The revolving door at theprison gate:Exploring the dramatic increase inrecalls to prison
NICOLA PADFIELD AND SHADD MARUNA
University of Cambridge, UK and Queen’s
University Belfast, UK
In this article we draw attention to the recent and extraordinaryincrease in the number of people in England and Wales recalled toprison during the licence period of their sentence (by examiningthe published Parole Board and prison statistics). This is followed bya description of the existing law and the recent changes to it,which we suggest will exacerbate the current trend. We seek thento explain the increase by looking primarily at the US experience(which reveals a system which is costly, discriminatory andapparently ineffective at reducing crime) and at recent judicialreview cases (which reveal a system which is increasinglyacknowledged to be unfair), concluding that current sentencing lawand practice puts inappropriate emphasis on ‘front door’sentencing practices rather than the equally important ‘back door’practices of release, supervision and recall. Unsurprisingly, thearticle ends with a call for much more research in this area.
new penology • parole • recall • re-entry • resettlement
Criminology & Criminal Justice© 2006 SAGE Publications
(London, Thousand Oaks & New Delhi)and the British Society of Criminology.
www.sagepublications.comISSN 1748–8958; Vol: 6(3): 329–352
The current trend in England and Wales is towards a sentencing frameworkbased on a detailed analysis of different levels of offence seriousness.1 Yetthis discussion largely ignores the reality of a flexible early release systemfor prisoners. As a result, despite a carefully calculated initial sentence, anoffender will actually serve a very different amount of time in prison. Thisis sometimes referred to as the difference between ‘front door’ and ‘backdoor’ practices in corrections, and ‘back door’ release decisions can have asmuch influence as ‘front door’ sentencing practices in terms of sentencelength and maintaining overall prison populations (Tonry, 2003).
This article focuses on an additional ‘door’ of concern to observers ofcorrectional trends: the so-called ‘revolving door’ involving releasedprisoners who are recalled to prison during the ‘licence’ period of theirsentence. On licence, former prisoners need not commit a new criminaloffence in order to be returned to prison, but instead can find themselvesback inside when there is evidence that they have violated the terms oftheir licence (for example, failed to turn up at specified meetings). Thisprocess of prison ‘recalls’ has received very little scholarly or publicattention in England and Wales. Yet, in the last four years, the number ofsuch recalls in England and Wales has increased three or fourfold. More-over, the body responsible for releasing all such recalled prisoners, theParole Board, rightly anticipates that this trend ‘can only increase’ withthe release provisions of the Criminal Justice Act 2003.2 This dramatic riseseems to necessitate a discussion regarding the theoretical, legal andpractical issues related to recalls, which we hope to initiate with thefollowing discussion.
Some statistics on prison recalls
The population in custody in England and Wales stood at 77,421 on 25November 2005.3 In 2002, the average prison population was 70,860,which itself represented an increase of 16 per cent compared to 1997 and55 per cent compared to 1992.4 While there has been much concern aboutthis increase, little comment has been made about the fact that a growingnumber of this population are there not only because they have beensentenced by a court, but because they have been recalled to custody aftera period on licence.
The public body that first highlighted this concern was the Parole Board;from its perspective the concern is primarily one of workload. Table 1,which is taken from tables in Annual Reports of the Parole Board, showsthe extraordinary increase in the overall workload for the Board in recentyears.
These figures show the total number of cases, which slightly overstates
Criminology & Criminal Justice 6(3)330
the position in terms of individual offenders since the same prisoner maychallenge his recall more than once: see Table 2.
Table 3 provides a summary of DCR5 recall cases.The Prison Statistics 2002 and the Offender Management Caseload
Statistics 2004 confirm the number of DCR recalls, but not the overallfigures. We are told that in 2004–5, 710 parolees were recalled, represent-ing 17 per cent of those on licence; in 2002–3, 420 paroles (13 per cent ofthose on licence) were recalled, of which under 6 per cent were recalled forcommitting a further offence. This does not include the numbers recalledwho were released under Home Detention Curfew.6 More details areoffered on these prisoners (perhaps because the Prison Service/NationalOffender Management Service (NOMS) make the decision to release?).Some 19,310 offenders were released early from prison on home detentioncurfew (HDC) in 2004, and a total of 3000 (16 per cent) were recalled toprison. (In 2002, 20,525 offenders were released on HDC, and only 7 per
Table 1. Parole Board workload (2000–5)
2000/1 2001/2 2002/3 2003/4 2004/5
Recalls 2457 4885 7246 9031 9320Oral hearings 272 466 495 1018 1341Mandatory lifer prisoners 531 513 915 1060 625Determinate sentence prisoners 5576 5514 6012 6038 7297Total caseload 8836 11,378 14,668 17,147 18,583
Source: Parole Board (2003: 4, 2005: 7)
Table 2. Summary of total recall cases 2004/5
DCRa (long-term)prisonersb Total
Recall requests 6378 1884 8262Reps after recall 749 309 1058Total cases 7127 2193 9320
Source: Sentence Enforcement Unit (from Parole Board (2005: 53)a Under the pre-Criminal Justice Act 2003 regime, which applied to those sentenced before 4April 2005, those sentenced to less than 12 months’ imprisonment were subject to AutomaticUnconditional Release at the halfway point in their sentence. Prisoners were then at risk ofbeing returned to serve the rest of their sentence if they were convicted of further imprisonableoffences before their sentence had fully expired, but are not subject to compulsory supervision.On the other hand, those sentenced to between 12 months and under 4 years were released atthe halfway point on Automatic Conditional Release (ACR). They were supervised on licenceuntil the three-quarter point in their sentence. Those sentenced to four years or more could begranted Discretionary Conditional Release (DCR) by the Parole Board from the halfway pointof their sentence. If early release is not granted, these prisoners are released automatically atthe two-thirds point in the sentence. They remain under supervision until the three-quarterpoint in their sentenceb Includes both prisoners recalled from parole licence and those recalled from non-parolelicence after automatic release at the two-thirds point of their sentence
Padfield & Maruna—The revolving door at the prison gate 331
cent were recalled to prison.) The most common reason for recall, whichaccounted for 57 per cent of all recalls (54 per cent in 2002), was breachingHDC conditions. This includes being absent from the curfew addressduring curfew hours, threatening monitoring staff, damaging the monitor-ing equipment or failing to be present for the installation of a newtelephone line or equipment. Others were recalled on the grounds that it isnot possible to monitor them: ‘change of circumstances’ (for example,where a subject has involuntarily lost their curfew address or has with-drawn consent to be monitored), which account for 26 per cent of allrecalls in 2002; ‘installation failure (where it is not possible to install themonitoring equipment or make the monitoring equipment fully opera-tional); or ‘monitoring failure’ (where it becomes impossible to continuemonitoring, for technical or other reasons). These last two accountedtogether for less than 1 per cent of recalls. In 2002, there was one HDCrecall on the grounds that the subject posed a risk of serious harm to thepublic. The key point for our purposes is that only 16 per cent of recalls inboth 2002 and 2004 were on the grounds of being charged with a newoffence.7
The small number of life sentence prisoners, who make up an increasingproportion of the prison population, who are recalled is also growing,8 asTable 4 shows.
Thus, both Prison Service and Parole Board data show very significantnumbers (and an increasing number) of prisoners recalled to prison,whether these are short-term or longer-term prisoners.
Table 3. Summary of DCR recall cases
Total no. ofprisoners recalledfor further offence
% of thoserecalled forfurther offence
Recalls as % ofaverage numberof parole
1997/8 79 3.4 190 8.21998/9 94 4.0 233 11.11999/0 93 3.8 250 10.12000/1 106 3.8 267 9.62001/2 90 3.0 329 10.92002/3 188 5.8 420 13.12003/4 252 7 601 16.62004/5 265 6.5 712 17.4
Source: Parole Board (2003, 2004, 2005)
Table 4. Life licensees recalled to prison
1999/0 2000/1 2001/2 2002/3 2003/4 2004/5
48 35 26 30 52 90
Source: Parole Board (2005: 59)
Criminology & Criminal Justice 6(3)332
The law on prison recalls
The current law: 1991–2005
Parole was first introduced in England and Wales in the Criminal JusticeAct 1967. It had two main objectives at that time: reducing the prisonpopulation and aiding in the rehabilitation of offenders by releasing theminto the community at the ‘right’ time and with the support and supervisionof a probation officer to whom they were required to report regularly.Recall was always a part of this system. A new system of early release wasintroduced by Part II of the Criminal Justice Act 1991 and, until theprovisions of Part 12, Chapter 6 of the Criminal Justice Act 2003 werebrought into force, the law on recalls to prison was governed by the 1991Act, as amended by the Crime (Sentences) Act 1997, and, more im-portantly, by the Crime and Disorder Act 1998 as well as by the Powers ofthe Criminal Courts (Sentencing) Act 2000, Schedule 9, para. 137.
We have already noted the system of Automatic Unconditional Release,Automatic Conditional release and Discretionary Conditional release.9
However, the apparent simplicity of the early release scheme was compli-cated by the fact that judges had a variety of sentencing options. Forinstance, some offenders received longer than commensurate sentences.Some sex offenders remained on licence until the end of their sentence onthe recommendation of the trial judge. There were complex rules applyingto such ‘extended sentences’, and, as demonstrated in the last section, anincreasing number received life sentences. The key provision on recall wass. 39 of the Criminal Justice Act 1991,10 which provided that:
(1) If recommended to do so by the Board in the case of a [short-term or]11
long-term prisoner who has been released on licence under this Part, theSecretary of State may revoke his licence and recall him to prison.
(2) The Secretary of State may revoke the licence of any such person andrecall him to prison without a recommendation by the Board, where itappears to him that it is expedient in the public interest to recall that personbefore such a recommendation is practicable.
(3) A person recalled to prison under subsection (1) or (2) above—
(a) may make representations in writing with respect to his recall; and
(b) on his return to prison, shall be informed of the reasons for his recall andof his right to make representations.
(4) The Secretary of State shall refer to the Board—
(a) the case of a person recalled under subsection (1) above who makesrepresentations under subsection (3) above; and
(b) the case of a person recalled under subsection (2) above.
(5) Where on a reference under subsection (4) above the Board—
(a) . . .
Padfield & Maruna—The revolving door at the prison gate 333
(b) recommends in the case of any person,
his immediate release on licence under this section, the Secretary of Stateshall give effect to the recommendation.
. . .
(6) On the revocation of the licence of any person under this section, he shallbe liable to be detained in pursuance of his sentence and, if at large, shall bedeemed to be unlawfully at large.
It was not long before the framework of the Criminal Justice Act 1991was amended. The Criminal Justice and Public Order Act 1994 made theParole Board a non-Departmental public body (see s. 149) and at the sametime took away the Board’s power to recall ex-prisoners (see s. 150). Recallbecame an executive decision, subject to reconfirmation (or otherwise) bythe Parole Board. The Crime (Sentences) Act 1997 which then followedwas a rather desperate attempt by a flagging Conservative government tobolster confidence in the criminal justice system by ensuring that offendersserved the whole of their sentences. As the White Paper Protecting thePublic stated, ‘offenders sentenced to custody should serve the full termordered by the court’ (Home Office, 1996: para. 9.1). But ss. 8–27 werenever implemented. The provisions, which would have given an un-structured power to prison officers to decide how many days’ early releasea prisoner might earn, were unworkable in practice. There was, of course,also a general election and the Conservative government’s 18 years inpower were over.
The Crime and Disorder Act 1998, ‘New’ Labour’s first attempt (ofmany) at criminal justice legislation then introduced extended sentenceswith longer periods of supervision (s. 58). These were available for avariety of violent offences, but clearly targeted sex offenders (Padfield,1998: 62). Sections 99–100 inserted new s. 34A, s. 37A, s. 38A into theCriminal Justice Act 1991 to provide the power to release short-termprisoners on a ‘home detention’ curfew licence (enforced by electronicmonitoring) for up to two months before their normal date of release. Thiswas first introduced in January 1999, governed by PSO 6700.
Since then, the rules have been changed more than once: since 14 July2003 home detention curfew (HDC) eligibility has been increased from 90days to 135 days.12 It is beyond the scope of this article to explore in detailthe way HDC works, but it should be noted that there are inconsistentviews on its purpose (reducing the prison population or rehabilitation/resettlement?), and wide variations in its use between prisons (local prisonshave a surprisingly low release rate, given their overcrowding) (Dodgsonand Mortimer, 2000; Dodgson et al., 2001). Section 103 transferredresponsibility for recalling short-term prisoners from the courts to theParole Board. Section 104 provides for offenders who are recalled to prisonwhile on licence to be subject to a period of supervision until the end oftheir sentence following their second release from prison at the three-quarters point of sentence.
Criminology & Criminal Justice 6(3)334
The Criminal Justice Act 2003
Part 2 of the Criminal Justice Act 1991 on the early release of prisoners isrepealed by s. 303 of the Criminal Justice Act 2003, which largely cameinto force on 4 April 2005, when Part 12, Chapter 6 of the Act (Releaseand recall of fixed term prisoners (prison sentences over 12 months)) cameinto force.13 The number of prisoners who will be recalled is likely to growsignificantly once these provisions begin to bite.
Sections 237–68 deal with the arrangements for prisoners’ early releaseon licence, recall to prison following breach of licence requirements andfurther re-release. There are also provisions for calculating (pre-trial)remand time, calculating how sentences should be served and drug testingrequirements on licence. Thus under s. 238, a court may recommendlicence conditions for those sentenced to more than 12 months. Sections244 to 253 provide for the release of offenders from custody: offendersserving sentences of 12 months or more will be released automatically onlicence at the halfway point of their sentence (subject to early release onhome detention curfew (HDC) which will be extended to the vast majorityof fixed-term prisoners (see s. 246)). Upon release, the second half of theirsentence will be subject to standard licence conditions and any combinationof the additional prescribed conditions that the Secretary of State maydetermine by order. For new custodial sentences of 12 months or more,licence conditions may be imposed right up to the end of the sentence.Offenders will be liable to recall up until the end of their sentence. TheSentencing Guidelines Council suggest that sentence lengths imposed bysentencers from this time should be reduced ‘by in the region of 15 per cent’to take account of the increased demands that sentences will make onoffenders. Time will tell whether they heed this advice. It seems likely thatnot only will the number of recalls increase; so will, in effect, sentencelengths.
Under ss. 254 and 255 recall becomes more clearly an executivedecision—by the prison and probation services—rather than by the ParoleBoard, as at present. The offender will have the right of appeal to theParole Board, and even if the offender chooses not to exercise this right theParole Board will none the less scrutinize all recall decisions. This removesthe anomaly by which the Parole Board currently both advises on recallsand acts as an appeal body against those same recalls.
Section 257 confirms the Secretary of State’s controversial power to adddays to prison sentences, under prison rules, where disciplinary offences arecommitted while in custody. Sections 259 to 262 then provides for a newearly removal scheme from prison for foreign national prisoners liable toremoval from the UK. Sections 263 and 264 set out the principles forcalculating the time offenders must spend in custody and on licence whereseveral sentences are passed on the same or different occasions, and areordered to be served concurrently (at the same time) or consecutively (oneafter the other).
Padfield & Maruna—The revolving door at the prison gate 335
These changes to the rules on recall need to be seen in the wider contextof the very different sentencing framework, which came into force on 4April 2005. Courts will be imposing ‘imprisonment for public protection’,an indeterminate sentence, on a much wider category of ‘dangerousoffender’ than would previously have been sentenced to a discretionary lifesentence: see ss. 224–36 of the Act. Thus, if there are many more peoplesentenced to indeterminate sentences, the number of them released and thenumbers recalled will also increase. It is here that the Parole Boardanticipates another major increase in its workload.14
Recalls in practice
A prisoner who is eligible to be released under HDC will be subject toeither ‘Presumptive HDC’, or will be subject to an ‘Enhanced assessment’.Either way, if he or she is recommended for release early on HDC, they willbe released subject to standard licence terms, plus any specific ones theBoard or Governor decide to impose. Similarly, a DCR prisoner releasedearly by the Parole Board may also have their licence conditions ‘tailored’to their particular case.
A standard licence includes the following conditions:
. . . While under supervision you must:
(i) keep in touch with your supervising officer in accordance with anyreasonable instructions that you may from time to time be given;
(ii) if required, receive visits from your supervising officer at your home atreasonable hours and for reasonable periods;
(iii) live where reasonably approved by your supervising officer and notifyhim or her in advance of any proposed change of address;
(iv) undertake only such employment as your supervising officer reasonablyapproves and notify him or her in advance of any proposed change inemployment or occupation;
(v) not travel outside the United Kingdom without obtaining the priorpermission of your supervising officer (which will be given in exceptionalcircumstances only);
(vi) be of good behaviour, not commit any offence and not take any actionwhich would jeopardise the objectives of your supervision, namely toprotect the public, prevent you from re-offending and secure your successfulreintegration into the community; . . .
Additional licence conditions may include:
• attendance at appointments with a named psychiatrist/psychologist/medicalpractitioner and co-operation with recommended care or treatment;
• not to take work or organized activities with people under a certain age;• a residence condition;
Criminology & Criminal Justice 6(3)336
• a requirement not to reside in the same household as children under aspecified age;
• a requirement not to approach or communicate with named people;• a requirement to avoid a particular area;• a requirement to address alcohol/drug/sexual/gambling/solvent abuse/anger/
debt/offending behaviour problems at a specified centre;• a drug testing condition.15
While the focus of this article is those who are recalled without havingbeen charged with another offence, it is worth pointing out the complexityof the law on the re-sentencing of those whose parole licence had beenrevoked administratively and who then appear for sentencing for anoffence committed during the second half of the sentence. There have beena number of successful appeals when offenders have essentially beenordered to serve a part of their sentence twice over.16 In these cases, asentencing court has failed to make an appropriate allowance for the timespent in custody following revocation in determining the period for whichthe individual is ordered to return to custody under s. 116 of the Powers ofCriminal Courts (Sentencing) Act 2000.
Explaining the increase
There is a variety of factors that probably contributed to the suddenincrease in recalls to prison in the last few years.17 One possible explana-tion, however, can be fairly safely ruled out: it is probably not the casethat the behaviours of ex-prisoners have changed substantially over the lastfew years.
Despite dozens of calls for reform of the release/resettlement processover the last century,18 the process of prisoner reintegration remains afraught, complicated and near-impossible ideal to realize. The systemremoves already vulnerable citizens from their spouses, homes and familiesof origin;19 takes away their legal means of employment (if they had anyprior to incarceration) and sends a clear signal of suspicion to any futureemployers;20 puts them in an environment in which they are surrounded byothers in similar circumstances; then, releases them back into the commun-ity with little more than a bus pass, £50 in gate money and a colossalstigma (see Petersilia, 2003 for a US-based analysis of this issue; for theBritish experience of prison release, see Maruna, 2001). Yet, all the while,they are expected to avoid any temptation to improve their lot throughillegal means. It is an almost impossible situation for many, but what isimportant for the present argument is that it is nothing new. According toVerne McArthur, in his book Coming Out Cold: Community Reentry froma State Reformatory, ‘the released offender confronts a situation at releasethat virtually ensures his failure’ (1974: 1). That was written in 1974. Inthe book After Prison—What? Maud Booth writes,
Padfield & Maruna—The revolving door at the prison gate 337
when one thinks that this prejudice and marking of discharged prisonersrobs them of any chance of gaining a living, and in many instances forcesthem back against their will into a dishonest career, one can realize howtruly tragic the situation is.
That was written in 1903. There is no reason to suspect that the dramaticrise in recalls over the last three years has anything to do with a dramaticchange for the worse in these release conditions.
Likewise, it would be difficult to blame the increase on a worsening ofex-prisoners’ actual behaviour. In his account of changes in recall rates invarious US states like California, Kevin Reitz is particularly insistent onthis point:
Simply put, it is a serious error to equate failure rates on post releasesupervision with the actual behavior of prison releasees. . . . In any jurisdic-tion, the number and rate of revocations depends to some degree on thegood or bad conduct of parolees, to be sure, but it also depends at least asmuch on what might be called the ‘sensitivity’ of the supervision system toviolations. Sensitivity varies with formal definitions of what constitutes aviolation, the intensity of surveillance employed by parole field officers, theinstitutional culture of field services from place to place, and the severity ofsanctions typically used upon findings of violations.
(Reitz, 2004: 215)
Similarly, in the British context, the source of the change in recall ratesis likely to be found in enforcement practices rather than releasee behav-iour. Most obviously, the change from a court-based process to an exec-utive function has made prison recalls much easier to administer, savingthe State the time and trouble of due process considerations. In addition,the State has greater surveillance technologies, like urine testing andelectronic monitoring, at its disposal for the detection of violations.Likewise, the increase in multi-agency work, in particular partnership workwith the police, has sharply improved the surveillance capabilities oftraditional aftercare (Nellis and Chui, 2003).
The most substantial change, however, has probably been the centrallyled and politically driven transformation of the culture and practice ofprobation from a social service orientation to a surveillance-led focus onpublic protection. First, in recent years, there has been a sharp increase inthe number and stringency of licence conditions. Obviously, the moreconditions, the more likely that someone will be found to be in breach ofthose conditions.21 At the same time, there has been an explicit de-emphasisof the agent–client interpersonal relationship and social work methodology,and a decrease in the flexibility in case management (Nellis, 1999: 302–23).In a recent press statement, for instance, a representative of the NationalAssociation of Probation Officers blamed the growth in recalls to prison inthe UK on the introduction of strict enforcement rules by the Home Officesince 1999:
Criminology & Criminal Justice 6(3)338
For the last four years, Probation staff have been obliged to follow strictnational standards on enforcement. Previously, there was greater pro-fessional discretion. The clear consequence of the change in policy has beena huge rise in the number of recalls and breaches, and, therefore, the prisonpopulation.
(Fletcher, 2003: 1)
As indicated by Fletcher’s remarks these changes are being imposed fromabove rather than emerging from above. Changes in practice are the resultof a growing climate of managerialism in the service alongside a morepolitically motivated attempt to present a ‘tough’ image to the public(Nellis, 1999). In times of increasing individual and departmental account-ability, probation naturally becomes a more risk-averse enterprise. Better toerr on the side of over-caution than to risk the media attention that mightsurround a probationer who is released from prison early and commits aheinous crime while allegedly under state supervision. Indeed, risk assess-ment and risk management have been said to replace the traditional casemanagement model in probation. In fact, some have argued that theincreased use of actuarial risk prediction instruments, facilitated by asystem-wide computer network, has encouraged probation staff to viewpersons on licence less as individuals and more as members of un-differentiated risk groups (see Lynch, 1998). Those deemed ‘high’ risks areimmediately suspect and allowed little leniency in behaviour.
In his history of parole in California, Jonathan Simon describes many ofthese trends as a move towards a ‘waste management model’ of penality orthe ‘New Penology’ (Simon, 1993). According to this model, a dangerousclass of ‘lifetime correctional clients’ with ‘no realistic potential’ to reformis ‘treated as a kind of toxic waste’ to be contained. In addition to thisinstitutional cynicism about the redeemability of this criminal class, the‘New Penology’ is also thought to be characterized by three primaryfeatures:
1 discourse emphasizing risk rather than reformation or justice;2 objectives of offender management and classification rather than punishment
or rehabilitation;3 new techniques of classifying and managing risks from widespread drug
testing to the use of statistical/actuarial risk prediction instruments.
Central to the ‘New Penology’ is the assumption that there is ‘a specialsubpopulation of dangerous offenders whose identification and neutral-isation would result in dramatic reductions in the overall crime ratewithout resulting in a massive increase in the [prison population]’ (Simon,1996: 26).
Importantly, the New Penology concept originated in California, whichis unique even in terms of the United States and is most certainly animperfect theoretical framework for understanding the contemporary scenein the United Kingdom. In particular, the renewed focus on rehabilitationand reducing re-offending in the UK makes the situation here almost the
Padfield & Maruna—The revolving door at the prison gate 339
opposite of what has been happening in California over the last threedecades. That said, certain echoes can be heard in the criminal justicesystem of England and Wales with the move to a National OffenderManagement Service. Indeed, in the key Home Office report behind thismove, Managing Offenders, Reducing Crime, Patrick Carter cites with onlysomewhat guarded enthusiasm (and no research evidence) the idea that100,000 persistent offenders commit 50 per cent of all crime and ‘if wecould identify and incapacitate the 100,000 persistent offenders, crimecould fall dramatically’ (Carter, 2003: 15).
Why recall matters: lessons from the USA
How far could the rise in the use of recalls in England and Walesconceivably go in a climate such as this? As with many trends in criminaljustice and elsewhere, one can answer this question partially by looking tothe experience of the United States where the use of prison recalls hasreached remarkable new heights. In 1985, around 70 per cent of paroleessuccessfully completed their term of supervision following prison; however,by 1997, that number had dropped to 44 per cent. As a result, by the late1990s parole violators constituted nearly two-fifths of prison admissionsacross the country (compared to around 18 per cent in 1980) (Travis andPetersilia, 2001: 291–313). Fewer than one in three of these recallsinvolved the commission of a new crime, and three out of five admissionswere reported as ‘technical violations’ of parole conditions (usually a ‘dirty’urine test) (Sabol et al., 2001).
As with most such discussions, it is important to distinguish betweendifferent states within the United States. In this case, California is theextreme outlier. In 1997, two-thirds of the individuals admitted to prison inCalifornia were parole violators, compared to less than 15 per cent ofadmissions in many other comparable states (Petersilia, 1999). The state ofMassachusetts, for instance, had a parolee ‘success’ rate of over 80 percent, whereas in California only one in five parolees avoided being sentback to prison during their time on parole supervision (Glaze, 2002). Ofthese California recalls, the majority (around 57 per cent) were fortechnical parole violations and only 12 per cent for new felony convictions(Travis and Lawrence, 2002). Because California has such a large popula-tion, these numbers can skew the overall US averages masking the fact thatthe rest of the country has not undergone anything like the experimenttaking place in the Golden State, where Simon based his original examina-tion of the ‘New Penology’.
This substantial diversity in practice across states provides an idealopportunity for comparative research on the effects of differential super-vision practices. Although this research is only beginning, early resultssuggest that the usage of high rates of prison recalls as a strategy for crimecontrol in states like California has been highly costly, discriminatory and
Criminology & Criminal Justice 6(3)340
apparently ineffective as a strategy for crime reduction (see especially Re-Entry Policy Council, 2005).
There has been considerable discussion about the way that the excessiveuse of incarceration in the United States over the past 25 years hasoverloaded parole departments. When over 2 million citizens are in-carcerated at any given time, one should not be surprised when 600,000 ormore prisoners are released each year into the care of badly stretchedprobation and parole offices (Petersilia, 1999). Because of severe budgettightening at the state level, the systems of probation and parole have notbeen able to expand to meet this increasing need. As a result, the averageparole officer’s caseload has swelled from around 45 parolees in the 1970sto around 70 per parole officer in the current climate, making oldercasework models a near impossibility (Travis and Petersilia, 2001).
However, the parole system in states like California has not simply beena helpless victim to the growth in prisoner numbers, but has itself been oneof the main drivers of this growth. Indeed, the number of new prisonadmissions (i.e. those sentenced by the courts) largely levelled off inCalifornia as it did across the country (even declining in several states)during the 1990s. The growth in prisoner numbers, then, is a product oflonger sentences, tighter release policies and, of course, the explosion inrecalls to prison. More than one-third of California’s prison populationnow consists of parolees who have been returned to prison, with nearly29,000 beds occupied by the technical parole violators alone. Unsurpris-ingly, then, high rates of prison recall have been hugely expensive. Researchin California revealed that the State paid almost $900 million to re-incarcerate parole violators. The study’s authors estimated that reducingthe share of parole violators by half to one-third of all admissions wouldsave California $500 million a year (Travis and Lawrence, 2002).
As in the current practice in England and Wales, in most US states the recallprocess is administrative and non-judicial. Parole officers typically have thepower to remand an individual to prison almost immediately on suspicionof a rule violation. Although this flexibility allows the State to avoid thedelays of the court system, a substantial cost is being paid in civil liberties.Parole revocation in many US states essentially involves a parallel system ofcriminal adjudication with lower burdens of proof and lesser adversarialprocess.
Unsurprisingly, research has demonstrated a class and racial bias in these‘technical violations’ of probation and parole. For instance, in a large-scaleexamination of probation violations, M. Kevin Gray and his colleaguesfound that although race was not a significant predictor of whether aprobationer would commit a new crime, it was highly significant in
Padfield & Maruna—The revolving door at the prison gate 341
predicting which probationers would be technically violated (Gray et al.,2001: 537–57). Minority probationers and those with low educationalattainment were more likely to be found in violation of probation ordersthan other probationers, and this relationship was disproportionate to therelationship between actual recidivism and these characteristics (see alsoIrish, 1989; Whitehead, 1991).
Despite these costs (in resources and fairness), one could argue that a highrate of prison recalls could be an effective tool for reducing the re-offendingrates of prisoners. Like ‘broken windows’ policing, premised on thesymbolic value of prosecuting minor crimes of disorder (vandalism,vagrancy), the high use of recalls sends a clear message of zero tolerance tothose released from prison that might also impact recidivism (Wilson andKelling, 1982, but note that the evidence in support of this theory ofpolicing is mixed at best: Taylor, 2001). In a ‘broken windows’ probationmodel, the recall process might allow the State to identify and incapacitatethose persons who, by virtue of engaging in high-risk behaviours, are mostlikely to commit a new crime (Kleiman, 1999).
Yet, there is very little empirical evidence in support of this hypothesis.According to the recently released, 650-page Report of the Re-Entry PolicyCouncil: ‘Notably, there is currently no conclusive research indicating thatnoncompliance with technical conditions of release signals a person’slikelihood of criminal behaviour or that returning such individuals toincarceration might prevent future crime’ (Re-Entry Policy Council, 2005:391). In their essay ‘Does Supervision Matter?’, Piehl and LoBuglio explainwhy this is:
Ideally, to test this proposition, researchers would randomly assign a pool ofsoon-to-be released prisoners to either a treatment group that would providepost-release supervision or to a control group that would have no super-vision, and compare the rates of criminal activity across the two groups.Unfortunately, there is an inherent problem with this design: the outcome—recidivism—is intrinsically linked with supervision. In practice, increasedsupervision will likely lead to greater detection of rule violations and of newcriminal offences.
The best existing evidence to test a ‘broken windows’-type hypothesis,however, is not at all encouraging for the theory’s advocates. Petersilia andTurner’s nine-state random-assignment evaluation of intensive supervisionprobation found considerable evidence that increased surveillance leads togreater numbers of technical violations. However, the authors found noevidence that this higher rate of violations led to a decrease in recidivism oran increase in public safety. For instance, states like Washington that usedfar fewer technical violations to send parolees back to prison did not see an
Criminology & Criminal Justice 6(3)342
increase in parolee dangerousness as a result (Petersilia and Turner, 1993:281–335).
Why recall matters: prisoners’ rights
The current system applied in England and Wales can be profoundly unfair.Offenders are recalled to prison, of course, without notice. We have notedsome of the reasons, which range from a failure to keep appointments withprobation officers to committing a serious crime. But what then happens?They are taken back to their local prison, from where they may seek tochallenge the lawfulness of their detention, although this is not an easy taskfor someone now back inside. Insight into the complexity and protractednature of these attempts to challenge is provided by a reading of recentjudicial review cases. We’ll look at three.
Justin West22 was sentenced in October 2000 to three years’ imprison-ment for affray. He was therefore a short-term prisoner, subject to ACR,and was released at the halfway stage in his sentence on 6 August 2001 (hehad presumably been remanded in custody for some 8 months prior tosentence). The licence conditions (largely in the standard form) requiredhim to:
• keep in touch with his supervising officer in accordance with any reasonableinstructions;
• live where reasonably approved by his supervising officer;• be of good behaviour, not to commit any offence and not to take any action
which would jeopardize the objectives of his supervision, namely to protectthe public, prevent him from re-offending and secure his successful re-integration into the community.
Ten days after his release a senior probation officer reported that MrWest had allegedly assaulted his former partner, but the victim would notconfirm the incident. He then failed to keep his appointment with hissupervising officer, and on 22 August 2001, the Home Secretary revokedhis licence under s. 39(2). He was arrested on 24 August and returned toprison. The Home Secretary referred the case to the Parole Board unders. 39(4)(b). On 27 September his solicitors made written representations tothe Board under s. 39(3) urging the Board to hold an urgent oral hearingto be attended by witnesses whose evidence should be heard on oath. TheParole Board rejected these representations in writing dated 2 October2001. He was therefore detained in prison for another eight-and-a-halfmonths, until 9 May 2002.
The Parole Board’s decision was unsuccessfully challenged by way ofjudicial review before Turner J (26 April 2002) and then the Court ofAppeal (October 2002). The majority concluded that Article 6 (the right toa fair trial) had no application because of,
Padfield & Maruna—The revolving door at the prison gate 343
the critical fact that when a parole licence is revoked and its revocation issubsequently confirmed this is solely with a view to the prevention of riskand the protection of the public and not at all by way of punishment.
(Simon Brown LJ, at para. 23)
Only Hale LJ recognized, in her dissenting judgment, that ‘to the personconcerned it is experienced as punishment, whatever the authorities maysay’ (at para. 52). But even the majority recognized the dangers of theprocess and the need for fairness: Simon Brown LJ stated, for example, thatthe Parole Board ‘should be altogether readier than presently they are tohold oral hearings if in truth their determination is likely to turn upon theresolution of important issues of fact’ (at para. 40).
Finally, on 27 January 2005, in R v. Parole Board, ex parte West UKHL 1, the House of Lords decided unanimously that he should havebeen allowed an oral hearing:
In his representations against revocation the appellant West offered theBoard explanations, which he said he could substantiate, of his failure tokeep an appointment with his probation officer and of the incident at his ex-partner’s hostel. The Board could not properly reject these explanations onthe materials before it without hearing him. He admitted spending one nightaway from his approved address, staying (he said) with a cousin. While thiswas a breach of his licence conditions, it is not clear what risk was therebyposed to the public which called for eight months’ detention. His challengecould not be fairly resolved without an oral hearing and he was not treatedwith that degree of fairness which his challenge required.
(per Lord Bingham, at para. 45)
Our second example is the case of Trevor Smith,23 who was convicted inMay 1998 of rape and of making threats to kill. He was sentenced to eightyears’ imprisonment, reduced on appeal to six-and-a-half years, and he wasmade subject to an extended licence under s. 44 of the 1991 Act. He wasnot granted parole, but was released on licence after serving two-thirds ofhis sentence on 7 November 2001. He tested positive for cocaine in anumber of tests while living in a probation hostel, and on 25 January 2002his supervising probation officer recommended revocation of his licence onthe basis of his drug use. That recommendation was supported by thedistrict manager of the probation service, and the case was referred to theParole Board to decide whether to make a recommendation under section39(1). On 4 February the Parole Board recommended revocation, and on 6February Smith was recalled by the Home Secretary to prison.
Smith made written representations under s. 39(3) on 20 March, and on3 April these were rejected by letter from the Parole Board. His originalapplication for judicial review was turned down by Hooper J. However, atan oral renewal of the application before Silber J, he was granted permis-sion to seek judicial review, but only in relation to Article 6 (right to a fairtrial) and the common law. The case was listed on 24 June 2003 beforeGoldring J who refused to allow him to rely also on Article 5 (right to
Criminology & Criminal Justice 6(3)344
liberty). He appealed against that refusal, and on 30 June 2003 the Courtof Appeal (led by Lord Woolf CJ) ordered that he be permitted to rely onArticle 5 (right to liberty) as well as Article 6 (right to a fair trial) andcommon law. The Court also ordered that the case be heard in the Courtof Appeal. The case was then heard by Kennedy, Brooke LJJ, Holman J on29 July, who held that Mr Smith had no right to an oral hearing. Here toothe House of Lords on 27 January 2005 allowed his appeal:
The resort to class A drugs by the appellant Smith clearly raised seriousquestions, and it may well be that his challenge would have been rejectedwhatever procedure had been followed. But it may also be that the hostels inwhich he was required to live were a very bad environment for a manseeking to avoid addiction. It may be that the Board would have beenassisted by evidence from his psychiatrist. The Board might have concludedthat the community would be better protected by encouraging his self-motivated endeavours to conquer addiction, if satisfied these were genuine,than by returning him to prison for 2 years with the prospect that, at the endof that time, he would be released without the benefit of any supervision.Whatever the outcome, he was in my opinion entitled to put these points atan oral hearing. Procedural fairness called for more than consideration ofhis representations, on paper, as one of some 24 such applications routinelyconsidered by a panel at a morning session.
(per Lord Bingham, at para. 46)
Finally, we have the case of Mr Sim who was convicted of two offencesof indecent assault and an offence of indecency with a chid under 14, thevictim of all 3 offences being the 7-year-old daughter of a friend. He wassentenced on 17 February 2000 to an extended sentence24 of two-and-a-half years and an extension period of five years. He was released on 11January 2001, but his supervising officer requested his recall on 17 July2001 because of two warnings about his use of alcohol and because of anarrest for alleged indecent exposure to teenage girls. His licence wasrevoked and he was recalled on 20 July.
Mr Sim made representations to the Parole Board, who granted him anoral hearing on 15 April 2002. They refused to direct his release, despitethe fact that it was agreed by then that he was no longer suspected of theindecent exposure. He challenged this decision before Elias J,25 who heldthat there was no ground for quashing the Parole Board’s decision.However, he also made two declarations, which the Home Secretaryappealed to the Court of Appeal:
1 The decision to continue to detain a prisoner who has been subject to recallduring an extended licence period is a decision which attracts the safeguardsof Article 5; accordingly, the detention must be consistent with the aims andobjectives of the original sentence and must be subject to regular supervisionby reviews which are compliant with Article 5.4.
2 Section 44A(4) of the Criminal Justice Act 1991 must be construed so thatthe Parole Board is obliged to conclude that it is no longer necessary to
Padfield & Maruna—The revolving door at the prison gate 345
detain the recalled prisoner unless the Board is positively satisfied that theinterests of the public require that he should be confined.
The Court of Appeal26 (unanimously) dismissed the appeal, agreeingwith Elias J on the first issue (‘detention during the extension period mustbe supervised by a judicial body’) and on the second (‘the Board has to bepositively satisfied that continued detention is necessary in the publicinterest if it is to avoid concluding that it is no longer necessary’). However,it also dismissed a cross-appeal: the admission of hearsay evidence does notrender proceedings unfair or in breach of the prisoner’s Conventionrights.
These cases are described at some length here, not only to expose thereasoning of the courts in this developing area of administrative law, but asexamples of the ‘stories’ of those who have been recalled to prison. It is ofcourse only a partial story. Further research might reveal, for example,whether the provision of legal advice to prisoners is declining in qualityand/or quantity. Tables 2 and 3, for instance, indicate that although theoverall workload of the Parole Board is increasing, the number of ‘repre-sentations after recall’ has fallen in recent years. It is beyond the scope ofthis article to explore this further: we seek simply to initiate debate andresearch into this important area.
Conclusions and future directions
There is a growing interest in the United Kingdom and in the EuropeanUnion in sentencing law and practice. Thus, although it has long beenaccepted that judges and magistrates may have a number of purposes whenimposing a sentence, and that these purposes may indeed conflict, Parlia-ment in section 142 of the Criminal Justice Act 2003 legislated the‘Purposes of sentencing’:
(1) Any court dealing with an offender in respect of his offence must haveregard to the following purposes of sentencing—
(a) the punishment of offenders,
(b) the reduction of crime (including its reduction by deterrence),
(c) the reform and rehabilitation of offenders,
(d) the protection of the public, and
(e) the making of reparation by offenders to persons affected by theiroffences.
We would argue that this list serves little practical function. The differentpurposes are not prioritized and judges will continue to choose from themenu of options. More importantly, this focuses yet again on ‘front end’sentencing decisions. Different purposes may be applied by those whodecide to recall prisoners, and it is these later decisions which need moreopenness and transparency.
Criminology & Criminal Justice 6(3)346
In the European Union there is much movement towards the approxima-tion of the rules on criminal law and sentencing.27 Thus, the Commission ofthe European Communities published a Green Paper on ‘the approxima-tion, mutual recognition and enforcement of criminal sanctions in the EU’in 2004 in which the key question was ‘whether national differencesregarding criminal penalties are an obstacle to attaining a high level ofprotection in an area of freedom, security and justice’ (Commission of theEuropean Communities, 2004: 2). We still know far too little about thesenational differences to be able to attempt to answer the questions raised.The Green Paper sets out reasons why the approximation of criminalpenalties may be desirable: to help give the public a shared sense of justice,for example, or to enhance mutual trust. These are valuable ambitions, ofcourse, but as the Paper also acknowledges ‘the formula used to harmonisepenalties has not been so much to determine effective, proportionate anddissuasive penalties as to set minimum penalties’ (2004: 15). Increasingsentence levels are not likely to reduce levels of crime. Indeed, suchincreases bring with them serious fairness, justice, cost and effectivenessconcerns. Finally, for our purposes, there is little recognition that harmon-ization will only make sense if early release and recall rules are also broughtinto line.
Little is known about best practices in prisoner recalls worldwide.Because of the costs and perceived inefficiencies of the recall process, mostinternational observers recommend (and indeed most jurisdictions claim toutilize) a system of graduated, community-based sanctions for probationviolators, culminating in a return to prison in only the most extreme cases(Carter, 2001). The US state of Georgia has had some success with theimplementation of a systematic and consistent policy for graduated sanc-tions, increasing the percentage of successful parole completions from 61per cent in 1998 to 72 per cent in 2002 (Re-Entry Policy Council, 2005:392). Much more research is still needed in order to design these escalatingsanctions in ways that are the most effective at encouraging compliance.
Some have suggested, for instance, that responses to rule breakingshould be therapeutically directed (Gendreau et al., 1994: 173–84). Forinstance, a probationer who is found to have a dirty urine sample might beordered to attend some form of addiction counselling rather than returnedto prison. Another probationer who regularly misses scheduled meetingsmight be referred to a structured day programme or life-skills coursedesigned to encourage responsible time-management. In the most seriouscases, probationers could be remanded to halfway houses or probationhostels where counselling is available. More research is needed on theeffectiveness of all such programmes, and care would be needed to ensurethat the deployment of these additional options actually reduced thenumber of recalls, and did not merely widen the net for the sanctioning ofthose on licence.
Finally, much more research is needed into the decision to recall and inparticular the role of the supervising probation officer. Richard McCleary’s
Padfield & Maruna—The revolving door at the prison gate 347
now almost 30-year-old ethnographic field observations of parole officerdecision making in a Midwestern US city, would be an ideal model forsuch research (McCleary, 1978). While there is a growing literature (e.g.Gendreau et al., 1994: 173–84) on the use of discretion within the criminaljustice system, the sociology of prisoner recall remains a badly under-researched area.
1 See Sentencing Guidelines Council (2004), or Sentencing Advisory Panel(2004) which suggests 13 degrees of seriousness for offences of sexualassault!
2 See their evidence to Home Affairs Committee, House of Commons (2005:236). The Criminal Justice Act 2003 release provisions came into force inApril 2005.
3 Population in Custody, November 2005 (London: Home Office, RDSNOMS, 2005).
4 Prison Statistics 2002 (Home Office, 2003: 3).5 See note (a) to Table 2.6 Early release before the ‘normal’ release date, first introduced in January
1999 (see later).7 Since this can only be used in respect of curfewees whose original offences
were committed on or after 1 January 1999, the figure may well grow infuture years.
8 The reason that relatively few lifers are recalled may well be explained bytheir low reconviction rates: Kershaw et al. (1997) indicate that of thosereleased between 1972 and 1994, less than a tenth (9%) were reconvictedof a standard list offence within 2 years; this group includes 1 per cent whowere reconvicted for a grave offence. A smaller proportion of mandatorylife sentence prisoners (8%) than discretionary life sentence prisoners(11%) were reconvicted within 2 years. The low numbers recalled shouldnot disguise the injustice that many individuals may suffer (Padfield,2002).
9 See note (a) to Table 2.10 Section 40 was repealed by the Powers of Criminal Courts (Sentencing) Act
2000, Sch. 12(I) para. 1.11 Words inserted by Crime and Disorder Act 1998, s. 103(3).12 See the Release of Short-Term Prisoners on Licence (Amendment of Requis-
ite Period) Order 2003 SI 2003 No. 1602.13 For a review of the changing role of the Parole Board, see Padfield
(2006: 4).14 See Padfield (2006).15 See Probation Circular No. 28/2004. The Management of Offenders and
Sentencing Bill 2005, lost with the calling of the general election in April2005, would have added a polygraph condition (see clauses 47–50).
16 Stocker  Crim LR 293; Teasdale  Crim LR 657.17 The authors posed the question of how to explain the dramatic increase in
Criminology & Criminal Justice 6(3)348
the use of prison recalls to a group of Cambridge University graduatestudents, all of whom hold senior positions in either probation or theprison service. This group was able to list over two dozen possible causes,several of which informed our own analysis.
18 Some of the most recent include Morgan and Owers (2001) and SocialExclusion Unit (2001).
19 The Public Accounts Committee, House of Commons, reported that fourout of ten prisoners were homeless on release, and that over 40 per centlose contact with families in the course of a prison sentence (PublicAccounts Committee, 2002: paras 2 and 31).
20 The Home Affairs Committee, House of Commons (2005) reports that66.6 per cent of prisoners have no job on release and only 16 per centreceive any advice or guidance about finding a job.
21 For a parallel literature on the increasing use of conditional bail, seeHucklesby (1994, 2002).
22 R (West) v. Parole Board  1 WLR 705; R v. Parole Board ex parteWest  UKHL 1. See also Padfield, N., ‘Back Door Sentencing’: IsRecall to Prison a Penal Process?  Camb LJ 276.
23 R (Smith) v. Parole Board  1 WLR 421; R v. Parole Board ex parteSmith  UKHL 1.
24 Under what was then s. 58 of the Crime and Disorder Act 1998 and nows. 85 of the PCC(S)A 2000.
25  1 WLR 2548.26  EWCA Civ 1845 (distinguishes R (Giles) v. Parole Board 
UKHL 42;  1 AC 1; cf.  PL 813).27 Under Article 29 (ex Article K.1) of the Treaty of European Union.
Booth, M.B. (1903) After Prison—What? New York: Fleming H. RevellCompany.
Carter, M.M. (2001) Responding to Parole and Probation Violations: AHandbook to Guide Local Policy Development. Washington, DC: USDepartment of Justice, National Institute of Corrections.
Carter, P. (2003) Managing Offenders, Reducing Crime. London: HomeOffice.
Commission of the European Communities (2004) The Approximation,Mutual Recognition and Enforcement of Sanctions in the EU: Green Paper.European Union: Commission of the European Communities.
Dodgson, K. and E. Mortimer (2000) Home Detention Curfew Orders: TheFirst Year of Operation. Research Finding No. 110. London: HomeOffice.
Dodgson, K., P. Goodwin, P. Howard, S. Llewellyn-Thomas, E. Mortimer,N. Russell and M. Weiner (2001) Electronic Monitoring of ReleasedPrisoners. Research Study No. 222. London: Home Office.
Fletcher, H. (2003) ‘Prison Licence, Community Order Breaches and the PrisonPopulation’, Napo News 149: 1.
Padfield & Maruna—The revolving door at the prison gate 349
Gendreau, P., F.T. Cullen and J. Bonta (1994) ‘Intensive Rehabilitation Super-vision: The Next Generation in Community Corrections?’, Federal Proba-tion 58: 173–84.
Glaze, L.E. (2002) Probation and Parole in the United States, 2001. Wash-ington, DC, Bureau of Justice Statistics.
Gray, M.K., M. Fields and S.R. Maxwell (2001) ‘Examining Probation Viola-tions: Who, What and When’, Crime and Delinquency 47: 537–57.
Home Affairs Committee (2005) Rehabilitation of Prisoners: Volume II, Oraland Written Evidence (HC 193-II). London: The Stationery Office.
Home Office (1996) White Paper: Protecting the Public. Cm. 3190. London:Home Office.
Home Office (2003) Prison Statistics 2002. Cm. 5996. London: HomeOffice.
Hucklesby, A. (1994) ‘The Use and Abuse of Bail Conditions’, The HowardJournal of Criminal Justice 33: 258–70.
Hucklesby, A. (2002) ‘Bail in Criminal Cases’, in M. McConville andG. Wilson (eds) The Handbook of the Criminal Justice Process, pp. 115–36.Oxford: Oxford University Press.
Irish, J.F. (1989) Probation and Recidivism: A Study of Probation Adjustmentand Its Relationship to Post-Probation Outcome for Adult Criminal Offen-ders. Mineola, NY: Nassau County Probation Department.
Kershaw, C., P. Dowdeswell and J. Goodman (1997) Life Licensees: Recon-victions and Recalls by the End of 1995. Home Office Statistical BulletinIssue 2/97.
Kleiman, M. (1999) ‘Community Corrections as the Front Line in CrimeControl’, UCLA Law Review 46(6): 1909–25.
Lynch, M. (1998) ‘Waste Managers? The New Penology, Crime Fighting andParole Agent Identity’, Law & Society Review 32: 839–69.
McArthur, V. (1974) Coming Out Cold: Community Reentry from a StateReformatory. Lexington, MA: Lexington Books.
McCleary, R. (1978) Dangerous Men: A Sociology of Parole. Beverly Hills,CA: Sage.
Maruna, S. (2001) Making Good: How Ex-Convicts Reform and RebuildTheir Lives. Washington, DC: American Psychological Association Books.
Morgan, R. and A. Owers (2001) Through the Prison Gate: A Joint ThematicReview by HM Inspectorates of Prisons and Probation. London: HerMajesty’s Inspectorate of Prisons.
Nellis, M. (1999) ‘Towards “the Field of Corrections”: Modernizing theProbation Service in the Late 1990s’, Social Policy and Administration 33:302–23.
Nellis, M. and W.H. Chui (eds) (2003) Moving Probation Forward: Evidence,Arguments and Practice. Harlow: Longman/Pearson Education.
Padfield, N. (1998) A Guide to the Crime and Disorder Act 1998. London:Butterworth.
Padfield, N. (2002) Beyond the Tariff. Cullompton, UK: Willan Publishing.Padfield, N. (2006) ‘The Parole Board in Transition’, Criminal Law Review 3:
Criminology & Criminal Justice 6(3)350
Parole Board (2003) Annual Report 2002–3. HC 1182. London: HomeOffice.
Parole Board (2004) Annual Report 2003–4. HC 1100. London: HomeOffice.
Parole Board (2005) Annual Report 2004–5. HC 518. London: Home Office.Petersilia, J. (1999) ‘Parole and Prisoner Re-Entry in the United States’, in
M. Tonry and J. Petersilia (eds) Prisons, pp. 479–529. Chicago, IL: Uni-versity of Chicago Press.
Petersilia, J. (2003) When Prisoners Come Home: Parole and Prisoner Reentry.Oxford: Oxford University Press.
Petersilia, J. and S. Turner (1993) ‘Intensive Probation and Parole’, in M. Tonry(ed.) Crime and Justice: An Annual Review of Research, vol. 19, pp. 281–335. Chicago, IL: University of Chicago Press.
Piehl, A.M. and S. LoBuglio (2005) ‘Does Supervision Matter?’, in J. Travisand C. Visher (eds) Prisoner Reentry and Public Safety in America, pp. 105–38. New York: Cambridge.
Public Accounts Committee (2002) Reducing Prisoner Re-offending. Fifty-thirdReport of Session 2001–02. London: House of Commons.
Re-Entry Policy Council (2005) Report of the Re-Entry Policy Council. NewYork: Re-Entry Policy Council, Council of State Governments (EasternRegional Conference).
Reitz, K. (2004) ‘Questioning the Conventional Wisdom of Parole ReleaseAuthority’, in M. Tonry (ed.) The Future of Imprisonment in the 21stCentury, pp. 199–235. New York: Oxford University Press.
Sabol, W.J., W.P. Adams, B. Parthasarathy and Y. Yuan (2001) OffendersReturning to Federal Prison, 1986–9. NCJ 182991. Washington, DC: USDepartment of Justice, Bureau of Justice Statistics.
Sentencing Advisory Panel (2004) Consultation on Sentencing under the SexualOffences Act 2003. London: Sentencing Advisory Panel.
Sentencing Guidelines Council (2004) Guideline: Overarching Principles:Seriousness. London: Sentencing Guidelines Council.
Simon, J. (1993) Poor Discipline: Parole and the Social Control of theUnderclass, 1890–1990. Chicago, IL: The University of Chicago Press.
Simon, J. (1996) ‘Criminology and the Recidivist’, in D. Shichor and D.K.Sechrest (eds) Three Strikes and You’re Out: Vengeance as Public Policy,pp. 24–50. Thousand Oaks, CA: Sage.
Social Exclusion Unit (2001) Reducing Re-Offending by Ex-Prisoners. Lon-don: Home Office.
Taylor, Ralph B. (2001) Breaking Away from Broken Windows. New York:Westview Press.
Tonry, M. (ed.) (2003) Confronting Crime: Crime Control Policy Under NewLabour. Cullompton, UK: Willan Publishing.
Travis, J. and S. Lawrence (2002) California’s Parole Experiment. Washington,DC: The Urban Institute Justice Policy Center.
Travis, J. and J. Petersilia (2001) ‘Reentry Reconsidered: A New Look at anOld Question’, Crime & Delinquency 47: 291–313.
Whitehead, J.T. (1991) ‘The Effectiveness of Felony Probation: Results from anEastern State’, Justice Quarterly 8: 525–43.
Padfield & Maruna—The revolving door at the prison gate 351
Wilson, J.Q. and G. Kelling (1982) ‘Broken Windows’, Atlantic Monthly211(March): 29–38.
Crime and Disorder Act 1998.Crime (Sentences) Act 1997.Criminal Justice Act 1967.Criminal Justice Act 1991.Criminal Justice Act 2003.Criminal Justice and Public Order Act 1994.Powers of the Criminal Courts (Sentencing) Act 2000.
R v. Parole Board ex parte Smith  UKHL 1.R v. Parole Board ex parte West  UKHL 1.R (Giles) v. Parole Board  UKHL 42.R (Smith) v. Parole Board  1 WLR 421.R (West) v. Parole Board  1 WLR 705.Stocker  Crim LR 293.Teasdale  Crim LR 657.
NICOLA PADFIELD is a senior lecturer in the Law Faculty, University ofCambridge. A barrister by training, she has written widely on criminal lawand sentencing, and sits as a recorder in the Crown Court.
SHADD MARUNA is a reader in Criminology at Queen’s University Belfast.His book Making Good: How Ex-Convicts Reform and Rebuild Their Lives wasnamed Outstanding Contribution to Criminology by the American Societyof Criminology in 2001.
Criminology & Criminal Justice 6(3)352