Abstract
In this article, we examine the operation of shame in the alcohol use habits of pregnant women and the responses of their families and associated institutions. Using a narrative–discursive approach, we interviewed 13 women, living in a low-resource setting in South Africa, who had consumed alcohol while pregnant. Narratives showed how both the act of drinking and “inappropriately” timed pregnancy (early and out of wedlock) were judged to be unacceptable. Women who engaged in these activities were positioned as bad mothers or promiscuous. Their actions were seen as resulting in the suffering of others—the future child, the family, and even the community. These narratives were underpinned by cultural and religious discourses. Women managed the shame accruing to them through avoidance and concealment; families instructed women to self-exclude or distanced themselves from the women’s behavior; and institutions subtly or overtly excluded women. The shaming of these women, and the mechanisms by which such shame was managed, did little to decrease drinking or to increase maternal health and welfare. Overall, this article demonstrates how the shame of drinking alcohol during pregnancy produces avoidance behavior, concealment, and exclusion, which are not constructive in terms of maternal health and well-being. The implications for a feminist narrative approach to drinking during pregnancy are outlined: moving beyond a focus on individual behavior change to locating personal stories within the meta-narratives and social discourses that shape pregnant women’s lives.
Drinking alcohol during pregnancy can result in children experiencing lifelong developmental delays, disabilities, and physical anomalies that are collectively referred to as fetal alcohol spectrum disorders (FASDs). As a result, the policy in many countries, including South Africa where this study was conducted, is that alcohol should be avoided entirely during pregnancy. In some instances, including in South Africa, authorities recommend that if women are not able to stop, reduction in consumption should be encouraged (O’Leary et al., 2007).
Research conducted on alcohol use during pregnancy in South Africa has identified a number of risk factors. These include sociocultural factors (e.g., limited access to social resources, physical access to alcohol, the social tolerance of drinking), behavioral factors (e.g., alcohol initiation at an early age, drug abuse, having multiple sexual partners, and smoking), lower educational attainment, familial issues (e.g., parents who consume alcohol excessively, cohabitating with a partner or spouse who is a heavy drinker, and being in a violent relationship), and pregnancy-related factors (e.g., high gravity or parity or an unwanted pregnancy; Brown et al., 2005; Desmond et al., 2012; May et al., 2007, 2013; Morojele et al., 2010; Viljoen et al., 2005).
Knowledge of these risk factors is important in terms of designing interventions to reduce the harmful effects of alcohol consumption during pregnancy. Researchers are, however, increasingly indicating that understandings of the experiences of women who drink alcohol during pregnancy are also important. For example, findings from qualitative studies conducted in Canada (Kruk & Banga, 2011), the United States (Zabotka et al., 2017) and South Africa (Eaton et al., 2014; Watt et al., 2014) highlight the role played by stress and trauma in substance use during pregnancy, with pregnant women indicating that they turn to alcohol and/or drugs in an attempt to cope with past and present stressors in their lives.
In this article, we extend this body of work by reporting on narratives of shame in the talk of South African women in a low-resource setting about their drinking of alcohol during pregnancy. Using a narrative–discursive approach, we show how these stories of shame are grounded in cultural and religious discourses and how they limit the ability of women to access the supportive help that they may need. We conclude with recommendations for interventions that are grounded in feminist narrative therapy principles: recognizing how alcohol use during pregnancy is embedded in power; surfacing counter discourses, narratives, and social practices; and assisting women in storying their lives in positive ways.
Alcohol Use in South Africa
South Africa has high rates of alcohol consumption. According to Vellios and van Walbeek (2018), one in three South Africans report drinking, and one in seven report binge drinking on an average day of drinking. The history of alcohol use in South Africa is closely linked to the history of apartheid (Rataemane & Rataemane, 2006). During this time, access to alcohol by the majority of black people was limited, and this led to the proliferation of homebrews (especially of sorghum beer) sold in unregistered shebeens also referred to as taverns. This resulted in excessive alcohol use despite shebeens being illegal. In addition to the history of apartheid, excessive alcohol use in South Africa can be traced back to the early years of Dutch settlement. Dutch farmers coming into the Cape implemented the dop 1 system in which farmworkers were paid in alcohol in addition to their wages, a practice that continued into the apartheid era (Olivier et al., 2016). While this system no longer exists, alcohol consumption in this region remains pervasive.
As a result of the above historical influences, binge/heavy drinking has become a common practice in South Africa, where people refrain from drinking alcohol during the week and then drink excessively on weekends (Friday–Sunday) and at parties. The practice of binge drinking is further exacerbated by the continued existence of multiple shebeens where alcohol is easily accessible and affordable. Evans (2015) further argues that the two historical influences have resulted in a “drinking culture” within the South African context and, as such, excessive alcohol use has become a norm. That is, this drinking culture promotes the use of alcohol during leisure time, as celebratory, cultural, and social events are organized around imbibing alcohol. As a result of this history and current circumstances, communal and individual identities have come to be shaped by alcohol use.
Studies on FASDs in three of the nine provinces in South Africa (Western Cape, Gauteng, and Northern Cape) show high rates of these disorders, ranging from 29 to 290 per 1,000 live births (Olivier et al., 2016). Research conducted in Cape Town with women attending antenatal clinics revealed that 20.2% drank alcohol, with most of these drinking at risky levels (Vythilingum et al., 2012). A similar figure (18%) was found in a later study conducted in an obstetric unit in a low-resource area of Cape Town (Onah et al., 2016). These figures are, however, higher than a household survey conducted in a similar area (8.3%; Davis et al., 2017) and a survey conducted in Mpumalanga (6.6%; Louw et al., 2011).
Given resource constraints in South Africa, clinical interventions aimed at stemming alcohol abuse tend to be brief and include motivational interviewing, in which clients are encouraged to take steps toward the reduction of drinking. Such an approach has shown promise in some studies but not in others. As argued by Hunting and Browne (2012), it is important for interventions to consider cultural, social, and gendered dynamics underpinning alcohol-related harms. Hunting and Browne’s (2012) argument is extended in this study by considering shame as an important cultural, social, and gendered dynamic, which underpins women’s alcohol use during pregnancy in South Africa.
Shame
Shame has received increasing attention in feminist literature. As indicated by Shefer and Munt (2019) in the editorial of a special issue on shame, Shame is closely entangled with gender subjectification and normative gender binarisms, which are raced, classed and enmeshed with other forms of intersectionalities. However, shame also holds multiple possibilities for social justice efforts when deployed for resistance, activism and critical pedagogies. (p. 146)
Shame is a powerful social mechanism that can be deployed in multiple ways. Parker and Pausé (2019), for example, show how fat pregnant people in Aotearoa/New Zealand manage their shamed maternal identities through a range of strategies. The authors argue that while fat-shaming is productive in constituting self-governed maternal subjects, it is not constructive in terms of improving maternal and infant health. Instead, it negatively impacted on participants’ experiences of their pregnancies, emerging maternal selves, and newly born children.
Method
Using a narrative–discursive approach (Taylor & Littleton, 2006), we analyzed women’s stories of the personal and interpersonal circumstances under which their consumption of alcohol took place. Our aim was to elucidate both the narratives told and the social discourses drawn upon in narrating these stories. In this article, we home in on stories of shame (viz., as noted above, being judged by others or the self as unworthy or wrong, as acting in unacceptable ways, or as being complicit in the suffering of others) and the discursive resources underpinning these narratives.
Convenience and purposive sampling were used to select women who were 18 years and older (for reasons related to the ability to give consent) and who drank during a previous pregnancy. Our sample focused on alcohol consumption meeting the criteria of moderate or harmful drinking based on the Alcohol Use Disorders Identification Test for Consumption (AUDIT-C) questionnaire. The AUDIT-C 2 is a three-item screening tool used to help identify persons who are (or were) hazardous drinkers, abusing, and/or dependent on alcohol (Babor et al., 2001). This was administered after the first subsession of the narrative interview (see later discussion).
Participants were recruited from a non-governmental organization (NGO) in the Eastern Cape that works in communities to reduce alcohol use during pregnancy. Prior to approaching gatekeepers from the NGO, ethical clearance was obtained from the Rhodes University Ethical Standards Committee. Permission to conduct the research at the NGO was granted by the organization’s director and social worker after they reviewed the proposed research. We used a two-phase approach to recruitment. Mentors from the NGO gave all potential participants a research information card that briefly outlined what the research was about and requested them to indicate that they gave the mentor permission to provide their names and numbers to the researchers. Upon receiving a list of names and contact details from the mentors, we conducted an information session with potential participants (14 people). Those who agreed to participate in this study (13 participants) then signed a consent form. A participant demographic form was compiled and used to obtain demographic data from the women.
All 13 women identified as “black” 3 and were unemployed at the time of interviews. Ten had a high school (Grades 8–11) level of education, two Grade 12, and one primary school (Grade 7). With regard to AUDIT-C scores, three of the women engaged in lower risky drinking during their pregnancies, while the remaining 10 women drank at risky or harmful levels.
The method of data collection used for this study was the narrative interview as delineated in the work of Wengraf (2001, 2004). The narrative interview, in contrast to structured and semistructured interviews, aims to limit the researcher’s influence (although the influence of researchers in any qualitative study can never be completely removed). Simply put, narrative interviews, as argued by Anderson and Kirkpatrick (2016), “provide an opportunity to prioritize the storyteller’s perspective rather than imposing a more specific agenda” (pp. 631–632).
In the first or main interview, two subsessions took place. In the first subsession, the interviewer (first author) posed one narrative question or what Wengraf (2001) calls a single question aimed at inducing narrative (SQUIN). For this study, the SQUIN did not make reference to alcohol use so as to help establish rapport with the participants, avoid upsetting them and making them feel embarrassed or judged for drinking during pregnancy. The SQUIN used for this study was as follows: “Please tell me the story of your pregnancy from before you were pregnant through to when your child was born including the events and experiences that were important to you during this time.” Before the SQUIN was posed, the interviewees were told that (1) the coresearcher (third author) would take notes that the two researchers would discuss, (2) in telling their stories, they would not be interrupted, (3) there were no right or wrong answers to the question, and (4) they could take all the time they needed to answer the question and could start wherever they liked (Wengraf, 2001, 2004).
In the second subsession of the main interview, the narrative follow-up, the interviewee was asked to remain in the room, and the two researchers, in a separate room, spent 15 minutes composing questions to ask the interviewee based on the notes taken by the coresearcher. In this subsession, three guidelines were adhered to (Wengraf, 2001, 2004). First, the interviewer asked the interviewee narrative-pointed questions or rather, questions related to the story that was told. Second, the questions asked were based on the topics raised by the interviewee in the first subsession. Finally, in asking about the topics raised, the interviewer did so in the order in which they were raised, one at a time, using the exact words of the interviewee.
The third subsession took place after the data from the first interview (the first and second subsessions) had been transcribed (verbatim) and read over (Wengraf, 2001). The process took approximately 2 weeks. In this subsession, the interviewee was reinterviewed, and the interviewer was given the opportunity to ask further and/or additional questions (narrative, nonnarrative, or other kinds of questions) that emerged from what was said or not said in the first interview (Wengraf, 2001, 2004).
All interviews were conducted in a small room in a community hall close to where the participants lived. Interviews were conducted predominantly in isiXhosa as all of the participants were isiXhosa speakers, although all the women combined isiXhosa with some English in their interviews. The interviews were audio-recorded with the permission of the participants. Being first language isiXhosa speakers and identifying as “black” women meant that the first and third authors were similar to the women who participated in this study. These aspects of our identity provided us with insider status, thus easily facilitating the development of rapport between the authors and the women. This status also aided in the level of openness of each of the women when it came to sharing their stories.
The first author translated and transcribed the isiXhosa audio recordings directly into English, leaving phrases and words about which she was unsure in isiXhosa. Thereafter, the transcripts were checked by the third author to ensure equivalence and to identify differences in interpretations. Where differences in interpretation occurred, the two authors discussed the differences and decided on the most plausible translations. This process of checking the transcriptions was also adopted for checking the translations of the research information card, study information sheet, and consent forms. Because the data were translated, transcription conventions were kept to a minimum: round brackets with number inserted, for example, (2) indicates pauses in speech with the number of seconds in round brackets; round brackets with full stop (.) indicates pauses in speech that last less than a second; double colon :: indicates an extended sound in the speech; and underlining indicates emphasis in speech.
We used a narrative–discursive approach to analyze the data. As a synthetic approach, various methods of analysis are drawn upon in a narrative–discursive analysis, namely narrative analysis and discourse analysis (Taylor & Littleton, 2006). We coded the data by searching for microstories of sequence and consequence (e.g., if a woman drinks alcohol during pregnancy, she is a bad mother and deserves to be disciplined). Thereafter, we identified discursive resources within individual narratives and across narratives. Discourses are interrelated systems of statements that cohere around common meanings and values; they construct objects and enable particular subject positionings.
Findings
We have organized our findings according to two key themes: narratives of shame and narratives of managing shame. Narratives of shame centered around three behaviors or identities: drinking alcohol during pregnancy, conceiving at an early age, and being a “bad mother.” These narratives drew from cultural, religious, and “good mothering” discourses. Narratives of managing shame spoke to avoidance, concealment, and exclusion. Overall, these narratives evidence how our participants were led to conceal their alcohol use and how, in turn, this impacted their health and that of their infants. In each theme discussed below, we provide key examples from the data.
Theme 1: Narratives of Shame
Narratives of shame were clearly evidenced in participants’ talk. In the exemplar Extracts 1–20 presented below, words such as “embarrassing,” “wrong,” “be in trouble,” “not right,” “not good,” and “ashamed” are used in the stories told by participants. As indicated above, shame refers to being judged by others or the self as unworthy or wrong (identity ascription), as acting in unacceptable ways (behavior ascription), or as being complicit in the suffering of others (relationship ascription). In these data, the shameful identity ascription was bad mother, shameful behavior ascription was drinking during pregnancy or conceiving early, and shameful relationship ascription was bringing the family into disrepute. These are discussed below.
Participants spoke clearly about the shame of drinking during pregnancy, which they knew was wrong. This shame was seen by participants as connected to understandings of what it means to be a good mother, with the social label of bad mother being something to be avoided.
Extract 1
In the Xhosa culture I did a lot of things that are embarrassing (.) The biggest one is that when I got pregnant I drank.
Extract 2
Drinking when you are pregnant in our culture is wrong. You can be called a bad mother.
Extract 3
I didn’t want them to talk and call me a woman who does not care about her child.
Pretty and Lucy clearly indicate how drinking during pregnancy is culturally shameful (wrong). This enables the ascription of the identity of bad mother. Morongwa talks about attempting to avoid this shameful subject positioning. These narratives are in line with other research that shows pregnant women who drink or have drunk during their pregnancies are viewed as not adhering to cultural norms and rules related to their gendered reproductive role as mothers (Greaves & Poole, 2005; Reid et al., 2008).
Shame accrued to the nature of the pregnancy itself. “Inappropriately” timed and out-of-wedlock pregnancies were depicted as taboo. Women in such circumstances were positioned as bad mothers from the outset and as bringing shame to the family. Nina, Pretty, and Lucy share their experiences in the extracts below, including experiences of violence:
Extract 4
There is a right time for having a child in our culture [the Xhosa culture] (.) first I must get an education (.) get::: get married (.) then have sex (.) and get pregnant then have a child. If this does not happen in this order like with me (.) you will be in trouble.
Extract 5
I got pregnant when I was still young and still in school (.) my father was a strict Xhosa man and he was
Extract 6
Being pregnant when you are young in our culture is not right (.) many people in the community and even some of my family members said that I was a loose girl [participant crying] (10) and this happens to many girls that get pregnant when they are young (.) they did not even know how I got pregnant (.) I was raped by my boyfriend sisi [sister] [participant crying] (5) and my mother did not believe me (.) she said I wanted to have sex and this still hurts me.
Participants in the above extracts describe a cultural injunction against early, out-of-wedlock childbearing. Nina lays out in clear terms the acceptable order to life stages, indicating that if these are violated, there will be “trouble.” Trouble for Pretty came in the form of a beating and being subjected to silence. For Lucy, it consisted of being called derogatory names and not being believed about her sexual assault. In Extract 7, Pretty also shows how shame accrues not only to the young woman but also to her family. The rupture of acceptable conjugal relations for childbearing, she states, led to the father’s standing in the community being diminished, and his capacity to raise a child who would not rupture cultural sanctions questioned. Similar findings emerged in the study conducted by James et al. (2012) with amaXhosa families. Teenagers’ significant others spoke about how the young girls failed to meet cultural expectations centered on the timing of the pregnancy. Some pregnant teenagers spoke about how they were resented by their families as the pregnancy brought shame and embarrassment to the family’s name.
Shame, thus, operates in a systemic way, with the consequences of culturally unacceptable practices—in this case, the unacceptable timing of pregnancy—being visited not only on individuals but also their families.
In narrating experiences of shame, the women we interviewed frequently spoke about how their actions ruptured cultural mores or norms. A feature of this talk is the use of the phrase “in our culture.” Participants, thus, drew on a “cultural” discourse, in which culture is referred to as a “real thing” that governs what is and is not acceptable.
Extract 7
In our culture it has always been said that drinking during pregnancy is not right because it harms the baby. I used to drink and I knew it was wrong and I knew that the elderly people were going to want to discipline me but I did it [I drank] anyway.
Extract 8
Like I said (.) in our culture it has always been said that (.) a pregnant woman
Extract 9
In the Xhosa culture I did a lot of things that are embarrassing…and old people in the community would swear and (.) shout at me when they saw me drunk or drinking (.) they said it is not right to do this [drink when you are pregnant].
The above extracts show the homogenization of culture in such a way that it becomes objectified. “Xhosa culture” is seen as a systematically harmonized whole comprising of a shared and stable system of beliefs, knowledge, sets of practices, and values. The injunction against drinking alcohol during pregnancy is seen by participants as firmly embedded in the culture to which they belong. This embeddedness makes their breach of this injunction that much more shameful.
Elders are seen in these extracts as the custodians of culture and thus as enforcing cultural norms. They know what is right and are tasked with disciplining those who rupture cultural imperatives (the imperative being, in this case, to not drink alcohol during pregnancy). In the narrative follow-up, Khethiwe mentioned that she was indeed disciplined for her alcohol consumption: She was called to a meeting with a group of women from the community organized by her mother, where she was told what was expected of her during her pregnancy. Similarly, Pretty speaks about being sworn and shouted at by elders.
Shame accrues not only to the individual and family but also to the entire community. For example, in Extract 8, Rosey outlines how the culturally embedded shame associated with drinking during pregnancy is communal. As such, Rosey sees herself as a representative of her people. Thus, when she engages in actions that in her culture are constructed as shameful, this not only threatens her own social position but also that of her people.
The connection between cultural shaming and religion also emerged in the data. Specifically, cultural shaming was seen by some participants as premised on religious discourse. This finding must be viewed in the context of Christianity being the dominant religion in South Africa with just under 80% of the population identifying as Christian.
Extract 10
Our culture says that a pregnant woman should not drink. Even the Bible says it [drinking during pregnancy] is not right.
The implication of Pearl’s statement is that women who drink during their pregnancies are sinners. Pearl’s close association of “culture” and Bible teachings is in line with the observation that in many African societies, “customary practices have been incorporated into religion, and ultimately have come to be believed by their practitioners to be demanded by their adopted gods, whoever they may be” (Okome, 2003, p. 71).
In the extract below, Nonny, who shared that she lost her first child as a result of her alcohol consumption, shows how the religious discourse operates:
Extract 11
I come from a family that goes to church a lot. At church we :: (.) the Bible says that a baby is a gift from God and the Ten Commandments (.) one of them says
By making reference to the Bible, specifically the fact that a potential child is a “gift from God” and that one of the Ten Commandments is “thou shalt not kill,” Nonny positions herself as directly culpable for the potential child’s death. As such, it is she who must take on the mantel of guilt and the associated shame for engaging in unacceptable practices. This position is, however, emotionally difficult to sustain (as noted by the long silences in her narrative and her crying), and she engages in justificatory labor to alleviate some of the blame and responsibility: “I knew it [the drinking] was wrong but I had problems and I did not know of another way to solve them.” In doing so, Nonny emphasizes that she did not kill her potential child on purpose.
Theme 2: Managing Shaming—Narratives of Avoidance, Concealment, and Exclusion
As indicated in the previous section, participants were very aware of the cultural injunctions they were rupturing, and the potential shaming that they faced. Hence, they sought to manage interactions in such a way as to reduce the chances of being shamed. Concealment or avoidance was, in some instances, self-imposed; in others, women were instructed to do so by family; in yet others, formal or informal exclusion measures were adopted against them by institutions. In addition to hiding their consumption habits, some of the women described hiding their pregnancies.
In the extracts below, participants narrate stories of self-managed avoidance and concealment as a way of avoiding shame.
Extract 12
Firstly sisi (.) I want to say that when I was pregnant I
Extract 13
I bought myself a bottle [of brandy] from next door (.) next door is a she[been] a tavern (.) I got myself my bottle and my carry-pack [six pack of ciders] and I put it somewhere and I knew that no one was going to find them because I used to hide them in shoe boxes in my cupboard (.) when I got back from work he [my partner] was not around because he worked night shifts (.) I would go to my spot [the cupboard] get my cider or brandy or both and I would drink until I sleep so no one knew I was drinking when I was pregnant because no one saw me (.) I drank when I was on my own (.) I didn’t drink next door [at the tavern] because people in the community who know me drink there and I didn’t want them to talk and call me a woman who does not care about her child.
Extract 14
Okay (.) so you say you wore a windbreaker [to school].
Yes.
Can I maybe ask why?
I felt
In Extracts 12 and 13, Dineo and Morongwa narrate their efforts to avoid being shamed through their positioning as an uncaring or bad mother. Both women do this by keeping their drinking a secret—Dineo hides her drinking from her mother, while Morongwa does so from her partner. They drink alone, in private, when no one is around instead of at the shebeen. Unlike Morongwa, Dineo resists the interactive subject position of bad mother when she says many women who are pregnant and who drink in the community are called bad mothers and I don’t want that. Concealment took the form not only of avoidance of particular spaces but also of disguise of the pregnancy itself. In Extract 14, Cindy narrates how she managed to keep shameful pregnancy a secret from her teachers and schoolmates through concealing it.
Additionally, participants indicated that avoidance and concealment were implemented not only by the women themselves but also by their families too. Extract 15 I was told to leave school when I got pregnant (.) my father said that I must stop going to school because I had already embarrassed him in the community and now I was going to embarrass him at school.
Here, Pretty talks about being forced to leave school by her father, who wished to avoid further shaming within the school environment. Pretty’s absence from the school ensures that, at least in this space, the shame of an “untimely” pregnancy is overcome.
A few participants also highlighted informal exclusionary practices associated with their religious institutions:
Extract 16
After what happened to my baby because of my drinking (4) I am still scared of going to church because I am worried about what people will say about me…some people do not speak to me at all because of what happened and I think (.) I think it’s (.) it’s because they know what I did.
Extract 17
When I didn’t go to church on a Sunday the next time I go everyone would be questioning me and asking if I had been drinking. Sometimes I feel like I must leave the community and go live somewhere else and start a new church where no one will question me.
Extract 18
I used to go to church during the time when I was pregnant but then I stopped because at church (.) at the church where I go (.) when you are a woman (.) you cannot be a woman who has a child while you are not married so I just stopped [going to church] after that [I got pregnant] and I haven’t gone back yet.
The women in the above extracts highlight how they decided to avoid their current church or find a new one as a result of the shaming (negative statements, invasive questions, and exclusion from conversation) they experience or anticipate they will experience. In each case, they felt ostracized by a community meant to provide support. In Extract 18, Pearl depicts the church as laying out certain “rules” that have to be followed by its members: In this case that a churchgoing woman cannot be pregnant outside of marriage. The use of the word “cannot” shows how Pearl experiences this injunction as set in stone. For Pearl, the best way of dealing with her rupture of this injunction is to avoid the church space altogether.
As indicated above, shame operated in complex ways, with the women and their families being implicated in shameful acts. In some cases, family members’ and women’s management of the shame conflicted, as seen in Dineo’s micro-narrative below, which includes an experience of violence:
Extract 19
When my mother came back from work one day she said to me that she heard from one of the people in the community that I went to buy alcohol at the shebeen and I was drinking at the house with friends. She asked me if this was true and she searched the whole house and found the alcohol where I was hiding it. She beat me up so much.
Dineo talks about how her shame is extended to her mother who responded by beating her publicly. Dineo manages her identity as a shameful pregnant drinker through hiding the alcohol and drinking with friends. Her mother, having been alerted to the drinking, takes a different tack. In beating Dineo publicly, we argue that the mother potentially alleviates any shaming accruing to herself, as she is visibly opposing Dineo drinking during pregnancy. This allows her to recuperate the valued identity of a good mother, which was potentially destroyed through people in the community knowing that Dineo was drinking and who may blame her (the mother) for this drinking. Although the South African Constitutional Court banned parents from using corporal punishment against children in 2019, the practice is common (Mathews & Benvenuti, 2014) and often condoned. Thus, Dineo highlights how people laughed and remembered the beating and how this compounded her (Dineo’s) shame. But, no shame accrued to the mother for the beating. This conflict in shame management (concealment vs. public beating) disrupted family relationships, according to Dineo, as she and her mother “were not on good terms” as a result.
Discussion
Our research shows that the women we interviewed in the Eastern Cape community understood the negative implications of drinking alcohol during pregnancy. The injunction against drinking, they indicated, is deeply embedded in Xhosa culture. Consuming alcohol while pregnant fractured, in their renditions, cultural and religious norms, resulting in the shaming of those who persist with drinking while pregnant.
The shaming of those who drink during pregnancy was closely associated with understandings of good mothering, bolstered by cultural and religious discourses. A good mother was understood as somebody who conceives at the appropriate time (when older and married) and whose primary function is to care for the children. The shame for failing to meet the criteria for good mothering (through inappropriately timed pregnancies and/or drinking alcohol during pregnancy) accrued not only to the woman in question but also to her family and potentially her community.
Both the pregnant women and their families spoke about engaging in actions aimed at avoiding the ascription of the shameful identity of bad mother to them. However, this shame was productive but not constructive (Parker & Pausé, 2019). The shaming-produced behavior focused on diminishing the shame, but these practices were not necessarily conducive to women’s health or to the health of their fetus. Instead of stopping drinking, women hid their drinking, drank in safe places (e.g., with friends), concealed their pregnancies, and ceased to attend places that could potentially assist (school and church). Family members likewise insisted on women avoiding places where shame might accrue (e.g., school) and engaged in overt behavior to ensure that the shame accrued only to the woman and not to themselves. None of these actions were, in our assessment, conducive to women receiving the help that they may need to reduce their consumption of alcohol during pregnancy.
The association of drinking alcohol during pregnancy with bad mothering is not unique to this context. Pregnant women and/or mothers who have alcohol or any other substance use problems are often labeled as “bad” and/or “unfit” mothers (Gueta & Addad, 2013; Reid et al., 2008). The internalization of the label of bad mother may lead to women feeling overwhelmed by guilt and shame (Lupton, 2011), which may serve as a barrier to women’s accessing the services they need (Greaves & Poole, 2005; Green et al., 2015). However, this label fails to take into account the contextual and structural factors that contribute to “bad mothering,” framing the behavior as an individual choice without locating it within the broader social context. These contexts are, we assert, essential in addressing the consumption of alcohol during pregnancy. Below, we outline a feminist narrative approach that does just that.
Implications for Interventions
Interventions aimed at preventing and/or reducing FASDs generally take the form of clinical interventions or universal interventions. Clinical interventions are targeted at women who drink alcohol, are dependent on alcohol, or already have a child with a FASD, with the aim of helping them to achieve abstinence from alcohol during pregnancy. Universal interventions are the most common type of intervention and are aimed at entire populations regardless of individual risk status to raise awareness about the risks of alcohol use during pregnancy at a population level (Barry et al., 2009; Hankin, 2002).
The difficulty is that many of these interventions focus on individual behavior change. Hunting and Browne (2012), for example, show how FASD interventions with Aboriginal women in Canada obscure the structural, social, and historical processes (such as urbanization, racialization, and colonialism) that give rise to women’s health and social inequities. Likewise, the NGO with whom we partnered mostly focused on women ceasing drinking with little attention being paid to the gendered and social power relations within which the women were located (we provided this feedback to the NGO as part of our engagement with them). In these situations, it is easy for objective health messages to be transformed into moral judgments of the women, as found by Snertingdal (2013) regarding prenatal alcohol interventions in Norway.
Given these limitations, what are the implications of our findings for feminist-based clinical and universal interventions in relation to drinking during pregnancy? We draw here on the principles of feminist narrative therapy, as outlined by Lee (1997), to make suggestions for such interventions. Narrative therapy is an approach that involves the dual task of helping people work with their personal (contradictory and multiple) stories and to locate these within the meta-narratives and social discourses that shape their lives (or in Lee’s (1997) words, “the socially and historically specific cultural scripts through which our subjectivities have been formed” (p. 4)). Lee outlines three concepts underlying feminist narrative therapy: The first involves a keen awareness of the misogynous meta-narratives of society that mold the storying process. Second is an understanding of narrative reciprocity whereby notions of the self-create themselves through stories that are reciprocally interwoven with the stories of others…Third is the issue of practitioner reflexivity which implies that therapists pay attention to their own self-narratives and are in touch with their own meaning constructions…[in relation to] race, class, and or sexual orientation, etc. (Lee, 1997, p. 12)
An example of an alternative narrative is provided below. Nonny surfaces an alternative narrative based on a religious discourse: A narrative of redemption from the shame associated with drinking during pregnancy. Despite the fact that Nonny constructed drinking during pregnancy as an act of killing, she highlighted the redemption made possible by a forgiving God. She said:
Extract 20
Even if I killed my child I know that God was punishing me for drinking at that time [when I was pregnant] (.) God has forgiven me now because I prayed (.) I continue to pray (.) I made a mistake but He forgives those who sin.
In this instance, the same God who can punish is also seen to have mercy. Nonny acknowledges the fact that she did something wrong, but she recognizes her mistake and repents through prayer.
The potential of a narrative approach to shift ways of being was surfaced in our study. Our research was not intended as an intervention. Nevertheless, the women were presented with a chance to have their stories listened to and witnessed in a nonjudgmental environment. Women were unanimously positive about the interviews. For example, Pearl reflected:
Extract 21
I felt alright bantasekhaya [my people],…it’s a bit better ever since [I took part in this research]. I was able to speak about my problem[s], you understand? At least I am now a person who is a bit better.…It goes like this and like this, but at least I do get some sleep. It’s a bit better because I could wake up at around twelve and just think and just think and just think about things.
Pearl’s reflection shows the potential of a narrative approach in shifting problem-saturated stories. The “narrative reciprocity” referred to by Lee (1997) comes into play here. In speaking to the researchers about her problems, Pearl’s personal narrative shifts somewhat to her feeling “a bit better.” By comparison, the participants’ narratives about families and partners tended to be problem-saturated. Apart from narratives of punishment or exclusion because of shame, stories of abandonment, lack of support, and paternity denial were told.
Shifting problem-saturated narratives requires deep listening and questioning that follows the logic of the client’s story while at the same time searching for a glimpse of alternative constructive stories that can be extended or elaborated upon. This is illustrated below. Again, this is not an intervention but rather an interaction where the interviewer is probing for better understanding; nevertheless, the narrative principle is in evidence here: Extract 22
When you discovered you were pregnant nhe [right] how did you feel?
I was shocked and felt scared.
Please tell us more about your fear/being scared. What were you scared of or fearful of maybe?
I was scared about what they were going to say at home because I was still in school at that time.
So you say you were scared about what they were going to say at home. So after you shared the news at home that you were pregnant what did the people say? How did the people at home react?
But my mother when she heard from the neighbours she didn’t have a problem and she didn’t show [that she had a problem] she didn’t swear at me and everything.
In the above extract, Nonny talks about anticipating that her mother (and possibly other people in her family) would react to her news negatively as becoming pregnant while still in school is regarded as inappropriate. However, this was not the case. Nonny’s family, particularly her mother, supported her pregnancy rather than problematizing it, thus undermining a “good mothering/appropriate pregnancies” discourse. In an intervention, this alternative story could be further explored and expanded upon: Why did the mother not have a problem? Are there others in the community who feel the same way? and How does this alter how Nonny sees her own pregnancy? In universal interventions, the practitioner does not have direct face-to-face contact with women. The requirement, thus, is for the reflexive practitioner to immerse themselves in contextual dynamics. They must research, through reading and community engagement, the power relations underpinning gendered inequities specific to the location and must hone messages accordingly.
In both clinical and universal interventions, the first two concepts outlined above are important. Alternative less “problem-saturated” stories are enabled through reference to the gendered discourses and meta-narratives operative in the particular context as well as narrative reciprocity (where personal stories interweave). Working with these in both clinical and universal interventions would be crucial for shifting the stories of women who drink alcohol during pregnancy in positive directions. In relation to the participants in our setting, this could include, in line with Lee’s (1997) approach:
Addressing the “good mother” myth by normalizing, in both clinical and universal interventions, stories of ambivalence, stress, and shame in relation to mothering;
Externalizing the problem of the bad mother to the structural, cultural, and social issues that shape these women’s lives. For example, in clinical interventions, this can be practiced through questions such as “What does thinking of yourself as a bad mother tell you about the expectations your community has for you?” Likewise, in universal interventions, this can be practiced through portraying culturally relevant interactions in which women resist the bad mother subject position;
Encouraging alternative narratives or counterplots that speak to positive aspects of the women’s identity or interactions. For instance, in clinical interactions, this can be practiced by helping clients read their experiences through different discursive resources in order to renegotiate personal stories that have the possibility of opening up new opportunities. Similarly, in universal interventions, this might involve honing in on the positive aspects of women’s lives separate from their identities as drinkers.
The formation of constructive alternative stories may require labor to unpick clients’ taken-for-granted assumptions. For example, participants in this study represented Xhosa culture as static, homogenous, and judgmental. As opposed to the religious discourse, no narratives of redemption or support were evidenced in the data in relation to the cultural discourse. Where clients did draw on such negative representations of culture, the counselor would need to pose questions that highlight the dynamic, fluid, and contradictory nature of culture and the possibilities of support—in general and specifically through contextual examples.
In sum, both clinical and universal interventions premised on feminist narrative principles should (1) recognize how alcohol use during pregnancy is embedded in power relations that intersect around gender, class, culture, religion, race, and age; (2) surface the discourses of meta-narratives that serve to position women as solely responsible for the outcomes of their pregnancies; (3) counter discourses, narratives, and social practices that underpin local gender norms and that lead to actions that are not constructive to the health and well-being of the woman and the fetus; and (4) assist women in storying their lives in positive ways, in which their identity as a drinker is counterbalanced with positive stories that enable health-related behavior and interactions.
Footnotes
Acknowledgments
We thank the women who participated in this research for their time and frankness; our colleagues on the broader project, Pieter Bredenkamp, Nicola Jearey-Graham, Katlego Molokoe, Nqobile Msomi, and Charles Young for their support and input; and the NGO partners (names omitted for the sake of anonymity) who assisted us in accessing the participants and who conduct work in the communities.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Eastern Cape Liquor Board and the South African Research Chairs initiative of the Department of Science and Technology and National Research Foundation of South Africa (87582).
