Modernisation of British Government

This article critically reviews the approach to evidence based policy-making (EBPM) advanced in the Labour government's modernisation agenda. The article contends that EBPM must be understood as a project focused on enhancing the techniques ofmanaging strategic policy-making as opposed to improving the capacities of the social sciences to influence the 'practices of democracy' as envisaged by Lasswell, or facilitating the kind of systems thinking advocated by Schon and by more recent students of 'complexity'. Introduction: Social Science and the Task of Improving Policy Making From its beginning the policy sciences movement was concerned about the relationship between knowledge, policy-making and power. This issue was at the heart of the work of Harold Lasswell, a founding father of public policy as a field of study. Lasswell believed that democratisation was an ongoing process and that the particular challenge facing modern democracies was how to ensure that policy-making could be informed by a new kind of interaction between knowledge producers and users (Torgerson, 1985). It is a matter of historical record that Lasswell's hopes for the 'policy sciences of democracy' were not to be realised. Indeed, if anything, the experience has been more consistent with the rise of the 'policy sciences of tyranny' than of democracy (Dryzek, 1989, p.98). There is scant evidence that policy-making has, in Britain, America, or elsewhere, been informed by the kind of relationship envisaged by Lasswell (Fischer, 1998; Lindblom, 1990; De Leon, 1997). Policy-making in liberal democracies has, for the most part, been more about 'muddling through' rather than a process in which the social or policy sciences have had an influential part to play. Hope, however, has a habit of springing eternal. With the election of a Labour government in 1997, a new chapter in the troubled relationship between government and the Public Policy and Administration Volume 17 No. 3 Autumn 2002 43 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from social sciences appeared to be opening in Britain. A defining moment for this new chapter came with David Blunkett's speech to the ESRC in February 2000: Influence or Irrelevance: Can Social Science Improve Government? In his speech Blunkett called for a new relationship between social science and government which would bring to an end the 'irrelevance' of social science to the policy-making process. This new relationship would be built on the territory marked out by the secretary of state: in this brave new world government would rely on social scientists to 'tell us what works and why and what types of policy initiatives are likely to be most effective' (Blunkett, 2000). This defining of the role of social science as about 'what works' set out a clear and unequivocal agenda for the future: public policy had to be driven by 'evidence' and policy research focused on finding out 'what works'. On the one hand, New Labour's 'what works' philosophy constitutes a real opportunity for social science to exercise 'influence' in the policy process after many years of 'irrelevance'. On the other, this desire for an evidence based approach is also something of a threat to the possibilities of realising the Lasswellian vision for the role which the policy sciences could play in democratising (and modernising) the policy process. The 'policy sciences of democracy', he maintained, must be 'directed towards knowledge to improve the practices of democracy' (Lasswell, 1951, p.15). This paper argues that evidence based policy-making (EBPM) is a missed opportunity for improving government and has only served to make the relationship between knowledge and policy-making in a democratic society more muddled rather than less confused. In its belief in the existence of 'firm ground' in (what Schon termed) the policy 'swamp', EBPM must be understood as a project focused on enhancing the techniques of managing and controlling the policy-making process as opposed to either improving the capacities of the social sciences to influence the 'practices of democracy' as envisaged by Lasswell, or facilitating the kind of systems thinking advocated by Schon in the 1970s and by more recent students of 'complexity'. Mapping and Occupying the Policy High Ground The emergence of EBPM as an approach to improving policy-making came to prominence a few years into the first Blair government, in 1999. It was an important year for EBPM in Britain. Two important conferences were held. One under the auspices of the Association of Research Centres in the Social Sciences, and the other at University College London in association with the Cochrane Centre at Oxford. In addition the journal Public Money and Management (Vol 19, No 1) produced a special number on evidence based policy-making. Alongside these developments the ESRC were also working on their plans to establish a national resource centre for evidence based policy. Meanwhile, the government signalled its commitment towards an evidenced based approach in the Modernising Government White paper published in March 1999 (Cabinet Office, 1999a), and this was followed up Public Policy and Administration Volume 17 No. 3 Autumn 2002 44 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from by a Cabinet Office report (Cabinet Office, 1999b). The government also set up an number of new units Performance and Innovation Unit, Social Exclusion Unit and the Centre for Management and Policy Studies (CMPS) within the Cabinet Office to drive forward the commitment to evidence based policy. In due course, when a special evidence based policy edition of Public Money and Management was published as a book, Professor Ron Amann, the director of CMPS, wrote a foreword. He concluded: Donald Schon perhaps went too far in once describing the policy process as 'a swampy lowland where solutions are confusing messes incapable of technical solutions'. Like the contributors to this volume, I believe that there is firmer ground. We should like to map it out and occupy it' (Davies, Nutley and Smith, 2000, p.vii). This is the central underlying assumption of EBPM: that, contrary to Schon's argument, there is indeed a firm high ground in the policy swamp, and the task is to 'map it out and occupy it'. The swamp idea is what Schon himself would have described as a 'generative metaphor' (Schon, 1979) since EBPM is inspired by the belief that, despite the mess, there does indeed exist some firm ground upon which can be laid 'hard facts' to support the grand edifice of modernised policy-making. The mess can be mapped and occupied. It is possible and desirable to move policy-making out of the realm of muddling through to a new firm ground where policy could be driven by evidence, rather than political ideology or prejudice. It was, despite the rhetoric of 'new' Labour and 'modernisation', a return to the old time religion: better policy-making was policy-making predicated on improvements to instrumental rationality. Schon's swamp is a world of change, and full of complexity, uncertainty and ignorance which suggests a very different approach to policy-making to that envisaged by EBPM. The metaphor issue raised by Amann signals that there are profound ontological, epistemological and methodological differences between those who believe in EBPM and those who have doubts as to its feasibility or the values it embodies. EBPM's conceptualisation of policy reality is that it is a territory capable of being 'mapped' and 'occupied', whereas from the Schonian point of view, Amann and company are chasing a notorious (positivist) will o' the wisp long associated with this particular swamp: a will to power. Knowledge for EBPM is a means of controlling the mess, and draining the swamp. In this respect EBPM marks not so much a step forward as a step backwards: a return to the quest for a positivist yellow brick road leading to a promised policy dry ground somewhere, over Charles Lindblom where we can know 'what works' and from which government can exercise strategic guidance. If policy-making is to go beyond the muddling through of the past, then, argued Mr Blunkett, 'sound evidence' is vital for government to secure 'overall strategies' (Blunkett, 2000). There, ensconced on the policy high ground, policy makers demand the big picture of what works. This meant, for Blunkett, that: Public Policy and Administration Volume 17 No. 3 Autumn 2002 4-5 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from We're not interested in worthless correlations based on small samples from which it is impossible to draw generalisable conclusions. We welcome studies which combine large scale, quantitative information on effect sizes which allow us to generalise, with in-depth case studies which provide insights into how processes work. (Blunkett 2000) It is, apparently, 'self-evident that decisions on Government policy ought to be informed by sound evidence' (Blunkett, 2000). Policy should not guided by 'dogma', but knowledge of 'what works and why'. It follows, with a certain inevitability, that the kind of knowledge/evidence that policy makers need is: 'to be able to measure the size of the effect of A on B. This is genuine social science and reliable answers can only be reached if the best social scientists are willing to engage in this endeavour'. The policy-making process needs 'A-B' knowledge if policy makers are to 'be tough on problems and tough on the causes of problems'. This 'knowledge as power' model contained in the Blunkett speech is the leitmotif that runs throughout the documents which provide the key texts for EBPM in the Blair government. These include, Professional Policy-Making for the Twenty First Century (Cabinet Office, 1999b; Parsons, 2001); Better Policy-Making (Cabinet Office, 2001); and Modern Policy-Making, (National Audit Office, 2001). Professional policy-making, according to these publications, must be driven by 'evidence' of 'what works' (Cabinet Office, 1999b, 7.1). If government is to 'deliver' and go beyond muddling through, policy-making has to be redesigned so that policy 'professionals' can have access to policy analysis which can facilitate the steering process (Hogwood, 2001). 'Evidence' is portrayed as essentially a problem of how knowledge can be utilised and managed. EBPM as expressed in the major official statements (Cabinet Office, 1999a) and subsequent documents involves the management of two forms of knowledge: academic research and professional! institutional experience. * Academic/ research evidence (Episteme). Knowledge produced by academics/ policy researchers which provides 'causal' or theoretical knowledge of what works. Here the challenge is to influence the production of policy-relevant knowledge so that it is reproduced in a form which can be utilised by policy makers. As is the case with evidence based medicine, the bias is overwhelmingly towards quantitative approaches. * Professional and institutional experience (Techne). Knowledge about 'what works' in practical policy-making. The focus here is the development of systems which can facilitate better management of 'learning' and knowledge within government departments and agencies. The major characteristic of both of these forms of manageable knowledge is that they are capable of being made explicit and codified. What counts as evidence is that which can be aggregated and disseminated: added up, joined up and wired up. 'Mapping' and 'occupying' the firm ground require that 'evidence' must be 'managed'. At one level this means that clear structures and procedures have to be put in place so that institutional modes of Public Policy and Administration Volume 17 No. 3 Autumn 2002 46 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from 'knowing what works' and 'learning' can be extracted, stored, retrieved and communicated, whilst on the other, that research or academic knowledge must be produced and disseminated in accordance with clear specifications of what counts as systematic A -B knowledge. Meanwhile, Back in Schon's Swamp The problem with EBPM is that it can run, but it cannot hide from Schon's point about the existence of the swampy lowland. The result is that EBPM, in trying to move beyond muddling through, succeeds only in creating more muddle surrounding the issue of how the making of public policy can be a process of organisational and public learning. Schon's work directs our attention to the central role of uncertainty, flux, unpredictability and change in human affairs, and the impossibility of knowing very much. Schon argued that change was here to stay: the idea that there is ever a state of 'stability' and 'unchangeability' was a dangerous illusion. The full quote from Schon's Reflective Practitioner cited by Amann reads: There is a high, hard ground where practitioners can make effective use of research-based theory and technique, and there is a swampy lowland where situations are confusing 'messes' incapable of technical solution. The difficulty is that the problems of the high ground, however great their technical interest, are often relatively unimportant to clients or to the large society, while in the swamp are the problems of greatest human concern. Shall the practitioner stay on the high, hard ground where he can practice rigorously, as he understands rigor, but where he is constrained to deal with problems of relatively little social importance? Or shall he descend to the swamp where he can engage the most important and challenging problems if he is willing to forsake technical rigor? ;...] There are those who choose the swampy lowland. They deliberately involve themselves in messy but crucially important problems and, when asked to describe their methods of inquiry, they speak of experience, trial and error, intuition, and muddling through (Schon, 1983, pp.42-3). The great task for those concerned with the future of government was (as he argued earlier in his Reith Lectures, Beyond the Stable State) to 'develop institutional structures, ways of knowing, and an ethic for the process of change itself' (Schon 1973, p.10). However, although problems constantly change and mutate, institutions tend to remain rather fixed and immutable. Schbn drew attention to the gap between institutions and problems: institutions, he argued, have an 'inertial life of their own' and the problems of today all too often take place in institutional contexts that are fifty years and more out of date. Second, Schon argued that as change is so important a process to understand, the critical question to ask was how can we develop systems which best provide for learning and adaptation? There was, he maintained, not an 'information gap': there was no shortage of evidence, information and data. The deficit was less to do with information than our capacity for public and private learning. Schon focuses on the issue of learning rather than the idea of knowing: on the learning rather than the information or evidence gap and the gap between institutions and problems. Public Policy and Administration Volume 17 No. 3 Autumn 2002 47 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from And what followed from this was that we had to understand government and policy-making as a process of learning. For Schon the answer to the question of improving government as a learning system involved radically re-thinking and redesigning the policy processes of increasingly more complex information societies. Sch6n's prediction was that institutions and professions (and individuals) would come under increasing pressure to change and would have to learn to adapt to a world which was ever more uncertain and unpredictable. We had to think about how we could re-design our institutions so that they could 'confront the challenges to their stability without freezing and without flying apart at the seams'; we had to think about how our values and sense of selfidentity could be maintained in conditions of continuous transformation; we had to think about our response to the challenges of guiding, influencing and managing these transformations. In short, he believed that: We must ... become adept at learning. We must become able not only to transform our institutions, in response to changing situations and requirements; we must invent and develop institutions which are 'learning systems', that is to say capable of bringing about their own continuing transformation. (Schon, 1973, p.28) Public policy, in Schon's sense, is really the study of how societies learn (or fail to learn) about those problems they define as being public and how they seek to solve (or fail to solve) their problems. The aim of government should not be to control and direct, but to facilitate the growth of individuals, organisations and communities that are capable of managing their own continuing transformation. Central government should lay less stress on its dubious and doubtful claim to know what is best for a particular organisation, in a particular place at a particular time, and more emphasis should be placed on organisations making the best use of local knowledge and their learning experiences. For government to become a learning system ... Government cannot play the role of 'experimentor for the nation' seeking first to identify the correct solution, then to train society at large in its adaptation. The opportunity for learning is primarily in discovered systems at the periphery, not in the nexus of official policies at the centre. Central's role is to detect significant shifts at the periphery, to pay explicit attention to the emergence of ideas in good currency, and to derive themes of policy by induction. The movement of learning is as much from periphery to periphery, or periphery to centre as from centre to periphery. Central comes to function as a facilitator of society's learning, rather than as society's trainer. (Schon 1973, p.166) This means that government should definitely not be in the business of deciding what works, so much as it should aim to be informed about (what Schon described as) ideas in good currency. Government acting as a 'facilitator of society's learning' rather than giving strategic guidance and exercising strategic control. Government which does not see itself or indeed portray itself as a 'problem solver', because the paradoxes which frame Public Policy and Administration Volume 17 No. 3 Autumn 2002 48 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from complex problems can never be solved or fixed. Given the nature of social change identified by Schon, the claim that government can actually possess the kind of knowledge sufficient for it to arrive at 'national solutions' to policy problems is both erroneous and dangerous to democracy. Government cannot fix things because things are never fixed: all is flux and uncertainty. We cannot put our feet into the same policy river (or swamp) twice. In a world which is, as Schon described it, unstable and uncertain, we cannot know in the way assumed by those who advocate the kind of instrumental rationality embodied in new public management. Schon's whole point was that due to the growing complexity of human problems we had to improve our capacities to learn and facilitate adaptation in conditions of rapid and cascading change rather than pursue the utterly facile belief that we can increase our capacities to know and to control simply because we have increased our capacities to manage information and 'evidence'. For any one problem there are a multiplicity of incompatible perspectives and solutions and a vast array of different local conditions and forms of knowledge. Evidence, in itself, as Schon argued, cannot help us to resolve conflicts of value in an uncertain world (Schon, 1973, p.199). This did not mean, however, that because we do not have the kind of scientific knowledge and 'facts' to which managerialists and EBPM lays claim, that we are doomed to a kind of 'epistemological nihilism' (Schon, 1973, p.213). The shift to communicative modes of policy-making involves the recognition that although we cannot know, we can learn and the role of government is to facilitate private and public learning. That is, government should be concerned with the task of designing policy processes and institutions that can enable citizens and consumers, private and public organisations, to become capable of bringing about 'their own continuing transformation' (Schon, 1973, p.28). The challenge for the twenty-first century is, as Schon argued in the 1970s, to create learning schools, learning hospitals, learning police services, learning universities and so on. Institutions which are open, transparent and accountable to taxpayers and users, but also institutions which are designed to be self-organising and self-transforming, but not self-regarding and selfregulating. Institutions which are designed to deal with change, instability, uncertainty, ignorance and chaos, rather than for stability, determinacy, knowledge and predictability. Schon's point about the swampy lowlands was, given that policy-making does indeed take place in such territory, our strategy should be to take account of the very forms of knowledge which EBPM seems to consider as irrelevant. Above all, it means that the role of the centre is to facilitate a process of learning in which the centre promotes periphery-periphery learning rather than central steering. The task of improving policy-making was fundamentally to do with improving communication and reducing distortions rather than improving control through better use of evidence. Policy disputes and conflicts, as he later emphasised, cannot be resolved by better evidence per se, but require the Public Policy and Administration Volume 17 No. 3 Autumn 2002 49 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from active participation and reflective learning of those who design and implement specific programmes in specific contexts (Schon and Rein, 1994). Policy conflicts are about meaning rather than simply 'evidence'. That is what finding out 'what works' really means (on this point see Yanow, 1996). EBPM, Self Transformation and Self Organisation. From a Schonian perspective, therefore, the idea that it is possible, or desirable, to map and occupy some policy high ground through a EBPM strategy is highly questionable. The key issue, for Schon, is how public policy-making can become a genuine learning process. This involves the notion that government should aim to facilitate the growth of organisations which are capable of their own self-transformation. In contrast, the approach to 'evidence' in the modernising government agenda, is predicated on strategic steering. EBPM has to be placed within the context of reforms to the public sector which place considerable stress on audit, monitoring, performance, strategic planning and quality management systems. Knowledge, from the perspective of (so-called) new public management, constitutes organisational capital and ways have to be found whereby this capital can be tapped and organised so as to improve the intellectual resources available for policy makers. This can be done through the dissemination of best practices or best case studies and by developing methods of organising or 'pooling' knowledge so that it can be accessed by everyone in the organisation. Knowledge management is a critical, if not the critical aspect of managing for 'delivery'. But, can knowledge in and for the policy process be managed in the way assumed by EBPM? Given the complexity which confronts policy makers, it could be argued that the belief in knowledge management and strategic guidance which underpins EBPM is an assumption which needs to be questioned. Ralph Stacey, for example, using the insights of the complexity sciences, argues that there are significant problems with the kind of mainstream knowledge management approach adopted by EBPM. Knowledge management..becomes a matter of getting knowledge out of people's heads and into something which can be owned by the organisation. A need is thus created to codify knowledge and record it, or somehow locate it in a system accessible to everyone in the organisation. The first problem with this is that people know far more than they can possibly convey to others. One response is to draw a distinction between explicit knowledge and implicit knowledge ... The task then becomes one of transforming the tacit knowledge into the form of explicit knowledge (Stacey, 2002, p.50). In turn, the problem of making tacit into explicit knowledge become a problem of transforming organisations so that they can systematically monitor and regulate the performance of knowledge workers themselves. Knowledge management consequently is inextricably bound up with the redesign of organisational systems to facilitate more effective control and steering to secure given objectives and policy 'targets'. Knowledge Public Policy and Administration Volume 17 No. 3 Autumn 2002 50 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from management systems are, he argues, vital instruments for governmental 'control freakery'. Systems thinking is to blame for this state of affairs: That idea, which is perfectly sensible in relation to central heating systems, has been applied to people and has colonised thinking about organisations, as well as sociology and psychology. It treats our minds as essentially mechanistic devices (Stacey, 2002, p.50). Human organisations are, however, complex systems. They do not operate in the mechanistic way assumed by those who contend that knowledge can be managed and systematised. Policy-making takes place in conditions of uncertainty, flux, unpredictability and variation such that, as Stacey argues, knowledge management is something of a 'chimera'. Knowledge inheres in interactions between people. It is the product of relationships, not something discrete, locked away in individual heads. This is not to suggest that the act of planning is somehow futile, nor that organisations and processes can never be managed with any degree of surety, but merely that we need to think differently about how we manage our organisations ... It highlights, rather, the fact that people's intentions are necessarily subject to the push and pull of the intentions of others, and whatever the aim original aims of a 'knowledge management process might be, its results will be unpredictable and different, to a greater or lesser degree. We can never gain anything like total control of knowledge (Stacey, 2002, p.51). The implications of viewing systems in a complex, rather than mechanistic way is that 'less is more'. This means less steering via information/knowledge systems, and more emphasis on decentralisation and self-organisation of the kind we find in Schon. From Schon's perspective, the centre should be concerned to promote the growth of reflective organisations which are capable of their own self-transformation. EBPM, from the standpoint of complexity theory, fails to acknowledge the profound limits of steering in a world in which prediction and control is so difficult and in which 'evidence' is so problematic. Writers such as Stacey, (Stacey 1996; 2002; and Stacey, Griffin and Shaw, 2000) therefore, strongly echo Schon's argument for designing organisations for self-organisation. The relevance of complexity to improving government was not entirely lost on one of New Labour's leading policy advisors, Geoff Mulgan, (a cofounder of the think tank DEMOS) who after serving in the number 10 policy unit, became in due course, director of the PIU in September 2000. In his book Connexity: How to Live in a Connected World (1997), Mulgan considers the important insights which ideas about complexity and selforganising systems provide, but he is not convinced by the argument for a more limited role for government. Connexity (sic), despite its sympathy for complexity approaches necessarily involves top-down steering. In societies with access to richer communication it does indeed become easier to organise things laterally or in parallel, rather than in hierarchies. Since complexity grows from simpler systems, it is inherently hard to unravel, design or adapt a social system from the top, and futile to try ... But the limit to the more naive Public Policy and Administration Volume 17 No. 3 Autumn 2002 51 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from accounts of self-organisation is simply this: that complex adaptive systems need some hierarchy of organisation because challenges ... may beyond the capacity of subsystems to respond. The role for higher authorities is not to engineer the system, or to monopolise power and knowledge, but rather to perform the roles that the lower elements are unable to perform: like watching out for threats, averting disasters, resisting parasites or taking responsibility for the future. Putting too much faith in self-organisation can lead to the conclusions that there is no need, or no possibility, of an overarching common perspective and a capacity for steering ... human systems are neither so unknowable that we are unable to act, nor so self-organising that we have no need to (Mulgan, 1997, pp.188-9). This acknowledgement of the usefulness of a complexity approach is, however, tempered by the need for strategic guidance. The 'lower elements' are deemed to have a poor capacity to 'watch for threats', or for 'averting disasters' or 'resisting parasites' or indeed for 'taking responsibility for the future'. Those charged with steering have, by dint of their ability to 'know what works' had best do the thinking for the lower elements in the policy subsystem. Mulgan's 'connexity' thesis, it appears, acknowledges the existence of complex systems, but is very uncomfortable with the conclusion that policy can be 'delivered ' without recourse to managing knowledge and directing subsystems. This tension between a complex systems perspective, which stresses self-organisation, and EBPM which is fundamentally predicated on a command and control model has been evident in several PIU documents. PIU has argued the case, en passant, for a (soft) systems approach in its report on Adding it Up, (Cabinet Office, 2000c), and in a working paper on Better Policy Design and Delivery (PIU, 2001). Mulgan also re-emphasised his commitment to systems thinking whilst at PIU (Mulgan, 2001). However, even though these various PIU documents recognise the importance of non-linearity, continuous adaptation and learning, it is difficult to see how these ideas can be reconciled with the actual practice of policy makers or the philosophy espoused in Professional Policy-Making for the Twenty First Century. This mismatch between the conclusions of complexity approaches which tend to endorse Schon's Beyond the Stable State thesis and Blairite modernised policy-making is a point forcefully made in a DEMOS pamphlet by Jake Chapman. Pressing the case for thinking of policy-making in terms of complex adaptive systems he shows how 'the current model of public policy-making, based on the reduction of complex problems into separate, rationally manageable components, is no longer appropriate' (Chapman, 2002, p.1 1). What this means is that improving policy-making is (as Schon argued) about learning, rather than command and control. It involves democratic innovation, based on an appreciation of government as a complex adaptive system, rather than as a mechanistic and linear process concerned with better techniques of command and control. So called 'soft systems methodologies' suggest that, given the complex and adaptive quality of organisations, government has to learn to let go and learn to how to learn. In Public Policy and Administration Volume 17 No. 3 Autumn 2002 52 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from Schon's terms, making policy-making 'better' or 'modern' or more 'professional' involves less centre-periphery direction and more peripheryperiphery, and periphery-centre interaction. Schon's approach would see 'evidence' informed policy-making as a process whereby knowledge is transacted and exchanged rather than 'pooled' and utilised by the policy professionals at the centre. It would mean less emphasis on knowledge as power, and more emphasis on learning as means of self-transformation. But, of course, all of this fits uneasily within target setting and the inherent topdowness of 'new' approaches to public management. As Tom Bentley, Mulgan's successor as director of DEMOS, argued in an article in Renewal: Command and control is a framework unsuited to the complex. Unpredictable demands of contemporary organisational life ... In seeking to stimulate and influence ... change, the underlying goal of political intervention should be seen as developing or supporting systems of self-governance in a complex and fluid environment, rather than simply establishing institutional control or imposing a simplified set of priorities on existing systems ... The challenge is to recreate public institutions and governance regimes as open, porous and decentred systems which can thrive on diversity, adapt to radical innovation and still maintain coherent purpose and progress (Bentley, 2002, pp.1 1-17). As these interventions from Chapman and Bentley well illustrate, what it is unclear is how the knowledge management which underpins EBPM can be reconciled with what Schon has to say about the kind of knowledge/evidence which exists in the messy, complex policy swamp. A complexity perspective means recognising that government does not know best, and that all policymaking takes place in conditions of uncertainty and flux. It means accepting diversity, the possibilities of failure and recognising the importance of context. Above all it means that the profound limitations of the kind of reductionism which informs EBPM has to be acknowledged. The goal, as Schon argued in his 1970 lectures, should be a policy-making process in which there is an ongoing process of learning at all levels and between all levels. As Chapman concludes, improving policy-making has to be focused on 'the creation of a system of government that can learn for itself, on a continuous basis, guided by democratically legitimated goals and priorities' (Chapman, 2002, p.78). Out of the Manageralist Cul-de-Sac: From Blair's Dome to Lasswell's Social Planetariums. The aim of the policy sciences for Lasswell was to contribute to the democratisation of the policy-making process. This theme of democratisation was also central to Schon's approach to policy-making. By contrast, the aim of EBPM is the modernisation of the policy-making process. Modernising policy-making in turn involves the idea of 'professionalising'. The 'professional model' (Cabinet Office, 1999b) defines various stages which can be broken down into clear and distinct sets of management tasks. The policy-making process is thus perceived as being Public Policy and Administration Volume 17 No. 3 Autumn 2002 53 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from a highly mechanistic system which necessitates improvements in systemic communications and co ordination (wiring-up and joining-up). It means that policy makers have to be trained in the right skills and policy protocols and provided with appropriate 'tool kits' (Cabinet Office, 2002). Evidence is critical to this professional model. More professional/rational policy-making is, to a large extent, dependent on the better use of evidence and the management of knowledge. From a Laswellian and Schonian perspective this constitutes a very peculiar and wholly inaccurate representation of both the policy-making process and the challenge of actually improving it. To begin with, it excludes some pretty central aspects of policy-making: people, power and politics. It will be a matter of historical research to discern how much of the Blair government's policy-making was in fact driven by evidence, and how much by key actors seeking to secure their ideas and policy recommendations. It is curious, for example, that for a government which has made such an unprecedented use of special advisors, the model completely ignores the role which their ideas may have as opposed to 'evidence' in policy-making. Even more curiously, policy-making is characterised as an ideology/ value free zone in which professional policy makers are only interested in what works. Policy-making is viewed as quintessentially a problem of designing management systems and decision making protocols to promote learning about what works. This is all in complete contrast to Lasswellian project, which focused on the fact that policy-making took place in conditions of power inequalities and recognised that knowledge is utterly embedded in power and value contexts and relationships. The task of policy analysis was not to produce 'evidence' to drive policy, but to facilitate the clarification of values and contexts. This involved democratisation, rather than simply modernisation. Values are at the heart of the Lasswellian policy sciences movement, whereas for EBPM, values, like naughty children, must be seen, but never heard: EBPM is about what works rather than what you believe. It is about efficiency, effectiveness and economy in delivery, rather than ethics. Values are the messy things which have to be extracted to give good objective knowledge of what works. Lasswell argued that, if democratic policy-making were to meet the challenges of the modern world, the relationship between power and knowledge was an absolutely critical consideration. This meant that policy-making had to be opened up and made more democratic: distortions had to be removed, and assumptions and perceptions had to be tested through deliberation, argument and discussion. Public policy-making ought to be a process of public learning. An example of this was his idea that, just as we have planetariums to study and facilitate understanding of the universe, so we should construct 'social planetariums' to enable communities to observe themselves and reflect on their problems and the kind of futures they would like to see. Lasswell's vision was not simply one of improving the management of policy-making, but a grander, far more inspiring one of renewing democracy in an knowledge based Public Policy and Administration Volume 17 No. 3 Autumn 2002 54 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from society. This meant enhancing governmental decision making through such techniques as decision seminars which sought to provide a way in which policy makers could question and challenge their 'facts' and assumptions, but also of enriching social and political life by facilitating a wider engagement of citizens in the clarification of their values. Perhaps an appropriate symbol for EBPM is the ill-fated 'Dome' at Greenwich. Instead of providing an opportunity for the country to reflect on the past, and envisage its futures, and thereby open up critical areas and issues of public policy to argument and debate and the exploration of possibilities, the Dome was a project which had a mission to amuse and entertain rather than a mission to promote public learning. A Lasswellian dome at Greenwich on the other hand, would have been but one of numerous 'social planetariums' in which citizens all over the country would have had the opportunity to be 'liberated from the perceptual caves of the present'. Just as decision seminars provided an opportunity to explore alternatives in the policy-making process, so social planetariums would give equal weight to 'alternative versions of the future and the past.' In principle, every community can build its own social planetarium where stress is placed on local objectives, local history, local prospects..[It is] ..a means of giving importance to the institutions that in many places are struggling feebly for public recognition ... The contextual frame of reference the orientation towards the future and towards decision making is a 'shot in the arm'. It makes the past pertinent to the present and the future. In many circumstances ... it will prove expedient to amalgamate the social planetarium with a program that encompasses all the museums and related cultural resources of the capital of the locality, city, region, nation or the transnational areas involved ... Gradually society can be changed until people learn to live as much as in the imagination of the future as in the reminiscence of the past ... In such a configurative setting ... individual choices can be made at a respectable level of rationality (cited in Parsons, 1995, p.536). Such is the chasm that separates the managerialism of EBPM and the Lasswellian vision for a revitalisation of democracy through improving the capacity of decision makers and citizens to clarify their value choices. Rationality for Lasswell was about building a communicative rational capacity in government and society as a whole rather than simply improving the instrumental rationality of policy makers. It is, therefore, not the least bit surprising that the key texts of EBPM makes no reference to those approaches which draw on the Lasswellian inheritance rather than those which derive from the more positivistic/instrumentalist modes of policy analysis associated with the 1960s. No reference do we find to the challenges posed by the 'argumentative turn' to the positivist mindset in policy analysis which had emerged by the 1990s (Fischer and Forester, 1993). EBPM fails to grasp, indeed wilfully ignores, some of the issues raised by post-positivist approaches to the theory and practice of policy-making regarding the relationship between policy analysis and the task of democratising the Public Policy and Administration Volume 17 No. 3 Autumn 2002 55 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from policy-making process. And, of course, there is no reason why there should be. The EBPM approach is framed by the idea that policy delivery involves improving the management of policy-making. Better policy-making, equals better policy delivery. EBPM is remarkable for the extent to which it frames 'evidence' around the concepts of knowledge as episteme and techne, whereas the Lasswellian orientation would stress the vital role of another kinds of knowledge identified by Aristotle: phronesis, or practical wisdom (Flyvbjerg, 2001). A key difference as between the EBPM project and the Lasswellian (and the more recent post-positivist) orientation in policy studies is that the latter is more sensitive to the strengths and weaknesses and limitations of these different modes of knowledge or 'frames'. Indeed, for Lasswell, the great challenge for the policy sciences of democracy was to promote the integration of these different forms of knowledge. It would be democracy which would be the means of this integration. Hence, his promotion of decision seminars and social planetariums. The Lasswellian orientation, in stark contrast to EBPM, points towards the need to develop a more communicative rather than a purely instrumental approach to the production and use of policy relevant knowledge. The Lasswellian approach, above all, is about ensuring that knowledge is politicised, whereas for EBPM, the aim is quite the opposite: to de-politicise and managerialise knowledge production and its utilisation. Lasswell's policy sciences of democracy are concerned with clarifying values, and in Schon's sense promoting 'reflective' policy-making and a reflective and more participative democratic practice, as opposed to establishing 'hard facts' on which to erect a platform in the policy high ground. EBPM is narrowly focused on policy 'delivery': whatever works, works. The Lasswellian turn in policy analysis, on the other hand, is far more broadly focused on how the policy process relates to the continuing democratisation of society. Hence, for all their rhetoric of 'joined up' policy-making, the proponents of EBPM fail to address the most important issue raised by the existence of Schon's swamp: how can the redesign of modern policy-making be 'joined up' to the urgent and vital task of renewing politics and democratic society? Fearing Greeks and Looking Gift Horses in the Mouth: Can Government Damage Social Science? When David Blunkett made his speech in 2000 about a brave New Labour world that would see social scientists at the heart of the policy-making process it appeared to signal that the days of irrelevance could indeed be over, and an era of relevance was about to begin. 'Things, could only get better' as the party's election song put it in 1997. Social Science could improve government. Bliss it was to be alive! However, as was all too apparent, the Blunkett agenda for the social sciences in the 21st century had the distinct smell of twentieth century Vienna about it. Relevant knowledge was knowledge of causes. Like a latter day Mr Gradgrind, Mr Blunkett's Public Policy and Administration Volume 17 No. 3 Autumn 2002 56 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from argument was simple enough to reduce to a sound bite: 'Facts Sir, give me facts about what works, all else is dross!' The speech turned out, therefore, to be less of a new chapter in the history of public policy, so much as 'deja vu all over again'. This is not unique to the UK: as Morcol's research into the beliefs of policy makers shows, the supposed demise of positivism has been greatly exaggerated (Morcol, 2001). As Blunkett's speech, and the Cabinet Office 'professional model' illustrate, the belief in the fact-value dichotomy, objectivity, rational analysis, and quantificationism is alive and in fiercely robust health. In its crude and naive way, Blunkett's speech exposed the profoundly positivistic bias of the government's philosophy of evidence based policy-making. Making social science more 'relevant' consequently may prove to be more of a Trojan, than a gift horse. In EBPM what is to count is what can be counted, measured, managed, codified and systematised. Evidence based policy was, from the outset, a magnificent misnomer: the model that is developed in the documents produced by the Cabinet Office was in truth ECMLP, evidence controlled, managed and legitimated policy, rather than evidence based or indeed informed policy. The main problem with EBPM, qua 'What Works', is that it is rooted in a wholly managerialist and mechanistic way of thinking about policy-making. This has meant that the government's call for a policy-making process which is 'evidence based' inevitably narrowed and constrained the way in which 'evidence' has been conceptualised and operationalised. The idea that policy-making should not be evidence based is hardly a proposition which even the most severe critics of EBPM would advance. The big question is, however: whose evidence and what counts as evidence in the first place (Solesbury, 2001)? In one sense the 'what works' approach has clarified the situation, in a Humpty Dumptyish sort of way. When Blair and Blunkett et. al., use the word evidence it means whatever they want it to mean: evidence means 'what counts is what works'. As Lasswell argued, the policy orientation is, above all, contextual. Lasswell would have considered the question of 'what works?' as arrant and dangerous nonsense. 'What works' is about: what works, for whom, when, and how?; or what kind of evidence works for what kind of problem/ policy in what context, and for whom? Context, as the complexity approach taken by the likes of Stacey, Jackman and Bentley and as the realist approach also acknowledges is all (Pawson and Tilley, 1997). 'Evidence', as post-positivist approaches to policy analysis argue, is inextricably interconnected with the problem of participation, power and inequalities in power (Fischer, 1998, 2001; Torgerson, 1986; Dryzek, 1982). If 'evidence' is to really to have more of an influence on policy-making then this involves tackling the thorny Lasswellian question of 'who gets what, when and how'. Whose evidence gets what influence, when and how? To acknowledge these issues, however, would quickly precipitate the proponents of EBPM falling from the walls of the high firm ground into the complex and messy policy lowlands where facts are not so hard and causes not so easily determined and power is Public Policy and Administration Volume 17 No. 3 Autumn 2002 57 distribution. © 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorized by guest on April 12, 2008 http://ppa.sagepub.com Downloaded from unequally distributed. And it is unlikely if the EBPM project could ever beput back together again.For Lasswell, the aim of the policy sciences movement was thereinvigoration of democracy. The aim of the EBPM movement is, however,very different. And it is a defining and critical difference. Whereas theLasswellian and Schonian approaches recognise that 'facts' are imbedded inthe swampy world of values and politics and competing frames, EBPMwishes to extricate them from the political / value quagmire. Lasswell sawpolicy analysis as about clarifying values and contextualising problems:improving the relationship between knowledge and power was consequentlya fundamentally political and democratic task. Schorn too, believed that therelationship between policy-making and knowledge involved finding ways inwhich society could better understand competing 'frames' within which factsand evidence are constructed (Schon and Rein, 1991, 1994). EBPM sees therelationship between knowledge and policy-making as fundamentally amanagerialist rather than a democratic or political project. Blunkett's call forsocial science relevance, and the mantra of 'what works', points less towardsa renewal of social science or politics or democracy, than a trip back to thefuture to revive an intellectually muddled form of positivistic andmechanistic managerialism.Schon and Lasswell, thou shouldst be living at this hour! ReferencesBentley, T. (2002), 'Letting Go: complexity, individualism and the left', Renewal,Vol 10, No 1, pp.9-26.Blunkett, D. (2000), Influence or Irrelevance: can social science improvegovernment?, Swindon: ESRC, and Department for Education and Employment,February.Cabinet Office (1999a), Modernising Government, Cmnd 4310, (London: CabinetOffice).Cabinet Office(1999b), Professional Policymaking in the Twenty-first Century: Areport by the Strategic Policy Making Team, (London: Cabinet Office).Cabinet Office (2000a), Peer Review of Cabinet Office, (London: CabinetOffice). Cabinet Office(2000b), Modernising Policy Development, Centre for ManagementStudies, (London: Cabinet Office).Cabinet Office (2000c), Adding itUp: Improving Analysis and Modelling in CentralGovernment, (London: CabinetOffice).Cabinet Office (2001), BetterPolicy-Making, (London: Centre for Management andPolicy Studies).CabinetOffice, (2002), Beyond The Horizon: International Comparisons inPolicy-Making: Toolkit, (London: Centre for Management and Policy Studies).Chapman, J.(2002), Systems Failure:Why Government Must Learn to ThinkDifferently, (London: DEMOS).Davies, H.T.O., Nutley, S.M. andSmith, P.C.(2000), What Works? evidence basedpolicy and practice inpublic services, (Bristol: The Policy Press). De Leon, P.(1997), Democracy and the Policy Sciences, (New York: StateUniversity of New York Press). Public Policy and Administration Volume 17 No. 3 Autumn 200258 distribution.© 2002 SAGE Publications and PAC. All rights reserved. Not for commercial use or unauthorizedby guest on April 12, 2008http://ppa.sagepub.comDownloaded from Dryzek, J. (1982), 'Policy Analysis as a Hermeneutic Activity', Policy Sciences,Vol 14, pp.309-329.Dryzek, J. (1989), 'Policy Sciences of Democracy', Polity, Vol 22, No 1, pp.97-118.Fischer, F. (1998), 'Policy Inquiry in a Postpositivist Perspective', Policy StudiesJournal, Vol 26, No 1, pp.129-146.Fischer, F.(2001), Citizens, Experts and the Environment, (London: Duke UniversityPress).Fischer, Fand Forester, J. (eds) (1993), The Argumentative Turn in Policy Analysisand Planning, (London: Duke University Press/UCL Press).Flyvbjerg, B. (2001), Making Social Science Matter: Why Social Inquiry Fails andHow it can Succeed Again, (Cambridge: Cambridge University Press).Hogwood, B. (2001), 'Beyond Muddling Through -Can Analysis assist in designingpolicies that deliver?' Appendix 1, Modern Policy-Making, (London: National AuditOffice).Lasswell, H.D. (1951), 'ThePolicy Orientation', in D. Lerner and H.D. Lasswell(eds), The Policy Sciences, (Palo Alto:Stanford University Press).Lindblom, C.A. (1990), Inquiry and Change, (New Haven: YaleUniversity Press)Morcol, G. (2001), 'Positivist Beliefs Among Policy Professionals: anempirical investigation', Policy Sciences, Vol35, Nos3-4, pp.381-401.Mulgan, G. (1997), Connexity: How to Live in a Connected World, (London: Chatto and Windus).Mulgan, G. (2001), 'Systems Thinking and the Practice of Government', Systemist,23 November, pp.22-9.National Audit Office (2001), Modern Policy-Making: Ensuring Policies DeliverValue for Money, Report by the Comptroller and Auditor General, HC 289Session 2001-2002: November 2002, (London, National AuditOffice). Performanceand Innovation Unit (2001), Better Policy Delivery and Design: ADiscussion Paper, (London: PIU). Parsons, W.(1995), Public Policy, (Cheltenham: Edward Elgar). Parsons, W.(2001), 'Modernising policy-making for the Twenty First Century:The Professional Model', Public Policy and Administration, Vol 16, No3,pp.93-1 10.Pawson, R. andTilley, N.(1997), RealisticEvaluation, (London: Sage).Schon, D.A.(1973), Beyond The Stable State: Public and PrivateLearning in aChanging Society, (Harmondsworth Penguin Books).Schon, D.A.(1979), 'Generative Metaphor: APerspective onProblem-Setting inSocial Policy', in A.Ortony, (ed.), Metaphor andThought, (Cambridge:Cambridge University Press).Schon, D.A.(1983), TheReflective Practitioner.(New York: BasicBooks) Schon, D.A andRein, M.(1991), 'FrameAnalysis', in P. Wagner, B.Wittrock, C. Weiss and H.Wollman (eds), Social Sciences and Modern States: NationalExperiences and Theoretical Crossroads, (Cambridge: Cambridge University Press).Schon, D.A and Rein, M.(1994), FrameReflection: Toward the ResolutionofIntractable Policy Controversies, (New York: Basic Books).Solesbury, W.(2001), 'Evidence BasedPolicy: Whence it came and where it'sGoing', Working Paper, (London: ESRC Centre for Evidence BasedPolicy andPractice).Stacey, R.D.(1996), Strategic Management andOrganisational Dynamics, (London: