Reference re Prov. Electoral Boundaries (Sask.), [1991] 2 S.C.R. 158

 

The Attorney General for Saskatchewan                                         Appellant

 

v.

 

Roger Carter, Q.C.                                                                           Respondent

 

and

 

Attorney General of Canada, Attorney General

of Quebec, Attorney General of British Columbia,

Attorney General of Prince Edward Island,

Attorney General for Alberta, Attorney General

of Newfoundland, Minister of Justice of the

Northwest Territories, Minister of Justice of

the Yukon, John F. Conway, British Columbia

Civil Liberties Association, Douglas Billingsley,

Wilson McBryan, Leonard Jason, Daniel Wilde,

Alberta Association of Municipal Districts &

Counties, City of Edmonton, City of Grande Prairie,

Equal Justice For All   Interveners

 

Indexed as:  Reference re Prov. Electoral Boundaries (Sask.)

 

File No.:  22345.

 

1991: April 29, 30; 1991:  June 6.

 

Present:  Lamer C.J. and La Forest, L'Heureux‑Dubé, Sopinka, Gonthier, Cory, McLachlin, Stevenson and Iacobucci JJ.

 

on appeal from the court of appeal for saskatchewan

 

                  Constitutional law ‑‑ Charter of Rights ‑‑ Right to vote ‑‑ Electoral boundaries ‑‑ Variances in size of voter populations among constituencies ‑‑ Whether Charter right to vote infringed ‑‑ Canadian Charter of Rights and Freedoms, ss. 1, 3 ‑‑ Electoral Boundaries Commission Act, S.S. 1986‑87‑88, c. E‑6.1, ss. 14, 20.

 

                  The Saskatchewan Court of Appeal, on a reference dealing with the provincial electoral distribution, found that proposed changes to the electoral boundaries infringed s. 3 of the Canadian Charter of Rights and Freedoms.  The Electoral Boundaries Commission Act imposed a strict quota of urban and rural ridings and required that urban ridings coincide with existing municipal boundaries.  The resulting distribution map, unlike the one it replaced, revealed a number of ridings with variations in excess of 15 percent from the provincial quotient and indicated a problem of under-representation in urban areas.

 

                  Two questions were stated for the court's opinion.  The first queried whether the variance in the size of voter populations among those constituencies infringed Charter rights guaranteed by the Charter and if so, in what particulars.  It also queried whether any such denial of rights was  justified by s. 1 of the Charter.  The second queried whether the distribution of those constituencies among urban, rural and northern areas infringed Charter rights and if so, in what particulars were these rights infringed and in what particulars were they justified.  This question too queried whether any such denial of rights was justified by s. 1 of the Charter.

 

                  Held (Lamer C.J. and L'Heureux-Dubé and Cory JJ. dissenting):  The appeal should be allowed.

 

                  Per La Forest, Gonthier, McLachlin, Stevenson and Iacobucci JJ.:  At issue here was whether the variances and distribution reflected in the constituencies themselves violated the Charter guarantee of the right to vote.  The validity of The Representation Act, 1989 in so far as it defined the constituencies, was indirectly called into question.

 

                  The definition of provincial voting constituencies is subject to the Charter and is not a matter of constitutional convention relating to the provincial constitution which is impervious to judicial review. Although legislative jurisdiction to amend the provincial constitution cannot be removed from the province without a constitutional amendment and is in this sense above Charter scrutiny, the provincial exercise of its legislative authority is subject to the Charter.  The province is empowered by convention to establish its electoral boundaries but that convention is subject to s. 3 of the Charter.

 

                  The content of the Charter right to vote is to be determined in a broad and purposive way, having regard to historical and social context.  The broader philosophy underlying the historical development of the right to vote must be sought and practical considerations, such as social and physical geography, must be borne in mind.  The Court, most importantly, must be guided by the ideal of a "free and democratic society" upon which the Charter is founded.

 

                  The purpose of the right to vote enshrined in s. 3 of the Charter is not equality of voting power per se but the right to "effective representation".  The right to vote therefore comprises many factors, of which equity is but one.  The section does not guarantee equality of voting power.

 

                  Relative parity of voting power is a prime condition of effective representation.  Deviations from absolute voter parity, however, may be justified on the grounds of practical impossibility or the provision of more effective representation.  Factors like geography, community history, community interests and minority representation may need to be taken into account to ensure that our legislative assemblies effectively represent the diversity of our social mosaic.  Beyond this, dilution of one citizen's vote as compared with another's should not be countenanced.

 

                  The history or philosophy of Canadian democracy does not suggest that the framers of the Charter in enacting s. 3 had the attainment of voter parity as their ultimate goal.  Their goal, rather, was to recognize the right long affirmed in this country to effective representation in a system which gives due weight to voter equity but admits other considerations where necessary.  Effective representation and good government in this country compel that factors other than voter parity, such as geography and community interests, be taken into account in setting electoral boundaries.  Departures from the Canadian ideal of effective representation, where they exist, will be found to violate s. 3 of the Charter.

 

                  The actual allocation of seats between urban and rural areas  closely followed the population distribution between those areas and effectively increased the number of urban seats to reflect population increases in urban areas.  In general the variations between boundaries in the southern part of the province appeared to be justifiable on the basis of factors such as geography, community interests and population growth patterns.  The northern boundaries were appropriate, given the sparse population and the difficulty of communication in the area.  A violation of s. 3 of the Charter was not established.

 

                  Per Sopinka J.:  The reasons of McLachlin J. were substantially agreed with, although the interpretation of s. 3 of the Charter was differently approached. 

 

                  The framers of the Charter did not intend to create a new right and accordingly the primary inquiry was to determine on what principles the right to vote was based.  Historically, the drawing of electoral boundaries has been governed by the attempt to achieve voter equality with liberal allowances for deviations based on the kinds of considerations enumerated in s. 20 of The Electoral Boundaries Commission Act.  Deviations were avoided which deprived voters of fair and effective representation.  Under the Charter, deviations are subjected to judicial scrutiny and must not be such as to deprive voters of fair and effective representation.

 

                  The Charter guarantee in s. 3 does not extend to the process.  The legislature was not required to establish an electoral commission or to ensure that a commission, when established, was able to fulfill its mandate freely without guidelines imposed by the legislature.

 

                  The constitutional validity of the factors in s. 20 or s. 14 was not at issue but rather the effect that their application produced.  The extent of deviation from strict voter equality and the reasons for those deviations were not such as to deny fair and effective representation.

 

                  Per Lamer C.J. and L'Heureux-Dubé and Cory JJ. (dissenting):  In Canada, each citizen as a minimum must have the right to vote, to cast that vote in private and to have that vote honestly counted and recorded.  Equally important, each vote must be relatively equal to every other vote; there cannot be wide variations in population size among the 64 southern constituencies.  Deviations from equality will be permitted where they can be justified as contributing to the better government of the people as a whole, giving due weight to regional issues involving demographics and geography.

 

                  The restrictions placed on the Electoral Boundaries Commission by The Electoral Boundaries Commission Act were unknown to previous commissions.

 

                  The electoral map at issue which resulted from the impugned legislation must be considered even if the Charter infringements involved might be thought to be relatively minor.  This is because the right to vote is fundamentally important to a democracy and to its citizens.

 

                  The problems with the impugned distribution were almost entirely a function of the two conditions placed on the Commission by the Act and were unacceptable and, given the eminently fair riding map of the previous distribution, quite unnecessary.  The first imposed a strict quota of urban and rural ridings and the second required that the boundaries of the urban ridings coincide with the existing municipal boundaries effectively "quarantining" them from the others.

 

                  The fundamental importance of the right to vote demands a reasonably strict surveillance of legislative provisions pertaining to elections.  Scrutiny under s. 3 attaches not only to the actual distribution in question but also to the underlying process from which the electoral map was derived.  While the actual distribution map may appear to have achieved a result that is not too unreasonable, the effect of the statutory conditions interfered with the rights of urban voters.  Once an independent boundaries commission is established, it is incumbent on the legislature to ensure that the Commission was able to fulfill its mandate freely and without unnecessary interference.  The right to vote is so fundamental that this interference is sufficient to constitute a breach of s. 3 of the Charter.

 

                  The creation of the two northern ridings met all the requisite conditions of the Oakes test and was justified under s. 1 of the Charter.  Geography and demography demonstrated a pressing and substantial need and the creation of these constituencies was rationally connected to the concept that they have effective representation.

 

                  The southern ridings were in a different position legally and geographically.  While the differing representational concerns of urban and rural areas may properly be considered in drawing constituency boundaries, the voter population of each constituency should be approximately equal and the type of mandatory conditions imposed here are therefore precluded.  Given the initial premise of equality, the Commission should be free to consider such factors as geography, demography and communities of interest in drawing constituency boundaries and allocating ridings between rural and urban areas.  No explanation was given why the balancing of relevant factors could not be left to the Commission and instead had to be mandated by the legislature.  The less equitable distribution that resulted because of these legislatively mandated conditions was, absent a reasonable explanation, suspect.  There was no basis for concluding that the legislature's objective in imposing these conditions was pressing and substantial.  Even assuming a pressing and substantial need, the legislation did not affect the rights of urban voters as little as possible.  Earlier and more equitable distributions indicated that the rights of urban voters could be interfered with to a lesser extent.

 

Cases Cited

 

By McLachlin J.

 

                  Referred to:  Dixon v. B.C. (A.G.) (1986), 1986 CanLII 770 (BC SC), 7 B.C.L.R. (2d) 174; R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295; Hunter v. Southam Inc., 1984 CanLII 33 (SCC), [1984] 2 S.C.R. 145; Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486; Edwards v. Attorney‑General for Canada, 1929 CanLII 438 (UK JCPC), [1930] A.C. 124; Law Society of Upper Canada v. Skapinker, 1984 CanLII 3 (SCC), [1984] 1 S.C.R. 357; R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713; McGowan v. Maryland, 366 U.S. 420 (1961); R. v. Schwartz, 1988 CanLII 11 (SCC), [1988] 2 S.C.R. 443; United States of America v. Cotroni, 1989 CanLII 106 (SCC), [1989] 1 S.C.R. 1469; R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103; Dixon v. B.C. (A.G.), 1989 CanLII 248 (BC SC), [1989] 4 W.W.R. 393; Baker v. Carr, 369 U.S. 186 (1962); Karcher v. Daggett, 462 U.S. 725 (1983); Kirkpatrick v. Preisler, 394 U.S. 526 (1969); Attorney-General (Aus.); Ex rel. McKinlay v. Commonwealth (1975), 135 C.L.R. 1; Reference Re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313; Canada v. Schmidt, 1987 CanLII 48 (SCC), [1987] 1 S.C.R. 500; Canada (Auditor General) v. Canada (Minister of Energy, Mines and Resources), 1989 CanLII 73 (SCC), [1989] 2 S.C.R. 49.

 

Statutes and Regulations Cited

 

Act to re-adjust the Representation in the House of Commons, S.C. 1872, c. 13.

 

Canadian Charter of Rights and Freedoms, ss. 1, 3.

 

Constituency Boundaries Commission Act, 1972, S.S. 1972, c. 18, s. 16(1).

 

Electoral Boundaries Commission Act, S.S. 1986-87-88, c. E-6.1, ss. 14, 20.

 

Electoral Boundaries Commission Act, 1991, S.S. 1991, c. E-6.11, s. 9(2), 11(1), (2).

 

Representation Act, 1981, S.S. 1980-81, c. R-20.1.

 

Representation Act, 1989, S.S. 1989-90, c. R-20.2.

 

Authors Cited

 

Canada. House of Commons Debates, Vol. III, 4th Sess., June 1, 1872.  Ottawa: Robertson, Roger & Co., 1872.

 

                  APPEAL from a judgment of the Saskatchewan Court of Appeal (1991), 1991 CanLII 8030 (SK CA), 90 Sask. R. 174, 78 D.L.R. (4th) 449, [1991] 3 W.W.R. 593, on a reference by the Lieutenant Governor in Council.  Appeal allowed, Lamer C.J. and L'Heureux-Dubé and Cory JJ. dissenting.

 

                  Robert G. Richards and  Thomson Irvine, for the appellant.

 

                  Roger Carter, Q.C., for the respondent.

 

                  Ivan G. Whitehall, Q.C., and Susan D. Clarke, for the intervener the Attorney General of Canada.

 

                  Louis Rochette and  Marise Visocchi, for the intervener the Attorney General of Quebec.

 

                  E. Robert A. Edwards, Q.C., and Frank A. V. Falzon, for the intervener the Attorney General of British Columbia.

 

                  Rosemary Scott and  Gordon L. Campbell, for the intervener the Attorney General of Prince Edward Island.

 

                  Peter M. Owen, Q.C., and P. Jon Fauld, for the intervener the Attorney General for Alberta.

 

                  B. Gale Welsh, for the intervener the Attorney General of Newfoundland.

 

                  Bernard W. Funston and Elizabeth J. Stewart, for the intervener the Minister of Justice of the Northwest Territories.

 

                  T. Murray Rankin, for the intervener the Minister of Justice of the Yukon.

 

                  Peter T. Costigan, for the intervener the Alberta Association of Municipal Districts and Counties.

 

                  John F. Conway, for the intervener John F. Conway.

 

                  Robert D. Holmes, for the intervener the British Columbia Civil Liberties Association.

 

                  Donald J. Boyer, Q.C., for the interveners Douglas Billingsley, Wilson McBryan, Leonard Jason and Daniel Wilde.

 

                  Timothy J. Christian, for the interveners the cities of Edmonton and Grande Prairie.

 

                  Larry W. Kowalchuk, for the intervener Equal Justice for All.

 

//Cory J.//

 

                  The reasons of Lamer C.J. and L'Heureux-Dubé and Cory JJ. were delivered by

 

                  Cory J. (dissenting) -- This appeal is concerned with the most fundamental of our rights in a democratic society, the right to vote.  I have read with great interest the reasons of my colleague Justice McLachlin.  Although I agree with many of the principles she has set forth, I have come to a different conclusion and would dismiss the appeal.

 

Something of History and Background

 

                  The right to vote is synonymous with democracy.  It is the most basic prerequisite of our form of government.  In a democratic society based upon the right of its citizens to vote, the right must have some real significance.  In Canada it is accepted that, as a minimum, each citizen must have the right to vote, to cast that vote in private, and to have that vote honestly counted and recorded.

 

                  There is, I believe, a further, equally important aspect of the right, namely that each vote must be relatively equal to every other vote.  That is not to say that there cannot be variations in population size between constituencies.  These variations or deviations from equality will be permitted where, in the words of my colleague at p. 000, they "can be justified on the ground that they contribute to the better government of the populace as a whole, giving due weight to regional issues within the populace and geographic factors within the territory governed."

 

                  Free people have always striven for relative equality of voting power.  The Americans of 1776 sought recognition of the reasonable principle that there was to be no taxation without representation and further, that representation was to be based upon the equal weight of every ballot. 

 

                  It is argued, quite correctly, that our Canadian background is different from that of our American neighbour.  It is said that we have never insisted upon precise equality of voting power, but instead have traditionally placed greater emphasis on the representation of community interests and given wider recognition to geographic considerations.  I agree with these submissions.  In Canada we have recognized that the vast, sparsely settled regions in the north must be adequately represented even where their population is less than half of that of a constituency in the south.  To recognize this is to recognize the reality of Canada and Canadian geography.  At the same time, in the rest of Canada there has been a conscious and continuing move towards greater equality among constituencies.

 

                  Saskatchewan became a province in 1905.  The early electoral maps of the province show a wide divergence in riding populations.  However, this tendency has changed greatly over the years, particularly since 1972 when the first electoral boundaries commission was established.  Unlike the Electoral Boundaries Commission upon whose recommendations the impugned distribution is based, the Constituency Boundaries Commissions of 1973 and 1979-80, established pursuant to The Constituency Boundaries Commission Act, 1972, S.S. 1972, c. 18, were not bound by any fixed allocation of ridings between urban and rural areas, nor were they required to draw urban boundaries so as to coincide with municipal limits.  These restrictions placed upon the Electoral Boundaries Commission by the provisions of The Electoral Boundaries Commission Act, S.S. 1986-87-88, c. E-6.1 (the "E.B.C.A."), restrictions unknown to previous boundaries commissions, are in my view of great significance to the disposition of this appeal.

 

                  The maps for 1981 and 1989 are attached as schedules to these reasons.  Putting aside the two northern constituencies, which are in a class by themselves, the distribution map of 1981 illustrates how equitably electoral boundaries can be drawn.  A glance at the 1981 map demonstrates that each and every southern riding, whether urban or rural, is within 15 percent of the provincial quotient -- the figure obtained by dividing the southern voting population by the number of southern ridings.  It seems that the figure of 15 percent is an eminently reasonable accommodation of the greater difficulties that may be encountered in representing some of the large rural ridings of Saskatchewan.

 

The Impugned Distribution

 

                  The E.B.C.A., which set into motion the process which eventually resulted in the impugned distribution, imposed two conditions on the Boundaries Commission.  First, a strict quota of urban and rural ridings was imposed.  Second, the boundaries of the urban ridings were required to coincide with the existing municipal boundaries.  It is as though the urban ridings were to be "quarantined" from the others.  The conditions imposed are both unacceptable and, particularly in light of the eminently fair riding map achieved in 1981, quite unnecessary. 

 

                  The absence of the mandatory conditions in the earlier legislation resulted in an electoral map that was fairer to all electors.  The legislation then in force (The Constituency Boundaries Commission Act, 1972) provided:

 

                  16.--(1) In determining the area to be included in, and in fixing the boundaries of, any constituency in the portion of the province lying south of the line described in section 14, the commission shall be governed by the following rules:

 

 

 

(a)special geographic considerations including in particular sparsity, density or relative rate of growth of population of various regions of the portion of the province lying south of the line described in section 14, the accessibility of such regions or the size or shape thereof appear to the commission to render such a departure necessary or desirable;

 

(b)any special community or diversity of interests of the inhabitants of various regions of the portion of the province mentioned in clause (a) appears to the commission to render such a departure necessary or desirable;

 

(c)physical features of any area or any other similar and relevant factors, including variations in the requirements of the population of any constituency in the portion of the province mentioned in clause (a) appear to the commission to render such a departure desirable;

 

but in no case shall the population of any constituency in the province as a result thereof depart from the constituency quotient to a greater extent than fifteen per cent more or fifteen per cent less.

 

                  It is interesting that the boundaries commission legislation currently in force in Saskatchewan, which in April 1991 replaced the legislation now before the Court, omits any reference to the numbers of urban and rural ridings.   Further, it does not fix the boundaries of urban ridings so as to coincide with municipal limits.  The relevant provisions of the new legislation (The Electoral Boundaries Commission Act, 1991, assented to April 16, 1991) are as follows:

 

9 . . .

 

(2)               In fixing the boundaries of proposed constituencies, the commission shall:

 

                                                                    ...

 

(b) divide the area of Saskatchewan south of the dividing line into 64 constituencies.

 

11(1)           In determining the area to be included in a proposed constituency south of the dividing line and in fixing the boundaries of that constituency, the commission shall ensure that the voter population of each proposed constituency be, as near as possible, equal to the constituency population quotient.

 

(2)               Notwithstanding subsection (1), the commission may depart from the requirements of that subsection where, in the opinion of the commission, it is necessary to do so because of:

 

(a) special geographic considerations, including:

 

(i) sparsity, density or relative rates of growth of population in various regions south of the dividing line;

 

(ii) accessibility to the regions mentioned in subclause (i); or

 

(iii) the size and shape of the regions described in subclause (i);

 

(b) a special community of interests or diversity of interests of persons residing in regions south of the dividing line; or

 

(c) physical features of regions south of the dividing line.

 

                  The difference between the old boundaries commission legislation and that implicated in the present appeal is of fundamental importance.  The presence of the two mandatory conditions in the E.B.C.A., on the basis of which the impugned distribution was established, is a serious cause for concern.  The impact of these conditions on the quality of the distribution is demonstrated by comparing the 1981 and 1989 distribution maps.  The deterioration in the quality of the distribution from 1981 to 1989 is evidence of the effect of the two mandatory conditions, in terms of the potential for inequality and unfairness to which they give rise.  (I would also observe in passing that the problematic -- and unnecessary -- nature of the two mandatory conditions is underscored by the notable absence of those conditions from the most recent boundaries commission legislation.)

 

                  The riding map which resulted from the E.B.C.A. is not as fair as the 1981 riding map.  The 1989 distribution map reveals a number of constituencies with variations in excess of 15 percent from the provincial quotient.  The problem of under-representation is most acute in the major urban centres of the province.  Thus, for example, based on its population, the growing city of Saskatoon should be entitled to elect another 1.1 members.  I hasten to add, however, that it is not for a court to get into the details of the riding boundaries set by the Boundaries Commission.  That work has been conscientiously performed by its eminent members.  Rather, a court can only determine if there has been an infringement of the s. 3 Charter right to vote.  A comparison of the 1981 map to that of 1989 convinces me that there has been such an infringement.

 

                  It is said that the current map reflects such a minor infringement that it is not worth considering.  I cannot accept that argument as correct for two reasons.  First, the right to vote is fundamental to a democracy.  If the right to vote is to be of true significance to the individual voter, each person's vote should, subject only to reasonable variations for geographic and community interests, be as nearly as possible equal to the vote of any other voter residing in any other constituency.  Any significant diminution of the right to relative equality of voting power can only lead to voter frustration and to a lack of confidence in the electoral process.  The 1981 distribution map demonstrates that relative equality can be achieved in all of the southern Saskatchewan ridings.  This degree of equality should be maintained. 

 

                  Second, the reason for the departure from riding equality must be considered.  In my view, the problems with the impugned distribution are almost entirely a function of the shackling of the Boundaries Commission by the two conditions imposed by the underlying legislation, the E.B.C.A.  These conditions prevented the Commission from sufficiently accommodating the changing demographic reality.  By stipulating a mandatory rural-urban allocation of ridings and by confining urban ridings to municipal boundaries, the E.B.C.A. has led to greater variances than would otherwise have been the case had the Boundaries Commission been completely free in making its recommendations.

 

                  Specifically, the mandatory rural-urban allocation may have prevented the Commission from taking sufficient account of the diminishing rural population and the corresponding urban growth in the province.  The requirement of conformity of urban ridings to municipal limits similarly poses a potential obstacle to the necessary accommodation of demographic realities, particularly since municipal boundaries often fail to reflect urban development.  On this point, it was said that there was no such thing as dormitory communities in Saskatchewan cities.  I accept that as correct.  Yet the demographic material filed indicates that there continues to be a movement of population away from the province and, more significantly, a movement of people from the rural areas to the urban centres.  This indicates a need for flexibility, not only in the allocation of seats between urban and rural areas, but also in the fixing of the boundaries of urban ridings.  It also highlights the steadily growing need for the fair representation of the urban resident.

 

                  The fundamental importance of the right to vote demands a reasonably strict surveillance of legislative provisions pertaining to elections.  While I agree with my colleague McLachlin J. regarding the meaning of the s. 3 right to vote and the relevant criteria to be considered in assessing whether a given distribution violates that right, I am of the view that the inquiry cannot be restricted solely to the ultimate result achieved.  We are concerned in this appeal not only with results but also with process.  In my view, s. 3 scrutiny attaches not only to the actual distribution in question, but also to the underlying process from which the electoral map was derived.  It is this process that concerns me.

 

                  Thus, while the actual distribution map may appear to have achieved a result that is not too unreasonable, I am of the view that the effect of the statutory conditions has been to interfere with the rights of urban voters.  Once an independent boundaries commission was established, it was incumbent upon the Saskatchewan legislature to ensure that the commission was able to fulfill its mandate freely and without unnecessary interference. The public would, quite properly, perceive the Commission to be an independent and trustworthy body.  It would be an affront for the legislature to undermine the jurisdiction and authority which members of the public would reasonably expect the Commission to possess.  I should add that, had the Saskatchewan government chosen to legislate the boundaries directly rather than by establishing an independent boundaries commission, the s. 3 right would still be engaged.

 

                  The right to vote is so fundamental that this interference is sufficient to constitute a breach of s. 3 of the Charter.  To diminish the voting rights of individuals is to violate the democratic system.  Such actions are bound to incur the frustration of voters and risk bringing the democratic process itself into disrepute.  The haunting spectre of "rotten boroughs" is not that far removed as to be forgotten.  The right to vote is too important to be diluted in the absence of some valid justification.  No such justification exists in this case.

 

Is the Infringement Justifiable under s. 1 of the Charter?

 

                  The northern regions are in a class by themselves.  The geography of these sparsely settled regions clearly demonstrates a pressing and substantial need for two northern constituencies.  The creation of these constituencies is certainly rationally connected to the concept that these vast, underpopulated areas need effective representation.  In short, the creation of the two northern ridings meets all the requisite conditions and they are justified under s. 1 of the Charter.

 

                  The southern ridings are in a different position legally as well as geographically.  I readily agree that the differing representational concerns of urban and rural areas may properly be considered in the determination of constituency boundaries.  However, any body charged with creating an electoral map should commence with the proposition that, to the extent that it is reasonable and feasible, the voter population of each constituency should be approximately equal.  In my view, this necessarily precludes the type of mandatory conditions imposed in the present case.

 

                  Proceeding from the initial premise of equality, the Commission should, in determining constituency boundaries and allocating ridings between urban and rural areas, be free to consider such factors as geography, demography and communities of interest.  In any given distribution, the degree of variance between constituencies and the allocation of ridings between urban and rural areas will depend on the nature of the constituencies under consideration and the extent to which these factors are present.

 

                  For instance, the 1981 map provides proof that it is possible in Saskatchewan to achieve equality within 15 percent of the provincial quotient for all southern constituencies while still addressing other relevant considerations such as the differing nature of rural and urban interests.  In other provinces, these concerns will be balanced differently.  Depending on the particular characteristics of each province, non-population factors may require greater or less deviation.  Thus, for example, a 25 percent variation has been found to be necessary and acceptable in British Columbia, whereas the legislation in Manitoba limits the variation to 10 percent.

 

                  In Saskatchewan, the basic requirement of reasonable equality was met when the 1981 constituency map was drawn.  No reason has been provided as to why it was no longer possible to achieve the degree of equality reflected in that distribution.  Moreover, no explanation has been given as to why the balancing of the relevant factors could not, as it was previously, be left to the Commission rather than being mandated by the legislature.  The province has failed to justify the need to shackle the Commission with the mandatory rural-urban allocation and the confinement of urban boundaries to municipal limits.  The effect of these mandatory conditions was to force the Commission to recommend a distribution which departs from the higher degree of equality achieved in 1981.  In the absence of a reasonable explanation as to why this was necessary, the distribution in question is suspect and there is no basis upon which to conclude that the legislature's objective in imposing the mandatory conditions was pressing and substantial.

 

                  However, even assuming that the mandatory conditions were enacted in pursuit of some pressing and substantial need, it cannot be said that the legislation affected the rights of urban voters as little as possible.  The earlier Constituency Boundaries Commission Act, 1972 and the maps resulting from that legislation demonstrate that significantly less intrusive means can be utilized to provide good and proper rural representation.  The 1981 map demonstrates not only that it is possible to achieve a greater degree of electoral equality than exists in the impugned distribution, but also that the goal of ensuring adequate representation of rural areas can be met without imposing restrictions on the boundaries commission.  Thus, the earlier legislation and resulting constituency maps clearly demonstrate that there are means of drawing the constituency boundaries which interfere with the rights of urban voters to a lesser extent.

 

                  I wish to emphasize that this is not a matter of a court entering the domain of the legislature.  Rather, it is no more than a requirement that the legislature refrain from infringing Charter rights.  It requires no more of the Saskatchewan legislature than that it comply with either its earlier or subsequent enactments on the same subject.

 

                  In summary, it has not been established that there was a pressing or substantial need either to rigidly fix the number of urban and rural ridings in southern Saskatchewan or to confine the urban ridings to existing  municipal boundaries.  It follows that the first requirement of s. 1 has not been met.  Even if it had, I would think it impossible to find that the rights of urban voters had been interfered with as little as possible.  The impugned legislation cannot therefore be justified under s. 1 of the Charter.

 

Conclusion

 

                  The fundamental right to vote should not be diminished without sound justification.  To water down the importance and significance of an individual's vote is to weaken the democratic process.  Here no sound basis has been put forward to justify legislation which clearly has the effect of diminishing the rights of urban voters and reducing the representation of urban residents in the legislature.  Democracy can all too easily be eroded by diluting voters' rights and representation.  Voting is far too important and precious a right to be unreasonably and unnecessarily diluted.

 

Disposition

 

                  In the result I would dismiss the appeal and answer the reference questions in the same manner as the Saskatchewan Court of Appeal.

 

//McLachlin J.//

 

                  The judgment of La Forest, Gonthier, McLachlin, Stevenson and Iacobucci JJ. was delivered by

 

                  McLachlin J. -- This appeal involves a constitutional challenge to provincial electoral distribution in the province of Saskatchewan.  My conclusion is that the electoral boundaries created by The Representation Act, 1989, S.S. 1989-90, c. R-20.2, do not violate the right to vote enshrined in s. 3 of the Canadian Charter of Rights and Freedoms. 

 

                  I reach this conclusion through consideration of a number of subsidiary issues:

 

I  The Question to be Answered

 

II  Application of the Charter

 

III  Defining the Right to Vote

 

IV  Is the Right to Vote Violated by the Saskatchewan Boundaries?

 

V  Section l and Justification

 

I                  The Question to be Answered

 

                  This case comes to us as an appeal from a reference to the Saskatchewan Court of Appeal (1991), 1991 CanLII 8030 (SK CA), 90 Sask. R. 174.  The reference requested that court's opinion on the following questions:

 

In respect of the constituencies defined in The Representation Act, 1989:

 

(a)Does the variance in the size of voter populations among those constituencies, as contemplated by s. 20 of The Electoral Boundaries Commission Act, S.S. l986-87-88, c. E-6.l,  and recommended in the Sasketchewan Electoral Boundaries Commission 1988 Final Report, infringe or deny rights or freedoms guaranteed by the Canadian Charter of Rights and Freedoms?  If so, in what particulars?  Is any such limitation or denial of rights or freedoms justified by s. 1 of the Canadian Charter of Rights and Freedoms?

 

(b)Does the distribution of those constituencies among urban, rural and northern areas, as contemplated by s. 14 of The Electoral Boundaries Commission Act, S.S. 1986-87-88, c. E-6.1, and recommended in the Saskatchewan Electoral Boundaries Commission 1988 Final Report, infringe or deny rights or freedoms guaranteed by the Canadian Charter of Rights and Freedoms?  If so, in what particulars?  Is any such limitation or denial of rights or freedoms justified by s. 1 of the Canadian Charter of Rights and Freedoms?

 

                  Different views have been expressed as to what issues these questions raise.  The appellant asserts that what is at issue is the constitutional validity of The Representation Act, 1989.  The respondent contends that the question is not whether the Act was unconstitutional, but whether the electoral boundaries created pursuant to the Act violate the Charter.

 

                  I am of the view that it is the boundaries themselves which are at issue on this appeal.  The questions focus, not on the Act, but on the constitutionality of "the variance in the size of voter populations among [the] constituencies" and "the distribution of those constituencies among urban, rural and northern areas".  In so far as The Representation Act, 1989 defines the constituencies, the validity of that Act is indirectly called into question.  And in so far as The Electoral Boundaries Commission Act provides the criteria by which the boundaries are to be fixed, that Act may affect the answers given to the questions posed. But the basic question put to this Court is whether the variances and distribution reflected in the constituencies themselves violate the Charter guarantee of the right to vote.

 

II                 Application of the Charter

 

                  A preliminary question arises of whether the definition of provincial voting constituencies is subject to the Charter.

 

                  The Minister of Justice of the Northwest Territories submits that the Charter does not apply since the legislation whereby constituencies are created is part of the constitution of Canada and hence not subject to the Charter. He submits that the provinces have had the right to establish electoral boundaries since joining Confederation.  In his view, the place of voter equality in this determination is a matter of constitutional convention which is impervious to judicial review.  The right of the provinces to create electoral boundaries as they see fit "must be taken as being an inherent limitation on the right to vote in s. 3."

 

                  I cannot accept this submission.  Although legislative jurisdiction to amend the provincial constitution cannot be removed from the province without a constitutional amendment and is in this sense above Charter scrutiny, the provincial exercise of its legislative authority is subject to the Charter; as McEachern C.J. observed "[i]f the fruit of the constitutional tree does not conform to the Charter... then it must to such extent be struck down": Dixon v. B.C. (A.G.) (1986), 1986 CanLII 770 (BC SC), 7 B.C.L.R. (2d) 174, at p. 188.  The convention for which the Minister contends goes no further than to empower the province to establish its electoral boundaries.  The particular exercise of that power is subject to s. 3 of the Charter, which binds Saskatchewan as it does every province and territory of Canada.

 

III               Defining the Right to Vote

 

                  Section 3 of the Canadian Charter of Rights and Freedoms reads as follows:

 

 

                  The question is simply stated: What is meant by "the right to vote" in s. 3?  Before addressing this question it is necessary to address the way the Court should go about determining the content of the right.

 

A.               General Principles Applicable to Defining the Right

 

                  The content of a Charter right is to be determined in a broad and purposive way, having regard to historical and social context.  As Dickson J. (as he then was) said in R. v. Big M Drug Mart Ltd., 1985 CanLII 69 (SCC), [1985] 1 S.C.R. 295, at p. 344:

 

                  In my view this analysis is to be undertaken, and the purpose of the right or freedom in question is to be sought by reference to the character and the larger objects of the Charter itself, to the language chosen to articulate the specific right or freedom, to the historical origins of the concepts enshrined, and where applicable, to the meaning and purpose of the other specific rights and freedoms with which it is associated within the text of the Charter.  The interpretation should be, as the judgment in Southam emphasizes, a generous rather than a legalistic one, aimed at fulfilling the purpose of the guarantee and securing for individuals the full benefit of the Charter's protection.  At the same time it is important not to overshoot the actual purpose of the right or freedom in question, but to recall that the Charter was not enacted in a vacuum, and must therefore, as this Court's decision in Law Society of Upper Canada v. Skapinker . . . illustrates, be placed in its proper linguistic, philosophic and historical contexts.

 

                  From this general statement of principle I turn to more particular considerations which bear relevance to this appeal.

 

                  The first of these is the doctrine that the Charter is engrafted onto the living tree that is the Canadian constitution: Hunter v. Southam Inc., 1984 CanLII 33 (SCC), [1984] 2 S.C.R. 145, at pp. 155-56; Reference Re B.C. Motor Vehicle Act, 1985 CanLII 81 (SCC), [1985] 2 S.C.R. 486, at p. 509.  Thus, to borrow the words of Lord Sankey in Edwards v. Attorney-General for Canada, 1929 CanLII 438 (UK JCPC), [1930] A.C. 124, at p. 136, it must be viewed as "a living tree capable of growth and expansion within its natural limits."

 

                  The doctrine of the constitution as a living tree mandates that narrow technical approaches are to be eschewed: Law Society of Upper Canada v. Skapinker, 1984 CanLII 3 (SCC), [1984] 1 S.C.R. 357, at p. 366.  It also suggests that the past plays a critical but non-exclusive role in determining the content of the rights and freedoms granted by the Charter.  The tree is rooted in past and present institutions, but must be capable of growth to meet the future.  As Dickson J. stated in R. v. Big M Drug Mart Ltd., supra, at pp. 343-44:

 

...the Charter is intended to set a standard upon which present as well as future legislation is to be tested.  Therefore the meaning of the concept of freedom of conscience and religion is not to be determined solely by the degree to which that right was enjoyed by Canadians prior to the proclamation of the Charter. [Emphasis in original.]

 

This admonition is as apt in defining the right to vote as it is in defining freedom of religion.  The right to vote, while rooted in and hence to some extent defined by historical and existing practices, cannot be viewed as frozen by particular historical anomalies.  What must be sought is the broader philosophy underlying the historical development of the right to vote -- a philosophy which is capable of explaining the past and animating the future.

 

                  This appeal also engages the general principle that practical considerations must be borne in mind in constitutional interpretation: R. v. Edwards Books and Art Ltd., 1986 CanLII 12 (SCC), [1986] 2 S.C.R. 713.  Courts must be sensitive to what Frankfurter J. (McGowan v. Maryland, 366 U.S. 420 (1961)) calls "the practical living facts" to which a legislature must respond: per La Forest J. in Edwards Books, supra, at pp. 794-95, approved in R. v. Schwartz, 1988 CanLII 11 (SCC), [1988] 2 S.C.R. 443; United States of America v. Cotroni, 1989 CanLII 106 (SCC), [1989] 1 S.C.R. 1469.  This is nowhere more true than in considering the right to vote, where practical considerations such as social and physical geography may impact on the value of the citizen's right to vote.

 

                  Of final and critical importance to this appeal is the canon that in interpreting the individual rights conferred by the Charter the Court must be guided by the ideal of a "free and democratic society" upon which the Charter is founded.  As Dickson C.J. stated in R. v. Oakes, 1986 CanLII 46 (SCC), [1986] 1 S.C.R. 103, at p. 136:

 

The Court must be guided by the values and principles essential to a free and democratic society which I believe embody, to name but a few, respect for the inherent dignity of the human person, commitment to social justice and equality, accommodation of a wide variety of beliefs, respect for cultural and group identity, and faith in social and political institutions which enhance the participation of individuals and groups in society.

 

                  The first task on an appeal such as this is to define the scope of the right to vote under s. 3 of the Charter.  The second is to evaluate the existing electoral boundaries in the light of that definition to determine if they violate s. 3 of the Charter.  If a violation is found, a third task arises -- determining whether the limitation on the right is "demonstrably justified in a free and democratic society" and hence saved under s. 1 of the Charter.  The general principles to which I have referred, while bearing particularly on the task of defining the ambit of the right, also animate the second and third steps of the analysis.

 

B.               The Focus of the Debate

 

                  The question for resolution on this appeal can be summed up in one sentence: to what extent, if at all, does the right to vote enshrined in the Charter permit deviation from the "one person - one vote" rule?  The answer to this question turns on what one sees as the purpose of s. 3.  Those who start from the premise that the purpose of the section is to guarantee equality of voting power support the view that only minimal deviation from that ideal is possible.  Those who start from the premise that the purpose of s. 3 is to guarantee effective representation see the right to vote as comprising many factors, of which equality is but one.  The contest, as I see it, is most fundamentally between these two views, although the submissions before us vary in the emphasis they place on different factors and hence on where they would draw the line.

 

                  The Saskatchewan Court of Appeal, as I read its reasons, fell into the camp of those who see the purpose of s. 3 as guaranteeing equality of voting power per se. It suggested that the only deviation permissible from the ideal of equality under s. 3 is that required by the practical problems of ensuring that the number of voters in each constituency is mathematically equal on the day of voting (at pp. 21, 24).  On the basis of this definition, it found that the electoral boundaries in Saskatchewan violated s. 3 of the Charter.  Other considerations, such as geography, historical boundaries and community interests, fell to be considered under s. 1.  The court found that the boundaries were not justified under s. 1, except for the two northern ridings where population is extremely sparse.

 

                  In this Court, the respondent, supporting the judgment of the Court of Appeal, urged that the goal of s. 3 is equality of voting power, as nearly as may possibly be achieved.  The appellant, while not going so far as to deny the importance of equality in a meaningful right to vote, urged that equality was but one of many factors relevant to the right to vote enshrined in s. 3 and that the fundamental purpose of s. 3 was not to ensure equality of voting power, but effective and fair representation conducive to good government.  The interveners tended to ally themselves with one of these two positions, stressing their own particular perspectives.  For example, Equal Justice for All urged no deviation from equality, except as might be justified in aid of disadvantaged groups, while the Attorney General for Alberta went so far as to deny equality's place as a "core" or "fundamental" value in assessing the right to vote.

 

C.               The Meaning of the Right to Vote

 

                  It is my conclusion that the purpose of the right to vote enshrined in s. 3 of the Charter is not equality of voting power per se, but the right to "effective representation".  Ours is a representative democracy.  Each citizen is entitled to be represented in government.  Representation comprehends the idea of having a voice in the deliberations of government as well as the idea of the right to bring one's grievances and concerns to the attention of one's government representative;  as noted in Dixon v. B.C. (A.G.), 1989 CanLII 248 (BC SC), [1989] 4 W.W.R. 393, at p. 413, elected representatives function in two roles -- legislative and what has been termed the "ombudsman role".

 

                  What are the conditions of effective representation?  The first is relative parity of voting power.  A system which dilutes one citizen's vote unduly as compared with another citizen's vote runs the risk of providing inadequate representation to the citizen whose vote is diluted.  The legislative power of the citizen whose vote is diluted will be reduced, as may be access to and assistance from his or her representative.   The result will be uneven and unfair representation.

 

                  But parity of voting power, though of prime importance, is not the only factor to be taken into account in ensuring effective representation.  Sir John A. Macdonald in introducing the Act to re-adjust the Representation in the House of Commons, S.C. l872, c. 13, recognized this fundamental fact (House of Commons Debates, Vol. III, 4th Sess., p. 926 (June 1, 1872)):

 

...it will be found that,... while the principle of population was considered to a very great extent, other considerations were also held to have weight; so that different interests, classes and localities should be fairly represented, that the principle of numbers should not be the only one.

 

                  Notwithstanding the fact that the value of a citizen's vote should not be unduly diluted, it is a practical fact that effective representation often cannot be achieved without taking into account countervailing factors.

 

                  First, absolute parity is impossible. It is impossible to draw boundary lines which guarantee exactly the same number of voters in each district. Voters die, voters move.  Even with the aid of frequent censuses, voter parity is impossible.

 

                  Secondly, such relative parity as may be possible of achievement may prove undesirable because it has the effect of detracting from the primary goal of effective representation. Factors like geography, community history, community interests and minority representation may need to be taken into account to ensure that our legislative assemblies effectively represent the diversity of our social mosaic.  These are but examples of considerations which may justify departure from absolute voter parity in the pursuit of more effective representation; the list is not closed.

 

                  It emerges therefore that deviations from absolute voter parity may be justified on the grounds of practical impossibility or the provision of more effective representation.   Beyond this, dilution of one citizen's vote as compared with another's should not be countenanced.  I adhere to the proposition asserted in Dixon, supra, at p. 414, that "only those deviations should be admitted which can be justified on the ground that they contribute to better government of the populace as a whole, giving due weight to regional issues within the populace and geographic factors within the territory governed."

 

                  This view of the meaning of the right to vote in s. 3 of the Charter conforms with the general principles of interpretation discussed at the outset.

 

                  The first and most important rule is that the right must be interpreted in accordance with its purpose.  As will be seen, there is little in the history or philosophy of Canadian democracy that suggests that the framers of the Charter in enacting s. 3 had as their ultimate goal the attainment of voter parity.  That purpose would have represented a rejection of the existing system of electoral representation in this country.  The circumstances leading to the adoption of the Charter negate any intention to reject existing democratic institutions. As noted in Dixon, supra, at p. 412: "There is no record of such fundamental institutional reform having been mentioned at the conferences that preceded the adoption of the [proposed] Charter".  Nor was the issue raised by any of the plethora of interest groups making submissions in respect of voting rights during the prolonged Joint Senate and House of Commons Committee Hearings on the proposed Charter.  The framers of the Charter had two distinct electoral models before them -- the "one person - one vote" model espoused by the United States Supreme Court in Baker v. Carr, 369 U.S. 186 (1962), Karcher v. Daggett, 462 U.S. 725 (1983), and Kirkpatrick v. Preisler, 394 U.S. 526 (1969), and the less radical, more pragmatic approach which had developed in England and in this country through the centuries and which was actually in place.  In the absence of any supportive evidence to the contrary (as may be found in the United States in the speeches of the founding fathers), it would be wrong to infer that in enshrining the right to vote in our written constitution the intention was to adopt the American model.  On the contrary, we should assume that the goal was to recognize the right affirmed in this country since the time of our first Prime Minister, Sir John A. Macdonald, to effective representation in a system which gives due weight to voter parity but admits other considerations where necessary.

 

                  I turn next to the history of our right to vote.  As already noted, the history of our right to vote and the context in which it existed at the time the Charter was adopted support the conclusion that the purpose of the guarantee of the right to vote is not to effect perfect voter equality, in so far as that can be done, but the broader goal of guaranteeing effective representation.  As I noted in Dixon, supra, at p. 409, democracy in Canada is rooted in a different history than in the United States:

 

Its origins lie not in the debates of the founding fathers, but in the less absolute recesses of the British tradition.  Our forefathers did not rebel against the English tradition of democratic government as did the Americans; on the contrary, they embraced it and changed it to suit their own perceptions and needs.

 

I went on to describe the Canadian tradition as one of evolutionary democracy moving in uneven steps toward the goal of universal suffrage and more effective representation, which even in its advanced stages tolerates deviation from voter parity in the interests of better representation:

 

                  What is that tradition? It was a tradition of evolutionary democracy, of increasing widening of representation through the centuries.  But it was also a tradition which, even in its more modern phases, accommodates significant deviation from the ideals of equal representation.  Pragmatism, rather than conformity to a philosophical ideal, has been its watchword.

 

                  Other Commonwealth countries have affirmed the same tradition.  Thus the Australian High Court rejected a "one person - one vote" approach in favour of an approach which permitted consideration of countervailing factors: Attorney-General (Aus.); Ex rel. McKinlay v. Commonwealth (1975), 135 C.L.R. 1. Stephen J. wrote, at p. 57:

 

                  It is, then, quite apparent that representative democracy is descriptive of a whole spectrum of political institutions, each differing in countless respects yet answering to that generic description....

 

                  To contend that the presence of what is described as "as near as practicable equality of numbers" within electoral divisions is essential to representative democracy, to a legislature "chosen by the people", is to deny proper meaning to language and to ignore long chapters in the evolution of democratic institutions both in this country and overseas, in which, representative democracy having been attained, its details have undergone frequent changes in response to community pressures but have failed to possess this feature of equality of numbers on which the plaintiffs now insist.

 

See also Gibbs J., at p. 45 and Barwick C.J., at p. 25.

 

                  To return to the metaphor of the living tree, our system is rooted in the tradition of effective representation and not in the tradition of absolute or near absolute voter parity.  It is this tradition that defines the general ambit of the right to vote.  This is not to suggest, however, that inequities in our voting system are to be accepted merely because they have historical precedent.  History is important in so far as it suggests that the philosophy underlying the development of the right to vote in this country is the broad goal of effective representation.  It has nothing to do with the specious argument that historical anomalies and abuses can be used to justify continued anomalies and abuses, or to suggest that the right to vote should not be interpreted broadly and remedially as befits Charter rights.  Departures from the Canadian ideal of effective representation may exist. Where they do, they will be found to violate s. 3 of the Charter.

 

                  I turn finally to the admonition that courts must be sensitive to practical considerations in interpreting Charter rights.  The "practical living fact", to borrow Frankfurter J.'s phrase, is that effective representation and good government in this country compel those charged with setting electoral boundaries sometimes to take into account factors other than voter parity, such as geography and community interests.   The problems of representing vast, sparsely populated territories, for example, may dictate somewhat lower voter populations in these districts; to insist on voter parity might deprive citizens with distinct interests of an effective voice in the legislative process as well as of effective assistance from their representatives in their "ombudsman" role.  This is only one of a number of factors which may necessitate deviation from the "one person - one vote" rule in the interests of effective representation.

 

                  In the final analysis, the values and principles animating a free and democratic society are arguably best served by a definition that places effective representation at the heart of the right to vote.  The concerns which Dickson C.J. in Oakes associated with a free and democratic society -- respect for the inherent dignity of the human person, commitment to social justice and equality, respect for cultural and group identity, and faith in social and political institutions which enhance the participation of individuals in society -- are better met by an electoral system that focuses on effective representation than by one that focuses on mathematical parity.  Respect for individual dignity and social equality mandate that citizen's votes not be unduly debased or diluted.  But the need to recognize cultural and group identity and to enhance the participation of individuals in the electoral process and society requires that other concerns also be accommodated.

 

                  In summary, I am satisfied that the precepts which govern the interpretation of Charter rights support the conclusion that the right to vote should be defined as guaranteeing the right to effective representation.  The concept of absolute voter parity does not accord with the development of the right to vote in the Canadian context and does not permit of sufficient flexibility to meet the practical difficulties inherent in representative government in a country such as Canada.  In the end, it is the broader concept of effective representation which best serves the interests of a free and democratic society.

 

IV               Do the Saskatchewan Boundaries Violate the Right to Vote?

 

A.               The Issue

 

                  It is important at the outset to remind ourselves of the proper role of courts in determining whether a legislative solution to a complex problem runs afoul of the Charter.  This Court has repeatedly affirmed that the courts must be cautious in interfering unduly in decisions that involve the balancing of conflicting policy considerations: see Reference Re Public Service Employee Relations Act (Alta.), 1987 CanLII 88 (SCC), [1987] 1 S.C.R. 313, per Le Dain J., at p. 392; Canada v. Schmidt, 1987 CanLII 48 (SCC), [1987] 1 S.C.R. 500, per La Forest J., at pp. 522-23; Canada (Auditor General) v. Canada (Minister of Energy, Mines and Resources), 1989 CanLII 73 (SCC), [1989] 2 S.C.R. 49, per Dickson C.J., at pp. 90-92.  These considerations led me to suggest in Dixon, supra, at p. 419, that "the courts ought not to interfere with the legislature's electoral map under s. 3 of the Charter unless it appears that reasonable persons applying the appropriate principles ... could not have set the electoral boundaries as they exist."

 

                  Before turning to the constituencies in question, it is necessary to clarify what is being tested.  As noted at the outset, the issue in this appeal concerns the "variance" in voter populations among constituencies and the "distribution" of constituencies among urban, rural and northern areas.  This wording suggests a focus on the result obtained rather than the process, although, as will be seen, I conclude that the process used here did not in fact violate s. 3.

 

                  The Court of Appeal focused on the constitutionality of The Electoral Boundaries Commission Act.  It held at p. 189 that the Act was unconstitutional because it did not "direct the Commission to be guided by the fundamental principle of equality of voting power", and in fact barred the electoral commission from applying the principle of equality by requiring a specified number of urban, rural and northern seats.  As the second stage of its analysis, the court considered the effect of the legislation.  After pointing out a number of discrepancies between various ridings, the court concluded that the electoral boundaries themselves violated the right to vote guaranteed under the Charter.  In arriving at this conclusion, the court applied a s. 3 test which I have suggested is wrong -- the test of voter parity in so far as it is possible of achievement.  It is therefore necessary to address this question anew in the light of the test for s. 3 which I have proposed.

 

B.               The Boundaries

 

                  The Electoral Boundaries Commission Act required the electoral commission to create 29 urban, 35 rural and 2 northern ridings.  The 64 urban and rural ridings fall roughly into the south half of the province, while the two northern ridings make up its north half.  In the southern half of the province, the voter population of each constituency is within plus or minus 25 percent of the provincial quotient.  The Act specifically permitted the two northern ridings to vary from the provincial quotient by up to plus or minus 50 percent.  The Court of Appeal, under s.1, found that special treatment for northern ridings was constitutionally acceptable, and no issue is taken on that point.  The main focus is on the southern ridings, which the Court of Appeal found violate s. 3 of the Charter.

 

                  The question is whether the deviations from voter parity in southern ridings can be justified on the basis of valid considerations.  The respondent suggests that the voter population disparities between ridings cannot be justified and violate s. 3. In support of this he argues: (l) that the electoral commission, constrained as it was by the legislation, acted arbitrarily; and (2) that in fact there are discrepancies in the population of various ridings which are unjustified.  The first argument is concerned mainly with process; the second with result.

 

                  I turn first to the proposition that the electoral commission acted arbitrarily and without due regard for the need to maintain relative voter parity.  The argument in support of this position, accepted by the Court of Appeal, is that the electoral commission was improperly prevented from giving due weight to voter equity because The Electoral Boundaries Commission Act required that it produce an electoral map with a specified number of urban, rural and northern seats.

 

                  This argument overlooks the genesis of the stipulation in the legislation and the actual population distribution that underpinned the allocation of urban and rural seats.  The allotment of seats to the various urban centres in The Representation Act, 1989 flows logically from the electoral map that it replaced.  Under The Representation Act, 1981 the number of seats in the various urban centres was as follows:  Moose Jaw - 2; Regina - 10; Saskatoon - 10; Swift Current - 1; North Battleford - 1; Prince Albert had one seat within city limits and another, Prince Albert-Duck Lake, which was comprised of part of the City and a small part of a neighbouring area.  The City of Yorkton was part of a riding comprised of the City itself and a very small surrounding area.  This map was made by an impartial commission not required to establish a particular number of rural or urban ridings.

 

                  The configuration of urban seats in The Representation Act, 1989 simply reflects population growth in those areas from the time the earlier map was put in place.  The 1989 map includes additional seats for the growing centres of Saskatoon and Regina.  The increased size of the City of Yorkton is reflected in the fact that it became a self-contained riding shorn of the surrounding rural area.  Similarly, Prince Albert, another rapidly growing city, was divided into two seats comprised exclusively of the urban area itself.  This is not an "arbitrary" allocation of constituencies.  It is founded on the electoral map made by an impartial and unfettered commission in 1981 and the population growth that has since occurred.

 

                  Overall, the electoral map put in place by The Representation Act, 1989 admits of a tendency for "urban" seats to have more voters than "rural" seats.  Urban ridings generally have somewhat more voters than the quotient and rural ridings generally have somewhat fewer.  The discrepancies, however, are not great and there are a number of exceptions.  Several rural seats are larger than a number of urban ones.  Moreover, "rural" seats are not necessarily "farm" or "agricultural" seats.  A number of relatively major centres including Weyburn, Estevan, Melville, Nipawin, Melfort and Lloydminster are included in these "rural" areas.

 

                  The actual allocation of seats between urban and rural areas is very close to the population distribution between those areas.  The rural areas have 53.0 percent of the seats and 50.4 percent of the population.  Urban areas have 43.9 percent of the seats and 47.6 percent of the population.  The rural areas are, therefore, somewhat over-represented, and the urban areas somewhat under-represented, but these deviations are relatively small.  Similar deviations occurred on the redistributions proposed by the Constituency Boundaries Commissions of 1974 and 1979-80.  For example, the 1979-80 Report allocated only 40.6 percent of seats to urban areas even though they had 42.6 percent of the population.  In the 1974 redistribution, urban areas had only 36.1 percent of the seats but 38.9 percent of the population.  It is thus seen that the effect of the allocation of seats to urban and rural ridings in the 1989 legislation was mainly to increase the number of urban seats to reflect population increases in urban areas.  This belies the suggestion that the 1989 Act was an unjustified attempt to adjust boundaries to benefit the governing party.

 

                  The argument that the commission was arbitrarily constrained by the governing legislation may also be criticized on the ground that it assumes an unduly constrained view of The Electoral Boundaries Commission Act.  Section 20 of the Act sets out the criteria which must govern the electoral map:

 

20  A commission, in determining the area to be included in and in fixing the boundaries of all proposed constituencies:

 

(a)  shall determine a constituency population quotient by dividing the voter population by the number of constituencies, from which:

 

(i)  no proposed southern constituency population shall vary, subject to section 14 and subsection 15(1), by more than 25%;

 

(ii)  no proposed northern constituency population shall vary, subject to section 14, by more than 50%;

 

(b)  may use the allowable variation from the population quotient mentioned in clause (a) to accommodate:

 

(i)  the sparsity, density or relative rate of growth of population of any proposed constituency;

 

(ii)  any special geographic features including size and means of communication between the various parts of the proposed constituency;

 

(iii)  the community or diversity of interests of the population, including variations in the requirements of the population of any proposed constituency; and

 

(iv)  other similar or relevant factors.

 

                  The Commission adhered to these criteria in setting the boundaries, applying a test consistent with s. 3 of the Charter as I have interpreted it, noting at p. 4 of its Final Report:

 

                  In the opinion of the Commission, the principle of representation by population is recognized in the legislation by the establishment of a constituency voter population quotient.  Clearly the Act by necessary inference implies that such voter population quotient must be the benchmark for all constituencies.  The right of the Commission to depart from that quotient is not an absolute one.  It is entitled to depart therefrom only for the reasons set forth in the Act and only to the extent that the special circumstances properly permit, and the legislation requires.  This was the interpretation followed by the Commission in submitting its Interim Report and in reviewing the representations made whether written or oral, in respect to that Report. [Emphasis added.]

 

                  I am satisfied that the proposition that the Commission was unduly constrained by the governing legislation and consequently failed to take into consideration the appropriate factors must fail.  The process, viewed as a whole, was fair.  The original division between urban and rural ridings was the work of an unimpeded commission; the subsequent adjustment largely reflected population changes, and gave due weight to the principle of voter parity.  The fact that the legislature was involved in the readjustment does not in itself render the process arbitrary or unfair, in my view.

 

                  I turn then to the contention that the distribution of seats itself violates s. 3 of the Charter.  As already noted, variances between southern seats fall within plus or minus 25 percent of the provincial quotient.  Moreover, the distribution between urban and rural seats closely approximates the actual split between urban and rural population.  It remains, however, to consider whether unjustifiable deviations exist with respect to particular ridings in the southern half of the province.

 

                  Before examining the electoral boundaries to determine if they are justified, it may be useful to mention some of the factors other than equality of voting power which figure in the analysis.  One of the most important is the fact that it is more difficult to represent rural ridings than urban.  The material before us suggests that not only are rural ridings harder to serve because of difficulty in transport and communications, but that rural voters make greater demands on their elected representatives, whether because of the absence of alternative resources to be found in urban centres or for other reasons.  Thus the goal of effective representation may justify somewhat lower voter populations in rural areas.  Another factor which figured prominently in the argument before us is geographic boundaries; rivers and municipal boundaries form natural community dividing lines and hence natural electoral boundaries.  Yet another factor is growth projections.  Given that the boundaries will govern for a number of years -- the boundaries set in 1989, for example, may be in place until 1996 -- projected population changes within that period may justify a deviation from strict equality at the time the boundaries are drawn.

 

                  Against this background, I turn to the boundaries themselves.

 

                  The Commission did not address deviations on a riding by riding basis in its report, contenting itself with a general description of the factors it relied on in establishing the boundaries.  It did, however, point out the importance of geography in drawing boundaries in the sparsely populated southwestern areas, where river banks often serve to demarcate distinct regions and communities and additionally affect transportation and the ease of servicing the populace.  The Commission also commented specifically on the two ridings showing the greatest deviation, Morse Constituency and Humboldt Constituency.  In each case, it provided good reasons in its Final Report, at p. 7, for the degree of variation:

 

                  The two proposed constituencies to which criticism was primarily directed were those of Morse and Humboldt.  Morse has the smallest voter population of any rural constituency and Humboldt has the largest.  This is understandable.  Morse lies in that area of rural Saskatchewan where there is a sparsity of population whereas Humboldt encompasses an area of the province in which there is a much denser population.

 

                  The Commission sees no benefit to be gained by altering the boundaries of Morse constituency.  A study of the map will show that the adjoining constituencies are also sparsely settled areas.  To add voters to the proposed constituency of Morse would only reduce the voter population in the surrounding constituencies for no beneficial purpose.  As well, the northern boundary of the Morse constituency is the South Saskatchewan River, a true natural boundary.  Moreover, the constituency of Morse surrounds the City of Swift Current.  If satellite villages should develop outside the City of Swift Current, as some people have suggested, the voter population will increase accordingly.

 

                  The recommendations made in this Final Report go some distance in reducing voter population in the constituency of Humboldt.  The Commission feels that this constituency has reached its optimum population.  It is also of the opinion that eventually the central and park area of the province will experience to some extent, the same changes that have occurred in the prairies areas.  With the increased use of large equipment, it is likely that farms in that area of Saskatchewan will become larger with consequent loss of population in the future.

 

                  A third riding which was criticized was Saskatoon Greystone, with a variance of plus 23 percent, which adjoins Saskatoon Sutherland - University with a variance of minus 24 percent.  A view of the electoral map for Saskatoon reveals that Saskatoon Greystone is entirely built up, while Saskatoon Sutherland-University is not.  It may be, as the appellant suggests, that the potential for future increases in the population of Saskatoon Sutherland - University is a factor in the discrepancy.  On the other hand, the respondent has presented no evidence apart from population figures supporting the contention that the variance between these two ridings is illogical or arbitrary.

 

                  I have earlier suggested that population discrepancies between urban and rural ridings are not great.  In so far as the election map may separate certain dormitory communities from adjacent rural ridings, it is not self-evident that such communities should be joined with the communities where the residents worked.  Their interests may differ from those of the community in the urban riding, and their inclusion might sweep in truly rural residents.

 

                  In summary, the evidence supplied by the province is sufficient to justify the existing electoral boundaries.  In general, the discrepancies between urban and rural ridings is small, no more than one might expect given the greater difficulties associated with representing rural ridings.  And discrepancies between particular ridings appear to be justified on the basis of factors such as geography, community interests and population growth patterns.  It was not seriously suggested that the northern boundaries are inappropriate, given the sparse population and the difficulty of communication in the area.  I conclude that a violation of s. 3 of the Charter has not been established.

 

                  In these circumstances, it is unnecessary to consider s. 1.

 

V  Conclusion

 

                  I would allow the appeal and answer both Reference Questions in the negative.     

 

//Sopinka J.//

 

                  The following are the reasons delivered by

 

                  Sopinka J. -- I have read the reasons of my colleagues, Justice Cory and Justice McLachlin, and while I agree with the result reached by McLachlin J. and substantially with her reasons, I would approach the interpretation of s. 3 of the Canadian Charter of Rights and Freedoms differently.

 

                  In my opinion, in using the simple words in s. 3 that "[e]very citizen ... has the right to vote ...", the framers did not intend to invent or give birth to a right not previously enjoyed by the citizens of Canada.  Indeed, it was frankly conceded in argument that the right to vote had existed in Saskatchewan prior to 1982.  Accordingly, in interpreting s. 3, the primary inquiry is to determine on what principles the right to vote, which has existed in this country for many years, was based.

 

                  A review of the historical background shows that not only in Saskatchewan, but in other provinces as well, the drawing of electoral boundaries has been governed by the attempt to achieve voter equality with liberal allowances for deviations based on the kinds of considerations which are enumerated in s. 20 of The Electoral Boundaries Commission Act, S.S. 1986-87-88, c. E-6.1.  Deviations were avoided which deprived voters of fair and effective representation.  Under the Charter, the deviations are subjected to judicial scrutiny and must not be such as to deprive voters of fair and effective representation.

 

                  The questions raised in this appeal require us to opine on the legislative result of the process embodied in The Electoral Boundaries Commission Act.  The product of this process is contained in The Representation Act, 1989, S.S. 1989-90, c. R-20.2.  My colleague Cory J. is of the view that once an independent boundaries commission was established, it was incumbent upon the Saskatchewan legislature to ensure that the commission was able to fulfill its mandate freely without unnecessary interference such as that contained in s. 14 of The Electoral Boundaries Commission Act.  With respect, I cannot agree.  Cory J.'s position assumes that there is some kind of constitutional guarantee for the process.  It was not necessary for the Saskatchewan legislature to create an independent commission, and, had it simply legislated the impugned boundaries, the process itself would not have been subject to judicial scrutiny.  Having chosen to delegate the task to the commission, there is no reason why the legislature should be prohibited from laying down tight guidelines delineating the powers to be conferred on the commission.

 

                  With respect to the guidelines, they are set out in ss. 14 and 20 of The Electoral Boundaries Commission Act.  The factors in s. 20 could be applied in such a way as to produce deviations that deprived voters of fair and effective representation but equally, they could be applied to achieve that objective.  Similarly, while s. 14 of The Electoral Boundaries Commission Act, which mandates a fixed number of rural and urban ridings, could have resulted in producing variations from the objective which were so extreme as to amount to a breach of the right to vote, it did not have that effect in this case.  We are not, therefore, concerned in this case with the constitutional validity of the factors in s. 20 or s. 14 but with the effect that their application has produced.

 

                  I am in agreement with the finding of Cory J. that the electoral boundaries established by the 1981 map are fair and did not violate the right to vote.  The boundaries proposed in The Representation Act, 1989 adopt the existing electoral map of Saskatchewan with the addition of two urban ridings.  The addition of the two urban constituencies reflects the increase in voter population in the relevant areas.  The extent of deviation from strict voter equality, as well as the reasons for those deviations, are comparable to those which inspired the 1981 map.  In these circumstances, it cannot be said that the deviations established by The Representation Act, 1989 are so extensive as to deny fair and effective representation.  In these circumstances, they do not infringe the right to vote entrenched in s. 3 of the Charter.

 

                  Accordingly, I would dispose of the appeal as proposed by McLachlin J.

 

                  Appeal allowed, Lamer C.J. and L'Heureux-Dubé and Cory JJ. dissenting.

 

                  Solicitor for the appellant:  Brian Barrington‑Foote, Regina.

 

                  Solicitor for the respondent:  Roger Carter, Saskatoon.

 

                  Solicitor for the intervener the Attorney General of Canada:  John C. Tait, Ottawa.

 

                  Solicitor for the intervener the Attorney General of Quebec: The Attorney General of Quebec, Ste‑Foy.

 

                  Solicitor for the intervener the Attorney General of British Columbia: The Attorney General of British Columbia, Victoria.

 

                  Solicitors for the intervener the Attorney General of Prince Edward Island:  Stewart McKelvey Stirling Scales, Charlottetown.

 

                  Solicitors for the intervener the Attorney General for Alberta:  Field & Field Perraton Masuch, Edmonton.

 

                  Solicitor for the intervener the Attorney General of Newfoundland:  Paul D. Dicks, St. John's.

 

                  Solicitor for the intervener the Minister of Justice of the Northwest Territories: The Minister of Justice of the Northwest Territories, Yellowknife.

 

                  Solicitors for the intervener the Minister of Justice of the Yukon:  Arvay, Finlay, Victoria.

 

                  Solicitor for the intervener John F. Conway:  John F. Conway, Regina.

 

                  Solicitor for the intervener the British Columbia Civil Liberties Association:  Robert D. Holmes, Vancouver.

 

                  Solicitor for the interveners Douglas Billingsley, Wilson McBryan, Leonard Jason and Daniel Wilde:  Donald J. Boyer, Edmonton.

 

                  Solicitors for the intervener the Alberta Association of Municipal Districts and Counties:  Brownlee Fryett, Edmonton.

 

                  Solicitors for the interveners the cities of Edmonton and Grande Prairie:  Timothy J. Christian and June M. Ross, Edmonton.

 

                  Solicitor for the intervener Equal Justice for All:  Larry W. Kowalchuk, Saskatoon.