, ‘ Risk and Human Rights in UK Prison Governance ’ ( 2007 ) 47 ( 5 )

Risk and human rights discourses dominate the landscape of prison governance in the United Kingdom. For the most part, however, criminologists have focused only on concepts of risk and lawyers on human rights; there has been little overlap in the scholarship of these disciplines. In this article, we problematize this separation. We argue for a new stream of academic enquiry which recognizes the co-existence of different types of risk and rights discourses, and which draws upon more interdisciplinary understandings of risk, human rights and regulation.

In 2004, O'Malley appealed for 'more nuanced analysis of the ways in which, almost everywhere that risk appears, it is assembled into complex configurations with other technologies… '(2004: 26-7).This article responds to that appeal by focusing specifically on prison governance in the UK.It aims, first, to draw attention to a range of intersections between risk and human rights, and argues that these intersections require analysis.Secondly, it suggests some key features of a 'risk and rights' analysis by drawing upon insights from within criminology, human rights law and regulation scholarship.
UK prison governance provides the catalyst for our argument because, in recent years, analyses of penal policy and prison law have been very heavily influenced by discourses of risk and human rights (Sparks 2000a, Kemshall 2003;Livingstone et al. 2003).For the most part, however, risk discourses have been used only by criminologists and not by lawyers: equally, although human rights has featured prominently in the work of lawyers, it is rarely discussed in criminology.This divide amongst scholars who share a common interest in prison governanceprovides one of our starting points.The other starting point is a concern about the role of human rights law in an era in which public sector regulation and service delivery in the UK are increasingly organised around the concept of 'organisational risk' (Power 2004). 1   While it is well-recognised that 'risk management' has become 'a key organizing principle of contemporary correctional practice and offender management' (Maurutto and Hannah-Moffat, 2006: 438), there is less scrutiny of how the prison sector is responding to demandsfrom both internal and external sourcesthat it assess and manage 'organisational risk' (such as financial or legal risk) (Black 2005;Hutter 2006).What interests us most is how risk-based developments intersect with rightsbased scrutiny of prison administration, the latter prompted both by the legal obligations placed on all public authorities by the Human Rights Act 1998 (Lester and Pannick 2004;JCHR 2004) and, more broadly, by the growth of prisoner litigation and rights-consciousness at the devolved, national and European levels (Foster 2005;Owers 2004; Whitty et al. 2001).
In order to explore these issues, we have divided the article into three parts.The first part speculates on British criminology's (non) engagement with human rights and queries why the growth in prisoner rights litigation has not registered in criminology scholarship on prisons. 2 Then, in the second part, we highlight the widespread lack of engagement with risk amongst (human rights) lawyers, both at the theoretical level and in terms of the uses of risk assessment and management in legal practice.In the final part of the article, we argue that criminologists and lawyers should be examining the co-existence of risk and rights discourses in UK prison governance and we suggest some key features of a 'risk and rights' stream of academic enquiry.

Part One: Criminologists and Human Rights
1 As copious official documents attest, the 'handling of risk' (Strategy Unit of the Cabinet Office, 2002: 4) is claimed to be an 'overarching concept' of UK administrative governance (Fisher 2003: 455). 2 Two caveats: first, our argument concerns the UK experience, though we make some reference to scholarship from Canada and elsewhere.Secondly, we draw a sharp distinction between criminologists and lawyers.This seemed the best way to convey the basic point that both lawyers and criminologists have neglected the relationship between risk and rights.The downside is that we will be guilty of a degree of misrepresentation: in practice, individuals and organisations do not map neatly onto a 'criminology versus law' divide For example, the feminist reform campaigns of the Canadian Association of Elizabeth Fry Societies combine the use of human rights standards and a critique of risk assessment and management in the Canadian criminal justice system (see CAEFS 2002).
The simplest explanation for the lack of engagement by British criminologists with the relationship between risk and rights would run as follows: only a handful of criminologists ever mention human rights.As McEvoy (2003: 39) points out, the significance of human rights 'appears to have made little genuine inroads into [the] conceptual or practical frameworks' of criminology (see also Cohen 1998;Jamieson 1999).This neglect of rights is surprising given British criminology's close engagement with state power: if '[f]or most of its existence, criminology has been located, for all practical purpose, within the institutions of the criminal justice state' (Garland and Sparks 2000: 201), it seems strange that the growth of prisoners' rights would be overlooked by criminologists.Yet, apart from McEvoy's work (2001) examining the role of law in the context of paramilitary prisoners in Northern Ireland, and Liebling's discussion of human rights norms in the creation of a 'moral performance' framework for the assessment of prison environments (2004: 452-3), criminological scholarship on UK prisons and prisoners generally does not seem interested in engaging with rights.3How, then, might this continuing absence be explained?
It could be that criminologists (like lawyers (McCrudden 2006)) adhere strongly to disciplinary boundaries.Moreover, criminologists may be all the more conscious of these demarcations as a result of the ongoing debates over the fragmented nature and status of their discipline (e.g., Cohen 1998;Braithwaite 2000;Walklate 2001;Garland 2002;Zedner 2003), including for example contemporary appeals 'to reconnect criminology to sociology' (Mythen and Walklate 2006: 380;Young 2003). 4Many commentators have suggested that the disciplinary closure of British criminology can be traced to its historical bias towards administrative and empirical work and its closeness to established centres of power.So, for example, Brown and Pratt have claimed that: 'Born as it was to meet the needs of governance, criminology reaffirms itself not by internal reflection but rather by reference to the material demands of penal administration ' (2000: 3;Loader 1998).This claim points us towards another possible explanation for the general absence of human rights from the British criminology literature: throughout the 1970s and 1980s, neither Home Office penal policy, nor Home Office-funded research programmes, foregrounded human rights as a relevant focus for academic inquiry. 5And, even though human rights compliance has become a key consideration for the Home Office following the success of prisoner litigation in both the UK and Strasbourg courts (Livingstone et al. 2003), the contemporary position appears much the same (Hillyard et al. 2004).Indeed, two factors suggest that the influence of research funders may be even more significant today in foreclosing human rights questions: first, 'the vast amount of funding has been directed to positivist/administrative projects that seek answers to crime causation or aim to improve existing apparatuses of crime control' (Walters 2003a: 20-1;2003b;Carlen 2002).Secondly, as Loader's interviews with Home Office personnel highlight, there is both increased politicisation of criminal justice research and reduced official interest in research-based policy-making, especially in relation to 'the politics of security in England and Wales today ' (2006: 23;Zedner 2003).Taken together, these factors do go some way towards explaining why British criminologists have not engaged with human rights.That said, it remains hard to understand why the 1990s criminological turn towards the rise in punishment (policies) did not prompt a companion interest in the growth of rights-based legal constitutionalism in the UK or, more specifically, the impact of the European Court of Human Rights on UK law in relation to matters such as prisoner release dates, disciplinary hearings or prisoner access to legal advice and the courts.
Perhaps (part of) the answer to the omission lies elsewherespecifically, in the relationship between criminology and law.Freeman has pointed out that, historically, human rightsnot just human rights lawwas foreign territory for non-lawyers: 'Before the 1970s almost all academic work done on human rights was done by lawyers, and most articles were published in law journals ' (2002: 78).This blanketing of human rights by lawyers could well have been particularly off-putting for criminologists, especially in light of historic tensions between criminal law and criminology concerning both the institutional origins of criminology, and the perceived negative influence of law and lawyers on criminology's intellectual and political agendas (Finnane 2006: 399-400). 6It is, of course, also the case that wariness of law is both widespread and relatively deep-rooted amongst those outside law; moreover, lawyers tend to nurture or encourage this wariness.Indeed, as Valverde et al. have observed, the commonplace assumption that 'scientists are the only authorized custodians of scientific information, so that if courts use scientific facts there is some kind of obligation to use these facts in a scientific manner', has an obvious parallel in law: [namely,] that lawyers are the ones who authorize or deauthorize the use of legal knowledge resources such as case law and legal doctrine.Lay uses of legal doctrine and other legal resources tend to be dismissed as uninformed and inaccurateas if lawyers owned not only the power to represent clients but also the intellectual machinery of law itself.(2005: 88) The paradox here is that discussion of human rights only started to feature prominently in mainstream legal literature in the UK in the 1990s, following indications from the Labour Party that, if elected, it would enact a charter of rights.
Prior to the Human Rights Act 1998, civil liberties, ECHR jurisprudence and international human rights law were subjects of specialist, not general, interest within both law schools and legal practice (Whitty et al. 2001: 1-18).Relatedly, debate about human rights amongst lawyersthat is, arguing about the pros and cons of rightsbased approaches to law in a broad and diverse manneris of even more recent vintage (e.g., Dembour 2006;Gearty 2004;Murphy and Whitty 2006).Moreover, there is still very little empirical literature within law on the impact of human rights in the UK, both in terms of the initiation of rights claims and the implementation of rights norms (Halliday and Schmidt 2004;Clements and Thomas 2005).
reasons for not mentioning human rights (see Simpson (2004) on the history of the European Convention on Human Rights (ECHR) 1950).
Yet, even if we tally all of these various explanations for the current foci of British criminology, 7 it is still surprising that human rights analysis remains so absent.To put it bluntly, the general shift towards rights-based constitutionalism in the United Kingdom, and the specific impact of human rights law on both prisoner rightsconsciousness and prison governance, needs to be recognised in prison-focused criminology.

Part Two: (Human Rights) Lawyers and Risk
The non-engagement of criminologists with human rights provides one explanation for the absence of a scholarship on the relationship between risk and rights.But, as noted at the outset, there is a second explanation that needs also to be considered: namely, lawyers in the UK haven't been greatly interested in the concept of risk, nor have they made much reference to the extensive risk literature (Steele 2004;Giddens 1999). 8We suspect that, for non-lawyers, these omissions will seem surprising given that law is often about using expert knowledges and managing risk: consider, for example, the use of 'risk of harm' tests in the child protection and mental health contexts, and the use of 'risk of re-offending' criteria in sentencing, parole and civil preventative order contexts.Yet, even in the areas of environmental protection and consumer health and safety law, where the 'precautionary principle' has an established presence in both national and European regulatory approaches, UK legal scholars have done remarkably little work on risk (Feintuck 2005;Chalmers 2005).
Most notably, in those areas of human rights law, such as the absolute prohibition on torture, where explicit dichotomies have been drawn between concepts of rights and risk, analysis also remains scant (Zedner 2005).9As we discuss further below, it is only in the field of regulation studies that lawyers have been involved in generating detailed critical accounts of risk assessment and management, both in public and 7 It is of course also important to break down general claims about 'British' criminology in light of the legal and political histories of the different jurisdictions within the UK, including the different histories of human rights activism (see McEvoy and Ellison (2003) in relation to Northern Ireland).
8 But see O'Malley (2004) for an historical account of the role of risk, and the related concept of uncertainty, in the development of contract, tort and insurance law.
private sector organisations (e.g., Fisher 2000;Hood et al 2004).To begin with, though, we consider if there are any general characteristics of the risk literature that could help to explain the typical stance of British lawyers towards risk.
One feature of the scholarship on risk that might be off-putting to lawyers is the range and complexity of risk theorising and the tendency to not differentiate sharply between perspectives.As Garland has argued, 'the risk literature' is in fact 'several distinct literatures, involving different projects, different forms of inquiry, and different conceptions of their subject matter, all linked tenuously together by a tantalizing four-letter word ' (2003: 49).This may present a particular challenge for British lawyers: the long-standing pragmatic ethos and practitioner focus of British lawyers (Twining 1994;Cownie 2004) is arguably ill-suited to a risk literature wherein 'myriad versions of social theory [are] operating under the conceptual umbrella of risk' (Chan and Rigakos 2002: 744).Indeed we suspect that a survey of the law school curriculum would find practically no reference to the three main critical groupings in the risk literaturenamely, the 'risk society' perspective associated with Giddens (1991) and Beck (1992Beck ( , 2002)); the 'governmentality' perspective which draws upon Foucault's writings and views risk as a complex disciplinary tool, operating through both coercion and voluntary compliance (e.g., Rose 1999); and the socio-cultural perspective on risk associated with Douglas (1992) and Sparks (2000a;2000b;2001), which insists that culture is a crucial factor in any social group's perceptions of risk and its politicisation.
A second factor explaining lawyers' non-engagement with the risk literature is that practices of 'risk assessment' are generally represented as requiring expert scientific processes of judgement.This representation can be traced to the techno-scientific focus of the fourth main grouping in the risk literature.That grouping tends to be clustered in disciplines such as medicine, economics and engineering: its hallmark is that it treats risk as a 'taken-for-granted objective phenomenon' (Lupton 1999: 2) and it aims to identify, map, predict and regulate different types of risk according to expert criteria.The key significance of this approach is that it generates a strong sense of risk as a non-legal knowledge which, in terms of legal practice, has important consequences.To put it crudely, although it is routine for practising lawyers and judges to work with expert evidence, the representation of risk as a scientific measurement means that, amongst academic lawyers, the critique of risk may be seen as belonging more properly to other disciplinessuch as economics, chemistry or psychiatry.Thus, evidence law scholars, for example, may scrutinise the assembly and uses of expert evidence, especially its reliability in terms of the evidentiary rules of proof, but they very rarely raise questions concerning the contingency and uncertainty of scientific knowledge (Jasanoff 1995).This phenomenon, furthermore, is compounded by the fact that the professional status of lawyers, and the adversarial nature of litigation, has historically resulted in mutually-reinforcing associations with certain expert professions (most obviously, medicine).
A third factor which may explain the pattern of legal scholarship on risk is that the political repercussions of the relationship between different perspectives on risk and different governing approaches remains relatively unexplored in both risk and regulation literatures.A belief in scientific definitions of risk and the importance of technical expertise will tend, for example, to lead to the claim that regulatory priorities ought to be established with reference to technical criteria.By contrast, an understanding of risks as socially constructed will lead to the rather different view that 'regulatory priorities and policies cannot be left to the "objective" evaluations of experts but have to emerge from democratically legitimate processes of debate and consultation' (Baldwin and Cave 1999: 142;Kahan et al. 2006).Some lawyers have engaged with the legal and political implications of risk-based regulatory approaches (e.g., Baldwin 2005;Black 2005;Feintuck 2005;Hood et al. 2004).But, in general, public lawyers have been very slow to respond to questions about whether and how we are 'governed in the name of risk' (O'Malley 2000: 458).Despite the seismic changes in the nature of UK state power in recent years, public law textbooks generally ignore the risk literature.Instead, the focus remains on 'meta-constitutional principles such as the rule of law and separation of powers' (Fisher 2003: 472) and now, most visibly, on the pros and cons of rights-based adjudication by courts.
Significantly, the work of those lawyers who have generated a regulation scholarship, exploring the ways in which regulatory techniques are now used to assess and control risks in both the public and private sectors, continues to be viewed largely as a specialist subset of public law.But a shift could occur now that several regulation scholars have begun to examine the values, including the protection of human rights, that underpin different models of regulation (Brownsword 2004;Prosser 2006).
Up to this point we have spoken either of UK lawyers in general or of public lawyers in particular.But, as emphasised in the Introduction, our particular interest is in human rights lawyers.Why is it that risk has not registered as a concept of general critical concern amongst human rights lawyers in the UK? Human rights law is replete with co-existences of risk and rights: looking specifically at the areas of detention and imprisonment, consider for example the fact that the statutory duty of parole boards to identify risk factors and make predictive judgments about the release of prisoners must be performed in light of the legal obligation not to act contrary to Convention rights under the Human Rights Act 1998, notably Article 5 ECHR right to liberty standards (Padfield 2006).Similarly, where a person in detention is considered at risk from other prisoners, or is a known suicide risk, the procedures for assessing It could, of course, be argued that these (and many other) contemporary human rights contexts demonstrate that the reason why human rights lawyers in the UK have not focused on the relationship between risk and rights is that, as outlined above, risk is widely viewed as a non-legal knowledge: it is, in other words, viewed as just another species of expert evidence.Support for this argument can be found, for example, in the case-law on prisoner challenges to parole board decisions.This jurisprudence demonstrates that there is a longstanding appellate court reluctance to become involved in adjudicating on expert opinion in relation to risk assessments of prisoners: as one judge put it, '[i]t is not for the court to second-guess the judgment of a specialist tribunal' (Watson 1986: 916; see also McLean 2005).There is, however, one notable (and historically-unprecedented) exception to this general trend: the post-9/11 national security context.In a series of judgments, most controversially A v Secretary of State for the Home Department (the 'Belmarsh Detainees' case) (2004), senior UK judges have drawn upon human rights jurisprudence (notably, the principles of proportionality and equality) to condemn the use of executive powers to detain alleged terrorist suspects indefinitely (Arden 2005), or to impose 18-hour curfew orders (JJ 2006).In reaching these decisions, the judiciary has had to directly confront national-security agencies' assessments of the level of terrorist risk and the necessity of particular counter-terrorism measures (Feldman 2006).
Future challenges over detainees' rights could place these judicial understandings of the risk/rights relationship in an even more intense spotlight.For example, there is an ongoing political debate about the European Court of Human Rights judgment in Chahal v UK (1996), 10 which reinforced the absolute nature of the international ban on torture.Current UK government policy of using 'diplomatic assurances' to justify deporting foreign terrorist suspects to countries which practice torture may result, it has been alleged, in new legislation requiring courts to balance the risk of torture against the risk to national security (Human Rights Watch 2006).
These stark conflicts between concepts of risk and rights have led to claims that human rights law and principles must play a normative role in risk assessment and management.So, for example, Hudson has argued that the way to avoid 'rushing into no-holds-barred risk control' is to encourage 'a whole-hearted embrace of the ideas of human rights: not just … the Human Rights Act, but also embrace of a rights culture' (2001: 110).Zedner also argues for 'adherence to legalism': 'To the extent that riskbased measures threaten individual liberty, it is all the more important that they be bound by legal strictures enshrining basic values such as equality, fairness, and the preservation of basic human rights ' (2006: 425).In the final part of the article below, we draw from scholarship within criminology and law to argue that the combined operation of risk and rights discourses may however be more complex and unpredictable than these commentaries appear to suggest.In particular, we take issue with the apparent dichotomy between risk and rights knowledges.

Part Three: The Co-Existence of Risks and Rights in Prison Governance
Our general argument so far has been that risk and human rights have become dominant discourses in the UK and are likely to remain so.In the area of prisons and prisoners, these are predominant discourses: hence in what follows we argue for a new stream of academic enquiry that focuses on how co-existences of risk and rights are affecting UK prison governance.In making this proposal, we have been strongly influenced by a number of scholars.One of these is Pat O'Malley: as noted at the outset, he has appealed for 'more nuanced analysis of the ways in which, almost everywhere that risk appears, it is assembled into complex configurations with other technologies… '(2004: 26-7).Another key influence has been Richard Sparks' work on risk.This work does not engage directly with risk and rights; its value lies in the fact that it calls for enquiry into how risk-based knowledges and practices intersect with other structuring principles of penal systems and penal politics.Sparks argues that '[w]hatever else we may say about contemporary penality and its associated politics, it seems clear that the discourse and practice of risk management do not have the field all to themselves'.Then he asks the following important question; '[b]ut on what terms do they co-habit with the existing occupants of that terrain?' (2000b: 129-30).Drawing on Mary Douglas' work on risk, Sparks goes on to emphasise two further points: first, the concept of risk is inherently plural and contingent, and we should expect risk discourse to be a mixed discourse and to work out differently in different contexts.Second, 'risk' is now a cultural key-word for holding people accountable, with the result that risk controversies expose questions concerning the competence and legitimacy of decision-makers (2001: 168-69).In explaining these points, Sparks notes that: To suggest that some constructions of risk in the penal realm have been unduly singular and one-dimensional is also to say that they have neglected what is most interesting about it, namely that like the language of rights, justice and legitimacy, with which it so closely intersects, it is a site of struggles for influence, credibility and recognition (2001: 162, our emphasis).
We find these observations especially useful.Positioning risk and human rights as 'closely intersecting', Sparks notes obvious parallels between the language of rights and constructions of risk in penal policy and practice: both are sites of 'struggles for influence, credibility and recognition'.This resonates with findings within regulation scholarship that emphasise the ways in which risk management can be a cover for disputes over institutional legitimacy and competence (Black 2005;Hood et al. 2004).
More generally, the explicit linkage of risk with questions of rights, justice and legitimacy also provides a useful reminder of the need for criminology to 'embark on the reconnection of penological research with normative moral and political reflection' (Sparks 2001: 172). 11  In what follows, we aim to build on these observations.We argue for increased recognition of different forms of risk and rights knowledges, and of the fact that these operate in different institutional and cultural settings, and can become combined or co-exist in unexpected ways.We identify four overlapping themes, drawn from a range of literature on risk and on rights, which we see as particularly pertinent for future research in this area.These are: the social construction of risk; the diversity of legal cultures; the nature of legal knowledges; and the 'framing' of risk and rights compliance.

The social construction of risk
As outlined earlier, many accounts of risk suggest that risk assessment and management do not involve socio-political choices.This representation is reinforced in legal literature when there is an uncritical acceptance of law's historical and mutually-reinforcing associations with certain other expert practices and knowledges (such as medicine and psychiatry).In order to challenge this mindset in the prison governance context, greater attention needs to be paid to those risk analyses that do recognise the variability of risk technologies and knowledges (for example, in relation to gender (Chan and Rigakos 2002;Hannah-Moffat 2004, 2005)).This type of risk scholarship not only provides examples from very diverse fields of how risks are socially constructed (e.g., Lupton 1999;Levi 2000;Valverde et al. 2005;Hudson and Bramhall 2005;Mythen and Walklate 2006), it also highlights the importance of scrutinising the 'ways in which risks are communicated and how they are politicised' (Sparks 2000a: 132;Malloch and Stanley 2005;Thirlaway and Heggs 2005.Furthermore, as Moore and Valverde have pointed out, the use of 'rational' criteria (such as statistical correlations or previous convictions) as techniques of risk governance means that there may be even more reason to investigate how 'myths, symbols and non-rational fears can also be shown to shape the "risks" in question' (2000: 515).Accounts of risk which privilege expert knowledges and quantifiable formats may mislead not only as to the character of the information used, they may also be misleading in relation to the methods of risk assessment: for example, 'risk calculations and predictions are in fact often carried out by non-scientific personnel using very subjective tools ' (2000: 521).

The diversity of legal cultures
Secondly, future research on risk and rights needs to pay particular attention to the diversity of legal cultures (Nelken 2004).This could begin by looking, on the one hand, at differences in rights cultures and, on the other, at differences in risk regulation regimes: thereafter it should be easier to build accounts of intersections between risk and rights.Differences in rights cultures emerge in part because of the distinctiveness of legal-professional cultures and also because human rights adjudication operates at intersecting levels (devolved, UK and European).This means that rights norms can be interpreted and articulated within, and across, distinct legal and political orders in ways that may lead to very different outcomes (Morison and Lynch 2007).The post-devolution Scottish legal context provides a good illustration: The diversity found in legal cultures of rights seems to have a counterpart in risk regulation.So, for example, research on the approaches of US and European agencies to risk management has highlighted substantial differences in regulatory style: Operating in a fishbowl of transparency, with significantly less protection from civil service traditions or legal insulation than their European counterparts, American regulators were not free to justify their actions by simply invoking delegated authority or superior expertise; they had to establish through explicit, principled argument that their actions fell within a zone of demonstrable rationality.Numerical assessments of risks, costs, and benefits provided compelling evidence.European regulators, by contrast, seemed generally better able to support their decisions in qualitative, even subjective terms.Expert judgment carried weight in and of itself as a basis for action, the more so when backed by negotiations among relevant parties; there was on the whole less need to refer to an exogenous method, model, or logic to support policy decisions.(Jasanoff 2005: 18) In order to investigate the validity of these claims in the prisons context, it would be necessary first to identify the nature of governance relationships and, second, the extent to which organisational risk awareness, management and compliance actually exist in practice.These are likely to be complex tasks.For example, for contracted-out prisons in England and Wales, the network of accountability mechanisms includes a range of external institutional actors who can influence the assessment of organisational risks: courts, National Audit Office, Prisons Inspectorate, Prisons Ombudsman, Independent Monitoring Board, and 'an on-site regulator (called a controller), appointed by the Prison Service to monitor compliance with contract specification' (Scott 2000: 53).Moreover, the assessment of organisational risk can be complicated by the involvement of some of these actors in prison litigation: 'the development of litigation strategies has been both supportive of and supported by the work of the prisons humanity regulators, and notably the inspectorate and the ombudsman, the regulators providing better information which may be used in litigation, litigation providing more robust definitions of appropriate norms relating to the treatment of individual prisoners ' (2000: 56).Of course, in addition to a focus on the range of external sources affecting risk management, what happens internally within organisational cultures is also key: especially, as Hutter points out, when it is 'not at all clear how many government departments have bought into risk based initiatives or to what extent ' (2006: 220).A relevant example here is the case of Napier (2005), where the apparent prioritisation of 'business risk' assessment in the publications of the Scottish public sector (e.g., Scottish Executive Public Finance Manual 2004) did not seem to be matched in the decisions of the Scottish Ministers, the Law Officers and the Scottish Prison Service to 'run the risks' of suspending the introduction of new sanitation facilities in Barlinnie Prison (Murphy and Whitty 2007).

The nature of legal knowledges
This brings us to our third theme.The dominance of a static model of law means that the 'dynamics of knowledge production and circulation' (Valverde et al. 2005: 87) amongst legal networks and actors can be neglected.To remedy this, questions need to be asked about which actors use particular knowledges or combinations of knowledges; when and where these knowledges are used; and with what 'legal, social and epistemological effects?' (2005: 87;Valverde 2003;Moore and Valverde 2000).
Multi-agency public protection panels in England and Wales provide a good example of the hybrid quality of risk assessment and management knowledges, and the variability in actual organisational forms and practices (Kemshall and Maguire 2001).
These panels (bringing together police, probation, prison, social services and other agencies) are designed to provide a forum for the exchange of confidential information about sexual and violent offenders, and a process for the classification, implementation and monitoring of individual 'risk management plans.'The key point for present purposes is that their construction of risk does not depend on a single distinct knowledge base: typically it will combine a mixture of actuarial, clinical, professional and common sense views, while giving a low priority to considerations of offenders' rights: The character of the risk assessment debates … was often anything but 'scientific' or 'technology-driven'.On the contrary, researchers noted that many discussions were unstructured, even rambling, and that close attention was paid to the views and 'instincts' of members who knew the offender in question, even if unsupported by hard evidence.It was not unusual for panels to revise instrument-derived risk classifications, in essence backing their abilitythrough a combination of 'gut feelings' and professional experienceto make a better prediction than one based purely upon actuarial risk. 13(ibid: 248) Similar conclusions have been drawn about the use of risk knowledges in courts.So, for example, Valverde et al. (2005: 87) have argued that, although 'literature on risk and law tends to counterpoise expert knowledge to law and legal reasoning', it is more 'useful to not assume that everything that goes in as 'expert witness testimony' is epistemologically homogenous ("science" or "expertise")'.They argue that the focus should be on the different types of knowledges that operate within both legal and pre-legal processes, and how and why such knowledges come to be categorised as legal/non-legal, expert/everyday, or some other hybrid form.In their case study on the ways in which courts in New Jersey have 'translated' expert assessments of risk of reoffending under Megan's Lawa community notification statute authorising public access to information about the identity of convicted sex offenders who are considered to present a risk of re-offendingthey identify the ways in which knowledges migrate between legally-trained personnel and extra-legal professionals.
One example of this phenomenon (they call it 'swapping knowledges') is the prosecutorial and judicial practice of relying on reports of acts, or alleged acts, that have not been the subject of a conviction in order to determine the risk level of particular individuals.'[D]espite privileging legal knowledge in the risk assessment process', courts 'then rely on non-legal paradigms to renegotiate legal practices' (ibid: 106).Hence: The judge may rely on documentation he or she considers relevant and trustworthy in making a determination … This may include but is not limited to criminal complaints not the subject of a conviction but which are supported by credible evidence, victim statements, admissions by the registrant, police reports, medical, psychological or psychiatric reports, pre-sentencing reports and Department of Corrections discharge summaries. 14e would argue that two general conclusions can be drawn from these empirical examinations.First, we should be sceptical of monolithic characterisations of (risk and rights) knowledges and, secondly, it should be expected that different types of knowledge will circulate and migrate between different actors (organisational, expert, judicial, lay, etc) in dynamic and unpredictable ways.In other words, if judges, psychiatrists, probation officers or prison staff draw upon the concepts of risk and rights as part of a (pre-) legal process, we should expect some type of mixed discourse and the character and effects of this discourse cannot be assumed in advance.

The 'framing' of risk and rights compliance
Our fourth and final theme concerns how organisations 'frame' questions of compliance with regard to risk-based and rights-based obligations or expectations. 15his question is complicated by the fact that regulation of the prison sector is not based exclusively on traditional legal sources (statutes, caselaw) and institutions (government departments, courts), even though legally-enforceable norms often receive privileged emphasis.Public sector regulatory obligations derive from both legal and non-legal rules and standards (for example, UK Treasury guidelines or Home Office circulars), as well as the demands of a range of regulation, audit, inspection and grievance-handling agencies and actors (Scott 2000;Black 2004).
Additionally, private organisations such as firms and NGOs have an increasing role in public sector governance, both in the creation and the implementation of policy (for example, private security firm provision) (Scott 2002).
Another complicating factor is the lack of research addressing the relationship between risk-based and rights-based regulatory demands onto use the terminology of the Human Rights Act 1998 -'public authorities' in the UK.Regulation scholarship, which has examined risk management in the public and private sectors, has not yet engaged with the growth of rights awareness or litigation.And socio-legal scholarship, which has started to examine human rights compliance practices, has not addressed issues of risk-based governance.What is needed are accounts of whether, and how, 'legal risk' is being framed at the different levels of the prison sector in light of both legal (rights compliance) and non-legal (organisational risk compliance) demands. 16n developing these accounts, the scholarship produced by public lawyers who have looked at the social and legal effects of risk-based regulation and, in particular, at how risk discourses operate within particular organisational cultures will be of central relevance (Fisher 2003;Black 2005;Feintuck 2005).Its importance lies in explaining how different types of organisational risk (such as financial or legal risk) are generated and must be 'managed' according to various regulatory models and goals.
While there is still very limited research on how organisations actually respond to governance by legal and non-legal rules, and the relationship between these sets of rules (Parker et al. 2004;Baldwin 2004), it is obvious that the prison sector has to respond to the twin requirements of organisational risk management and compliance with rights norms (Fisher 2003;Foster 2005;Murphy and Whitty 2007).
Empirical studies of the prison sector are required to develop an account of framing.
Here we make two preliminary observations.First, risk and/or rights issues will be 'framed' differently in different contexts, and political culture influences both normative and scientific discourses (Jasanoff 2005).For example, accounts of regulatory governance demonstrate that 'organisations adopt structures and follow procedures not just, or not even, to achieve goals, but to gain legitimacy in the widest sense' (Black 2005: 19).And this emphasis on legitimacy has important political effects: The rhetoric of 'risk management' and 'risk-based' approaches combines a sense of strategy and control in a way which is politically compelling; moreover, framing one's actions as 'risk-based' is, in the current climate, a useful legitimating device.But the framing of the regulatory task in terms of risk has the potential to have more than a rhetorical effect: it imports particular conceptions of the problem at hand, and leads to the framing of a solution in a particular way.(ibid) Secondly, 'risk colonization'whereby 'risk increasingly comes to define the object, methods, and rationale of regulation' (Rothstein et al. 2006: 93) has implications for attitudes towards compliance.For example, organisational cultures may turn towards 'defensive compliance' -'actors think, act, and communicate within the four-square corners of risk classification schemes and internal procedures, and they avoid making hard decisions and expressing opinions that are more honest' (Ericson 2006: 352) or viewed as 'political' (Feintuck 2005: 388;Fisher 2003).17 A second branch of scholarship of relevance in understanding 'framing' is that which examines the extent of human rights compliance.The literature on public sector awareness of Human Rights Act norms portrays very variable patterns of legal knowledge, political commitment, exercise of professional judgment, allocation of legal risk posed by prisoners as other litigants (e.g.NGOs) are possible: furthermore, an organisation's assessment of legal risk will also have to take account of operational, financial and reputational risks.required resources and fear of sanction (Clements and Morris 2004;Clements andThomas 2005: DCA 2006). 18It illustrates that, in some contexts, human rights law will be viewed predominantly as a legal risk and hence a technical problem to be managed, rather than a source of normative values.In other words, the official commitment to promoting a 'human rights culture' in the UK is given a multitude of meanings within the public sector, with ongoing disagreement over whether greater attention should be devoted to enforcing legal or non-legal compliance methods.The perspective of the prisoner or pressure group (potential) litigant is different: the Human Rights Act and other rights norms tend to be interpreted uniformly, as a key symbolic, if not always practical, resource for challenging prison standards and administrative decisions (Valier 2004;McEvoy 2001).Lastly, and often of greatest significance in assessing the actual impact of rights norms, close attention needs to be paid to socio-cultural representations and understandings of 'deserving' and 'undeserving' rights-holders.For example, as Kemshall and Maguire have noted, there is an accepted devaluing of the rights of some groups, such as (suspect) sexual offenders: The notion of offenders' rights took low priority in the thinking of [public protection] panel members; concern was rarely expressed about possible violations of rights to privacy, or of the fundamental distinction between those who are under statutory supervision or control (e.g. on probation or conditional release licence) and those who are not (including those merely suspected of offending).Similarly, although police officers recognized that they had no right to enter the homes even of registered sex offenders, they often deliberately gave them the impression that they did have such rights.

Conclusion
Our main purpose in this article has been to draw attention to cleavages in scholarship on risk and on rights.UK prison governance has become enveloped by discourses of risk and of rights: what is little understood, however, is how these discourses are interacting.In suggesting that there is a need for an analysis of how these discourses co-exist, we have advocated that both criminologists and lawyers engage more fully with scholarship which recognises the social construction of both risks and rights, investigates public sector regulatory models, and pays closer attention to the apparent mobility and hybrid quality of legal knowledges.We have also argued for greater interdisciplinarity in order to be able more accurately to describe the current prison landscape. 19Taken together, these measures may help to guard against the threat of 'an irresistible logic of risk' (O'Malley 2004: 150) and, in some of the legal literature, what we would term a parallel 'logic of rights'.As we hope to have demonstrated, the relationship between risk and rights is far more complex and unpredictable than these logics suggest. 19Wider questions such as when concepts of risk should be defended or risk management practices deployed for progressive ends (O'Malley 2004: 8;Maurutto and Hannah-Moffat 2006: 451;Rothstein et al. 2006;Zinger 2004) also remain to be explored, especially when considered in the context of combined risk and rights discourses.
and managing that risk must be Convention-compliant with Article 2 right to life standards (Edwards 2002; Keenan 2001; Van Colle 2006).Asylum and deportation case-law provides another striking example: it foregrounds tests to assess risk of torture (Mamatkulov 2005), or risk of threat to mental health (J 2005; Tozlukaya 2006), flowing from the Article 3 ECHR prohibition against torture and inhuman or degrading treatment.In short, references to risk are now almost commonplace in core areas of human rights jurisprudence.
several of the leading cases challenging Scottish Executive action under the Scotland Act 1998 have involved prisoner litigants.To an unexpected degree, prisoners' rights claims have been centre stage as judges develop a new Scottish legal constitutionalism (see generally O'Neill 2006).Of course, this discourse has also generated strong anti-prisoner rights sentiments within the Scottish Parliament and amongst the wider public on the ground that prisoners are availing of 'special rights'.Another illustration of this point about differences in rights cultures is provided by the fact that the Scottish and Northern Irish courts have opted for different resolutions of human rights-based claims for damages arising from 'slopping out' and unsanitary prison conditions (see, respectively, Napier 2005 and Martin 2006).Both courts found violations of Convention rights but, in Martin, the court cited the public interest as a reason for not awarding any damages: Having regard to the wider public who have an interest in the continued funding of a public service one cannot lose sight of the financial consequences of even a modest award … to the large number of prisoners going through the prison system at Magilligan [Prison]. 12

18
An example of the human rights training provided to Northern Irish prison staff is described in Martin (2006: para 22), where prison conditions were found to be in violation of Convention standards: 'The powerpoint documentation used at the training showed a general education on the implications of the 1998 [Human Rights] Act and the Convention.The [Northern Ireland] Prison Service witnesses, however, revealed a somewhat cursory and unparticularised knowledge of the Convention.'