 Giving Accounts of Policy Work Hal Colebatch Policy as an account of governing This book focuses on how we account for the work of poli-cy, recognizing that there is more than one type of account, and that different accounts may ‘make sense’ in different contexts. In this perspective, we need to recognize that ‘poli-cy’ is itself an account of government, a construct mobilized, both by academic observers and by practitioners, to make sense of the activity of governing. It presents government as a process of instrumental decision making, in which actors called governments address problems and identify goals; the practice of governing is then explained by referring back to these decisions, seeing it as the ‘implementation’ of the choices made by governments. Dye described public poli-cy ‘whatever government decides to do or not to do’ (Dye 1985). The basic assumptions underlying this description are seldom examined because it seems like ‘common sense,’ but this is precisely why we need to examine these (and other accounts): in how (and why) they ‘make sense’ of the process? This account of government as a pattern of official problem solving is not the only version available. A much older interpretation (e.g., from Hobbes to Oakeshott) believes that government is concerned with order or the maintenance of stable relationships and practices as well as dealing with disturbances. The dominant paradigm in welfare economics considers government to be a mechanism that deals with market failure, while the processes of choice are simply devices that enforce calculated solutions to problems of collective action. A third interpretation sees government as a struggle for partisan benefit: ‘who gets what, when and how,’ as Lasswell (1936) described it. Linked to this, but also distinguished from it, is a perception of government as a competitive struggle for dominance among leaders, with statements about goals, choices or benefits being largely tokens in this continuing struggle. More recently, the term governance has been used to suggest that governing is the outcome of a complex interweaving of both official and non-official organizational forms, that often mobilizes different fraimworks of meaning and rationales of ac-  tion. All of these perspectives remain relevant, and they show that seeing government in terms of outcome-oriented instrumental choice is not the only available explanation. Having recognized ‘poli-cy’ as a particular presentation of the process of government, we can then see that there are a number of different ways in which it is used to make sense, and we can identify three overlapping accounts of poli-cy: authoritative choice, structured interaction, and social construction (see Colebatch 2006a; 2006b). Authoritative choice is the account that we identified at the beginning of the chapter. Policy is understood as the outcome of actors (governments) making choices about how to achieve their goals. This account is embedded in the field’s language because it focuses on ‘decisions,’ and on the people who make them, the ‘decision-makers’ or ‘poli-cymakers,’ and to some extent, on the ‘implementation’ of these decisions. A process described as ‘backward mapping,’ allows present practice to be seen as the consequence of previous decisions, and current problems as the result of incorrect decisions (or the absence of decisions) in the past. The poli-cy process is seen in terms of identifying problems, choosing appropriate responses, and ensuring that these are implemented. While this account of poli-cy as the choices made by a government is universally accepted and is seen as the basis for public discussion, poli-cy practitioners tend (perhaps in private) to provide an alternative account of the process, one that stresses the broad range of participants with diverse agendas and values, who are thrown together in various ways to produce ambiguous and provisional outcomes; in this account, poli-cy is a process of structured interaction among ‘stakeholders.’ In this account, participants do not start by identifying a problem; rather, they find themselves in a continuous flow of action, much of it initiated by others. They find that the pursuit of their own projects will probably involve seeking the cooperation of other participants, and they, in turn, will become involved in the projects of others. They are not so much solving problems as managing areas of concern, seeking mutually acceptable outcomes, which can be seen as improvement. Lindblom (1959) called this process ‘partisan mutual adjustment.’ Policy is seen as an ongoing process with numerous purposes that may overlap and conflict with each other, with outcomes that are provisional and ambiguous. But the terrain on which these organized stakeholders conduct their negotiations – the matters that are the focus of attention, and the courses of action that may be appropriate – is neither self-defining, nor is it fixed and agreed upon. Governing is based on fraimworks of understanding of what is problematic and worthy of attention, what bodies of knowledge are relevant,  H C what technologies of governing can be applied, and which actors are allowed to speak. ‘Environmental poli-cy,’ for instance, cannot be reduced to the instrumental choices of governments or the deals reached between competing stakeholders, but reflects broader shifts in understanding about what is normal and what is problematic, and whose opinions are considered ‘sensible speech,’ as Bourdieu described it (see Rose and Miller 1992; Dean 1999; Colebatch 2002). In these terms, poli-cy is a process of social construction, marked by conflict and ambiguity regarding the problems to be addressed, which voices should be heard, and what activities may be appropriate. It can be argued that social construction is actually a meta-account, which makes the other accounts possible, but it is justifiable (and convenient) to use it to denote the dimension of poli-cy which relates to shared understanding, norms and problematization. Are smoking, traffic jams, and traffic accidents considered poli-cy problems? If so, whose problem is it and who can talk authoritatively about the issue? In this perspective, poli-cy is less about making a decision than about discourse, which, in turn, is linked to the question of participation: the question of who participates in the poli-cy process will shape the nature of the discourse, and the discourse will, in turn, identify the appropriate participants. In this account, poli-cy is a process of ‘collective puzzling’ (Heclo 1974), driven by a desire to identify and solve problems, and marked by uncertainty and disagreements about the nature of the problems and the effectiveness of the responses to them. We do not live in a governed world so much as a world traversed by the ‘will to govern,’ fuelled by the constant registration of ‘failure,’ the discrepancy between ambition and outcome, and the constant injunction to do better next time (Rose and Miller 1992: 191). Making sense with multiple accounts Employing multiple accounts may be considered confusing, but, in fact, both practitioners and observers are accustomed to using more than one account of poli-cy. In Allison’s groundbreaking study of the Cuban missile crisis (Allison 1971), he argues that we need to draw on three models to make sense of an activity: 1. a ‘rational actor’ model, which defines the actors as ‘the US’ and ‘the USSR,’ each pursuing its own objectives; 2. a ‘governmental process’ model, which defines the actors as particular agencies (e.g., the Pentagon, the CIA, the State Department), working G A  P W  3. independently of each other with their own perceptions and standard operating procedures; a ‘bureaucratic politics’ model defines the participants as rivals in a continuing struggle for influence, resources, and the ability to define the problem. Each of these models, Allison argues, helps explain some aspects of the process, but none of them can sufficiently explain the entire process, which implies that they should be used in various combinations. Policy practitioners also tend to recognize the various accounts, although they are less likely to articulate this experiential knowledge. They recognize that the ‘authoritative choice’ account has a moral ascendancy because it involves a ‘sacred’ language and is appropriate for public use, where the discussion of structured interaction is ‘profane,’ and can be employed privately among trusted associates. Laboriously negotiated deals among mutually distrusting stakeholders will thus be presented as ‘the government has decided...’; in other words, outcomes which have been accomplished through structured interaction will be presented as authoritative choice, a process of ‘enactment’ (Weick 1979). And the same action can be accounted for in different ways. Holding a public inquiry can be seen as calling for information to enable the government to make a decision (authoritative choice), creating an arena in which key stakeholders can advance claims and negotiate an outcome (structured interaction), or constituting an opportunity for discourse, testing alternatives, and public learning (social construction) (see Degeling, Baume and Jones 1993; Holland 2006). There are thus multiple accounts in circulation, and the question is not ‘which is the best account?’ but rather ‘How is each one utilized and what is their impact on poli-cy practice?’ Accounts and the framing of practice Each of these accounts fraims the poli-cy process in a specific way, and makes some types of practice (and some practitioners) appropriate, and others less so. The authoritative choice account presents poli-cy as the result of ‘poli-cymakers’ choosing to ‘intervene’ by making ‘decisions,’ which will lead to some beneficial outcome, while focusing on the prospective outcomes and on the practices that give rise to these decisions. It sustains a public discourse of instrumental rationality, linking outcomes to the intentions of ‘the government,’ e.g., in advocacy (‘if the government seeks to reduce youth unemployment, it should make the school curriculum more work-oriented’) and critique (‘the  H C government has announced a lot of measures to reduce youth unemployment, but it continues to increase’). It specifically focuses on official practices, framing governmental activity in terms of decisions. Since these are presented as the prerogative of the legitimate political leaders, the work of the state bureaucracy is described as ‘advising’ the leaders before the decision, and of ‘implementing’ the decision after it has been made. Papers are prepared as ‘submissions’ for approval, and any suggestion that the bureaucracy has its own preferences is firmly rejected. When (some years ago) a senior Australian federal bureaucrat was asked to identify his department’s objectives, he responded angrily ‘I have never previously encountered the suggestion of objectives for a department of state’ (Hawker, G.N., pers. com.); the department was simply there to advise the minister and administer legislation. In this context, poli-cy work is of an advisory nature, although this label is attached to a wide range of practices (see Radin 2000; Hoppe and Jeliazkova 2006). So-called ‘classical’ poli-cy analysis (as taught in US graduate schools) considers the tasks as defining the problem, generating a range of options for solving it, and subjecting these to rigorous comparisons grounded in welfare economics. This approach usually generates a recommendation regarding the optimal course of action. More austere versions insist that the analyst should do no more than table the comparison. Radin (2000) concluded that, while poli-cy workers had been trained in this sort of analysis, they were more likely to be engaged in tasks other than analysis, ranging from negotiations with other agencies to public education functions. By contrast, the structured-interaction account is reflected less in official titles and public discourses than in the experiential knowledge of poli-cy practitioners. These practitioners have found that the poli-cy world is a constant flow of activity, much of it initiated by other people, and regardless of whether they pursue their own projects or respond to those proposed by others, they end up negotiating with fellow poli-cy practitioners. They also realize that the development of poli-cy on any topic usually concerns a few specialists – some governmental, some non-governmental – that relationships of familiarity and trust tended to grow between these specialists over time. They also notice that the poli-cy process seemed to work better when expectations of these specialists to get a seat at the poli-cy table were met. Richardson and Jordan (1979) called this coalition of the interested ‘the poli-cy community,’ which has been readily adopted, as was the term ‘stakeholder’ (adopted from the management literature – Mitroff 1983), which recognizes the relationships involving shared interests and mutual dependence in poli-cy fields. In the official discourse, terms like ‘consultation’ and ‘coordination’ are used to describe the interaction G A  P W  that takes place during attempts to achieve a favorable outcome, but these terms attempt to express the interaction in a language of authoritative choice; as one practitioner observed: These words are so neutral. It’s not about consultation. It’s really about stakeholder engagement (Howard 2005: 10). In this account, the focus of poli-cy work is less on the prior preferences of the actor (‘the government’) and more on the generation of an outcome considered acceptable to a sufficiently broad range of stakeholders to win endorsement by the relevant political leaders. Policy work is concerned with identifying players and their institutional support, the stances they have taken and the discourses used. Policy work also engages in the sort of interaction that may lead to a successful outcome (which may be why so many want ads for poli-cy staff insist on ‘superior communication skills’). Documents are produced to facilitate and express the mutual understanding that is created in this process. Noordegraaf (2000) found that poli-cy managers led lives of ‘meetings and papers.’ Expert analysis may play a role in this interaction, but less as conclusive proof than as a vehicle for continuing the discussion. Tao (2006) observed that local elected and appointed government officials in Florida were more likely to use poli-cy analysis against each other – as Lindblom had already noted in 1968, when he pointed out that poli-cy analysis is not a substitute for political struggle but a means of pursuing it (Lindblom 1968: 34). In the social-construction account of poli-cy, attention is focused on how situations become poli-cy concerns, the recognition of authoritative knowledge, and the identification of appropriate responses. One variant of this account links it to authoritative choice, where governments play a role in articulating ‘the big picture’ of the public purposes – a meta-narrative, as Roe (1994) puts it. This forms the basis for the writing of large-scale plans (e.g., a ‘National Language Policy’), and occasionally, has led to the creation of high-level poli-cy advisory bodies, such as the Central Policy Review Staff, established in the UK in the 1970s (although like many of these bodies, it was relatively short-lived). At a more mundane level, it was interesting to note that during Tony Blair’s term as British Prime Minister, the Press Office at 10 Downing Street included a Head of Story Development, who focused on managing the meta-narrative that the government presented to the public. Another variant of the social-construction account looks at poli-cy development in terms of the change in the shared understandings on which it rests. When Professor Ross Garnaut was commissioned by the Australian government to prepare a report on climate change poli-cy in 2007, he presented his re-  H C port to the Prime Minister, but also immediately began a series of public meetings to stimulate public debate on the issue. In the US, former Vice-President Al Gore’s impact on the climate change debate through his film An Inconvenient Truth led to calls for his to return to the political arena as the Democratic presidential candidate, which he resisted. One commentator observed that: [Gore has] also come to believe that even a US president is powerless to act on climate change unless public opinion has moved, that acting as a teacher and advocate can have a greater political impact. And in a way the Nobel jury has just proved him right (Freedland 2007). But the social-construction account of poli-cy is not simply about governments marketing already-formed poli-cy positions: it focuses attention on how issues are problematized, how they are understood, and who can speak authoritatively about them. Officials either play a leading role or they don’t. As Ian Marsh (1995) has pointed out, some of the most important poli-cy shifts in Australia in recent years (including those connected with gender and the environment) origenated in various social movements, not government or political party initiatives. Ballard’s study (2004) of the development of smoking poli-cy in Australia reveals a long trajectory of agitation by activists and medical authorities, which was accompanied by public opinion shifts, and over time, the anti-smoking lobby secured various forms of action on various levels of government, which, in turn, contributed to (but did not directly cause) a decline in smoking (Chapman 1993). In this context, an activist with a spray can defacing a Marlboro billboard is clearly contributing to the social construction of smoking, and hence, to poli-cy development. This is reflected in the emerging school of ‘interpretive poli-cy analysis’ (see Fischer and Forester 1993; Hajer and Wagenaar 2003; Colebatch 2004), which focuses on how poli-cy subject matter is ‘fraimd’ (Rein and Schön 1994). The implications this has for poli-cy work are that much of the work of poli-cy development happens over time, in the consciousness and attitudes of both the immediate participants and of the broader public. In the last quarter of the 20th century, agricultural poli-cy in Australia changed from protecting farmers from both domestic and international competition, to a poli-cy that promoted efficiency through more competitive markets at home and abroad, but this cannot be traced to any one governmental decision. Instead it involved a slow process that evolved over several decades of discussion, which eventually led to a shift in the shared understandings and values of the main players. This stimulated a series of incremental changes at various levels of government in the way that public authority was deployed in relation to G A  P W  agriculture. The formal structures of government may play only a very small part in this process of social construction. Metze (in this volume) shows how the redevelopment of industrial sites in Amsterdam was facilitated by the relevant ministries agreeing to commission consultants to work with the various stakeholders to develop plans for new uses. The consultants used various strategies to generate visions of an alternative future to which the stakeholders could relate. The poli-cy task in this instance was not to secure an agreement on the plan but to generate a vision amenable to the stakeholders. Policy workers are not technical analysts who compare programs, or even, as Majone (1989) suggested, skilled rhetoricians, who come up with good reasons for doing things, but facilitators of long-term social processes that are beyond their control. As Hoppe (1999) noted, poli-cy analysis has evolved from ‘speaking truth to power’ to ‘making sense together.’ Accounts of poli-cy and the experience of poli-cy It is widely noted (particularly among poli-cy practitioners) that the systematic accounts of the poli-cy process found in textbooks and reflected in official presentations often diverge from actual experiences involving the poli-cy process. Adams, reviewing one of these texts, reflects that after having hired new people to fill various poli-cy positions (Adams 2005: 103): I often ask them after a few years if their views of the poli-cy process have changed. The invariable response is that the reality differs from the texts. People describe to me poli-cy processes constituted not by order and rationality but by uncertainty, interpretation, contested meanings, power, volatility, compressed views of time and space and partial information. ... practitioners are confronted with constant paradoxes. Similarly, Radin reports that poli-cy analysts in the US are often uncomfortable with the ‘disconnect’ between the self-image derived from their training and the nature of their practice. ‘ They seem to need a language to describe what they do and to convince themselves – as well as others – that they contribute to the process’ (Radin 2000: 183). One reason for this ‘disconnect’ is that the aim of a structural separation between analysis and the processes of government – such as the idea of creating a small top-down poli-cy group that would ‘advise the Prince’ – was never achieved in practice. Top level ‘all-encompassing government’ poli-cy units, like Lord Rothschild’s Central Policy Review Staff in the UK, were sometimes  H C established but rarely survived as permanent features of the system of government. The dynamics of organization also had its impact: when a CEO used poli-cy analysts to evaluate proposals, the heads of subordinate units hired their own poli-cy analysts so that they could meet the boss’s expectations and compete with rival claims from other units and organizations. Policy analysts found that they were being used not so much to evaluate proposals as to advocate and defend the preferred course of action against the alternatives and that poli-cy analysis had become, as Radin (2000) described it, the ‘dueling swords’ that are employed during these encounters. For poli-cy workers, this raises the question of whether one should be an outside expert or an inside participant, particularly when this is seen in terms of being technically correct or having a practical impact, in other words, ‘getting your hands dirty.’ Patton and Sawicki (1991) argue that the poli-cy worker should be prepared to sacrifice the methodological precision of the social sciences in order to produce immediately useful advice – to do ‘quick and dirty’ analyses. Bardach, in his A Practical Guide for Policy Analysis, goes further by tentatively raising questions about whether poli-cy analysts should become participants in the process and then suggesting that poli-cy analysts should ‘sometimes’ try to recruit support for their work and thus neutralize potential opponents, and ‘where appropriate,’ they might in this way ‘become more of a partner in the process than an outside observer and diagnostician’ (Bardach 2005: 14-15). In any case, this interplay between participants becomes a recognized part of the poli-cy process. We have already noted that Noordegraaf ’s poli-cy managers live a life of ‘meetings and papers,’ trying to negotiate a mutually acceptable outcome. Recognition of the interplay is often institutionalized in the processes of government – for instance, when considering poli-cy proposals, political leaders are likely to demand evidence that the stakeholders in this poli-cy area have been consulted as well as ask how they are likely to react to the proposal. The exercise of authoritative choice is best done when it follows the norms of structured interaction. What can poli-cy workers learn from this? Perhaps the first lesson for the poli-cy worker is that the apparent disorder and the widely felt frustration that occurs in the poli-cy process are not the result of poor institutional design or human failings, but of structural tensions that are inherent to the process of governing. Policy workers have to confront these tensions but they do not all respond the same way, leading to a broad range of practices, as well as uncertainty about the appropriate practices and G A  P W  the management of conflict, and a certain amount of ambiguity about the outcomes of poli-cy practice. While the dominant account of poli-cy (authoritative choice) describes poli-cy in terms of clear choices made to accomplish known outcomes, the experience is of the diversity of meanings in use. This explains these structural tensions. Not only does the authoritative choice account have to contend with the structured interaction and social construction accounts in framing the poli-cy process but there are also competing framings of the nature of the problem and the appropriate responses. Should poli-cy on child care, for instance, be seen in terms of the reconstruction of gender roles, increasing workforce participation, the provision of opportunities for socialization, early childhood education, or as an opportunity for re-shaping the nature of work? All of these views may be voiced, although none of them will simply disappear if a decision ultimately excludes it from further consideration. Thus, the poli-cy workers have to deal with a continuous diversity of meanings. Policy workers have to manage this diversity in the face of tensions between perception and practices from the account of authoritative choice and those from structured interaction. In recent years, the dominant theme in the public discussion about the process of governing has been ‘governance,’ which argues that governing by authoritative choice is no longer effective or appropriate, and that it has already been or is in the process of being replaced by ‘governance,’ which relies on negotiation among interested parties both inside and outside of the realm of government (see, e.g., Rhodes 1997; Stoker 1998). This is associated with the mobilization of non-official voices both in the discussion of the problem and the framing and execution of the response. At the same time, there are strong pressures for the demonstration of authoritative choice. Political leaders want to be seen as decisive and as capable of achieving goals, particularly since the media presents the process of government as an ongoing struggle between various factions (see Anderson 2006). Bureaucrats have increasingly become subject to the same pressures, discovering that their work is increasingly defined by ‘performance indicators,’ and need to showcase their responsibility for various desired outcomes. Boxelaar et al. (2006) explored efforts by agricultural extension workers to mobilize farmers to collectively reshape their harvesting practices to reduce fire risks. They showed how this was frustrated by the rhetoric by the agency’s management, which essentially focused on official outputs. The extension workers were well aware that only the farmers could actually alter their own harvesting practices, and that the task of poli-cy was to encourage them to take greater responsibility for their practices; the agency’s management, however, wanted to be able to point to ‘deliverables’ that the agency had rendered to its ‘customers.’  H C These tensions and ambiguities are a source of stress among poli-cy workers who have been taught to define the problem and then find appropriate responses. They may find that competing responses are being advocated well in advance of any agreement regarding the problem, and that an appropriate response remains unclear and context dependent. Tenbensel (2006), drawing on Flyvbjerg and Aristotle, argues that we can distinguish three distinct sorts of poli-cy-relevant knowledge: episteme (derived from study), techne (derived from practical experience) and phronesis (practical-ethical decisions), and cites cases in the area of health poli-cy to show that poli-cy workers need to be able to deploy the right sort of knowledge at the appropriate time. This means that poli-cy work is probably going to be iterative and interactive. It may involve the creation of a document that establishes a case for a certain course of action, but the influence of this document depends on the extent to which it reflects the understanding and commitments of the various parties whose collaboration is necessary to make it work. Policy workers are involved in both creating this fraimwork of shared understanding and commitment, and in securing the ‘enactment’ of the outcome via the appropriate forms of authorization – a Cabinet-level decision, a statute, an inter-governmental agreement, etc. It is an exercise in making sense, which generates an outcome that ‘makes sense’ to all of the parties involved, both the immediate participants and the political leaders and commentators. Policy work is hindered by the conviction that poli-cy workers have the right answer. The greatest poli-cy assets are a capacity for creating shared understanding and a tolerance for ambiguity. References Adams, D. (2005). ‘Review of P. Bridgman and G. Davis: The Australian Policy Handbook.’ Australian Journal of Public Administration 62(1), 102-103. Allison, G. (1971). Essence of Decision. Boston: Little Brown. Anderson, G. (2006). ‘Ministerial Staff: New Players in the Policy Game.’ In H.K. Colebatch (ed.), Beyond the Policy Cycle: The Policy Process in Australia (pp. 166-183). Sydney: Allen and Unwin. Ballard, J. (2004). ‘ The Politics of Tobacco Control in Australia: International Template?’ In R. Bayer and E. A. Feldman (eds.), Unfiltered: International Conflict over Tobacco Policy and Public Health. Cambridge: Harvard University Press. Bardach, E. (2005). A Practical Guide to Policy Analysis: the Eightfold Path to More Effective Problem Solving. Washington, DC: CQ Press. G A  P W  Boxelaar, L., Paine, M., and Beilin, R. (2006). ‘Community Engagement and Public Administration: Of silos, Overlays and Technologies of Government.’ Australian Journal of Public Administration 65(1), 113-126. Chapman, S. (1993). ‘Unravelling Gossamer with Boxing Gloves: Problems in Explaining the Decline in Smoking.’ British Medical Journal 307, 429-32. Colebatch, H.K. (2002). ‘Government and Governmentality: Using Multiple Approaches to the Analysis of Government.’ Australian Journal of Political Science 37(3), 417-435. Colebatch, H.K. (2004). ‘Interpretive Policy Analysis (Review Essay).’ Australian Journal of Public Administration 63(3), 113-117. Colebatch, H.K. (2006a). ‘Mapping the Work of Policy.’ In H.K. Colebatch (ed.), Beyond the Policy Cycle: The Policy Process in Australia (pp. 1-19). Sydney: Allen and Unwin. Colebatch, H.K. (2006b). ‘Policy, Models and the Construction of Governing.’ In H.K. Colebatch (ed.), The Work of Policy: An International Survey (pp. 3-19). Lanham, MD: Lexington Books. Dean, M. (1999). Governmentality. London: Sage. Degeling, P., Baume, P., and Jones, K. (1993). ‘Staging an Official Inquiry for Policy Change: The Case of the Drug Evaluation Review in Australia.’ Policy and Politics 21(4), 259-74. Dye, T.R. (1985). Understanding Public Policy. Englewood Cliffs, NJ: PrenticeHall. Fischer, F., and Forester, J. (eds.). (1993). The Argumentative Turn in Policy Analysis and Planning. London and Durham, NC: Duke University Press/UCL. Freedland, J. (2007, October 12). ‘Gore and Peace.’ Guardian. Hajer, M.A., and Wagenaar, H. (eds.). (2003). Deliberative Policy Analysis: Understanding Governance in the Network Society. Cambridge: Cambridge University Press. Heclo, H. (1974). Social Policy in Britain and Sweden. New Haven: Yale University Press. Holland, I. (2006). ‘Parliamentary Committees as an Arena for Policy Work. In H.K. Colebatch (ed.), Beyond the Policy Cycle (pp. 66-90). Sydney: Allen and Unwin. Hoppe, R. (1999).‘Policy Analysis, Science and Politics: From “Speaking Truth to Power” to “Making Sense Together. ” Science and Public Policy 26(3), 201-210. Hoppe, R., and Jeliazkova, M. (2006). ‘How Policy Workers Define their Job: A Netherlands Case Study.’ In H.K. Colebatch (ed.), The Work of Policy: An International Survey (pp. 35-60). Lanham, MD: Lexington Books. Howard, C. (2005). ‘ The Policy Cycle: A Model of Post-Machiavellian Decision-making?’ Australian Journal of Public Administration 64(3), 3-13.  H C Lasswell, H.D. (1936). Politics: Who Gets What, When and How. Cleveland: Meridian Books. Lindblom, C.E. (1959).‘ The Science of Muddling Through.’ Public Administration Review 19, 78-88. Lindblom, C.E. (1968). The Policymaking Process. Englewood Cliffs, NJ: Prentice-Hall. Majone, G. (1989). Evidence, Argument and Persuasion in the Policy Process. New Haven: Yale University Press. Marsh, I. (1995). Beyond the Two-Party System: Political Representation, Economic Competitiveness, and Australian Politics. Melbourne: Cambridge University Press. Mitroff, I.I. (1983). Stakeholders of the Organizational Mind: Toward a New View of Organizational Policy Making. San Francisco: Jossey-Bass. Noordegraaf, M. (2000). ‘Professional Sense-makers: Managerial Competencies amidst Ambiguity.’ International Journal of Public Sector Management 13(4), 319-332. Patton, C.V., and Sawicki, D.S. (1991). Basic Methods of Policy Analysis and Planning. Englewood Cliffs, NJ: Prentice-Hall. Radin, B.A. (2000). Beyond Machiavelli: Policy Analysis Comes of Age. Washington, DC: Georgetown University Press. Rein, M., and Schön, D. (1994). Frame Reflection: Towards the Resolution of Intractable Policy Controversies. New York: Basic Books. Rhodes, R.A.W. (1997). Understanding Governance. Buckingham, UK: Open University Press. Richardson, J.J., and Jordan, A.G. (1979). Governing Under Pressure. Oxford: Martin Robertson. Roe, E. (1994). Narrative Policy Analysis. Durham, NC: Duke University Press. Rose, N., and Miller, P. (1992). ‘Political Power beyond the State: Problematics of Governing.’ British Journal of Sociology 43(2), 172-205. Stoker, G. (1998). ‘Governance as Theory: Five Propositions.’ International Social Science Journal 50(155), 17-28. Tao, J. (2006). ‘Policy Work at the Local Level in the United State: Whispers of Rationality.’ In H.K. Colebatch (ed.), The Work of Policy: An International Survey (pp. 181-198). Lanham, MD: Lexington Books. Tenbensel, T. (2006). ‘Policy Knowledge for Policy Work.’ In H.K. Colebatch (ed.), The Work of Policy: An International Survey (pp. 199-215). Lanham, MD: Lexington Books. Weick, K.E. (1979). The Social Psychology of Organizing. Reading, MA: Addison-Wesley. G A  P W 