Canadian Parliamentary Review

Current Issue
Canadian Region CPA
Upcoming Issue
Editorial and Stylistic Guidelines

HomeContact UsFranšais

Constitutional Amendment and Constituent Assemblies
Clyde Wells

At the time this article was written Clyde Wells was Premier of Newfoundland and Labrador. This article is based on his statement to the Special Joint Committee on the Processing of Amending the Constitution on April 9, 1991.

While our current constitutional problems have a variety of facets they reflect two essential problems, both of which require us to focus on the constitutional amending process:

First, there is a substantial degree of dissatisfaction with many of our existing constitutional provisions particularly in Quebec but also in Western Canada and Atlantic Canada. There are also widely divergent views as to what constitutional changes are appropriate to address the present dissatisfaction. The differences are substantial and reflect disagreement as to the fundamental nature of the country.

Second, the existing amending formula is not adequate to deal with the problem. Unanimity is not achievable for those kinds of amendments requiring the agreement of all provinces, and there is a reluctance to use the general amending formula for those amendments that do not require unanimity because of Quebec's position that it would neither negotiate nor participate in the process. There is no formal provision in the Constitution at the moment that would provide for the use of a constitutional convention, or a national referendum.

While it is not impossible, it is improbable that acceptable constitutional changes can be negotiated and implemented under our present amending procedures. As matters now stand alteration of those amending procedures will require the unanimous approval of all of the provinces and that too is improbable in the present circumstances.

It is therefore important to re-examine the current amending procedures in our Constitution as part of the ongoing constitutional debate. Indeed, the drafters of the 1982 Constitution Act recognized that such a re-examination would be necessary and provided in section 49 for a constitutional conference of first ministers to be convened by the Prime Minister within fifteen years of 1982 to review the provisions dealing with constitutional amendment...

Unanimity found its way into the Constitution as a means of avoiding special status for Quebec. In lieu of giving only Quebec a constitutional veto, one was given to all provinces by requiring unanimity. Neither special status nor unanimity is appropriate because each is capable of producing constitutional paralysis. In the case of special status it can result from the intransigent position of the particular province having the right of veto and in the case of unanimity from the intransigence of any one province.

My personal view is that the requirement for unanimity should be eliminated entirely. Our amending formula is absurd insofar as it allows a single province to hold up important reforms. That absurdity would have been exacerbated under the Meech Lake Accord which would have extended the unanimity requirement to even more areas.

No one province, whether it is as large as Ontario or as small as Prince Edward Island, should be in a position, ever, to hold up the constitutional development of the nation. The general amending formula should be used for all amendments.

Just how ludicrous the requirement for unanimity is was obvious in the case of the Meech Lake Accord. The citizens and governments of Newfoundland and Manitoba were castigated by some as nation-wreckers when, acting in accord with the sincere conviction of the overwhelming majority of their people, they did not approve of the constitutional changes proposed in the Meech Lake Accord thereby denying the unanimity required. Yet, if the amendments in the Meech Lake Accord had required only the general amending formula, the actions of Manitoba and Newfoundland in refusing to approve would have been entirely acceptable. It is an absurdity to suggest that all provinces have a right to approve or refuse approval, as their people in conscience see fit, but small provinces dare not withhold approval in cases where unanimity is required and the larger provinces have approved. I cannot think of a position more offensive to the democratic process than that.

While there is no justification for unanimity, I believe there is justification for a limited constitutional veto for Quebec. Even while recognizing that our aboriginal people were the first citizens of this land, and that Canada is today, as it will be in the future, the beneficiary of contributions of people from a variety of cultural and ethnic backgrounds, we cannot ignore the historical fact that the Canadian nation was founded on the basis of an understanding between the French and English- speaking peoples of the colonies in North America, then administered by Britain, to build a nation encompassing their two cultures and using their two languages and two legal systems. If that understanding is to be honoured today, I believe that it can only be fully honoured by a commitment to promote those two cultures, accommodate the two legal systems and build, in the decades ahead, a bilingual nation from coast to coast. If this is accepted as the constitutional precept I believe it to be, then it must also be appropriately reflected in Parliament and the functioning of our national institutions, and appropriately accommodated in our constitutional amending formula.

That accommodation can be achieved through the implementation in a reformed Senate or even in the existing Senate, of a mechanism of separate linguistic voting on constitutional changes affecting language, culture, and the civil law system. Such changes would require the approval of the majority of senators from Quebec separately from, and in addition to, the approval of the majority of the senators from the other provinces collectively. This double majority principle could also be extended in appropriate cases to certain types of federal legislation specifically altering language or cultural rights.

If the ideal were being pursued, the Senate would be divided into two divisions with all Senators declaring themselves to be francophone in one division and all Senators declaring themselves to be anglophone in the other division, without regard to province or territory of origin. However, one must recognize that Quebec with eighty-five percent of its population living and working in the French language is the primary repository of the French language and culture. As well, only Quebec has the civil law system. These facts, together with a desire to resolve the constitutional dilemma facing the country today, justify establishing a limited double majority veto in the Senate by dividing the Senate into Quebec and non-Quebec senators.

A reformed Senate and the linguistic voting divisions could be used as well to approve appointments to the Supreme Court of Canada with an effective veto over appointment of civil law judges given to Quebec senators and over appointment of common law judges given collectively to the senators from the other provinces.

Other than amendments affecting culture, language and the civil law system, I believe every amendment to the Constitution should become effective upon receiving the approval of the Senate, the House of Commons and the legislatures of seven of the ten provinces having fifty percent of the population. Our constitutional development should never be held captive by the straightjacket that results from either unanimity or a special status right of veto for a particular province.

It has also been suggested that if there had been a one-year time limit, the Meech Lake constitutional changes would have been approved. Many others would say that is in itself the soundest possible argument against a shorter time limit. Those who blame the changes in provincial governments within the three-year period for the failure of the Meech Lake Accord avoid facing the real reason. The truth is that the impact and nature of the constitutional changes proposed in the Meech Lake Accord became known to, and sufficiently if not perfectly understood by, the Canadian people during those three years, and by the end of the period the proposed changes were totally unacceptably to an overwhelming majority of Canadians. One can only marvel at the insensitivity to democratic rights displayed by those who would suggest the time limit should be shortened in order to avoid a similar opportunity to assess and understand major or complex amendments in the future.

Whether we have unanimity for some amendments and the general amending formula for others or the general amending formula for all, proposed changes should be implemented if, as and when the appropriate level of approval is achieved. If any legislature thinks the proposal is outstanding for an undue length of time, it can take action to rescind its resolution of approval. Either the House of Commons or Senate could do likewise. The United States, the original federal state, has never had any difficulty with an amending formula without such a constitutional time limit. On occasion, the congressional resolution to amend has itself incorporated a time limit, and this has served very well in any circumstance where a time limit may for some reason have been desirable. Thus, instead of being shortened, I would suggest the time limit should be eliminated. It serves no purpose and could be detrimental.

Constitutional Conventions

I do not think there is any merit whatsoever in suggesting that our existing legislative procedure could be replaced with either a constitutional convention or a referendum procedure or even with both. The three methods have different functions. The legislative procedure provides a means of approving or rejecting a proposal to issue a proclamation to amend the constitution. A referendum is primarily a means of ascertaining the wishes of the people but it can also be used as a means of directly approving or rejecting approval for the issue of the proclamation. A constitutional convention is a gathering of representatives of the people, either appointed or elected, or a combination of both, whose mandate it is to consider all of the alternatives and put forward a specific proposal or alternative proposals for approval or rejection by either the legislatures or the people directly in a referendum.

A constitutional convention cannot be used for an approval process because it would amount to a delegation of authority to "actually amend" the Constitution without regard to the wishes of the people or the legislatures, and without being answerable for its actions. Its function is to develop proposals or achieve a compromise that can subsequently be submitted either to the legislatures or to the people directly in a referendum for approval. While a referendum can be used to directly approve or reject a proposal, in my view it is unwise to eliminate the existing legislative process and replace it with a referendum process because a referendum is inappropriate to deal with routine or ordinary amendments where there is no real doubt or strong opposition. Instead, the legislative model should remain the basic constitutional amending process with a referendum or constitutional convention being used to supplement the normal legislative process where circumstances are such that the legislative process cannot adequately deal with the magnitude of the proposed changes or the divergence of opinion on the issue.

A referendum might be called where there is great controversy as to whether or not a particular proposal should be accepted. A constitutional convention might be used where major issues might be involved and there is great divergence of opinion as to what proposal should be put to the legislatures or the people for approval, in a referendum.

In the current debate, a constitutional convention would provide the two things essential to achieve a legitimate and enduring compromise that could then be presented to Parliament and the Legislatures for entrenchment or submitted to a referendum. Those two things are: first, a means of exchange of views and adjustment of positions that could lead to development of a compromise between Quebec and the other provinces of Canada and amongst Canadians generally and, second, the legitimacy that can only come from a compromise identified after open public debate of issues, positions and proposals where the debaters will be aware of and feel the pressures of public opinion.

Neither the Spicer Commission nor individual provincial commissions can provide those two essential elements. In fact, the provincial commissions not only will not provide an opportunity for such dialogue, their focus will tend to be on matters that are exclusively or predominantly of concern to the particular province.

Clearly the outstanding constitutional issues in Canada today are of such a magnitude that they cannot be dealt with on a simple amendment basis. As well as all of the major issues involved in responding to Quebec's proposals to address its legitimate concerns and the amending formula issues set out in the discussion paper, there is also the major question of Senate reform. When you add to this the fact that there are strong differences as to the basic concept of the nation as a federal state, and recognize that a level of acrimony does exist, it is difficult to imagine any other process that could be used to resolve such problems and achieve an acceptable compromise.

Our Constitution, at the moment, takes no provision for a constitutional convention but that should not be an impediment. The current amending procedures do not specify what process is to be pursued in arriving at the wording of constitutional amendments to be submitted to the legislatures and Parliament. It could just as well come from a constitutional convention as from a first ministers' conference. Any proposal coming out of the convention would still require the approval of Parliament and either seven or all ten legislatures to be effective. Even if a compromise were achieved, feelings are running so high and opinion so divided, a referendum may still be necessary in order to determine if the proposal would meet with an acceptable level of approval across the country.

One difficulty that might arise is in achieving agreement on the manner in which the convention should be constituted. Because it would have no jurisdiction to enact or implement a recommendation or decision, the voting of its members would only be a factor in determining the final recommendations it would put forward. An argument cold be made for having equal representation for each province but then again an argument could be made to have representation weighted to take into account population. Perhaps a fair solution might be to have one half of the membership divided equally amongst the provinces and the other half based on population. There would remain the question of whether they should be elected or appointed by the provincial governments. Perhaps they should be partly appointed and partly elected. In our present circumstances I would suggest that at least half should be elected in order to enhance credibility and chances of public acceptance of the results of their deliberations. The following is one possible model but I emphasize that any one of a variety of possible models may be equally valid:

The number of members of a national constituent assembly should be not less than 100 and not more than 200. At least half of the members should be elected in order to ensure the credibility of the assembly, and the acceptability to the electorate of its recommendations.

About half of the elected component should be elected in equal numbers from each province. The other half should be elected on the basis of population. Assuming approximately one hundred elected members, this means that each province could elect 5 members for a total of 50. Each territory could elect one member. The distribution of elected members on the basis of population could be as follows: Newfoundland 1, Prince Edward Island 1, Nova Scotia 2, New Brunswick 2, Quebec 13, Ontario 18, Manitoba 3, Saskatchewan 2, Alberta 5, British Columbia 6, Yukon 1, Northwest Territories 1, for a total of 54. Under such a proposal the total elected members would be 106.

The remainder of the members could be appointed by the governments and legislatures. Particular effort should be made to ensure adequate representation of aboriginal Canadians. Each provincial government and the federal government could appoint its first minister or attorney-general or an alternate, together with two other government appointees.

Similarly, each territorial government could appoint its leader or attorney-general or an alternate. This would add thirty-two provincial and territorial appointed members.

Thirty-five other members could be appointed on the basis of population by the provincial legislatures. The distribution could be as follows: Newfoundland 1, Prince Edward Island 1, Nova Scotia 1, New Brunswick 1, Quebec 9, Ontario 12, Manitoba 2, Saskatchewan 1, Alberta 3 and British Columbia 4. Each legislature would determine its own rules for such appointees provided that they should not be members of the provincial cabinet and that they need not be legislators.

Finally, twenty other members should be appointed by the federal government. Again, the federal government would determine its own rules for these appointees and would of course have to satisfy Parliament.

These issues should be able to be resolved. In the event that agreement could not be reached the matter could be resolved by Parliament. Clearly, as matters now stand Parliament would have jurisdiction to establish and set the groundrules for such a convention in any event.

In conclusion, I suggest that it is vitally important to establish a constitutional amending process, for the long term, which is more sensitive to the views and opinions of the people of Canada than our existing procedures.

Canadian Parliamentary Review Cover
Vol 14 no 3

Last Updated: 2020-03-03