Introduction
In Canada, women constitute a small but growing subgroup of the federal correctional population (Correctional Service Canada 2014; Zinger Reference Zinger2018), both in prison and under community supervision (McConnell et al. Reference McConnell, Rubenfeld, Thompson and Gobeil2014). The topic of women’s correctional and supervisory experiences, including women’s interpretations of re-entry and reintegration, is a growing area of inquiry (Berg and Cobbina Reference Berg and Cobbina2017; Carlton and Segrave Reference Carlton and Segrave2016; Morash et al. Reference Morash, Kashy, Bohmert, Cobbina and Smith2017; Heidemann, Cederbaum, and Martinez Reference Heidemann, Cederbaum and Martinez2016). Within this literature, researchers have examined different facets of reintegration, including, for example, the barriers women face upon re-entry (Carlton and Segrave Reference Carlton and Segrave2016; Berg and Cobbina Reference Berg and Cobbina2017; Morash et al. Reference Morash, Kashy, Bohmert, Cobbina and Smith2017), factors associated with or that promote desistance (Stone et al. Reference Stone, Morash, Goodson, Smith and Cobbina2018; Opsal Reference Opsal2012; Brown and Ross Reference Brown and Ross2010), and, relatedly, how parole governance shapes post-release experiences (Johnson Reference Johnson2015; Opsal Reference Opsal2015).
Regarding experiences of parole, researchers have illuminated the tensions inherent to supervision and reintegration and examined how parolees navigate parole governance in light of the realities of re-entry (Johnson Reference Johnson2015; Opsal Reference Opsal2015; Werth Reference Werth2012). In particular, parole conditions, which are binding rules set at release by parole agencies, are noted to structure post-release experiences in ways that may take precedence over other social objectives (e.g., Binnall Reference Binnall2019; Turnbull and Hannah-Moffat Reference Turnbull and Hannah-Moffat2009). In some cases, researchers have found that parole officers (POs) use parole conditions as the framework for parole governance, to the potential detriment of other case management objectives, such as the provision of support and assistance in relation to parolees’ reintegration efforts (Bonta et al. Reference Bonta, Rugge, Scott, Bourgon and Yessine2008).
In the current study, we examine how release conditions shape parole experiences, focusing specifically on the effects of non-association conditions for women under federal community supervision in Canada. Qualitatively analyzing the parole files of forty-three women, we consider how restrictions on association shaped participants’ reintegration efforts. We suggest that restrictions on associations can operate in a paradoxical manner: on the one hand, attempting to steer women away from negative influences and, on the other, potentially alienating women from supports, activities, opportunities, and roles that support reintegration efforts.
Women and Reintegration: Recidivism and Desistance
Previous researchers have found certain factors to be linked to the movement towards desistance from crime, including aging (maturation), marriage, stable employment, and positive identity change (Laub and Sampson Reference Laub and Sampson2001). “Integrated” theories of desistance point to the roles of individual-level factors (e.g., identity change; King Reference King2013; Healy and O’Donnell Reference Healy and O’Donnell2008; Vaughan Reference Vaughan2007) and socio-structural factors (e.g., housing and employment opportunities; Farrall, Bottoms and Shapland Reference Farrall, Bottoms and Shapland2010) in shaping post-release outcomes (LeBel et al. Reference LeBel, Burnett, Maruna and Bushway2008; Bottoms et al. Reference Bottoms, Shapland, Costello, Holmes and Muir2004; Laub and Sampson Reference Laub and Sampson2001).
Despite an original focus on men, desistance scholarship has come to produce a wealth of knowledge on the factors that support or impede successful reintegration among women (Cobbina Reference Cobbina2010; Brown and Bloom Reference Brown and Bloom2018). Berinbaum (Reference Berinbaum2009) has identified several factors that affect the release outcomes of women, including access to housing and education/employment, substance use issues, childcare options, mental health issues, and other individual factors (e.g., problem-solving skills). More recently, Rodermond et al. (Reference Rodermond, Kruttschnitt, Slotboom and Bijleveld2016) studied factors that promote desistance among women; identified factors included motherhood (having children), supportive relationships, economic independence, abstinence, and agency. Access to higher education may also positively affect women’s desistance (Brown and Bloom Reference Brown and Bloom2018).
Researchers assert that social bonds, via avenues such as family and work, are key to promoting social conformity and pro-social living (see Maruna Reference Maruna1999). Many of the factors deemed supportive of desistance are fundamentally connected to social bonds. However, the influence of social networks and associations on women’s desistance is mixed and nuanced, likely reflecting the complexity of, and variation in, relationships. This is particularly true given that pathways to incarceration for women are often tied to familial/relationship factors and rooted in unequal social dynamics; thus, social networks may not have a simple or straightforward effect on reintegration (Brown Reference Brown2006).
In the case of men, researchers have found that marriage and spousal relationships are supportive factors in desistance; some researchers suggest this may hold true, albeit to a lesser extent, for women (e.g., Bersani, Laub and Nieuwbeerta Reference Bersani, Laub and Nieuwbeerta2009). Regarding familial ties, Barrick, Lattimore, and Visher (Reference Barrick, Lattimore and Visher2014) found contact with family during and following incarceration serves as a protective factor for women, whereas non-contact serves as a risk factor. Upon release, Opsal and Foley (Reference Opsal and Foley2013, 271) note “family can provide economic resources, housing, childcare or emotional support in the precarious post-incarceration months when re-entering offenders are typically less able to rely solely on themselves.”
Familial relations may also be important for desistance at the level of personal identity. In the Canadian context, Doherty et al. (Reference Doherty, Forrester, Brazil and Matheson2014, 574) identified family support as an essential element of reintegration and the ability for women to enact personal change. Based on interviews with released women, they found that family members play a supportive role in overcoming the shame associated with a criminalized identity. The authors argue: “[t]ransforming one’s life does not occur in isolation but in connections with families and friends and with mutual belief that change is possible.” In a similar vein, Leverentz (Reference Leverentz2011, 262), analyzing the experiences of Black women in the United States, notes that familial ties may be important to women’s identity, even “when they are a source of strain or when they come into conflict with desistance efforts.” As caregiving roles are central components of identities, breaking familial ties may be “neither feasible nor desirable” for many women.
Some researchers suggest that, for women, relationships are often tied to pathways to incarceration and/or re-incarceration (Brown Reference Brown2006; Leverentz Reference Leverentz2006; De Li and MacKenzie Reference De Li and MacKenzie2003; McNeeley Reference McNeeley2019). As Brown notes, unequal relations of power—tied to relationships and social structures—mean that familial factors promoting desistance among men, including marriage and childrearing, may interact differently with women’s crime and desistance experiences. In the case of childrearing, the effect on re-entry is indeed not straightforward; some research suggests motherhood may be key to women’s experiences of desistance (Sharpe Reference Sharpe2015; Michalsen Reference Michalsen2011), although the presence of children may also precipitate stress, which in turn may drive a return to crime (Michalsen Reference Michalsen2011) or may otherwise not prevent certain types of offending (Brown Reference Brown2006).
How women experience social bonds and other aspects of reintegration may be shaped by elements of parole governance, in turn laced with gendered assumptions (Wyse Reference Wyse2013) and heteronormative ideals (Kerrison Reference Kerrison2018). Analyzing the experience of men and women on parole and probation in the United States, Wyse (Reference Wyse2013) found constructions of rehabilitation are mediated by gendered paradigms; whereas for men, individual deficits are the site of intervention, women’s interventions are linked to social boundaries with others. As a result, “officers’ rehabilitative efforts center on policing these boundaries, encouraging emotional independence, and discouraging romantic relationships” (Wyse Reference Wyse2013, 249–250).
For both men and women, studies of parole governance have emphasized the increasing salience of parole conditions (i.e., rules and requirements of release as set by parole bodies) in shaping how parolees are governed and managed in the community (Travis and Stacey Reference Travis and Stacey2010; Bonta et al. Reference Bonta, Rugge, Scott, Bourgon and Yessine2008; Turnbull and Hannah-Moffat Reference Turnbull and Hannah-Moffat2009). One set of conditions relates to restrictions on associations—that is, with whom and under what circumstances a parolee can interact with others (Opsal Reference Opsal2015; Binnall Reference Binnall2019). Binnall (Reference Binnall2019) has explored the implications of restrictions on “inter-offender relationships” among parolees in the United States. He argues that such conditions, restricting how, and with whom, individuals associate, are based on the notions that criminal justice-involved persons are perpetually and characteristically anti-social, and that associate-related risk is universal across individual parolees. Such assumptions thus obscure the particularities of individual histories and circumstances and depart from evidenced-based practices rooted in tailored interventions. He notes that such restrictions are typically applied in a blanket fashion, even when associates are not a criminogenic factor; in such cases, “categorical offender no-association conditions are superfluous and potentially detrimental to their reintegration” (Binnall Reference Binnall2019, 47).
Similar themes are evident in Opsal’s (Reference Opsal2015, 200) analysis of women’s reintegration and parole governance in Colorado. Speaking to the important role of social bonds in reintegration, she found that conditions that restrict associations with other criminal justice-involved persons can prevent women from accessing the social support they needed to remain crime free: “most participants explained their networks of support were made up mostly or entirely of people with a history of involvement in the criminal processing system.” Opsal found some women ignored non-association conditions, while others alienated themselves from their networks in attempts at adherence. In the latter cases, Opsal finds “parole conditions stipulating what appropriate relationships looked like meant that they established no bonds at all, thus isolating themselves from people, relationships, and sources of emotional support.”
The works of Opsal and Binnal speak to the tensions that can emerge between parole governance and the social realities of releasees. While intended to reduce criminogenic factors, restrictions on one’s social interactions and relationships can lead to social isolation rather than reintegration, and may unintentionally impair factors that are supportive of desistance. Further, parole governance may be shaped by normative assumptions regarding gender, relationships and other facets of social life—thus implicating the substantive scope of community supervision (Kerrison Reference Kerrison2018). In sum, women’s relationships when released under community supervision may be shaped not only by inherent tensions (e.g., elements of support and strain; Leverentz Reference Leverentz2011), but also by the way in which associations are governed by supervisory agents.
Current Study
Contributing to the literature on parole conditions and reintegration, we explore the impacts of non-association and no contact orders on women’s community reintegration. We theorize how associations are integral to reintegration yet can also be linked to returns to criminal behaviour, illustrating the challenge women face when navigating the social terrain of their post-release realities, and a tension that shapes parole governance. Specifically, while recognizing that certain associates may, in certain ways, undermine positive reintegration efforts, we identify how restrictions on association may also limit women’s abilities to access support, move forward in pro-social pursuits (e.g., obtaining employment and housing), re-establish familial and parenting roles, and pursue intimate relationships. Furthermore, we consider how restrictions of social relationships may serve as an impediment to responsive case management relationships. In line with the findings of Opsal (Reference Opsal2015), we suggest that parole governance can operate in tension with the larger goal of reintegration.
Method
The data for the current study comes from a larger project on desistance and reintegration among self-identifying men and women serving federal prison sentences in Canada. Federal prisons in Canada are run by a central agency, the Correctional Service of Canada, and primarily house individuals who have been sentenced to a prison term of two years or more. The vast majority of federal sentences include a component of community supervision. Statutory release is typically automatic in the case of those serving determinate (fixed) sentences and occurs when an individual has served two-thirds of their sentence in custody (Correctional Service Canada 2019). Individuals serving determinate and indeterminate sentences may also apply for day or full parole at specified datesFootnote 1; however, release is not automatic and must be granted by the Parole Board of Canada (PBC). Conditional releases include standard parole conditions (e.g., reporting requirements, adherence to the law) (Correctional Service Canada 2019), as well as ‘special conditions’ as determined by the PBC. In the community, parolees are supervised by parole officers (POs); however, their case management team may also include other actors such as program officers, mental health care professionals, social service workers (e.g., employment counsellors) and halfway house staff.
Participants for the current analysis include women referred to and recruited by staff at a day reporting centre in a large urban setting. The centre is non-profit and provides case management support (e.g., in finding housing or employment) to individuals under community supervision who are deemed to require additional assistance. It is therefore important to note that the sample may not be reflective of the federal population of women parolees, given that referral was based on perceived elevated case management need (i.e., they were experiencing difficulty in at least one element of reintegration). While specific results are confined to this sample, we present general implications that may have wider applicability in understanding social bonds, reintegration, and parole governance.
Recruitment for the current study typically occurred upon enrollment at the centre; specifically, centre staff informed the women about the study and provided an opportunity for the inclusion of their data. Participation in the research was voluntary and participants provided informed consent; there were no benefits or incentives to participation. Data was accessed through a collaborative research partnership with the Correctional Service of Canada, which enabled access to participants’ case files. As participation in the study was confidential and data was anonymized, we omit names and identifying information in the presentation of findings.
To understand women’s release trajectories and reintegration experiences, we analyzed parole documents (including case notes), decision documents, and other case file documents (e.g., criminal profiles, correctional plans). Methodologically, we employed documentary analysis, which involves the analysis of documents not originally intended for empirical inquiry (Ahmed Reference Ahmed2010). As Ahmed (Reference Ahmed2010, 6) notes, in conducting qualitative documentary analysis, “the structure used must first be derived from the data, which means systematically analysing it so as to tease out themes, patterns and categories.” Rather than analyzing data according to pre-determined themes, we considered the themes that emerged “from the data,” particularly those that were pronounced both within and across cases.
We first organized themes within cases, paying attention broadly to women’s experiences of reintegration; we then systematically analyzed and coded case summaries for emergent themes, using a semi-grounded constructionist approach (Charmaz Reference Charmaz2006; Glaser and Strauss Reference Glaser and Strauss1967; Ricciardelli, Clow, and White Reference Ricciardelli, Clow and White2010). Specifically, one author coded all data into emergent themes manually and the authors then collectively engaged in data analysis and interpretation. The current analysis focuses on one prominent theme that appeared salient within and across cases—namely, the role of non-association conditions on women’s reintegration experiences.
Methodologically, the study provides a unique lens through which to examine parole governance, as we are unaware of any other Canadian research examining parole using case file documentation or with a particular focus on non-association conditions. However, the current analysis, which relies on parole documentation, does not capture the perspectives of parolees or case managers and should therefore not be interpreted as representing their voices. We recognize that our document-based approach is based on formal parole notes and decision documents and therefore does not capture all elements of parole and case management experiences. Like any methodological approach to social inquiry, the study provides a certain vantage point through which to examine social experiences; only select events and perspectives are captured. Notwithstanding these limitations, we suggest that parole and case file documentation provides a rich chronological account of release experiences, and, given the centrality of such documentation in parole-related decisions, an insightful object of study.
Results
Overview of Sample
For the current analysis, we used information for participants’ current sentence (i.e., the sentence for which they were referred to the day reporting centre). The sample included forty-three women, of whom nineteen (44%) identified as black, eleven (26%) as white, four (9%) as Indigenous, and the remaining identified as “other.” The majority of women (54%) were serving time for a drug-related offence, 86% were serving their first federal sentence, and all except one was serving a determinate (fixed) sentence. Around half (51%) were identified as having a mental health condition, 44% had a documented history of suicidal/self-injurious behaviour, and 40% had a history of substance misuse. Around half (51%) of the sample had not completed high school. Following their initial release, twenty-one of the forty-three women had their release suspended, either for breaching their conditions or due to another indication that their risk level had escalated (i.e., their risk was deemed no longer manageable in the community for reasons such as re-offending, exhibiting behaviour or actions suggesting re-offending would reoccur, or absconding from supervision).
Non-Association Conditions
A majority (53%) of participants were initially released on day parole. Day parole typically requires residency at a halfway house, as well as adherence to conditions of release (Correctional Service Canada 2019). In addition to standard conditions, women also had an average of five special conditions attached to their release (i.e., determined by the Parole Board of Canada in each individual case; Correctional Service Canada 2019). In all but one case, special conditions included restrictions on associations with others, such as the requirement to “report all relationships,” restrictions with specified individuals, and general restrictions with categories of people. We discuss each in turn.
The requirement to “report all relationships” typically requires releasees to report emerging or desired interpersonal relationships, including romantic relationships, friendships, and associates. Typical wording includes: “immediately report all intimate sexual and non-sexual relationships with females to your parole supervisors” (participant 87); or “immediately report all intimate sexual and non-sexual relationships and friendships with males to your parole supervisor” (participant 157). Such conditions require the use of discretion among POs/supervisors in that they must determine whether or not to permit the association. The condition to report relationships was noted in sixteen of forty-three (37%) cases.
Conditions requiring the releasee to “avoid certain persons” had both specific and general iterations. Specific iterations prohibited contact with identified individuals (e.g., victims, co-accused, or persons deemed to present a risk for re-offence). For example, five individuals had conditions not to associate directly or indirectly with victims of their offences, nine had conditions not to associate with their co-accused, and six had conditions not to associate with other specified individuals (such as family members or spouses). Regarding specified individuals, an example was restricted contact with one’s family: “Not to associate with any members of your immediate family specifically your mother, father, and brother, unless in the context of family counselling sessions” (participant 106). Of note, individual categories often overlapped (e.g., victims or co-accused were also family members).
The most general iteration of “must avoid certain persons” involved restriction on associations with criminal justice-involved individuals. This condition was by far the most common condition related to association, noted in thirty-nine of forty-three (91%) cases, and typically read “not to associate with any person you know to be involved in criminal activity or have reason to believe is involved in criminal activity” (participants 63, 159, 32, and several others). In some cases, general non-association conditions also restricted association with categories of people, such as children, drug-involved persons, and sex trade workers; for example, “Not to be in the presence of any children under the age of 16 unless accompanied by a responsible adult who knows your criminal history and has been previously approved in writing by your parole supervisor” (participant 86, similar for 119 and 158); “Avoid sex trade worker” (participant 110).
While the wording of general conditions would appear to restrict association with persons who were actively involved in criminal activity, POs often interpreted these conditions to include individuals with a criminal record, even if the criminal offending was historic. In cases where parolees wished to have contact with individuals restricted by the condition, the PO could perform an assessment on the contact, with the purpose of determining whether the association would be supportive or detrimental to the individual’s reintegration. Therefore, POs have discretion when it comes to determining approved associates. While there were cases in our sample where a contact’s criminal record did not preclude association (i.e., the relationship was permitted, albeit often under certain conditions), the tendency was for POs to interpret general non-association conditions in a broad fashion, thereby attempting to limit parolees’ exposure to individuals who had any form of criminal justice involvement.
Community Assessment of Contacts
In addition to explicit restrictions, parolees’ associations are restricted by virtue of the fact that any community contacts (e.g., friends, family members, partners) with whom they wish to visit or reside may have to be approved by the PO through a process known as a community assessment. A community assessment involves the PO meeting with a proposed associate and reviewing their circumstances and relationship with the parolee to determine the influence (e.g., supportive, criminogenic) they might have. Thus, the community assessment process requires the releasee’s associates to undergo a degree of scrutiny. While the process is intended to discern the impact of distinct influences, the process can make access to community supports difficult for the released person. Obstacles to access include case backlog (resulting in delays in community assessments being performed), hesitancy on the part of associates to participate, and negative outcomes resulting from assessment.
Conditions and Reintegration
Non-Association with Criminal Justice-Involved Persons
We first examine the most common special condition among women, namely restrictions on associations with those involved in crime. The general non-association condition often affected women’s peer networks, as previous researchers have noted (Opsal Reference Opsal2015; Binnall Reference Binnall2019). While some women wished to “move on” from past associates, others expressed a desire to access support from individuals facing similar life situations. When it came to incarcerated friends, POs typically denied requests for permission to maintain contact. Likewise, contact with other former prisoners (i.e., those on supervised release) was prohibited or strongly discouraged. Given their social realities (e.g., having spent years in prison, or having many friends with criminal records), the general non-association condition could result in sudden social disconnect.
For instance, a PO informed participant 158, who was serving a five-year sentence, that upon release she could no longer have contact with her incarcerated friends. Her files demonstrate that accompanying the sudden and abrupt loss of contact with her longstanding friends from prison was restrictions on who she could associate with in the community. After breaching her conditions by contacting her friends inside, she noted the difficulty of unexpectedly and immediately cutting ties with individuals who had played key roles in her life. The termination of prison relationships in the context of her supervision served as a key “pain of release” (Durnescu Reference Durnescu2019; McKendy and Ricciardelli Reference McKendy and Ricciardelli2020). Her decision to breach the non-association condition, thus putting her release status in jeopardy, demonstrates how she prioritized social bonds when navigating parole requirements.
Somewhat paradoxically, non-association conditions sometimes meant women could not associate with fellow halfway house residents. Participant 37, despite her actions appearing non-criminal in nature, was disciplined and threatened with suspension after she associated with co-residents in public. Found in the same vehicle as another resident, and, in another instance, calling a resident from her place of employment, the PO dubbed these actions serious and warned the woman about the prohibited nature of such interactions. Subsequently, after being stopped by police in a vehicle with another halfway house resident, the participant’s release was suspended (despite documents indicating the police identified no wrong-doing and that the resident was apparently providing her with a drive to work). Although her suspension was cancelled (allowing her to return to the community), her PO contacted her employer to verify her schedule, which “outed” her parole status and resulted in her termination. Her seemingly non-criminal interactions with fellow halfway house residents thus had considerable repercussions on her reintegration efforts. Again, her decision to associate with fellow halfway-house residents suggests that parole requirements were at odds with, and perhaps de-prioritized by, competing re-entry efforts.
General association conditions could also affect women’s efforts to pursue employment opportunities. During re-entry, employment is a key area of concern; however, participants typically faced barriers to employment (i.e., many lack formal education or employment experience).Footnote 2 While employment goals were often central to the PO’s case management approach, clear instances emerged across participant files where conditions tied to association further restricted participants’ already constrained employment opportunities. For example, information-sharing among parolees regarding “felon friendly” employers, i.e., those who do not perform a criminal record check or otherwise exclude job candidates due to a criminal background, sometimes occurred among fellow halfway house residents. However, POs sometimes restricted fellow residents from working together (e.g., 106, 110, 157) or, in certain cases, established restrictions that made it virtually impossible to be co-employed (e.g., imposing the condition that co-residents could not work the same shifts). In practice, participants, as prospective employees, are left needing to make unrealistic demands on their employers (e.g., requesting their shifts not overlap with other employees). In such instances, perceived associate-related risk appears to outweigh case management objectives tied to employment.
Association restrictions also interfered with employment options in cases where women sought to work in places deemed risky because of location or circumstances (e.g., 110, 116). To exemplify the latter, participant 116’s romantic partner offered her an administrative position in his company; however, as he was a non-approved contact, she was prohibited from taking the position. Parole notes indicate that she described the non-association condition as preventing her from obtaining employment and thus working towards a pro-social existence.
Housing and accommodation were also at times constrained by non-association conditions—particularly for participants released into urban centres with notoriously high rental costs who needed low-income housing options, such as rooming houses, subsidized housing, or rooms offered for cheap or modified rent (e.g., in exchange for cleaning services). Parole officers could disapprove housing options on different grounds, such as the presence of possible negative associates. For example, participant 32 applied for and received a room in a women’s transitional house with subsidized rent; however, her PO would not approve the housing given concerns about her risk and the presence of three other roommates. In another case, participant 110 was not approved to move into her mother’s residence (where her young child resided) because her mother had a criminal record. Ironically, the denial of housing opportunities forced women to remain at a halfway house—where all residents have criminal backgrounds.
Restrictions on associations with crime-involved persons are based on the assumption that individuals with a criminal record are a criminogenic influence (Binnall Reference Binnall2019). Yet the assumption can obscure the fact that incarceration shapes (and is shaped by) an individual’s social realities and circumstances, including one’s social network and friends (Opsal Reference Opsal2015). As illustrated above, such restrictions often had wide-reaching effects on parolees’ social situations, as participants often had many associates with criminal records, including friends they had done time with, halfway house co-residents, family members, spouses, and other acquaintances. Adhering to non-association conditions may lead to social disconnection, although clear instances emerge where social ties are prioritized above parole requirements. Thus, general non-association conditions can present a multitude of challenges for women attempting to rebuild post-institutional lives and draw support from their networks, although women vary in how they navigate social bonds in light of non-association conditions.
Non-Association Conditions and Familial Roles
General and specific non-association conditions can also affect familial and parental relationships. In some cases, participants had direct non-association orders that restricted contact with specified family members (e.g., children, siblings, partners, and parents). In other cases, access to family members was constrained by virtue of non-association conditions that restricted association with criminal justice-involved persons. The ways women navigated contact with family members in light of association conditions varied, but in some cases, POs denied a positive community assessment for contact, meaning parolees were not permitted to see or spend time with family members. Such decisions could limit access to social support—a factor that can be vital for reintegration (Opsal and Foley Reference Opsal and Foley2013). In other cases, POs allowed contact under specified conditions, offering a more balanced but still restrictive approach to family reconnection.
Some women had association conditions that indirectly limited their ability to see their children, as Opsal (Reference Opsal2015) similarly found. Participant 106’s conditions stipulated that she could not see her parents or siblings, who had custody of her child, until family counselling sessions were completed and a psychologist determined that familial interactions were no longer a source of criminogenic risk. In this way, the parole experience implicated (and burdened) the family, and the woman faced considerable difficulties in attempting to see her child. Her central focus upon release was appealing the condition and regaining access to her child. Thus, on top of dealing with the normal challenges of reintegration, such as finding employment and housing, she had to navigate the complexities of a non-association condition that affected her entire family. Her PO and case management team recognized the tension that existed between supervision conditions and the process of reintegration, and attempted to relieve it in favour of the latter. Her PO advocated several times on her behalf to have the condition removed so she could access family support and, most importantly to her, reassume her mothering responsibilities.
Participant 32’s child was also in the custody of her mother, who had a dated criminal record. The participant’s general condition restricting contact with criminal justice-involved persons was interpreted to include her mother. As in the case of participant 106, the condition made it difficult for her to see her child (e.g., a third party was required for visitation). While participants in both cases were not outright denied access to their children, the complexities underpinning access meant that they could not assume basic parenting roles and responsibilities. Thus, these women were constrained from enacting the mothering identities that could potentially promote the identity work associated with desistance (Sharpe Reference Sharpe2015) or that are otherwise important during re-entry. In addition, while a PO’s actions are likely underpinned by the objective of limiting exposure to perceived risk, the hyper-scrutiny of and restricted access to family members appear to undermine rapport and trust between the PO and parolee, potentially to the detriment of responsive case management.
No contact orders with current or former spouses who were co-parents also complicated access to children. In some cases, contact between co-parents had to be indirectly facilitated through grandparents, as direct contact was prohibited. For example, participant 172 had a non-association order with the father of her children, who regularly visited the children (at her mother’s residence) and provided material support. Despite his seemingly supportive role, all contact and interaction was required to be strictly through the child’s grandmother (i.e., no direct interactions).
While less common, parolee contact with their children or grandchildren can also be explicitly prohibited, typically in cases where safety considerations are at play. Participant 158 was denied access to her children by virtue of a parole condition that restricted contact with both her minor and adult children. In this case, the inability to see her children became a salient “pain of release” for her (Durnescu Reference Durnescu2019; McKendy and Ricciardelli Reference McKendy and Ricciardelli2020), and establishing permission to see her children became her primary focus, again illustrating how appealing parole conditions can dominate post-release efforts (at the potential cost of other reintegration undertakings). After unsuccessfully attempting to appeal the condition, she tried to work with her PO and the children’s father to establish contact—a drawn out, emotionally draining process complicated by tensions between herself and the children’s guardians. When she was eventually permitted to have regular, albeit restricted, contact with her children, her case management team viewed her parental interactions as highly supportive of her reintegration.
Another woman was restricted from seeing her grandchildren due to having the condition: “not to be in the presence of any children under the age of 16 unless accompanied by a responsible adult who knows your criminal history and has been previously approved by your parole supervisor.” Upon release, she immediately expressed a desire to spend time with her grandchildren, but was informed that, to do so, the Children’s Aid Society would need to be involved. Not wanting to implicate her adult daughter, she decided to forgo contact—making contact with her daughter, a central goal upon release, difficult since she cared for young children. Following approximately two years on release, she was granted permission to visit her grandchildren with her daughter’s supervision. Her PO ultimately deemed her interactions and involvement in her grandchildren’s lives to be supportive of her reintegration.
The community assessment process could also constrain access to one’s family, even when not restricted by non-association clauses. For instance, PO caseloads may lead to backlog, which in turn can result in interim restrictions on contact with individuals who might otherwise serve as positive supports. Participant 113, for instance, wished to spend a holiday at her adult child’s house; however, she did not provide her PO with sufficient time to conduct a community assessment. Not wanting to let her family down, she lied to her PO about her whereabouts and attended the unapproved residence. Once it was discovered that she was not at an approved location, the police were informed and she was arrested in front of her grandchildren, which had considerable fallout on her familial relationships. Here, the prioritization of social bonds, specifically familial roles, parallels findings by Opsal (Reference Opsal2015, 198), who found that mothering identities could take precedence over adherence to parole conditions. Consequently, conditions restricting or complicating access to family may put women “in a lose–lose situation where, ultimately, they become more at risk of failing as both mothers and parolees.”
Non-Association Conditions and Intimate Relationships
Restrictions on associations also affected the governing and experience of romantic relationships—as they too were restricted by both specific conditions (e.g., that prevent contact with one’s co-accused, victim, or other specified individual) and general conditions (e.g., not to have contact with crime-involved persons). Parole officers appeared to be rather restrictive with the interpretation of association conditions that affected romantic relationships—likely due to the perceived possible criminogenic effect of such intimate associates.
Some releasees were prevented by specific or general conditions from associating with their current or former spouse/partner (e.g., participants 21, 25, 74, 106, 109, 110, 111, 116, 159, 172), which, perhaps not surprisingly, resulted in violations and returns to custody. For instance, participant 109 and her spouse, with whom she was formerly incarcerated, were prevented from having any contact with each other, as the Parole Board viewed the two as negative influences on each other. Participant 111 was also prevented from seeing her previously incarcerated partner by way of a general non-association condition. She viewed her partner as a key source of support; however, her PO advised her that contact would not be permitted under any circumstances. In both cases, the women were unsuccessful in amending conditions that prevented them from seeing their spouses and, consequently, breached the condition.
In another case, participant 159 had a non-association condition with her spouse and father of her child, who was an accomplice in her index offence. Communication between the Children’s Aid Society and her PO revealed she was lying about her living situation and had been residing with her spouse. Although there was no evidence that the two were involved in criminal activity—in fact, her performance in a post-secondary educational program indicated positive progress—she was arrested for breaching her non-association condition. Here, breach of the non-association contact again speaks to the prioritization of family ties above parole requirements when tensions between them emerge (Opsal Reference Opsal2015; Leverentz Reference Leverentz2011).
A more balanced approach to the interpretation of non-association conditions in relation to spouses was evident in the case of participant 25. Consistent with cases 109 and 111 above, the participant had begun a relationship with an individual she met while incarcerated. When the relationship came to light the woman indicated she would not terminate the relationship even if it meant returning to custody. In this case, the PO consulted with a lawyer regarding interpretation of the non-association condition, who advised that it should not be interpreted in an overly broad fashion. The PO recognized that the couple could potentially provide each other with mutual support, although also felt the relationship may serve as a risk factor given both partners faced issues with substance use. In response, the PO supported the relationship, allowing association insofar as it served as a stabilizing force, but required the partner to attend supervised visits and provide evidence of receiving substance abuse treatment—thereby, in a sense, expanding the scope of supervision to the releasee’s partner.
Non-association conditions could also affect new relationships. For instance, participant 116 began a new relationship with an individual she felt was supporting her reintegration efforts. However, after he refused to give his personal information to the PO, a no contact order was established. He eventually consented to a criminal record check, which revealed previous and outstanding criminal charges; this in turn left the no contact order in effect. Case files indicate the participant described this no contact order as undermining the support and opportunities he had been providing her (e.g., use of his vehicle, obtaining employment at his business). After the couple continued to push to have contact, her parole was suspended on the grounds that the new relationship served to increase her risk.
In some cases, women concealed their romantic relationships with criminal justice-involved persons until incidents of re-offence revealed such relationships. For example, participant 21, despite appearing to be doing well for several months on release (e.g., she was employed, adhering to conditions and receiving psychological counselling), was arrested with a former prisoner on drug-related charges. Following arrest, her PO learned she had been in a secret relationship with the individual for several months; thus, in addition to her new drug charges, she was found to be in violation of her non-association condition. It is evident that the non-association condition here did not deter involvement with negative associates; rather, it prompted considerable deceit on the part of the releasee, likely to the detriment of responsive case management practices. Similar scenarios, whereby new criminal offences revealed forbidden relationships, were similarly observed in the cases of participants 172 and 87.
Discussion
Prior researchers indicate that associates and social networks play complex roles in both offending and desistance for women. On the one hand, researchers have found that social relationships marked by power and control are often tied to women’s pathways to incarceration and re-incarceration (Brown Reference Brown2006; Leverentz Reference Leverentz2006; De Li and MacKenzie Reference De Li and MacKenzie2003; McNeeley Reference McNeeley2019). Further, researchers have found women are often involved in crime as accomplices or co-offenders (McNeeley Reference McNeeley2019) and that associates are a contributing factor to both offending and recidivism (Law Reference Law2004). At the same time, researchers have found that social supports can be invaluable to women’s reintegration (Doherty et al. Reference Doherty, Forrester, Brazil and Matheson2014). Social networks and familial relationships may not only provide women with forms of material, practical, and emotional support; they may also play an important role at the level of identity (Opsal Reference Opsal2015; Opsal and Foley Reference Opsal and Foley2013; Doherty et al. Reference Doherty, Forrester, Brazil and Matheson2014; Leverentz Reference Leverentz2011).
Given the complex role of family and social networks in women’s pathways to incarceration and re-incarceration, it is perhaps not surprising that the governing of women’s relationships during community supervision becomes fraught with tension. We found that non-association conditions could have considerable impacts on several facets of women’s re-entry experiences. While intended to steer individuals away from negative influences, such conditions obscured the social realities that characterized the lives of parolees. Specifically, we noted illustrations of how non-association conditions placed considerable restrictions on participants’ social networks (with implications on social support, employment, and housing opportunities), limited access to familial and parenting roles, and restricted relationships between partners. While women may navigate the tensions in different ways, non-association conditions undoubtedly contribute to “pains of release” (Durnescu Reference Durnescu2019; McKendy and Ricciardelli Reference McKendy and Ricciardelli2020), and, at times, constrain positive reintegration efforts.
Noteworthy in relation to cases involving child custody was how POs took different approaches in distinct cases (which could reflect the idiosyncrasies of the case as well as the PO), ranging from an advocacy role (participant 158), a more balanced approach (participant 86), to a more restrictive orientation (participant 113), all areas requiring additional research and study. Similarly, the interpretation of the non-association condition in relation to intimate partners varied, likely reflecting distinct case circumstances as well as case management styles.
Association restrictions that prevent contact with one’s social network may set releasees up to fail during re-entry by compromising their support network and inhibiting transparent and honest PO-client dynamics—which are key elements to effective case management practices. While the logic underpinning no contact orders with associates and spouses may be rooted in considerations of criminogenic risk, unrealistic conditions, such as preventing contact with one’s spouse, present a new form of risk that can limit PO capacity. For example, fear of returning to custody may prevent parolees from disclosing key personal information to their PO, including situations where they may be in need of genuine support and guidance (e.g., in leaving an intimate relationship, for example). Insofar as POs and case management staff are unaware of the actual social circumstances of releasees, they cannot provide responsive case management that addresses the needs of the client “where they are.”
Previous researchers indicate that key to positive working relationships in correctional services is the principle of responsivity, in which service delivery incorporates and reflects client needs, as well as the “working alliance” dynamic, whereby the bond between service providers and clients is integral to case management (see Hart and Collins Reference Hart and Collins2014). Parole conditions, which may serve as the framework through which releasees are governed (Opsal Reference Opsal2015), may undermine the potentiality of the working alliance by forbidding relationships and prompting secrecy, which in turn may inhibit responsive case management.
Overall, the centrality of associations in both offending and reintegration raises difficulties for case management, an area warranting future inquiry given that both are compounded by the tension associated with parole governance. Parole officers are required to simultaneously perform a surveillance/control function (e.g., ensuring compliance with parole requirements) and support releasees in their reintegration efforts (Ricks and Eno Louden Reference Ricks and Louden2015; Alexander Reference Alexander2017; Bonta et al. Reference Bonta, Rugge, Scott, Bourgon and Yessine2008). As we demonstrate, POs navigate the tension in different ways, which may reflect the circumstances of the particular case as well as the characteristics and work orientation of the officer. As previous researchers have found, POs can sometimes veer too greatly on the enforcement side of the continuum, which may contribute to poor release outcomes (Chamberlain et al. Reference Chamberlain, Gricius, Wallace, Borjas and Ware2018; Morash et al. Reference Morash, D. A., S. W. and J. E.2016; Vidal et al. Reference Vidal, Oudekerk, Reppucci and Woolard2015). Yet, POs can also have a positive influence on release outcomes and can serve as key sources of support (Holmstrom et al. Reference Holmstrom, Adams, Morash, Smith and Cobbina2017; Morash et al. Reference Morash, Kashy, Smith and Cobbina2018; Ricciardelli Reference Ricciardelli2018; Smith et al. Reference Smith, Cornacchione, Morash, Kashy and Cobbina2016; Roddy et al. Reference Roddy, Morash, Adams, Holmstrom, Smith and Cobbina2019). In the current study, participant cases evidenced the clear potentiality of POs to support their clients’ reintegration efforts as they navigated association conditions.
The current study is limited in that it is exploratory and based on document analysis rather than direct research with parolees or case managers. Future research unpacking perspectives on non-association and other parole conditions could shed further light on how such conditions affect both reintegration and case management at the level of experience. Our study is also limited in sample size, regional concentration, and non-random sampling method. As such, we caution, as with all qualitative studies, against widespread generalizability. Future research examining parole conditions in distinct locales/geographical settings may also shed light on if and how interpretation and experience of conditions is mediated by certain contextual factors. Moreover, the rationale for the conditions of release imposed on the women in our sample was beyond the scope of the current study, although a necessary area of future inquiry. Finally, subsequent research may examine more systematically the effect of association conditions in promoting intended effects (i.e., reducing criminogenic risk) as well as unintended effects (i.e., increasing criminogenic risk).
Conclusion
Our analysis suggests that association conditions may be detached from the social objectives of conditional release insofar as they constrain releasees’ attempts to construct post-institutional lives and identities. Non-association conditions and the restrictions on social contact can have unintended consequences that may actually enhance criminogenic risk by preventing opportunities tied to support, employment, and housing, and by constraining women’s ability to enact social roles geared toward promoting desistance (e.g., parenting). While recognizing that associates may serve as an influence in women’s pathways to (re)incarceration, we suggest that non-association conditions may be interpreted in an overly broad fashion that, at times, departs from the underlying spirit and meaning of conditions of release. This form of interpretation contributes to release frameworks that become self-referential, i.e., guided by technical conditions rather than the broader social objectives associated with supervised release.
