Reach and relevance of prison research

In this contribution I reflect on the changes in the penal landscape and how they impact on prison research. I do this from my experiences as a prison researcher in a variety of roles, in both Europe and Australia. The growing dominance of managerialism has impacted on both corrective services and universities, in ways that have changed the relationship between current prison practices and academically oriented research. Therefore, academics have to question how their contemporary prison research can bridge the emerging gap: how they can not only produce research that adheres to the roots of criminology and provides a base for a rational penal policy, but also how they can develop strategies to get recognition of and funding for this broader contextual work which, although it might not produce results that are immediately identifiable, can be of relevance in indirect ways and in the longer term.


Introduction
If the integrity and the quality of the research are good enough, and the sense of timeliness and social relevance is acute enough, the criminologist can influence policy whatever his formal position in the overall scheme of things and however he chooses to bring his work to the public voice. (Harding 2003: 483) Prison populations have been a focus of my research for over two decades. Over time, I have seen the penal landscape change, which is reflected in two main developments that have informed the content of this special issue and my contribution to it. Firstly, prison populations have grown, which has led to a boom in research analysing this phenomenon from the outside. Secondly, and at the same time, conducting (qualitative) prison research from the inside is increasingly facing challenges. While it would be incorrect to state that the heydays of penological research are over -a growing body of prison research, including the contributions to this special issue, contradicts that -a main thread throughout these writings concerns the difficulties prison researchers now face: more hurdles to be overcome and more restrictions in terms of access and dissemination of research. These difficulties might be most acute in the IJCJ&SD 5 Online version via www.crimejusticejournal.com © 2015 4(1) United States (US) where, as Reiter (2014) describes, there is a huge discrepancy between the growing prison population and the scarce research investigating the impact this has inside prisons. The times when sociologists and criminologists were allowed to conduct in-depth research inside prisons for extended periods of time, resulting in the classic readings by Clemmer (1940), Sykes (1958), Irwin (1970), Jacobs (1977) and many more, are gone. This led Simon (2000) and later Wacquant (2002) to plead for more American qualitative prison research. According to them, after the collapse of the rehabilitative ideal in the US, the focus of prison research turned away from the sociology and criminology of prison life to the more quantitative analysis of crime trends, prison populations and their underlying (societal) causes. The situation in Europe, including the UK, is different from the US, with qualitative prison researchers producing new scholarship. The need for in-depth, longitudinal and qualitative prison research -measuring the 'quality of prison life' in different prisons within one country (Liebling 2004), comparing prison conditions and regime aspects in several jurisdictions (Pratt 2008a(Pratt , 2008bPratt and Eriksson 2013), or re-assessing the contemporary 'pains of imprisonment' including the 'depth', 'weight', 'tightness' and 'breadth' of imprisonment (Crewe 2011(Crewe , 2015 -has been argued by many. New trends have developed, including a rapid growth of the use of ethnographic and narrative methodology, and a growing interest in the emotive aspect of (ethnographic) prison research (see Jewkes 2012Jewkes , 2014Liebling 1999;Scheirs and Nuytiens 2013). However, from their writings derive other issues that also urge us to think about the aims and purposes of contemporary prison research, such as those developed in the contributions to this special issue.
In this contribution I reflect on the changes in the penal landscape and how they impact on prison research. I do this from my experiences as a prison researcher in a variety of roles, in both Europe and Australia, which has provided me with a broad perspective on the interaction between prisons and research. It has been an interesting journey during which I have been confronted by dilemmas on how my chosen profession might be of relevance to both corrective services and prison research. The purpose of this article is to stimulate the dialogue between the research world and prison practice, in the hope of contributing to a constructive and valuable research environment. As it is a rather personal account, I follow the trend set by other articles on this topic (see Ugelvik 2014), replacing the traditional third person with the personal pronoun and using my own experience as a 'research tool'. I begin with the 'bigger picture' of a changing penal landscape and then move on to discuss how this impacted on both the world of corrections and academia. Following this environmental sketch, I focus on the different roles prison researchers can fulfil in this setting, describing possible pitfalls and how to deal with them. Finally, I argue why I think researchers need prisons, and why prisons need researchers.

The changing penal landscape
As a point of departure for my description of the changes in the penal landscape, I start from the concept of 'punitiveness'. One aspect of this concept that has been widely and deeply discussed (however, still contested) is the trend of increasing prison populations in most western societies. The picture doesn't look as bleak as initially suggested, with prison populations actually going down in some European countries in recent years (Netherlands, Germany, Sweden, Finland), 2 and also in several states of the US for three consecutive years (2009)(2010)(2011)(2012). 3 Nevertheless, in comparison to the 'progressive 1970s', overall growth rates in prison populations have persisted in more recent decades. Since this growth trend started in the early 1980s, an expanding body of literature has analysed the underlying reasons for this development, either within one country, or as comparative studies between countries. Evidence from these studies has revealed that broader socio-economic developments impact on penal policies. There was the waning belief in the welfare state with this model losing credibility as crime rates increased in the 1970s, despite the investment in welfare provisions. The welfare model was increasingly criticised for actually making things worse, causing unacceptable discrimination, promoting dependency and reducing individual responsibility (Downes and IJCJ&SD 6 Online version via www.crimejusticejournal.com © 2015 4(1) Hansen 2006). This disenchantment, together with the global economic recession of the 1970s, led to a new economic model labelled neo-liberalism, with advocates pleading for free markets, privatisation, deregulation and welfare state retrenchment (Brown 2011). Growing neoliberalism created societies that became more exclusionary, moving from the view of crime as a societal problem that needs to be addressed by supporting people and trying to rehabilitate them, to one premised on individual responsibility, for which more severe punishment is the appropriate answer, with incapacitation and retribution as primary aims. Increasing prison populations not only brought the topic to the attention of (comparative) criminologists but also put it on the political agenda, and made it a leading matter in electoral discussions and campaigns. This was exemplified by increasing discourse on what has been called a 'law and order' approach where politicians from both the left and the right argued for harsher punishment and supported forms of 'expressive justice' (Garland 2001), claiming that the use of imprisonment was legitimate and that 'prison works'. The law and order discourse became most apparent in majoritarian democracies with bi-partisan political systems, and has been linked to their electoral structures, bureaucratic organisation, and their relationship with the judiciary, the media, and the perceived punitive public opinion (for an overview of this literature, see Tubex 2014).
From a cynical point of view, one could consider this a positive development for prison research; it creates more issues to investigate, and the matter becomes more complex as growing prison populations spawn new generations of problems, such as prison overcrowding, impoverishment of regimes, changes of relationships within prisons, and concerns about human rights violations. But the context as described here did not impact solely on penal policy. Neoliberalism has not only been associated with growing punitiveness; it has also been linked to managerialism, and this has impacted on both aspects of our research field: the world of corrective services and that of universities. In the following paragraphs, I explore more deeply these two developing trends, firstly by considering how correctional managerialism has impacted on prison practice, and secondly by turning to the academic side of the coin, discussing how universities relate to research in a managerial environment.

Managerialism and corrections
In his account of the major trends that have impacted on the criminal justice system over the last few decades, Freiberg (2005) claims that the development of managerialism, or new public management, has had 'the most powerful and influential impact on public administration generally, and criminal justice in particular' (Freiberg 2005: 12), although he considers it has remained 'relatively unremarked' in criminology.
Managerialism stemmed from the private sector and was introduced in the public sector in the 1980s and 1990s within the broader context of modernising government agencies. As a consequence of its infiltration, criminal justice agencies need to demonstrate effectiveness in their outcomes, and prison management has become a 'business', illustrated in the growth of private prisons. To this end, their approach needs to be 'evidence-based': corrections agencies are forced into measures of transparency and accountability, and prison performance has to be evidenced in the meeting of key performance indicators that can be quantitatively measured over time and benchmarked.
We agree with Freiberg (2005) that managerialism is not, per se, a bad thing. It is preferable that the resources available to prison managers focus on achieving well-articulated penal policy aims. The problem arises when managerial goals appear to eclipse penological objectives, with the spotlight shifting to the mechanics of operational systems to the exclusion of the careful articulation of the penal aims they attempt to achieve, and objectives become cast in terms of easily measurable system variables. This is a trend that has been exacerbated in many jurisdictions by a master objective to reduce government expenditure. More recently, Cunneen et al. (2013) considered that the influence of managerialism on the criminal justice system is significantly underrated in criminology and, more particularly, that this style has impacted on the level of human interactions within corrections. They see correctional managerialism as a product of neo-liberal governmental logics, reducing the 'welfarist rehabilitation ideal' to a system that overemphasises re-offending rates and promotes the rise of risk thinking. This risk thinking developed as a double-edged sword: on the one hand, it aimed for a more rational risk analysis but, on the other, it engendered 'a justification for an irrational and risk-averse decision-making' (Cunneen et al. 2013: 70). This risk thinking was not only centred on the offender but spread to the possible risks to the agency in the interpretation of their duty of care, and prompted corrections to feeling pressured to reduce the risk that something might happen. Prison in particular then becomes the solution for those considered 'risky, dangerous or simply unpleasant characters' (Cunneen et al. 2013: 76), with a predominantly warehousing function. Both Freiberg and Cunneen et al. point to the link between the rise of managerialism/risk thinking and a politicisation of public services.
This development has been associated with a number of unwelcome outcomes. When it comes to offenders, the focus moved away from rehabilitation and problem solving to performance management, order management and short term objectives, in which the perpetrator becomes more an object of treatment rather than a subject with which the practitioner enters into a therapeutic relationship (Freiberg 2005).
For staff working in a managerial context, this change resulted in a completely different relationship between staff, their clients and colleagues. When working within a managerial environment, staff have reported feeling evaluated rather than valued, driven to focus on compliance rather than outcomes, and suffering a decline in morale, which in turn impacts on their motivation (Freiberg 2005).
Therefore, Freiberg (2005) acknowledges that managerialism is open to some serious criticisms: 'it is not an end it itself: it should be there for a purpose. A process should be efficient, economic and effective, but only to achieve a further goal…' (Freiberg 2005: 32). In other words, as the Inspector of Custodial Services in Western Australia pointed out in one of his presentations on this topic, corrective services are profoundly human businesses, and some things are just not open to metrics and key performance indicator-type measurement. This is because the aim of accountability and transparency is improvement, not just measurement. Instead the focus should be on outcomes, on what you are trying to achieve and on the longer term goals (Morgan 2013).
But there are also consequences related to research in a managerial environment. The whole performance-related approach puts a lot of pressure on corrections agencies by consuming a great deal of their research potential and, importantly, impacting on the kind of 'evidence' they are interested in. The main interest of prisons as a business has become tailored around the concept of 'what works', demanding evaluations that measure the effectiveness of programmatic interventions and therefore with a mainly quantitative focus, this to the detriment of thorough qualitative research, also in a comparative perspective. This development, in a context of shrinking budgets and subject to the pressure of an unforgiving media and a public assumed to be strongly punitive, has made government agencies more risk averse towards potentially negative research findings. Prison policies are driven by not wanting to 'end up on the front page of the local newspaper' and by 'what makes them look good'. Therefore a situation has been created in which government agencies are already committed to many forms of reporting and where they feel increasingly vulnerable to exposure of their weaknesses, even for factors that are beyond their control. This situation carries the risks that statutory agencies and academic researchers might drift apart, jeopardising both the future of prison research and the evidence base of penal policy.

Managerialism and universities
Following from the impact of managerialism on corrections, one can see a parallel and equivalent trend in universities, particularly in the English speaking world. Universities these days are also subjected to the private sector motto of competition and profitability. The ranking of universities is becoming paramount, and climbing up the rankings becomes the primary goal and initiator of universities strategic planning. Measurable outputs are a matter of survival because the university's budget depends on it. While there is agreement within academia that international rankings are not an objective discipline and tend to discriminate against human sciences, they are the tools we currently have to work with. This puts the researcher in a dilemma of cost-efficiency: where to invest ones time? The highest goal -and the expectation of university leaders -is for researchers to publish in highly ranked journals. However, in the area of expertise we are referring to, there are few journals within these rankings where our writings can find a home. Consequently, supply greatly exceeds the capacity to publish, with long waiting lists a consequence. This time lapse can reduce the relevance of the research findings as they may require timely reporting. Further, in aiming for these highly ranked journals, there is the risk we are only writing for ourselves and hence end up in 'navel-gazing' (Loader and Sparks 2011: 27). Very few practitioners will ever glance over these journals or might not even have the economic resources to access them. 4 Moreover, this goal of aiming for the highly ranked journals endangers the value of publishing in ones that are of a more practical nature with the attendant risk of losing contact with people working in the corrections field. Reporting research findings in publications intended for an audience of non-academic practitioners should not be seen as a hobby for researchers to pursue in their own time as currently appears to be the case.
Further, there is the essential matter of funding. In the Australian context, the main funding source after the Federal government changed the funding base for universities in the 1980s (see Israel 2000) is the Australian Research Council (ARC). In general, the major grants are Discovery and Linkage grants, and Future Fellowships, with the latter being rarely awarded in criminology. Alternatively, there are the more modest funding budgets of the Criminology Research Grants (CRG), which can be considered little more than 'seeding funds'. Requirements for applications are laborious: one of our recent Discovery applications consumed 105 pages, and a large Linkage grant even more (219 pages). That represents a huge investment of time and money by academics throughout Australia, given the fact that, in their latest rounds, the success rate of Discovery Grants was 18 per cent; of Linkage Grants 36 per cent; 5 of Future Fellowships 18 per cent; and of Criminology Research Grants 10 per cent. Moreover, most of these grants do not pay for salaries, except for PhD students or research assistance, although they might provide for teaching relief. But even then the budget tends to be restrictive: grants simply fund the direct costs of research activities. Thus, while (inter)national collaboration is essential, these collaborations rely on the goodwill of people becoming involved, and on the willingness of host organisations to invest. In these circumstances, research actually costs universities money, forcing them to rely disproportionately on student fees for their income and challenging the financial viability of research centres.
Finally, some funding bodies expect/recommend ethics approval and/or agency approval to have been sought before submitting an application. While this is understandable from their point of view because they want to ensure that the grant will eventually be used, this creates a significant Catch 22 situation. Seeking approvals from both named sources is known to be a lengthy process, and problematic in light of the low success rates of grant applications. If ethics and agency approval are obtained but the project does not get supported, many people's time has been wasted. Conversely, if the project achieves grant support but the corrections agency does not allow access, the problem is magnified and might harm researchers' reputations with the funding body. In the event that research is related to Indigenous issues, even more hurdles need to be cleared, such as the preference for involving (an) Indigenous researcher(s), with setting up an Indigenous reference group, and with seeking approval from Indigenous ethics IJCJ&SD 9 Online version via www.crimejusticejournal.com © 2015 4(1) committees, state-wide and/or local. In light of this, a 'fatal fatigue' created by meeting approval requirements might result in certain aspects of prison research -more particularly those related to Indigenous over-representation in prisons, one of Australia's biggest penal problems -remaining under-informed by lack of much-needed qualitative research to deliver understandings of its dynamics.

The changing role of the prison researcher
In the second part of this contribution, I reflect on how prison researchers can respond to the changing conditions in the light of the above, based on my own experiences. To find a framework for doing so, I was inspired by the classification of possible roles criminologists can adopt, as more generally discussed by Loader and Sparks (2011) in their publication Public Criminology?. Loader and Sparks start their thinking from the idea that criminology these days is 'a successful failure' (Loader and Sparks 2011: 10). On the one hand, criminology is booming; on the other hand, the influence of its growing knowledge is shrinking. The same might be happening with prison research. While still a flourishing business within academia, we hear and read signs of a dividing gap between prison research and penal policy, and policy makers being more concerned about their public image than being steered by expert advice. In the following, I apply this model to the more narrow discussion of the role of the prison researcher and will use some of their prototypes 6 to explore how I think that prison research can contribute to the development of penal policy in a way that is beneficial for both academia and government organisations. It is my argument that our role as a (prison) researcher can be miscellaneous, and that each position taken has its own merits. The different roles involve a different proximity to decision making bodies, and this can impact on the uptake of the findings. However, as demonstrated by Harding (2003) in an Australian context, influencing policy through research is possible, and the uptake of research findings can be quite independent from the varying proximity to official agencies. 7 The scientific expert Loader and Sparks (2011: 29) describe the scientific experts' interpretation of the meaning of criminology as a source of 'valid, reliable and useful knowledge' in a way that is valuable for politicians, policy makers, practitioners and the general public in their decisions/visions on how to deal with crime. The idea is to provide solid evidence, to make crime-interventions effective, in a rational and evidence-based way.
While appealing at first sight, evidence-based research in the 'what works' tradition raises some methodological and fundamental concerns, as I have summarised elsewhere (Tubex 2010). It is very difficult in a criminological/penological context to attribute a (behavioural) change to a programmatic intervention because the sequential occurrence of two phenomena does not necessarily mean that there is a causal relationship. It also ignores the heterogeneity of offenders as individuals and the reasons why and how people achieve changes in their lives, as well as the social contexts in which this happens (McNeil 2004). Evaluation studies only measure what is measurable, possibly overlooking significant changes below the surface.
Further, because additional subsidies/support for programmes often rely on a positive evaluation of earlier outcomes, success can be stimulated by programme managers being selective about who is allowed to participate in an intervention. In the Australian context, this problem is exemplified by the limited access for Indigenous people to culturally appropriate programs. These programs are scarce and, while corrections might be well aware of these limitations, their budgets and research potential might simply not allow for program expansion. journey, I mainly worked in an academic context, conducting 'curiosity-driven research' on topics such as long-term imprisonment, parole, violence in prisons, sex offenders and the offer of welfare services to prisoners. The experience of these projects has demonstrated the importance of research that looks beyond the 'what works' tradition. The first research mentioned here investigated if the increase in numbers of long-term prisoners was due to real increases in the prevalence of seriousness of crime or due to changing punishment practices; the parole research project investigated not only the needs of the offender but also the desires of the victim; the violence in prison project described why prisons possibly don't work under the threat of unhealthy internal relationships and how the prison environment can address these; the research work on sex offenders investigated how their narratives lead them into the criminal justice system/prison and what the possible impact of early intervention could be; and, finally, the topic on welfare research assessed how we can meet other needs than the ones addressed by programmatic interventions, but which are about well-being and desistance.
Therefore, I argue that the wider context of penological research can provide added value to the evidence-based model. It is important that this background knowledge is gathered, that we continue to try to improve its quality, and that we provide a comprehensive and clear view of the whole set of factors that lead to imprisonment. This knowledge is one source that informs criminal justice policy, with due observance of its limitations. But penological research should never be reduced to mere effectiveness measurement. Analysing the complexity of criminal behaviour and continuing to question the whole criminal justice process and the implementation of punishment is, in my view, the core business of penological research and can thus make an important contribution to the evidence-based model.

The policy advisor
While criminological research in this perception remains independent, researchers being in charge of their own research agenda, within their research environment and with access to independent budgets to allow for less policy related investigation, there is also need for a more practical dimension. Research needs to 'inform debate and action', and this can be done by giving evidence and advice on a national and international level (Loader and Sparks 2011: 31).
I found myself in the position of 'policy advisor' when called upon as a scientific expert for the Council of Europe on various occasions, and as an advisor to the Belgian Minister of Justice, with both roles attached to my academic appointment. While these positions were measured as being less productive than academic research, they were rewarding from the perspective that they were possibly the best positions for an academic to influence change, in both direct and indirect ways.
Recommendations from the Council of Europe are not binding but they are still important levers for penal reform as demonstrated, for example, in the European Prison Rules. They become more directly relevant when experts are conducting fieldwork in member states of the Council of Europe in assisting those states to meet the European standards. I had the opportunity to be involved in penal reform in several Central and Eastern European countries in that respect. This experience enabled me to look beyond the borders of national research, to gain new insights, and rethink our own practices. While we generally consider our Western European standards as progressive by comparison with Eastern Europe, some aspects of penal reform are well and truly embedded in prisons in the latter jurisdictions. For instance, I was impressed with the generous opportunities that existed for family visits to prisoners in Eastern Europe. These provisions were, in my experience, well ahead of practices I had seen in some prisons in my native country of Belgium, and I was therefore pleased to be in a position to argue for improvement in my home settings. In that respect, international comparative prison research can provide leverage for improved practices across the board. Understanding prison practices in a foreign context requires researchers to familiarise themselves with local practices, as described by Nelken in his 'researching there' approach (Nelken 2010: 94-95). However, this can be challenging. As I have described elsewhere (Tubex 2013), comparative criminology is a relatively young discipline and this is even more the case for comparative penology, which was initiated by Downes' (1988) pioneering work. There are many barriers -language to name but one -and the need for broader historical and cultural knowledge to allow for correct interpretation of what shows on the surface. The comparative works of Pratt (2008) and Pratt and Eriksson (2013) have in this respect been widely discussed, both positively and negatively. From a more practical point of view, this kind of knowledge requires time and money, which in an era of shrinking budgets is far from readily available. Over the years, Council of Europe experts were given less time to accomplish fieldwork, mainly becoming fly-in, fly-out experts, with just enough resources to get the job done but with no additional time to spend in the country to read and breathe local circumstances. This has endangered the potential value of the work of expert advisors. Possible ways to address this problem could be to invest in local experts to become advocates for reform in their own country. In the light of the discussion above, however, my preference would be for stimulating broader (European/international) collaboration between supranational bodies that can carry the costs of such an undertaking.
Some other forms of pragmatism infused my role as a political policy advisor to the Belgian Minister of Justice. As this arena is highly volatile, realism sometimes gets in the way of principles. As post-release supervision services were limited, a decision had been made at some stage to reserve these services for the monitoring of long-term offenders on early release (those with sentences of more than three years), while shorter-term offenders were given an automatic form of release, with limited conditions and supervision. This flies in the face of criminological literature and, understandably, engendered some academic criticism. When it comes to the role of the policy advisor, it is my personal opinion that prison researchers too often start from the assumption that we still need to convince politicians of the fact that prisons are expensive and that they don't work. From my experience, most senior policy makers in this area are well aware of the basics of criminological knowledge. What they might not have available and need from academics is, firstly, knowledge presented in ways that allows them to communicate this message and, secondly, advice about achievable alternatives within the reach of their portfolio and timeframe (see also Indermaur 2009).

Observer turned player
In this interpretation of a third role, that of 'observer turned player', researchers decide to do something about the chasm between criminology and government by 'getting their hands dirty' (Loader and Sparks 2011: 32) and also by engaging with government through working inside government agencies, even if that comes at a price: 'one is often fighting lonely and losing battles' (Loader and Sparks 2011: 33).
This position is described with reference to my role as a Team Leader Research and Evaluation at the Department of Corrective Services (DCS) in Western Australia. When I joined DCS, it was still a young agency, having recently been separated from the larger Department of Justice. This name change is significant in itself: the move from 'justice' to 'corrections' is symbolic, reflecting a change of perspective about what a prison is and what it is meant to achieve. This also points to the Anglo-Saxon predominance in this development, as several European prison services now describe themselves in English as 'correctional services' although that is not their title in their own language. 8 While DCS had a substantial research team and a considerable number of external/commissioned pieces of research going on, there was not a strong governance framework around research. During the years I worked for DCS, we built on this by developing IJCJ&SD 12 Online version via www.crimejusticejournal.com © 2015 4(1) more rigorous procedures within the Research and Evaluation Committee (REC) -the body overseeing all external (and later internal) research and evaluation projects -that aimed for a fair and research-friendly process. This was an interesting experience for an academic researcher such as myself and, while initially it felt like 'sleeping with the enemy', it allowed me to gain better insights into considerations prejudicing the perception of prison staff against the benefit of allowing external researchers access to prisons.
Firstly, there may be misconceptions by prison staff of the extent of researchers' knowledge about and appreciation of what goes on in the prison world. Prison staff sometimes feel, or fear, they are being treated disdainfully and dismissively. One element of this might derive from a misunderstanding of the need for researchers to maintain a professional distance from staff, which might lead to staff becoming resistant to assisting researchers whose attitudes appear to prejudice their understanding of workplace complexities. Secondly, the time scales to which researchers work are different from the needs or expectations of correctional management. The problem with thorough and sound research, as required for scientific work, is that it takes time.
As a consequence, findings are sometimes reported well after the fieldwork was conducted, and is therefore of limited immediate practical relevance. The problem for policy is that it needs to move on and that implementation of a new or revised practice cannot always wait until a review is completed. One possible way to handle this is to share findings during the research process. This method is not without risk: incomplete results might be misleading, or might be generalised and therefore incorrect. Yet it is important that key stakeholders be informed during the course of a project so that they know what activities have occurred, what the results are, and how this eventually leads to certain conclusions. The third point is related to this: when research is facilitated by a prison, a regular complaint from the prison staff and prisoners is that, once researchers have harvested their data, they disappear into a black hole and are not heard from again. A lack of giving feedback on research findings undermines the reputation of the research. Research is valued where all parties maintain their relationship throughout the process but at the same time respect relative professional positions.
From this personal experience, I plead for prison researchers to take these aspects of prison life into consideration when researching and reporting. Revealing weaknesses without acknowledging efforts already in place to rectify associated problems or identifying areas and ways for improvement is less than helpful. In fact, this can be very discouraging for people who must work in the field with the daily reality of tight budgets and pressing problems. Prison research will benefit from perceptive interpretation and constructive suggestions, without this compromising the integrity of the research and the authenticity of the findings.

The lonely prophet
Research needs to look at the bigger picture, and instead of being small scaled and theoretically ungrounded, investigate economic, social, cultural and technological changes that impact on crime. It is macro explanations that are needed in this field, to understand matters such as mass imprisonment, and enlighten alternatives for thinking and acting accordingly (Loader and Sparks 2011: 34).
The 'lonely prophet' position is best described as relating to the current stage of my journey through prison research. Being back in academia, at this stage of my career and at the other side of the world from where my journey began, I decided to somewhat move away from fieldwork on the inside and look at 'the bigger picture' instead. I found this was like entering another world again, not only because of the contextual differences but, even more strikingly, due to the 'ideological' contrasts. Over recent decades, various drivers of prison populations have been identified and the bigger picture has been unravelled at many levels. At the same time, this kind of macro perspective is being challenged in some prison research, including articles in this special issue, for ignoring the reality of the 'lived prison experience'. The two strings of IJCJ&SD 13 Online version via www.crimejusticejournal.com © 2015 4(1) literature hardly seem to communicate and, if they do, it is often to criticise each other's methodology and the limitations thereof. In my view, it is not an 'either/or discussion'. Empirical fieldwork is essential because every prison, every penal setting is different and merits its own analysis. At the same time, to interpret what is going on in all these pieces that compose the jigsaw, we need the bigger picture. For change is needed at different levels of punishment, and knowledge needs to contribute to both: on the ground, to improve daily prison practices, but also on a societal level, to rethink punishment theories.

The need for academic prison research
In this last section, I first consider possible objections that may hamper academic prison research. Then, based on the experiences described above, I make a case for why I think it is worth building bridges between the research world and prisons, and why we need each other.
From a corrections point of view, allowing prison research raises some concerns. Corrections have a duty of care towards offenders, their being considered 'vulnerable', a 'captive audience' in need of protection, including against research. Some research is, by its nature, intrusive and upsetting and this reality needs to be appropriately accommodated. However, the need for protection of potential research subjects can also be overstated by prison officials, and many prison researchers report that their respondents appreciated the opportunity to talk to people from the outside, if only as a break from daily prison life. Further, the protection of prisoners needs to be balanced against their right to express themselves and be heard, just because they are vulnerable. Conversely, prison research also has its 'fashion trends' and sometimes certain groups within the prison population tend to become over-researched, which is problematic. In these cases, the duty of care might include putting a temporary stop to popular research topics. Additionally, corrections regimes are also in charge of security and order in a prison, and moving prisoners forwards and backwards for research purposes does have an impact, particularly when conditions of overcrowding exist in a prison. This requires understanding and respect from researchers, and flexibility in their approach.
So, why should government agencies engage in research conducted by external researchers? A first motivation is a very neo-liberal one: they get most of the research work done for them, for free, by (mostly) experienced researchers in whose training they don't have to invest. More fundamentally, academic researchers might be able to identify significant topics or trends that are out of the scope of daily practice but that might be of relevance to it: in other words, sometimes you don't know what you might need to know. Moreover, government agencies can gain from academic research which can keep them abreast of new/international developments and more fundamental or theoretically oriented research questions. Academic research can also assure that pro-active, innovative, conceptual and enduring research is conducted, which complements research of an applied nature, and from which correctional and penal policy might gain in the longer term.
Working with governments can come with some benefits for academic research, as it might lower the threshold for access and engender (financial) collaboration, but it comes with its own issues. There can be restrictions regarding the research agenda, about what can be researched and how, and about the ownership of results and publications. No organisation is keen on exposing itself by revealing negative findings or matters that are subject to improvement and, in a way, this is understandable. Related to the accountability of government agencies, there is, however, a need for independent scrutiny, and I see a willingness to identify shortcomings, to acknowledge them, and to indicate how they will be addressed in future as a matter of maturity. So, why should academic researchers continue to do prison research? There are many reasons, one of the most important possibly being that we like doing it. The carceral world remains a fascinating environment and an almost inexhaustible source of research. But there is more. I think that we, as prison researchers, have a moral obligation to keep questioning and investigating prisons and all closed and total institutions, so as to provide an outsiders' report of what is going on. It is a scientific duty for prison researchers to contribute to gathering evidence for developing correctional and penal policy, aiming for insight and improvement.

Conclusion
Researchers these days need to serve many masters, with the risk of losing their autonomy and integrity. However, I am convinced that criminology and, for this part, penology has a role to play in shaping penal policy. In this respect, I place myself more so in the camp of Roger Hood (1987) than of Stanly Cohen (1985), as discussed by Harding (2003). I hope that this special issue will contribute to the debate between academia and practice. With this in mind, I have tried to outline how I think we can cooperate within the current penal landscape.
When it comes to gaining the collaboration of government organisations, academic researchers need to meet, but not be limited to, their needs. We need to offer them research that is interesting for them, and more. While an evidence-based (penal) policy might sound as something of an oxymoron, I see the undeniable usefulness of a policy that is informed by research, taking into account the above concerns. Quantitative measurements, evaluations, audits and monitoring need to be complemented with qualitative research which by its very design can add significant value to the evidence base. For a healthy relationship between research and policy, I agree with Freiberg and Carson (2010) in their preference for 'the enlightenment model', as described by Young et al. (2002). In this model, research stays at a distance from policy, providing a framework for thinking about the problem. Rather than science finding solutions to policy problems, it provides information and a conceptual framework within which a problem can be studied and understood. This model also allows for the emotional aspect of implementing evidence, which is identified by these authors as an important driver of public policy (Freiberg and Carson 2010). In that sense, an evidenceinformed model is a more accurate description.
But it is also within our own home base that more hard work needs to be done. While I absolutely agree with, and have been arguing for, making the results of our research work available and of value to the wider community, there will always be a need for curiosity-driven research, which might be less attractive for government agencies, but fundamental to achieve academic excellence. Therefore, a plurality of funding sources is needed to allow for a plurality of research environments. Further, for the sake of preserving high quality research and output, we need to convince academic leadership of the importance of 'slow science', allowing us to think, explore, find, and rethink before scoring ('Bear with us, while we think' 9 ). The increasing problem of scientific fraud is a symptom of an academic world where producing outcomes overtakes scientific rigour. We should plead for 'free range researchers' instead of the caged version of us, even if this takes more time and costs more money. Human science cannot always or solely be measured in output, as one of our 'key performance indicators' should be what we can mean to people living and working in difficult environments and how we can possibly make a positive difference.