This small book packs a considerable theoretical and practical punch. Alan Ware challenges much received wisdom about the dynamics of two party politics. In the process, he adds considerably to contemporary discussion of the intersection of structure and agency in the development and adaptation of political systems. Ware picks out two party systems for concentrated attention because of their relative tractability – in his words: ‘these systems are ideal for analysing the capacity of parties to pursue their interests in the face of both other actors within the political system and also of elements within the party itself’.
A threshold point concerns the presentation of this study: it is both highly condensed and presumes ready familiarity with a wide range of political systems. This monograph could as easily be a book of several hundred pages. This reviewer at least had to work to come to grips with the nuances of the argument, but that had its own rewards.
Ware first places his analysis in the context of extant explanations of regime dynamics. Social class (and its analogues) was early offered as the primary explanatory mechanism. Citizen political allegiances and orientations were ‘determined’ by their class identity. Hence, party approaches would be a reflex of these underlying conditions. More recently, median voter theory suggested that electoral incentives would dictate the disposition of parties. Again, the dispositions of voters were determinative. For its part, historical institutionalism suggested that present dynamics were reflexes of a past settlement: here the inertial power of structure served to replicate an original pattern − it has no present dynamic role. Finally, heresthetics pointed to the scope for individual initiative by party leaders quite independently of systemic features: in other words, it suggests agency can be exercised virtually without qualification.
Ware's account challenges or substantially qualifies all these propositions. His story is of the distinctive and independent potency of a two party structure in day-to-day dynamics and the way this configuration qualifies the exercise of agency.
Ware sets the scene with a striking empirical observation: in two party systems, major established parties rarely lose their dominant role. In the eleven two party systems in his sample, there are indeed only three instances of major party collapse. These involved the Progressive Conservatives who were reduced to a parliamentary rump in the 1993 Canadian election and thereafter imploded; the Liberal Party in Britain, which progressively lost its major party role after 1916 and which nevertheless continued as a minor party; and the Whig party in the United States in the 1850s, which also imploded.
In analysing major party collapse, Ware first identifies the three circumstances in which this possibility arises. The first − surge and collapse – would occur when a new party rises to prominence on an immediate electoral surge. What systemic factors inhibit the durable success of the upstart? Next, if a new party does manage to seize the ground from an established major party, when does the latter dissolve and when does it survive as a minor party? Finally, in the face of a third party rival, a major party can seek to expand its support by fusion. Under what conditions does this occur?
In analysing these dynamics, Ware shows that established parties are generally sheltered by at least two structural factors: first, what he terms their ‘focal arbiter’ role; and, second, the resources available to the party in preserving the loyalty of activists and supporters. The ‘focal arbiter’ concept refers to the capacity of an established dominant party to determine the range of alternatives that legitimately figure in public debate. Moreover, in holding the loyalty of its primary supporters, the resources available to a major party are also a prime asset. It can better satisfy their career or other ambitions. In other words, a party's presence as a major party in a two-party system is its primary defence. But electoral upheavals do occur. What determines whether a major party survives this event?
This discussion is introduced with an overture – a survey of the experience of parties in Canada's provinces. Here there are 13 cases of major party collapse compared to one at the national level. These all followed a significant electoral turnaround. However in 17 other cases of equivalent electoral implosions (vote decline by at least 16%), the losing party survived as a major party. In explaining these differences, Ware identifies four capacities that can consolidate a major parties focal arbiter role. These are: the existence of a career path linking various elements of a party through to national politics; the partisan loyalty of voters; the resources required for contesting an election; and the patronage available to party leaders. He concludes: ‘An earthquake type election is no more than an enabling condition.’
But could durable electoral realignment still be the decisive factor? Turning to the three cases of major party collapse in national systems, Ware shows why this standard explanation is inadequate. At an abstract level, a number of qualifying observations are pertinent. First, realignment assumes the pertinent group of voters is unified. In practice, the unity of particular social interests cannot be taken for granted. This is particularly true in the case of the support base for major parties. And finally realignment is not a one-off event. Party support needs to be consolidated over several elections, which creates serial not one-off tests.
US experience further affirms these propositions. Major party collapse in the US accompanied only one of its three major instances of realignment. Realignments occurred in the 1850s, 1930s, and (marked by the slow shift of white southerners to the Republicans) from roughly 1960 to 1985. But on only one occasion did a major party collapse – the Whigs in 1854.
So what does precipitate major party collapse? Ware's answer: ‘when party elites face more than one source of conflict, which could take the form of more than one party seeking to mobilise part of its support or more than one line of division within the party, with these divisions cross-cutting, rather than complementing each other’. In the Canadian PC case, the party's national base included Québec separatists and Western states economic liberals. As new parties emerged (Reform and the bloc Quebecois, the PC became unsustainable. In the case of the Whigs in the US, divisions over slavery and divisions over the rights of immigrants played the same roles. And in the Whigs case, a third factor came into play. The incidence of elections in the US required the Whigs to frame local responses immediately after their national defeat in the presidential election of 1853. Candidates in different states faced markedly different pressures. Meantime, the Democrats survived the rupture amongst their own supporters because their control of the presidency created resources which provided immediate rewards for supporters.
In a brief coda, Ware explores another proposition, namely do more footloose voters (and hence weaker party ties) lead to more frequent party collapse − or at least complicate major party efforts to manage their environments? Such tendencies are now evident throughout the liberal democratic world. For evidence, he returns to Canadian provincial experience. Here separation between national and provincial parties is symptomatic of the underlying weak links between voters and parties. Yet even in this context major parties have been able to recover from sudden reverses in their electoral fortunes and resume their dominant roles. In the face of more footloose voters, Ware envisages a turn to more ‘impure’ forms of two-partism, but not the transcendence of two party structures.
Ware then turns to the issue of party fusions. How does structure influence the behaviour of parties when approaching either temporary or more permanent merger. This discussion is introduced with a thought experiment. He posits a three party system with Conservative and Liberal major parties and a Green minor party. What factors might condition the approach of any of these protagonists to a pre-election pact? Ware again points to the many strategic and tactical complexities and uncertainties and concludes: ‘It is here that the significance of leadership becomes evident . . . leaders can stretch constraints.’
This leads into an empirical discussion of how major parties have formed electoral coalitions with other parties. One pattern involves temporary coalitions. These were common in the US after the Civil War. But the conditions under which these might arise are highly qualified. Another context might be where one of the parties is so weak that an enduring alliance is essential for its survival. The case of the Liberal Unionists in Britain in the late nineteenth century is illustrative. Again, the weaker group possesses minimal leverage. A third pattern is permanent party coalitions. The Liberal-National coalition in Australia illustrates this. But it is sustained by the AV voting system. Finally, parties might bargain with each other to create a larger merged unit. Ware discusses in detail cases in Canada, Costa Rica and Minnesota to and concludes: ‘All three cases expose the point that the distribution of political benefits matters, and that distributional factors – who is going to get what out of a merger – complicate any drive to expand the base of a party.’ In other words, coalitions with other parties are hard to achieve.
In sum, in two party systems, structure is a powerful dynamic influence - the scope for agency is qualified and circumscribed.
Could there be any qualification of this finding? One possibility might be to doubt the basic categorisation. Is the description ‘two party’ system itself sufficiently nuanced? For example, thinking of the evolution of many western systems, other analyses use the terms mass, catch-all and cartel to describe succeeding moments. In conjunction with the cartel form, the literature on citizen disaffection has mushroomed. In this context, is the outlook for third parties so negative? We might also doubt if the US system can be similarly analysed. As Richard Katz once observed, is it two parties or three hundred? Moreover, the legal structure plays a role in the preservation of the US system that is absent from the others. Then there is the Japanese system: how, in more nuanced terms, might it be characterised? None of the foregoing removes the inertial power of the systemic forces that Ware identifies – but perhaps there is more scope for adaptation than his study admits.
