Introduction

The term polygamy is a Greek word meaning “the practice of multiple marriages” and is used in related ways in social anthropology, sociobiology, and sociology. Polygamy can be defined as any “form of marriage in which a person [has] more than one spouse.” (Al-Krenawi and Slonim-Nevo 2008a). It has also been used synonymously with polygyny, there are several forms of polygamy: polyandry refers to the marriage of one woman to two or more husbands, polygyndry is a situation in which two or more women are at the same time married to two or more men, and polygynandry is used to refer to group marriage (Al-Krenawi and Graham 2009). The other type of polygamy is an informal polygamy which describes relationships characterized by the simultaneous existence of a legal marriage of one man to a woman and an affair with a second woman that has become a stable feature of the family structure (Rivett and Street 1993).

Polygamous communities exist in Algeria, Benin, Chad, Congo, Ghana, Togo, Tanzania, Thailand (Bao 2008), Saudi Arabia, United Arab Emirates (UAE), Egypt, Kuwait, Jordan and other societies including Mormons in the United States (Al-Krenawi and Graham 2006a; Al-Shamsi and Fulcher 2005; Al-Sherbiny 2005; Jankowiak 2008; Lesthaeghe et al. 1987). The socio-cultural significance of polygyny has been long appreciated (Al-Krenawi et al. 2001). The prevalence of polygyny represents a way of life which is profoundly embedded in a religiously and culturally motivated sense of obligation (Zeitzen 2008).

Particularly in the Arab Middle East, polygamy is often characterized by competition and jealousy among co-wives as is commonly observed within plural marriage communities (Al-Krenawi et al. 1997, 2001; Al-Krenawi and Lev-Weisel 2002; Chambers 1997; Madhavan and Sangeetha 2002; Starr and Brilmayer 2003; Wing 2001). Co-wives likely have very limited private time with the lone husband they share, and thus might vie for his attention and favour. In some polygamous communities, women’s self-worth is linked to the number of children they bear and, therefore, having time with their husband is also critical to their status within the family and community (Committee on Polygamous Issues 1993; Thompson and Erez 1994; Ware 1979; Welch and Glick 1981). Studies showed that in certain contexts, jealousy between co-wives can escalate to intolerable levels, resulting in physical injuries sustained by women and suicide attempts amongst women; families often live together in crammed and overcrowded conditions, creating an environment that aggravates stress and conflict between co-wives. (Bissuel 2002; Starr and Brilmayer 2003; Lauras-Lecoh 1990; Simons 1996). There are significant implications regarding children’s lower academic achievement, and men’s psychological problems, amongst polygamous marriages (Al-Krenawi and Graham 2009). The practice has implications for entire familial structures, including current as well as future families and communities.

Despite robust international research on polygamy, no research, to date, has considered the attitudes and experiences of women in Jordan. As a corrective, the present manuscript is the first to examine the psychological, family function, marital satisfaction, life satisfaction and the degree of agreement with the practice of polygamy among senior wives in polygamous marriage with a control group from monogamous marriage in Jordan.

The Practice of Polygamy in the Middle East

Several studies based in the Middle East show the association between psychosocial problems and polygamy among women and children from polygamous families (Al-Krenawi and Graham 2006b; Al-Sherbiny 2005). Al-Sharnebi (2002) claimed that the first wife in polygamous marriage is affected psychologically and these women are also more likely to visit mental health practitioners. A Jordanian study found that polygamy influenced the rate of school drop out among children, increased alcohol addiction, juvenile delinquency and led to low self-esteem (Alfaqer wa Tashqeel Alatfal fe Alordon 2002). Zeitoun (2001) found that polygamy and divorce were associated with almost 95 percent of cases of juvenile delinquency. This study found that one hundred percent of youth aged between 13 and 17 from Sharjah and the Northern Emirates were placed in care because of crimes such as theft or indecent practices associated with sexuality. These children came from families whose parents were divorced and the UAE national father was married to more than one wife (Zeitoun 2001). Al Kobesi’s UAE study (2001) showed that children from monogamous families adjusted to school better than children from polygamous families, and that children from different wives rarely developed positive sibling relationships, with sibling rivalries and jealousy reported instead.

Previous research in Egypt, Kuwait, Palestine, Saudi Arabia, and Turkey likewise confirms the considerable psychological and health risks that are prevalent amongst senior wives in polygamous marriages (Al-Krenawi et al. 2001; Al-Shamsi and Fulcher 2005; Al-Sherbiny 2005; Chaleby 1985). Ours is among the few Arab world studies that utilize standard research instruments, and therefore the validity of our findings is particularly compelling. There are therefore several implications that practitioners and future researchers might usefully take into account.

From an Islamic perspective there are several rules that must be followed by men who choose to practice polygamy. The Koran says “Marry women of your choice two, or three, or four; but if you fear that you shall not be able to treat justly with them, then only one… That will be more suitable to prevent you from evil” (Koran, 4:3). If a man cannot treat each of his wives equally, then he should only take one wife (Bewely 1999). Another verse says “You will never be able to deal justly between wives however much you desire (to do so). But (if you have more than one wife) do not turn altogether away from (from one) leaving her in suspense” (Koran, 4:129). Several studies point out the association between the practice of polygamy and psychosocial and familial problems this may be result of men acting without reference to the teachings of Islam—and in particular the imperative to treat all wives equally, and to assume a second wife only if economically feasible (Jameelah 2001). Abdu Salaam (1997) pointed out that 71 percent of Kuwaiti female respondents reported that men could not do justice or be fair between their wives; the same study showed that 50 percent of the men agreed that they cannot do justice between wives. In Islam, it is important that a husband tell his first wife whenever he plans to marry again; but in many instances, this consultation does not occur (Al Kobesi 2001).

Jordan

Jordan is a country with a population of nearly 6 million (Department of Statistics (DOS) 2009). Some 98 percent are Arabs, with a few small communities of Circassians, Armenians, and Kurds; each of which has adapted to the Arab culture; about 96 percent of the population is Muslim (Al-Krenawi et al. 2000). Approximately 70 percent of Jordan’s population is urban; less than 6 percent of the rural population is nomadic or semi-nomadic. Jordan shares its borders with the West Bank, Israel, Saudi Arabia, Syria, and Iraq. Jordan is considered to be a peaceful and stable country in the middle of a stormy, problematic region, and as a consequence of this it has become a shelter for a large number of refugees from neighbouring countries. Contemporary writers suggest that this movement of people has created a demographic crisis in Jordan. For example, Jordan has within its borders about 1.7 million persons registered as Palestinian refugees and, in addition, many other displaced persons reside in Jordan as citizens (CIA Factbook 2009).

Several refugee waves contributed to the current social and demographic changes in Jordan. Jordanians tend to refer to Palestinians as persons who fled or were driven from Palestine during the Arab–Israeli War of 1948 and the 1967 War (CIA Factbook 2009). The latest wave of refugees was from the Gulf area after the crisis in 1990 when a large number of people who had been working out of the country returned to Jordan. In addition, there has been a gradual demographic change from rural and nomadic to urban living (Al-Krenawi et al. 2000).

Jordanian families, like other Arab families, tend to be collectivist and patriarchal in nature. The predominant family structure places the family ahead of individuals (Al-Krenawi et al. 2000) and the male is considered the first authority and the head of the house. Obedience, respect and conformity to parents are learned from an early age. In Jordan, through the influence of modernization and urbanization, the traditional family structure is being challenged to transform into a more democratic model (Barakat 1993). Exposure to the media, industrial capitalism (Al-Krenawi et al. 2000), democratization, the liberalization of women, and paid employment for women are also challenging the structure of the Jordanian family (Lowrance 1998). Despite these changes, as with the rest of the Arabic world, Jordan is a traditional society and continues to hold a strong family orientation (Barakat 1993). Economically, poverty, inflation and unemployment are fundamental problems although recent broad reforms are expected to improve living standards in the long-term (CIA Factbook 2009). Research by the Jordanian government shows that upwards of 40% of women of marriageable age have not wed, including 100,000 women who are over the age of 30. Possible factors include men staying in school longer, and therefore delaying marriage; or men or their families concluding that they do not have sufficient wealth to marry yet.

Method

Sample

Sampling followed a purposive, snowball approach. The sample consisted of 199 women, 93 from polygamous families and 106 from monogamous families. All women from polygamous families were “senior”—that is, their husband’s first wives; and their husbands all had only two wives.

Data were collected from the following cities Ajlon, Salt, and Madaba, these cities represent the north, middle and south of Jordan. A snowball method of sampling was used, conducted by female undergraduate students from a local university trained to collect data according to culturally competent methods. In order to facilitate the research, the data collectors tended to be from, or near to, the cities in which data were collected. Questionnaires were structured; data collectors were present during the interview, completed the questionnaire forms with the respondent, and for those with limited reading or writing skills the interviewers read to the respondent and filled in the questionnaire according to the given responses. Data collection occurred from May to September 2009. The data collectors contacted the women prior to the interview and explained to them the goal of the study, the issue of confidentiality and that no identifying information would be used in this study. After they received the consent of the woman to participate in the study, the interview was conducted in a comfortable place and during the day time while she was alone with no disturbances. All respondents were told that their participation was voluntary and that they could withdraw their consent during the interview. Some respondents had limited reading or writing skills and in these instances the responses were filled out by an interviewer. Therefore some of the data are not self-rating scales but are interview rating scales; our analysis does not distinguish between these response types.

Research Instruments

The instruments were translated into Arabic, and back translated to ensure accuracy. It should be noted that most of these instruments were used in previous research with other Arab people in the Middle East (Al-Krenawi and Graham 2006b; Al-Krenawi and Slonim-Nevo 2008a). They assessed 6 areas, as follows:

Sociodemographic Variables

The variables were as follows: a woman’s age, her age at the time of marriage, the wife’s education, husband’s age when married, husband’s education, husband’s age, number of children, wife’s economic status, type of family (polygamous or monogamous marriage), blood relationships between the woman and her husband and her degree of agreement with polygamous marriage.

Family Function

We used the McMaster Family Assessment Device (FAD) as developed by Epstein and colleagues (Epstein et al. 1983; Miller et al. 1985). It includes 60 items on seven dimensions of family functioning: problem solving, communication, roles in the family, emotional involvement, behaviour control, emotional responses and general functioning. All subscales range from 1 to 4, with higher scores indicating more problems in a family’s functioning. Section points discriminating between ‘clinical’ and ‘normal’ families in American populations are available. The scale has satisfactory reliability (Cronbach’s alpha = 0.72–0.92), good test–retest reliability (r = 0.66) and high validity, as indicated by comparing the scale’s scores with other measures of the same matters (Epstein et al. 1983; Miller et al. 1985). A recent study (Ridenour et al. 1999) found that these 12 items give a satisfactory picture of a family’s general functioning, and there is no need to use all 60 questions. In the current study we used only the 12 items that assess the family’s general functioning. The reliability of the subscale was satisfactory (Cronbach’s alpha = 0.72).

Marital Satisfaction

We used the ENRICH questionnaire, whose original details were selected following a comprehensive overview of the literature on marital problems and interpersonal conflicts (Fournier and Olson 1986, as cited in Lavee et al. 1987). The questionnaire, which measures satisfaction with marriage and quality of marital adjustment, is divided into eight parts, each containing 10 items. Several studies (as cited in Lavee et al. 1987) found that it has a rather high reliability (Cronbach’s alpha = 0.88–0.89). Other studies indicated a high degree of discriminating validity and concurrent validity. Research that used the instrument in Arab society (Al-Krenawi and Graham 2006a) found a satisfactory level of internal reliability (Cronbach’s alpha = 0.96). In this survey, we used the shortened version of the ENRICH questionnaire composed by Lavee that includes 10 items, each rated on a Likert scale ranging from 1 (lesser satisfaction) to 5 (greater satisfaction). The internal reliability of the shortened version among the women in the current study is very high (Cronbach’s alpha = 0.89).

Symptoms Checklist (SCL-90)

The SCL-90 is a self-report questionnaire originally oriented towards symptomatic behaviour of psychiatric outpatients (Derogatis et al. 1973). It has since been applied as a psychiatric case-finding instrument, as a measure of symptom severity, and as a descriptive measure of psychopathology in different populations (Derogatis 2000). The SCL-90 is intended to measure symptom intensity on nine different subscales: somatization, interpersonal sensitivity, obsession-compulsion, depression, anxiety, hostility, phobic anxiety, paranoid ideation and psychoticism. The 90 items of the questionnaire are scored on a five-point Likert scale, indicating the rate of occurrence of the symptom during the time reference. The instrument’s global index of distress is the Global Severity Index (GSI), which is the mean value of all of the 90 items (Derogatis 2000). Reliabilities of the 9 subscales were high (Cronbach’s alpha = 0.79–0.89) as was the reliability of the GSI (Cronbach’s alpha = 0.98).

Self-Esteem

The Rosenberg (1979) Self-Esteem (SE) scale consists of 10 items, with ranges from 1 to 4, higher scores indicating higher self-esteem. It has high internal consistency (Gutman measurement of reconstruction 5.92) and high test–retest validity (r = .85). The scale was translated into Arabic. The Arabic version yielded a satisfactory level of internal consistency (Cronbach’s alpha = 0.76). It is adapted from the 1965 original scale.

Life Satisfaction (SWLS)

We used the Diener et al. (1985) scale, which consists of five items examining life satisfaction. It uses a Likert scale ranging from 1 (low) to 7 (high satisfaction); the scale has high internal reliability (Cronbach’s alpha = 0.87) and good stability examined by test–retest reliability (r = 0.82). Diener et al. (1985) tested the validity of the scale by comparing it with existing scales finding good validity. The internal reliability in the current research was satisfactory (Cronbach’s alpha = 0.74).

Results

Table 1 presents the demographic characteristics of participants from polygamous and monogamous families. Women from polygamous families, as well as their husbands, were older than women and husbands from monogamous families. However no significant difference was found regarding their age at marriage. Husbands’ ages at marriage in polygamous families were found to be older compared to monogamous families. No statistical difference in women’s or husbands’ education was found between polygamous and monogamous families. Participants reported more children in polygamous families. Monogamous families were more satisfied with their economic state. No difference was found between the two familial structures regarding the women’s blood relations to their husbands. In addition, the two familial structures differed in the women’s agreement with polygamy, with more women from monogamous families opposing polygamy.

Table 1 Sociodemographic characteristics of the sample
Full size table

Analyses of the differences between monogamous and polygamous families in family function, well being and mental health symptoms were conducted by t test analysis. The results are presented in Table 2. As demonstrated in Table 2, women from polygamous families experienced more problems in family functioning as well as more problems in their marital relations. Moreover, women from polygamous families reported low self-esteem and less satisfaction with life. In addition, women from polygamous families experienced more somatization, interpersonal sensitivity, depression, anxiety, hostility, paranoid ideation, psychoticism and their general severity index was higher (GSI).

Table 2 Familial structure differences in family function, well being and mental health symptoms
Full size table

Next, regression analysis was used in order to assess the effect of family structure on the various measures of family function, well-being and mental health symptoms, while controlling for sociodemographic variables. Family structure was entered as the independent measure while controlling for the effects of age, education and economic state. Regressions were conducted for each of the study’s dependent measures. Standardized effects of the independent variables and R-squares are presented in Table 3. Results supported the research hypothesis. Family structure was found to be a major predictor of family functioning, marital relationship, mental health symptomatology and low self-esteem. Specifically, it was demonstrated that women from polygamous families experienced more problems in family functioning and more marital problems compared to their monogamous counterparts. Polygamous women were also less satisfied with their life, although this finding was only marginally significant. Following this line of results, polygamous women reported low self-esteem and experienced more somatization, interpersonal sensitivity, depression, anxiety, hostility and psychoticism. In addition the general severity index (GSI) of polygamous women was higher than the GSI of monogamous women indicating that polygamous women experienced more mental health symptoms.

Table 3 Family structure and sociodemographic variables as predictors of the study’s dependent measures: standardized regression effect and R-square
Full size table

Discussion

There are several obvious limitations to the study. As exploratory research, the sample is necessarily small, and the sampling was snowball convenience. And so the generalizations are thereby limited. But at the same time, the research provides new inroads on which further scholarship and practice insights can be built. Between senior wives in a polygamous marriage, and wives in a monogamous marriage, there are significant differences in family functioning, marital satisfaction, and to a lesser but still significant extent, self-esteem and life satisfaction; in these instances, as the data show, women in polygamous marriages tend to fare worse than their counterparts in monogamous marriages. Likewise, as Tables 2 and 3 indicate, many of the mental health symptoms were different; particularly noteworthy were depression, hostility, psychoticism, somatization, and the GSI. Again, women in polygamous marriages tend to fare worse than their counterparts in monogamous marriages.

How to understand the experiences of our female respondents? Previous scholarship—much of it on Arab communities outside of Jordan—provides some clues. The economics of polygamy are particularly problematic. The modest level of Jordanian economic development no doubt exacerbates the potential for economic problems amongst polygamous families—families who are already vulnerable to economic problems. Even in the oil rich Persian Gulf region, for instance, Al Toniji (2001) found that 75 percent of participants agreed that polygamist husbands faced economic problems because of having to pay for two houses. However, there are demographic imperatives that sometimes encourage the practice (Al-Darmeki 2001). Amongst Palestinian polygamous marriages, there is compelling evidence that senior wives are often less favoured by their husbands, have fewer economic resources, and receive less conjugal support and attention than junior wives. Within that society, senior wives are typically married to men through arranged marriages based on exchange, while subsequent, junior wives are chosen by husbands, and their marriages are based on love matches (Al-Krenawi et al. 2001, 2002). Amongst Jordanian communities, future research could certainly explore these possibilities.

Further secondary literature in the Arab world provides deeper insight into the possible contexts of Jordanian female respondents’ lives. We interpret our findings on women’s somatisation, GSI, and like complaints in a broader cultural context. As previous research outside of Jordan points out, Arab women are prone to experience depression, anxiety disorders, somatization, and eating disorders (Hamdan 2009; Hamdan et al. 2008). In the present research, respondents in polygamous marriages were all first, or senior wives; this factor bears particular emphasis. Research in the United Arab Emirates, Kuwait, Egypt, Jordan, Palestine and with Arabs in Israel showed that first wives in polygamous marriages reported psychosocial, familial and economic problems (Al-Darmeki 2001; Al-Krenawi and Slonim-Nevo 2008a, b; Al-Shamsi and Fulcher 2005; Al-Sherbiny 2005). First wives in polygamous marriages in Bedouin-Arab society in the Negev suffered more than monogamous wives from low self-esteem, loneliness, and other emotional problems (Al-Krenawi 2001). Outside of the Arab world: a study conducted in rural Cameroon (Gwanfogbe et al. 1997) revealed that junior wives were more satisfied with their marriage than were senior wives. Chaleby (1987) points out that in the psychiatric service in Kuwait there are more senior than junior wives under psychiatric treatment. In another study, Chaleby (1985) observes that senior wives relate their psychiatric symptoms to their husbands’ subsequent marriages.

The next question to consider is the cultural context of somatisation, which is one of Arab women’s major ways of conveying psychological distress (Al-Issa 2000; Al-Krenawi 2001). Previous research confirms that senior wives in polygamous marriages may exhibit body aches, headaches, insomnia, fatigue, and nervousness (Al-Krenawi and Graham 2006b). When a Jordanian woman expresses somatic and related psychological complaints, the practice of polygamy may be a causal factor. Likewise, somatization may be evidence that a variety of psychological problems could be underlying. Particular means through which somatization is conveyed varies across and within communities and cultures; the decoding process is therefore vital (Al-Krenawi and Graham 2009). It is essential for practitioners to be able to recognize and decode these symptoms, particularly in relation to the underlying possibility of polygamous family structure as an implicating factor. Likewise, particularly when they are working in cross-cultural contacts, with recent immigrant families, or both; clinicians could usefully ask the marital status and the attitudes of the patient towards that marital status. It is essential to determine whether the patient, or family members associated with the patient, may have or have had a polygamous marriage and had to choose one wife to bring as the official legal wife. The effects of being chosen and those on being left behind should be made explicit for mental health treatment.

From a sociological perspective, women’s self-conceptions are also an important potential vehicle for understanding the present findings. Previous scholarship on Arab communities shows that gender constructions of women as self-sacrificing wives and mothers who do not complain may in turn exacerbate the likelihood of the sorts of symptomatology revealed in our study (Abdu Salaam 1997). Women are frequently not consulted when a man opts to assume a junior wife; the powerlessness of that lack of choice, and the possibility of fewer familial, social and economic resources can be distressing (Al-Darmeki 2001; Al-Krenawi and Graham 1999). In Egypt, Philips (2001) found that while permission is required from the first wife, few women actually give their husbands permission to marry a second wife. In Kuwait, many men married again without consulting or telling their wives, and roughly half from a recent survey did not agree with telling their wives before remarrying (Abdu Salaam 1997).

The present research points out some concerning prospects both for women in polygamous marriages as well as for the societal construction and expectation of women. In Table 1, a little more than one-third of polygamous respondents, and 12 percent of monogamous respondents, agreed that under certain circumstances or no circumstances at all, polygamous marriage could occur. That there were differential responses here suggests a different extent of legitimation. Particularly in more traditional, collectivist societies such as Jordan, community persuasion may be a strong factor in people’s acceptance of practices that are nonetheless problematic (Al-Krenawi and Graham 1999). Likewise those women who practice it may seek to legitimize polygamy as a way of coping with the associated problems in their lives. Moreover, the notion of self-sacrifice has a cultural dynamic, and the need to maintain a relationship for the sake of the children is a significant motivator for many women (Al-Krenawi and Graham 2009). While not examined in the present study, there is a potential relationship between socioeconomic status and attitudes towards polygamy. Some practitioners speak of a “first wife syndrome” that reflects the risks that our previous studies point out and our research confirms (Al-Sherbiny 2005). Khadijah (2002) found that most women in Egypt did not accept polygamy and those who did were generally from poor families.

Conclusion

A number of points bear emphasis. First is the need for practitioners and policymakers to appreciate the significance of polygamy to family structures and the nature of somatic and other forms of complaints as potential indicators of a problematic polygamous relationship. Second is the stress that polygamy places upon senior wives, and therefore the prospects for negative role modeling to children and other community members is possible. Policymakers in education, health, and social services need to be aware of the psychosocial and related needs of women and children in polygamous relationships. The findings may also motivate those community leaders who sanction or encourage the practice to investigate polygamy further. As the results point out, family functioning may be lower and marital distress higher in a polygamous family, which may in turn exacerbate negative role modeling and impede children’s growth and achievements. Thus polygamous families need to be understood longitudinally within the context of multigenerations. The present manuscript gives voice to women—who are frequently unheard and whose views are often overlooked. In Jordan, as with many countries that practice polygamy, there are, indeed, profound issues of gender construction that are embedded within the practice (Mernissi 1994). But as previous research shows, the appreciation of rights for women leads to many social improvements in areas of demographic governance, academic achievement, and the like (Boserup 2007). The future of any society is dependent upon its children. The well being of mothers and their families is crucial to Jordan, as it is to any society; may the present manuscript contribute to families’ improvements in this country and in any other wherein polygamy has been practiced.