publisher colophon
John S. Dryzek - A Post-Positivist Policy-Analytic Travelogue - The Good Society 11:1 The Good Society 11.1 (2002) 32-36

A Post-Positivist Policy-Analytic Travelogue

John S. Dryzek

Policy analysis involves creating, compiling, and applying evidence, argument, and interpretation in scrutinizing, evaluating, and improving the process and content of public policy. In terms of its relation to the good society, one can think of application in instrumental terms, as identifying the best means to ends that are in turn consistent with improving society by ameliorating social problems. However, there are dangers in this instrumental approach that have been highlighted by critics of the policy analysis mainstream. Sharing these criticisms, I begin as a proponent of what has come to be called the post-positivist tendency in policy analysis. (No policy analysis has ever actually measured up to the canons of logical positivism as philosophy of science and practice, so in this sense the term post-positivist is misleading. However, even proponents of the traditional approach such as Weimer (1999) now recognize the distinction made in this language, so the usage can be accepted, though personally I would rather call the poles "technocratic" and "critical").

The case for post-positivism has to differ quite substantially between the kind of political system in which I live, Australia, and that familiar to most readers of The Good Society, the United States. Here I will develop and compare the cases that need to be made. The differences help to illuminate the inadequacy of what passes for explanatory theory of the policy process, which policy analysis as an applied activity must also take on board in order to account for its own application.

Post-positivist policy analysts start with a critique of the role of technocratic analysis and search for subtle influences such as material forces, discourses, and ideologies that act so as to condition the content of policy. While a few analysts might consider these forces pervasive and overwhelming, most would regard them as in some degree contingent, such that it is worthwhile to explore escapes from them. In light of such considerations, post-positivists are interested in conceptions of rationality in society and policy that are more expansive and subtle than the instrumental, means-ends rationality that pretty much defines the technocratic alternative. Many (but not all) post-positivists are interested in a more authentic democratization of the policy process. Thus one finds calls for more genuinely participatory policy analysis (see Fischer, 1993 among others), though process values other than democratic ones could also be brought to bear. Many post-positivists draw on social and political theory in an effort to illuminate aspects of policy and society missed by narrower models of policy analysis.

Post-positivism does not, however, connote a well-defined recipe book for doing policy analysis, and in this it is at a disadvantage with more traditional kinds of policy analysis, especially when it comes to curriculum design. It has perhaps done better when it comes to critique of its traditional opponents than in providing such recipes. But most of its proponents would say that the whole point is to replace the illusion of certainty with recognition of the reality of contention and so avoid simplistic recipes. However, some practitioners have indeed thought about and developed methodological guidelines (for example, Fischer, 1995), which are quite demanding and require substantial epistemological self-consciousness. 1

Post-Positivism Amid Incrementalism:
The United States

In some ways it is relatively easy to be an advocate of post-positivist policy analyst in the United States. There, the post-positivists' case can begin by pointing out that technocratic policy analysis founders in the face of political reality, and, relatedly, that it produces work that policy makers can never use. Further, the technocratic image is untrue to what analysts actually do—especially when it comes to doing anything that might have an application. U.S. political reality features of course the separation and sharing of power across the branches and levels of government, undisciplined parties within the legislative branch, and widespread access to veto power on the part of organized interests. Technocratic policy analysis proceeds in the image of an omniscient benevolent decision maker, a situation in which there is no politics (Majone, 1989, calls this "decisionism"), let alone politics of the complex American sort. In a complex political system, instrumentally rationalistic policy making is possible at best only in rare moments of consensus amid crisis, or in the occasional area insulated from more pluralistic control (such as diplomacy or national security policy). [End Page 32]

The normal condition of U.S. policy making is impasse, from which only incremental deviations are possible. The classic statement of this condition is of course Lindblom's (1959) "science of muddling through," though Lindblom erred in treating disjointed incrementalism as a universal condition, as opposed to a contingent feature of the U.S. political system. From the post-positivist perspective, Lindblom erred too in failing to recognize the systematic distortions and biases to which incrementalism and (by association) alleged pluralism were subject, though in fairness he later corrected this oversight (see especially Lindblom, 1990). But irrespective of these finer points, technocratic policy analysis fails because in a U.S.-style context it has an inadequate implicit model of the policy making process.

Given that the policy-making system is generally decentralized, and decentralization is one characteristic of democracy, it is but a short step for proponents of critical or participatory policy analysis to argue that the process should be held up to principles either of communicative rationality (for example Healey, 1993) or of more authentic democracy. Communicative rationality may be defined as the degree to which interaction is engaged by competent actors, free from domination, deception, self-deception, and strategizing. There is no real conflict between these two sets of principles, more a question of emphasis and level of abstraction. Communicative rationality resonates with the now-dominant deliberative approach to thinking about democracy (Dryzek, 1990). The idea of policy analysis is then to act as one kind of critical input, possibly in the form of argument, to an interactive process. Such policy analysis can also attend to the condition of the process itself. Both sorts of inputs can be grouped under the heading of the "argumentative turn" in policy analysis (Fischer and Forester, 1993).

Post-positivist policy thinkers can further buttress their position here by noting that the technocratic model is untrue to what practicing analysts actually do with their time. So Forester (1989) has demonstrated at length that most of what planners do is actually a kind of communicative action. It involves questioning and directing attention, not technocratic manipulation. Perhaps surprisingly, Lynn (1999) in his savage attack on the post-positivists also points out that rigidly analycentric means-ends calculation forms a small part of what analysts actually do in the United States and the world. But while Lynn uses this empirical point to claim that the post-positivist critique mischaracterizes and oversimplifies the reality of traditional sorts of policy analysis, Forester uses it to argue that policy analysis as a field ought to jettison the positivist model.

Even if analysis in the technocratic idiom stands no chance of being acted upon in efficient and unchanged fashion, the mere presence of technocratic policy analysis serves to reinforce a discourse of disempowerment for those who are not part of the technocratic specialization being deployed or are not the targeted "policy makers" for the analysis in question. Correspondingly, it buttresses an ideology of a managerial-interventionist state, whose center of gravity is left-centrist rather than conservative or market liberal, that engages in economic management and the provision of social programs via the welfare state. So Lynn's (1999) defense of traditional policy analysis against the post-positivist critique fails because though he is right that traditional-technocratic policy analysis captures only a small part of what analysts actually do, the technocratic image of analysis remains powerful, even if it is only window dressing. This kind of analytical image and self-image has real negative consequences—even if it is untrue to how analysts actually spend most of their time.

Post-Positivism Confronts Real Technocracy: Australia

Let me now turn to the very different context offered by Australia, where there is actually a possibility that technocratic analysis really can be acted upon and so very directly help constitute technocratic policy making. This is in large measure a consequence of the Westminster system in which a prime minister and cabinet control a majority in Parliament and face only weak judicial scrutiny (though Australia also has a strong upper house elected by proportional representation, where the government rarely commands a majority, and federalism complicates the picture). When I moved to Australia from the United States, I continued to teach public policy and its analysis. Soon it became apparent that just about everything I thought true about the way the policy process worked was true only for the aberrant polity of the United States. Worse still, what passes for explanatory policy theory—models of disjointed incrementalism, garbage can, bureaucratic politics, pluralism, advocacy coalitions, and so forth—turns out to be theory mostly applicable only to the United States. In Australia, rational comprehensive policy making driven by social science findings can and does occur, and some of those engaged in policy making are quite self-conscious of this fact. My students had little difficulty in coming up with counterexamples that disproved the conventional [End Page 33] American "theoretical" wisdoms, and indeed they often found it quite hard to grasp the importance of that material.

The rational-comprehensive model was for example widely applied in social policy development under a federal Labor government in Australia in the 1980s and early 1990s. Edwards (1997) describes in detail how a child support scheme that used the income taxation system to transfer payments from income earning non-custodial parents to sole parents was developed and implemented in this idiom, largely under her own guidance. She also notes that other initiatives such as the higher education contribution scheme (whereby students pay back tuition fees as a percentage of post-graduation income) and programs to benefit the long-term unemployed have followed the same model.

One other prominent example of rational-comprehensive policy making in action in Australia is microeconomic reform. Since the mid-1980s under both Labor and conservative (Liberal-National) governments, Australia has been transformed from a statist, "colonial socialist" economy to a deregulated, open system. Government itself has been made more competitive through devices such as purchaser-provider splits and compulsory competitive tendering for service delivery. This reform process has been guided by public choice theory and free market neoclassical microeconomics more generally, and has proceeded despite opposition from powerful and previously protected sectors of the economy, as well as public sector unions.

Whether this kind of policy making is good or bad in any particular instance would on the face of it depend a lot on the content of the ends of that policy making. In Australia, those who applauded effective social policy that more effectively delivered child support benefits to single parents from their ex-partners or helped return the long-term unemployed to work would often be the same people who were aghast as the textbook prescriptions of public choice led to across-the-board cuts in social spending and the wholesale privatization and contracting out of government services—and perhaps vice versa. But if we are thinking about the place of policy analysis in the good society, we have to go beyond the contingent and transient content of the ends of particular policies.

This fact that technocratic policy analysis can be quite consistent with the way the Australian policy making process works and actually be used instrumentally within the process means that the case for post-positivist policy analysis is harder to make than in the United States. Advocates of critical, argumentative, post-positivist, or participatory policy analysis must establish the relative desirability of their favored kinds of analysis in light of the availability of a seemingly effective technocratic alternative. How might this be done?

Arguments That Should Work Anywhere

First, one might question whether technocratic policy analysis is truly effective when it comes to the resolution of complex social problems. Most of the Australian social policy successes that involve the application of rational-comprehensive analysis-driven policy making are actually in response to quite simple and well-bounded problems. The applications of public choice microeconomics have had numerous unanticipated and unwanted effects. (For example, compulsory competitive tendering has opened up major new opportunities for corruption in the form of payoffs from providers to purchasers.) Australia actually has a number of intractable problems of substantial complexity and conflicting values that have defied the efforts of governments to tackle them. Foremost among these would be the question of reconciliation with the indigenous peoples of Australia and the ecological devastation that has attended intensive agriculture (on which Australia's export base depends). The more complex a problem, the greater the number of frames that can plausibly be brought to bear upon it; in this light, Schön and Rein (1994) recommend frame-reflective policy analysis that tries to sort out the frames at issue and explore the possibilities for productive interchange across their adherents. It is also possible to argue in anti-Weberian terms that it is participatory analysis engaged by communicatively competent actors that can best bring to bear all sides of a complex problem (Dryzek, 1990, ch. 3).

Second, there is an issue of legitimation that relates to the intrinsic value of democracy. As deliberative democrats (Cohen, 1989, among many others) now point out, democratic legitimacy is secured largely to the extent those affected by a policy decision have the ability or right to participate in deliberations about its content. Technocratic policy analysis either rules out such a process, or reinforces discourses and ideologies that denigrate it. Here, there are numerous synergies between post-positivist policy analysis and deliberative democratic theory. (On the idea that policy design should serve democracy, see Ingram and Smith, 1993). Pragmatically, to the extent legitimacy is achieved, then the likelihood that powerful actors will attempt to obstruct the implementation of policy is diminished.

Third, and relatedly, democracy has problem-solving worth as well as intrinsic value. A more participatory policy process [End Page 34] helps to create more effective and competent citizens, who are also more effective problem solvers, within the policy process and beyond. They are also more capable of constructing productive relationships with others concerned with different facets of complex problems. Liberal scholars such as Lindblom and Karl Popper have long perceived the "intelligence" that decentralized democratic processes can achieve. Lindblom (1990) eventually recognized the degree to which this intelligence could be impaired, thereby constraining the capacities of individuals within the process. Critical analysts do not stop with recognition of impairment but also contemplate liberation from it. To the extent citizens are liberated from impairment, they will be better able to contribute to complex problem solving. There is a role here for those Torgerson (1997) calls "dissenting professionals" to change the system from within. Torgerson describes the efforts of such professionals in cases of environmental policy making.

A further benefit of more participatory democratic policy making is that engaged and competent citizens contribute to the creation of social capital, which in turn is conducive to the effective functioning of both political and economic systems. (Social capital analysts such as Putnam (2000) stress non-political schools of association, but arguably political and public association and interaction are equally if not more important.) Without social capital and associated trust, the schemes of microeconomic policy analysts are likely to fail (for example, marketization will produce only corruption or, at an extreme, a mafia economy, as in Russia).

To use the language of problem solving as I have done has its dangers. Does it not turn the alleged critical alternative into just another instrumental activity, ultimately in the service of technocracy? To escape the danger here, I should stress that post-positivist policy analysis is not just, or perhaps even mainly, about solving problems. It is also about defining problems, and questioning and destabilizing accepted definitions. A purist critical position here would take us into the critique of political economy and society more generally. While there is nothing wrong in drawing such connections, post-positivist policy analysis just has to retain the connection to social problem solving. No apology is necessary in a world that features poverty, inequality, violence, and ecological devastation, none of which are merely social constructions.


I conclude that the post-positivist project is universally desirable, but the case for it, and how one goes about pursuing it, depend a great deal on the kind of political system within which one is located. In a system like the United States, the case for post-positivist policy analysis is easier to make—but the institutional innovations necessary for that kind of analysis to prosper are hard to introduce. Thus modest analytical innovations may be more attractive to the post-positivist (Durning, 1999). Radical institutional re-shaping is more feasible in a country like Australia (as a decade and a half of privatization and marketization of government make clear)—but that possibility is itself indicative of the degree to which technocratic analysis can be put into policy practice. This situation is further corroboration for the recognition that, as Harold Lasswell argued long ago, the policy sciences of democracy must be contextual as well as multimethod- and problem-oriented. (For an excellent post-positivist interpretation of Lasswellian policy science, see Torgerson, 1985.)


John Dryzek is a professor of social and political theory in the Research School of Social Sciences at Australian National University.


Cohen, Joshua. "Deliberation and Democratic Legitimacy." The Good Polity: Normative Analysis of the State, Alan Hamlin and Philip Pettit, eds. Oxford: Basil Blackwell, 1989.

Dryzek, John S. Discursive Democracy: Politics, Policy and Political Science. Cambridge: Cambridge University Press, 1990.

Durning, Dan. "The Transition from Traditional to Postpositivist Policy Analysis: A Role for Q Methodology." Journal of Policy Analysis and Management 18 (1999): 389-410.

Edwards, Meredith. Australian Social Policy Development Processes. Melbourne: Centre for Public Policy, University of Melbourne, 1997.

Fischer, Frank. "Citizen Participation and the Democratization of Policy Expertise: From Theoretical Inquiry to Practical Cases." Policy Sciences 26 (1993): 165-87.

Fischer, Frank. Evaluating Public Policy. Chicago: Nelson-Hall, 1993.

Fischer, Frank and John Forester, eds. The Argumentative Turn in Policy Analysis and Planning. Durham, N.C.: Duke University Press, 1993.

Forester, John. Planning in the Face of Power. Berkeley: University of California Press, 1989.

Healey, Patsy. "Planning Through Debate: The Communicative Turn in Planning Theory." The Argumentative Turn in Policy Analysis and Planning, Frank Fischer and John Forester, eds. Durham, N.C.: Duke University Press, 1993.

Ingram, Helen and Steven Rathgeb Smith. Public Policy for Democracy. Washington, Brookings, 1993.

Lindblom, Charles E. The Science of Muddling Through. Public Administration Review 19 (1959): 79-88.

Lindblom, Charles E. Inquiry and Change: The Troubled Attempt to Understand and Shape Society. New Haven, Conn.: Yale University Press, 1990.

Lynn, Lawrence E. Jr. "A Place at the Table: Policy Analysis, Its Postpositive Critics, and the Future of Practice." Journal of Policy Analysis and Management 18 (1999): 411-24.

Majone, Giandomenico. Evidence, Argument, and Persuasion in the Policy Process. New Haven, Conn.: Yale University Press, 1989.

Putnam, Robert D. Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster, 2000.

Schön, Donald and Martin Rein. Frame Reflection. New York: Basic Books, 1995.

Torgerson, Douglas. Contextual Orientation in Policy Analysis: The Contribution of Harold D. Lasswell. Policy Sciences 18 (1985): 241-61.

Torgerson, Douglas. "Policy Professionalism and the Voices of Dissent: The Case of Environmentalism." Polity 29 (1997): 345-74.

Weimer, David. "Comment: Q-Method and the Isms." Journal of Policy Analysis and Management 18 (1999): 426-9.


1. Post-positivism now finds an institutional manifestation in the Conference Group on Theory, Policy and Society, affiliated with the American Political Science Association (http://www.cddc.

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.