Politics, Values, And Public Policy

The Problem Of Methodology

Frank Fischer

• Publisher: Routledge
• Year: 2019
• Topic: Politics & International Relations
• eBook ISBN:
  9781000307627
• Subtopic: Politics
• Index: Politics & International Relations

Addressed to the growing concerns about norms and values in policy assessment, this study develops a methodology for the political evaluation of public policy. It is designed to move policy evaluation beyond its current emphasis on efficient achievement of goals, focusing instead on the assessment of the acceptability of the goals themselves, emplo

1
Introduction: The Problem of Methodology

The main evidence that a methodology is worn out comes when progress within it no longer deals with main issues.

—Alfred N. Whitehead¹

In recent years social scientists have emphasized the study of public policy and the development of a methodology for policy analysis.² Policy analysis has grown up under the influence of the positivistic methodology of the behavioral sciences and constitutes a collection of approaches that rely on the scientific method and its techniques: cost-benefit analysis, survey research, mathematical simulation models, experimental design, input-output studies, multiple regression, and systems analysis. By adapting highly developed scientific techniques to policy studies, the field of policy analysis has established itself as a rigorous, applied social science in a relatively short period of time.³ Over the developmental period, however, the methodology of behavioral science has itself come under criticism. There has been growing dissatisfaction in the social and political sciences over the use of scientific methods in the study of normative political and social phenomena. From the outset, some theorists complained about the application of positivistic methodology in the social sciences, and, more recently, many mainstream writers as well have begun to question some of behavioral science’s basic methodological principles, particularly its separation of facts and values.⁴ More recently, these questions have filtered into the policy analysis literature, where the normative dimension of political behavior is frequently a crucial issue.

The fact-value separation is based on positivism’s adherence to the metaethical theory known as “value noncognitivism.” According to this theory, value judgments are essentially emotional responses to life conditions. As subjective commitments, they contain no verifiable truth content. To qualify as objective knowledge, statements must be verifiable by formal scientific methods. Although important aspects of value statements can be investigated scientifically, particularly statements about the conditions that lead to adoption of specific values or statements about consequences that result from the acceptance of value positions, there is no way of scientifically establishing the categorical truth of a value judgment. Thus, in the final analysis, value judgments must fall beyond the reach of rational methods that are defined as the formal rules and procedures of science. Even though in practice there are various degrees of adherence to the “fact-value dichotomy,” at the methodological level the debate remains a dominant governing principle in the behavioral sciences. To be judged as proper, all research must—at least officially—pay its respects to the principle.

The practice of public policy analysis often appears to be far removed from such basic epistemological questions. In reality, however, the problem is nowhere more urgently important. Public policies are essentially political agreements designed for the practical world of social action where facts and values are inextricably interwoven. Policy science, as an applied discipline designed to empirically analyze value-laden policy issues, thus uniquely straddles the fact-value problem. Given the value-laden character of public policy, it is difficult methodologically in policy science to separate facts and values without distorting the basic purposes. Although many policy writers have tried to avoid conflict with the fact-value separation (sometimes by merely ignoring it) an increasing number believe that major difficulties encountered by the development of policy science are inherently linked to this underlying principle governing the methodological treatment of values.

There have been attempts to deal with the normative dimensions of public policy; in fact, the number of essays on the subject is beginning to grow steadily. Within policy science, however, few writers have been willing to completely divorce themselves from the grip of positivism and the fact-value dichotomy. In general, this bias has restricted attempts to deal with values in policy analysis to an overly simplistic concept of value clarification. While there can be no disagreement about the importance of clarifying value orientations, the positivistic approach leads to misleading assumptions about the nature of values and the role of normative discourse. Under positivism and its modern-day variants, values tend to emerge as external to or detached from specific social or material conditions rather than as inherently tied to the particular life understandings that gave rise to them. Normative relationships are generally treated as formal, static, noncomplex relationships rather than changeable, multidimensional, dialectical processes. Normative analysis is limited to the methods of formal logic and verification, ruling out the informal discursive processes that mediate the construction of social reality.

In large part, positivism’s insistence that value judgments be submitted to the formal methods of scientific verification is the source of the problem. Preoccupation with establishing the theoretical and scientific validity of value noncognitivism has led positivists to either reject or ignore the study of normative discourse in everyday life. Since fundamental values, in this view, are ultimately arbitrary and irrational, everyday normative discourse must necessarily rest on irrational foundations. Such discourse, therefore, cannot have truth content in the formal sense of the term. As subjectively based discourse grounded in a particular social context, normative knowledge must be relegated to an epistemological limbo—if one accepts noncognitivist premises. However, even if it proves true that fundamental values cannot be established by the formal methods of science, noncognitivism fosters a distorted understanding of the basic nature and process of normative discourse. In actual substantive terms, it does not follow that normative discourse proceeds without recourse to rational methods. This viewpoint does not deny that normative deliberation ultimately rests on fundamental values or argue that these values can in fact be proven scientifically; rather, it recognizes the complexity of normative discourse. Not all levels of normative discourse are designed to fulfill the same purposes, nor are they governed by the same epistemological requirements.

Some contemporary schools of thought in epistemology and ethics, particularly variants of ordinary-language philosophy, indicate several aspects of normative discourse that facilitate rational discussion of practical value judgments. First, actors in everyday life seldom confront the lofty intellectual task of justifying fundamental values. In fact, most deliberation about practical affairs is conducted below the level of fundamental values and carried out within a general social consensus about high-level values. Second, within this framework, the task of the policymaker is to choose between competing policy alternatives designed to facilitate the achievement of accepted ideals. The process by which the policymaker arrives at these choices will not conform to the canons of scientific rationality but is nonetheless an identifiable process with rational elements. In fact, it is the possibility of such normative decision making that provides the basis for the construction of society. Positivists have overlooked the rational elements that establish the framework for such deliberation; as a result, they have failed to adequately explore the normative inferential methods employed in policy augmentation. From the perspective of policy analysis, it is possible to argue that positivism has thrown the proverbial baby out with the bath.

This has placed the social and policy sciences in an unfortunate all-or-nothing dilemma with respect to normative inquiry. While noncognitivism may confront important problems that arise in ethical inquiry, misinterpretation or overextension of the fact-value separation has impeded the development of useful methods for improving rational judgment in practical deliberation. This study does not purport to resolve the meta-ethical disputes that have long surrounded value noncognitivism, but it does attempt to present the problem in a new and different light.

The development of an alternative political methodology for the evaluation of policy decisions requires an exploration of the wealth of material that in recent years has begun to appear in the philosophy of social science and related fields. Much of this work has developed as part of an assault on the traditional, positivistic conception of the physical or natural sciences. By the 1950s and 1960s, a significant group of physical scientists had begun to question the adequacy of positivistic explanations of scientific method. Since then, growing numbers of social scientists and philosophers have begun to trace the implications of this methodological rethinking for a social science constructed on traditional scientific methodology. Particularly important in the philosophy of the social sciences have been the contributions of phenomenological sociology, ordinary-language analysis, and the revitalization of the speculative methods of political philosophy. My purpose is to show that a synthesis of elements drawn from these disciplines can supply an alternative foundation for a political methodology for policy evaluation by organizing the competing metanormative approaches into a unified methodological system. Compared to the positivistic approach, the policy methodology derived from this logical foundation is better suited to the policy questions analysts confront and more consistent with the procedures analysts already employ in practice.

The Policy Literature

It is not difficult to locate the problem of methodology in policy literature. Generally, analysts agree that the purpose of public policy analysis is to intelligently consider the decisionmaking problems of governmental organizations. The decisions of public organizations, like all organizations, can be divided into two fundamental categories: instrumental and integrative.⁵ The instrumental decision is concerned with efficient achievement of organizational goals. It relates primarily to empirical questions about provision and delivery of services and, as such, is the principal concern of economics. The integrative decision is more the subject of political science and sociology and is aimed at the normative political problem of “conflict maintenance.” The objective of the integrative decision is to hold the organization together so that it can solve its instrumental problems. Although the two types of decisions are separated for analytical purposes, in the real world they are usually closely interwoven.⁶

Effective policy decisions result from considering both of these problems in the deliberative process. Even though governmental decisions are ultimately decided by political bargaining between participants, policymakers can turn to policy analysts for analytical assistance in both instrumental and integrative phases of the decision processes. In the analytical language of decision making, March and Simon have labeled these two methodological phases “problem solving” and “persuasion.”⁷ Problem solving is directed at the instrumental decision. In this phase, participants assume they share relevant criteria and goals and attempt to determine a course of action through collective investigation of the problem. This involves gathering empirical data and proposing and assessing alternatives until all concerned feel content with the scope of the inquiry. If the participants discover that they do not share the same criteria and goals, decision making shifts to the second analytical phase, persuasion. In this phase, the task is to examine and test the discordant beliefs and values producing the discrepant criteria and goals. Emphasizing the beliefs and values of the participants instead of the empirical aspects of the problem, the policy analyst proceeds by searching for more general criteria shared by the disputants that enhance the possibility of reconciling conflicting views. To be useful to the decision maker, the policy analyst should be able to provide information in both of these analytical phases.

Although the policy literature acknowledges these two analytical phases of decision making, emphasis on the fact-value separation in the behavioral sciences has channeled most of the attention and energy of policy writers into the instrumental problem-solving aspect of evaluation. Although scientific policy techniques lend themselves to the problem of instrumental efficiency, they are not suited to dealing with the subjective problems that characterize political persuasion. Questions of which problem or action is relevant or which standard or criteria ought to be employed in a decision cannot be answered by scientific analysis. The implications of this methodological impasse are reflected in the policy literature. Not only is there a dearth of studies about the normative political aspects of public policy, but, in addition, the political dimension itself is at times denigrated. All too often in the policy literature, politics has been described in negative terms such as “pressures and expedient adjustments,” or “haphazard acts—unresponsive to a planned analysis of the needs of efficient decision design.”⁸ These characterizations clearly capture an attitude reflecting the primacy of scientific policy methods over substantive political problems: If politics doesn’t fit into the methodological scheme, then politics is the problem.

Failure to deal with the normative political dimensions of policy analysis is also reflected in governmental affairs. The limitations of a policy science dominated by efficiency were demonstrated dramatically in the late 1960s and early 197Os: In response to the domestic crises of the period, the federal government and some states initiated a number of reform measures that emphasized policy evaluation. Spurred by the federal adoption of program budgeting and a vast amount of Great Society legislation carrying policy evaluation requirements, policy analysis developed into a small industry, both inside and outside the universities. However, early enthusiasm about the utilization of policy evaluation gave way to much pessimism in a surprisingly short period of time. James Schlesinger, in front of a congressional subcommittee examining the uses of policy analysis in defense budgeting, conceded that everyone is, in principle, in favor of policy evaluation but few are hopeful that its conclusions will be utilized in real-world policymaking processes.⁹ Much of the political failure encountered by policy evaluation can be attributed to its narrow emphasis on the technical evaluation of means. The crises of the late 1960s were more than a problem of inefficient programs; first and foremost, they represented basic conflicts in social values.

Increasingly it has become apparent that a significant part of the problem confronting policy evaluation is attributable to the failure of policy methodologists to directly confront the task of political evaluation. Writers like Alice Rivlin have expressed doubt as to whether giving the highest priority in policy analysis to instrumental techniques such as cost-benefit is the most effective approach:

Politicians and decision makers are unlikely to pay attention to them. They and their constituents have strong, intuitive ideas about the relative importance of health, education, and social well-being, that are not likely to be shaken by cost-benefit estimates. The ratios are unlikely to sway the choice of a congressman between a reading program and a cancer program…. Both cancer research and literary pr...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Contents
• List of Tables
• Preface
• 1. Introduction: The Problem of Methodology
• 2. The Fact-Value Debate and the Search for Methodology
• 3. Political Evaluation and the Problem of Rational Judgment
• 4. The Logic of Evaluation: Evaluation Research and Phenomenology
• 5. The Logic of Evaluation: The Systems Perspective and Political Philosophy
• 6. The Logic as Policy Methodology
• Appendix
• Index