Doing Marginalized Motherhood: Identities and Practices among Incarcerated Women in Mexico

This study examines the mothering practices and identities of incarcerated women in Mexico. Data gathered from repeated life-story interviews with 12 women, were analyzed to describe mothering practices in the different phases of incarcerated women’s’ lives. We argue that knowledge of the Latin American context is crucial to understand their experiences of motherhood. In a society based on familism and marianismo identities that suffers from a lack of welfare institutions, motherhood provided a way for socially and economically excluded women to escape destructive family environments and gain autonomy. Motherhood also provided a way to cope with the stigma of delinquency. Using the framework of Southern Criminology, we explore the importance of marginalized motherhood in this tradition. The results reveal the tragic paradox of motherhood for incarcerated women and the importance of studying marginalized mothering beyond the


Introduction
Studies of crime and incarceration tend to emphasize problematic behavior, such as drug use and violence, or prison conditions. However, for many, and perhaps mothers in particular, the most important element of their lives is neither their crimes nor their drug use, but their children. In interviews conducted with mothers in prisons, the main topic of conversation often relates to their relationship with their children (e.g., Baunach 1985). Imprisoned women struggle with the impact that their criminal lives and their subsequent incarceration have on their mothering ability. To compensate, they try to be "good mothers," despite their imprisonment, drug use and criminal offences (e.g., Hoskins and Cobbina 2019;Stearns 2019). Given incarcerated mothers past and present situations, this is a challenging task.
Engaging in crimes, drug use and sexual relationships outside marriage are often a rejection of or challenge to traditional ideas of gender and femininity (Roberts 1993). In the present study, the incarcerated Mexican mothers simultaneously fulfilled and rebelled against highly traditional gendered expectations, such as getting married, securing a provider or being a homemaker. The tension between their lives and what can be described as normative motherhood (raising their own children within a heterosexual marriage and conforming to the sexual division of labor) was particularly striking within the highly gendered and family-oriented Mexican society. Mexico is a country torn between "traditional" and "modern" values. Despite having enclaves in which equality is the norm, in large areas a notable division between the genders abounds (Herrera and Agoff 2018). The latter was particularly evident in this study as it is in many disadvantaged neighborhoods in Latin America.
This article aims to expand existing scholarship on how structurally disadvantaged mothers negotiate their maternal identities (for a review, see Grundetjern 2018) and to include the experiences of imprisoned women in the Global South. Latin American criminology has examined the experience of imprisoned women in general (Larroulet et al. 2020;Lemgruber 1999;Salazar and Cabral 2012), but less research has been conducted on the experience of mothers in prison. We draw on the life stories of 12 incarcerated mothers to answer the following questions: What characterizes early entry into motherhood in Mexico? What are the motherhood identities and practices of incarcerated mothers? And what role does the Mexican context play? We aim to contribute to the bourgeoning framework of Southern Criminology (Carrington et al. 2016) by addressing these research questions and exploring and discussing some widespread dilemmas facing marginalized and imprisoned mothers in Latin America today.

Motherhood in Prison
In both modern and traditional societies, motherhood is closely intertwined with ideas of femininity and what it means to be a woman. While there has been increasing emphasis on how multiple social identities are in constant flux, being a mother is arguably one of the more stable social identities. For most women who have a child, being a mother becomes a crucial part of their identity (Rich 1986). West and Zimmerman (1987: 125) famously described how doing gender was a "routine accomplishment embedded in everyday interaction." They further emphasized how all identities and statuses are achieved through interactions and should be seen as a product of a system of relationships closely connected to accountability. For example, many marginalized, drug-using or imprisoned women are expected to do motherhood both through accounts of their motherhood and by adopting compensating motherhood practices even when they do not have custody of their children. We refer to this as doing marginalized motherhood. Imprisonment usually involves the separation of mothers and children, which makes it more difficult to fulfill the ideal of being a "good mother" (Luther and Gregson 2011: 85). Serving a prison sentence hinders the enactment of a social role and limits a mother's contact with her children, making it more difficult for mothers to play an important role in their children's lives (Jensen and Du Deck-Biondo 2005). The lack of liberty and autonomy impedes regular maternal practices, such as "nurturing, protecting and training" (Arendell 2000) and limits the capacity of mothers to make decisions in their children's lives. Traditional gendered expectations, such as creating a sense of family and home, are also hard to achieve while imprisoned (McGraw and Walker 2004). Studies of mothers in prison have emphasized feelings of guilt (Imber-Black 2008), depression, anxiety, and stress (Arditti and Few 2008). Such studies have also highlighted how social expectations of the mothering role exert an important influence on social identity and self-image (Enos, 2001). Most importantly, difficulties in meeting the expectations associated with the role of being a mother are combined with a new stigmatizing identity as "criminals" and later as "prisoners" (Easterling et al. 2018;Siegel 2011). For example, Easterling and Feldmeyer (2017) observed that women share a widespread feeling of shame and assume a new spoiled identity as a result of their incarceration. Women in prison are often viewed as doubly deviant or are twice stigmatized (Grundetjern 2018) for being both "bad mothers" and "bad women." This is also the case in Mexico (Montoya 2015).
Mexico City has a prison population of 33,309, of whom 1,732 (5.2%) are women (INEGI 2016). The prison population reflects the prevalent poverty and inequality divisions in Mexico. In relation to the overall prison population, 39.3% have been sentenced to prison for property crimes, thefts from passers-by and businesses, and thefts of vehicles, auto parts, and so on. Among the female prison population in Mexico, 89.1% have dependent children, compared to only 62.9% of men (Montoya 2015). Mothers in prison have some specific rights; for example, children under the age of three can stay with them in prison, and mothers cannot be subject to punishments that involve isolation measures when they are pregnant or live with their children in prison. However, the living conditions of mothers and children in prisons are poor; for example, only 51.4% receive paediatric care and only 32.4% receive psychological care (Reinserta 2016). It has also been reported that only 32.1% have access to appropriate medicines for their age, and only 22.9% benefit from nutritional services. Notably, only 76% of the minors of mothers in prison are legally registered. Additionally, only 65.4% of penitentiary centres have play areas for children (64.2% have outdoor play areas) and only 48.6% provide specific dormitories for mothers (Reinserta 2016). Notably, many Mexican prisons offer extensive opportunities for personal development courses and programs, including courses in mistreatment prevention, abuse prevention and early stimulation. These programs are taught by non-government organizations and organized by the undersecretary of the penitentiary system.
Extensive research has been conducted on how imprisoned mothers in the Global North do motherhood, but less is known about how imprisoned mothers in the Global South negotiate their motherhood identities and practices (but see Fleetwood and Torres 2011;Mello and Gauer 2011;Montoya 2015;Ojeda 2015;Verea 2004;Villalta et al. 2019). Many of these studies describe motherhood as a source of pride and a means to overcome both the constraints of prison (Bosworth 1999) and the difficult living conditions outside prison. The present study sought to examine how doing marginalized motherhood becomes a resource for imprisoned women and how this is closely connected to their social and cultural contexts.

Motherhood in Mexico
Cultural specificity (Collins 2000;Segura and Pierce 1993) is crucial to any understanding of marginalized motherhood. For example, it is difficult to understand mothering practices and identities without an analysis of the nature, shape, and dynamics of the family institution in Mexico. Here, a demanding labor market, where jobs are often secured through informal connections, combined with the absence of a welfare state, make family networks crucial. In the face of scarcity, domestic social networks play a fundamental role in the provision of resources. They also serve to reinforce and uphold traditional gender norms and expectations, leaving women vulnerable in a gendered hierarchy of familism. Put simply, familism is often defined as placing the family "ahead of individual interest and development" (Ingoldsby 1991: 57, see also Ruiz and Ransford 2012).
In Mexico, as in many other Latin American countries, socioeconomically marginalized families are often characterized by unplanned pregnancies, teenage marriages, cohabitation, and a patriarchal structure under which women's paid labor is considered secondary to that of men (Echarri Cánovas 2009; Lomnitz IJCJ&SD 10(1) 2021 and Pérez-Lizaur 1987;Robichaux 2002). Family and motherhood are important. Notably, reproductive labor, including caring for elderly or sick family members, siblings, and underage relatives, is almost exclusively the responsibility of women. Having a married couple reside with or near the husband's parents, which is referred to in anthropology as patrivirilocal residency, is another characteristic of Mexican families (Dáubeterre 2002). Generally, these situations foment the crucial role of family in mothering in the form of "shared mothering" as well as "othermothers" and "bloodmothers" (Glenn 2016;Rosenman and Klaver 2008). This tradition originally comes from indigenous communities, but it continues to exist within popular urban sectors. This type of residency enables a classic patriarchal model, which results in "property, residence, and descent through the male line" (Kandiyoti 1988), and also ensures the subordination of women. Consequently, women in Mexico are oppressed in both their family life and in the labor market. This oppression is even more prevalent in the most marginalized sections of society.
Like many other transitional societies with a strong traditional and collective nature (Haj-Yahia 2002;Yount 2011), women in Mexico are the guarantors of family honor. They must manifest an ideal of virtue associated primarily with sexual behavior and sexual reputation (Hirsch 2003) and self-sacrifice. Researchers have observed that "in traditional Latino families, marianismo is set as the ideal, where women are selfless, submissive and subordinate; an object of reverence, or 'elevated motherhood'" (Ruiz and Ransford 2012: 56). Such expectations are difficult for imprisoned women from disadvantaged sectors of society to fulfill.
The present study explores how imprisoned Mexican mothers describe their mothering identities and practices. The emphasis is on motherhood as a symbolic resource for marginalized women and on their struggle to construct favorable identities in a situation characterized by stigmatization (Paat et al. 2018) and power imbalances. Specifically, we study motherhood as an escape from difficult family environments, fluctuating mothering practices and identities, and as a form of ongoing work to repair spoiled identities (Goffman 1963). The results reveal how incarcerated mothers in Mexico strive to achieve normative ideals of mothering, and how motherhood can be a (sometimes tragic) resource for marginalized populations.

Method
The data in this study were gathered from 12 life-story interviews that were conducted with incarcerated mothers in Mexican prisons. The incarcerated mothers were interviewed at a Mexico City prison between January and July of 2019. All were Mexican nationals aged 18-50 (with an average age of 24 years). For all of the women, except one, the highest level of education attained was secondary school and the last job they held required only low qualifications (e.g., a street vendor, waitress, employee, or self-employed). The women had been in prison for between 3 and 16 years (with an average of 8 years). The incarcerated women had been sentenced for a variety of crimes including theft, robbery, homicide, manslaughter, sexual crimes, and kidnapping. Their sentences ranged from 4 to 60 years (with an average of 22 years). The women interviewed had an average of 2.16 children each. At the time of the interviews, five of the women were living with their children in prison.
After obtaining permission from prison authorities, the participants were recruited by the guards following the principles of purposeful sampling (Patton 1990). The aim was to have a heterogeneous sample of women who had at least one child, but who were of different ages and had been convicted of different types of crime. In all cases, informed consent was obtained, and anonymity was assured. No compensation was paid to the women for their collaboration. The interviews, which were semi-structured and organized as life-story interviews (Atkinson 1998), covered participants' lives from their childhoods to the present. The interviews were conducted in visiting rooms or in inmates' cells in Spanish and were translated into English by the authors. On average, each participant was interviewed three times and each session lasted approximately two hours. A flexible research design was adopted and repeated interviews were conducted to allow the participants to emphasize the topics that were most important to them.
Notably, one important topic that arose in the interviews was family relationships, particularly the interviewee's relationship with their own children. Extensive data were gathered about pregnancies, mothering practices outside and inside prison, and what we describe in this study as mothering identities or mothering ideals. The interviewer audio recorded the interviews (with the participants' permission) and then transcribed them. Any identifying information, such as participants' names and particularities about their crimes or other identifying information, was removed.
A thematic analysis was conducted to identify the primary analytic themes (Strauss and Corbin 1990). The data collected in the interviews were analyzed by an interdisciplinary group of researchers, and transcripts were entered into ATLAS.ti to facilitate data management. Categories for analysis and their inclusion and exclusion criteria were consensually agreed upon by the researchers. The themes were inductively drawn from the interviews, but also emerged from concepts identified in the research literature on marginalized motherhood. Thus, the categories and analysis were both data-driven and theoretically informed. In analyzing the interviews, we first coded broadly for a variety of themes, including pregnancy, relationship with children before incarceration and relationship with children while incarcerated. The initial basic categories included all references to pregnancies and children in the transcripts. The second stage of coding was more detailed and involved categorizing participants' mothering practices and identity work.
A study of only 12 participants has some limitations. Notably, we cannot claim representativeness. Further, selection bias and the risk of random results are problematic. However, the strength of the data should be acknowledged. Repeated life-story interviews provide time for participants to reflect on their lives between interviews, for the interviewer to prepare follow-up questions, and for both to get to know each other better than they would in one-off interviews. This is especially important when themes include highly stigmatized practices, such as mothering under difficult circumstances. In such instances, building trust and establishing resonance is crucial to data quality.
A 48-year-old woman conducted the interviews. She had previous ethnographic experience from her doctoral work, and extensive knowledge of the dynamics and codes of, and jargon used by, women in prison. The interviewer developed a good rapport with the incarcerated women, but positionality, class, gender and cultural identity in particular, must be taken into account when interpreting data. For example, the interviewer could have been seen as representative of "the system," or middle-class society, which may have influenced participants' responses. Even though the data have some limitations, we believe they indicate something important about marginalized motherhood in Mexico and Latin America.

Doing Motherhood in Adolescence through Breadwinning, and in Prison
We distinguish between three phases in the lives of the incarcerated mothers: life leading up to their pregnancy (for many, while in their teens), life with children before prison (sometimes in the streets, and other times with abusive husbands and/or families), and life as mothers in prison. We study these phases to better understand the backgrounds to participants' pregnancies and explore how they did marginalized motherhood under difficult circumstances.

Escaping Destructive Families through Motherhood
Most of the women in this study had been abused, emotionally, physically, or sexually, during their childhoods or early adolescence by their neighbors or family friends, among others. This abuse was related to their incarceration and early pregnancies, and defined how they lived and perceived their relationships with men. The abuse they experienced during their childhoods and early teenage years was sometimes related to family honor and familism. For example, one participant's (Ximena, aged 36) problems began when she ran away from home at the age of 14 and began using illegal drugs on the streets. The events that triggered her to run away from home was her exclusion from the family and the abuse that she was subjected to by her father after she was discovered kissing a boyfriend: Ximena: After he beat me up, I hid in the closet, bleeding, all swollen. I stayed there until it was dark. He had beaten me up at around 3-4 pm, I was there, feeling stuck, until 2 or 3 in the morning crying like that, very anxious … everything broke down, everything. The bubble I was trapped in, where I only watched Walt Disney movies, everything was nice, but everything broke there.
Interviewer: And in the following days, did your father say anything to you?
Ximena: He didn't say a word to me. He told my sisters that they weren't allowed to speak to me either because I was worthless, that he didn't want them to be like me. That I was the family's rotten apple and that they couldn't speak to me. I was only going to be the maid of the house.
Witnessing or being the victim of domestic violence is one of the main reasons children and young people abandon their homes (e.g., Anooshian 2005;Giraldo et al. 2006), and this, together with sometimes living on the street or with juvenile friends, often leads to crime (Chesney-Lind and Shelden 1998). For Ximena, it was the start of a criminal and drug-using career, and an early pregnancy.
Another participant (Irma, aged 30) assumed much of the responsibility at home from an early age. Due to the family's needs and in accordance with her socialization into a marianismo identity, she was left in charge of the household and was responsible for her younger brothers and sisters. She described how this created an emotional void: I was so angry when they left me alone with my siblings, I was in charge of raising my siblings. I was run over when I was six years old, and since then my mother had to go out and work and I had to stay with my sister and then with my brother, so I didn't really see my parents and when I did they were resting or wanted to sleep. They didn't want us to make noise, we did house cleaning, I mean home stuff and nothing more, we didn't have much time together as a family, and so each one of us got used to the voids, the holes in the family.
As a young girl, Irma fell in love with a much older man. Given the privation of affection in her family life, and the absence of a caring father figure, she sought to create her own family. She describes how the man was initially "very nice," and that she fell in love with him because he gave her "his attention, his time." Irma noted that even though the man did not have money "it felt like a dream come true … I saw myself living in the family that I wanted, united, with love and attention." A difficult family environment can drive women towards relationships that might not be good for them. Similarly, motherhood can be seen as a solution to neglected teenagers who seek love and connection (Kaplan 1997).
The families of the young women sometimes initiated romantic relationships that led to early pregnancies. Another participant (Carina, aged 40) began working when she was 15 years old. This is where she met the father of her children, who was almost 10 years older than her. Carina's case highlights the role of the patriarchal family in Mexico. For example, it is often up to the parents (usually the father) to make decisions about early marital unions or cohabitation: Carina: Later, he was convinced and asked me to be his girlfriend, and I immediately said "yes" because it was soon going to be February the 14th and how was I going to be alone, without a gift?
Interviewer: How old was he?
Carina: He was 24 … He came in July, he spoke with my father because he wanted me to be his girlfriend officially, and he said he really wanted me, and he wanted to take me out of my house. He didn't want to fool around. He said, "I know I'm older than her, but I want to be serious with her." And then my father asked me, "What do you think mija [my daughter], do you feel capable enough to have a boyfriend?" and I was like, "Ground swallow me up," and I told him "I think so." The strong familism culture meant that the adults made the decisions for Carina, even the important ones related to marriage and cohabitation, which was accepted from an early age. The families of the women in this study often allowed or facilitated them to marry or live with older men as teenagers. Although the initiative sometimes came from the girls, and they were often consulted, the final decision was left to the family. Following the patriarchal logic, the women simply went from living under the rule of their fathers to that of their often much older husband or, in cases of patrivirilocal residency, their husbands' families. The combination of their young age and the fact that other women often took care of their children (e.g., their mothers or mothers-in-law) meant that their identities as mothers could be suspended in time.
The life stories of the incarcerated mothers in this study reveal an embodied vulnerability (Willen 2012). The countless threats and risks to which they had been exposed contributed to their early pregnancies. In most cases, these early pregnancies were usually not interrupted due to a strong religious (Catholic) influence, or because motherhood in large sections of Mexican society is perceived as a destiny rather than a choice (Herrera and Agoff 2018). Moreover, abortion is illegal in most states, and only a few states (e.g., Mexico City) provide easy access to abortion clinics. The absence of functioning welfare institutions, combined with a society in which women are expected to live under the protection of men, makes motherhood one of the few ways a young woman can gain autonomy. For many, motherhood is "a source of pride and a reason for women to maintain a sense of purpose" (Fleetwood and Torres 2011: 134). Relationships with (often older) men were occasionally used as a way to replace or escape problematic family environments. However, often the same types of environments were recreated, as the abusive men bound the women to new, destructive family environments.

Doing Motherhood through Breadwinning
After the young mothers gave birth, many of them went to live with their own mothers, with other family members, or with older friends (mainly due to their age). The women relied on others for the daily care of their children, but this did not mean that they gave up their motherhood identities. The participants' way of doing motherhood differed to more mainstream norms of motherhood. They used gendered symbolic resources from local parenting practices and adopted parenting practices often associated with men.
One participant (Elsa, aged 28) used to work as a table dancer for which she earned a considerable income (at least in comparison to the other mothers in this study). She was still underage when she began working and supporting her mother. She explained how her mother played an essential role when she had her first daughter: Interviewer: Who helped you take care of her?
Elsa: My mother, yes, my mother has made a lot of mistakes as I've told you, but she has always been there for me. After all, she has always stood beside me, helped me. She took care of my daughter, so when I couldn't take care of my daughter, I told her: you know what mom, I'm going somewhere, I'll be right back.
Elsa left her daughter with her mother to make money from table dancing and prostitution. When asked what she had done with the money she earned, she replied: It was to support my mother, who provided for my daughter. She bought everything my daughter wanted, I would buy dresses, and dance shoes, because I liked that job so much, I thought, "I'm staying here." So, I went out, I earned money, took it back home and then got high with a friend at a hotel. My life has always been like that. I've never been with my daughter, to be honest.
By the time she reached the age of 10, Elsa's daughter had spent eight years with her grandmother and only two with her mother.
Another participant, (Bety, aged 28), who had a child when she was only 13 years old, shared a similar story. She noted, "I think it's terrible to say that I'm incompetent to raise her, but I need help. At least for her to be well raised, even if she's far from me, I need to know she's safe, secure and well." She sold food at a street stall to contribute to the household while her mother assumed most of the responsibility for raising her child.
These young, marginalized mothers often felt powerless and incompetent regarding the care of their children, and frequently entrusted their children to their own mothers (i.e., the children's grandmothers), other women in their households or even their friends. This is a common practice in many low-income neighborhoods in Mexico in which the daily care of children is often shared among generations of women. Caretaking includes staying at home with children, providing food and clothing and, in some cases, protecting children from sexual abuse by other family members. These norms of child nurturing and caretaking do not involve emotional education or guidance, but focus instead on the more concrete and material aspects of childrearing.
The mothers in this study claimed a motherhood identity and did motherhood by meeting the material needs of the child rather than by providing daily care and fulfilling the emotional needs of their children. Following Marks and Palkovitz's (2004) categories of manhood, when these women could not be involved mothers, they tried to be good providers. The participants felt economically responsible for their children, and often defined their mothering role in terms of that of the typical male breadwinner-they worked to provide the necessary resources for their families. In this way, they were still able to consider themselves "good mothers" despite not being responsible for the daily care of their children. When Elsa had a son a few years later, the same pattern continued: Elsa: I gave him to my best friend who took care of him for me, I gave money to both, but I only saw my children sporadically.

Interviewer: And your son?
Elsa: He was with my friend, he has always been happy with her, always. But I know someday I'm going to get out, I know confinement isn't forever, my daughter needs me, my son who's always here and there, he isn't vaccinated.
Interviewer: Hasn't your friend taken care of that?
Elsa: He's not registered, I've already told her, so I need to get out of here or have him brought here in order to register him. My friend loves him a lot, she takes care of him, she feeds him, but he doesn't go to school.
Holding on to a maternal identity by taking care of the material needs of a child is a fragile way of doing motherhood and depends on these needs actually being met. This may be difficult everywhere, but is especially so in Mexico, where poverty rates are high and female employment is usually very precarious (INEGI 2020). Elsa's portrayal of herself as a "good mother" fractured when she spoke of how her son is not officially registered, does not have his vaccinations, and does not go to school, which is the case for many children born at home in disadvantaged neighborhoods in Mexico. Without welfare institutions that secure minimum services for children, families rely heavily on the income and resources of mothers, while the caretaking of children is provided by relatives or friends. The mothers in this study were torn between the demands of their role as mothers and the temptations of street life, including drug use, crime, and multiple sexual partners. It is not uncommon in deprived neighborhoods in Latin America for women to leave their children with their families and accept work elsewhere to contribute to the larger family household. However, the reason these women did not have daily contact with their children was judged harshly by the society around them and they were usually viewed as failed mothers. One way to negotiate the stigma of this spoiled identity (Goffman 1963) was to assert their motherhood by emphasizing the financial support they provide for their children. This is not generally considered sufficient according to mainstream parenting ideals (Schmidt 2017), but it is in line with both local practices of shared mothering and masculine ideals of providing for one's family.

Learning to Become "Good Mothers" in Prison
Being incarcerated is a life-changing experience, but some of the mothering practices of the incarcerated women did not change much when they went to prison. Most were not in charge of the daily care of their children before they went to prison nor once they were inside. Arguably, the greatest change in terms of mothering thus related to the new maternal identities they developed in prison (Sufrin 2018). Additionally, many incarcerated women continued to support their children financially in prison, which they were able to do because of the large legal and illegal labor markets in Mexican prisons (Montoya 2015). The prospect of work enabled the incarcerated mothers to continue to be the family breadwinners despite being in prison. The income they earned from their jobs was used to pay expenses and allowed the inmates to send money to their families. Elsa described how she was still able to cover her child's expenses: I send 500 pesos home, 300 pesos for debt (…) But yes, my mother's very stressed out, because after I hung up with my brother, I spoke with my mother and I asked her "What's up, why do you have that tone of voice?," "Nothing," I asked again, "What's wrong?" and she says, "When are you going to send more money?" and I hung up, I hung up, I mean yeah, she has my daughter, but she's free, she's free, I'm imprisoned.
Elsa's mother did not have a job and Elsa was supporting her from prison. This served not only as a source of pride or respectability, but also provided Elsa with a way to hold on to her motherhood identity. In Mexico City women's prisons, formal work is a privilege granted by authorities. Not all female inmates are afforded this opportunity (Montoya 2015: 190), and even when they are, it is extremely difficult to support a large family on the outside with the meager income earned from working in prison. Doing motherhood by providing for the family, both outside and inside prison, is thus a fragile and vulnerable project.
Motherhood exercised in prison is subject to great criticism. The incarcerated mothers are singled out for having made two fundamental mistakes: violating penal laws and disobeying the social norms of what it means to be a woman (Montoya 2015). Consequently, they are viewed as both "bad women" and "bad mothers." Such views are also shared outside prison; however, they are intensified when the women are institutionalized, as they have to do motherhood under public scrutiny. As Ximena, who has one of her children with her in prison, explained: Ximena: Yes, it's been hard for me because there's a lot of criticism from everyone.
Ximena: All the people who know you, who see you. I once hit my son on the hands, and they told me, "He's ill-mannered because of you, that boy, no, he's terrible, he's one of the worst IJCJ&SD 10(1) 2021 in the prison." It is very difficult, everyone criticizes you: "Your rude kid!" Or, "Oh, your mother doesn't teach you to behave!" Ximena said that she "did not know how to be a mother," but that she internalized this criticism and worked on being a better mother. She stated, "I am learning … from the people who are close to me, I listen to them." She had also attended courses and programs intended to teach her to be a better mother. Other participants told similar stories. Notably, Elsa discussed how prison had changed her and given her a new perspective on life, stating: Today I'm conscious that I have two beautiful children and I wish with all my soul I could be with them right now, instead of selling chicharrones [cracklings] and roll calling (…) I'm telling you, I've learned so much here in prison. What I've learned here is knowing that there's responsibilities to fulfill, if you have children, you're not going to just throw them away, no, if you've made mistakes, you opened your legs, you partied, you fucked up, but they're your children.
The incarcerated mothers spoke of how prison had changed their view on parenting and made them realize that they were "bad mothers." They felt they were constantly learning the value of having children and taking care of them, both as a result of educational programs, self-reflection, and seeing other mothers with kids inside prison. In their own words, their time in prison had "taught them how to be mothers." However, these learned perceptions were grounded in the understanding that what they were doing, or had done, was "bad mothering." Thus, the incarcerated women had a positive view of how their time in prison had influenced their mothering practices and identities.
In deprived and marginalized neighborhoods in Mexico, as elsewhere in Latin America, motherhood is a crucial, if not dominant, symbol of feminine identity. Women are primarily perceived as mothers and wives, and maternity and conjugality are the two main pillars for the conformation and organization of their lives (Lagarde 2014;Montoya 2015: 193-4). This is also the case in prison. Paradoxically, the same motherhood that represents a burden and important source of stigmatization in prison, can also become a source of redemption. For example, Irma, explained how she spent time with her daughter when she was visiting her in prison: I try to talk, to play with her, to give her quality time. Not all day but the moments I'm with her, I dedicate myself solely to her. I forget about TV, my knitting, my chores, everything but her. I dedicate those 20 minutes or whatever it is only to her. If something falls or the food gets burnt, I don't care, I'm with my daughter, 15 or 20 minutes alone.
When asked to reflect upon what her child meant to her, Irma replied: A lot of things, opportunities I couldn't give myself before. Knowing I'm not a bad person, knowing I can give more. I've been wrong, yes, I've been wrong in many things with my daughters previously and from those things I'm learning, I've fallen and now I'm getting up.
Prison reminds women of their shortcomings as mothers, but doing motherhood also enables them to atone for their guilt (Montoya 2015:188). Doing motherhood while following the ideals and standards they have learned in prison enables the women to recreate themselves and negotiate the stigma of their spoiled identities (Goffman 1963). Being a mother represents an alternative identity to the stigmatized identities of being a prisoner, drug user, or criminal. Motherhood may, in this way, be a symbolic resource in a "normalcy project" (Frederick 2017) both inside and outside prison. Due to their imprisonment however, normative motherhood is frequently a dream or future project that they claim will be realized as soon as they are released from prison. Miller (2006: 15) emphasized that "the practiced, recognizable gendered and embodied self, which makes up our identity, is challenged by the experiences of first-time motherhood: over time, a new social self as mother has to be learned." The present study showed that the shaping of motherhood identities is an ongoing project related not only to first-time motherhood, but also to other changes in a mother's life, such as imprisonment. Early pregnancies and leaving their children in the care of others represent a continuity of the women's experiences of family relationships (first having, then becoming, absentee parents). The penitentiary institution may arguably become the parental figures they never had or never were. This changes their fundamental motherhood identities, if not their mothering practices.

Discussion
The intersection of various forms of inequality, age, gender, and socioeconomic condition (Crenshaw 1991) have placed these women and their children in a vulnerable position. As adolescents, they often lacked control and support mechanisms. Many got involved in early marital unions or cohabitations in an attempt to either escape their often tragic circumstances, or to fulfill their ideal of a family they never had. Early marital unions and cohabitation, and early motherhood thus became a way to escape or to move from childhood into adulthood (Carbone-Lopez and Miller 2012). Motherhood became a resource for gaining autonomy, but often resulted in the tragic paradox of merely moving from one destructive family environment to another.
When the young women had children, other people (most often the women's mothers, but also their mothers-in-law, or other family members or friends) often assumed the traditional mothering responsibilities. The practice of leaving children with extended family is generally accepted by mothers in disadvantaged areas in Latin America, as it allows the women to migrate to obtain work. The practice is also supported by working-class parenting values, under which providing for the material needs of children is a primary goal. The women in this study thus did motherhood, and countered the stigma of being "bad mothers" by assuming and emphasizing their role as family providers. However, their "immoral" lives and work often made this a difficult "normalcy project" (Frederick 2017). In a highly gendered and traditional society, it is difficult for women to gain respect through practices associated with the paternal role and when engaged in work that is largely viewed as immoral.
Most of the mothers in this study did not have daily contact with their children who lived outside prison. Consequently, the changes in their relationship with their children following their imprisonment mostly related to their identities as mothers. The women often described how prison had changed their views on parenting. The mothers tended to hold on to their role as providers, but expanded this by emphasizing the emotional care provided and expressing a desire to spend time with their children. In prison, they were exposed to new norms for doing motherhood. As Sufrin (2018: 57) states "incarceration disrupts reproduction and motherhood, while simultaneously promoting an idealized, normative motherhood." These more middle-class norms of motherhood can serve as a source of respectability and generate renewed self-respect. At the same time, these norms provided another way of doing motherhood that is doomed to failure, as most of the incarcerated women cannot follow up with their children on a daily basis. To offset this, many of the women accepted the new ideals of doing motherhood, and adopted middle-class norms of motherhood to be accepted as "good mothers" within the prison system, but postponed their realization as something that could and should be done after their release from prison.
The motherhood discourse that the women learned in prison was problematic in a number of ways. Notably, it neglected many of the mothering experiences the women had had outside prison, and was fundamentally based on a paternalistic view of incarcerated women (Sufrin 2018). The penitentiary system tends to infantilize incarcerated mothers and views them as immature and irresponsible women who need to be instructed to become "decent mothers" (Montoya 2017: 111). Consequently, the motherhood ideals of imprisoned women change. Rather than viewing themselves as providers who share the burdens of motherhood with family members across generations as they did outside prison, they adopt IJCJ&SD 10(1) 2021 a more individualistic and middle-class view of maternal identity based on emotional presence, guidance, and responsibility for the daily care of their children. Arguably, these motherhood ideals are difficult (sometimes impossible) to achieve for imprisoned women, and problematic for children who may benefit from more communal mothering practices.

Conclusion
A motherhood identity is not something people can "choose" in the same way as fathers can choose a fatherhood identity (Rich 1986). It is imposed upon many women by a patriarchal society. This is particularly prevalent in Latin American societies that are characterized by an ideology of familism and marianismo identities. However, this does not mean that there is only one way to do marginalized motherhood. We have shown how different motherhood practices and identities influence, and compete to define, the lives of these marginalized mothers. One of the main contributions of the present study is an insight that doing marginalized motherhood differs widely depending upon which life phase the mothers are in and the institutional and societal context of their mothering. Our study also reveals the importance of including situated experiences of culture (Machado et al. 2010), and the importance of studying marginalized mothering beyond gendered Northern foundations (Walklate 2018) or beyond the Global North (Carrington et al. 2016).
The importance of family (which in many cases replaces welfare institutions), patrivirilocal residency, and the sharing of childcare among female members across generations are key to understand mothering among marginalized Mexican mothers. When they are imprisoned, these mothering practices and identities are challenged by educational programs based on more middle-class values, which have often been inspired by similar programs in the USA and Europe. These programs are founded on norms and ideals that assume that children are raised in a dyadic relationship (between a woman and a child), and that care is the sole responsibility of the mother. The emphasis on individual responsibility does not reflect the incarcerated women's lived experiences of shared mothering in an extended family. This creates tensions and, in many cases, introduces new motherhood ideals that have little connection with either local maternal practices and life worlds, or mothers life in prison. Having a better understanding of marginalized mothering practices could help bridge the gap between educational programs and the lived realities of imprisoned mothers in Latin America.