Vulnerability to violence of girls of the street in Mauritania

Children and Youth Services Review 33 (2011) 656–662

Contents lists available at ScienceDirect

Children and Youth Services Review j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c h i l d yo u t h

Vulnerability to violence of girls of the street in Mauritania Jérôme Ballet a, Nicolas Sirven b, Augendra Bhukuth c,⁎, Sophie Rousseau c a b c

FREE & IRD Madagascar, 101 Antananarivo, BP 434, Madagascar IRDES, 10 rue Vauvenargues, 75018 Paris, France FREE & University of Versailles Saint-Quentin-en-Yvelines, 47, Boulevard Vauban, 78047 Guyancourt Cedex, France

a r t i c l e

i n f o

Article history: Received 9 April 2010 Received in revised form 7 November 2010 Accepted 8 November 2010 Available online 11 November 2010 Keywords: Girls of the street Street children Vulnerability Mauritania

a b s t r a c t This article is concerned with an empirical study of girls of the street in Mauritania. This study is original for three main reasons. First, it investigates Mauritania, a country where there have been very few studies of the phenomenon of children of the street. Secondly, it seems to us that though the academic literature on street children sometimes provides comparisons between girls and boys, very few studies focus specifically on girls of the street. Thirdly, the literature focusing on children of the street generally elucidates the mechanisms that lead some young people to live on the streets. On the contrary to previous literature, our study does not directly focus on mechanisms that lead some young people to live on the streets. Our study focuses on the question of the socio-economic determinants of the risk of violence. In other words, whereas the literature, following the hypothesis of aberrant families, uses violence as an explanatory factor for the phenomenon of children of the street, we attempt here to highlight the explanatory factors for the risk of violence and we treat this risk as an explained variable. A novel finding emerges from this study: there is less risk of violence facing girls of the street if the mother does not work. Conversely, the risk of violence is greater if the mother works and the father stays at home. © 2010 Elsevier Ltd. All rights reserved.

1. Introduction Everywhere in the world, but especially in developing countries, street children constitute a population particularly affected by deprivation and absolute poverty (Bartlett, Satterthwaite, De La Barra, & Missair, 1999; Gordon, Pantazis, Pemberton, & Townsend, 2003). The international community became aware of the phenomenon of street children, in 1979 (International Year of the Child), and child-specific human rights were adopted in 1989 (U.N. Convention on the Rights of the Child). Improving the condition of children thus came to be an important principle of the Millennium Development Goals in the year 2000, and is now incorporated into most international development strategies. Despite this, the population of street children continues to grow, in particular in developing countries (Kerfoot et al., 2007). UNICEF (2006) estimates the number of street children to be between 100 and 120 million in developing countries. Obviously these figures are not to be taken at face value, and this for various reasons. Firstly, it is not easy to estimate the phenomenon accurately. Secondly, estimates are a way for international agencies (mainly UNICEF) to draw the attention of people and political decision-makers to their activities. The figures used to support the phenomenon are symbolic rather than accurate (Ennew, 2000).

⁎ Corresponding author. E-mail address: [email protected] (A. Bhukuth). 0190-7409/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.childyouth.2010.11.004

Thirdly, the expression “street children” has several different meanings. The general notion of “street children” encompasses both working children living at home and street children who work in the streets but do not live with adults (Aptekar, 1994). Therefore the estimate can have different values depending on the meaning adopted. According to UNICEF (1984), children on the street (also sometimes called street-working children) include categories of children who spend a considerable part of their day on the streets, but who still have numerous connections with their family as well as a home to return to, especially for meals and rest periods. The distance separating these children from their families is quite relative. Moreover, the street phenomenon represents an important aspect of these children's expression and activity. For them, the street constitutes a living space in which they spend time, though it cannot be regarded as their sole living space. Children of the street (also sometimes called street-living children), on the other hand, refers to categories of children for whom the street is not only a living space, but also an existential space in the sense that it constitutes either their only living space or an essentially dominant space in which their life is entirely constructed. Relationships with the family are either nonexistent—which is generally the case—or sporadic and do not provide for these children's subsistence. So, while the situation of children of the street is connected with a social marginalisation, that of children on the street may be interpreted as being vulnerable with regard to marginalisation (Ballet, 2006). Between children on the street and children of the street a third category is generally considered, those of children of street

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

families. It is an intermediary category since children work on the street, have regular relations with their family, but at the same time, street constitutes an existential space (Aptekar, 1994; Lusk, 1989). Nevertheless this dichotomy between children on the street and children of the street arose out of the policy literature and is far from being an accurately defined categorisation. According to Hall and Montgomerry (2000) the dichotomy is not neutral, and is used as much to alert the public to the living conditions of such individuals as to describe their stages of life. However, according to Lalor (1999), making a distinction between children on the street and children of the street sheds light on the factors that drive children to migrate onto the streets. Furthermore, Panter-Brick (2003) considers that classifying these children is useful as long as the categories are not perceived as discrete and homogeneous. The use of the term “street children” certainly poses problems, but few alternatives are available—apart from local terms—to refer to this particular group of children (Ennew, 2000; Panter-Brick, 2003). Being aware of the non-homogeneity of categories, in this paper we look at girls of the street. Indeed, in our case study, girls living in the street have only sporadic relationships with their family. We attempt to analyze the factors that increase the risk of violence in the family; in other words, we will analyze the vulnerability of girls of the street to family violence. We use "vulnerability" in the meaning given to it by the World Bank as "the likelihood that a shock will result in reduced wellbeing" (World Bank, 2001, p.139). The shock considered here is family violence. Then, following Hutz and Koller's (1999) suggestion, a risk-related dimension is introduced. In this study, we propose to use data gathered in Mauritania in 2004 among an exclusively female population of children of the street. As the subjects being studied are girls of the street, the data concern past relationships within the family. This study is original for three main reasons. First, it investigates Mauritania, a country where there have been very few studies of the phenomenon of children of the street (Ballet, 2006). Secondly, it seems to us that though the academic literature on street children sometimes provide comparisons between girls and boys (Aptekar & Ciano-Federoff, 1999; Conticini & Hulme, 2007; Raffaelli et al., 1993; Rizzini & Lusk, 1995), very few studies focus specifically on girls of the street (see also Lalor, 1997). Thirdly, the literature focusing on children of the street generally elucidate the mechanisms that lead some young people to live on the streets (In the African context, see for instance Aderinto, 2000; Aptekar, 1994; Beyene & Berhane, 1997; Bourdillon, 1994; Ebigbo, 2003; Ennew, 2003; Kilbride, Suda, & Njeru, 2000; Le Roux, 1996; Mambwe, 1997; Marguerat & Poitou, 1994; Matchinda, 1999; Muchini & Nyandiya-Bundy, 1991; Olley, 2006; Sampa, 1997; Taçon & Lungwangwa, 1992; Veale, 1996; Veale & Donà, 2003). Two competing explanations are often discussed. The most widespread explanation is that children are urged by their family to work on the streets in order to support themselves, or even provide the household with a supplementary income (Aderinto, 2000; Alexandrescu, 1996; Aptekar, 1988; Basu & Van, 1998; ILO, 1996; Lusk, 1992; Olley, 2006; Peacock, 1994; Rizzini & Lusk, 1995 for instance). Then, children on the street gradually become children of the street. However, the part played by economic poverty has been challenged (see for instance Aptekar, 1988; Blanc, 1994; Conticini & Hulme, 2007; Lalor, 1999; Marguerat, 1999; UNICEF, 2000). The studies challenging the economic poverty factor can be grouped under the “aberrant families” hypothesis (family disintegration, family conflict, alienation from parents, abandonment, neglect, physical abuse, or incestuous relations) (Aptekar, 1994). On the contrary to previous literature, our study does not directly focus on mechanisms that lead some young people to live on the streets. Our study focuses on the question of the socio-economic determinants of the risk of violence. In other words, whereas the literature, following the hypothesis of aberrant families, uses violence as an explanatory factor for the phenomenon of children of the street,

657

we attempt here to highlight the explanatory factors for the risk of violence and we treat this risk as an explained variable. Then we do not attribute causality to family violence for explaining girls of the street phenomenon. We take for granted the causal mechanism. On the contrary we consider family violence as an explained variable. In other words we analyze factors contributing to violence. The article is structured as follows: after the Introduction, Section 2 presents the Mauritanian context and the method used to gather data. The ensuing statistical results are expounded in Section 3. First we report the results using descriptive statistics. Amongst other things, these results are intended to find out whether the girls of the street have been the victims of violence. We then carry out a multiple correspondence analysis in order to identify the risk factors for violence. Section 4 draws the conclusions from the study, and proposes a few guidelines for public policy. 2. Data overview In a seminal paper on the analysis of street children in the developing world, Aptekar (1994) warns about the methodological problems that may be encountered when studying this phenomenon. This author suggested providing the reader with a contextual analysis of the data (explaining how the data were gathered and how the methodological problems were overcome). Of course, the need for contextual analysis is not specific to studies on street children, but it appears very crucial in this case. Ballet (2006) shows for instance that responses to a questionnaire vary depending whether they are coming from children, families or educators. In this section we will focus especially on the context, and the method used to gather data. 2.1. The Mauritanian context For the past forty years or so, the Mauritanian context has been marked by a very strong migratory movement towards the cities. The 1961–1962 census thus revealed that the nomadic population represented 76.9% of the whole population, compared to 13.3% for the sedentary rural population, and 9.8% for the urban population. In 1977, the proportions were of 33.2% for the nomadic population, 44.1% for the sedentary rural population and 22.7% for the urban population. By 1998, the nomadic population only corresponded to 5% of the population, while the urban population accounted for 53% (Ignegongba, 1992; ODHD, 2000).1 This means that the settling down process in rural areas occurs alongside a relatively new but huge phenomenon of urbanization. Furthermore, there is a large foreign population from neighboring countries migrating into Mauritania either to find a job (for example in the fishing sector), or who pass though to get to Europe. Migration flows into the big cities are therefore strong. Migration has been presented as a factor that may lead to the existence of street children in Africa (Mambwe, 1997; Sampa, 1997), and is often advanced in the local context to explain the phenomenon (Mohamed Salem, 1996). For instance, the change in life style of people who were formerly nomads, or the arrival of foreigners from neighboring countries can explain the phenomenon of children living in the streets with their families, due to the difficulty of finding accommodation and employment. However, the mere fact of living in the street does not mean that children necessarily become children of the street. The increased migration of nomadic populations may increase the phenomenon children on the street, but not necessarily that of children of the street. If, for example, we cannot rule out the possibility that certain formerly nomadic populations find it difficult to change their way of life, no evidence has established any link between migration and the presence of children of the street in that country. On the contrary, we are entitled to doubt the reality of this relationship. In particular, the analysis of the children's ethnic origins 1

The new census has only just been completed and the data are not yet available.

658

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

and family origins does not allow us to conclude that there is a link between the phenomena of migration and children of the street. First, the children of the street come from very diverse ethnic origins, which are less true of the nomadic population; second, the families of these children have often been living in the town for a fairly long time (Ballet, 2006). In fact, we should point out that the studies establishing a link between migration and children of the street do not make any distinction between children on the street and children of the street, which leads to some confusion between the categories of children. In addition, Mauritania is an Islamic republic where the role of religion is preponderant in the organization of society. We should emphasize here, for example, that few of the children of the street are orphans. Indeed, the Muslim religion and traditions in Mauritania provide for a certain number of mechanisms intended to take charge of orphans. Sharia law (Islamic law) lays down strict codes for the care of orphans. In particular, three aspects have been identified in particular: hadhana which is concerned with their affective acceptance, nefaka which concerns providing them with accommodation, food and clothing, and wilaye, which concerns the moral responsibility of the guardian (Lafdal & Sow, 2006). The lot of these children may be deplorable and call for special care, but they do not automatically swell the numbers of children of the street. The Muslim religion has another important role to play in the children of the street phenomenon. In the case of Ethiopia, Veale (1996) reveals that there are far fewer street-living girls in the areas where the Muslim population is predominant. There seems to be inverse relationship between the number of street-living girls and the size of the Muslim population. One reason could be the very strong control that girls are subjected to in those countries. However, the resonance of religion in Mauritania is such that the recognition of girls of the street is a critical issue. The National Council for Children (the highest authority concerning children issues) first acknowledged their existence in 2003. But the identification of girls of the street is also difficult for other reasons. One of the main reasons is that the apparently small number of girls of the streets results from the fact that it is difficult to recognize them. Girls, find shelter for the night more often than boys, mainly because they are more frequently picked up by adults who either abuse them directly in return for accommodation, or who hire them out to brothels or prostitution networks. So, on the whole, girls of the street tend to be less visible, which does not mean that they do not exist. 2.2. Data collection It is difficult to gather data on children of the street in Mauritania, mainly on account of the small number of children of the street there, compared to other countries. Furthermore, no census of children of the street exists in this country. And as census are generally led in family they fail to cover children of the street phenomenon. The parties involved in the field consider that the number of children of the street (as defined above) barely exceeds a few hundreds. By way of comparison, the number of children of the street is, for example, estimated to be more than 10,000 in Uganda (Witter, 2002) and about 500,000 in the larger cities of Bangladesh (ARISE, 2001). The small number of children of the street in Mauritania does not make this a phenomenon of any less importance in a country where the total population is estimated to be less than three million inhabitants. Another problem arises when we look at an exclusively female population of children of the street: boys constitute the majority of children of the street everywhere (Aptekar, 1994; Lalor, 1999). And, as we have already mentioned, Mauritania is no exception in this regard. Two main reasons are generally put forward to explain this under-representation of girls. First, the families who urge their children to search for a supplementary source of income on the streets prefer girls to stay at home in order to help with the housework. Girls are more useful in the home (Connolly, 1990), and

working on the streets involves more risks for them than for boys (Chatterjee, 1992; Veale, 1996). Then, boys are more likely to gradually become children of the street. Secondly, many street-living girls are not counted either because they are involved in occupations that are not very visible (working in restaurant kitchens or as domestic servants for instance), or because they are put into the same category as the prostitutes and, as such, are no longer regarded as children of the street (Aptekar, 1988). This explains why one of the few studies conducted on this topic (Lalor, 1997) is only based on a small number of observations (69 street girls in Addis Ababa, in Ethiopia). To allow us to observe the phenomenon of girls of the street, data were gathered at the two reception centers of a NGO devoted to helping destitute children—the Association Enfance et Développement en Mauritanie (AEDM), in the two largest cities in the country: Nouakchott, which is the capital, and Nouadhibou, the main financial center. The protocol and consent process of the study were defined with the president of AEDM, and with the participation of social workers of the NGO. In total, the sample examined consists of 77 girls of the street, 32 in Nouakchott and 45 in Nouadhibou. The data gathered concerns with the girls’ living conditions when they were in their family, their past relationships with their family, and past family characteristics, such as the parents’ occupation, their marital status, etc. The small size of the sample investigated could seem to be problematic. However, as we have pointed out previously, the number of children of the street in Mauritania barely exceeds a few hundred, and girls form a minority. The 77 girls who make up the sample in fact constitute most of the girls of the street listed in these two cities by the NGO. The other towns in the country are considerably smaller and the phenomenon of children of the street is limited (except in the town of Atar, where this phenomenon is beginning to develop according to information provided by people involved in this problem in the field). Our sample is of course not exhaustive, therefore appears at least to be representative of the phenomenon during the period covered by the survey, at least in regard of girls of the street identified by the NGO. Of course, first, the NGO has certainly failed in indentifying all girls in the street (for reasons previously mentioned); second, the involvement of respondent girls in the NGO programme may have had potential influence on the sample selection and then findings. Girls are firstly identified by the NGO workers as potential girls of the street, using usual methods (identification of girls sleeping in the streets, begging and working in the street without their family…). Secondly, a selection is done between all girls identified through discussions with them. Only girls of the street are retained for the third step. Then, thirdly, the NGO workers gradually talk girls into engaging within the NGO program. Of course, the process may let aside girls that have not been identified as girls of the street. Meanwhile, we should note that the NGO is working in the field for more than 20 years and has a great experience of the phenomenon. We should also note that the information collected is the fruit of much hard work by the educators working in these reception centers with a view to (i) establishing the children's family origin and finding their parents and ii) getting in touch with the parents with the aim of restoring links with their children. We are of the view that this fundamental work, which has contributed towards developing links of confidence over time, guarantees the quality of the information used in the present study. 3. Empirical findings 3.1. General descriptive statistics The 77 girls in the sample were asked about their past living conditions: did they suffer from affective assaults? Was violence involved? If violence was involved, were the affective assaults

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

combined with any form of economic exploitation? By affective assaults, we mean physical violence and moral and psychological harassment. We have divided affective assaults into three categories. The first category involves moral and psychological harassment, the fact of being ill-treated by ones family even if no physical violence is perpetrated. This is of course a subjective assessment by the children, based on whether they feel that they have been rejected by their family or not. The second category also includes physical violence. Once again, we base our study on what the children say, and so on their subjective views. The fact of being "smacked" may, for instance, be interpreted as physical violence by some of them, but not by others. However, we did try to define a degree of physical violence that is not only an educational act considered to be normal within Mauritanian population as a whole. Thus, when we described children as having undergone physical violence category, this meant that they had received violent blows that have left marks and injuries, and not simply that they had been "smacked" from time to time. For these children, family relationships had been considerably affected by regular violent acts. The third category adds a dimension of economic exploitation. By this we mean that the family used the child to earn money. However, this does not involve just working with the rest of the family, which could rather be seen as an act of socialization and apprenticeship, but to a used that the children perceive as being an abuse. The children assigned to this category said that they had to bring enough money back to the families, otherwise they were beaten violently. They therefore constitute a specific category, distinguished from the second category, because here the violence is intimately linked to child labor, which was not the case in the previous category. Other questions are also related to the parents’ characteristics: did the father work? If so, what kind of work did he do? Did the mother work? What kind of work did she do? Were the parents divorced or separated? Was the father absent from the household? All the data have been collected by educators who were caring for the children in the two centers mentioned above. This information was not obtained during a single interview, but a set of information collected over a long period of work, often lasting several months, during which the children tell the educators about their problems. This information was completed by information about the parents obtained by the educators. All the information was reported in order to evaluate the work of the centers during the year 2003. We subsequently clarified the data with the educators and the children. Of all the information available, a total of six qualitative variables (including a dummy variable for each city) are retained here: a) children's living conditions (types of affective assaults), b) family situation (parents living together, divorced, and deceased), c) presence or absence of the father in the household, d) father's employment, e) mother's employment, f) the dummy variable for each city. Two details should be mentioned concerning the employment of the father and of the mother. For both parents, there was such a wide variety of the work reported that we pooled all categories of work apart from two: the fact that the father was a soldier, or the mother a prostitute (we did not make any moral judgment about the mother's work, and so we included prostitution amongst the statistical category of jobs). The qualitative variable of the father's job it therefore classified as one of three possible categories: unemployed, employed, and soldier. Similarly, the mother was classified as: unemployed, employed, and prostitute. We used three specific categories because the language (used by politicians, social workers and the children of the street themselves) about children of the street in Mauritania highlights the part played by fathers who are soldiers and abandon their children, and prostitute mothers who also abandon them. The language used has probably been influenced considerably by a previous report (Lefort, 1988), compiled for UNICEF and for the Mauritanian Ministry of Social Affairs, about twenty years ago, that excessively highlighted the role of soldier fathers and prostitute mothers. Thus, while keeping these two categories, our

659

Table 1 Descriptive statistics. Total N Children's living conditions No affective assaults Affective assaults With violence With violence and exploitation Parents Living together Divorced Deceased Father in the household Present Absent Father's employment Unemployed Soldier Other Mother's employment Unemployed Prostitute Other

77 14 35 13 15 76 24 39 13 76 9 67 65 19 12 34 72 36 6 30

Nouakchott

Rel. freq. N 0.182 0.455 0.169 0.195 0.316 0.513 0.171 0.118 0.882 0.292 0.185 0.523 0.500 0.083 0.417

32 0 13 9 10 32 15 15 2 32 5 27 32 13 3 16 30 7 0 23

Nouadhibou

Rel. freq. N 0.000 0.406 0.281 0.313 0.469 0.469 0.063 0.156 0.844 0.406 0.094 0.500 0.233 0.000 0.767

45 14 22 4 5 44 9 24 11 44 4 40 33 6 9 18 42 29 6 7

Rel. freq. 0.311 0.489 0.089 0.111 0.205 0.545 0.250 0.091 0.909 0.182 0.273 0.545 0.690 0.143 0.167

study gives a clearer picture of the role of these two determinants in the children of the street phenomenon in Mauritania. Table 1 provides some simple descriptive statistics. The general trend in the sample indicates that over 80% of the girls suffered from affective assaults, and that there were indeed three categories of different kinds of phenomena as mentioned above. We therefore use the term “affective assaults” to distinguish all forms of violence (whether physical or not), from purely economic factors such as poverty. This phenomenon is particularly important in Nouakchott, where every child surveyed reported undergoing affective assaults. In contrast, Nouadhibou girls seemed to be less subjected to affective assaults, with 31.1% of the population not complaining about any affective assaults. This difference between the two cities raises an interesting question: why are the children's living conditions so different from one place to another?2 A closer look at the other variables may shed light on some possible causes. First of all, it appears that the proportion of girl children whose parents were separated is higher in Nouadhibou (79.5%) than in Nouakchott (46.9%). These two figures are both quite high, and should be put into perspective by remembering that, in most cases (88.2%), the father was not present in the household. As a consequence, we may say that women were usually the parent with whom girls had links. A second main difference between the two cities is that 69% of the mothers of Nouadhibou girls were unemployed, and 14.3% worked as prostitutes (of course this percentage is not significant since it represents a very small number of mothers), whereas none of the mothers of Nouakchott girls was a prostitute, and 76.7% of them had an occupation. These preliminary findings lead us to an observation and raise a question. The observation is that in contrast to what one tends to hear in the country, soldier fathers and prostitute mothers play no significant role. The question which then comes to mind is: are working mothers less able to protect their children from affective assaults, and if so, why would this be so?

2 Of course we could envisage that this was due to selection bias due to the fact that the two ONG centers do not work in the same way. However, we should not forget that the girls interviewed correspond to most of the girls of the street in Mauritania. However, the Nouakchott center may have a wider concept of what constitutes girls of the street. However, we were not able to detect any significant differences in how the centers identify these girls or how they work. A possible explanation could reside in the differing economic contexts of the two cities. In particular, Nouadhibou is marked by the fact that its economic activity is essentially centered on fishing and port activities, whereas Nouakchott has a wider range of work.

660

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

Table 2 Multiple correspondence analysis (MCA) summary. Factor Eigenvalue Percent of variance Cumulative percent F1 Children's living conditions No affective disorder Affective disorders With violence With violence and exploitation Parents Living together Divorced Deceased Father in the household Present Absent Father's employment Unemployed Soldier Other Mother's employment Unemployed Prostitute Other City Nouadhibou Nouakchott

F1 0.390 20.296 20.296 Coord.

F2 0.281 14.646 34.942 QLT

F3 0.279 14.545 49.487 t-stat

−0.877 0.161 −0.049 0.286

0.172 0.021 0.000 0.020

−3.605* 1.285 −0.191 1.228

0.913 −0.382 −0.750

0.397 0.166 0.108

5.357* −3.374* −2.948*

1.169 −0.198

0.186 0.246

3.707* −4.464*

0.381 −1.088 0.389

0.050 0.224 0.126

1.899 −4.076* 3.015*

−0.630 −1.078 0.881

0.349 0.099 0.527

−5.145* −2.732* 6.134*

0.869 −0.684

0.578 0.595

6.387* −7.068*

F4 0.202 10.512 59.999 CTR (%) 7.495 6.237 0.528 0.018 0.712 19.119 11.588 3.295 4.236 8.636 7.118 1.518 12.799 1.593 8.228 2.978 25.783 8.270 4.037 13.475 26.168 13.987 12.181

*Significant at b 5%.

3.2. Multiple correspondence analysis In order to give an overview of the data available, we used a Multiple Correspondence Analysis (MCA), which sheds light on the various relationships between the variables. Table 2 provides a summary of the results, and Fig. 1 below proposes a more userfriendly presentation of the findings. The categorization of the variables as suggested by the MCA confirms the previous observations: the two cities are quite different insofar as they lie on different sides of the F1 axis. Nouakchott girls in the sample generally come from relatively close-knit households (Parents Together) where the mother worked and the father was present.3 The father's presence here has a strong influence on the children's well-being, since it is systematically associated with affective assaults (despite the small number of cases in which the father was present compared to the whole of the population surveyed, this factor was significant to the 5% level for violence in Nouakchott). These assaults also occur if the father worked, but are aggravated (by violence and economic exploitation) if he was unemployed. Nouadhibou girls in the sample generally come from single-parent households (parents divorced or deceased, father soldier and/or absent) where the mother was more likely to be unemployed. What is striking is that being a child whose mother was unemployed seems to be associated with a relative absence of affective assaults. These relationships are confirmed in Table 3, which clearly shows that the odds ratio associated with Nouadhibou girls is significantly more than 1 for “Separated Parents”, whereas it is of 7.33 for “Unemployed Mother”. Table 4 shows the results of the hypothesis test, in which unemployed mothers were expected to be more likely to protect their children from affective assaults. Among the four variables that are suspected of influencing the children's exposure to affective 3 We should note here that father is the biological father. Even if divorced parents are frequent in Mauritania, due to the pressure of religion, re-marriage is an exception.

assaults (separated parents, missing father, unemployed father, and unemployed mother), only the last is significant. We may therefore say that the fact that the children's parents were separated do not affect their well-being, since there was at least one parent (usually the mother) who took care of the children—and fathers were generally absent anyway (see Table 1). We should point out here that the father's absence also explains why whether he was unemployed or does not make much difference. So, it appears that girls with an unemployed mother are less likely to be subjected to affective assaults (OR = 0.118 b 1). 4. Concluding remarks Our study of the risks of violence faced by girls of the street in Mauritania raises three questions, in comparison to the existing literature. Firstly, family composition appears important, especially family unity and the father's presence. Girls from close-knit families in which both father and mother were present were more likely to be subjected to more forms of violence than the others. These results call into question those that are usually reported about street-living children in Africa. For example Moberly (1999) on Angola, Onyango, Sudaa, and Orwa (1991) on Kenya, Lalor (1999) on Ethiopia, have all found that most street children tend to come from single-parent families headed by the mother, which can explain that children are neglected. Most of the fathers were absent in our study, which was consistent with previous results, but it was still clear that their presence significantly increased the risk of violence. Secondly, to understand this finding, it is necessary to look at the mother's status and, to a lesser extent, at the father's status. In families in which both parents were present and both worked, girls were neglected or subjected to violence. In fact, when the mother worked, we can assume that she had less time to look after her children. The fathers seem to take relatively little responsibility for their children's education. This means that when the father was present in the home and did not work, girls were more often the victims of acts of violence and economic exploitation. The mother's role is therefore crucial, in that her availability in the household certainly limits violence against girl children. So, girls who undergo the least violence, usually come from households where the mother was unemployed, and was thus in a position to look after her children. For instance in households where the mother was a street vendor, and the father was a craftsman the mother was away a great part of the day and more specifically late in the evening till the sunset, which is when people buy food for the evening meal. In contrast the father has finished his work in the evening. The mother was not present at home, and the children had to fend for themselves, because the father did not take good care of them. Since the mother was absent, violence was all the more probable. It then appears that fathers are the perpetrators of violence. Nevertheless, it should also be noted that some working-class families (mothers and fathers) still chain up their children to ensure that they stay at home. The more unruly the children are or if they want to leave the house, the more likely they are to be chained up, and this is obviously also more frequent if the parents are absent. Thirdly, the results presented above are to be linked to the question of the mother's allocation of time, which may produce some unexpected results. Thus, paradoxically, girls who did not undergo any violence tended to have mothers who were prostitutes. We would, of course, have expected that the mother's prostitution would not be conducive to reducing violence. The issue of the mother's availability certainly explains this paradox. On the whole, mothers who were prostitutes had more flexible working hours than those who were shopkeepers, laundresses, etc. As a result, they were in a position to devote more time to their children. Furthermore, the fact that they alone were in charge of their daughters explains why the latter were protected from father's violence. Thus, in contrast to the

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

661

Prostitute Mother

-- axis F2-->

Disorders without Violence Working Father Divorced Parents

Nouadhibou Soldier Father

Deceased Parents

Affective Disorders & Violence

Working Mother

Nouakchott

Missing Father

Parents Together

Father Present Unemployed Mother No Affective Disorder

Unemployed Father Violence & Exploitation

-- axis F1 --> Fig. 1. Multiple correspondence analysis.

typical case mentioned above with a mother who is a street vendor, and the father a workman or a craftsman whose children are subject to violence, prostitutes have more time and work from home. They receive their customers at home, and between two customers they have time to do housework and look after their children. In our study, all the mothers who were prostitutes were in fact their own bosses and worked at home. This phenomenon seems to be the rule in Mauritania, where prostitution is not organized in networks by agents who exploit girls, but in groups of self-organized and relatively united women (Sall, 2006). Nevertheless, two questions remain after this study. Firstly, Conticini and Hulme (2007) highlight that in the case of Bangladesh, boys are more often subjected to a higher degree of violence than girls. Girls, however, are more often the victims of (i) sexual violence and (ii) several simultaneous forms of violence. Kabeberi (1990), Korbin (1981), Aptekar and Ciano-Federoff (1999) for instance point out that girls are generally more likely to be abused than boys. Ballet (2006) also emphasizes the major role of violence and emotional deprivation in Mauritania where, on the whole, boys are subjected to greater violence than girls. However, in this last study, in contrast to previous studies, girls did not appear to be particularly exposed to sexual violence. This appears to be attributable to the strong presence of Muslim religion in the country which, even though it accepts discrimination against girls, does condemn sexual

violence. Our study did not reveal whether the girls were particularly victims of sexual violence. Secondly, the differences between the two towns remain unexplained. We can attribute these differences either to bias resulting from a wider concept of what constitutes a girl of the street used at the Nouadhibou center than at the Nouakchott center; or to a difference in the context of the two centers, which focus on different types of work. We cannot provide a full answer to this question, although we did not detect any differences between the interpretation by the two centers of the concept of girls of the street. Finally, to conclude, we would like to address the implications in terms of policy. In the case of Rwanda, Veale and Donà (2003) have highlighted the need for community-based support for children in alternative guardianship care, and for policies to support the reintegration of male youths within the framework of post-conflict welfare strategies intended to prevent migration onto the street. Our analysis suggests the need for a policy promoting women's education. Unlike industrialized countries, developing countries are confronted with a context of poverty in which working mothers do not earn enough money to hire someone else to look after their children while they are out at work. In other words, poor households cannot substitute time for money when it comes to child care. Furthermore, family solidarities do not necessarily come into play, and this can result in children's experiencing some degree of “abandonment”. Emotional deprivation, i.e. not receiving much emotional investment (or at least suffering severe affective disorders) constitutes a major cause of the phenomenon of street-living children. Of course, apart

Table 3 The characteristics of children's social environment in Nouadhiboua. Variable

Household status Separated parents Missing father Parents’ occupation Unemployed father Unemployed mother Affective disordersb Without violence With violence

Odds ratio

Signif.

95% confidence interval

3.431* 1.852

0.017 0.389

1.251 0.456

9.414 7.528

76 76

Variable

0.325 7.330**

0.051 0.000

0.105 2.515

1.007 21.357

65 72

1.398 0.171**

0.474 0.001

0.559 0.062

3.494 0.472

77 77

Household status Separated parents Missing father Parents’ occupation Unemployed father Unemployed mother

Lower bound

N valid Table 4 Determinants of affective disorder risksa.

Upper bound

* Significant at b 5%, ** Significant at b 1%. a Dependent variable: city [Nouadhibou = 1, Nouakchott = 0]. b 68.8% of the children have affective disorders in Nouadhibou.

Odds ratio

Signif.

1.257 1.310 0.673 0.118**

95% confidence interval

N valid

Lower bound

Upper bound

0.713 0.755

0.371 0.241

4.256 0.863

76 76

0.570 0.008

0.172 0.024

2.636 0.574

65 72

*Significant at b 5%, **Significant at b 1%. a Dependent variable: Affective Disorder [Yes = 1, No = 0].

662

J. Ballet et al. / Children and Youth Services Review 33 (2011) 656–662

from emotional deprivation, the main problem is domestic violence. Violence may also be linked to the parents’ lack of availability, in addition to the prevailing cultural context.4 Parents who do not have time to look after their children more often have recourse to violent punitive methods to impose their authority. Likewise, neglected children have more opportunities to behave badly, and thus be subjected to violent punishment from their parents. Educated women get married later, have fewer children, work outside the home, and have greater autonomy from their husbands. Therefore, having fewer children, they can take a better care of them, and having a job they can cater for their needs for food, education and health care; they can also keep them at home. They have higher status in the home, and are given by more respect by their husbands. An educated woman will be more independent, and will generally suffer less conjugal and domestic violence. One major contribution to breaking the vicious circle of violence would certainly be to promote education for mothers, which would enable them to earn more and be given higher status in the family. Our conclusion does not exclude of course other forms of action, in particular political action and NGOs’ action. On the first, promotion of the rights of children appeared an essential way for changing mentalities. On the second, we can note that the NGO with which we worked was the only one NGO engaged in the rehabilitation of street children in that country. A lot is needed to make the population conscious of the negative effects of domestic violence.

References Aderinto, A. (2000). Social correlates and coping measures of street children: A comparative study of street and non-street children in Southwest Nigeria. Child Abuse & Neglect, 24(9), 1199−1213. Alexandrescu, G. (1996). Programme note: Street children in Bucharest. Childhood, Global Journal of Child Research, 3(2), 267−270. Aptekar, L., & Ciano-Federoff, L. M. (1999). Street children in Nairobi: Gender differences in mental health. New Direction for Child and Adolescent Development, 85, 35−46. Aptekar, L. (1988). Street children of Colombia. Journal of Early Childhood, 8(3), 225−241. Aptekar, L. (1994). Street children in the developing world: A review of their condition. Cross-Cultural Research, 28(3), 195−224. ARISE (2001). Baseline survey of street children in six divisional cities of Bangladesh. Dhaka: Department of Social Services, Ministry of Social Welfare, Government of Bangladesh. Ballet, J. (2006). Les enfants des rues: pauvreté monétaire et pauvreté affective. Formes sociales de la pauvreté en Mauritanie (pp. 77−112). Paris: L'Harmattan. Bartlett, S. R. H., Satterthwaite, D., De La Barra, X., & Missair, A. (1999). Cities for children: Children's rights, poverty and urban management. London: United Nations Children's Found-Earthscan. Basu, K., & Van, P. H. (1998). The economic of child labour. The American Economic Review, 88(3), 412−427. Beyene, Y., & Berhane, Y. (1997). Characteristics of street children in Nazareth, Ethiopia. East African Medical Journal, 74(2), 85−88. Blanc, S. C. (1994). Urban children and distress. Luxembourg: Gordon and Breach Publishers. Bourdillon, M. F. C. (1994). Street children in Harare. Africa, 64(4), 516−532. Chatterjee, A. (1992). India: The forgotten children of the cities. Florence, Italy: UNICEF. Connolly, M. (1990). Adrift in the city: A comparative study of street children in Bogota, Columbia and Guatemala City. Child and Youth Services, 14, 129−149. Conticini, A., & Hulme, D. (2007). Escaping violence, seeking freedom: Why children in Bangladesh migrate to the street. Development and Change, 38(2), 201−227. Ebigbo, P.O. (2003). Street Children: The Core of Child Abuse and Neglect in Nigeria. Children, Youth and Environments, pp.13(1). Ennew, J. (2000) Why the Convention is not about the Street Children. In: D.Fotrell (ed), Revisiting Children's Rights: 10 years of the UN Convention on the Rights of the Child (pp.169-82). (The Hague/Boston, Kluwer Law Int). Ennew, J. (2003). Difficult circumstances: Some reflections on ‘street children’ in Africa. Children, Youth and Environments, 13(1). Gordon, D. S. N., Pantazis, C., Pemberton, S., & Townsend, P. (2003). Child poverty in the developing world. Bristol: The Policy Press.

4 The cultural context plays of course an important role in domestic violence. It is particularly marks by the Muslim religion. But religion cannot be considered as an explaining factor of violence. Furthermore, women enjoy a certain degree of freedom compared to some other Muslim countries.

Hall, T., & Montgomerry, H. (2000). Home and away: ‘Childhood’, ‘youth’ and young people. Anthropology Today, 16(3), 13−15. Hutz, C., & Koller, S. (1999). Methodological and ethical issues in research with street children. New Directions for Child and Adolescent Development, 85, 59−70. Ignegongba, K. (1992). Fécondité et ethnie en Mauritanie. Bamako: CEPROD. ILO (1996). Child labour surveys: Results of methodological experiments in four countries 1992–1993. Geneva. Kabeberi, J. (1990). Child Custody, Care and Maintenance. Oxford: Oxford University Press. Kerfoot, M., Koshyl, V., Roganov, O., Mikhailichenko, K., Gorbova, I., & Pottage, D. (2007). The health and well-being of neglected, abused and exploited children: The Kyiv Street Children Project. Child Abuse & Neglect, 31, 27−37. Kilbride, P., Suda, C., & Njeru, E. (2000). Street children in Kenya: Voices of children in search of a childhood. Westport: Bergin & Garvey. Korbin, J. (1981). Child abuse and neglect: Cross-cultural perspectives. Berkeley: University of California Press. Lafdal, M., & Sow, A. (2006). Stratégies de prise en charge des enfants orphelins et autres enfants vulnérables en Mauritanie. In J., & B. (Eds.), Formes sociales de la pauvreté en Mauritanie (pp. 113−143). Paris: L'Harmattan. Lalor, K. J. (1997). Victimisation of street children in Addis Abeba: Factors of resilience and susceptibility. Unpublished PhD. Thesis: University College Cork, Ireland. Lalor, K. J. (1999). Street children: A comparative perspective. Child Abuse & Neglect, 23(8), 759−770. Lefort, C. (1988). Enquête sur les enfants de la rue de Nouakchott, report for UNICEF and Ministère des Affaires Sociales. Nouakchott, May: UNICEF. Le Roux, J. (1996). Street children in South Africa: Findings from interviews on the background of street children in Pretoria, South Africa. Adolescence, 31(122), 423−431. Lusk, M. W. (1992). Street children of Rio de Janeiro. International Social Workers, 35, 293−305. Lusk, M. W. (1989). Street children programs in Latin America. Journal of Sociology and Social Welfare, 16(1), 55−77. Mambwe, A. (1997). Stress and coping strategies amongst street children in Lusaka. Journal of Psychology in Africa, South of the Sahara, the Caribbean and Afro-Latin America, 2, 24−33. Marguerat, Y. (1999). Les actions en faveur des enfants de la rue au Cameroun. Relation d'une visite (Novembre 1998). Cahier de Marjuvia, 9, 128−136. Marguerat, Y., & Poitou, D. (1994). A l'écoute des enfants de la rue en Afrique Noire. Paris: Fayard. Matchinda, B. (1999). The impact of home background on the decision of children to run away: The case of Yaounde city street children in Cameroon. Child Abuse & Neglect, 23(3), 245−255. Moberly, C. (1999). Creating policies which address the ‘voluntary separation’ of children in Angola. In A., & A. (Eds.), Prevention of street migration: Resource pack (pp. 36−50). (London, Consortium for Street Children, U. K). Mohamed Salem, M.L. (1996). Enquête sur les jeunes et les enfants en situation difficile. (Nouakchott, Mauritania, Direction des Affaires Sociales & UNICEF). Muchini, B., & Nyandiya-Bundy, S. (1991). Struggling to survive: A study of street children in Zimbabwe. Harare, Zimbabwe: UNICEF. ODHD (2000). Rapport sur le développement humain en Mauritanie. Nouakchott: UNDPODHD. Olley, B. O. (2006). Social and health behaviors in youth of the streets of Ibadan, Nigeria. Child Abuse & Neglect, 30, 271−282. Onyango, P., Sudaa, C., & Orwa, K. (1991). A Report on the Nairobi Case Study on Children in Especially Difficult Circumstances. Florence: UNICEF. Panter-Brick, C. (2003). Street children, human rights, and public health: A critique and future directions. Children, Youth and Environments, 13(1), 147−171. Peacock, R. (1994). Street children. Africa Insight, 24(2), 138−143. Raffaelli, M., Campos, R., Payne Merritt, A., Siqueira, E., Autunes, C. M., Parker, R., Greco, M., Greco, D., & Helsey, N.the street youth study group. (1993). Sexual practices and attitudes of street youth in Belo Horizonte, Brazil. Social Science and Medecine, 37, 661−670. Rizzini, I., & Lusk, M. W. (1995). Children in the streets: Latin America's lost generation. Children and Youth Services Review, 17(3), 391−410. Sampa, A. (1997). Street children of Lusaka: A case of the Zambia Red-Cross Drop-In Center. Journal of Psychology in Africa, South of the Sahara, the Caribbean and AfroLatin America, 2, 1−23. Sall, A. (2006). La prostitution à Nouakchott, enquête auprès d'un groupe de femmes à risque. In J., & B. (Eds.), Formes sociales de la pauvreté en Mauritanie (pp. 53−75). Paris: L'Harmattan. Taçon, P., & Lungwangwa, G. (1992). Street children in Zambia. Lusaka, Zambia: Institute for African Studies. UNICEF (1984). Latin American seminar on community alternatives for street children. (Brasilia, Brazil, November 12–15). UNICEF (2006). The state of the world's children. New York: UNICEF. UNICEF (2000). Les enfants domestiques. Florence, Italy: UNICEF, Innocenti Digest. Veale, A. (1996). An Empirical and Conceptual Analysis of Street Children in Sudan and Ethiopia. PhD dissertation, (University College, Cork). Veale, A., & Donà, G. (2003). Street children and political violence: A socio-demographic analysis of street children in Rwanda. Child Abuse & Neglect, 27, 253−269. Witter, S. (2002). The silent majority: Child poverty in Uganda. London: Save the Children. Bank, World (2001). World Development report 2000/2001. Attacking poverty. New York: Oxford University Press.