Abstract
This article explores an ostensibly strange legitimacy gap between two forms of indefinite preventive detention in Norway and in England and Wales – jurisdictions, respectively, considered progressive and punitive outliers in Europe. Despite their differing penal traditions, both jurisdictions have employed indefinite post-conviction preventive detention, through the Imprisonment for Public Protection (IPP) sentence in England and Wales and forvaring in Norway. A key distinction is that IPPs were legislatively abolished after widespread criticism, while forvaring retains high levels of legitimacy, representing a pocket of indefinite punishment within Norway's penal-welfare state. Our comparison analyses four contributory explanations for this: burden of proof; scale, resourcing and delivery; position in the penal hierarchy; and different penal cultures. We conclude that Norway's emphasis on rehabilitation fosters a more permissive context for change-based sanctions, while the availability of formal life sentences in England and Wales facilitated the abolition of IPPs without a major penal shift.
Keywords
Introduction
Indefinite preventive detention is punishment potentially without end, placing it among the most severe sanctions imposed by the state. This article explores an ostensibly strange legitimacy gap between two such forms of detention in Norway and England and Wales – jurisdictions respectively considered progressive and punitive outliers in Europe. Wacquant (1999) identified England and Wales as a conduit for US-style punitive policies in Europe, while Norway has been framed as a progressive counterpoint to the punitive turn in scholarly literature (Pratt, 2008; Pratt and Eriksson, 2014) and popular imagination. 1
Despite their differing penal traditions, both jurisdictions are – perhaps surprisingly in Norway's case – sites of indefinite post-conviction preventive detention, through the Imprisonment for Public Protection (IPP) sentence in England and Wales and preventive detention (forvaring) in Norway. Both are informal life sentences, whereby the court imposes ‘a sentence that it does not call life imprisonment, but which could actually result in the persons being held in prison until they die there’ (Van Zyl Smit and Appleton, 2019: xi). A key distinction is that IPPs were legislatively abolished after widespread criticism, while forvaring retains high levels of legitimacy, representing a ‘punitive pocket’ (Taxhjelm, 2025) of indefinite punishment within Norway's penal-welfare state. This article explores why indefinite preventive sentences hold greater legitimacy in Norway than in England and Wales, analysing internal systemic factors and broader penal-cultural dynamics.
While previous comparative work between Norway and England and Wales has focused on sites of deep-end confinement (Crewe et al., 2022b), this article focuses on comparing two indefinite sanctions. Crewe et al. (2022b) compared the institutions holding people serving preventive detention in Norway with Close Supervision Centres (CSC) in England and Wales. However, individuals detained in these comparison sites are not necessarily serving similar types of sanctions. In England and Wales, the CSC is intended to protect prison order and the safety of other prisoners and staff, with entry based on behaviour while already serving a sentence (HM Inspectorate of Prisons, 2024). In contrast, our comparison examines Norway's forvaring and England and Wales's IPP – two types of indefinite preventive detention aimed at protecting society from so-called ‘dangerous’ individuals.
The contrast in legitimacy is striking and counterintuitive, given the significant differences in political economy (Cavadino and Dignan, 2005; Lacey, 2008), media landscape (Green, 2012) and the (de)politicisation of justice politics (Lappi-Seppälä, 2008; Shammas, 2020; see also Brangan, 2022; Todd-Kvam et al., 2025). Our comparison aims not only to highlight the specific challenges of indefinite punishment, but also to use a dual heuristic of exceptionalisms to analyse two different trajectories of indefinite preventive sentencing. We seek to understand how the substantive nature of these exceptional sentences interacts with the political, public and cultural dynamics that shape their levels of (de)legitimacy within these two distinct penal contexts. We analyse legal, parliamentary, policy documents and debates, and statistical data on forvaring in Norway and IPP in England and Wales to profile these sanctions in policy and practice. Our comparative analysis focuses on four contributory explanations: (i) burden of proof; (ii) scale, resourcing and delivery; (iii) position in the penal hierarchy; and (iv) different penal cultures. We conclude that Norway's emphasis on rehabilitation and resettlement fosters a more permissive context for change-based sanctions, while the availability of formal life sentences in England and Wales facilitated the de-legitimatisation and abolition of IPPs without a major penal shift.
Conceptualising legitimacy
Stillman (1974) defines legitimacy as the ‘compatibility of the results of governmental output with the value patterns of the relevant systems’, the ‘governmental output’ in question here being indefinite punishment. Stillman identifies multiple levels of legitimacy (individual, group, societal and international), albeit emphasising the societal level. This view of legitimacy comes then from an external comparison of policies and values. Coicaud (2013: 40), meanwhile, sees legitimacy in a more Weberian sense as ‘the recognition that those who are not in a position of power grant to those in commanding positions to have the right to hold and be in power’, framing legitimacy as a feedback loop between rulers and the ruled. This links to Bottoms and Tankebe's (2012) conceptualisation of legitimacy as dialogic, as well as Schoon's (2022) dyadic model whereby legitimacy is a relational concept formed between an object of legitimacy and an audience that confers legitimacy.
Despite their different nuances, these perspectives complement each other: conflicts between output (or object of legitimacy) and values (or audience expectations) often result in diminished recognition of authority. Wallner (2008) makes similar arguments with regard to the legitimacy of specific policies, affirming that policy legitimacy is a product of two relationships: between government and stakeholders and government and the public. We would argue that there is also a relationship between stakeholders and the public, whereby stakeholders can lobby for (de)legitimising particular polices/practices – in part by highlighting how such policies and practices align or clash with values and expectations, but also from emphasising particular aspects of a value set or, over longer periods, altering such values and norms (for example around the death penalty).
Analysing legitimacy also illuminates penal culture, because differences in the compatibility of indefinite punishment with societal values in England and Wales and Norway reflect underlying variations in their respective penal cultures.
Methodology
Nelken (2016) identifies three ideal-typical contexts for comparison: ‘disciplined comparisons’ (advancing inquiry by understanding social variation in the context of developing a given intellectual discipline); ‘foil comparisons’ (intentionally learning lessons from perceived better/worse practices elsewhere to inform policy or advocacy); and ‘standardising comparisons’ (using global indicators to create rankings according to prescriptive standards in order to influence choices and/or exert pressure for reform). Table 1 summarises these different goals and the role of indicators (we classify case studies as a type of indicator).
Three contexts of comparison.
For the purposes of this paper, we conducted a disciplined comparison of two indefinite sanctions or ‘punitive pockets’ (Taxhjelm, 2025) in contexts that have been positioned at opposite ends of the progressive-punitive spectrum. We contend that reading one jurisdiction against the other helps to draw out specific dynamics that otherwise risk remaining too ‘natural’, too familiar, and therefore unexamined. As Nelken (2016: 400) argues, foil comparisons that construct the comparison from one's own situation risk missing the significance of wider structural and cultural conditions. The exceptionalism thesis launched by Pratt (2008), and expanded upon by Pratt and Eriksson (2014), had the laudable policy goal of showing punishment can be delivered more humanely in some penal contexts. As Nelken (2016: 400) observed: ‘The debate about differences in punitiveness is geared not only to advancing the explanatory agenda of the social sciences, but also to finding a way of reducing levels of imprisonment (and perhaps also finding a stick to beat neoliberalism)’. This links to both a criticism of the exceptionalism thesis made by Brangan (2020) – that its Anglocentric starting point risks eliding particular pains of punishment – and to a refinement by Crewe et al. (2022a), who argue exceptionalism is best suited for making relative rather than absolute claims about Nordic penality. 2 We hope that having two starting points or ‘pockets’, and authors based in both jurisdictions, can help to highlight the different dynamics at play in each context, without essentialising either jurisdiction. In concrete terms, this means combining both familiarity with our resident jurisdictions but also an element of ‘marginal gaze’ (Neumann, 2008), whereby not everything appears as natural and taken for granted.
Our comparative analysis draws on legal, parliamentary, and policy documents and debates that set the parameters for practices of indefinite preventive punishment in both jurisdictions. In addition, quantitative data for Norway were obtained from a 20-year national dataset derived from forvaring sentencing records. 3 This unique national register encompasses every sentence imposed during the period 1 January 2002–1 January 2022, the first 20 years of forvaring. The records were obtained from the Norwegian Correctional Service's registration and case management system (Kompis). The dataset was updated at 6-month intervals to ensure all forvaring sentences were recorded and sentence progression was tracked (see Appleton et al., 2025). The quantitative data for England and Wales were drawn from the Ministry of Justice official statistics. 4 We use these data in two ways. First, we provide an overarching comparative account of the history, policy and practice of preventive punishment in both jurisdictions. Second, these data provide a foundation for analysis of internal systemic factors and broader penal-cultural dynamics that help explain the contrasts in legitimacy.
Indefinite preventive sentencing in policy and practice
Norway was one of the first European jurisdictions to legislate for indefinite preventive detention. Its 1902 Penal Code sought to protect society from individuals considered at high risk of committing future serious crimes by imprisoning them, ‘as long as deemed necessary’ (Almindelig borgerlig straffelov, 1903: 45). However, extensive critique led to a lengthy policy process aimed at replacing ‘sikring’ – a preventive safeguarding measure encompassing both detention in prison and preventive supervision in the community. The critique centred around disproportionate sentencing, scepticism towards rehabilitation/treatment effectiveness, harms of indeterminate sentencing, and the prominent role of forensic psychiatrists and risk prediction (Appleton et al., 2025; Lappi-Seppälä, 2016). Unlike England and Wales, Norway abolished formal life sentences in 1981 but retained sikring as a sentencing option (Todd-Kvam et al., 2025).
5
A government White Paper argued that abolition was justified: …in part because the threat of life imprisonment was not real, and would be very inhumane if it was, in part because this form of punishment seems to involve elements of indeterminacy, [and] in part because new offences [in prison] can no longer be met with sanctions of any significance or deterrence. (Justis- og Politidepartementet, 1981: 5)
It was also anticipated that sikring would wither away of its own accord due to steadily decreasing usage (Justis- og Politidepartement, 1978), but this did not prove to be accurate. In 1990, the government proposed a new form of forvaring – indefinite preventive detention – for individuals deemed too dangerous to be released after a fixed-term sentence (NOU, 1990), which was implemented on 1 January 2002. Preventive detention was one of three new indefinite ‘special criminal reactions’ alongside two compulsory psychiatric care orders (Norwegian Penal Code, 2005: Section 12). 6
England and Wales have experimented with different forms of indefinite sentencing for almost as long, albeit in more limited forms. The Prevention of Crime Act 1908 introduced preventive detention for ‘persistent offenders’, lasting 5–10 years with conditional release considered every three years. Critiques over the next three decades included its underuse, unclear purpose, and the lack of difference between the physical conditions of prison and preventive detention. Despite reform attempts, preventive sentences were abolished in 1967, criticised for failing in their aims and misapplication to less serious crimes than intended (Samuels, 1968). From then until 2003, formal life imprisonment was the only indefinite sentence. The IPP sentence was introduced by the Criminal Justice Act (2003), following mounting political concern about so-called ‘dangerous offenders’ whose crimes did not warrant life imprisonment (Annison, 2015; Appleton and Van Zyl Smit, 2016), and was implemented from 2005.
Forvaring and IPP are both ‘pure’ forms of indefinite preventive detention: they have an indefinite point of release and may last for the rest of a person's life. However, they differ in sentence progression. For forvaring sentences, courts set a ‘minimum’ and ‘maximum’ timeframe. The minimum period is usually no more than 10 years but can be set at 14 or 20 years. After serving the minimum period, individuals on forvaring may apply for release on parole. The maximum period typically does not exceed 15 years but can extend to 21 or 30 years for particularly serious crimes (Norwegian Penal Code, 2005: Section 43). Shortly before the end of the maximum period, the prosecuting authority can request extensions of up to five years at a time, but without a timely request, the individual is released at the term's end. In contrast, IPP sentences only specify a minimum term. After this, the individual becomes eligible for parole every two years, but there is no maximum limit on how long they may be detained (Criminal Justice Act, 2003).
The sentences also diverge in post-release structures. Individuals serving forvaring may be conditionally released on parole and potentially be recalled for any violations. This period typically lasts up to five years, after which the sentence ends unless extended by the court (Storvik, 2021). Conversely, IPP individuals can be released on an indefinite ‘life licence’ and may be recalled for breaches. Probation supervision can be suspended after five years, but recall remains possible for license violations. Until recently, the Parole Board reviewed IPP sentences 10 years post-release to decide whether to leave the sentence in place or end it. Following reform campaigns, that period has been reduced to three years from February 2025, with additional provision (if unsuccessful) for an IPP to be automatically removed after a further two years if no recalls occur (Beard et al., 2024; Victims and Prisoners Act, 2024).
The criteria for imposing forvaring and IPP differ. Forvaring can be imposed for certain serious offences if the risk of a further serious offence is imminent. It may also apply to less serious crimes if (i) a serious crime was previously committed or attempted, (ii) there is a close connection between the two, and (iii) the risk of reoffending is deemed particularly imminent (Norwegian Penal Code, 2005: Section 40). By contrast, IPP had no requirement of imminence. It targeted individuals who, when sentenced for one of 153 specified offences, were deemed ‘dangerous’ by the judge – i.e. posed a significant risk of causing serious harm through committing further specified offences (Criminal Justice Act 2003: Section 225). The judge was told that they ‘must’ assume dangerousness unless ‘unreasonable’ to do so, and if the individual had previously committed one of the specified offences, then the judge was compelled to impose an IPP. Judicial discretion, however, was increased by reforms introduced through the Criminal Justice and Immigration Act 2008 (Annison, 2015; Appleton and Van Zyl Smit, 2016).
Between 2002 and 2022, approximately 300 people received forvaring sentences, accounting for 0.16% of all custodial sentences during that period (Statistics Norway, 2024b). Between 2005, when IPP was first introduced, and its abolition in 2012, 8711 IPP sentences were imposed — 0.81% of all custodial sentences during that time. (Ministry of Justice, 2016). As of September 2023, 151 people were serving forvaring, comprising 5% of the prison population (Statistics Norway, 2024a). On 30 September 2024, 2694 people remained imprisoned serving IPP, despite its abolition – around 3% of the prison population (Ministry of Justice, 2024). The number of individuals serving IPP grew rapidly until its abolition, followed by a gradual decline until around 2019, when numbers stabilised. This plateau resulted from a “recall merry-go-round” with yearly recalls approximately matching or exceeding yearly releases (House of Commons Justice Committee, 2022). Meanwhile, the forvaring prison population has grown steadily since its inception, particularly since 2017 (Appleton et al., 2025) – see Figure 1.

Forvaring and IPP population over time.
Both sanctions are underpinned by a philosophy of public protection, relying on risk assessment/reduction to fulfil their aims (see for example, Annison, 2015; Johnsen, 2006). Within a risk framework (where progression towards freedom is contingent on reduced risk), around 50% of people on forvaring have been released (Appleton et al., 2025), and around 62% of IPP sentences have resulted in at least one period of release (House of Commons Justice Committee, 2022). However, there are around 640 IPP recalls annually, with 60% of the remaining individuals serving IPP having been recalled (Ministry of Justice, 2024). Of the 40% never released, 99% have exceeded their minimum term, two-thirds by over 10 years and 147 by 15 years or more (Ministry of Justice, 2024). In the forvaring population, just over half of those currently serving a forvaring sentence in prison have yet to reach their minimum timeframe, around 15% are between their minimum and maximum timeframes, and nearly one-third are beyond their maximum timeframe. The maximum length of extension past maximum term in the period 2002–2020 was 17 years (the mean extension length was 5.6 years) – such extensions cover both time in prison and under probation supervision (Appleton et al., 2025).
A final – and arguably the most crucial – difference between the two sentences is their legal standing. IPP was abolished in 2012, deemed by the UK government as ‘not defensible’ (Annison, 2015: 222), but also practically unsustainable due to rising numbers sent to prison with no release date (Beard et al., 2024). 7 Pressure also came from a European Court of Human Rights ruling deeming the sanction ‘arbitrary and therefore unlawful’. 8 In contrast, forvaring continues, with around 20 sentences handed down per year (Statistics Norway, 2021).
The IPP sentence lasted just seven years before it was abolished. As early as 2010, ministers described it as ‘indefensible’ (Hansard, 2010: col. 730), having ‘almost lost its credibility’ (Annison, 2015: 161). The Prison Reform Trust, in collaboration with academics (Jacobson and Hough, 2010), were encouraged by political actors to ‘continue producing the background noise of evidence’ to help achieve reform (Annison, 2015: 161). Then-Prime Minister David Cameron publicly called IPP ‘unclear, inconsistent and uncertain’ (Cameron, 2011). These critiques of IPP's legitimacy indicate that it had offended certain natural principles of justice, as well as being opaque to public understanding. Other politicians criticised IPP for its 'totally unfair' approach to time served (Hansard, 2014: col. 678). On a more pragmatic level, the Prison Governors’ Association pointed to the 'resentment and frustration' that IPP caused, through perceived unfairness and delays from the bureaucratic strain of delivering the sentence as intended (Howard League for Penal Reform, 2013: 4). This caused increased disciplinary issues, security threats, and negative impacts on health and wellbeing of prisoners. In remarkably short order, IPP lost legitimacy on multiple fronts. It was antithetical to liberal retributivist principles, yet failed to deliver as a crime control measure, and poor resourcing also led to high levels of bureaucratic illegitimacy.
IPP was not abolished retrospectively, creating unique ongoing legitimacy problems. Campaigning has continued beyond abolition, with critique only sharpening as the remaining IPP prisoners serve ever longer beyond their minimum term. Successive governments have defended their lack of action based on pragmatism rather than any pretence of legitimacy. As well as a struggle to meet the scale of need, practitioners have observed feelings of injustice, anxiety and hopelessness impacting people's willingness and ability to engage with services (House of Commons Justice Committee, 2022). Many face becoming ‘over-programmed’ from repeatedly being asked to complete updated programmes, which, along with parole rejections, fosters deep mistrust, hopelessness and withdrawal from activities. Sentence progression setbacks are a suicide risk factor among those serving IPP (Prison and Probation Ombudsman, 2023). People serving IPP have also described feelings of injustice and despair upon seeing people convicted of the same crime post-abolition progress through the system on a determinate sentence, while they remain incarcerated (Edgar et al., 2020; UNGRIPP, 2022).
Following an inquiry by the Justice Committee in England and Wales, the Prison Service has committed to more ‘bespoke’ interventions for people serving IPP (HM Prison and Probation Service, 2024), but it remains unclear if/how these will address IPP's distinct legitimacy deficits. Campaigners and academics have argued that the ‘toxic’ mental health effects resemble those found among wrongfully convicted prisoners, and that injustice needs to be taken more seriously as a driver of mental health deterioration in people serving IPP (Mooney et al., 2021).
The government argued that ‘mass release’ would create public protection issues (House of Commons Justice Committee, 2023), though some MPs have questioned whether the IPP sentence had itself created the risks that some individuals may now pose (Hansard, 2023: col. 440WH). In recent years, campaigning efforts by family members have gained increasing prominence, with extended media coverage and mentions of individual constituents by politicians. Families’ accounts of their own suffering as well as their loved ones, have further framed IPP as a manifestly unjust sentence. People serving IPP themselves have also made their voices heard, with hundreds submitting testimonies to a 2022 inquiry, and increasing coverage in both traditional and social media. Their contributions reinforce critiques of IPP as unjust, as they cannot be readily dismissed as merely resentful or unwilling to accept punishment. A recurring narrative – ‘I absolutely deserved to go to prison – but not like this, or for this long’ – highlights the perceived moral illegitimacy of the sanction's indefinite nature (see UNGRIPP's archive of letters from people serving IPP sentences 9 ).
Psychiatrists and psychologists, as well as probation officers – who all play a role in risk management – have collectively supported IPP reforms, criticising the sentence as psychologically damaging, and questioning its effectiveness (British Psychological Society and Probation Institute, 2023). They also highlight how the psychological impact of indeterminacy can undermine the very conditions needed for risk reduction (Stevenson and Ellis, forthcoming). This additional criticism of IPP's effectiveness means the sentence's remaining justification as a public protection measure appears increasingly fragile. In sum, while forvaring and IPP share similarities, such as their indefinite nature and the requirement to demonstrate change for release, they differ significantly in legal status and legitimacy. The section below explores four key factors that may explain why this ‘pocket of punitiveness’ (Taxhjelm, 2025) persists in Norway with high levels of legitimacy, whilst IPP has been subject to a significant and relatively successful campaign of delegitimisation.
Explaining the legitimacy gap
Ashworth and Zedner (2014: 170) proposed a set of constraining principles for indefinite preventive detention as a means to ‘mitigate possible abuses of power on grounds of public protection’, and thus increase its legitimacy (perhaps particularly in the stakeholder-government axis as identified by Wallner, 2008). These principles include requiring the state to prove continuing risk, limiting additional detention to the shortest period necessary, regular reviews, and adequately resourced rehabilitation. These foundational requirements provide a useful framework for understanding the differing levels of legitimacy between forvaring and IPP, which can be analysed through four factors: (i) an inverted burden of proof; (ii) scale, resourcing and delivery; (iii) position in the penal hierarchy; and (iv) differences in penal culture.
Inverted burden of proof
Ashworth and Zedner (2014: 169) emphasise the need for ‘strong caution’ in assessing dangerousness, arguing the state ‘should bear the burden of proving that the person presents a significant risk of harm’, with the risk threshold proportional to severity of predicted harm. For indefinite preventive sentences, regular reviews with fair trial rights, including proper procedures and legal assistance, are essential. From the outset, both forvaring and IPP incorporated oversight mechanisms aimed at ensuring continued detention remained justified and proportional. However, a key difference lies in the placement of the burden of proof, impacting the perceived legitimacy and effectiveness of each sanction.
In Norway, forvaring prolongations require regular judicial review, where courts reassess the individual's risk to society (Johnsen and Storvik, 2020; Appleton et al., 2025). Once a person serving forvaring reaches the end of their maximum timeframe, the burden falls upon the prosecution authority to prove the individual poses an imminent risk of committing further serious offences to justify extended detention. Requests for extensions, up to five years each, must be submitted at least three months before the timeframe expires (Norwegian Penal Code, 2005: Section 43). Without such a request, the individual is automatically released. This process requires the state to substantiate the need for further detention, thereby acting as an independent mechanism that regularly reassesses an individual's risk. The review process safeguards against arbitrary or disproportionate use of indefinite detention by requiring evidence-based assessments (Storvik, 2021).
By contrast, under the IPP regime, this inversion of the burden of proof does not occur. Instead, the default position is that a person remains detained indefinitely, subject to a parole hearing every two years where: The Parole Board must not give a direction [for release] unless the Parole Board is satisfied that it is no longer necessary for the protection of the public that the prisoner should be confined [in prison] (The Parole Board for England and Wales, 2013: 3).
The same test applies to life-sentenced individuals, but there is an important principled distinction here: formal life sentences are justified by the seriousness of the offence. IPP sentences were often given for less serious offences (on average, they were ‘worth’ a tariff of three years), based mainly on predicted future risk. While a separate non-penal system of detention could have been used, policymakers preferred mimicking the life sentence, viewing the possibility of a distinct system as ‘not British’, reminiscent of wartime internment (Annison, 2015: 57).
An amendment tabled during passage of the Victims and Prisoners Act proposed a presumption of release for severely post-tariff IPP individuals unless the state could prove continued risk. The government opposed it, claiming the parole process applies an impartial test rather than a formal burden of proof (Hansard, 2024b: col. 1965), and stated explicitly that proportionality was irrelevant: The Government's present view is that proportionality should not be a factor for the Parole Board. It is a very difficult ask of the Parole Board to weigh things up; we should give it one task and one task only: to decide the question of public protection. We should have that test, and that should be the right test for all IPP offenders, however long they have served and whether they are over tariff or not. (Hansard, 2024c: col. 990)
However, the process has been criticised for its rigidity and inaccessibility, placing a substantial burden on IPP individuals to prove reduced risk and readiness for release, thus undermining fundamental principles of proportionality (Ashworth and Zedner, 2014). Many have reported significant obstacles in meeting release criteria, including limited access to rehabilitative and mental health care due to overcrowding and resource shortages (Edgar et al., 2020; UNGRIPP, 2022). This was exacerbated by what has been described as ‘a systemic failure’ to provide adequate rehabilitation (HM Chief Inspector of Prisons and HM Chief Inspector of Probation, 2008: 3) and, latterly, the demoralising impact of loss of hope on engagement (House of Commons Justice Committee, 2022).
As mentioned, the Victims and Prisoners Act introduced a significant change, requiring the Parole Board to presume in favour of removing an IPP licence unless retention is deemed necessary (Victims and Prisoners Act, 2024: Section 66). This change has been interpreted as a ‘presumption for release’, and a step forward in inverting the burden of proof. However, it is not yet clear whether these changes, and/or whether the test for release is burdened one way or another, actually affect release or licence termination rates.
Whilst the IPP has been acknowledged as a state failure, that has not automatically translated to more positive outcomes at parole hearings. From a risk perspective, moral accountability struggles to find a role. Refusal to engage may count as an indicator of increased risk, but state failure is not an indicator of reduced risk: the salient question is whether a risk factor has been addressed, not why that has or has not occurred. With a starting presumption of non-release, the absence of room for considerations about moral fairness (whether time spent imprisoned relative to the original offence, or systemic failures to help individuals address the issues they have been told they must) is experienced as highly illegitimate by IPP prisoners and their supporters (House of Commons Justice Committee, 2022).
In contrast, the starting point for forvaring is based on the default presumption of release, unless the state can identify compelling evidenced reasons for continued detention. By making release the default outcome, the forvaring framework (government output in Stillman's model) aligns closer to Ashworth and Zedner's principles of proportionality and human rights (values), enhancing legitimacy. Although much relies on how this is operationalised, it holds the potential, at least, to be perceived as a more legitimate approach relative to shared conceptions of fairness.
Scale, resourcing and delivery
In terms of scale, when comparing jurisdictions that considerably differ, absolute numbers do not tell us much. The number of people serving IPP sentences in England and Wales is, unsurprisingly, much higher than the number of people serving a forvaring sentence in Norway (2734 versus 140 in 2024). 10 These numbers represent approximately 3% and 5% of the respective prison populations. When taken on a per capita basis though, England and Wales subjects proportionally more of its polity to the IPP than Norway does to preventive detention (4.5 per 100,000 population in E&W versus 2.5 per 100,000 in Norway in 2024). 11 This is particularly significant given the IPP sentence has been abolished for 12 years, yet still proportionally more people are serving this sentence even today. Scale considerations become particularly acute when considering the numbers serving beyond their tariff/timeframe – in England and Wales 1083 are currently beyond tariff on an abolished sentence (Ministry of Justice, 2024). Annison (2018) has described the political challenge of handling an abolished, yet persisting, sentence as a Gordian knot. We see this as a key factor in undermining the legitimacy of IPPs, because it both breaches with considerations of fairness and is evidence the sentence is not ‘working’ in terms of reducing risk. Ashworth and Zedner (2014: 169) affirm that, ‘adequately resourced risk-reductive rehabilitative treatment and training courses should be made available’ from the beginning of the sanction (not just at the end) to provide the opportunity to work towards release. In the case of IPP, the effects of early lack of provision persist in the number of people now post-tariff, and their consequent mistrust and hopelessness.
There is a broader point here about the greater policy emphasis and resourcing put towards rehabilitation and resettlement in Norway (Todd-Kvam, 2022; Todd-Kvam and Ugelvik, 2019), which may act to enhance the legitimacy of a sentence based on the logic of risk reduction. Both sentences are built on the premise that a reduction in risk is necessary to achieve release. Such change-work is contingent upon the availability of programmes, mental health and addiction treatment, education and training, and constructive and meaningful activity. These are aspects of the prison experience prioritised by policymakers and practitioners in Norway, including specifically for those serving forvaring, where the regime was intended to be implemented and resourced differently with greater focus on rehabilitation, progression and a clear pathway to release (Forskrift om særreaksjonen forvaring, 2004). The relatively significant ambitions for rehabilitation and resettlement – and access to health and welfare services – create a more conducive enabling context for a change-contingent indefinite sanction.
Ashworth and Zedner (2014: 169) also affirm that 'any preventive detention going beyond the proportionate sentence should be served in non-punitive conditions with restraints no greater than those required by the imperatives of security'. A further point of difference regarding resourcing and delivery is the existence of two specific forvaring institutions for men: (i) Ila Prison and Detention Centre in Bærum near Oslo and (ii) Nermarka prison unit in Trondheim (there is no standalone unit for women, who make up about 5% of the forvaring population). These units are therefore themselves separate, concrete pockets of indefinite punishment within the carceral geography of the Norwegian penal-welfare state. In England and Wales, there were initially no separate regimes and no additional resources for those serving IPPs. There has recently been an uptick in IPP units within prisons, though they are still not widespread, nor nationally mandated. The dedicated forvaring units, therefore, remain an important point of difference that may contribute to forvaring's greater legitimacy. There is, though, a concerning development here.
The number of persons serving forvaring rose sharply in the period 2015–2022, and there are now more sentenced individuals than places in designated forvaring institutions, leaving people waiting for rehabilitative resources in other high-security prisons not necessarily suited to their needs. This rise indicates that the rehabilitative policy ambitions have proved difficult to realise in practice (Appleton et al., 2025; Sivilombudet, 2024). In addition, ebbing political interest in rehabilitation and resettlement work – and steadily reduced resourcing – means such work is becoming increasingly challenging to deliver (Anderson and Gröning, 2017; Todd-Kvam, 2020, 2022). These developments have not, though, made a noteworthy impact on the legitimacy of forvaring to date. Similar concerns have been raised regarding people serving IPPs. The government revealed that approximately one-third of individuals serving IPP sentences were incarcerated in prisons that did not provide the rehabilitation programmes specified in their sentence plans, effectively preventing them from making progress towards release (Hansard, 2024a: col. 241WH).
There are, then, crucial differences in how the state is able to deliver services, both rehabilitative and more broadly, to people serving indefinite sentences in England and Wales and in Norway. These differences may create a more permissive context for indefinite, change-based sanctions because it seems more plausible that the Norwegian penal-welfare state will be able to effect the necessary change to enable release. However, any sanction where freedom is predicated upon ‘the system’ working carries deep implications and responsibilities that may not be possible to deliver upon, particularly in times of budget cuts and service delivery under pressure.
Position in penal hierarchy
The position of forvaring within Norway's penal hierarchy, in comparison to the IPP sentence, offers a further perspective on perceived legitimacy. Norway abolished formal life sentences in 1981, based on the rationale that nobody had ever served a lifetime in prison. Forvaring has since been established as, ‘the law's severest sanction’, 12 and is generally reserved for serious crimes that invoke the strongest public safety concerns. It has been applied in some of the most high-profile and serious cases, including Anders Behring Breivik following his conviction for the far-right terror attacks on 22 July 2011 (see Johnsen and Storvik, 2020; Sørensen and Johnsen, 2021), reinforcing its position as Norway's apex punishment.
In contrast, since the abolition of the death penalty in 1965, England and Wales have continued to use formal life sentences as the system's ultimate penalty. Whole life orders (genuinely life-long sentences) are possible, although the vast majority of individuals serving formal life sentences are given a ‘tariff’ or minimum term – the retributive portion of the sentence that must be served in prison – after which they become eligible to be considered for release on (life) licence (see Appleton, 2010; Van Zyl Smit and Appleton, 2019). The reliance on formal life sentences for the most serious crimes means IPP sentences have occupied a lower position within the penal hierarchy. The existence of formal life sentences, with their permanence and severity, may then have played an enabling role in the eventual de-legitimisation of IPP sentences, a type of informal life sentence. 13 In practice, IPP sentences became a problematic tier in the hierarchy, disproportionately affecting those who were not among the most serious cases, but who often faced prolonged imprisonment with little hope of release. This dynamic contributed to calls for reform and IPP becoming viewed as an excessive and extreme punishment, more akin to internment than humane punishment (Annison, 2015; Appleton and Smit, 2016).
The relatively higher levels of legitimacy of forvaring in Norway may thus, in part, be explained by its status as an apex punishment, acting as it does as a ‘pocket’ in which to place the most high-profile and ‘dangerous’ people for the protection of society. 14 Its position at the ‘top-end’ of the punishment system arguably grants forvaring greater legitimacy as it symbolises a statement of society's boundaries regarding punishment and public safety (cf. Reiter and Blair, 2018). As Norway's ultimate penalty, forvaring appears easier to justify, signalling society's stance on balancing punishment with protection and emphasising its role as an exceptional, rather than routine, measure of justice.
Different penal cultures
Turning to broader penal-cultural dynamics, comparisons incorporating Norway and England and Wales have highlighted differences going back over 30 years (e.g. Crewe et al., 2022a; Green, 2007; Kommer, 1994; Kuhn, 1993). There is a higher level of trust in the state and its ability to deliver public services in Norway (48% of Norwegian respondents have high/relatively high trust in government compared to 27% in the UK (OECD, 2024: 23)). This trust extends to the penal sphere. A belief in state intervention (from many perspectives a good thing) can, as Smith and Ugelvik (2017) have observed, create acceptance for relatively intrusive measures or ‘pockets of punitiveness’ (Taxhjelm, 2025), such as forvaring. This also extends to an acceptance of the role of experts, with courts relying heavily on risk assessments from forensic psychiatrists when deciding whether to impose forvaring (Appleton et al., 2025; see also Johansen and Laursen, 2023; Løvlie, 2006). A belief in state expertise and intervention underpins the premise of a change-based indefinite sentence. This belief is exemplified by the use of expert committees in guiding policy, including a recent report that evaluated forvaring as well as the compulsory psychiatric care orders (NOU, 2025).
Linked to this issue of trust in institutions, there is less febrile political debate and discussion about issues of crime and punishment in Norway compared to England and Wales. Whilst there are media demands that ‘something must be done’ about a given concern regarding crime and punishment, the policymaking process has been more insulated from such impulses in Norway (Shammas, 2016; Shammas, 2020; Todd-Kvam et al., 2025). Whilst it would be going too far to claim Norway is immune to moral panics or populist rhetoric (Todd-Kvam, 2018), politicians are less beholden to the media, and crime is less employed as a wedge issue in political contestation than in England and Wales (Green, 2007). This relative lack of political salience is exemplified by such topics barely featuring in Norwegian party manifestos or election debates. Even when criticisms are made by independent bodies like the Parliamentary Ombudsman (Sivilombudsmannen, 2017, 2019), these tend not to be used to launch sustained political attacks.
In overarching terms, a more regressive penal culture, a more punitive approach and a more dysfunctional prison system in England and Wales has long been observed and critiqued. The punitive instincts of the dominant right-wing print media often drive the political and policy agenda, with the IPP sentence a key example (Annison, 2015). However, the sustained campaign of de-legitimisation, where activists have built alliances with academics and politicians, is a demonstration of a more contested and diverse penal culture than might first appear. The change in the media narrative over the course of this campaign has been remarkable. Rather than being framed as a ‘dangerous offender’, as a terrifying ‘other’ in our midst (Brown and Pratt, 2000), there has been a notable re-framing of such individuals as victims of a failing system (Annison and Condry, 2022). This is an example of how legitimacy can be affected by engagement between stakeholders and the public as well as stakeholders and government.
In 2024, the UN Special Rapporteur on Torture publicly stated: ‘IPP sentences are inhuman treatment and, in many cases, amount to psychological torture’ (OHCHR, 2024). Lord Blunkett, who was Home Secretary at the time of their introduction, went so far as to call them ‘the biggest regret’ of his political career, acknowledging they have, in some cases, resulted in injustice (Bowden, 2024). These statements are an interesting example of authority figures (one international, one national) invoking an incompatibility between a government output and values (c.f. Stillman, 1974), undermining the legitimacy of IPP sentences. This contrasts with IPP's introduction, when several high-profile cases drove the rise of the idea of risk and an acceptance of the need for – and viability of – indefinite punishment (Annison, 2015). 15
Though legitimacy levels remain higher in Norway, there are significant questions regarding the extent to which policy ambitions regarding individualised treatment can be fulfilled and concerns about those currently sentenced but waiting to be transferred to a forvaring institution (see Appleton et al., 2025). Overall, assumptions about state benevolence and efficacy in a context where concerns about societal protection increase provide an enabling aspect to Norwegian penal culture in which indefinite preventive detention can expand.
Conclusion
This article compares two forms of indefinite punishment: Norway's forvaring and the IPP system in England and Wales, jurisdictions considered ‘exceptional’ in rather different ways. We identified four relevant dynamics that clarify the contrasting legitimacy of similar measures in different contexts, and what this reveals about each nation. In Norway, a severe punishment appears more legitimate within a less punitive system, while in England and Wales, a comparable sanction is widely seen as illegitimate. This contrast can be explained by: (1) Norway's greater balance between public protection and rights/fairness in forvaring's structure and procedures; (2) greater early investment in rehabilitation; (3) forvaring's clear position at the top of the penal hierarchy; and (4) deeper mistrust of the state in England and Wales, alongside coordinated critique from campaigners, politicians, practitioners, academics and individuals serving the sentence.
The general emphasis on rehabilitation and resettlement may create a more permissive context for change-based sanctions, strengthening the Norwegian penal-welfare state's change-making credentials. Meanwhile, the continued availability of ‘top end’ formal life sentences in England and Wales might have made the de-legitimatisation and legislative abolition of IPP easier to achieve. A ‘top-end’ abolition might be interpreted as a more significant shift in the penal Overton window 16 than abolishing a slightly lower tier penalty in the penal hierarchy. Finally, whilst there is a more punitive penal culture in England and Wales, stoked by a dominant right-wing print media, there is also a strong campaigning sector, independent local and national journalism, benefactors willing to fund campaigns, and the BBC remain, as public service broadcaster, an important reasoned interlocutor in tricky policy issues. These elements have been important in the process of de-legitimisation.
In reflecting on the lessons that may be drawn, we can return to Nelken's terminology, albeit moving slightly away from a disciplined comparison, instead asking how might our chosen nations, the nature of their ‘pockets of punitiveness’ and attendant politics, be utilised as a ‘foil’ for the other? At the level of substance, there is clearly force to the argument that England and Wales is a more punitive nation with more problematic penal policy, if understood on accepted metrics (most obviously imprisonment rates) and by reference to its imposition of long and indefinite sentences. More specifically as regards the IPP sentence, it is widely accepted to have been a disaster in practice, even on its own terms as a putatively risk-based sentence: this was a potentially life-long sentence that could be imposed on ‘dangerous offenders’ identified only on the basis of judicial ‘common sense’ rather than expert risk assessment (Jacobson and Hough, 2010). It was a sentence whose only route to release – and its legitimising justification – was by these ‘dangerous offenders’ engaging in a range of programmes, education and support, but which was afforded no additional resources. Indeed, many practitioners did not even understand what the IPP sentence was and could not explain it to those sentenced (Jacobson and Hough, 2010).
From this perspective, the Norwegian forvaring sentence sharpens our recognition of the importance of a lengthy policy development process, the tight legislative drafting of such ‘pockets of punitiveness’, their sufficient resourcing in a manner that takes seriously the intentions underlying a risk-based sentence, and the establishing of systems and processes that ensure the appropriate balancing of public safety with individual rights.
As regards the penal cultures of the respective nations, there is certainly much to bewail about the political and cultural dynamics that led to the IPP sentence (Annison, 2015). Is there, however, also a more positive lesson to draw out regarding the political and cultural dynamics that have accompanied the IPP sentence as it has become increasingly contested, and even outright discredited? As much as England and Wales can plausibly be characterised as a punitive, fearful society with a fierce retributive bent (e.g. Garland, 2001), public opinion appears more nuanced. Debates around the IPP sentence have provided evidence for the outsized centring of public protection in political discourse, and the ease with which people who offend are ‘othered’, treated as ‘weeds in the garden’ to be removed rather than fellow citizens (Annison, 2014). However, debates have increasingly been successful in framing the IPP sentence in quite different ways. Those serving IPP are no longer ‘dangerous offenders’, but ‘damaged victims’; and those affected are not faceless ‘others’ or their stigmatised family members, but mothers or partners enduring unbearable pains (Annison and Condry, 2022; Mackenzie et al., 2023).
The increasing force of these arguments has been striking, and they have begun to substantively benefit those sentenced to IPP. When understood as a ‘foil comparison’ for Norway, does the ongoing contestation around the IPP sentence in England and Wales – powered by family campaigners, charities (and their committed donors), concerned reporters, parliamentarians, academics and others – exemplify Christie's (1973) call 50 years ago for the pursuit of a ‘quarrelling society’? Rather than leading us towards compliance and pacification, Drake and Walters (2015: 415) argued that, ‘Christie's quarrelling society serves to remind us that effective and productive policy often emerges from the contestations and struggles between individuals, governments, interest groups and communities when debating responses to complex social problems’. This, in terms of our legitimacy model, is an argument for strengthening the ‘feedback loop’ of consent between governed and governing to provide a more democratic dialogue regarding the compatibility between politics and policy on one hand and values on the other.
In terms of further work regarding the legitimacy of indefinite punishment, a broader international comparison including those abolitionist nations that prohibit all types of indefinite preventive detention in their criminal justice systems 17 would, we believe, be beneficial.
Footnotes
Acknowledgments
The authors are grateful to Hilde Dahl and Dirk van Zyl Smit for reading and providing useful comments on the draft manuscript. In addition, the editors of the special issue helped strengthen the piece considerably through their thoughtful engagement with our ideas.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: JTK and CA's contribution to the article was written as part of The Implementation and Impact of the ULTimate PENalty in Norway (ULTPEN) project, funded by the Norwegian Research Council (Grant ID 315221) and based at the Department of Mental Health, Norwegian University of Science and Technology in Trondheim.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
