I Introduction Citizens have long wanted a reliable measure of police performance. They want to know whether the police are producing something that is valuable with the assets entrusted to them. They want to be able to hold the organization to account for its performance. Police managers, too, have needed a measure of police performance-partly to meet external demands for accountability, and partly to establish a form of accountability inside their organizations that could focus attention on achieving valuable results rather than simply reliably executing established policies and procedures. A crucially important task for both citizens and police managers, then, is to out an understanding of how the produced by policing can be recognized and assessed. Generally speaking, developing appropriate measures of police performance has been treated as a managerial or technical problem to be solved through some combination of the practical methods of business management (develop a bottom line for policing) or the statistical methods of social science (conduct a program evaluation of police departments focusing on the ultimate outcomes produced by the police). And it is true, of course, that developing suitable performance measures do raise important managerial, technical, and scientific issues. The managerial issues focus on how different measurement systems can be used to guide, motivate, or enable the learning of an organization. The technical issues concern the development of the statistical measures and instruments that can reliably capture the dimensions of performance that the police deem important. The scientific inquiry focuses on the extent to which performance measurement systems can help the police find out what particular programs work to accomplish various police objectives. In the end, though, it seems self-evident that the development of police performance measures is also, and perhaps most fundamentally, a normative and political question; not only or even primarily a management or technical issue. After all, to develop a standard for assessing the performance of an organization as or bad, improving or deteriorating, is to make a normative, value claim as well as a positive, scientific claim. One has to have an idea of the or the as it applies to police operations to be able to defend a standard of police performance. One also has to have a theory about whose views of the and the should count in setting a standard for policing--more particularly, whether it is the views of individuals deciding for themselves what they think is and as they encounter the police as individuals seeking assistance or individuals who are stopped, cited, or arrested; or whether the idea of the good and the right in policing emerge from some kind of collective political process in which the body politic gathers itself together to say what is good and right for the whole, and where that view trumps the views of individuals. One may also have to have an idea about the kinds of deliberative processes that should, ideally, go into the formation of a collective judgment about the good and the right as it applies to public police departments. In short, the important question that lies at the core of developing any adequate measure of police performance is for citizens and their elected representatives to decide what it is that is intrinsically valuable, or what it is that we as a political community value in the activities and operations of a public police department. That is simultaneously an issue for political philosophy and practical politics, as well as for professional management and science. Indeed, once one sees the issue as a normative, philosophical, and political question, a great many additional questions arise, including the following: * Who does the valuing of public sector operations? Are police operations properly evaluated by the customers of the police (those who receive services from the police and/or pay the costs of keeping the police operating)? …