A determined executive may often make initially unconsidered assumptions that agents and citizens will accept the policies they put forward or may develop underfunded or badly structured policies that allow agents to undermine the policy author’s intentions. However, several treatises in the 1980s, such as Mazmanian and Sabatier (1983) or Barrett and Fudge (1981), pointed out that policy implementation is a longer-term game in which principals over time react to the difficulties encountered with implementation and amend their ideas accordingly. The idea of evolution within the social sciences is far from a new development and in

economics was certainly recognised by Joseph Schumpeter and further developed by Nelson and Winter (1982). An organisation will evolve by learning from its successes and mistakes, building up expertise and knowledge about what strategies and innovation may work. It is a process involving trial and error that may be built into a government’s machinery for seeking effective policy change. Governments show an increasing awareness of chances of failure as much as success in their policies and create systems to guard against the possibility of rectifying mistakes before they have created too much damage. It is frequently the case that policies are borrowed from experience in other organisations or countries and tried in pilot studies and then subjected to longterm evaluation before becoming part of the accepted policy structures of the state. The argument that public policy is in a continuous state of flux through incremental change in both the form of policies but also how it is interpreted and implemented over time has strong resonance with Lindblom and Braybrooke’s ideas on muddling through. Later studies have further developed such ideas by arguing that the public policy

process and its outputs must effectively be seen not as completed events but, as Peter John (1998) pointed out and the ACF framework shows, abstracted points in an evolving process. It is of value therefore to consider the social forces that determine such constant change and also the possibilities that have been suggested to manage and cope with such change. Peter John (1998: 182-8) suggests that the various elements within the policy process should be seen as the evolution of ideas and action:

Single level approaches focus on one key principle as the foundation for the political system, and they often assume that political systems are stable as a result. What may be called evolutionary theory regards all these elements as continually interacting with each other over time. The institutions are sets of constraints upon actors that can sometimes be reshaped if actors so choose. Interests are important because they structure the sets of choices for other actors. Economic power structures similarly offer alternative choices. Individual self-interest drives the machine

and interacts with the constraints. But what drives policy is the continual interaction between these features. And evolution is the direction in which integrative accounts of public policy are heading.