989); M. Walzer, Spheres of Justice: A Defense of Pluralism and
Equality (New York: Basic Books, 1983).
McGILL LAW JOURNAL
[Vol. 39
The problem, though, is that the use of the concept of communities of interest
in distribution biases the very system of representation from the outset by pro-
viding special status to particular communities over others. Representation
based on the equal representation of individuals qua individuals does not negate
the various communal concerns held by those individuals; rather, it enables
those concerns to continue to exist and influence political life free from insti-
tutionalized representational favouritism.
D. Deviation Quotients
Closely related to the concepts of minority representation and community
of interest is the questionable practice of allowing the electorates of ridings to
deviate from a general quotient by a set percentage figure so as to take account
of factors such as geography and community of interest. Such a permissive rule
was upheld by the Supreme Court of British Columbia in Dixon and the
Supreme Court of Canada in Reference Re Provincial Electoral Boundaries as
the constitutionally correct approach through which to give effect to consider-
ations pertaining to the territorial size of ridings and community interests within
ridings, while maintaining a level of concern for the principle of substantial
equality. Pursuant to the rule, ridings are allowed to vary in electorate size up
to the given percentage figure. Deviations in excess of the percentage figure and
lacking any exceptional justification, as was the case in Dixon, are to be viewed
as unconstitutional violations of the principle of substantial elector equality.
Of note is that differing authorities, both political and judicial, have upheld
differing allowable deviation figures. In federal distribution law, as well as that
of Newfoundland, Quebec, Ontario, Alberta and British Columbia, for example,
deviations of 25% are generally allowed, with greater deviations being permis-
sible in exceptional circumstances. Nova Scotia has made allowance for devi-
ations of up to 33%, while New Brunswick and Prince Edward Island have no
restrictions on permissible deviation. At the other extreme, Manitoba generally
restricts deviations to a maximum of 10%.72 The impugned Saskatchewan leg-
islation allowed for deviations up to 25% in southern Saskatchewan, with var-
iances up to 50% being acceptable in the two northern ridings. Given the dis-
parity within the approaches to this matter taken by legislators, it is not
surprising that various members of the judiciary have likewise offered differing
views as to what should be an acceptable deviation figure. Though the members
of the Saskatchewan Court of Appeal asserted that it would have been inappro-
priate for them to set a de minimus acceptable deviation figure, they stressed
that any deviation should be as limited as possible.’ At the Supreme Court of
Canada, Madame Justice McLachlin and the majority upheld her reasoning in
Dixon in which the general 25% deviation figure was accepted as valid.74 It was
partly their inability to agree with this deviation figure which led three of the
72See Supreme Court of Canada, Pleadings and orders, supporting materials filed by Counsel in
Saskatchewan (A.G.) v. Carter, [1991] 2 S.C.R. 158, 81 D.L.R. (4th) 16, vol. 1 at 124, 141, 151,
156; vol. 2 at 170, 173, 179, 184, 187.
73Reference Re Electoral Boundaries Commission Act, supra note 14 at 462-64.
74Reference Re Provincial Electoral Boundaries, supra note 1 at 190, 194.
1994]
CASE COMMENTS
members of the Court to issue a dissenting opinion. In this dissent, the minority
upheld the validity of the factors of geography and community of interest as
being important, but stressed that these factors could be accommodated within
a general 15% deviation figure.75
The range of options respecting an appropriate percentage deviation repre-
sents more than simple differences of opinion concerning how best to operatio-
nalize interests associated with geography and community; rather, these differ-
ences illustrate the major problems of principle which flow from the nature of
a percentage deviation rule. If a 25% deviation is considered acceptable, why
not 26% or 27% or 33%? What qualitative difference in principle arises in the
interstices between these quantitative figures? Of course, the argument is made
that increasingly large deviations impugn the integrity of the principle of elector
equality, leading to the proposition that at some point this principle must take
precedence. But at what point? Why not at the point of a 20% deviation, or 15%
as suggested by the minority, or 10% as enshrined by the relevant Manitoba leg-
islation?
These questions go to the heart of the dilemma posed in Reference Re Pro-
vincial Electoral Boundaries. The case pits the egalitarian moral principle of
“one person –
one vote” against various social factors and interests which are
also viewed by many to reflect principles important to the quality of represen-
tation. An equality principle rooted to a conception of individual rights thus
confronts competing principles rooted to collective interests. The dilemma is
created, however, through the combined recognition by most actors concerned
with representation that, on the one hand, the principle of absolute equality
implicit in the pure conception of “one person –
one vote” is impossible to
operationalize given practical constraints in distribution-making and mainte-
nance. On the other hand, concern for the collective interests in competition
with the principle of elector equality cannot be allowed to vitiate the general
principle of elector equality implicit in the concept of “one person –
one vote”.
The principle of elector equality is thus viewed as unattainable in absolute
terms, while remaining the key value needing recognition and support in all dis-
tribution systems.
This interplay of practical considerations with moral principle then
becomes highly problematic. The creation of percentage deviation figures rep-
resents the creation of devices designed to quantify a moral principle, or more
precisely, the extent to which a moral principle can be modified while still
retaining its legitimacy. The very attempt to quantify a qualitative condition,
though, is questionable given the subjective decision-making surrounding any
attempt to delineate, in scalar form, qualitative changes in a moral condition.
Differing actors will bring differing perspectives to bear in assessing the nature
of the moral condition, the extent to which the condition can be modified while
remaining true to principle, and in defining instruments to measure qualitative
changes to the condition under-analysis. This final point is not to be underesti-
mated. A gulf exists between the blunt nature of quantitative analysis and the
75 Ibid. at 172-74.
REVUE DE DROIT DE McGILL
[Vol. 39
complexities of moral conditions. Given that the nature of the moral condition
in this case is open to interpretation, so too are all statistical indicators designed
to measure the nature of the condition.
Beyond this general theoretical problem, moreover, lies the problem of
practical justification. Pursuant to the pluralist approach, deviations from the
principle of equality are justified on the basis of serving other practical consid-
erations, with these considerations being viewed as relating to matters of prin-
ciple. As argued above, however, these practical considerations rooted to con-
cerns over territorial size of ridings, communications and the representation of
communities of interests or minority representation, are all subject to deep crit-
icism. For example, it has been argued that either the practical consideration
giving rise to the concern can be met with practical alternative policy instru-
ments, as is the case with the issue of territory and communications, or the prac-
tical consideration is rooted to a concept so vague and indeterminate or ques-
tionable as to be analytically worthless, as is the case with the concept of
community of interest, or theoretically and practically dubious, as is the case
with minority representation. If the practical considerations giving rise to the
claims for special treatment via a deviation rule are found to be either suscep-
tible to remedy by other means, or are found to be philosophically questionable,
then the practical justification for the deviation from the substantial equality
principle falters.
Given the .foregoing analysis, the one practical factor which does stand as
a valid concern militating against the principle of elector equality, and which is
devoid of alternative treatment other than through a deviation rule, is that of the
impossibility of devising and maintaining electoral constituencies composed of
absolutely equal electorates. The practical impossibility of the realization of
absolute elector equality in distribution necessitates that some deviation from
absolute equality be recognized and endorsed as a practical necessity of bound-
ary drawing. Yet, given the practical and mechanical nature of this rule of
exception, the exception hardly stands as a principled countervailing factor to
the principle of elector equality. Substantial elector equality thus stands as the
fundamental principle structuring the democratic nature of the electoral system
in this country, while also representing a practical method of distribution which,
in contrast to the pluralist approach, is relatively uncomplicated and easily
understood and operationalized. As such, it is a principle and practice which
should be circumscribed as little as possible. Thus, the opinion of the Saskatch-
ewan Court of Appeal is commendable for its recognition that any rule of devi-
ation should be as limited as possible, while still allowing for a modicum of
pragmatism to be exercised by boundaries commissions in drafting constituen-
cies and by courts in reviewing the constitutionality of electorate variances in
existing constituencies.
Final Reflections
Judicial review of systems of electoral distribution is of crucial importance
to the structure and functioning of representation and, thus, of immediate rele-
vance to the nature of democracy in this country. It is then unfortunate that the
19941
CHRONIQUES DE JURISPRUDENCE
quality of the decision in Reference Re Provincial Electoral Boundaries was not
equal to the importance of the matter under review. To their credit, the different
judgements of the Supreme Court of Canada stressed the importance of the egal-
itarian principle of “one person –
one vote” as being integral to the working
of democracy in this country. The Court, however, failed to rigorously address
the validity of the main claims made based on the pluralist approach to distri-
bution to constrain the operationalization of the principle of elector equality.
The Court failed to critically evaluate the merits of such concepts as special rep-
resentational entitlement to territorially large areas, special minority interests,
and communities of interest. By failing to address the many issues enveloped
in the concept of community of interest, the Court also failed to devote serious
attention to the merits and demerits of allowing distribution commissions to
vary the electorates of constituencies by set percentages. On account of these
many failings, the decision of the Court is open to great criticism; indeed, by
adhering to the basic tenets of the pluralist approach, the decision does damage
to the working of democracy within this country.
These weaknesses suggest that the Court’s work in Reference Re Provin-
cial Electoral Boundaries, far from providing authoritative guidance in distribu-
tion matters and far from foreclosing litigation on such matters, may but
enhance questions amongst those concerned with distribution, thereby encour-
aging future litigation. Furthermore, it is not hard to foresee future challenges
being made to electoral distribution legislation on the grounds that concerns
respecting effective representation of territorially large areas can be met through
the use of policy instruments which are less violative of the Charter than ter-
ritorial over-representation. Similarly, future challenges could be rooted to argu-
ments that the general concept of community of interest is inherently problem-
atic and unprincipled, thus leading it to be inoperative as a legitimate guide to
distribution. The 25% deviation rule is likewise open to attack, while the majo-
rity’s reference to the concept of minority representation is questionable.76
In Reference Re Provincial Electoral Boundaries the Supreme Court of
Canada rendered an important and controversial decision respecting electoral
distribution in this country. In light of this, and as a result of parliamentary rep-
resentation becoming increasingly an issue of popular political discourse, it is
fair to suggest that this case represents but the first of a number of equally major
decisions which have yet to be crafted by the Court respecting electoral distri-
bution. It is hoped that in future adjudication the Court’s decision-making will
do justice to the seriousness and complexity of the subject matter. It is hoped
that in future consideration of these matters, the Court will affirm and promote,
as did the Saskatchewan Court of Appeal, the egalitarian approach to distribu-
tion founded upon the theory and practice of substantial elector equality as the
core principle defining representative democracy in this country, and thus being
the principle which should guide all future redistributions in this country.
76It is interesting to note that the current distribution of seats in the Legislative Assembly of
Prince Edward Island has been ruled unconstitutional by the trial division of the provincial
Supreme Court on the basis that the distribution violates Charter guarantees of equal representa-
tion. A new distribution is currently underway (MacKinnon v. Prince Edward Island (16 February
1993), (P.E.I. S.C.) [unreported]).
