Normalisation by default and normalisation by design: A systematic review of empirical studies on the normalisation of prison life

Normalisation is an important principle when shaping life in prison. The conceptualisation and operationalisation of normalisation are not straightforward, which merits a systematic review of the literature. This article reports the results from such a review, drawing on 16 empirical studies from a range of countries. Four relevant domains of normalisation in prison were identified: design, facilities and services, atmosphere, and identity. Additionally, the origins and dominant norms underpinning normalisation were found to differ depending on the local context: ‘normalisation by default’ resulted from resource scarcity and the importation of pre-existent roles, hierarchies and goods, while ‘normalisation by design’ was an intentional effort to shape prison conditions in line with conventional norms. These different manifestations were associated with different limitations and possible benefits. The article also discusses how normalisation is related to resocialisation and responsibilisation, and offers underexplored areas that deserve attention in future research.


Introduction
Normalisation is an important concept in determining life in prison.Normalisation is generally defined in international soft law as the ambition to shape life in prison according to life outside prison and is urged in both the United Nations Standard Minimum Rules and European Prison Rules as a guiding principle for prison policy (Council of Europe, 2006;United Nations, 2015).The principle of normalisation has strongly influenced Nordic prison policy and practice (Engbo, 2017), and is often linked to the 'penal exceptionalist' prison culture characterised by low imprisonment rates and humane prison conditions (Pratt, 2008;Smith and Ugelvik, 2017).For example, the prison island Bastøy in Norway and architecturally acclaimed Storstrøm in Denmark do not resemble 'traditional' prisons and are therefore considered more normalised environments (cf.Jewkes, 2022).Bastøy, referred to as a 'village prison' has no walls; prisoners are free to roam the island and spend their days working 'normal' jobs such as working in a bike repair shop and operating the ferry to the mainland.Storstrøm stands out because it is designed to resemble a college campus with views overlooking the large courtyards.
Although these examples may suggest that shaping normalised prison conditions is straightforward, research shows that the implications of the principle of normalisation for policy and practice are not unequivocal, and are oftentimes conflicting in nature.First, on the one hand, normalisation supposes prisoners retain all their basic human rights, which is reflected in the obligation for prison authorities to provide sufficient (health) care, work and allow for prisoners to maintain contact with their families (Engbo, 2017).On the other hand, normalisation requires consideration of individual differences and respective needs (Snacken, 2002).In this context, Snacken (2002) distinguishes between collective normalisation and individual normalisation.Collective normalisation supposes prisoners should have access to the same welfare services as citizens in a free society.Individual normalisation on the other hand alludes to the various social roles one fulfils in society such as father, son and partner, and urges that prisoners can continue to fulfil those roles while imprisoned.Second, studies have attributed special pains of imprisonment to open and less restrictive prisons, which are typically characterised as representing a more normalised prison environment (De Vos and Gilbert, 2017;Shammas, 2014).Furthermore, researchers have argued that prisons possess fundamentally oppressive and damaging characteristics, irrespective of their appearance, security level or culture (see: Crewe et al., 2023;Jewkes, 2022).Third, well-meant normalisation efforts can become 'abnormal normalisation' in practice (Smith and Ugelvik, 2017: 519).For example, the self-catering system in Danish prisons does not necessarily allow prisoners to eat and prepare the kind of food that resembles their eating habits in a free society (Minke and Smoyer, 2017).Indeed, practices that are considered normal may diverge from practices preferred by prisoners (Smith and Ugelvik, 2017), and a person's normal in a free society might not be their desired normal.It thus remains largely unclear what elements in prison should be normalised in order to achieve normalisation of prison life.
Another ambiguity arises when normalisation is tied to other related principles, mainly resocialisation.Arguably, normalised prison conditions ease the transition back into society because the contrast with a free society is smaller and the incarceration experience is less harmful (Crewe et al., 2023;De Vos and Gilbert, 2017;Mjåland et al., 2021).This link between normalisation and reintegration, however, deserves further scrutiny.In much of the literature, the concepts of normalisation, resocialisation and responsibilisation are not clearly demarcated.It is assumed that normalised prison conditions create circumstances that allow prisoners to either retain or (re)gain skills which are needed when rebuilding their life after release (Council of Europe, 2006;De Vos, 2021;United Nations, 2015).By linking normalisation to reintegration, behavioural norms become intertwined with an intrinsic right of prisoners to humane prison conditions.A focus on resocialisation might sway the normalisation effort towards behaviour instead of prison conditions, resulting in efforts to normalise the individual as a form of responsibilisation (Van de Rijt et al., 2023;Van Ginneken, 2018).Here, normalisation transgresses into a form of discipline conform with Foucault's definition of the concept of normalisation (Foucault, 1977).From the above line of argumentation, it becomes clear that it is not possible to make a straightforward distinction between the rights-based principle of normalisation and disciplinary normalisation in the Foucauldian meaning of the word.We therefore treat these two forms of normalisation not as static opposites, but rather as a 'sliding scale' where both forms can appear as a result of the same effort (De Vos, 2021;Van de Rijt et al., 2023).
In other words, the normalisation of prison conditions is a delicate endeavour.What is considered normal depends on local living standards, cultural practices and individual preferences.An examination of the conceptualisation, operationalisation and application of normalisation in the prison literature will generate further insight into commonalities and differences regarding this important concept.This contribution will synthesise the current empirical literature on the normalisation of prison life, to answer the question: In what way is the concept of normalisation operationalised in the empirical literature, what are the notable findings, and how is normalisation demarcated from related concepts such as resocialisation and reponsabilisation?A systematic review of the literature reveals which studies have tackled normalisation in prison explicitly, how they have done this, and what we can learn from the findings.In doing so, the aim is not to reach an all-encompassing definition of the concept normalisation, but instead to expose the full range of aspects that may constitute the normalisation of prison life.We conclude the article with a discussion offering a critical reflection on the current state of the art and its limitations.

Systematic review Databases and search strategy
To gain a thorough understanding of the concept of normalisation in the prison context, only empirical literature on normalisation specific to the context of the general prison regime within adult prisons was reviewed.The initial search included articles published from 1945 to March 2022 in the English or Dutch language.The scholarly literature was searched using the databases Web of Science, PsycInfo and Criminal Justice Abstracts.Google Scholar was consulted to extract undetected grey literature.The following search terms were used: Normali* AND (imprison* OR prison* OR jail OR detentie OR gevang* OR 'correctional institution*' OR 'correctional facilit*' OR penitentiar* OR incarc*).Studies on closed prisons, irrespective of their security level, and (half) open prisons were included, albeit with their respective (explicit) differences and context taken into consideration.Empirical studies conducted in the setting of juvenile detention, forensic psychiatry, immigrant detention or penal measures executed outside an institution were not included in the study.Studies were also excluded when normalisation (noun) or to normalise (verb) was aimed at a psychological and/ or physical disorder, disability or disease such as HIV (i.e., the opposite of discrimination) or related to the statistical analysis in research (not related to the prison environment).

Search results
Figure 1 illustrates the search process.The initial search returned 440 studies, published in the English or Dutch language.After removing duplicates 341 studies remained.Screening of title and abstract resulted in 51 remaining publications, which were reviewed in full.In total, 14 articles met the criteria for inclusion.Two additional relevant studies were found through Google Scholar and scanning of the reference lists of the included articles.The final selection consisted of 16 articles.All included articles were published in peer-reviewed journals, apart from one PhD thesis (De Vos, 2021) and an article in a Dutch-language journal for a mixed scholarly and professional audience (Farahi and Van de Rijt, 2017).
The selected articles varied in terms of country of research, research method, setting and the way normalisation is studied.Notably, all included studies were conducted after the year 2000.The majority of literature stems from countries from the Global North: United States (4) Denmark (3), Belgium (2), Finland (1), Spain (1), The Netherlands (1) and Norway (1).Studies from the Global South were conducted in the Philippines (3) and Hong Kong (1).With regard to the research method, the vast majority of the articles used qualitative methods (14), mainly (semistructured) interviews and observations; one article used quantitative research methods and one used mixed methods.Looking at the setting of the research, 10 of the articles were conducted in male prisons, two in prison (units) for women and four studies included both female and male prisons or prisoners.Normalisation was not often studied as the main objective, which was the case in only six of the 16 studies; instead, it was usually explored as an outcome of research.Operationalisations of normalisation were therefore often implicitly described in the text.An overview of these results is given in Table 1.
Lastly, it should be noted that due to the variations in method, study sample and prominence of normalisation, the operationalisations synthesised in this study reflect the views of divergent (groups of) individuals such as policymakers, prison management, frontline staff, and incarcerated individuals.Nevertheless, it is important to emphasise that the voices of incarcerated individualsmen and womenwere well represented in these studies.

Analysis
The selected articles were systematically analysed and openly coded without any fixed constructs or domains of prison life in mind.Codes were written in the margins of the texts.The elements were then inductively abstracted until coherent domains arose (see Table 1).In order to verify the four identified domains, the operationalisations were then deductively organised in a table alongside the corresponding references (see Table 2).

Findings
To begin with, a shared understanding across the articles was that normalisation is the extent to which life in prison resembles life outside prison.A detailed analysis of the articles revealed that empirical operationalisations of the concept of normalisation were related to different aspects of prison life.More specifically, four domains were identified, namely 'design', 'facilities and services', 'atmosphere', and 'identity'.We describe the identified categories in more detail in the paragraphs below, starting with the general prison environment and ending with the normalisation on the individual level.Another important finding is that normalisation has a considerably different translation given the socioeconomic and cultural context.Although this is in itself not a new notion with regard to the concept of normalisation, what this study reveals is that this does not only result in differences in prison practice but also in differences in how and why normalisation of life in prison is achieved: by design or by default.Normalisation by design is the effort to intentionally create normalised circumstances.Normalisation by default, on the other hand, constitutes normalised circumstances that arise out of necessity, often originating from a shortage of resources.The systematic literature has uncovered that empirical research explicitly dealing with the normalisation of imprisonment is scarce.Furthermore, critical reflections on the conceptualisation and operationalisation of the term are mostly missing.As a result, our findings on the operationalisation of normalisation are a starting point for debate and further research, rather than an exhaustive account.

Domains of normalisation
Design (n = 5).The analysis of the articles showed the importance of design in the normalisation of prison conditions.Normalised prisons are typically small-scaled institutions (Reiter et al., 2018;Villman, 2023) with modular living units where categories of prisoners (men/women, adult/juvenile, remand/convicted) are housed separately (Cid, 2005).Small-scaled prisons create an environment more closely associated with home (Farahi and Van de Rijt, 2017), allow for more familiarity amongst staff and prisoners resulting in more normalised ways of interaction and offer possibilities for encouraging the autonomy of prisoners (De Vos, 2021).Generally, low-  (1997,1998,1999,2000,2001,2002) Narag et al., 2017;Sloan, 2012;Trinidad, 2020 (continued) security open prisons are said to offer possibilities for more normalised circumstances, for example by allowing prisoners a higher degree of freedom of movement (Villman, 2023) and freedom to have contact with other prisoners (Reiter et al., 2018).The security level does not only relate to architectural design but also to the degree of (direct or indirect) supervision from prison staff (Reiter et al., 2018).Open prisons depend more on dynamic security, where interaction between staff and prisoners partly serves as a form of (informal) control, in the absence of the more static forms of security and control that dominate in high(er) security prisons (De Vos, 2021).
Conversely, as will be discussed later, prisons with a higher security level in some cases offer opportunities for more normalised circumstances, for example in the contact between staff and prisoners (Reiter et al., 2018).In these prisons, the building is designed to regulate security, allowing staff to engage in more informal interaction.The degree to which prisoners enjoy freedom of movement also transcends the prison perimeter, for example opportunities for prisoners to have jobs, study or go on leave outside prison (Reiter et al., 2018).Therefore, the prison environment should not only resemble life in the community, but should also physically be placed within the community and prisoners should be housed in their own region (Cid, 2005).Although allowing prisoners to go outside the prison walls does not normalise life in prison, it normalises the life one has in prison and connects life inside the prison to life in society.Normalisation also pertains to the accommodation and living spaces of prisoners.Certain design and furnishing choices can enhance normalisation, such as furniture made of natural instead of 'vandal-resistant' materials that are not bolted to the floor, noise-reduction measures, and access to television (Tartaro, 2006).Vandal-resistant materials represent an institutional environment whereas natural materials reflect the furniture people own at home.Also, subtler aspects of the living spaces, such as the level of control prisoners have over switching the lights in their cell and the degree of privacy, are considered part of a normalised environment (Tartaro, 2006).
Facilities and services (n = 8).Normalisation is operationalised as prisoners having access to the same facilities and services that are available to citizens in a free society, including healthcare, work in prison and contact with family and friends (Farahi and Van de Rijt, 2017).Health care is not limited to physical care but also concerns healthcare for mental illness and treatment for addiction (Cid, 2005).Work in prison should be sufficiently available and generate enough earnings to 'satisfy economic needs' (Cid, 2005: 153).Regarding contact with the outside world, it is important that prisoners are provided with enough possibilities to maintain contact with family and friends through email, phone calls and visits (Farahi and Van de Rijt, 2017).The possibility to have private visits and overnight visits from family enhance normalisation because they contribute to the maintenance of family life while imprisoned (Minke, 2014;Villman, 2023).Furthermore, allowing prisoners to prepare their own meals, which reflects the daily activities of life in a free society, contributes to a normalised prison environment (Farahi and Van de Rijt, 2017;Minke, 2014).These facilities and services are related to normalisation in the context of retaining all human rights while imprisoned (Villman, 2023), and also are considered to contribute favourably to the rehabilitation of prisoners (Blakely and Bumphus, 2004).Mass media (television, books, newspapers, etc.) and digital media can also enhance normalisation (Reiter et al., 2018;Vandebosch, 2005;Villman, 2023).It is noted, however, that access to digital devices (e.g., mobile phones and internet) is largely lacking and does not reflect the conditions of a free society (Reiter et al., 2018;Villman, 2023).Two things seem noteworthy when it comes to the normalisation of facilities and services in prison.First, the absence of restrictions on facilities and services is an important stipulation for the normalisation of the prison environment (Reiter et al., 2018;Villman, 2023).This concerns restrictions on the types of facilities and services and the extent to which prisoners have access to them.Second, normalisation can also be enhanced by allowing prisoners to bring their own private goods into prison, including appliances (e.g., electric fan, TV, refrigerator) and personal possessions such as food, clothes, and money (Narag et al., 2017).
Atmosphere (n = 5).Normalisation does not solely concern physical and material aspects of prison life, as is the case in the two domains described above, but also pertains to more intangible aspects of the prison environment.One of these aspects is the atmosphere in prison, which is related to (social) contacts in prison, such as contact with peers, staff or even animals.Researchers have found that the presence of dogs promotes normalisation by creating a more homely and cosy atmosphere and a 'safe space' for prisoners within the prison walls (Minke and Smoyer, 2017;Smith and Smith, 2019).For prisoners who own or owned dogs at home, the dogs in prison were a reminder of a positive aspect of their life at home.This form of normalisation is subjective and related to (prior) experiences and feelings.In prisons in the Philippines, normalisation was related to a communal atmosphere created through the formation of gangs (brotherhood or 'pangkat') reflecting community life in a free society (Narag and Jones, 2017;Narag et al., 2017).The informal social control and the prisoner classification scheme that have formed as a consequence of the formation of these gangs imitate 'Filipino cultural practices of communitarian living and community spirit' (Narag et al., 2017: 355).As will be discussed later, the manifestation of normalisation of prison life discussed in these studies originated from necessity, as a result of a lack of resources, shortage of staff and inmate overcrowding, rather than intention.
Identity (n = 7) On an individual level, normalisation can pertain to a person's identity by respecting their dignity and autonomy.Farahi and Van de Rijt (2017) operationalise normalisation according to the definition of individual normalisation provided by Snacken (2002), where the variety of social roles the prisoner has in a free society, such as father/mother, brother/sister are recognised and respected.One way to achieve this is by allowing prisoners to wear their own clothes instead of a standard prison uniform (Minke, 2014).In the Philippines, a prisoner's identity is transferred from the community into the prison community by deploying the occupational skills prisoners bring into prison for the benefit of the prison community (Narag and Jones, 2017;Narag et al., 2017).Furthermore, their pre-prison identity is not just transferred into prison; prisoners are in some cases even able to continue part of their (occupational) life outside prison.The authors argue that prison authorities 'inspire inmates to keep their pre-prison identities' and mention successful but anecdotal cases where a prisoner built a political career from prison or recorded an album (Narag and Jones, 2017: 15).Normalisation of identity can also mean allowing prisoners to reshape certain aspects of their identity, which might benefit their life after prison.For example, the participation of prisoners in religious activities in the Philippines reportedly helps prisoners shape a new identity and find a sense of purpose (Trinidad, 2020).Importantly, such instances of normalisation are voluntary, but facilitated by the prison service.
The normalisation of identity can also arise in an attempt to distinguish oneself from the (abnormal) prison environment and the associated prisoner identity.Although prison is said to inherently limit autonomy (Reiter et al., 2018), this limitation can ignite distinct practices to (re-)claim control over one's life while in prison.Sloan (2012) describes this complex interaction between normalisation and identity by arguing that prisoners use cleanliness to assert control over their lives and their identity.Cleaning fulfils this function through manipulation of the self and personal space (i.e., tidying one's cell), which is used to set oneself apart from the perceived 'dirty' prison environment and 'inferior' prisoner identity (Sloan, 2012).Normalisation in this sense should not be understood in reference to norms outside prison, where cleanliness and cleaning may not be a defining part of a person's (male) identity or a typical masculine habit.In prison, however, cleaning practices may help maintain a (masculine) identity, where 'available mechanisms through which to perform one's gender are heavily restricted' (Sloan, 2012: 409).In other words, normalisation of identity means fulfilling a non-prisoner identity within the restricted institutional context.In contrast, Gray (2006) argues that normalisation is utilised by prisons in Hong Kong to shape the behaviour of imprisoned women along societal gender norms through the provision of specific education, vocational training, and activities (e.g., knitting and household chores).From a review of the studies, it thus appears that opportunities for identity normalisation are partly dictated by the prison, but also partly shaped (and even circumvented) by prisoners.
The influence of place: Normalisation as a universal concept?This systematic review exposed differences in the operationalisation of normalisation of prison life in the Global North and the Global South.There are important contextual differences between the Global North and Global South in the origins of normalisation in prison (intended vs. not intended), as well as the substantive manifestation (e.g., related to privacy, material standards, etc.).These differences are the result of the context of the prison setting, in this case, the country and the habits and customs of the population.
One of the areas where countries from the Global South seem to hold differing views from countries in the Global North is the level of privacy when 'implementing' normalisation in the prison environment.In the Global North, the use of single cells is promoted as an element to promote normalised circumstances in prison (Cid, 2005;Tartaro, 2006).In the Global South such as the Philippines, more importance is given to communion and communitarian living, as this reflects life in a free society (Narag and Jones, 2017;Narag et al., 2017;Sloan, 2012;Trinidad, 2020).The particularity of the context by which normalisation is defined is shown in one of the articles describing the attempted importation and application of Western inmate classification schemes in the Philippines (Narag et al., 2017).Western-inspired inmate classification schemes were implemented for purposes of housing assignment and protocols for supervision and control in prison.However, cultural and organisational factors such as a lack of trust from staff in the new classification scheme and insufficiently qualified staff resulted in the failure to implement the classification scheme.Norms may thus not 'travel' very well, and the place is important to the meaning of normalisation.
In many countries in the Global North efforts to normalise the prison environment are created intentionally due to prevailing penal philosophies about a humane prison climate and the reintegration of prisoners (Cid, 2005;Reiter et al., 2018;Vandebosch, 2005;Villman, 2023).Conversely, in the Philippines (and possibly other countries in the Global South, we return to this in the Discussion), normalisation of prison life is generated by default, which may be due in part to extremely limited funds and resources (Narag and Jones, 2017;Narag et al., 2017).As a consequence, prisoners share responsibility for administrative tasks and maintaining safety and order, and they can import their own goods (electrical appliances but also food, clothes and money) into prison.Prison culture is shaped by the importation of hierarchies and gangs.These elements promote the normalisation of prison life in the sense that they reflect life in the Philippine community.Here, normalisation is not emulated but rather emerges by default and, as a result, is accompanied by inequality and corruption.At the same time, the subsequent shared governance system in the Philippines also creates beneficial circumstances for both prisoners and staff.Staff is helped by the system of informal control that the gangs (Pankat) provide.Prisoners, in turn, feel empowered and treated as humans because they are respected for their (pre-prison) skills and knowledge (Narag and Jones, 2017: 15-16).
Inherent limitations of normalisation.In the previous paragraph, it was pointed out how resource limitations may shape opportunities for normalisation, but the analysis also revealed other ways in which the institutional environment shapes and oftentimes limits normalisation.Security concerns, in particular, can interfere with normalisation efforts.In some cases, normalisation can be explicitly regarded as undesirable in prisons.Alcohol use, for example, is legal in free society but is prohibited in prison (the same is true for the possession of other contraband items, such as mobile phones and medication that is not prescribed).Staff can also impose restrictions on participation in activities or freedom of movement in response to other behaviour not considered unlawful in a free society, such as verbal aggression or disobedience (Reiter et al., 2018).Paradoxically, more restrictive environments can sometimes offer greater opportunities for informal interactions between staff and prisoners.Limitations in movement may be offset (partially) by more informal and in-person contact with staff, as is discussed by Reiter et al. (2018) using the example of a teacher's experience in a highsecurity facility who had more contact with the incarcerated student.This suggests that normalisation may be realised in different domains of the prison experience, and that limits to normalisation in one domain (e.g., freedom of movement) may be compensated to some extent by greater normalisation in another domain (e.g., participation in activities, and respectful and friendly interactions with custodial and non-custodial staff).
Other limits to normalisation are tied to the deprivation of autonomy and liberty imposed by imprisonment.In the context of Danish prisons it is noted that, although many elements of life in prison have to a great extent been normalised such as the ability to cook one's own meal and respectful staff-prisoner relationships, prisoners still experience limited autonomy (Reiter et al., 2018).The restrictions that are listed mainly concern digital (communication) devices and restrictions to freedoms such as eating fast food and deciding your own daily schedule.With developments in technology that have significantly impacted life in the community in the last decades, life in prison has further diverged from life in the community.Arguably, prisoners experience frustrations because they are denied access to mobile phones, Internet and a variety of TV channels because of institutional rules and regulations that are not necessarily perceived as legitimate (Reither et al., 2018: 104).
Van de Rijt et al.

Normalisation and related concepts of resocialisation and responsibilisation
A utilitarian justification of normalisation is that it enhances prospects of successful reintegration after release, because the differences between life in prison and life in the community are minimal and, as a result, people would be expected to encounter fewer adjustment problems on release.Theoretically, the concepts of normalisation and resocialisation may be different; empirically, however, they are more difficult to distinguish.Each concept has a different ideological and temporal emphasis; while normalisation is associated primarily with humane prison conditions and is therefore present-oriented (i.e., time in prison), resocialisation refers to life after prison and is thus future-oriented.However, the practical implementation of normalised conditions also requires decisions on what is 'normal' (and often imparts some judgement on what is desirable or conventional).This is illustrated by the study by Cid (2005), who explicitly considers rehabilitation opportunities separately from normalisation efforts.The operationalisation of both concepts, however, shows considerable overlap, such as the availability of treatment programs for drug addiction and opportunities for temporary leave.The extent to which normalisation of the prison environment contributes to resocialisation is only researched in one article.Villman (2023) found that personal circumstances prior to imprisonment and attitudes towards life outside (or after) prison are related to the experience of normalisation in prison.That is, prisoners with a relatively conventional life before prison saw the normalised circumstances as 'unproblematic'.For example, prisoners appreciated access to mobile phones as it allowed them to stay in contact with family while in prison and valued the ability to go grocery shopping (Villman, 2023: 88).However, prisoners with a less stable life before prison and who aspired to a conventional life after prison considered the normalised circumstances insufficient to prepare them for life after release.Finally, some prisoners rejected the imposed conditions associated with the normalised prison environment, and described them as 'hypocritical' or 'patronizing' (Villman, 2023: 92-94), because they allowed some activities but prohibited others (such as the use of alcohol during prison leave).
The last example shows how the normalisation of prison life can develop into forms of (experienced) responsibilisation, even more so when (participation in) 'normalised' practices or activities (e.g., cooking or working) are obligatory or conditional on good behaviour.While the literature suggests that normalised circumstances encourage good behaviour of prisoners (Narag and Jones, 2017;Narag et al., 2017;Tartaro, 2006;Trinidad, 2020;Villman, 2023), it is problematic from a human rights perspective if normalised prison circumstances are conditional on good behaviour, and effectively become a reward that can be earned (e.g., more time out of the cell, more visits or a place in the 'honour block').In practice, however, these lines are often blurred, also due to security concerns (discussed previously), which are often tied to risk assessment and prior behaviour.This means that a normalised environment can simultaneously have a normalising function if it is a privilege to be gained or lost depending on whether behaviour is considered desirable and conventional.

Discussion
Beyond the definition offered in international soft law, normalisation has not been clearly and consistently operationalised for empirical study.In our systematic literature search, we were therefore unable to depart from a clearly demarcated operationalisation of the concept of normalisation and limit our analysis to outcomes of normalisation.The literature search revealed the wide-ranging scope of normalisation, including many different aspects of prison life.The lack of a clear operationalisation precluded a deductive analysis of normalisation, but our inductive analysis is intended to yield a helpful starting point for further research.Through a discussion of dimensions of operationalisation in conjunction with substantive findings, we aimed to contribute to a greater awareness of the concept of normalisation of life in prison and provide more depth to the discussion on the subject.Moreover, we aimed to provide more clarity on the relationship between normalisation, resocialisation and responsibilisation.In most research thus far, these concepts are considered separately and separatable (cf.De Vos, 2021), but in practice, their application and meaning overlap, resulting in a blurry line between the rights-based principle of normalisation and the normalisation of behaviour (i.e., disciplinary normalisation). 1  This review has exposed variations of normalisation between the Global North and the Global South, most notably how the normalisation of prison life can also emerge from weak institutional governance, as was described in the context of Philippine prisons.We refer to this as normalisation by default, as opposed to normalisation by design.Similar practices have been found in prison systems across South America, for example in Brazil and Uruguay (Boersma, 2016;Darke, 2022;Darke and Karam, 2016) and in prisons in African countries such as Ethiopia (O'Donnell, 2023).Birkbeck's (2011) comparative analysis of penal institutions in North and South America offers some further insight into the origins of the normalisation of prison conditions in the Global South.Normalisation by default is a product of what he describes as 'internment', or 'the absence of projects for doing something with or to inmates while in confinement' (Birbeck, 2011: 321).He does not find the availability of resources a satisfactory explanation for the difference between the Global North and Global South in terms of prison life and disciplinary control; rather, he argues that it should be understood in light of a different point of focus on penal social control.Venezuela, for example, offers many more gradual release opportunities from an earlier stage in a prison sentence, compared to the United States.Penal control and intervention are thus focussed on the release process rather than on behaviour inside the prison walls.It is important to note that a strict division between the Global North and Global South is too simplistic, as practices of communal prison life are also present in collective cultures in Eastern parts of Europe (Idrissi, 2020;Piacentini and Slade, 2015) and even among prison (sub-)cultures in Western European counties, for example the IRA in Ireland (O'Donnell, 2023). 2 In these contexts, it has been similarly observed that prison authorities intervene very little in the lives of prisoners, resulting in greater autonomy and extra-legal or shared governance.
The distinction between 'normalisation by default' and 'normalisation by design' thus reflects societal norms, cultural practices and the locus of penal power, but each has different benefits and pitfalls.A crucial difference between the two is the level of control exercised by the authorities.Greater official governance and staff presence can work in favour of prisoners lower in the prisoner hierarchy or with fewer financial means, because they may be better protected from exploitation and undue inequality in access to resources. 3Additionally, it may offer more opportunities for reinvention (Crewe and Ievins, 2020), rather than reproduction of a community lifestyle that was likely associated with one's offending behaviour.Shared governance, on the other hand, can also contribute to the empowerment and autonomy of prisoners in positive ways, including the utilisation of skills and knowledge, and the development of leadership skills (Narag et al., 2017).Less formal control also often extends to family visits, which have the character of 'open days' (Birbeck, 2011: 314) due to the mingling of prisoners and community members, and help to break down the isolating experience of imprisonment.It is worthwhile to continue the debate on whether the benefits of both manifestations of normalisation can be implemented in systems seeking to normalise without the disadvantages.It is important to keep in mind, however, that norms and ideas about what is 'beneficial' (and for whom) are context-dependent.
A limitation of this systematic review is that it resulted in a fairly small number of included publications, mainly because only studies were included that use the term normalisation explicitly.This does not mean, however, that normalisation of the prison environment has not been the subject of empirical research.Instead, the concept of normalisation has been studied through the use of related terms or concepts, such as the social or living climate in prison (see, e.g.Liebling, 2004;Van Ginneken et al., 2018), through architectural design (for an overview of studies on 'ethical architecture' see Engstrom and Van Ginneken, 2022) or differences in prisoner experience in open and closed prisons (Crewe et al., 2023;Mjåland et al., 2021;Shammas, 2014).This is also true for the four domains identified in this article; design, facilities and services, atmosphere and identity.We therefore cannot rule out the existence of additional studies on a similar or related topic.
In reflection on the findings, we identified areas that merit further examination.First, the empirical studies that were included in the review were based on a variety of samples.Further research could devote attention specifically to the question 'whose norms are the basis of normalisation efforts?' Second, the articles provided relevant but limited information on what the various domains mean for shaping the prison environment.In addition, they did not account for the different needs of specific groups of prisoners such as the elderly (Dawes, 2009;Ginn, 2012;Snacken, 2002).Furthermore, the articles did not address how prior health issueswhich are generally more severe for the prison population compared to the general populationshould influence standards of normality (Van Zyl Smit and Snacken, 2009).A third influential domain only marginally touched upon is the relationship between staff and prisoners.Staff-prisoner relationships are an important component of the social climate in prison (Boone et al., 2016;Liebling, 2004;Van Ginneken et al., 2018).Finally, digitalisation deserves more attention in relation to normalisation of prison life.If not addressed, the gap between the use of digital technologies in prison and free society will grow wider in time.

Declaration of conflicting interests
The ideas voiced in this article are the authors' and do not reflect the position of the Dutch Custodial Institutions Agency.

Figure 1 .
Figure 1.Flowchart of included articles systematic review.

Table 1 .
Summary of included articles.
SpainQualitative Official reports and data of the years 1996-2003 such as Defensor del Pueblo (Spanish Ombudsman) Infrastructure: are modular, based on the cellular principle, and provide for the full range of services stipulated by legislation (workshops, sports and cultural facilities, sick bays and others)(2) Accommodation: place where sentence is served (close to home or not), separation of prisoners (categories such as men/women, adult/ juvenile, remand/convicted) and sleeping accommodation (single cell)The normalization of prison conditions, mainly the security level (small 'campus-style' facilities).No separate perimeter and no immediate supervision, level of freedom of movement, facilities and services ability to work or study (in or outside the prison), access to mobile phones, the possibility for overnight visits with family and prison leave.
. Informe anual.Madrid: Cortes Generales and reports from the European Monitoring Centre for Drugs and Drugs Addiction (EMCDDA) (2001) and CPT reports.

Table 2 .
Overview of domains and main findings.