The Significance of Sheltering in the Lives of Four Women Affected by Abusive Relationships

The significance of sheltering in the

lives of four women affected by

abusive relationships

Ruth WrightDepartment of Psychology, University of the Witwatersrand, Johannesburg, South Africa

Peace KiguwaDepartment of Psychology, University of the Witwatersrand, Johannesburg, South Africa

Charles PotterDepartment of Psychology, School of Human and Community Development, University of theWitwatersrand, Private Bag 3, WITS, 2050, South [email protected]

Domestic violence is recognised as a pervasive problem in South Africa. This study focused on thenarratives of four abused women and attempted to establish the significance of sheltering in their lives. Aqualitative research design was used based on semi-structured interviews, which were analysedthematically. The findings supported past research, indicating that although the women’s experiences werediverse they contained many features similar to those reported in previous studies. Each of the four womenwas not a passive victim, having taken the decision to leave an abusive and violent relationship. Shelteringwas found to provide the protected space necessary to move beyond the abuse, and in addition providedthe structure and social support necessary to start to transcend the abuse. Sheltering was found to fill gapscreated by a society in transition, in which abuse and violence are often tolerated or condoned withinexisting social and family structures.

Keywords: abused women; agency; domestic violence; gender violence; empowerment; shelter

Although there have been many studies of domestic violence over the past three

decades, it remains a persistent and elusive social problem without reliable or lasting

solutions. The most publicly and professionally debated aspect remains the quandary

over why women remain in abusive relationships. Research has established that the

problem has no socioeconomic, racial, ethnic, or national boundaries, affecting one

woman in six worldwide (Lawson, 2003). In South Africa, sheltering specifically for

abused women is a comparatively recent phenomenon. The knowledge and models of

sheltering based on Western experience have created debate on the appropriateness of

shelters in different environments. Many African and developing nations do not use or

616

# Psychological Society of South Africa. All rights reserved. South African Journal of Psychology, 37(3), 2007, pp. 616–637ISSN 0081-2464

see the need for high-cost sheltering since the role of the traditional extended family is

to act as a mediator in disputes between partners (Park, Shaik, & Rasool, 2000). Yet

social conditions such as urbanisation, modernisation, and the patterns of migrant

workers in Africa have impacted on the traditional family deterrents to domestic

violence (Vetten, 2000).

In South Africa, Angless and Maconachie (1996) report that it is estimated that one

in four women is subjected to violence from her intimate partner. These are believed to

be under-estimates. South Africa has been described as ‘one of the gender violence

capitals of the world’ (Dangor, Fedler, & Park, 2000). The current theoretical

explanations for partner/intimate violence are divided into three major groupings. Each

of these theoretical positions emphasises the complex, multifaceted nature of the

phenomenon, which is conceptualised as both a social problem and an individual one

(Dwyer, Smokowski, Bricout, & Wodarski, 1996).

The psychiatric model covers all theories pertaining to individual pathology,

including early childhood development, intrapsychic characteristics, alcohol and drug

abuse, and interactional family theory (Foreman & Dallos, 1993; Lawson, 2003;

Wallace, 1998). The social-psychological model analyses the environmental factors and

social structures by focusing on social learning theory (Simons, Lin, & Gordon, 1998),

resource exchange theory (MacMillan & Gartner, 1999), ecological theory (Carlson,

1984), and evolutionary psychological theory (Wilson & Daly, 1998). Finally, the socio-

cultural model based largely on feminist theory is focused on the social roles of women

and the cultural attitudes towards viewing violence against women as acceptable. This

group includes theories such as violence in society as a whole and patriarchy theory

(Dobash & Dobash, 1992; Fedler & Tanzer, 2000; McWilliams, 1998; Vetten, 2000).

A review of the research literature on domestic violence (Wright, 2006) suggests that

different theoretical models have taken dominant positions at particular times in

history, and are dependent on fluctuating contextual and cultural influences. In

studying the types of abuse that women experience, it is thus essential to determine the

impact of the abuse on their lives, as this informs the choices women perceive as

available to them. Male abuse against women can be conceptualised as occurring on a

continuum of behaviours, involving explicit and implicit threats to misuse power over

women (Goodman, Koss, Fizgerald, Russo, & Keita, 1993). Traditionally, research

concentrated on the type, frequency, severity, and uncontrollable nature of the physical

abuse (Lawson, 2003). Contemporary research acknowledges that men use other forms

of controlling behaviours such as sexual, economic, and emotional or psychological

abuse (Kirkwood, 1997).

Given the high prevalence of violence in society, what is defined as abuse and what

are acceptable levels of abuse are debated both in the academic and public forums.

Sorenson (1996) points out that the attitudes concerning acceptable levels of abuse have

evolved over time and are determined by social norms and practices. Kirkwood (1997)

proposes that the different types of abuse do not operate in isolation, but are

617

The significance of sheltering in the lives of four women affected by abusive relationships

interconnected and therefore impact on intervention and the healing process for the

women.

This study focused on the narratives of four women in relationships in which they

experienced severe mental and physical abuse. It also explored the situations that led

them to leave the abusive relationships and seek shelter. In addition, the study

examined whether the women were able to move beyond and transcend their

experiences with abuse. The study was guided by the following research questions:

(a) Why do women seek assistance in sheltering? (b) Did the shelter experience meet the

expectations and needs of the women? and (c) Did the women experience empowerment

and agency as a result of their stay at the shelter?

RESPONSES TO ABUSE: THREE THEORETICAL FRAMEWORKS

The responses of four women to abuse were analysed using Lempert’s (1996) theoretical

framework, staying/leaving patterns (Kirkwood, 1997), and paradigmatic shift or last-

straw events (Rosen & Stith, 1995). Lempert (1996) outlines three progressive stages

that women use to cope with the abuse and its contradictions: keeping the violence

invisible, containing the violence and preserving agency, and making the invisible

visible.

Lempert’s stages of coping with abuse

Lempert (1996) provides a good starting point in her outline of three progressive stages

that women use to cope with the abuse and its contradictions. She argues that violent

episodes are often part of a very entangled and multifaceted relationship between two

people. In the first stage of the abuse, the overriding concern is with keeping the abuse

invisible. Caught up in contradictory and overly romanticised expectations of her role

as a ‘dutiful’ and ‘good’ partner, a woman in an abusive relationship will tend to

experience both shame and confusion about her abusive situation, and will therefore

hide the violence. Face-saving strategies as a way of preventing public exposure of the

abuse are then used to preserve some semblance of credibility as a happy couple.

According to Lempert (1996), the abusive male partner is also often complicit in this

charade and will use protective measures such as controlled assaults to hide any

evidence of beatings.

The second stage is characterised by a desire both to contain the violence and preserve

personal agency. This stage brings realisation of unhealthy and idealistic expectations

for relationships as possibly unattainable. The increasing unpredictability, frequency,

and severity of the abuse also contribute to this ‘turning point’ (Campbell, Rose, Kub,

& Nedd, 1998) whereby women will negotiate for a non-violent relationship, such as

problem-solving, self-preservation, and subordinating self strategies. These strategies

are influenced by the way in which the woman understands the violence and the

618

Ruth Wright, Peace Kiguwa, and Charles Potter

resources available to her. The strategy of leaving, however, is still not central to these

processes.

The third stage is dependent on both cultural definitions and contextual influences.

This stage is concerned with making the violence more visible, for example by taking

family members into close confidence about the abuse (Dobash & Dobash, 1992).

Legal, social, and medical agencies are also consulted. Kirkwood (1997) describes this

as an outward spiral movement whereby the woman becomes aware of the

consequences of the abuse and the importance of gaining personal resources as a

means of shifting the balance of control from her abuser, and to re-establish her own

needs, wishes, and eventual departure.

Staying/leaving patterns

Why women stay, leave, and then return to abusive relationships continues to be one of

the most debated research questions in domestic violence literature. Common

perceptions or misperceptions of women’s having the option to stay or leave do not

always take into account the woman’s own perceptions and contextual factors that may

limit her opportunity to do so (Choice & Lamke, 1997; Kirkwood, 1997). Numerous

theoretical positions have been advocated, however, to explain abused women’s staying

and leaving patterns. These include the notion of a learned helplessness (Walker, 1984,

cited in Choice & Lamke, 1997), cycle of violence (Walker, 1984, cited in Wallace,

1998), traumatic bonding theory (Wallace, 1998), psychological entrapment, investment

model, and reasoned action approach (Choice & Lamke, 1997; Rusbult & Martz, 1995),

normalising violence model (Eisikovits, Goldblatt, & Winstok, 1999), and exchange

theory (Johnson, 1992).

Paradigmatic shift or last straw events

Rosen and Stith (1995) identify the process of leaving an abusive relationship as ‘a

paradigmatic shift and last straw event’ (p. 158). This process can occur over an

extended length of time where there is an accumulation of numerous small events, or it

can occur suddenly and dramatically with one catastrophic event leading to immediate

leaving. When women assess the situation as dangerous or undeserving, they let go of

their relationship-maintaining rationalisations and begin to actively seek different

options (Rosen & Stith, 1995). A process of growing self-awareness and self-worth

begins to be nurtured more actively. However, although this shift brings about the

desire to leave the relationship, leaving is often a painful and difficult process.

Dobash and Dobash (1992) and Green (1999) argue that women are not passive,

docile victims as portrayed in popular literature, but are actively negotiating their

relationship in order to control or eliminate the abuse. Kirkwood (1997) presents the

most influential argument in emphasising the empowering and learning elements of the

staying/leaving actions in her spiral inward/outward concept. As two very prominent

619


concepts in community psychology, empowerment and human agency have generated

considerable debate, making them complicated concepts to operationalise in evaluating

their presence. The critics of empowerment find the concept problematic since it is

based on Western social and economic belief systems (Riger, 1993), while authors such

as Ramazanoglu and Holland (1993) argue that women’s socialisation and the

discourses surrounding women’s experiences with subordination compromise female

agency and actions.

Yet empowerment theory postulates that it is possible to empower women in a shelter

environment. In Kirkwood’s (1997) research, three levels were identified at which

sheltering can assist abused women to find empowerment and agency through the

acquisition of resources and knowledge, the opportunity to give voice to their

experiences, and actions towards broader social change. These issues are explored in the

current study by focusing on the decision made by four South African women to seek

safety in a shelter in preference to traditional community and family networks and

structures, as a basis for moving beyond the abuse and taking control of their lives.

However, the notion of empowerment in relation to abused women remains a difficult

one, for many reasons. The tensions that exist between notions of subjectivity and

agency highlight some of the contradictory positions that some women experiencing

intimate partner abuse will adopt, such as choosing to stay or even return to the abusive

relationship. Such contradictory positions are not accounted for in traditionally

essentialist constructs of subjectivity (Long & Zietkiewicz, 2006). Although it is not the

focal point of discussion in this study, we acknowledge this significance in moving

beyond traditional conceptualisations of women abuse.

METHOD

Research into domestic violence is influenced by the debate over the appropriate use of

qualitative versus quantitative research methodology. Although statistical data have

been instrumental in bringing public and academic attention to the plight of battered

women, Johnson (1998) argues that qualitative and ethnographic research studies are

required to document the real-life subtleties of human interactions implicated in

violence against women. This study is based in the qualitative research tradition,

concentrating on the narratives of four women, which are explored in depth. It thus

focuses on what Simons et al. (1998) has called the ‘science of the singular’, and

attempts to find commonalties across a small number of individual case studies (Lucas,

1974a, 1974b).

Design

For this research study, a qualitative naturalistic design was employed, using a semi-

structured interview technique. Due to the naturalistic format, the research design

proved to be an evolving process. Following Neuman’s (1997) suggestions on field

620


research, a technique of progressive focusing was used to identify themes as they

emerged in the interview process, which were then explored in depth. An interview was

also conducted with the manager of the shelter, focused on the particular context and

form of the sheltering provided to the residents.

Participants

The purpose of this study was to focus on individual experiences. The sampling

strategies used thus differed from those used in quantitative research, which would have

required a systematic representative statistical sample of the population. The study

maintains the qualitative research objectives of obtaining information and under-

standing a range of opinions and experiences, rather than counting those opinions and

people (Gaskell, 2000). Gaskell argues that selecting a random probability sample is

unsuitable for the purpose of qualitative studies because a given social milieu will

contain only a limited number of views, positions, or experiences. The number of hours

involved in analysing a single interview also limits the number of interviews that can be

completed. The aim was in-depth analysis of content.

Four residents from a shelter for abused women were recruited through direct

solicitation of volunteers by the shelter staff. These respondents represented a non-

probability, purposeful, convenience sample. Each woman’s admission into the shelter

was based on a claim that she was experiencing an abusive relationship, and that she

had decided to leave the relationship. The length of time women had been in abusive

relationships extended between almost one to eight years. Although each woman had

been subjected to physical, sexual, economic, and emotional abusive behaviours

(Kirkwood, 1997), in each case the escalation of the physical abuse combined with

another form of abuse was instrumental in the leaving decision.

The interviews were conducted in English, which was the interviewer’s home

language. Two of the women respondents were South African, of whom one spoke

isiZulu as her home language and the other one referred to herself as ‘white African’.

One woman was a refugee from the Congo, and the other one was a migrant worker

from Zimbabwe. All the women were able to communicate in English to different

degrees of proficiency. The Zulu woman spoke isiZulu and was worried that her

English was not good enough for an interview, while the other South African woman

spoke Afrikaans and English. The Congo refugee’s home language was French but her

proficiency in English and other South African languages was better than the

impression she initially gave. The Zimbabwe migrant worker referred to herself as

‘English Zimbabwean’ and she also spoke the language of her husband’s Shona culture.

The ages of the four respondents were 20, 27, 29, and 41, while the interviewer was in

her late 50s. The educational backgrounds of the participants were diverse. The woman

from Zimbabwe had not finished school and planned to work on her matric through

correspondence courses. For the Congo refugee, it was difficult to assess her level of

education, since she was vague, stating that she had a ‘hostess diploma’ from her

621


country. The Zulu woman had grade 9 education without further training. The white

African woman had a diploma in business administration; however, she had been

unable to get a job using these skills. The ages, educational, and ethnic backgrounds of

the women in the sample were thus broad, allowing for an inspection of different

experiences with and reactions to abusive relationships. It is important to reiterate that

generalisability of findings was not a focal objective of the study. Rather, what is

emphasised and discussed are the shared meanings of intimate partner violence

experienced by a specific group of women at a shelter, negotiation of the leaving

process, and their perceptions of the utility of a ‘safe house’ or shelter. Thus, while the

backgrounds of the participants may be diverse in relation to education, nationality,

race, and so on, there is still some homogeneity in their constructs and interpretations

of abuse. As expected, this homogeneity is not always consistent, but it is nonetheless

significant in its implications for ways in which women understand their own experience

of abuse and the staying/leaving process.

An interview was also conducted with the shelter manager, who volunteered to give

supporting information on the shelter and formed the fifth participant in the study. The

shelter manager was in her 30s and had been employed at the shelter for two years, first

as a social worker for the children, and then as the shelter manager. She had a degree in

social work. Culture, age, education, and language, as well as reflexivity, may thus have

influenced the quality of the data obtained in the interviews. These factors should be

borne in mind in interpreting the results of this study.

Instruments

Semi-structured interviews were used to collect data from all five participants. This

form of interviewing provided a flexible field research technique, enabling the

researcher to focus selectively on issues raised by each participant. It also enabled

the researcher, if necessary, to avoid probing issues that were sensitive to particular

participants, thus protecting the interests of the participants while maintaining a degree

of control over the topics discussed.

A topic guideline (see Appendix) for the interviews was constructed based on the

example presented in Kirkwood’s (1997, p. 159) research, which was adapted based on

a search of the literature. Further modifications and adaptations also took place during

the interview process, as experience with using the instrument was developed in the

field.

Initially, only one instrument was constructed, for use with the four abused women

respondents. However, a change in focus in the research design took place as a result of

the information obtained during the interviews with the four women. It was found that

interpretation was not possible in the absence of contextual information on the shelter.

An additional interview was thus negotiated with the shelter manager. The additional

interview topic guideline was constructed in a similar format to the one previously

created for the interviews with the four women respondents.

622


Setting

The shelter’s exact location and parent organisation need to be kept confidential for the

safety and privacy of current and future residents. It was a residential home located in a

major urban setting, and had been in operation for a number of years at the time the

research was conducted. Its capacity was about ten women with their children. The

shelter staff consisted of one manager (social worker), one social worker, and two

housemothers. The length of stay for those seeking shelter was normally between one

and six months. The cost of the shelter stay for those employed was 50 rands per

month, and no cost for those unemployed.

The advantages of conducting the interviews in this setting were that the women were

in a familiar, safe place to relive painful, traumatic experiences and feelings, and if they

required assistance, their counsellors would be available. The interviews took place in

the office building in the back garden.

Procedure

Prior to conducting the interviews, approval for the research and the instruments

constructed was secured from the appropriate university committees as well as the

shelter manager. Arrangements with the shelter manager were then made to conduct the

interviews at a time convenient for the participants and the shelter staff. Both verbal

and written consent was obtained from the participants in the study. In addition, the

participants were made aware of their right to terminate the interviews at any time, and

assurances of confidentiality were given. The purpose of the study and the recording

procedure were verbally explained. Copies of the information letter and topic guide

were then given to the participant, along with verbal explanations. The participants

were encouraged to voice any concerns and questions they had about the research.

Measures were taken in attempting to put the participants at ease to counteract social

norms encouraging abused women to be silent. As anticipated, the participants did at

various times test the interviewer’s sincerity, interest, and dependability.

All the interviews were conducted on a one-to-one basis, each interview being audio-

taped for later transcription. On average, the interviews with the women lasted

approximately ninety minutes. The interview with the shelter manager lasted over two

hours. This enabled probing to take place, and also ensured that sufficient time and

space was provided for the participants’ stories to emerge.

The narrative process emerged at different times within each interview, and the

involvement of the participant in telling her personal story was taken as an indicator of

a successful interview. Initially, a great number of yes and no answers were given in

response to questions. Once the ice had been broken, the personal narrative would

normally emerge, followed by probing, which in each case was thorough. Some

exasperation by the women was noted with the resultant length of time taken to

complete the interview topics. However, this was unlikely to have affected the quality of

623


the women’s narratives, which were recorded prior to subsequent probing, and

generally early in the interview process.

The participants were encouraged to seek help from the shelter staff if the re-telling of

their stories resulted in later emotional upheaval. The women were thanked for their

time and the sharing of their stories, and for participating in the research. Field notes

were then made immediately after each interview to supplement the audio-tapes in

recording non-verbal behaviours.

Data analysis

The audio-tapes of the interviews and field notes were transcribed to paper copies and

onto a portable computer disk, maintaining where possible the exact words used by the

interviewees, including pauses and non-verbal communications. The transcriptions

were organised in the same manner as the topic guidelines.

Data analysis was conducted through progressive use of three approaches, namely

interpretative phenomenology, content analysis, and social construction. These three

approaches were used to enable the researcher to familiarise herself with the data, prior

to answering the research questions.

The first step in the data analysis followed the steps for interpretative phenomen-

ological analysis, as outlined by Willig (2001), to focus initially on the impressions of the

content of the text. This involved examining the women’s narratives to establish how they

made sense of and interpreted their world as entangled in violence and abuse. In the

process, the specific context, time, quality, and texture of each woman’s experience were

identified, as outlined in her responses to the open-ended questions in the interviews.

It was assumed at this first stage in the analysis that the interactive characteristics of

the social world were important elements in each woman’s life-world, giving meaning to

events in her life. Additionally, it was acknowledged that the researcher’s own view of

the world and the relationship between the researcher and the participant had been

implicated in the process, limiting the researcher’s access to a full understanding. The

subsequent steps in the analysis thus involved progressive focusing on themes in the

narratives, which were then explored in depth.

The second step thus involved focusing on specific themes in the women’s narratives,

which were identified through a process of content analysis. The topic guidelines

formed the preliminary schemes. This was done as the questions in each interview

schedule had been based on prior categories obtained from analysis of the literature and

prior research. From these standard categories, additional categories were then

identified in each woman’s narrative.

Descriptive coding and latent coding were done through the use of paper/pencil

coding on code sheets. This was essentially a subjective process in which the researcher

first overviewed the content of the text and then conducted coding and interpretation.

In the process, clear word definitions were first established and then tested against the

data, being either modified or maintained. The themes identified at this stage of the

624


analysis were then compared with findings of a large South African survey (Rasool,

Vermaak, Pharoah, Louw, & Stavrou, 2002), to establish whether the specific

contextual themes identified in the participants’ narratives were similar to those of a

wider group of abused women in South Africa.

The third step in the analysis involved examination of the themes against a

background of social theory, in attempting to establish whether there were particular

social constructs or structures that were instrumental in maintaining or resisting

domestic violence in this particular social milieu. Throughout the research literature,

particular discursive positions are allotted to domestic violence and abused women

through culture, historical times, contextual factors, and language. The themes from the

interview data were thus examined for the presence or absence of the discursive

positions established in the literature review.

The final step in the analysis then involved establishing the relationship between the

themes identified through the different steps of the analysis and the research questions,

both within the narratives of the individual interviewees and across the different

interview protocols. This involved analysis of commonalties and differences in ways in

which the women described and voiced their experiences of sheltering, against the

background of theory.

To answer the first research question, numerous theories were drawn on to establish

aetiology of abusive relationships. This involved examining psychological, social-

psychological, and social-cultural factors, which could be possible explanations for the

significance attributed to the types of abuse experienced by the women. The theoretical

frameworks of Dobash and Dobash (1992), Lempert (1996), and Green (1999) were

used to establish how different issues and responses had influenced each woman’s

decision-making process, relative to the decision to leave the abusive male partner and

the decision to seek sheltering.

To identify the reaction of each of the women to the support provided by the shelter as

elicited in research question two, the work of Dobash and Dobash (1992), Loseke (1992),

Kirkwood (1997), andPark et al. (2000)was used as a theoretical framework for analysing

the functionality of shelters for abused women. Finally, for research question three, the

stages of empowerment identified in Kirkwood’s research were used to establish whether

empowerment could be attributed to the sheltering process for these women.

Ethical considerations

Prior to conducting the interviews, approval for the study and the instruments was

secured from the Human Research Ethics Committee and the research manager for the

shelter organisation. The recommended changes made by the ethics committee were

implemented. To guarantee informed consent for voluntary participants, two consent

forms were prepared and approved. An information letter was given to each woman,

outlining the purpose of the research and the procedures to be taken. This was

reinforced with a verbal explanation. The participants were given the opportunity to

625


read the topic guide for the interview and to ask any questions before starting the

interview. The women were also informed of the true nature and purpose of the study,

making sure each was fully aware of the purpose and aims of the study, and

understanding what she was agreeing to. Although none of the women interviewed

asked for therapeutic counsel, a skilled counsellor was available on the premises,

provided by the shelter. A debriefing session was also held after the interviews.

RESULTS

Why do women seek assistance in sheltering?

In order to examine why sheltering at this point became an option for the women, the

inquiry took three different but interconnected routes. The first route examined the

possible causes for the abusive relationships. The second investigated the significance of

the types of abuse experienced, while the third established what kinds of response were

available for each woman in the context of her particular aetiology of abuse. Table 1

shows a summary of the accounts of the women, indicating the presence of both

individual and shared aetiologies for abuse.

Table 1. Individual and shared aetiologies for abuse

Shared causes

Psychological

causes

Immature poor communication skills between partners in which violencewas used as a powerful coercion tool and a process to resolve conflict.The undermining of women’s self-confidence; self-esteem by male partners.

The women experiencing feelings of guilt, shame, embarrassment, and/orresponsibility, leading to further isolation and vulnerability.

Learned behaviour led to silence about beatings, defeating intervention.The men’s use of the women’s core identity as woman/wife/mother toattack them through criticism, humiliation, belittlement, ridicule, name-calling, and making fun of their lack of skills and knowledge.

Social-

psychological

causes

High level of tolerance for physical abuse by the women and in the socialenvironment, with low risk of intervention, therefore men beat theirpartners because they could.No attempts to hide evidence of beatings, instead the evidence appears totake on a badge of ownership.Forced isolation and privacy of family environment.Access to resources controlled by males and used as controlling lever.

Extreme jealousy and possessiveness, involving autonomy-limitingbehaviours and/or tight vigilance on women’s activities, were exploited asan illustration of love and reasons for the violence by both partners.Dependency on male partner for housing and financial support.

Poor knowledge and use of informal and formal social support networks.Presence of children.Urban, nuclear family setting with female isolated from extended family.

Influence of male partners’ extended families.

626


Social-cultural

causes

Predominance of patriarchal ideology: male supreme authority; right tophysically punish female partners; subordinate role of women, treated asproperty or possession; women subjected to expectations of male partner;women’s belief in roles and family structures defining ‘good wife andmother’; tactics of domination used by the male partners.Conflict over sex and children.

Trust and belief in romantic love and religious sanction of marriage used bymale partners to keep the women in the relationship.Unemployment or low wages of male partner.Poverty.Social conditions that promote violence against women such as culture ofviolence, high tolerance for violence, inability or reluctance of women touse legal/police resources, and ineffectiveness of police.

Individual causes

Psychologicalcauses

Individual pathology: depression, shyness, withdrawal, low self-esteem,insecurity, anxiety, hopelessness, suicidal, and/or dependency.

Experience with abuse in formative years, causing vulnerability later.Child in position of danger.Male partner suffering from alcoholism.

Fear of abandonment in male partner, causing inability to controlaggression and rage.

Sole financial provider for family a source of self-confidence and self-esteem but also instrumental in the continued abuse.

Social-

psychologicalcauses

Youth and inexperience with older male partner.For older women, years invested in the relationship and the problems ofstarting over.Ability to provide accommodation tied to husband’s continued demandsand abuse.Husband able to force his wife to live with him through threats of killing thewoman and their children.

Social-cultural

causes

Poor education opportunities for black women, leading to pooremployment prospects.Status of migrant worker and refugee leading to vulnerability.

With reference to patriarchal theory, all the sources of conflict leading to violence in

intimate relationships identified by Dobash and Dobash (1992) and Fedler and Tanzer

(2000) were present in the accounts of the women. These factors were aggravated by their

socialisation, which led them to believe and trust in the nurturing role of women, romantic

love, and the religious sanction of marriage. There was also evidence of a need to find

excuses for such behaviours so that the women could maintain their roles of ‘good wife/

mother’. Consequently, although the interview data portrayed images of coercive,

physically abusive relationships to one listening to their stories, the women themselves

didnot strongly identifywith the label of ‘abused/batteredwife’.All fourwomenconsidered

themselves tobe ‘survivors’ in the reality of their lives, and identified stronglywith this label.

Feminist literature has argued it is important for any abused woman to take on survivor

roles as part of the discursive shift that enables her to go beyond the abusive relationship to

re-establish herself as a human being worthy of love and self-autonomy:

627


He never treated me like he should treat a woman. (Clara)I want to spare this baby . . . be a good role model for my little girl. (Dora)

Responses to abuse

Theoretically, the women’s responses to abuse were consistent with Lempert’s (1996)

stages. Each of the women attempted to satisfy her partner’s demands and expectations.

Each woman would use cognitive, dissociation, subordination, non-confrontational,

non-threatening, and problem-solving strategies. Staying/leaving behaviours were used

as negotiating strategies by three of the women (Dobash & Dobash, 1992). However,

few traditional African strategies (Green, 1999) were identified, indicating that the

option of using their extended families as a resource was not available to them. The

presence of a paradigmatic shift or last-straw event (Rosen & Stith, 1995) for each of

the women was tied not to the increased severity of the physical abuse but to the

combining of other forms of abuse with the physical abuse; thus instigating an

immediate leaving decision:

This time my husband was beating my little girl so badly that the school phoned CPU[Child Protection Unit]. And they came and arrested my husband. I then I decided it wastime for me to leave . . . for my daughter. (Dora)Me, yes, I loved him but [. . .] No, not now, not now. I finished to love him . . . when hestarted forcing me to have sex. (Amy)

Did the shelter experience meet the expectations and needs of thewomen?

The nature of the shelter’s ability to provide support for the women was gauged from

the accounts of the four women respondents, examined in conjunction with the

interview with the shelter manager. For each of the four women respondents, the shelter

represented a safe place to stay. It met their basic needs for survival, accommodation,

food, and support. The availability of a shelter made the process to leave easier. Given

the traditions and cultural background of each of the women, it is of interest on a cross-

cultural level that the manner in which the women came to the shelter was similar to

that reported in a number of Western societies. However, what was apparently different

in South Africa was the absence of self-referrals to the shelter commonly found in

Western societies (Lawless, 2001; Loseke, 1992). Table 2 highlights further significant

differences between Western-style shelters and the particular shelter that is the object of

study.

628


Table 2. Similarities and differences between the shelter studies and Western-style

shelters

Similarities to Western-style shelters Differences from Western-style shelters

Safety — a place free from abuse and violence Shelter services not as widely known in thecommunity; therefore referrals made throughother community services rather than self-referrals.

Basic needs for survival, shelter, food, andsupport are met.

Increased government controls as a result ofrequiring increased government funding.

Children stay with the mother when possible. Flexible, evolving process in meeting socialneeds of the community, not a rigid feministor therapeutic ideology.

Staffed by professionally trained staff. Individual counselling as the main therapeuticobjective rather than combined withcompulsory group sessions.

Professional counselling for mother and child. Fewer restrictions.

Funding through government and non-government agencies.

Race, socio-economic, and ethnic diversitycatered for.

Funding problems. Little emphasis on ‘good client’, insteadallowing the women to decide what their needsare, and admitting any woman in dangerwhether or not she is ready to leave theabusive relationship.

House rules similar. ‘Time out’ or ‘cooling down’ acceptablereasons for admission to shelter.

Safety problems.

Although all the women respondents reported feeling safe in the shelter, the shelter

manager reported that security still posed a problem similar to that reported in other

shelters worldwide. Nevertheless, sheltering for these abused women was able to fill

gaps in social support and provision, which have been exacerbated in South Africa by

the dynamics of a society in transition. The women’s stories illustrated legitimate cases

of domestic abuse, the need for the women to escape the abusive environment, and the

community’s inability to respond on levels of family and traditional controls:

If you were married in my country, if you are fighting you go to your families. Bothfamilies would meet and they would talk about. (Amy)I trying to call people, neighbours, but no one help me. (Beth)

The results also confirmed past findings (Loseke, 1992) in indicating that women come

to shelters anticipating that the shelter will meet a wide diversity of expectations and

needs. The type of flexible response provided by the shelter staff appeared to contribute

to a positive shelter experience by each of the women interviewed. The data indicated

that the shelter adopted a mixture of therapeutic and activist approaches, involving not

just the containment of the social problem but also encompassing individual and social

change. It therefore created a specific experience for these women based on a mental

health approach in which professional counselling formed the main intervention

629


strategy. Each of the women reported benefiting from the counselling, suggesting the

value of this form of social support.

I am not a very forceful person. They help me to say no to something I don’t like. I talkabout it if something happens now. (Amy)When I am going to group sessions, sharing problems . . . I feel alright [. . .] express howfeel, if not right, you talk. (Dora)I want to continue with counselling. I still can come back when I need to talk. (Beth)Sometimes I get very frustrated . . . [. . .] I get counselling and it helps . . . just talking. (Clara)

Did the women experience empowerment and agency through theshelter stay?

In addition to their support functions, shelters have been theorised as significant

empowering and agency tools for abused women. Kirkwood (1997) has argued that

empowerment for abused women takes place in stages. The first stage is coping with the

abuse. The second is acknowledging its effects on freedom and control over one’s life.

The third stage involves entering a survival mode. The data from the interviews

indicated that, in the shelter environment, three of the women had entered the second

stage and were fluctuating between acknowledging the effects the abuse and attempting

to find the most appropriate survival mode for their particular needs. One woman

however, remained in the first stage (coping with the abuse):

I suffer in my house, and in my house I suffer. And he beat me. Suffer, wicked man, nomoney. When he does not pay me, you suffer without the money. Now, I don’t knowwhere I going when I finish this shelter. Where I am going, no money. I suffer. (Beth). . . fight, fight, every time fight. Go out and beat me. And go at me every time want tofight me . . . [. . .] He tell me wrong every time. It’s me, I am wrong. When I am talking, meI am wrong. [Silence] . . . marriage is not right for me. I not like it. No, another man, No,No, No, No [Very loud and firmly]. (Beth)I was not allowed to talk about it (the abuse). I was scared, scared that no one wouldbelieve me. But now . . . now I am divorcing him. It is a better life for me. (Amy)My husband said to me ‘you have to choose, it is either me or your family’, and he wanted anaffidavit from me. But I didn’t want to give him an affidavit. I said I would give him one. Iwould support him, just to keep my playing cards open because I was planning to move out.(Dora)

The data further indicated that eachwoman’s psychological presentations of helplessness,

depression, anger, hopelessness, and anxiety were possible influences in her abilities to

move beyond the abuse. Other factors noted in the interview data included the continued

influence of the abusive partner, limited supportive social network and social services, and

responsibility for children. The patriarchal subject positions of ‘good wife and mother’

influenced the way in which the women defined their positions within their relationships

with their partners, their marriage, and their community. However, it was also clear that

their abilities to challenge the culturally assigned subject positions were severely

obstructed by extreme fear and the need for survival.

630


Within the context of domestic violence, Kirkwood (1997) identified in her research

three levels in which abused women move towards empowerment. On the first level, the

women had been denied or restricted in obtaining practical resources and knowledge to

secure their independence. Maslow’s theory of self-actualisation (Kaplan & Sadock,

1988) points to the sequence of stages that must occur before the individual reaches

ultimate functioning levels. For the women interviewed, the inability to obtain the basic

survival-oriented needs such as shelter, food, and physical and emotional safety had

seriously hindered their abilities towards self-actualisation and eventually empowerment.

Kirkwood’s second level points to the need to break the silence imposed on women.

The ability to share stories of abuse among women is seen as an important empowering

aspect of sheltering. The breaking of the ‘silence’ is essential in negotiating for a non-

violent relationship or moving out of it. The women’s stories legitimised their admission

to the shelter and their descriptions as battered wives. Through the sharing, the women

developed awareness that they were not alone, but for these women there appeared to

be only a minimal decrease in their sense of isolation and embarrassment. Two of the

women were comfortable in sharing their stories within the shelter environment. The

other two stated that they were unable to share their experiences with the other women,

but had been able to do so with the shelter staff. At the beginning of the interviews,

three of the women demonstrated a certain amount of hesitation and shyness. However,

as the interviews proceeded, the women became more spontaneous, gave fuller answers,

and were more confident in expressing their opinions and describing their experiences.

Lawless (2001) claims that the act of speaking out allows women to reconstruct self in

moving beyond the identity of a battered woman to articulate a new empowered self.

Two of the women moved away from their responsibilities to their husbands to identify

their roles as mothers as the most significant position for their futures. One woman’s

enthusiasm in finding a new career path indicated her belief in an attainable new self:

I am taking my daughter home (to Zimbabwe) . . . so that I will be free to continue myeducation . . . I have a certificate now . . . and I am training to become communityresource worker . . . I am still going to counselling even when I leave this place (the shelter). . . but at least . . . I am taking a good step for myself now. (Amy)

DISCUSSION

Although the finding of a study based on self-reports of five volunteers cannot be

generalised, this study highlights a number of important issues in the complex nature of

domestic violence and its impact on women’s lives.

Empowerment and agency

A number of factors influencing empowerment and agency have been identified in

previous Western research (Kirkwood, 1997; Lawless, 2001). The data from this study

631


would suggest that the ability of women to take up sheltering opportunities in South

Africa is aggravated by severe social conditions connected to poverty and discrimina-

tion. The shelter environment offered access to resources and information not

previously accessible to each of the four women interviewed. It also offered the

opportunity of exposure to a different way of acknowledging and voicing the abuse,

which may in future allow for empowerment.

However, while the shelter played a successful role in enabling the women to obtain

the practical necessities they required to avoid a situation of severe domestic abuse, as

well as in providing a space for recovery and thus enabling the women to move beyond

their abusive relationships and gradually take control over their lives, it is difficult to

determine what longer term empowerment has been nurtured, if at all. Given the

obvious limitation of a small sample that is largely based on self-report responses, as

well as the multiplicity of identities that intersect, such as race, socio-economic position,

and culture, empowerment at an individual level remains difficult to fully ascertain. It is

hoped that the present study highlights the necessity of being more cognisant of

women’s responses to abuse that go beyond denial or rejection of the abuse, and may

rather involve also the active ‘taking on’ of identities that influence the staying/leaving

process.

LIMITATIONS AND INDICATIONS FOR FURTHER RESEARCH

This study has a number of limitations. The shelter sample was essentially selective,

volunteer, and non-random. The data collection procedure involved interviews, and

involved self-reports, which may have introduced bias into the data. As it was not

possible to involve the respective male partners in the study, it was not possible to

compare the women’s reports with their male partners’ perspectives. Cultural, age,

educational, and language factors, as well as issues of reflexivity, may also have

influenced the interview process.

The analysis of the data was subjective, and may thus have been open to alternative

interpretations. The transcriptions, for example, all contained a large number of ‘yes/

no’ answers. This was an indication that each of the subjects imposed limitations on her

responses to the interview questions for possibly different reasons, affecting the quality

and continuity of the stories, and the narrative quality of the data obtained.

Despite these limitations, this study indicates the importance of conducting similar

research in this area in the future. One avenue for future research would be to focus on

the staying-leaving-returning decisions of abused women. This would assist in

developing understanding of why women remain in abusive relationships, and the

factors in specific social, family, and societal contexts that contribute to decisions to

stay or leave. Such studies are necessary so that interventions can be designed to

support abused women in particular communities.

632


APPENDIX: TOPIC GUIDE FOR INTERVIEWS WITH PARTICIPANTS

Personal history

Age

Employment history

Parent’s employment

Ethnicity

Education

Training

Children: Ages & Sex

Who do they live with?

Are they children of abusive

partner?

Relationship status:

Married: living together / not

Divorced

Friend: living together / not

Age when met partner

How long together?

Objections to relationship:

From whom / why?

Able to maintain friendships

during relationship? Why /

why not?

Employed during

relationship

Any financial problems in

relationship

Type of housing: owned /

rented / lived with others

Abuse

How was conflict resolved:

In your family growing up

In your present relationship

How do you define abuse?

Acceptable not

What have you experienced

as abuse?

In childhood / dating /

adolescent / in your

present relationship

Mental abuse

Physical abuse

Sexual abuse

Financial abuse

Resulting injuries

When and why did the abuse

occur in your present

relationship?

First attack

First seen as abuse

How often Patterns in

abuser’s behaviour

Related to substance abuse

Changes over time and

marital status

Have your children

witnessed the abuse?

How did you feel about this?

Notice any change in their

behaviour?

Reaction to abuse

Silence — why?

Your reactions to the abuse:

Emotional / physical

Ways tried to stop attacks or

abuse: physical / verbal /

emotional

How successful were the

attempts?

Did the feelings change over

time –

About the abuse?

About self?

About partner?

Children’s reactions?

Whom did you tell about the

abuse?

Reactions of friends and

family?

People contacted for help

Service agencies contacted:

Responses of

Leaving

Why did you decide to leave?

Change in feelings

Still love your partner

Specific event or

circumstance

Sudden or planned for

Who contacted for help in

leaving?

Family / friends / agencies

What kind of help asked for?

What responses were most

helpful?

Were you aware of your

options?

Ever return to partner after

leaving?

Why?

Fear of reprisals?

From abuser / family /

friends

Shelter

Use of shelter services

before:

Why / When?

How did you find out about

the shelter?

Through other women

Other community services

Written information

Availability of shelter: did it

affect your decision to stay

/ leave? How?

Do you feel safe?

What services have been

most helpful?

Emergency accommodation

Counselling

Group discussions

Educational programmes:

Assertiveness training

Parenting skill

Problem-solving skills

Job / computer skills

Arts / crafts

Life skills

Job search / CV writing

Legal assistance

Medical referrals

Financial aid

Information: social

assistance

Women’s rights

Resources

Child care

633


Help with paper work

involvedWhat services not availablewould you like to have?

What information was mostuseful?

What services / informationgave you the feeling that

you could regain yourindependence / control ofyour life?

Do you feel free to talkabout your abusiverelationship in the shelter?

Does the presence of otherwomen with sameexperiences help?

Having difficulty:

With lack of privacy?With shelter rules?With sharing facilities?

With other’s lifestyles /cultural behaviours?

Some describe sheltering as

exchanging one prison foranother — is this so foryou?

How is conflict resolved in

shelter?

Effective / not?

Is there too much anti-male

prejudice?

Is the shelter a culture-

relevant environment for

your needs?

Language / food / religion

Are your privacy and

confidentiality

maintained?

Cultural beliefs

Role of women

Role of men

How are these beliefs formed

in your culture?

Does role definition

contribute to violence in

the family?

Are marriage bonds sacred,

not to be broken? Under

what conditions can be

broken?

Must you always listen to,

obey your partner?

Is marriage important to

you?

Is loyalty to family / husband

important?

How are decisions made in

your home?

Does your community

practise non-interference

in relations between

husband and wife?

Do you share household

chores / child care?

Religion: healing — source

of support / guilt / defines

your role as mother / wife?

Enforce marital controls?

Are culture values changing?

How?

Do you see yourself as

‘battered’?

Do you see yourself as a

survivor / victim / neither?

What are your hopes,

desires, ambitions for the

future?

REFERENCES

Angless, T., & Maconachie, M. (1996). Battered women: Problems and proposals. In L.

Glanz & A. Spiegel (Eds.), Violence and family life in contemporary South Africa (pp.

151–164). Pretoria: HSRC Press.

Campbell, J., Rose, L., Kub, J., & Nedd, D. (1998). Voices of strength and resistance: A

contextual and longitudinal analysis of women’s responses to battering. Journal of

Interpersonal Violence, 13(6), 743–762.

Carlson, B. (1984). Causes and maintenance of domestic violence: An ecological

analysis. Social Service Review (December), 569–587.

Choice, P., & Lamke, L. (1997). A conceptual approach to understanding abused

women’s stay/leave decisions. Journal of Family Issues, 18(3), 290–314.

Dangor, Z., Fedler, J., & Park, Y. (2000). Introduction. In Y. Park, J. Fedler, & Z.

Dangor (Eds.), Reclaiming women’s spaces: New perspectives on violence against

women and sheltering in South Africa (pp. 11–14). Johannesburg: NISAA Institute for

Women’s Development.

634


Dobash, R. E., & Dobash, R. P. (1992). Women, violence, and social change. London:

Routledge.

Dwyer, D., Smokowski, P., Bricout, J., & Wodarski, J. (1996). Domestic violence and

woman battering: Theories and practice implications. In A. Roberts (Ed.), Helping

battered women: New perspectives and remedies (pp. 67–81). New York: Oxford

University Press.

Eisikovits, Z., Goldblatt, H., & Winstok, Z. (1999). Partner accounts of intimate

violence: Towards a theoretical model. Families in Society: The Journal of

Contemporary Human Services, 80(6), 606–619.

Fedler, J., & Tanzer, Z. (2000). A world in denial: International perspectives on

violence against women. In Y. Park, J. Fedler, & Z. Dangor (Eds.), Reclaiming

women’s spaces: New perspectives on violence against women and sheltering in South

Africa (pp. 17–46.) Johannesburg: NISAA Institute for Women’s Development.

Foreman, S., & Dallos, R. (1993). Domestic violence. In R. Dallos & E. McLaughlin

(Eds.), Social problems and the family (pp. 7–45). London: Sage.

Gaskell, G. (2000). Individual and group interviewing. In M. Bauer & G. Gaskell

(Eds.), Qualitative researching with text, image, and sound: A practical handbook (pp.

38–56). London: Sage.

Goodman, L., Koss, M., Fitzgerald, L., Russo, N., & Keita, G. (1993). Male violence

against women: Current research and future directions. American Psychologist,

48(10), 1054–1058.

Green, D. (1999). Gender violence in Africa: African women’s responses. London:

Macmillan.

Johnson, H. (1998). Rethinking survey research on violence against women. In R. E.

Dobash & R. P. Dobash (Eds.), Rethinking violence against women (pp. 23–51).

London: Sage.

Kaplan, H., & Sadock, B. (1988). Synopsis of psychiatry: Behavioral sciences clinical

psychiatry (5th ed.). Baltimore, OH: Williams & Wilkins.

Kirkwood, C. (1997). Leaving abusive partners: From the scars of survival to the

wisdom for change. London: Sage.

Lawless, E. (2001). Women escaping violence: Empowerment through narrative.

Columbia, MO: University of Missouri Press.

Lawson, D. (2003). Incidence, explanations, and treatment of partner violence. Journal

of Counseling and Development, 81 (Winter), 19–32.

Lempert, L. (1996). Women’s strategies for survival: Developing agency in abusive

relationships. Journal of Family Violence, 11(3), 269–289.

Long, C, & Zietkiewicz, E. (2006). Going Places: Black women negotiating race and

gender in post-apartheid South Africa. In T. Shefer, F. Boonzaier, & P. Kiguwa

(Eds.), The gender of psychology (pp. 198–219). Cape Town: UCT Press.

Loseke, D. (1992). Battered women and shelters: The social construction of wife abuse.

Albany, NY: State University of New York Press.

635


Lucas, W. (1974a). The case-survey method: Aggregating case experience. Santa

Monica, CA: Rand Corporation.

Lucas, W. (1974b). The case survey and alternative methods for research aggregation.

Santa Monica, CA: Rand Corporation.

MacMillan, R., & Gartner, R. (1999). When she brings home the bacon: Labor-force

participation and the risk of spousal violence against women. Journal of Marriage

and Family, 61(November), 947–958.

McWilliams, M. (1998). Violence against women in societies under stress. In R. E.

Dobash & R. P. Dobash (Eds.), Rethinking violence against women (pp. 111–140).

London: Sage.

Neuman, W. L. (1997). Social research methods: Qualitative and quantitative

approaches. Boston, MA: Allyn & Bacon.

Park, Y., Shaik, F., & Rasool, S. (2000). Indigenous sanctuaries: A global view of

shelters for abused women. In Y. Park, J. Fedler, & Z. Dangor (Eds.), Reclaiming

women’s spaces: New perspectives on violence against women and sheltering in South

Africa (pp. 209–238). Johannesburg: NISAA Institute for Women’s Development.

Ramazanoglu, C., & Holland, J. (1993). Women’s sexuality and men’s appropriation of

desire. In C. Ramazanoglu (Ed.), Up against Foucault: Explorations of some tensions

between Foucault and feminism (pp. 239–261). London: Routledge.

Rasool, S., Vermaak, K., Pharoah, R., Louw, A., & Stavrou, A. (2002). Violence

against women: A national survey. Pretoria: Institute for Security Studies.

Riger, S. (1993). What’s wrong with empowerment. American Journal of Community

Psychology, 21(3), 279–293.

Rosen, K., & Stith, S. (1995). Women terminating abusive dating relationships: A

qualitative study. Journal of Social and Personal Relationships, 12(1), 155–160.

Rusbult, C., & Martz, J. (1995). Remaining in an abusive relationship: An investment

model analysis of nonvoluntary dependence. Personality and Social Psychology

Bulletin, 21(6), 558–571.

Simons, R., Lin, K., & Gordon, L. (1998). Socialization in the family of origin and male

dating violence: A prospective study. Journal of Marriage and the Family, 60, 467–

478.

Sorenson, S. (1996). Violence against women: Examining ethnic differences and

commonalities. Evaluation Review, 20(2), 123–145.

Vetten, L. (2000). Gender, race, and power dynamics in the face of social change:

Deconstructing violence against women in South Africa. In Y. Park, J. Fedler, & Z.

Dangor, (Eds.), Reclaiming women’s spaces: New perspectives on violence against

women and sheltering in South Africa (pp. 47–80). Johannesburg: NISAA Institute for

Women’s Development.

Wallace, H. (1998). Victimology: Legal, psychological, and social perspectives. Boston,

MA: Allyn & Bacon.

636


Willig, C. (2001). Introducing qualitative research in psychology: Adventures in theory

and method. Buckingham: Open University Press.

Wilson, M., & Daly, M. (1998). Lethal and nonlethal violence against wives and the

evolutionary psychology of male sexual proprietariness. In R. E. Dobash & R. P.

Dobash (Eds.), Rethinking violence against women (pp. 199–230). London: Sage.

Wright, R. (2006). The significance of sheltering in the lives of four women affected by

abusive relationships. Unpublished master’s research report, University of the

Witwatersrand, Johannesburg.

637


The Significance of Sheltering in the Lives of Four Women Affected by Abusive Relationships

Documents

Transcript of The Significance of Sheltering in the Lives of Four Women Affected by Abusive Relationships