Ch 18. Rehab

The whole process was torpedoed in the 1970s by a devastating study which looked back over 231 evaluations that had taken place between 1945 and 1967. In a robust summary for a political journal, one of the authors, sociologist Robert Martinson, concluded that with ‘few and isolated exceptions the rehabilitative efforts that have been reported so far have had no appreciable effect on recidivism’.

Robert Martinson, ‘What Works? Questions and Answers About Prison Reform’, Public Interest 35, Spring 1974, p. 25.




Martinson was a penal reformer1 who two years previously had written for a liberal magazine claiming nothing made much difference and therefore people languished too long in jail because the authorities played safe before releasing them.2

1 Martinson had been arrested as a civil rights ‘freedom rider’ and had spent time in a high-security Mississippi penitentiary. He once wrote that prisons cannot be reformed and must be gradually torn down.

2 Robert Martinson, Paradox of Prison Reform, The New Republic, 1972, 166, April 1, 6, 15 and 29.




… in the following years across the country maximum sentences became de facto tariffs as rehabilitation was phased out. There was talk of the ‘death of liberal criminology’.1 In 1979 Martinson tried retracting some of his more forthright comments2 but it was too late; and the following year he leaped to his death from a ninth-floor window of his apartment in New York. Finally rehabilitation was officially killed off by the US Supreme Court.3

1John Galliher, ‘The life and death of liberal criminology’, Contemporary Crises, Vol. 2, 1978, p. 245.

2 ‘… some treatment programs do have an appreciable effect on recidivism’. Robert Martinson, ‘New Findings, New Views: A Note of Caution Regarding Sentencing Reform’, Hofstra Law Review, 1979, 7 p. 244.

3Mistretta v. United States, 18 January 1989.





In fact Martinson’s flamboyant writing had been less guarded and more compelling than the facts that underpinned it. He had come late to the research, though his pithy articles pre-empted its academic publication and were written without consulting his co-authors.

Douglas Lipton, Robert Martinson and Judith Wilks, The Effectiveness of Correctional Treatment: A Survey of Treatment Valuation Studies, New York: Praeger Press, 1975. The authors compared an assortment of correctional-based programmes including intensive supervision, psychotherapy, group therapy, vocational training, educational approaches and medical interventions with a range of outcomes including recidivism, educational achievement, and drug use. They concluded that they had not ‘as yet’ found ways of cutting reoffending ‘by significant amounts’.




In the ensuing hullabaloo the US National Academy of Science appointed a panel to try to sort out what the evidence really was. They were appalled … Most rehabilitation programmes had been weak, piecemeal and poorly managed, and the research programmes used to appraise them had been unimaginative, unsystematic and lacking rigour.

L. Sechrest, S. O. White, E. D. Brown, Rehabilitation of criminal offenders – problems and prospects, National Academy of Sciences, Washington DC, 1979.[Abstract:]




Then there are official groups like the probation services where social workers, traditionally from the liberal left, feared ‘a real danger that probation is slipping away from us’.

A. Mullender, editorial, British Journal of Social Work, Vol. 29, 6, 1999, pp. 837–8.




Home Office researchers randomly divided 1,000 offenders so that half were left to their own devices while the other 500 had intensive help with family dynamics, housing practicalities, leisure pursuits and finding work. It made no difference (or rather, while there were slight differences for some offenders they were not statistically significant, which amounts to the same thing).1 A flurry of other reports around this time reached similar conclusions2 and an official review for British ministers, hoping to find some crumbs of comfort, finished up ‘confirming the pessimism of previous reviewers’.3

1M. S. Folkard, D. E. Smith and D. D. Smith, IMPACT: Intensive Matched Probation and After-Care Treatment, Home Office Research Study 36, 1976, HMSO, London.

2 e.g. Davies 1969; Wilkins 1969; Hood 1971; Sinclair 1971; Davies 1974; Dunlop 1974; Millham et al. 1975; Cornish and Clarke 1975; Brody 1976; Bottoms and McWilliams 1979; Thorpe et al. 1980.

3 S R. Brody, ‘The Effectiveness of Sentencing – a review of the literature’, Home Office Research Study No. 35, HMSO, London, 1975, p. iii.




Liberals were appalled – ‘harshness has replaced hope, retribution has replaced rehabilitation, and prevention has eroded proportionality’.

A. Freiburg and S. Ross, in A. Karpardis (ed.) ‘Sentencing: Some Key Issues’, Special Edition, Law in Context, 1995, Vol. 13, No. 2, p. 138.




… so much hope and effort had been invested in social interventions that supporters of the sociological tradition were bound to fight on, convinced their babies had been thrown out with the bathwater. They pointed out that no one could be sure that nothing works; the case was simply unproven. They argued that rehabilitation had not had a fair trial: many of the tested interventions had been ill defined and poorly funded, counsellors had often had little formal training, prisoners had sometimes been seen only intermittently (in one case five times in a year), and crude reconviction rates could fail to spot that an offender was scaling back to less serious transgressions.

Among the most-often cited are the psychologists, Don Andrews, James Bonta and Robert Hoge, ‘Classification for effective rehabilitation: Rediscovering psychology’, Criminal Justice and Behavior, 17, 1990, pp. 19–52. They claimed that rehabilitation benefits got watered down because treatments were applied haphazardly. According to their Risk Principle better results could be achieved (a) by matching the level of intervention to the risk that a prisoner will reoffend, and (b) by targeting specific needs of the offender which were likely to lead directly to criminal behaviour (such as drug addiction or unemployment). Note that as with CBT the interventions are fairly mechanistic, not concerned with less immediate concepts like dreams or childhood or lack of esteem.




Doris MacKenzie, an expert on experimental design, was put in charge of the penal rehabilitation section and her team came up with 184 studies which at least tried to make some comparisons between offenders who had been treated and those who had not. Of an astonishingly diverse field – from physical exercise through restitution and apology to group counselling1 – most could not be shown to work. Boot camps, ‘scared straight’ programmes and ‘tough love’ therapeutic camps came out badly – they are always found to make things worse.2

1Among the multitude of interventions, each of which needs testing on its own, and most both when applied in penal institutions and when used after release, are severe discipline, physical challenge, intensive supervision, various forms of apology or restitution, group counselling and therapy, individual counselling and therapy, family interventions, vocational training, job provision, academic training, CBT, a range of other behavioural and cognitive approaches, life skills training, probation and parole enhancements, intensive aftercare, diversion from crime projects, diversion from court projects, multimodal programs, and many others.

2 A meticulous review published by the Campbell Collaboration (December 2006) showed emphatically that ‘Boot camps do not reduce criminal behaviour’ []




What’s more the Maryland findings were soon supported by a Canadian review

Richard Seiter and Karen Kadela, ‘Prisoner Reentry: What Works, What Does Not, and What is Promising’, Crime & Delinquency, Vol. 49, No. 3, 2003, pp. 360–88.




Almost none of the tens of thousands of well-meant interventions across the whole of North America had reported any outcomes at all, and only two of all the schemes the Americans and Canadians had looked at had been properly tested by randomised trials.

Joan Petersilia, ‘What Works in Prisoner Reentry? Reviewing and Questioning the Evidence’, Federal Probation, Vol. 68, No. 2, September 2004.




Some statisticians thought we were trying to make a silk purse out of a sow’s ear.

John Copas, ‘What works? Selectivity models and meta-analysis’, Journal of the Royal Statistical Society, 162, Part 1, 1999, p. 107.




In 2003, in a hugely ambitious piece of work, two investigators from Missouri went through all the hundreds of published and unpublished academic literature on rehabilitation over the quarter-century between 1975 and 2001 and, excluding drug treatment programmes, found only nine credible academic papers.1 Three years later highly another rigorous review looked back even further, over half a century, and found only fourteen studies which they regarded as top-quality and well-controlled trials.2

1Richard Seiter and Karen Kadela, ‘Prisoner Reentry: What Works, What Does Not, and What is Promising’, Crime & Delinquency, Vol. 49, No. 3, 2003, pp. 360–88.

2 David Farrington and Brandon Welsh, ‘A half-century of randomized experiments on crime and justice’, Crime and Justice, Vol. 34, University of Chicago, 2006.




For a time community courts were the great hope of penal reformers. Based on a Community Justice Center in New York, they embody every reasonable liberal view of how to stop offending.1 Criminals are seen as problematic individuals who need help, and are embraced by as many agencies as possible to deal with what are thought to be their underlying problems. They go before dedicated courts and their drug use, housing, education or employment and their relationships are addressed, in the expectation that once they get their lives straight they will go straight. It didn’t work.2

1Founded in 2000 the Red Hook Community Justice Center in Brooklyn is an experiment in integrating criminal, civil and family courts into one along with a clinic and social work centre. A single judge hears evidence and can sentence an offender to restitution, short-term counselling, drug or mental health treatment or long-term psychotherapy. I have visited it and it looks impressive. The problem is there is no rigorous evaluation.

2 Lucy Booth, Adam Altoft, Rachel Dubourg, Miguel Gonçalves and Catriona Mirrlees-Black, ‘North Liverpool Community Justice Centre: Analysis of re-offending rates and efficiency of court processes’, Ministry of Justice Research Series 10/12, July 2012: ‘There was no evidence that the NLCJC was any more effective in reducing re-offending than other courts’, p. iii. []




Defenders claimed that ‘bald’ reconviction rates don’t tell the whole story. ‘We should look at other benefits,’ said the judge who presided over the experiment.

The Times, London, 28 July 2009, p. 18.




Some of the more impassioned supporters of rehab programmes are so resigned to negative results that they reject all sceptical evaluations. In Britain, some criminologists have denounced them as a capitalist excuse to impose more ‘social control and exclusion of those citizens deemed “intransigent” or “irresponsible”’.1 Others say it wholly unreasonable that treatment should be challenged by ‘methodological idealists’2 using ‘unimaginative evaluation [which] divert[s] resources from delivery’3

1Hazel Kemshall, ‘Effective Practice in Probation: An Example of “Advanced Liberal” Responsibilisation?’, Howard Journal of Criminal Justice, Vol. 41, 1, 2002, p. 41.

2 Quoted by Sarah Brown, Treating Sex Offenders: An Introduction to Sex Offender Treatment Programmes, Willan, Cullompton, Devon, 2005.

3Mark Israel and Wing Hong Chui, ‘If “Something Works” is the Answer, What is the Question? Supporting Pluralist Evaluation in Community Corrections in the United Kingdom’, European Journal of Criminology, Vol. 3, No. 2, 2006, pp. 181–200.





Another school of thought says quantitative assessments can never come up with ‘clinching evidence’ because of ‘a failure to recognise that work with offenders is a highly reflexive process [so] the effectiveness of interventions will be highly context-specific. What works in one culture at one time may be ineffective in other settings and at other times.’

In an unpublished and ‘unashamedly argumentative’ paper in 2008 for the National Audit Office, Reducing reoffending: getting off the treadmill, the criminologist Mike Hough put forward several plausible explanations for variable results, including that, too little attention is given to the ‘craft’ of working with offenders. He surmises that outcomes might be affected by differences in interventions, the quality of the prison, staffing continuity and the personal qualities of the staff, the mix of offenders, arrangements for resettlement, and other factors like resources available to the institution. While all this is valid, a similar list could be made for most medical studies, including the variable quality of doctors and nurses – so none of this is an excuse for bad science. Indeed, Mike Hough makes it clear that he is not just concerned with recidivism but with the ‘decency agenda’ and improving the quality of prison regimes.




Dr Beck refined his approach and called it cognitive behaviour therapy.

CBT is not the only such approach. Around the time that Aaron Beck was pioneering CBT, another US psychologist, Albert Ellis, was working on similar lines. He too suspected depressive patients had got into a vicious circle of negative thinking which was sabotaging their lives, and that they could be liberated by replacing those self-defeating thoughts with what he called ‘rational emotive behaviour therapy’, or REBT. Supporters of Dr Ellis claim he was first to come up with this very practical and outcome-focused form of psychotherapy. In fact there are now many flavours and brand names for improving rational thinking, but CBT is the term most widely used.




In the 1970s a charismatic psychiatrist, Samuel Yochelson, and a brilliant young clinical psychologist, Stanton Samenow, were investigating criminal behaviour with offenders in St Elizabeth’s psychiatric hospital in Washington DC … Offenders began to report changed attitudes and in gratitude some even threw parties in the hospital for the researchers. But, much to his consternation, Yochelson discovered the patients were continuing to commit crimes. In fact, the thank-you parties had been stocked with supplies stolen from the hospital.

For more on this, and a brief and immensely readable biography of Staunton Samenow, see C T Genre, Stanton Samenow, The Criminal Personality, online. []




It occurred to them that as we grow up some of us experience bad luck, or bad treatment at the hands of others, which gives us distorted views about ourselves. This often results in offenders regarding themselves as victims (‘the world is against me’) and blaming everyone other than themselves – even those they victimise (as in ‘he deserved it’).

That required getting offenders to reason differently, which led them to CBT.

The failure of psychoanalysis also led Yochelson and Samenow to rewrite the conventions on what caused criminals to behave as they do, and to make sure everyone knew about it. Their book, The Criminal Personality, was both perceptive and influential. The year it was published, in 1976, Samuel Yochelson died, but Stanton Samenow went on to write another persuasive volume, Inside the Criminal Mind, and continued to develop his theories of reform. See Samuel Yochelson and Stanton Samenow, The Criminal Personality (Vol. I, II, III), NY: J. Aronson, 1976, 1977, 1986; and Stanton Samenow, Inside the Criminal Mind, Random House, 1984, Revised and updated 2004.




… in 1979 two Canadian psychologists, Paul Gendreau and Robert Ross, reviewed the evidence and confirmed it seemed to work.

Paul Gendreau and Robert Ross, ‘Effective Correctional Treatment: Bibliotherapy for Cynics’. Crime & Delinquency 25: 1979 pp. 463–89.




Cognitive behaviour therapy was not much loved by psychoanalysts and is unsatisfying to those who think society, not the individual, is responsible for crime.

CBT is also deeply unsatisfying to those who believe antisocial behaviour stems from some long-forgotten trauma in the offender’s past, and for whom crime is a symptom of turmoil in what followers of Freud picture as a seething unconscious underworld. According to them CBT only treats the outward manifestation not the central problem.




When results were pooled from fifty-eight broadly comparable studies across the US and Canada, CBT showed a drop in rates of re-arrest or conviction, or revocation of probation or parole of 25 per cent.1 What is more, and rather surprising, is that the results were consistently better with serious offenders than with minor ones. A similar if smaller effect was found in Europe.2

1Nana A. Landenberger and Mark W. Lipsey. The positive effects of cognitive-behavioral programs for offenders: A meta-analysis of factors associated with effective treatment. Journal of Experimental Criminology, 2005, 1, pp. 451–76. Nana is a clinical psychologist who had conducted CBT with adult offenders in prisons. Mark is a public policy researcher and Director of Vanderbilt’s Center for Evaluation Research and Methodology in Nashville, Tennessee. [See also updated evidence from the Campbell Collaboration published April 2008:]

2Santiago Redondoa, Julio Sanchez-mecab & Vicente Garridoc, ‘The influence of treatment programmes on the recidivism of juvenile and adult offenders: An European meta-analytic review (sic)’, Psychology, Crime & Law, 1999, Volume 5, Issue 3. The paper reviewed thirty-two European studies of recidivism over an average follow-up of two years equivalent to a 12 per cent reduction in recidivism, mostly through behavioural and cognitive techniques. []




Whatever the reason, the assessors warned it was not clear that ‘the impressive effects on recidivism can be routinely attained under everyday circumstances’.

Nana A. Landenberger and Mark W. Lipsey study (see above), p. 472: ‘The amount of high quality research on CBT in representative correctional practice is not yet large enough to determine whether the impressive effects on recidivism can be routinely attained under everyday circumstances.’




But when CBT for recidivism was tried in Britain, a review conducted for the Home Office found its effect was ‘inconclusive… mixed and limited’.1 There were some positive outcomes but researchers warned that even those gains might not have been caused by the therapy2 and, in any case, ‘can be quickly lost’.3

1Gemma Harper and Chloe Chitty, ‘The Impact of Corrections on Re-Offending: A Review of “What Works”’, Home Office Research Study 291, Home Office, London, 2005, p. 47 and p. 77. [ ] This melancholy conclusion applies to a wide range of cognitive approaches and their attendant acronyms including ETS (Enhanced Thinking Skills), R&R (Reasoning and Rehabilitation), Think First, Priestley One-to-One , ASRO (Addressing Substance Related Offending), SOTP (Sex Offender Treatment in Prison and in the community), DID (Drink Impaired Drivers) and ART (Aggression Replacement Training in probation).

2John Wilkinson, ‘Evaluating Evidence for the Effectiveness of the Reasoning and Rehabilitation Programme’, Howard League of Criminal Justice, Vol. 44, Issue 1, 2005, pp. 70–85.

3 Ibid., Executive Summary, p. xx.




Even when results seem encouraging they tend to dissipate over time.1 Another wide-ranging official report, this time for the government in Northern Ireland, described the outcomes of all the British evaluations as ‘uniformly disappointing’.2

1A good example of an attempt at evidence-based practice was STOP (Straight Thinking on Probation) implemented by the Mid Glamorgan Probation Service in Wales in the mid-1990s – though sadly there was no randomised control group and so it was not possible to test the outcomes properly. At first the reconviction rates seemed lower than normal but after two years perceived gains had disappeared.

.2 Shadd Maruna, Vivian Francis, Mark Liddle, Lyndsay McAteer, Emma Watts, and Sam Wright, Reducing offending: A critical review of the international research evidence for the Northern Ireland Office, School of Law, Queen’s University Belfast and ARCS (UK) Ltd, Cambridge, March 2008.




But though the big Northern Ireland survey of British and American research concludes that the best schemes ‘can be at least modestly effective in reducing re-offending among ex-offenders’, as always when it comes to rehabilitation, not much research has ‘met the highest standards for programme evaluations’.

Ibid., p. 62. This review was compiled to inform policy in Northern Ireland where conventional crime has been rising since the end of the troubles. It is thorough and non-technical and a superb general reference on international and UK research into crime reduction, though sadly not widely available.




Whatever your anger at crime in general, you usually feel sorry for individual offenders when you meet them in their prison garb, and it’s plain that most of them could do with help. I have seen an audience melt with fellow-feeling and warmly applaud an orchestra made up of young offenders, some of whom were serving time for savage violence.

Public attitudes to crime are far more complex than tough or tender. Liberals can be full of forbearance about crime in general but vehemently angry and upset when confronted by an ugly incident of theft or violence. Conservatives can breathe fire and brimstone in the abstract but be full of sympathy for offenders they come to know individually and especially for friends or family who break the law.




It acknowledges that treatments can ‘increase reoffending rates’ and must be ‘evidence-based’, but then is selective in the evidence. It gives prominence to a US study that suggested reoffending rates can be cut ‘by an average of 17 per cent if delivered in custody and 35 per cent if delivered in the community’.1 But it chooses not to cite a study of its own which concluded that ‘there is a paucity of robust evaluations’ and that ‘the overall evidence remains tentative’.2

1 What Works in Offender Rehabilitation, NOMS, Ministry of Justice, London, May 2010. []

2 Veronica Hollis, Reconviction Analysis of Programme Data using Interim Accredited Programmes Software (IAPS), RDS NOMS, December 2007.

In sex offending the outlook is especially disappointing. In a review of official UK policy for the British Medical Journal a consultant forensic psychiatrist concluded gloomily, “No evidence from academic or policy research has shown that the treatment programme significantly reduces sexual reoffending. To be clear: convicted sex offenders are sent to prison, undergo this treatment programme, are deemed to have been somewhat rehabilitated, and are released to the public; however, they are as likely to offend as before receiving treatment. Their victims and the public deserve to know this.” David Ho, BMJ, 2015; 350 doi: (Published 27 January 2015), BMJ 2015;350:h199




A British psychologist, Iain Crow, summed up the state of affairs, and this from a man who has himself worked for years on community-based projects for offenders: ‘We are still a long way from being able to say what, if anything, does work. Indeed, it seems more likely that nothing works in a universal sense, but that some things can be effective in reducing the likelihood of reconviction for certain people in certain circumstances.’

Iain Crow, The Treatment and Rehabilitation of Offenders, Sage, London, 2001. ISBN:9780761960393




When you track what happens to twins or to adopted babies, you find inheritance moulds their personalities far more than had generally been supposed. Genetics accounts for the general tendency of antisocial parents to have antisocial children whether or not the children live with them.

Judith Rich Harris, The Nurture Assumption, Why Children Turn Out the Way They Do, The Free Press, New York, 1998. See also: Judith Rich Harris, ‘Socialization, Personality Development, and the Child’s Environments: Comment on Vandell’, Developmental Psychology, November 2000. Here Rich Harris makes a powerful rebuttal of her critics.[]

Politically motivated critics of genetic explanations for behaviour once routinely denounced heritability studies, calling for an end to the ‘dangerous idea’ that biology has anything to do with how people act. By the 1980s these complaints subsided in the face of overwhelming evidence for behavioural genetics but as recently as 2014 some sociologists were still claiming that twin studies in particular lead to ‘preposterous’ conclusions and ought to be abandoned (Callie Burt and Ronald Simons, ‘Pulling back the curtains on heritability studies: biosocial criminology in the postgenomic era,’ Criminology, 2014, vol 52, pp223-62). In a celebrated exchange their accusations provoked an even stronger riposte in the same journal which recruited mathematics, computer simulation and plain old-fashioned fact-checking to ridicule their analysis and accused the authors not just of logical flaws and dreadful selectivity but almost of academic dishonesty: of labeling ‘ideologues’ as ‘experts’, of concealing evidence which ‘flatly contradicts’ their allegations, and of ‘misrepresentations, anecdotes, distortions, and misquotes’ (J C Barnes, John Paul Wright, Brian Boutwell, Joseph Schwarz, Eric Connolly, Joseph Nedelec and Kevin Beaver, ‘Demonstrating the validity of twin research in criminology,’ Criminology, 2014, vol 52, pp588-626). Burt and Simons later conceded a ‘careless’ mistake but stuck to their own theory that crime is caused by ‘adverse social factors’ (Ibid, p655).




… as Steven Pinker puts it succinctly, ‘The greatest contribution parents make is in their genes and where they choose to set up home.’

Steven Pinker in conversation with the author, Cheltenham Science Festival, Brilliant Minds session, 8 June 2008.




As you might expect, risk factors include harsh discipline, poor supervision and lack of the mother or father’s involvement with their children, parental conflict, family breakdown, parental criminality or drug use, physical violence, sexual abuse and maternal deprivation. In particular mothers who were themselves brought up badly are likely to lack child-rearing skills with their own offspring.

These factors have emerged from several studies, not least the Cambridge Study in Delinquent Development which followed over 400 south London working-class boys born in 1953 until they were fifty, by which time 41 per cent of them had criminal convictions. Many factors were associated with criminality including low personal and socioeconomic attainment, but in terms of parenting, harsh home discipline was a strong predictor, followed by poor supervision. (David Farrington et al., Criminal Careers Up to Age 50 and Life Success Up to Age 48: New Findings from the Cambridge Study in Delinquent Development, Home Office Research Study 299, London, 2006.) This was reinforced by work in Nottingham by John and Elizabeth Newson, and in a classic study a long-term follow-up by Joan McCord in Boston found poor parental supervision was the best predictor of violent crimes as well as property ones.




Accordingly, pioneers in the 1970s tried teaching mums-to-be to be better at their task … Similar studies started in the 1990s in Tennessee and Colorado hint as similar improvements.





But the best support for early intervention comes from pooling fifty-five studies from across the world covering almost 10,000 under-fives who were exhibiting aggressive and troublesome behaviour. All of the trials were proper randomised controlled experiments – in each case arbitrarily assigning families to be left alone or to take part in a parent-training programme.

The three main interventions were the Incredible Years and Parent-Child Interaction Therapy, pioneered in the US, and Triple P which was developed in Australia. Some involved visits to the home; most were based round appointments at a clinic, school or place of work. They were all found to be equally effective.




The investigators concluded: ‘Family/parent training programs help reduce the number of little trouble-makers and in doing so, in the long term, the number of big trouble-makers.’

Alex Piquero, David Farrington, Brandon Welsh, Richard Tremblay, Wesley Jennings, Effects of early family/parent training programs on antisocial behavior and delinquency. Campbell Collaboration, 2008 []




The existence of so many rehab schemes is more a proof of hope than of effectiveness.

A friend has warned me against being so sceptical of rehab efforts which ‘almost everyone believes in … They can’t all be wrong.’ It reminded me of how William A. Tiller won the Randi Foundation’s first Pigasus Award in 1979 for saying that although the evidence for psychic events was very shaky and originates with persons of doubtful credibility, it should be taken seriously because there is so much of it.




You could try CBT and maybe helping with employment – and by all means do more research.

James McGuire, A review of effective interventions for reducing aggression and violence, Philosophical Transactions of the Royal Society, Vol. 363, May 2008, pp. 2577–97.




Or we could follow the advice of Shadd Maruna, a professor at my alma mater, the Queen’s University of Belfast: offer recidivists a wheelchair to renovate…
Many prisoners struggle to find a sense of meaning in their lives. They seek to fill empty lives with cocaine binges, joy rides and violence. Some find they can fill this void equally well with creative writing, raising children or even volunteering. This is how rehabilitation works. Or maybe not. If volunteer work fails as a panacea — as has every other magic bullet in offender rehabilitation — at least the world will have more respite care and more repaired wheelchairs.
Shadd Maruna, CSV annual Edith Kahn Lecture, House of Lords, London, 24 April 2007.




Doris MacKenzie, who led the US Justice Department ‘What Works’ inquiry on offenders in the 1990s, became even more gloomy, having spent a further fifteen years trawling through studies and statistics. She acknowledges that CBT might help, but for the rest she is now emphatic: ‘Deterrence, incapacitation and increased control do not reduce the future criminal activity of offenders and delinquents. Nor have programs targeting social opportunities such as employment.’

Doris Layton MacKenzie, ‘First do no harm: a look at correctional policies and programs today’, The 2011 Joan McCord Prize Lecture, Journal of Experimental Criminology, Vol. 9, No. 1, March 2013, p. 1. Professor MacKenzie concludes that we need ‘smarter ways to address crime problems’ with ‘new evidence-based solutions [rather than] the old philosophy’ (p. 15).




Those who had been given therapy grew up with more problems and committed more serious crime than those who had been left to their own devices.

Ibid. The classic study was by Joan McCord, the doyenne of US experimental criminologists who questioned the effectiveness of social programs championed by both liberals and conservatives by conducting a series of longitudinal studies, eventually concluding that many interventions can do unexpected harm and nothing demonstrably works. When, in 1996, the New York Times asked her what she would recommend for a twelve-year-old arrested for gang rape and murder she replied, ‘It’s almost impossible to imagine a good thing to do with a gun-toting twelve-year-old.’




Large-scale British and other European evaluations have proved as dispiriting as those in North America.

Johann Koehler, Friedrich Lösel, Thomas Akoensi, David Humphreys, ‘A systematic review and meta-analysis on the effects of young offender treatment programs in Europe’, Journal of Experimental Criminology, Vol. 9, No. 1, March 2013, p. 19–43.

Another review (for the Campbell Collaboration) compared data from over 9,000 anti-social youngsters, half of whom were randomly assigned to social skills training while the other half were not. Those who had been trained did develop more social competence but the positive effects were so small as to be almost inconsequential. Ominously, ‘some interventions produced outcomes that are even worse than in the untreated control group’. See Preventing Crime: What Works for Children, Offenders, Victims and Places, edited by Brandon Walsh and David Farrington, Springer (The Netherlands), 2006, pp. 33–54. The chapter by Friedrich Lösel and Andreas Beelmann is also an excellent reference source for other publications on child social skills training.




Changing people is difficult stuff, a far remove from simplistic slogans like ‘reforming’ or ‘correcting’ criminals. As with Humpty Dumpty, once people’s personalities have been damaged, all the king’s courses and all the king’s men find it hard to put them together again.1 Or as Andrew Bridges, England’s chief inspector of probation, put it, whatever we do we can only make a 10 per cent difference to reoffending rates, ‘at best, at very best’.2

It is not as though we haven’t tried to change people, even unethically. In 1977 the US Senate Select Committee on Intelligence revealed classified mind-bending programmes codenamed Bluebird, Artichoke and MKUltra which were led by the CIA and involved all three armed forces as well as scores of academics. For decades they tried drugs, hypnosis, sensory deprivation and electric shock to influence human behaviour. Nothing worked. Even the simple idea of a truth serum turned out to be fanciful. If people can’t be manipulated with all the might of the world’s biggest military machine, working without moral compunction and using anything and everything at hand, are we to believe that a dose of criminal justice rehab can do the trick?
John Lancaster and James P Rathmell, The Moralist, A Medical Revolutionary and his Secrets.

Or as Andrew Bridges, England’s chief inspector of probation, put it, whatever we do we can only make a 10 per cent difference to reoffending rates, ‘at best, at very best’.

Andrew Bridges, HM Chief Inspector of Probation, Partnership in Practice conference, Manchester, 9 May 2007.