Lessons in Flexibility: Introducing a Yoga Program in an Australian Prison

International research provides support for yoga as a wellbeing intervention in prison. Until recently, no systematic research had been undertaken in Australia to assess the effectiveness of a yoga program, or consider the challenges of implementation. In 2017, the authors, in partnership with Australian Capital Territory (ACT) Corrective Services and the Yoga Foundation, introduced a pilot yoga program at the Alexander Maconochie Centre in the ACT. This paper draws on comments from the prisoners who participated in the program and the yoga teacher, as well as the perspectives of a prison psychologist and the lead author, both of whom also participated in the program. The paper reflects on the lessons learnt from developing and delivering a prison yoga program and advocates for the expansion of such programs in Australian prisons.


Introduction
Readers are invited to imagine that it is a hot afternoon in Canberra, Australia in mid-January 2017. Ten male prisoners have gathered in the newly opened gymnasium at the Alexander Maconochie Centre (AMC), which houses all adult prisoners in the Australian Capital Territory (ACT). Many of these prisoners (commonly referred to as detainees in the ACT and generally referred to as participants in this paper) are heavily muscled and covered in tattoos. Despite such external appearances, there is no sense of threat in the gym. Instead, the yoga teacher (ME), firstauthor (AH) and a prison psychologist working at the AMC (LM), who have arrived to deliver and participate in the program respectively, are greeted with easy smiles and the slightly nervous laughter of those risking ridicule. Mats are rolled out. The teacher gives brief instructions. The laughter subsides as participants sit, close their eyes and begin to focus on their breath. These men have come to participate in an eight-week yoga program, 1 a first for the AMC and the first yoga prison program in Australia to be the subject of evaluation. This is the start of a research-driven pilot program designed to bring a less conventional wellbeing intervention to prisoners, to document the process of implementation and evaluate its impact. The pilot was unashamedly a scoping and relationship-building exercise and we acknowledge our bias in favour of such programs from the outset. Each of us was hopeful throughout the conception and delivery of the project that undertaking a yoga program would be effective in improving prisoners' wellbeing. This position is supported by international research that demonstrates the potential of yoga to enhance prisoners' mental and physical wellbeing (see Auty et al. 2017;Bilderbeck et al. 2013;Kerekes et al. 2017). Insofar as the program could be demonstrated to have a positive impact on prisoner wellbeing, we hoped the research would provide a foundation for making yoga available as a regular program at the AMC and in other Australian prisons. It would also provide a foundation for further study in Australian correctional centres.
This article describes the project and reports on the process of introducing a yoga program in an Australian prison; elsewhere, we draw on quantitative and qualitative data to consider the impacts on participants ; see also ). The article begins by surveying the international research on the impact of yoga in prisoner populations, providing justification for an Australian pilot. The paper then considers the evidence with respect to Australian prisoners' physical and mental health, which paints a stark picture of compromised wellbeing. A description of the program, its establishment, integral partnerships and its delivery follows, drawing on the perspective of prisoners, the teacher (ME) and LM and AH who attended and participated in each of the eight sessions run by ME. The purpose of this discussion is to identify key learnings for future programs, as well as strengths and weaknesses in the way the program operated. The article concludes by summarising the various perspectives to provide guidance for future yoga program providers and correctional institutions.

Yoga in prison: The international evidence
To our knowledge, this is the first research on a prison yoga program in Australia. As part of the preparation for this project, we contacted all correctional agencies in Australia to determine if they currently offer yoga programs to prisoners and whether they could provide any evaluation research on such programs. Although some agencies indicated that they offered or had offered such programs in the past (see also Rudd 2017;Wright 2011), none reported any research on these initiatives. This would appear to be consistent with Griera and Clot-Garrell's (2015, 155) reference to the 'camouflaged but increasingly prominent presence of holistic practices in the penitentiary context'. In this section, we summarise the recent international evidence in respect of yoga prison programs, although the methodological limitations of this research (see Auty et al. 2017;Karup 2016;Muirhead and Fortune 2016;Wimberley and Xue 2016) Auty et al. (2017) completed a systematic review and two meta-analyses to examine whether prison yoga and meditation programs are significantly related to increased psychological wellbeing and improvements in behavioural functioning, although most of the programs studied focused on meditation, rather than yoga, and their discussion tended to conflate the two. Auty et al. (2017) found that participants who completed a prison yoga or meditation program experienced small improvements in their psychological wellbeing and behavioural functioning. Programs of longer duration had a slightly larger positive effect on behavioural functioning, compared with more intensive programs. They suggested that yoga and meditation programs 'have the potential to provide a cost-effective supplementary treatment', while 'techniques that improve an individual's ability to cope with difficult emotions could be valuable' (692). They concluded that 'the evidence suggests that yoga and meditation have favourable effects on prisoners' (689). Muirhead and Fortune's (2016) literature review on yoga in prisons considered five programs, two of which were also examined by Auty et al. (2017) (Bilderbeck et al. 2013;Landau and Gross 2008). They found that the studies 'point towards the rehabilitative benefits of yoga and meditative practices' (62).
Wimberly and Xue (2016) examined 10 papers referencing nine studies, including all those included by Auty et al. (2017) and Muirhead and Fortune (2016). They found that the five hatha yoga programs examined demonstrated statistically significant decreases in stress, depression, anxiety and neuroticism, while the three programs that focused on philosophy and spirituality were associated with improvements in aggression, anxiety, impulse control, spirituality and reincarceration.
Two more recent studies adopted a randomised design. Kerekes et al. (2017) reported on the findings of a study of a 10-week yoga program in nine Swedish prisons, in which 77 participants were randomly assigned to a weekly 90-minute yoga class, while 75 were assigned to a control group, whose participants were asked to participate in 90 minutes of exercise each week. The authors found significant improvements on 13 of the 16 measured variables (e.g., less perceived stress, better sleep quality) among the yoga group, compared with improvements on two variables in the control group (perceived stress and psychological wellbeing). The yoga group was also more likely to show improved emotional wellbeing and improvements in computer tests that measure attention and self-control, as well as reductions in antisocial behaviour. The authors accordingly concluded that 'yoga practice can play an important part in the rehabilitation of prison inmates' (2017: 1).
Danielly and Silverthorne (2017) reported on female participants in a 10-week trauma-focused yoga program in two correctional facilities in South Carolina (United States). Participants were randomly allocated to the treatment group (n = 33) or a waitlist (i.e., control group, n = 17), with waitlisted participants joining the subsequent class. Anxiety and self-control scores decreased and increased respectively for the treatment group, although these findings were not statistically significant. In contrast, the control group reported a worsening or no change in these measures. The authors concluded that yoga 'is a relatively inexpensive intervention that could benefit both inmates and prison staff by reducing some negative behaviours and possibly mental health problems' (2017: 9). Karup (2016) presented qualitative findings from 11 semi-structured interviews. She also examined 30 letters written by 10 prisoners between 2008 and 2015 to the Prison Phoenix Trust, a charity whose mission is to introduce yoga and meditation to prisoners across the United Kingdom (UK) and Ireland. 2 Karup (2016: 40) concluded that participating in yoga 'helped prisoners by reducing stress and alleviating the tendency to ruminate in prison … [and] appeared to create opportunities for recovering a sense of self through its positive impact on feelings of control and self-worth' and suggested that yoga may offer a supplementary form of rehabilitation.  Griera's (2017) ethnographic study of yoga delivered in intensive mode in three prisons in Spain also provided a qualitative analysis, including surveys with 54 prisoner program participants and interviews with a subset of these participants, yoga teachers, social educators and other prison personnel (n = 25), supplemented by the researcher's own fieldnotes from participating in the yoga sessions. Although Griera did not seek to determine the impact of the yoga program on participants quantitatively, her findings show some of the meanings yoga can provide to practitioners in a prison setting. A related paper by Griera and Clot-Garrell (2015) involved interviews with prisoners and prison staff and described the yoga class in greater detail. The authors noted that '[t]he particular scenography of the yoga class-with the music, incense, mats laid out around the room and lights partially turned off-also aids inmates in crossing the threshold of the everyday reality of prison' (2015: 147).
Viorst (2017) also undertook qualitative research, conducting interviews with eight former prisoners who had participated in a prison yoga program in San Quentin Prison in the United States for at least 18 months. Viorst (2017:20) found that interviewees identified the nonphysical aspects of yoga, namely, breath control and meditation to be 'especially effective in rehabilitative efforts'. All participants described reduced aggression and 'greater emotional intelligence and awareness ' (2017: 24). The perceived positive effects would appear to be supported by the psychological tests, which indicated that all interviewees had low levels of depression, anxiety and verbal aggressiveness, coupled with a high degree of self-control and selfawareness.

Australian prisoners' physical and mental health
When we consider prisoners as a population and the data on their physical and mental health, we find a disturbing picture of compromised wellbeing, especially relative to the general Australian population. For example, Butler et al. (2006: 273) found that prisoners were far more likely to be diagnosed with a mental disorder; the 'overall prevalence of any psychiatric illness was 80% for prisoners and 31% (weighted) for the community. The contrast between prison and the community was most pronounced for substance use disorder (66% vs. 18%)'.
The most recent picture of the health of Australian prisoners was presented by the Australian Institute of Health and Welfare ([AIHW] 2019). Overall, the AIHW found that: … people in prison have significant and complex health needs, which are often long-term or chronic in nature. They have higher rates of mental health conditions, chronic disease, communicable disease, acquired brain injury, tobacco smoking, high-risk alcohol consumption, recent illicit drug use, and recent injecting drug use, than the general population' (2019, vi). In addition, because nearly all prisoners are ultimately released back into the community, improving prisoners' health and wellbeing 'benefits the entire community (2019: iv).
The data on those entering prison found that: • 29 per cent reported a long-term health condition or disability that limited their daily activities and/or affected their participation in education or employment; • 22 per cent tested positive to hepatitis C; • 75 per cent were current smokers and 69 per cent smoked daily; • 65 per cent had engaged in illicit drug use in the previous year; • 34 per cent reported a high risk of alcohol-related harm in the preceding 12 months; IJCJ&SD 51 Online version via www.crimejusticejournal.com © 2019 8(4) • 40 per cent had been told by a doctor, psychiatrist, psychologist or nurse that they had a mental health disorder; • 26 per cent reported high or very high levels of psychological distress; • 23 per cent were taking medication for a mental health disorder; and • 21 per cent reported self-harm.
The AIHW also found that 33 per cent of prison entrants had only completed schooling up to Year 9; 54 per cent were unemployed in the month leading up to their incarceration; 33 per cent were homeless in the period leading up to their incarceration; 18 per cent had had a parent incarcerated during their childhood; and 38 per cent had dependent children. Of those about to be discharged from prison, only 22 per cent had organised paid employment to commence within two weeks of their release, while 78 per cent were expecting to receive government social security payments. Taken together, this information not only conveys the poor health of Australian prisoners, but the challenges they face in their lives more generally.

The present study
The quantitative and qualitative methods and findings from our project are described in more detail in Bartels et al. (2019). 3 Pre-(n = 8) and post-program (n = 9) quantitative wellbeing data were collected with participants' consent, using a number of well-accepted psychological testing measures. 4 Although the limitations of the small sample size are acknowledged, our findings indicated that participants showed statistically significant improvement, including reduced levels of stress and depression, as well as increased self-esteem and goal-directed behaviour. Qualitative data collected through pre-and post-program audio-recorded interviews revealed positive impacts, including increased flexibility, improved sleep and mental wellbeing. No participants demonstrated negative outcomes.
We intended to undertake follow-up interviews two months after the program ended. Unfortunately, though invitations were extended to all participants, only two program completers attended. We were not in a position to determine the reason for this; however, as one of those who did attend explained, there had just been a death at the AMC and this may have affected participants' desire to speak on the day. We also invited those who had initially expressed interest and completed wellbeing testing to attend to talk to us about why they did not participate in the program. One person attended and we have included his feedback below.
In this paper, we discuss the process of developing and delivering the program, drawing on the observations of the prisoner participants. We have privileged their voices wherever possible to provide a richer sense of their subjective experiences. Both ME, who volunteered his time and expertise, and LM, who enthusiastically supported the program, were asked to provide their feedback on the program and this feedback is also included here. This paper also incorporates the observations of AH, who participated in the program as an embedded researcher.

Establishing the program
The project was conceived on the basis of our personal experience with yoga and meditation, as well as our professional experience with prisoner populations. It was informed by the emerging international literature supporting the efficacy of such interventions. Our aims were to introduce a yoga program at the AMC and investigate the impact of such a program on the wellbeing of prisoner participants, with the introduction of the yoga program itself part of the research initiative. This distinguishes the research from research undertaken to evaluate existing programs. Significantly, the literature review on yoga in prison above describes evaluations of existing yoga programs. Accordingly, our research provides a unique opportunity to explore and report on the process of establishing the program, identifying key learnings to support the introduction of similar programs elsewhere. In this section, we consider the crucial partnerships required for the program.

ACT Corrective Services
Establishing and running a wellbeing intervention in a correctional setting is a challenging, and often time-consuming, enterprise. This is particularly so if the impetus for such a program originates from outside the correctional centre itself. Correctional authorities have to be persuaded that there is merit in such a program and it will not harm the prisoners in their care or present a security risk. Further, although no financial contribution was sought from ACT Corrective Services, managing the logistics of internal programming and providing a staff member to facilitate delivery was a cost in itself.
The process of persuasion commenced with high-level meetings with ACT Corrective Services, AMC management and AMC Prison Psychological Services. AH and LB, who have worked on justice reform and research in conjunction with the ACT Government, drew upon their existing relationships to commence the process. Having initially proposed an intensive meditation program, it was ultimately agreed that a yoga pilot would be trialled instead.
To understand some of the logistical challenges faced in the delivery of the program, it is necessary to have a basic understanding of the correctional centre itself (see Bartels and Boland 2016). The AMC houses all adult prisoners in the ACT, remand and sentenced, women and men of all security classifications. This meant that movement within the centre had to be carefully managed, with prisoners often needing to be escorted from different sections across the facility. As participants in the program came from different locations, this occasionally led to delays as custodial officers faced competing priorities. However, ACT Corrective Services demonstrated their support for the program throughout and apologised for delays in bringing prisoners to be interviewed or attend the program.
The task of coordinating delivery from inside the AMC, as well as identifying and selecting the participant cohort, fell to LM, a clinical psychologist. LM is also a trained fitness instructor who, among other things, had taught mindful stretching. She participated in the classes alongside the prisoners, encouraged them and supported the group to practise in an unavoidable two-week interval discussed below. She was also integral in identifying the importance of holding a graduation ceremony on the final day of the program, at which time those participants who had completed the program were given their own yoga mats.

The Yoga Foundation and yoga community
Although we have practised yoga in our personal lives, we did not have the expertise to deliver the program. Accordingly, we approached the Yoga Foundation (2018), which delivers yoga programs in Sydney to marginalised communities and helped to locate a volunteer yoga teacher based in Canberra. ME had significant experience teaching yoga and, as it emerged, importantly, he also had professional experience working within the controlled environments of onshore asylum-seeker detention centres. Drawing on his contacts, ME sourced yoga mats from a Canberra yoga studio, which generously donated them to the program.

Participant selection and engagement
The project was conceptualised as a small-scale pilot, with the value of the research being both discussing the process of establishing and delivering the program and reporting on its outcomes. In circumstances in which there is an absence of published research on yoga (and meditation) programs in Australian correctional settings, laying a foundation for future program offerings and further research is critical. Given the small scale and research aims, it was not possible to identify a control group. Although we acknowledge that this is a limitation of the present project, it is broadly in keeping with the international experience (cf. Danielly  In total, 18 prisoners expressed interest in the program, with all 18 participating in initial interviews and completing wellbeing measures in December 2016. None of the participants had done yoga before, although some had limited meditation experience, principally through prison programs. Of the initial 18 prospective participants, eight completed the program. A ninth joined in the second week and also completed the program, though he was not one of the 18 who expressed initial interest. Completion was defined as having attended a minimum of five of the eight yoga classes taught by ME. Of the 10 who participated in the initial interview but did not complete the program, seven did not attend any classes. Two attended the first class and did not continue and one attended two classes before being released from prison.

Program delivery
As outlined in the introduction, the program commenced in January 2017. Each class ran on Monday afternoons for approximately one hour. The program ran for 10 consecutive weeks, although it was originally envisaged that it would last eight weeks. This occurred because ME had overseas commitments during the program. The unavoidable delivery gap occurred after two classes had been held, laying the foundation for practice. LM was concerned about the group losing momentum and interest. To address this, she invited the participants to attend at the same time during ME's absence so that they, with her guidance, could practise what had been learned to that point. This had not been planned into the program, but was an example of fluid adaptation and an effective partnership; it was very well received by the participants, who commented on this in their post-program interviews, noting the commitment LM showed to them and the program. The remaining six classes were held after ME's return.
The classes proceeded on the basis of an experiential or 'learning-by-doing' model. As ME explained in his post-program interview, his objective was to teach participants a basic 20-minute routine that they could practise by themselves. This included 'sun salutations', stretches, balancing poses and meditation. An instruction sheet was provided after the second class with basic movements and poses shown in picture form. Breath awareness was explicitly built into the learning of each movement and as part of routine progression from movement to movement. Classes also started and finished with periods of breath awareness meditation without movement. Practice outside formal classes was encouraged, though ME did not check in with participants to determine whether any practice had taken place.
No effort was made by the authors or ACT Corrective Services to define or control the way ME would deliver the program or the actual yoga program that he would deliver. There was no suggestion that he should focus on any particular aspect of yoga practice or theory or explicitly tailor his teaching for the environment. There was also no suggestion of the extent to which meditation or breath awareness would be a specific focus.
The final class concluded with a graduation ceremony. This was suggested by LM, who was cognisant of the benefits of celebrating prisoner success. The senior manager of Corrections Psychological Support Services attended the ceremony and made a short speech congratulating the participants. Following this, each participant was presented with a yoga mat donated by a local yoga studio.

Participants
It was generally acknowledged by participants that the yoga program was considered a unique, unconventional and 'weird' offering. It may be that this deterred some prisoners from indicating interest or participating. P3 stated that other prisoners who were not engaged in the program 'had a bit of a laugh, but stuff 'em'. He explained that this changed once the program commenced, as other prisoners 'wanted to know about it' and a few 'were even upset that they didn't sign up'. P6 expressed the view that while yoga was an unknown in the prison, there was 'no stigma or judgement' attached to participating. Indeed, in his view, there was 'a bit of envy from those who were a bit dirty they didn't keep going'. P5 similarly commented that there was no 'shame about it or nothing' and that, after the program started, 'a lot of them [other prisoners] wanted to do it'. P7 had explained what he was doing to other prisoners, stating: 'they would just laugh. Some people wanted to join up but it was too late'. P1 referred to other prisoners who learned about the program from him when he returned from the class as being 'dirty that they didn't put in for it'.

Cohort selection
In terms of selection of future cohorts of prisoners, P5 took the view that those suffering from depression should be a focus. He stated that the AMC should identify 'people that are really depressed that are going to be here for a long time', stating 'it might just help them ease their mind a bit … it would help bring them up'. P3 referred to the need to ensure cohesive groups were selected for any future program, bearing 'jail politics' in mind. In particular, he noted that participants were 'mains', rather than 'protection'. 5 In his view, the prison should consider offering separate programs for those in 'protection' and 'mains'.
P6 suggested that the future success of any yoga program depended on ensuring that participants selected to participate 'are really serious about changing themselves'. This was echoed by P2, who felt that some 'paperwork' might be appropriate 'to weed out those who don't really want to do it'. He referred to another prisoner who had not participated in the program, saying that he had been deterred by the notion that it would not be a serious program.

Practice outside class
After program completion, participants were asked if they did any practice outside the classes they attended. All but one participant engaged in limited practice outside class, while P4 indicated that he 'just wasn't motivated really'. P5 said that 'now and then I tried to give it a quick go'. He explained that he tried 'the stretching, sitting on your hands, downward dog, upward dog … because I had the sheet'. P3 explained that he used the stretches outside class for lower back pain, but 'it was not something you would really do in the yard'. P6 commented that he would do yoga stretching 'a couple of times a week' outside class after weight training. P2 also stated that he did 'a little bit' and 'tried some of the warrior poses and the leg stretches'. He stated that he intended to continue but indicated that he was restricted in his efforts to practice outside the classes because he did not have his own cell. He said it was 'a bit difficult having a cellie [cellmate] who has no idea and is looking at you like "what the fuck are you doing?" It was sort of awkward'. Similar to P6, he explained that it was not possible to do yoga in the yard because 'everyone looks at you'. P7 and P8 both stated that they used the stretching exercises outside class. P8 also did some breathing exercises, referring to the breath awareness exercises being consistent with another prison program he was involved with, 'Stress Less'. P1 explained that he had done yoga stretching outside class, using a blanket on the floor of his cell. It was clear from the participants' comments that they would have welcomed access to space to use for practice outside formal class. P3 indicated that this would 'definitely' make a difference to him.

Non-participating prisoner
The fact that 18 prisoners expressed interest in the program and completed pre-program wellbeing testing, but only nine completed the program, clearly raises questions about the reasons for attrition. It is possible that some of those who did not attend had in fact been released from custody, as occurred with a prisoner who attended the first two classes. In an effort to understand why some prisoners did not attend, we invited those non-attending prisoners who remained at the AMC to speak with us when we were conducting follow-up interviews with participants two months after the program. Only one non-participating prisoner attended, but his contribution is of value, even if it may not be representative of the nine in the original cohort who ultimately did not participate or the AMC population more generally.
When asked what prevented him from attending the program, NP1 responded: 'I just wasn't in the right frame of mind'. When asked if there was something that we could have done to help him attend, he explained that it was just his 'mindspace'. He went on to say that he was now ready to do the yoga program, because 'the other boys are doing it, it's a community thing'. This insight suggests that if yoga programs were held on a regular basis, prisoners would be drawn to participate as a consequence of observing the effect on those already participating. NP1 noted IJCJ&SD 56 Online version via www.crimejusticejournal.com © 2019 8(4) that participants 'came back all happy and that. They got a lot from it. It took stuff off their mind.
That's what I got told'.
Yoga teacher ME was interviewed after completion of the program to help develop an understanding of how and why the program worked and identify key learnings for future prison yoga programs. We had already conducted the post-program participant interviews at the time we spoke to ME and, though the data had not been analysed, it was clear that participants viewed the program as a success. Therefore, the interview with ME was conducted in light of this.
For ME, the key to success had been the work of LM, whom he described as having great 'enthusiasm' for the program, 'rapport' with the participants and a willingness to organise things from the inside. ME reflected on the fact that he had not realised just how difficult logistics could be within the prison and how such a program depended on the support of correctional staff responsible for the management and movement of prisoners. He noted that it was rare for him to be able to offer a full one-hour class, due to late arrivals. Indeed, he was surprised by how many competing appointments and commitments had to be managed.
When asked about the essential attributes of a yoga teacher in a custodial environment, ME explained: 'it's not a normal yoga class. You really have to be adaptable and flexible'. This required 'having a lot of different ways of communicating and presenting things; always adapting whatever plans you had, [or] just forgetting about it halfway through because you have a sense of how people are going'. In his view, it would help to be an experienced teacher, so that the poses and technical aspects are second nature, thereby enabling the teacher to focus on the 'environment [and] the mood'. Indeed, he suggested that knowledge of the poses only accounted for about 20 per cent of the teaching responsibility, the rest being taken up with reading the participants' engagement and mood. He also acknowledged that for those who have not worked in controlled environments, the experience could be quite intimidating. ME indicated that he had a lesson plan for each occasion, but invariably this would have to be adjusted during the class. Taking breath awareness meditation as an example, he explained that his objective might be to have participants sitting and meditating quietly on their breath for five minutes, but he found that he needed to adapt and build up to this. Moreover, this process of building capacity was non-linear, as the capacity of each participant to concentrate depended on unknown aspects of their prison experience at the time they attended the class. He referred to a particular participant who was very focused one week-like a model student in a communitybased yoga class-and the next week was pacing around and unable to settle.
In terms of starting and settling the group in, ME stated that he initially did not really know 'how things would go'. He noted that joking and banter punctuated the first class or two, but then seemed to disappear and was replaced by more focused attention, although he felt he could not rely on this. Reflecting on the challenge involved in keeping participants focused and motivated in class and preventing them distracting others, ME noted that no protocols or rules had been established at the outset. In one particular class, the problem of distraction became acute and he felt it would have been very helpful to use agreed rules or protocols as a way to address the challenge. Despite this, he noted that generally the participants were 'pretty decently behaved', even in circumstances in which they may have decided they could do no more in a particular class. He was clear that the issue of behaviour management would require further focus and attention for any future program offering. In reflecting on the program, ME indicated that he could have been given more advice with respect to aligning his teaching with the objectives of the research, in order to promote their achievement. We believe this presents a useful observation for any future such programs that seek to adopt a research framework from the outset. Noting the struggle many participants experienced with breath-work and stillness, ME reflected that they might benefit from a more direct introduction of meditation, whether introduced as part of yoga or separately. He noted: [T]hese people are anxious, suffering all sorts of things, and bored, so it is really hard for them to just sit. It would be really good to think about how other mindfulness meditation could be delivered … how to incorporate it if you're doing yoga or teach it separately.
In ME's view, the gym-an indoor basketball court-was not an ideal yoga space. It was large and amplified noise, muffling his instructions. It was also difficult to run the program in a space where prisoners would one minute be playing basketball and the next be asked to sit and focus their attention on breath and yoga poses. Accordingly, it would preferable to offer the program in a quieter space, not associated with boisterous sporting activities.
In considering whether a yoga program could be offered on an ongoing basis, ME reflected on the importance of prisoners experiencing a sense of accomplishment. In his view, whatever the length of the program, there needed to be a goal to enable participants to experience success after having committed themselves to the program. In this regard, he again noted the efforts of LM, who organised certificates for participants who successfully completed the program, together with instigating and facilitating the graduation ceremony, in which each participant was given their own yoga mat to keep.

Prison psychologist
As noted above, we interviewed LM about her perceptions of the program. She indicated that ME was 'a really good choice of instructor because he was very calm, unassuming and quiet', making him 'a really good match for that demographic'. She took the view that a male instructor was ideal for male prisoners. She also suggested that having a member of staff attend was important for behaviour management and to ensure respect for the teacher, noting that that staff member would have an ongoing relationship with the prisoners. She did not think that the participating staff member would have to be a member of the prison psychological services team and suggested that the program could run successfully with a 'fit uniformed staff member', for example, 'an activities officer'. In her view, prisoners 'just like seeing one of us be equal to one of them. I think that's really important'.
LM also reflected on the importance of adaptation, bringing in the concept of challenge and ensuring that sessions changed their rhythm to avoid the risk of boredom. She had been concerned about this at one point and encouraged ME to introduce a new challenge at the beginning of the next class, which helped to re-engage participants' interest. This is an important observation for future programs and could easily be incorporated, with a new practice or pose being identified in each class as the 'new challenge'.
For LM, perhaps unsurprisingly, the biggest challenge was logistics. She reported that it took her two-and-a-half hours each Monday to organise and arrange prisoner movements to the class, participate and then arrange prisoner movements from the class. She referred to this as 'operationally' a long process. She reflected that some movement challenges-including on a Monday that was a public holiday-could be addressed if each prisoner was given a letter that they could provide to officers in charge of their management indicating that they had been approved to participate. In her view, this might avoid the need for repeated personal communication of approval. Discussion and conclusions Viorst (2017: 30) recently commented on 'the potential benefits of yoga for prisoners and for society at-large through cost-effective mindfulness interventions in incarceration settings', while Auty et al. (2017: 706) concluded that 'there is sufficient evidence to date to suggest that yoga and meditation practices have promising effects on prison populations'. They also observed that: [T]he practicability of initiating yoga and meditation programs in prison settings should be considered. The various potential benefits of practicing yoga or meditation are easily achievable, due to favourable logistics in terms of program delivery. Yoga and meditation classes require little equipment and can cater for relatively large groups. Classes can be provided in a cost-effective manner. (2017: 694) The objective of this paper was to set out our experiences of establishing a pilot yoga prison program in Australia, in the hope that this may assist other yoga teachers and correctional agencies in following suit. To our knowledge, this is the first time that a detailed picture of the process of establishing such a program has been set out.
Several themes emerged from our experience. First, it is clear that relationships-among prisoners and with the teacher and correctional staff-are key to the success of such a program.
In particular, we would support the implementation of programs that promote prisoners' innate dignity and worth and treat them with respect, as would be the case in a community yoga program. We also understand better than we did at the outset how complicated the logistical aspect of delivering such a program is. Accordingly, even programs that are delivered on a voluntary basis require the commitment of correctional staff and, ideally, active participation in the program by a staff member. We believe LM's support for and participation in the classes, as well as her behind-the-scenes work to move participants to and from class, were crucial to it running smoothly and effectively. In the context of whether the program is run on a voluntary or paid basis, several participants remarked that they appreciated that ME was donating his time to run the program. We are aware that yoga programs are run on a voluntary basis in the UK by the Prison Phoenix Trust and in New Zealand by the Yoga Education in Prisons Trust (2018). Our experience working with the Yoga Foundation suggests that Australian yoga teachers may also be willing to donate their time to such a project. However, we would advocate paying yoga teachers for their time, out of respect for their training and profession and in recognition of the benefits of such programs.
We suggest that consideration be given to ensuring that the space in which any yoga program is held is conducive to the practice, including its quiet meditative aspects. That said, one benefit of running the pilot program in a gymnasium was the adaptation made by the teacher to move the class from a standard row format to a circle with him at the centre. This increased both the potential to communicate and also the sense of participation and equality, with no participant in any better position to receive instruction than any other. The choice of location should be made to best facilitate prisoner movement to and from class, addressing a key logistical challenge.
We suggest that rules and protocols be established at the outset, ideally in collaboration with program participants, to manage issues such as distracted or uncommitted participants. Providing a pictorial guide for yoga practice to assist prisoners motivated to practise outside class would also be of assistance. 6 We note that establishing a program of this nature may also take more than one iteration to engage participants across a prison; our experience suggests that as early adopters speak to other prisoners about the benefits of the program, interest will grow among those who may initially be wary of something 'weird' like yoga. We would also advocate the inclusion of a graduation program at the end of a set period, which should be communicated to participants at the outset. Some participants indicated that they had never completed a IJCJ&SD 59 Online version via www.crimejusticejournal.com © 2019 8(4) program in their lives and were excited and grateful to receive their certificate and yoga mat. Giving participants a mat to keep may promote ongoing practice, while the role of graduation rituals has been recognised in the desistance literature (Maruna 2001).
Finally, we recognise that working to advance prisoner wellbeing may be viewed to presupposing an acceptance of incarceration and the rapidly rising rates of incarceration in Australia. This is not our position (see e.g. Bartels 2016). However, whether the focus is on wellbeing, desistance or decarceration, there can be no turning away from the lives of those we incarcerate. Regardless of both their offending and the merits of imprisoning them, each prisoner is a person with inherent dignity, worthy of concern and compassion (Hopkins and Bartels 2019). Indeed, for far too many, their pathway to offending is very much a function of compromised wellbeing resulting from life circumstances not of their choosing. We hope that this article will be understood as joining in the body of scholarship that turns towards the lives and experiences of prisoners. We also hope that it may help open the door for more yoga and meditation programs in Australian prisons.