Bacchi

 A “what’s the problem represented to be?” approach to policy analysis, therefore, has as its goal opening up policies and policy proposals for the kind of critical analysis offered in the examples above. It produces a methodology to identify the problem representations implicit in policies and policy proposals, and to reflect on their ethical implications. The approach shares a basic insight with sociological framework theory – that how “problems” are “framed” matters to perceptions of the issues and the outcomes that follow. However, the differences between framing theory and a problem representation approach are significant. For example, a good deal of framing theory is concerned with intentional or strategic framing, either to accomplish a particular goal or to resolve a dispute. By contrast, a focus on problem representation moves us outside the realm of intentional “impression management” (Provis 2004, Chapter 9) to consider the deep-seated conceptual logics implicated in specific policies and policy proposals. Moreover, the overriding purpose of frame theory is to find ways to understand the dynamics of policy making (Hajer and Laws 2006, 260) in order “to learn how to make better decisions” (Goodin, Rein and The Ethics of Problem Representation - 17 Moran 2006, 11). Framing is understood as an “ordering device” that assists in understanding how policy actors deal with ambiguity. By contrast, a problem representation approach is an analytic method for scrutinizing the deep-seated assumptions and preconceptions that underpin particular policy directions. The intent is to point out “on what kinds of assumptions, what kinds of familiar, unchallenged, unconsidered modes of thought the practices that we accept rest” (Foucault 1981 in Pálsson and Rabinow 2006, 91).

Conventional approaches to public policy restrict the understanding of public policy to “what governments do, why they do it and what difference it makes” (Edwards 2004). In these accounts public policy is understood as simply the output of government. Some years ago Bachrach and Baratz (1963) expanded on this understanding of policy-making processes by noting that government policy includes not just what governments do, but also what they do not do, not only decisions, but also non-decisions. Steven Lukes (1974) famously pushed the boundaries of analysis to include three dimensions of power in policy making. A one-dimensional view of power is concerned with “who prevails in cases of decision-making where there is an observable conflict” (Lukes 1974, 10), that is, what gets on the policy agenda. A two-dimensional view of power, associated with the work of Bachrach and Baratz (1963), directs attention to the way power “may be, and often is, exercised by confining the scope of decision-making to relatively ‘safe issues’” (Lukes 1974, 18), limiting the issues that make it to policy agendas. Lukes’ three-dimensional view of power extends the analysis to consider how power is exerted by influencing, shaping or determining people’s wants. This way of exercising power prevents people “from having grievances by shaping their perceptions, cognitions and preferences in such a way that they accept their role in the existing order of things …” (Lukes 1974, 23-24). Developments in post-positivist policy analysis (see for example Shapiro 1992, 11) take these insights a step further, pointing out that the ways in which issues are represented have wide ramifications that need to be considered part and parcel of policy analysis. Governments in this interpretation are not reactive, addressing political “problems”; rather they are active in the creation of particular understandings of political “problems”. Because policies and policy proposals identify what needs to change, they also imply what the “problem” is. Hence representations of the “problem” are implicit in all policies and policy proposals (see Turnbull 2006, 20, fn 3). Crucially, these problem representations matter for what is done and not done, and for how people live their lives. Deborah Stone (1988, 106) describes how representations of a problem define “interested parties and stakes”, allocate “the roles of bully and underdog”, and affect power relations. These implications invite ethical reflection. In other work (Bacchi 1999) I have constructed an approach to policy analysis, which I call “what’s the problem represented to be?”, that assists in identifying 14 - Carol Bacchi the ethical dimensions of the problem representations that lodge within existing and proposed public policies. The approach consists of a set of questions. First, ask how the “problem” is represented in a specific policy proposal and/or in a specific policy debate: what is the “problem” represented to be? Then consider the assumptions and presuppositions that underlie this representation of the problem. Next, reflect on the impact these presuppositions and assumptions may have on possible outcomes: what is likely to change with this representation of the “problem”? What is likely to stay the same? Who is likely to benefit from this representation of the “problem”? Who is likely to be harmed? Identify who is held responsible for the “problem” and how this attribution of responsibility affects both those targeted and the perceptions of the rest of the community about who is to “blame”. Consider what is left unproblematic in this representation of the “problem” and how “responses” would differ if the “problem” were thought about or represented differently. Finally, examine the means of dissemination of identified problem representation/s as an exercise of power and scrutinize reflexively one’s own representation of the “problem”. A focus on problem representations directs attention to all three of Lukes’ dimensions of power. The analysis starts with policies that are proposed or have made it to the policy agenda (one-dimensional power) but the purpose is to identify what is missing from or suppressed (silenced) with those policies (a form of two-dimensional power). In addition, a specific goal is to scrutinize the implications of particular representations of the “problem” for political subjects, their positioning in society and their subjective reactions to that positioning (three-dimensional power). However, whereas Lukes (1974, 45, 34) seems to endorse the idea of false consciousness and talks about “interests” as if they are readily identifiable, in a problem representation approach there is no pure outside from which to exercise these judgments; hence, the insistence in a “what’s the problem represented to be?” approach to subject one’s own problem representations to reflexive scrutiny. In addition, in this approach there is a broadening of the understanding of public policy to include the roles played by experts, including doctors, psychiatrists, psychologists and indeed ethicists, in producing particular representations of social “problems” and particular kinds of governable subjects. That is, in this approach, governance extends beyond the institutions of government to include “an inventive, strategic, technical and artful set of ‘assemblages’ fashioned from diverse elements” (Dean and Hindess 1998, 8), including the human sciences (Dreyfus and Rabinow 1983, xxiv).

Carol Bacchi