Canadians should be grateful to the Canadian Alliance for providing them with a summer of blockbuster political intrigue. Every day seemed to bring a new twist in the plot, a new level of personal invective, and a new personality either jumping ship or defending it. At every turn the Alliance has teetered on the brink of becoming a spectacular political flame-out. Here was politics at its most personal and visceral—clashing personalities engaged in a game of chicken, an activity, as one wag put it, usually reserved for young children and great superpowers.

Whatever its entertainment value, the Alliance leadership conflict affords an opportunity to reflect on the evolution of intra-party governance in Canada, and in particular on mechanisms for leadership selection and review. The Alliance is a telling example of a trend evident in all Canadian political parties. More and more parties choose their leaders “democratically,” allowing each member a direct vote. Leadership review—the process by which a leader is formally evaluated and possibly dismissed—is also increasingly the members’ prerogative.

This trend is as paradoxical and disturbing as it is widespread. Democracy in the form of membership-based leadership review and selection procedures may sound good in principle but, as the Alliance crisis indicates, may be disastrous in practice, not only for the parties concerned but for democratic government in Canada more generally.

Leaders can always resign their position voluntarily after a dismal election result or when they judge that their time is up. No party’s constitution prevents this. But party rules have to anticipate the contrary situation, when the leader decides to stay on even though the party’s interests would be advanced by his or her departure. Such a situation is not hard to imagine: Leaders are highly ambitious people, at or near the top of the political food chain, and are surrounded by people who constantly tell them that no obstacle is insurmountable, no challenger unimpeachable. In this circumstance, with whom should the power to remove rest?

Before the 1960s Canadian parliamentary caucuses had the power to yank the leader all on their own. Such power was informal—that is, not stated in party rules—but no less real for it. Only in the 1960s did the party-in-convention acquire the power to review the leadership and in effect pass a vote of non-confidence. More recently, parties at the provincial level, either as a matter of principle (e.g., the Parti Québécois) or as an attempt at party renewal (e.g., the Alberta Progressive Conservatives), began to select their leaders through wide-open direct membership votes, not unlike American primaries. So the trend is clear: over the course of the 20th century, internal party decision-making, especially that pertaining to party leadership, has become increasingly “democratized.” Party notables, chief among them caucus members, have been left with less power.

To argue against these developments is to risk the worst of accusations—that one is undemocratic. As Tocqueville foresaw, democratic societies value equality highly and extend the principle widely and deeply into the institutions of civil society. But even in a democracy, not all institutions are run perfectly democratically. Many important institutions of all kinds have hierarchical governance structures; that is what allows them to operate, make timely decisions, and react to events. And so long as there is a distinction to be made between public and private realms, democracy as a public principle of governance should not automatically intrude into the organization of all the institutions that make up civil society.

Political parties are one such institution. They may be involved in public governance and even receive public financial and legal support; but they are eminently private in the sense that they quite properly enjoy the freedom to adopt whatever internal mode of governance best suits their purposes. There is nothing inherent in the idea of a political party that requires it to conduct its internal affairs in a certain fashion, “democratic” or otherwise.

Quite aside from this theoretical argument on behalf of a party’s freedom to choose its own mode of governance, it is simply unclear what “democratic” means in the context of party governance. As the Canadian Alliance’s trial illustrates, either or both sides can claim democratic legitimacy.

The central conceit of the Reform/Alliance party over the years has been to project the image of “the people” as some unvarnished, organic, authentic collectivity that can act, express itself, and be represented by an elected group. In fact, this concept of “the people” is mainly a convenience for political “outs” seeking a place on the electoral map.

In reality, “the people” is a complex and overlapping variety of constituencies that come together for different purposes. There is: the population of the country as a whole; the subset of eligible voters; the further subset of actual voters; voters divided into federal, provincial and municipal constituencies; voters who vote for the winning constituency candidate in a general election; voters who vote for losing candidates; people who are members (voters and non-voters) of a political party; members of a party who occupy positions in the local party constituency association and/or higher-level party governance structures; members of a political party elected to Parliament; and so on. Which part of “the people” does a leader or his or her challenger represent? And which part is it more “democratic” to represent? There is no obvious answer to this question. “The people” is the pea in the contemporary populist political shell game.

Like “the people,” “democracy” is one of those words George Orwell warned us about. It denotes an idea in whose name the most scurrilous things can be done. So even if we are committed to “democracy” as a principle of party governance, we need to specify exactly what we mean to say—and face the fact that competing concepts of democratic governance may simply be incommensurable: there may be no way to determine which version ultimately is best.

What we can do, however, is judge various versions of democracy by their consequences. We may now be coming to a point in Canadian political history when we have enough experience with different ways of doing things to consider the effect of internal party democracy on overall Canadian political democracy. My central contention is that membership-based party leadership selection processes, apparently designed to make leaders more accountable to their members, actually exacerbate the problem of leader domination in Canadian politics.

This idea is not new or radical. In his 1867: How the Fathers Made a Deal, Christopher Moore suggests that the first ministers who negotiated Confederation in the 1860s were actually under greater control by elected politicians than were the first ministers who in the 1980s concocted the patriation deal. Moore makes the unconventional but convincing argument that the 1867 deal was the product of a more democratic process than the 1982 deal: In 1867 the first ministers brought opposition politicians into the negotiations from the very beginning. The idea may have been to co-opt them, but the effect was to give broad sanction to the agreement that eventually was negotiated—this despite the absence of one-person, one-vote and many other accoutrements of modern democracy.

The inner workings of government
Keep track of who’s doing what to get federal policy made. In The Functionary.
The Functionary
Our newsletter about the public service. Nominated for a Digital Publishing Award.

Looking abroad, Australia and Britain remain quite viable democracies even though the fate of party leaders is still largely in the hands of caucuses. Remember the summary termination of Margaret Thatcher’s leadership of the Tories in 1990—while she was Prime Minister! (On the other hand, even the Tories have given party members some power in leadership selection, allowing them to choose between two candidates selected by the caucus.)

In Canada prior to the 1920s (even later for some parties), party leaders were at the mercy of their caucus colleagues. Ultimately, a party leader knew that his caucus colleagues had put him where he was and could remove him expeditiously. This bound the leader closely to their views. And the caucus, by virtue of its intimate knowledge of the workings of government and daily experience of the gifts and failings of the leader, had the information and the power necessary to pronounce upon party leadership. A bad leader could be dispatched quickly. A good leader knew that should he ever turn bad, his caucus would be on to him soon. In terms of power, the caucus had both information and authority sufficient to aspire to a relation of equality with the leader.

Increasingly, party leaders owe their jobs not to their caucus colleagues but to the party membership. An asymmetry has thus developed. The membership has the formal power to determine leadership, but the caucus has the intimate knowledge of the capacities of the leader. In the normal course of things, the leader rises above his or her caucus, claiming that the membership, not the caucus, is the group to which he or she is accountable (as Stockwell Day has claimed incessantly). Deprived of its constitutional authority within the party, the caucus cannot exert the leverage it once had.

When combined with the development of electronic media and centralized party fundraising and electioneering machinery, this supposed democratization of the party leadership selection and removal processes has in fact helped make Canadian politics highly leader-dominated. As Donald Savoie has argued, political executives in Canada are among the most powerful in the world. The trend toward internal party democracy has only increased this tendency.

By diminishing the effective power of elected caucus members and giving formal power to relatively inactive and unorganized party members, “democratic” leadership selection actually increases the political distance between political leaders and the people. When confronted with an untenable leadership situation, the enfeebled modern caucus has no formal recourse. In the face of a particularly stubborn, obtuse leader, frustrated caucus members must resort to drastic measures whose effect may ultimately be to destroy the party, as the events of summer 2001 have illustrated.

In practice, the party membership’s formal power to remove the leader cannot be exercised in an effective, timely way. And the membership lacks the intimate knowledge of operations in caucus, the legislative chamber and in-party management to pronounce intelligently on the incumbent’s suitability. We will have to await the spate of kiss-and-tell books on the Alliance disaster to confirm this, but for now it seems reasonable to assume that caucus members were aware of leadership problems even more severe than was implied by the torrent of gaffes, contradictions, and attacks on which the media have reported so extensively. After all, the rebel group contains some of the most experienced and intelligent MPs in the Alliance. Surely they know the political costs of party disunity. That they were driven to such extreme measures bespeak problems beyond what the membership at large can know.

From the perspective of the support base of the Reform/Alliance, the fact that greater democracy may end up centralizing power in a single person is particularly paradoxical. These two parties were the product of a tide of populist sentiment in Canada which was itself born of a resentment of elitist, allegedly anti-democratic politics. Putting power “back in the hands of the people,” and providing leadership according to the “common sense of the common people” were the parties’ founding refrains. Even the Alliance rebels, the self-styled Democratic Representative Caucus, continue the theme. On July 19th Deborah Grey referred to her group as dedicated to a “principled, populist” vision of Canada, apparently oblivious to the potential for self-contradiction in the very phrase.

Many Canadians seem quite happy to witness the self-immolation of a party that just 12 months ago appeared formidable enough to scare the Liberals into calling an early election, with all the risks that this entailed. But all Canadians should be concerned about the effects of internal party politics on the wider political order. Good parliamentary government requires effective opposition. Even Liberals benefit from conscientious, vigorous opposition. Moreover, no matter how private and autonomous political party operations may be, they have consequences for political life in general. Leadership requires accountability, and accountability is ultimately a matter of constitutionalism.

If we wish to make party leaders, both those in government and those in opposition, more accountable for their conduct, then we need to think intelligently about how they are to be selected and dismissed. By now in our history, it should be evident that internal party democracy does not necessarily translate into political democracy generally.

The Alliance leadership crisis clearly has more complex origins than leadership selection and removal mechanisms. Right-wing populist parties in North America have attempted to marry right-wing political ideology—which translates these days into market-oriented policies benefiting the business sector and upper middle-income earners—and left-wing party governance themes. The Alliance has a mode of party governance for the whole people but a platform for only part of it. This is a circle the Reform/Alliance has found it difficult to square, and is part of the reason for the Alliance’s difficulties. Like many populist movements, the Reform/Alliance has stumbled in its quest to become a competitive, disciplined party. It is also true that opposition parties are particularly vulnerable to factionalism and in-fighting. Consider the Progressives, the Social Crediters, and the history of the Progressive Conservatives chronicled in George Perlin’s The Tory Syndrome. Opposition party leaders have little patronage to dispense, and opposition status, by definition, means the incumbent leader has not delivered the holy grail to his or her followers.

But underlying these crosscurrents the lesson remains that parties which fail to place leadership removal power where it is most effectively exercised will court disaster. For this the Canadian Alliance will be known for generations.

Photo: Shutterstock

Thomas M.J. Bateman
Thomas M.J. Bateman is a Professor of political science at St. Thomas University. His main research interests include: Canadian politics and government, the Canadian Charter of Rights jurisprudence, federalism, and the politics of religion in liberal democracies. Bateman is also the co-author of The Canadian Regime: An Introduction to Parliamentary Government in Canada. 

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License