British Journal of Politics and International Relations, Vol. 2, No. 1, April 2000, pp. 89 94 The uses and abuses of evolutionary theory in political science: a reply to Allan McConnell and Keith Dowding PETER JOHN Public policy analysts need to be careful when they assess evolutionary mechanisms in dramatic events in public policy. The poll tax episode was special because of the explicit selection mechanism it invoked. In particular, the incisive intervention of public opinion during the policy s implementation was a unique event in British politics. As the poll tax did not represent a normal period of policy-making, it illustrates special, rather than general, features of evolution in politics. In my article, I acknowledged the need for a long period of analysis by locating the introduction and abolition of the tax in the evolution of policies from the mid-1960s onwards (John, 1999). However, as I did not have much space, this reply gives me the opportunity to set out how evolution operates in the longer time-frame and to point out the relationship between gradual policy change and dramatic and politicised events, such as the poll tax. I first deal with the specific points McConnell raises. He argues that I largely ignore wider institutional and economic structures because of my ideational account of policy change (McConnell 2000). This view misreads my work. In the article, and also in my book (John 1998), I criticise approaches in public policy that solely depend on ideas or interests. However, I fundamentally disagree with McConnell s economic determinism. Whilst acknowledging the power of economic interests, I dispute the assumption Political Studies Association 2000. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF and 350 Main Street, Malden, MA 02148, USA 89
Peter John that they win systematically; and this is McConnell s position. He believes that when powerful economic interests say they are in favour of a measure there is sufficient evidence that a policy is intended to buttress the hegemony of powerful economic interests (Connell 2000, 85). But McConnell needs to do much more than cite the public statements of businesses at the time of the poll tax s formulation. McConnell has to prove, rather than assume, that business influenced policy because business did not participate directly in the decision to introduce the tax. The study of anticipated reactions would be a necessary part of this empirical work. Moreover, as Marx and many Marxists contend, the capitalist class does not always act in its own interest. The preference of business for the tax may not even show that the policy was in the interests of the capitalist class. To show this, McConnell would need to find that the tax impinged on the fate of private enterprises and significantly affected wage costs and the regulation of employment. McConnell argues that I do not recognise the constraints on the decisionmaking process. But I argue that the process was limited by the character of the options available rather than economic interests. What happened in 1984 85 was that the menu of possible options altered, making the poll tax more attractive as a replacement for the rating system. The argument that the power of capital ruled out certain local tax options needs to be proved rather than assumed. McConnell needs to explain why local income tax exists in other advanced capitalist economies in Europe and that this difference results from capitalist interests rather than from ideas and institutions. As many commentators have discussed (e.g. Travers 1986), the reason why Britain never adopted a buoyant local tax was because of the low constitutional status of its local government. In any case, the failure of the poll tax is a neat demonstration of the non-marxist point of view for, once the tax became unpopular, the private sector chimed in with everyone else to propose its abolition, especially as some businesses were hostile to the National Non-Domestic Rate (Butler et al. 1994, 160). Had the measure been essential for business, there would have been an attempt to manage the agenda and to maintain the policy. Like much of Conservative opinion, businesses approved of the principles behind the tax; but once the policy did not work, they had no difficulty abandoning it. McConnell doubts that advocacy played a role in the poll tax decision and he cites the work of Crick and Van Klaveren (1991). However, one of the central features of advocacy is the ability to persuade other decisionmakers that the preferred option is the only one on offer. The study team needed to win over the sceptics within the government machine, who had successfully opposed the tax in 1981, and to convince senior policy-makers, 90 Political Studies Association 2000.
Uses and abuses of evolutionary theory such as the Prime Minister. Butler, Adonis and Travers (1994, 57 61, 65 69) provide an illuminating account of how the small group of civil servants and junior ministers persuaded others in government that the measure would work. The notion that public decision-makers reluctantly groped toward this option does not capture the activism of those who formulated the proposal. Finally, McConnell does not provide evidence for the power of Parliament in his attack. He merely shows how the Government was able to push its measures through the legislature and what he writes does not affect the conventional wisdom of British politics that the competition of ideas occurs in and around the core executive. In summary, I do not find McConnell s socioeconomic explanation compelling. The phenomena he cites to support his case can be subjected to many interpretations, not just those based on the interests of capital. Dowding takes issue with my object of selection because policy-makers intended to introduce the poll tax. The argument is that prominent public policies, such as the poll tax, cannot be the outcome of an evolutionary process because they emerge from public debate and express the objectives of public decision-makers. Defined in this way, conventional models of the public policy process, perhaps with the addition of ideas, adequately explain the choices that politicians and civil servants make. Dowding s model implies that the poll tax arose out of the deliberation by civil servants and ministers, where decision-makers ranked the possible courses of action and chose those they thought best met the objectives of the party in power. However, the literature on policy formulation shows there are almost no public policies that emerge in this fashion (John 1998, ch. 2). Such a stereotype of public decision-making belongs only in the textbooks. While policy-makers think they are in charge, they often articulate ideas and courses of action that are embedded in the institutions and networks of relationships they occupy. There are coalitions, maybe lasting for 20 or 30 years, making the influence of ideas on policy outcomes slow and incremental (Sabatier and Jenkins-Smith 1993). Complex relationships and exchanges exist between the agencies and groups at the levels of government. Such gradual and unreflective change is consistent with the evolutionary model that Dowding outlines, whereby forms of practice transmit themselves over time and equilibria emerge. For example, the period of policymaking in local government finance in the years since 1945 resembles such a pattern of transmission. Policy debates and decisions occurred in a stable framework of institutions, which ensured that the rates and central grants remained as the key features of the system of local government finance. The norms and practices of policy-makers maintained themselves during Political Studies Association 2000. 91
Peter John the period of the expansion of the central finance for local government since the 1950s. Even when local government finance became more politicised in the late 1960s and 1970s, much of its administration continued to be uncontroversial. There is no reason to believe that a long period of time however defined is a necessary condition of an evolutionary process both in the natural or in the political worlds. What occurs in evolution are periods of both stability and rapid transition. In public policy studies, Baumgartner and Jones show, through an examination of evidence from public finances and the adoption of public policies, that times of rapid policy change punctuate long periods when decision-making is in equilibrium (Baumgartner and Jones 1993). Such punctuations are non-linear because of the tendency for issues to gather momentum and because of the serial processing capacity of decision-makers. These are the times when major changes in public policy occur and set the patterns for the subsequent and much more settled periods of policy-making. The question for the evolutionary theorist is whether the shift from routine and institutionalised patterns of behaviour marks a break from evolutionary mechanisms. It may be the case that this is the moment when conventional politics takes over, where argument and deliberation really matter. The evolutionary take on this is that these arguments and ideas have their origin in the previous period, but the selection process and competition become far more intense in periods of policy choice. The claim is that, although politicians think they exercise control, they face limits to what they can do; they drift with events or try to guide one particular stream. In certain circumstances their actions may be crucial; but they do not define policy choices. As a disruption of a period of equilibrium, the poll tax was an extreme case of punctuation. The intensity of debate and concern made even local government finance officers public figures. Detailed administrative decisions became, for a time, national news. Normal forms of selection and competition became suspended. Yet the process of trial-and-error selection occurred none the less in a different form, even if magnified by the public glare. Once this period ended, it was no surprise that old ideas about local taxation re-emerged at the end of 1990. A full account of the evolution of local government finance would trace such a history and the poll tax would be just one chapter in it. This account would elaborate the claims of my article (which does not have the poll tax in its title). Dowding does not offer a reason to distinguish between discussing evolution in prominent public policies from elsewhere in social and political life. Both policies and other forms of social action have elements that are 92 Political Studies Association 2000.
Uses and abuses of evolutionary theory intentional and other aspects that are not. If policy-making fails to be evolutionary because of the intentions of decision-makers, so does the whole of social action. In this case the application of evolution in the social sciences would be restricted to the findings of sociobiology. Quite rightly, Dowding does not rule out evolutionary explanation in public policymaking, so he must show how controversial decisions differ on grounds other than intentions. In fact, both prominent public policies and those that emerge over generations have much in common and link together over time. Dowding complains that I confuse the familiar processes of politics with evolutionary mechanisms. However, an evolutionary explanation is not going to overturn a century of findings in political science. Evolutionary theory adds explanatory power by its focus on the replicator in this case, policy ideas. Researchers seek to find out how replication shapes the more familiar processes of politics. As such, analysts can examine a set of events, such as changes in public opinion, in a different manner than in standard accounts of politics, such as the tendency for change over time to be nonlinear and for there to be periods of equilibrium that lock certain units of selection into place. In a similar fashion, evolutionary explanation does not upset the theory and tests of rational-choice theory, in particular evolutionary game theory, but it places strategic choice in the context of competing ideas about the best courses of action in politics. Evolutionary theory produces testable models of public decision-making. By conceiving of public decision-making as dynamic, interactive and competitive, with replication as one of its central features, analysts are able to model and test different categories of political behaviour. Some of these models may reflect the gradualism of much of public life, such as the diffusion of innovations or the co-operative and competitive behaviour amongst local governments. Other models would examine case studies of the politics of organisations taken over a long time period. A further set of studies would be on the specific evolutionary mechanisms that take place in periods of dramatic policy change, such as the poll tax episode. Bibliography Baumgartner, F. and Jones, B. (1993) Agendas and Instability in American Politics (Chicago: Chicago University Press, 1993). Butler, D., Adonis, A. and Travers, T. (1994) Failure in British Government (Oxford: Oxford University Press). Crick, M. and Van Klaveren, A. (1991) Mrs Thatcher s greatest blunder, Contemporary Record, 5:3, 397 416. Political Studies Association 2000. 93
Peter John Dowding, K. (2000) How not to use evolutionary theory in politics: a critique of Peter John, British Journal of Politics and International Relations, 2:1. John, P. (1998) Analysing Public Policy (London: Cassell). John, P. (1999), Ideas and interests; agendas and implementation: an evolutionary explanation of policy change in British local government finance, British Journal of Politics and International Relations, 1:1, 39 62. McConnell, A. (2000) Local taxation, policy formulation and policy change: a reply to Peter John, British Journal of Politics and International Relations, 2:1. Sabatier, P. and Jenkins-Smith, H. (1993) Policy Change and Learning: An Advocacy Coalition Approach (Boulder, CO: Westview). Travers, T. (1986) The Politics of Local Government Finance (London: Allen and Unwin). Dr Peter John Department of Politics and Sociology Birkbeck College University of London Malet Street London WC1 7HX (Submitted June 1999, accepted for publication July 1999) 94 Political Studies Association 2000.