Canadian Parliamentary Review

Current Issue
Canadian Region CPA
Upcoming Issue
Editorial and Stylistic Guidelines

HomeContact UsFrançais

Constitutional Conventions and Election Campaigns
Andrew Heard

At the time this article was written Andrew Heard was Assistant Professor of Political Science at Simon Fraser University . On December 8, 1994 he appeared before the Senate Standing Committee on Legal and Constitutional Affairs which was considering Bill C-22 and on September 25, 1995 he appeared on a panel before the Special Committee on the Pearson Airport Agreements

Parliamentary government is based on a number of important but unwritten constitutional conventions often overlooked both in public debate as well as in academic circles. In the October 1993 general election, the Conservative Government was defeated. Shortly thereafter, a question was raised by the new Liberal Government as to whether the previous administration violated any constitutional convention by signing, just three weeks before the election, an agreement to turn over Terminals 1 and 2 of the Lester B. Pearson International Airport to the private sector. This article provides one view about the general nature of constitutional conventions and discusses the kinds of constraint that face a government in its last days. Another perspective on this issue will be published in the winter issue of the Review.

Constitutional conventions are a very important set of informal rules which regulate political behaviour. They are laws which essentially ensure that legal powers work in some fashion other than the way the letter of the law requires or stipulates. They arise principally through political practice but rely essentially on a level of general agreement upon them. This is the fundamental characteristic of constitutional conventions. They are informal rules, but their obligatory character arises from the degree of agreement in the general political community on those rules.

Some conventions are very important. They shape the character of our Constitution. The very basis of responsible government comes from convention. The transfer of the vast legal powers of the Governor General to the elected politicians of the day is achieved by convention and convention only. It is important that we understand that these rules have a fundamental role to play in the political system. We must be careful in what we describe as conventions because they have such an important role to play.

As an example of how important conventions can be, even though the Constitution Act of 1867 stipulates that the Governor General shall send copies of every law passed in Canada to the British government, a constitutional convention now nullifies that obligation. We have an obligation in the supreme law of the Constitution which is nullified by convention. That is an idea of the strength of these political rules.

The Traditional Approach

The traditional method of viewing constitutional conventions is one which is based on historical precedent. This approach was most explicitly formulated by Sir Ivor Jennings, a British constitutional writer. However, his views were adopted by the Supreme Court of Canada in the Patriation Reference Case when they endorsed the three-step test for conventions posed by Jennings. Sir Ivor Jennings said that when we look to see whether a convention exists, there are three questions we must try to answer: "First, what are the precedents; secondly, did the actors in the precedents believe they were bound by a rule; thirdly, is there a reason for the rule?"1

These are three steps which place precedence at the top of the line. Was there a precedent? Did the actors in those precedents believe themselves to be bound by a rule? Is there a reason we can root in constitutional principle that would support this rule as a convention?

Precedents are the absolute foundation of this approach. The late Senator Eugene Forsey embraced Jennings' approach and underlined the necessity of precedent. He wrote: "A constitutional convention without a single precedent to support it is a house without any foundation ... indisputably, at least one precedent is essential. If there is no precedent, there is no convention."2

If there is no precedent, there is no convention. This was Forsey's reformulation of the traditional view of conventions.

Using this approach to look at the matters surrounding the signing of the contracts with the Pearson Airport, we need to search for a relevant precedent. This is the key to establishing whether there was a convention regulating the behaviour.

I have searched widely in both Canadian and British authorities without finding any direct precedents about the issue of signing or not signing contracts during an election campaign. Nor did I did find any cases where discussions about contracts during campaigns gave rise to normative observations about specific constraints on a government of such a nature as to point to a constitutional convention.

The precedents which have been discussed by the authorities relate to limitations on governments after an election. This is a difficulty. The authorities discuss the behaviour of governments after an election, not during the election. They are concerned with occasions when the results of the election make it clear that the governing party has been defeated, and another party will be called upon to form a government. The key to these precedents is the defeat of the cabinet in the election and the impending appointment of a new cabinet.

The following examples review some of these precedents: Modern authorities point to and look with approval at instances where either a Governor General or Lieutenant Governor has refused to make appointments devised by cabinet after it has been defeated in a general election.

The notable federal precedent occurred in 1896 when Lord Aberdeen refused to make some appointments to the bench and to the Senate after Sir Charles Tupper was clearly defeated in a general election of that day.

There are also a number of provincial precedents where the Lieutenant Governor has refused the advice of the cabinet after its defeat in a general election. Twice in Nova Scotia, in 1878 and 1882, the Lieutenant Governor refused to make appointments; in New Brunswick in 1908, and Quebec in 1886 as well.

In all these cases, the Governors believed that cabinet should not try to make to the appointments. In the Aberdeen - Tupper incident, Lord Aberdeen had to make a report to the British government on his actions, and this was supported at the time by the Colonial Secretary.

There is one contrary incident in 1905 when the Ontario government was defeated in the general election and succeeded in having some appointments made. There was apparently a great outcry at the time and the authorities look at those negative reactions as an indication that this behaviour was contrary to a constitutional convention.

In my view, these precedents all reinforce a general principle of a responsible government, this being the reason for the convention. A cabinet has the right to have its advice acted upon, so long as it has not been defeated at the polls or in a clear vote of confidence.

If a cabinet loses the confidence of the House of Commons or the electorate, it must limit itself to conducting routine business. Using the traditional approach to conventions based on precedents, I must conclude that a government defeated at the polls must limit itself to routine affairs.

In summary, the traditional approach relies heavily on historical precedent for establishing constitutional convention, and since I was not able to find precedents about whether a government could sign important contracts during the election campaign, we must conclude, in the absence of precedents, that there is no convention.

To say that there is no convention that limits a government during the campaign period, however, is not to say that there are no constraints whatsoever on a government's behaviour.

I would suggest that the limits on a government's behaviour during an election campaign are ones of prudence, of good politics, not of constitutional convention.

Constitutional Principles as a Guide to Conventions

Historical precedents are useful, but they are not completely determinative in my view. They provide a helpful window to view the public debate over constitutional principles and rules. But should we be bound by precedents that come from the 19th century, when so many of our values have changed since then? What happens if there is a precedent and most people would argue it was a case of someone getting away with it? Do we say that the action created a new convention or it illustrated an instance of someone acting against the rule?

If we look back through history and find no evidence of the action, does that mean that no one thought of doing it, or that they were in fact following a rule? The particular problem comes in suggested conventions which relate to refraining from an action. The rule imposes an obligation not to do something. It is too unclear and too unsatisfactory in my view. I do not believe political actors should be free to do as they wish — and declare, "Well there is not convention because there is no precedent, so we can do as we please."

Since I am unsatisfied with the traditional approach which relies so heavily on historical precedent, I have developed another approach which is based more on constitutional principles than historical precedent.

The distinguishing feature of conventions, and one that separates them from mere habits and usages, is that conventions support important principles of the Constitution. The very nature and importance of conventions is that they are informal rules brought about to realize a generally agreed upon principle of the constitution. To refer back to Jennings, the reason for the rule is the principle which is involved.

In the absence of precedents, we can and do in practice turn to basic constitutional principles which bind our political actors.

In looking at the signing of the Pearson Airport contracts during an election campaign, I would want to examine what constitutional principles may be involved. The most fundamental group of principles directly related are those which support responsible government.

The legal authority of the Governor General is exercised in practice on the binding advice of a cabinet who can command a majority of seats in the House of Commons. This is the essence of responsible government. The cabinet has the authority to advise the Governor General because it is able to command a majority in the House of Commons.

If the cabinet loses the confidence of the House of Commons, it must either resign or advise that an election must be held. Until it wins back the confidence of the electorate, that cabinet can only carry out limited decisions. This limitation arises because it lost the confidence of the House of Commons. If a cabinet is defeated on a clear vote of confidence in the House of Commons, it is limited in what it can do until it wins back the confidence in a general election.

In another situation, if a cabinet clearly loses a general election, it, too, has lost its authority to advise the Governor General, and it must offer its resignation and limit itself until a successor is appointed.

In the case of the Pearson Airport contracts there was a cabinet formed by a new Prime Minister, Kim Campbell who had not met the House since her appointment. However, since she led a party with a majority in the House, there is no doubt in anyone's mind that she would have won any vote of confidence she faced.

Prime Ministers and premiers are frequently appointed to office and replaced between elections by party leadership conventions. When they become premier or prime minister, they simply carry on. This is the situation we faced in 1993 when Ms Campbell won the leadership convention of the Conservative Party and replaced Brian Mulroney as Prime Minister. I have not heard any discussion on past leadership changes that suggested a new premier or new prime minister is somehow limited until he or she meets the legislature or wins a vote of confidence.

There is one instance in which a related concern was raised, and that came with the spate of appointments which accompanied John Turner's succession to office in 1984. There was considerable debate, some of it heated, at the time, because of a number of appointments which Pierre Trudeau made prior to resigning, and because John Turner continued to make some appointments thereafter. The argument raised at the time was that the government was in its so-called dying days and should constrain itself.

Such an argument, I believe, was unfounded since that Parliament had a year to go before an election had to be called. Many people did not like that large grouping of patronage appointments, but that is a question of public policy and not constitutional convention.

I would argue that constitutional principles support the right of the government to advise the Governor General on any matter it wishes until such time as it loses the confidence of the House or it loses a general election. If I am mistaken in that conclusion and there are some limitations upon a government during an election period, we should consider what sort of limitations there might be.

For me, the clue lies in comments made by Peter Hogg in reviewing the Aberdeen - Tupper incident of 1896. He wrote: "The Governor General had a discretion to refuse to concur in an important and irrevocable decision which could await the early and inevitable formation of a new government."3

The key phrase here is "important and irrevocable", and I believe it conveys a very sensible distinction. What a government is restrained from doing after it is defeated in an election is making important, irrevocable decisions which cannot be undone by the new government.

Thus if there is a limitation on a government during a campaign period, it could not be anything more than that limitation. Defeated governments cannot make irrevocable decisions. They may make decisions which a new government can undo. If one concedes, for the sake of argument, that a cabinet must avoid implementing important decisions during a campaign period, then I would suggest that the sorts of decisions they are prohibited from making are ones that are important and irrevocable. However, the Pearson contracts are not irrevocable. They may be undone. We can debate the means chosen to undo those contracts, but the successor government and Parliament clearly have the power to undo them.


1. Sir Ivor Jennings, The Law & the Constitution, 5th ed., (London: University of London Press, 1959), p. 136.

2. Eugene Forsey, "The Courts & the Conventions of the Constitution," (1984) 33 UNB Law Journal, p. 34.

3. Peter W. Hogg, Constitutional Law of Canada, 3rd ed., (Toronto: Carswell, 1993), p. 253.

Canadian Parliamentary Review Cover
Vol 18 no 3

Last Updated: 2020-03-03