1.

'Responsibility' is always in danger of being a rather grey word; it can be used repressively and even menacingly. Because it has come to be the counterweight to talk about rights, it is liable to sound like the bad news after the good news - rights and responsibilities, we say, to remind ourselves that there are no free lunches. I suspect that this rather sombre colouring to the word is the result of understanding responsibility primarily as being accountable to others, to society, authority and so on. 'You must act responsibly' is what the world around says to us, to me; it is about restrictions, about the proper negotiating of our purposes in full awareness of the claims of others.

This is true as far as it goes, but the cost of thinking only in these terms is high. It leaves out of account that other aspect of responsibility to which I want to draw attention here, which is responsibility for: it isn't just that we are responsible to others, and that others have a right to demand certain kinds of behaviour from us; we are also responsible for each other. Each of us has to answer for someone else's welfare as well as our own - and that means that there must also be someone else who is there to answer for us, whose concern is our welfare. This is a theme deeply rooted in our Judaeo-Christian ethic. From the question of God to Cain - 'Where is your brother?' - to St Paul telling us to 'bear one another's burdens', this ethical tradition affirms that the most fruitful and peaceful common life depends on willingness to speak and act for each other in this way. It isn't simply that we are all responsible to a common and impersonal set of claims; the relation is horizontal as well as vertical, and we are in the business of speaking up for, defending and enriching the lives of other people. A good legal culture, on this definition, will be one that allows and encourages this kind of mutuality, not simply one that reinforces the general claims of the other and the community. It is a culture in which it is taken for granted that everyone should have a voice, and that where voices go unheard for whatever reason, there will be others who accept that they must make sure this silence is broken. It is not perhaps a self-evident feature of liberal or democratic society if those terms are thought of only as securing various kinds of freedom from outside interference or unchallengeable top-down authority. But if liberal and democratic societies are to be more than assemblies of atoms, we need something like this biblical dimension of responsibility for each other to come into play.

So yes, the doublet of rights and responsibilities is unavoidable. But it is not a matter of balancing what's due to me with what's due to society or other people in general. My responsibility to speak for you is inseparable from someone's responsibility to speak for me: no-one is left isolated, and no-one's welfare is finally to be separated from that of all others. Responsibility is good news as well as bad, if you must think in that sort of framework. And what a healthy legal culture promotes is - if you allow this general moral background, what you could call a 'liberal-plus' approach - individuals growing into better mutual awareness, so that their speaking for each other and taking on the task of another's welfare may be just, well-informed and solidly motivated.

In what follows, I want to apply these principles to the Criminal Justice System as we know it in the UK at present and to ask where the most urgent need lies for action to equip it better for this kind of purpose. If a healthy legal culture is defined as I have suggested, how healthy is our current climate? Reform in this area naturally asks the question of whose interests are suffering at any given moment, whose voice is unheard. And perhaps the strongest new groundswell of popular opinion in the last couple of decades has been to do with the rights of victims. The way in which our legal processes have often had the effect of 'revictimising' those who have suffered from crime has been a constant refrain, especially where crimes against women and children or otherwise vulnerable people are concerned. Stress is also laid on the need of victims for some sense of closure, and for some sense of involvement in a legal process which, after all, is only happening because they have suffered. This arises in part from a very natural reaction against sentimental attitudes to offenders on the part of well-meaning liberal reformers, reflecting a lack of any concern with that aspect of crime which is about more than the guilt of an individual but involves others because it is a breaking of relationship. It is all too easy to work from a notion of crime that is wholly bound up with the individual responsibility of the offender and the punishment of the offender as a means of asserting those impersonal social and communal claims that have been denied. But crime has effects that spiral outwards from the single act or group of actions at its centre, into the lives of many specific people; how do we attend to the mending of all that has been broken in this process?

2.

Hence the new emphasis on the welfare of victims and the creation of groups like Victim Support; hence the questions asked about how those aspects of legal process that intimidate victims might be modified; the calls for the introduction of an ombudsman for victims of crime, and for various benefits to be available for victims whether or not an offence has been through the criminal justice system. In terms of redressing a balance and attending to a seriously neglected dimension of criminal justice policy, all this was and remains welcome. International standards for the care of crime victims - the EU Framework Decision of 2001 is a good example - have been transformed. 'Most trial processes - whether civil or common law based - leave victims feeling cheated and too many offenders fail to take responsibility for their actions' wrote Helena Kennedy in her 2004 book, Just Law (p.224). Change was scandalously overdue.

But it is the same Baroness Kennedy who points out the problems associated with this new emphasis. It can be, she says, 'a Trojan horse' (223) for those who believe that offenders' rights, and indeed the rights of any accused person in the system even before sentencing, need to be reduced. It can stereotype the needs of victims and confuse the function of legal processes (which are not designed to be about compensation first and foremost). And in spite of concerns about the experience of victims in the legal process, some of the political advocates of victim rights have often sidestepped the immediate practical issues about how victims can be better prepared for the experience of the courts and what special provisions are desirable for the most vulnerable in particularly sensitive cases such as rape and abuse.

Most disturbing, though, is the insidious way in which the legal system, like so many other areas of our public life, comes to be consumerised in this climate. Brian Williams, Professor of Community Justice at De Montfort University, in an excellent recent study of victims and the justice system, says, of the developments in the nineties and the early years of this decade, 'Victims were increasingly being characterised in policy debates as users, or even as clients or consumers, of the criminal justice system, although a number of observers have argued that this is a false analogy: victims have little choice about using criminal justice services and most would prefer not to be in the position of having to do so' (Victims of Crime and Community Justice, 2005, p.25). The most extreme versions of this focus on the supposed needs of victims and on the 'client' model have been in evidence in the USA, with the involvement of victims in the parole process, and in some contexts the invitation to relatives of a murder victim to witness the murderer's execution. American jurists are beginning to worry about the disproportionate leverage exercised on the legal process by this degree of participation.

In the light of the general principles sketched out here, what does this say about responsibility? First of all, like every attempt to do justice to the perspective of the victim, it runs the risk of imagining the victim as perennially passive, someone things are done to, whether good or evil things. The shift towards the concerns of victims began with the recognition that the existing system left victims feeling powerless; the danger is of creating a culture in which they remain powerless - powerless to change or move forwards or exercise their freedom in a civic or communal way, and endowed only with a rather strange variety of consumer power, faced with choices that others have determined for them. A system that was at the mercy of organised lobbying on behalf of the victim would not serve the real interest of the victim because it could never break out of the stress on the victim role at just the point where someone might need help to shed that. And, as Brian Williams notes (p.26), there is a temptation to politicise the whole situation, with tough sentencing policies being presented as the best way to address victims' needs. Such research as there is on 'victim satisfaction' is very far from giving anything like clear support to this. Should we not be thinking about policies that looked towards the restoring to the victim of some renewed capacity to engage responsibly? To make fuller use of the empathy that can be nurtured through reflecting on experienced injustice and trauma?

3.

If some of the language around the needs of victims has these rather ambiguous implications, how do we begin to think about the needs of offenders in a way that avoids what I earlier called the sentimentality and individualism of some penal reformers in the not too distant past? If the underlying problem in crime is a breakage in relationship, this means that the offender has lost the active sense of being answerable for others. That sense is, as I've suggested, inseparable from the assurance of having others who are answerable for you. The most unhelpful and indeed damaging way of treating this is thus surely a system that leaves the offender without any grounds for believing that he or she is the object of anyone's responsibility.

This is emphatically the message that much of our present system still gives to the offender. The statistical likelihood is that an offender will be accommodated in overcrowded conditions, deprived of privacy; that contact with family will be vulnerable to unpredictable moves and varying policies in different institutions; and that informal personal support (such as the excellent work of the Samaritans, for example) will be at best patchy. Families of prisoners, including, very disturbingly, remand prisoners, who die in custody thankfully now have the assurance of transparent investigation into their deaths; but there is still no mention of prisons in the Corporate Manslaughter Bill, an omission that the Joint Home Affairs and Work and Pensions select committees have agreed to be indefensible. And of course prisoners are still deprived of the vote.

Since 2004, the National Offender Management Service has been developing, in the light of the 2002 report from the Social Exclusion Unit, a far more integrated approach to continuity within the custodial period and in the context of resettlement, and is addressing resettlement needs in far sharper focus than was once the case, commissioning more comprehensive service packages around housing, work, family support, substance abuse risks and so on. Likewise the Prison Service at the moment understandably prides itself on the quality of some rehabilitation and detoxification programmes in prisons, and on the fact that primary health care is supposed to work to the same level as the NHS in general. All this is welcome evidence of the recognition of the urgent need to provide stable and reliable services for offenders at every stage of their progress through the system and back into the community. But plenty of issues remain about delivering these new aspirations and standards, and much of this work is still embryonic. The risk of some fragmentation when services are contracted to private providers is still a contested aspect of the developing pattern of offender management. And the quality of services within prisons poses the uncomfortable question of why so many people with drug problems and (especially) mental health problems are in custody in the first place, and why prison is expected to supply what ought to be available in the community at large for vulnerable individuals. If certain kinds of care and treatment are in practice available to some people only in custodial conditions, the message is still that there is a deficit in responsibility somewhere.

The populist complaint that facilities are available in prison that are not accessible to people outside has, ironically, a point to it, though not exactly the point that the populist wants to make. It is not that prisoners are 'privileged', but that prison is being asked to plug gaps in an overstretched welfare regime. And given that the Prison Service itself is overstretched and that sustained attention to prisoners' health and education needs is in any case almost impossible to effect with shorter sentences (further reduced by remand delays), the standards that some institutions can properly boast do not mean, alas, that we can take it for granted that prison, corporately and generally, always involves attention to the most pressing needs of the offender. It is said so often that it is virtually a cliché, but it has to be said again: custodial sentencing means that you are punished by being deprived of your freedom, not that your subsequent welfare is a matter of indifference.

If we seriously want to address the problem of reoffending, it is clear that a penal culture in which there is no real attention to how offenders change is worse than useless - literally worse than useless, in that it reinforces alienation, low self-worth and the lack of any sense of having a stake in the life of a community. Once again, this has been recognised in recent policy developments, and programmes directed to changing offending behaviour are now a mainstream part of the overall 'paths to resettlement' defined by the NOMS and are beginning to show some effects in reoffending rates. The then Home Secretary said in his 2005 lecture to the PRT - a lecture in which he gave strong support both to community sentencing and to a greater involvement of prisons in the civic life of an area - that 'clear goals right from the start' and a commitment to a distinctive 'package of support and interventions for each offender' had to be the keystone of a really effective penal policy. It was a lecture that quite rightly signalled the need for a substantial culture change.

4.

To create another kind of culture in the Criminal Justice System, one that is committed to building responsibility, the first thing we must do is to get rid of the tacit assumption that managing the needs of victims and the needs of offenders is a zero-sum game. As I have outlined it, the history of recent policy debates shows how easy it is to become trapped in such a model. Isolate offenders' needs from those of victims, and there is a predictable reaction in the direction of treating victims as the primary beneficiaries of a process in which they are to be given something like a veto in some circumstances and are imagined as customers who must be satisfied. Isolate victims' needs from those of offenders and you have another reaction which will leave victims without information or interaction in regard to offenders, and which will risk making the rehabilitation of offenders a narrowly focused therapeutic exercise. As many people, probably the overwhelming majority of those who work with either offenders or victims, recognise, we are very unlikely to move away from this standoff within the framework of our current practice. Adversarial forensic processes and the custodial fundamentalism of a lot of new law will inexorably combine to keep appropriate responsibility from both parties. I say nothing - though a great deal could be said, and indeed has been said in the last week or so - of the way in which the custodial obsession, the creation of more and more offences with a custodial tariff, simply chokes the prison system and compounds all the failures in responsibility for prisoners that I listed a few minutes ago. Nor will I elaborate on the cost to taxpayers of an ineffectual and overloaded system, and the wider cost in patterns of reoffending because of the inadequacy and unevenness of responsibility-building services in such a context.

But naturally, new models do not spring to life ready-made. The main tools of a policy that will avoid the zero-sum deadlock are of course community justice and restorative justice. But, to paint with a very broad brush, community justice (non-forensic hearings and negotiations outside a court system, involving 'stakeholders', direct and indirect) can still be seen as a vehicle for addressing victim concerns at the expense of offenders; and restorative justice (mediation, conferencing, confrontation between offenders and victims, attempts to achieve some kind of emotional closure by apology or reparation) can be seen as giving offenders the chance to avoid appropriate punishment by going through certain motions, not too difficult to learn, which may or may not be of any use to victims and in fact risk some damage to them.

Yet it would be disastrous if such negative or suspicious perceptions stopped us thinking through both these models more carefully - models which, in fact, have a good deal in common. Both take seriously the notion that those most directly involved in a criminal event have the largest stake in containing and dealing with its consequences. Both assume that penal custody as the default solution to the effects of crime is inadequate - though both accept that it is in a good many serious cases a necessary part of the whole strategy. Both attempt to find ways of treating crime that move things on, that create new situations and relations, instead of simply removing people from the social scene. So let me try to outline some of the considerations that may help us keep both in focus together and that could answer some of the concerns about reintroducing the victim-offender standoff under a new guise.

The first is obvious, but needs articulating nonetheless. We must think about these models on their own moral merits, not as economic or political shortcuts. Community justice needs to be separated from any suggestion of franchising or privatising the operation of the law. The idea that offender management should be put out to tender is one that could sit very comfortably with some sorts of talk about community justice if we are not careful; and this buys into a very questionable understanding of genuine collective responsibility fully owned by the state - a properly common moral discourse about crime and punishment. And it is essential that both community and restorative programmes are seen as genuine responses to a problem, not dismissed as simply attempts to evade the punitive dimension, the necessary vindication of a society's values. It is an easy vote-catcher to describe alternatives to custody in this way. It would be welcome - though it feels at times like crying for the moon - if politicians and commentators could refrain from speaking as if punishment were essentially about the expression of disapproval and the infliction of legally controlled suffering and not much more. In itself, such an approach changes nothing; but any crime surely indicates that something needs to change in a person's awareness and conscience.

Secondly, we need clarity and honesty about what offences can and cannot be appropriately dealt with under the non-custodial and reparative model. Where serious crimes of violence and abuse are concerned, and where a continuing threat to public safety is involved, we have to think less about alternatives to custody and more about what can be achieved in the custodial setting, admitting that custody is an unavoidable element in society's response to the crime. But this means thinking very carefully about the tariff for many kinds of theft, for economic crimes more generally, for a good deal in the area of petty vandalism and drug offences - to offer a few examples. Custody certainly carries a distinctive stigma, and it is regularly argued that this on its own is a uniquely powerful deterrent. But there are problems with this. It is not at all obvious that the 'stigma' sanction works for those among whom petty criminality related to theft and drug usage is relatively common - those with educational deficit or mental health challenges. There is something of a vicious circle involved in repeated use of custody for those who manifestly do not see it as a sanction of great significance. And, paradoxically, brief custodial sentences are likely to be the least effective in changing or challenging behaviour, given those characteristics of the contemporary prison experience already listed - overcrowding, frequent movement, uneven provision of rehabilitative services and so on. There is good evidence that reoffending rates are proportionately higher among short-sentence prisoners.

It should be clear too in this general context that the impact of custody on women, especially women with heavy family responsibilities, is disproportionately severe and destructive. It is worth remembering that two thirds of women in custody have children under sixteen. Helena Kennedy, in the book already mentioned (pp.305-6), gives a heartrending case to do with a teenage Bengali girl pregnant as a result of rape within her family, who tried in panic to dispose of her newborn baby and was charged with attempted murder. She rightly asks why a child who is the victim of crime should be prosecuted in this way. In a case like this a custodial sentence for the crime of which the girl was accused might 'send a message' of a kind, but would both traumatise and criminalise a vulnerable individual, as well as leaving guilty persons untouched (fortunately the judge in the case agreed).

So we need to continue to work to define with greater and greater clarity where custody is going to be destructive, and where the primary need is to address some of the deficits in education, capacity and empathy that stand behind crime. We should also note in passing that the case Baroness Kennedy cites shows some of the potential pitfalls of an uncritically 'communitarian' approach: the terror of the young woman in this story was intensified by the attitudes of her family and immediate community to pregnancy outside marriage. Any community justice programme worth the name needs the closest monitoring and the most stringent training to prevent its becoming a vehicle for group or local prejudice. Community justice needs to work in such a way that it represents a local group of 'stakeholders' taking responsibility not simply for their immediate environment but for the application of standards, liberties and dignities agreed by the whole of their society - which is why the popular delegating and franchising (privatising) options need to be unmercifully scrutinised.

Third, community and restorative models need to gain consent from both offenders and victims. That is, they have to define and deliver outcomes that will be attractive enough to hold the confidence of those involved - some sorts of reparation and closure for victims, some sorts of new empowerment and opportunity for offenders. This takes time. What victim and offender believe they need may not in fact be what will most help them. There will be some victims for whom reparation can mean something like revenge; there will be offenders who confuse empowerment with an escape from guilt. Victims (or their advocates) may, as we have seen, want to control the fate of offenders, offenders may want to accumulate merit marks to secure better treatment (early parole, if they are within the custodial system). Faced with these difficulties, the temptation is to cut corners; Brian Williams in his study of both models of procedure refers to numerous studies in the last five years that will have a very sobering effect on anyone who is inclined to be messianic in their attitude to alternatives to custody. There can be community programmes that are just as careless of victims' needs as any classical forensic process; there can be restorative schemes that fail to implement or monitor agreements reached and become both hurtful for the victim and useless formality for the offender. To use that painfully well-worn phrase, there has to be a 'desire for change'. Victims may want to move on with the sense that they have been properly heard and attended to. Offenders may want to find the right kind of help in altering destructive patterns of behaviour. Those different sorts of desire then have to be taken seriously and worked with by those who administer such schemes, whether the police, the custodial institution, or whatever local consortium is involved.

5.

How this desire is fostered is the crucial question; dealing with this is the key to the other two considerations I have raised. The whole complex of penal issues is less likely to be abused and exploited by politicians and commentators if there is wider ownership of an approach that asks how the criminal justice system helps to bring about change. The law encodes our respect for human dignity in general; but this should not be taken to mean that the function of the justice system is only to flatten out a bumpy surface or to restore a disturbed situation to the status quo. It must also be the guarantor of possibilities, clearing the way for individual citizens to exercise their dignity by taking part in the processes of shaping the conditions of their lives.

This does not mean that the law in itself is the agent of moral change - a tempting but dangerous idea as you can see when it gets into the wrong hands. The law cannot prescribe reconciliation and it cannot effect forgiveness; what happens if a penalty is completed or remitted in some way is not forgiveness but only discharge. Mended relationships need more than this, and what they need the legal system does not and should not try to provide. But a system that actively works against reconciliation, against the development of those involved in a criminal event towards something more adequately and resourcefully human, is in its way just as dangerous. It can serve human dignity only in a formal and rather abstract way: punishment is calculated on the basis of its effectiveness in expressing social disapproval and in some measure on its deterrent capacity. These are components in any nuanced theory of punishment; but they do not create conditions in which a society can move forward in lessening the chances of disruptive and abusive behaviour. Somehow, the law has to play its part in energising such movement. To put it slightly differently, the legal system of a society cannot enforce morality; but a society that has no corporate energy for improvement is going to waste human and material resources in a way that undermines its credibility and legitimacy. And, on present showing, it will do nothing to halt rising levels of serious crime. The state through its legal processes will not make us good, but it will, if it's doing its job, give room to and support a range of other processes and factors that prompt and sustain moral decisions. If democracy is more than a dreary plebiscite for apathetic consumers, it should be nourishing empathy, awareness and critical change.

And this amounts to suggesting that the problems of the criminal justice system are inevitably problems generated by a wider context than this alone. In a setting where volunteering is not encouraged, where the ideals of public service are not reinforced and where the 'customer' paradigm prevails in most areas of public transaction, it is not surprising if we have a penal system that too often appears chaotic and ineffectual. It will not be possible to build responsibility in the criminal justice system unless the sense of being answerable for each other is strengthened in the whole of our social fabric. Which does not mean, of course, that prison reform can be postponed until we have solved the bigger problem; the special significance of this area of our difficulties is that it brings into such sharp focus the cost of a climate of non-responsibility. It pinpoints how desperately hard it becomes to spring people out of destructive cycles when that aspiration is not in the bloodstream of a culture. And - sadly - it reminds us also that it is almost impossible to move some things forward when they lack obvious electoral rewards; this is how a downward spiral of expectation is created.

That downward spiral of expectation is of course one of the things that sustains those assumptions about the needs of victims which can trap victims of crime in a reactive posture and allows concern about victims' needs to be exploited by those who want to reinforce alarmist messages about 'soft' treatment of offenders. In a different climate, well-managed processes of restorative justice have the potential to restore dignity to the victim as well as the offender. And in a community justice framework of the right kind, the chosen involvement of victims in dealing with the broader fallout of a criminal event means that they are able once again to exercise their dignity actively in what I earlier called the shaping of the conditions of their lives.

That raising of the level of expectation in relation to victims of crime necessarily entails a similar raising of the sights of what is expected from offenders. If crime can be not only punished but challenged, this puts the offender in a position where he or she has to make sense of their actions in some way; and this is a vital part of the fostering of responsibility. Making sense of your actions involves understanding where they come from and where they go to - motives and consequences; it means seeing what you do in a wider context than how you feel or what you want. Responsibility grows when actions can be planned that will play a part in this wider context, where goals are imagined that are more than just release of feeling. To be helped to be answerable for what you do is to be helped to understand deeper levels of your own worth or significance; there is some aspect of another person's good that depends on you. To allow people to see that realisation as the root of empowerment is one of the great challenges of a renewed criminal justice culture, and it is not clear that it can be achieved simply by the cognitive-behavioural therapies that tend to be the favoured option at the moment. In the most practical terms, it can be fostered also and importantly by further exploring volunteering programmes for prisoners along with the more protracted work that goes on in restorative conferencing, anger management and so on in a properly sustained rehabilitative scheme. And the potential for the 'civic presence' of prisons needs more flesh on it, not least in regard to the possible role of local groups, including faith communities, in sustaining community chaplaincy provision to facilitate the transitions between custody and freedom and to help balance the difficult tension between guaranteeing a community's safety and pushing the boundaries of what is possible for the ex-offender. If the care of offenders after release is focused too tightly on managing risk, there will be less opportunity for trust and therefore responsibility to be built up. It is a point worth pondering for the probation service.

In recent years, the restorative model has become more and more deployed in the Youth Justice System; and, in spite of some false starts and awkwardnesses, it is beginning to prove its worth. Now is the time for 'mainstreaming' it. The commitment of the Home Office to developing community prisons is in many ways welcome. But it has to go side by side with alternatives to custody - and therefore with some long hard thinking about the multiplication of custodial tariffs when new legal offences are created. This is a long shot, I realise - but I wonder if the time has not come for a comprehensive commission on penal policy and non-custodial options. We cannot for much longer manage with an expanding prison population whose levels of reoffending are so high.

In conclusion, though, let me repeat the point made a few minutes ago. The crisis in the penal system is bound up with the wider question of whether our social imagination in general is being fed by the vision of mutual responsibility. It is unhappily easy for the sceptic to suppose that the religious perspective on these matters is essentially and even exclusively about underscoring guilt and penalty. I have been trying to suggest that the most distinctive contribution such a perspective may bring is a stress on finding our adult liberty in carrying the responsibility for someone else's welfare. If at least that dimension of the religious ethic that has grown out of our tradition can be revitalised for our society, it is not only our attitudes to penal policy that will be regenerated and transformed.

© Rowan Williams 2007