Mobility, surveillance and control of children and young people in the everyday: Perspectives from sub-Saharan Africa

Abstract

Surveillance of children and young people in non-Western contexts has received little attention in the literature. In this paper we draw principally on our research in one African country, Malawi, to examine the ways in which their independent travel is shaped by(usually adult-directed) surveillance and control in diverse urban and rural contexts. Surveillance is interpreted very broadly, because our empirical data indicates a range of practices whereby a close watch is kept over children as they move around their community and travel out to other locations. In some cases we suggest that surveillance of children and young people becomes internalized self-surveillance, such that no external social control is required to police their movements. Our evidence, from eight research sites, brings together a wide range of source material, including findings from intensive qualitative research with children and adults (in-depth interviews, accompanied walks, focus groups, life histories) and a follow-up questionnaire survey administered to children aged 7-18 years [N=1,003]. Although many of the children in our study attend school, local economic circumstances in both urban and rural areas of Malawi commonly require children's participation from an early age in a much broader range of productive and reproductive work activities than is usual in Western contexts, with corresponding impact on daily patterns of movement. Children may have to travel substantial distances for school, in support of family livelihoods, and for other purposes (including social events): the necessity for independent travel is common, and frequently raises concerns among parents and other adults in their communities such that surveillance is considered essential. This is achieved principally by encouraging travel in groups of children. We show how young people's independent travel is mediated by (urban and rural) locational context, time of day, age and, in particular, by gender, and how adult efforts at surveillance may help shape resistances in the interstitial spaces which mobility itself provides. [PUBLICATION ABSTRACT]

Headnote

Abstract

Surveillance of children and young people in non-Western contexts has received little attention in the literature. In this paper we draw principally on our research in one African country, Malawi, to examine the ways in which their independent travel is shaped by(usually adult-directed) surveillance and control in diverse urban and rural contexts. Surveillance is interpreted very broadly, because our empirical data indicates a range of practices whereby a close watch is kept over children as they move around their community and travel out to other locations. In some cases we suggest that surveillance of children and young people becomes internalized self-surveillance, such that no external social control is required to police their movements.

Our evidence, from eight research sites, brings together a wide range of source material, including findings from intensive qualitative research with children and adults (in-depth interviews, accompanied walks, focus groups, life histories) and a follow-up questionnaire survey administered to children aged 7-18 years [N=1,003]. Although many of the children in our study attend school, local economic circumstances in both urban and rural areas of Malawi commonly require children's participation from an early age in a much broader range of productive and reproductive work activities than is usual in Western contexts, with corresponding impact on daily patterns of movement. Children may have to travel substantial distances for school, in support of family livelihoods, and for other purposes (including social events): the necessity for independent travel is common, and frequently raises concerns among parents and other adults in their communities such that surveillance is considered essential. This is achieved principally by encouraging travel in groups of children. We show how young people's independent travel is mediated by (urban and rural) locational context, time of day, age and, in particular, by gender, and how adult efforts at surveillance may help shape resistances in the interstitial spaces which mobility itself provides.

Introduction

This paper examines the independent daily mobility patterns of children and young people1 and their surveillance and control in sub-Saharan Africa. Whereas the surveillance of children in Western contexts has increasingly become a focus of research (see for example Fotel and Thomsen 2004, in an early issue of this journal, and a recent special issue), surveillance of children in non-Western contexts has received little attention in the literature. However, studies of rural children's mobility and autonomy by Katz in Sudan (1991; 1993; 1994; 2004) and Punch in Bolivia (2000), discussed below, offer important insights. Recent extensive and intensive field research into child mobility in three countries, Ghana, Malawi and South Africa, has drawn our attention to the extent to which (usually adult-directed) surveillance and control of children's daily mobility not only shapes their current lives but has the potential to strongly influence livelihood trajectories into adulthood with far-reaching, inter-generational implications (Porter et al. 2010b; 2010c; 2011 in press). Child mobility is a significant development issue. While it would be naïve to argue that mobility ensures freedom and opportunity, control over mobility clearly reflects and reinforces established patterns of power relations (Sheller and Urry 2006, 211).

Surveillance is interpreted very broadly in this paper because our empirical data indicates a range of practices whereby a close watch is kept over children and young people as they move around their community and travel out to other locations. Surveillance, we find, may operate in the form of direct watch by individual adults or by groups of children over those in their midst, or in the form of indirect watch, whether through requiring accompaniment by other children, or by requiring permission and knowledge of the child's whereabouts. In some cases, surveillance of children and young people appears to become internalized self-surveillance and associated self-regulation, such that no external social control is required to police their movements: this resonates strongly with Foucault's concept of governmentality, whereby surveillance takes place inside the body (Kallio 2007, 127). Self-surveillance may be encouraged through real or imagined threats to safety, not least stories and rumours in which supernatural forces play a significant role: the rhetoric of pressing, highly probable risk can transform 'every individual into a virtual victim, of others and/or of his or her own behaviour' (Vaz and Bruno 2003, 284). Young people also, however, have the potential to resist surveillance by employing political tactics of subterfuge-presenting themselves as docile bodies-in order to 'survive in the nets of discipline', while maintaining their 'unruly' bodies (see Kallio 2007, 127, citing de Certeau 1984, xiv).

Young people's daily mobility in Western countries has often been considered principally in the context of travel to/from school. Our discussion ranges well beyond the school journey, however, because although many of the children in our study are attending school, local economic circumstances in both urban and rural areas of sub-Saharan Africa commonly require children's participation from an early age in a much broader range of productive and reproductive activities than is usual in Western contexts, with corresponding impact on daily patterns of movement. The necessity for independent travel is common, and frequently raises concerns among parents and other adults in their communities: gender can play a significant role in the shaping of expressed concerns. Following Cresswell and Uteng (2008, 2) we take the term 'mobility' to mean not only geographical movement per se, but the potential to undertake movements as lived and experienced in the everyday, recognizing furthermore that 'narratives of mobility and immobility play a central role in the constitution of gender as a social and cultural construct'.

We focus on findings from eight sites (urban and rural) in Malawi,2 a small and very poor country in southern Africa with around 13.1 million people (Republic of Malawi 2008), poor infrastructure and one of the highest HIV prevalence rates in Africa (Kathewera-Banda et al. 2005; Peters 2006; UNAIDS 2010. Following a brief review of research methods we consider the broad context within which mobility and associated surveillance practices take place in Africa and Malawi, drawing on relevant literature, and outline key characteristics of each study site. We then examine patterns of daytime and night time movement, surveillance and control in rural and urban areas. This allows us to reflect on the ways that the mobility required of young people for schooling, family maintenance and other reasons in our Malawi sites is mediated on the one hand by family and community security concerns and on the other hand by young people's mobility desires. Surveillance in its varied forms-including self-surveillance-plays a complex role in the shaping of gendered mobility patterns. A key theme in the paper is the way group travel is widely utilised as a tool of mobility surveillance and control.

Utilising a mixed-method approach to researching young people's mobility patterns in Malawi

The work presented in this paper is drawn from a larger three-year, three-country inter-disciplinary research project focused on the daily mobility of children and young people up to the age of around 18 years. We utilised a two-strand approach, in which a more conventional adult academic research study was complemented by a less conventional child researcher strand. In Malawi 24 young researchers (secondary school pupils aged between 14 and 22 years) from six locations, following a basic training, selected research tools they preferred and conducted their studies among peers in places convenient to them: the methods they used included in-depth interviews, focus group discussions (FGDs), diaries, life histories, photographic journals of children's travel to school and at work (using disposable cameras), and accompanied walks. A full description of the recruitment and methodology in the Malawi child researcher component is available elsewhere (Robson et al. 2009). The young researchers' findings have fed into and complement the adult research strand. The latter included intensive qualitative research (in-depth interviews, life histories, Focus Group Discussions (FGDs), ethnographic diaries, accompanied walks) with young people, their parents and other key informants at the eight sites described below, and a quantitative survey among child respondents aged approximately 9-18 years in the same sites.

Mobile ethnographies3 have been a key method in our study (both for adult and young researchers, Porter et al. 2010a.). These have consisted principally of accompanied walks in which the adult or young researcher accompanied their (young) respondent to a place to which they travelled regularly. On some occasions, parents or guardians (approached in advance for consent) were reluctant to allow such walks, especially in the case of male researchers or research assistants accompanying young girls beyond their home compound: they perceived such interaction between young people of the opposite sex as antithetical to 'proper' conduct. Interaction with young people of the opposite sex is often constrained in an African social context, as young researchers stressed on a number of occasions (and as we discuss further below):

We girls, according to our tradition, we're not supposed to chat with boys; parents think you'll make friendships. Parents wouldn't allow you to go home with a boy. [16 year old girl, Malawi]

A questionnaire survey of 1,003 young people aged predominantly 9-18 years (approximately 125 respondents per site) provided additional quantitative data from the adult research strand. The questionnaire was designed following the in-depth qualitative research and has been analysed using SPSS. We draw on this data where appropriate.

Gendered mobility, sexuality and surveillance of young people in Malawi and elsewhere

In Malawi struggles for control over youth by an older generation trying to maintain hegemonic control, by the missions, and by political organizations, date back many decades (van Dijk 1998): surveillance has inevitably played a significant part in the maintenance of social control. Within the family context of modern-day Malawi, surveillance of young people remains a key component of control. The importance of surveillance and control can be explained not only in the fact that children are precious as individuals in themselves, but also because young people's inputs-particularly their labour contributions-are commonly perceived as vital to family survival (as is also widely the case elsewhere in Africa). Ensuring the security of young people's labour input is a major concern for many families, given the precarious circumstances within which they struggle to get by. However, basic education requirements and livelihood imperatives (such as the need for domestic supplies of water and firewood and the purchase and sale of goods at market), also require young people to be mobile and this mobility is widely considered to present substantial risks, especially for girls, as we will show. In such circumstances, keeping a close watch over young people as they move around can be presented as not only sensible, but also an ethical imperative.

Care and control, the two faces of surveillance (Lyon 2001; 2003), are complexly intertwined in the stories which unfold below. This is particularly evident with respect to gender, a key element in our discussion. In all our Malawi study sites we show how the imposition of mobility surveillance and control is highly gendered, especially once girls reach puberty, a finding which resonates substantially with Katz's detailed study of rural children's mobility and autonomy in Sudan (1991; 1993; 1994; 2004), and recent mobility studies in urban Ghana and Zambia (Gough 2008; Langevang and Gough 2009).

Katz (1993, 90) suggests that, in the Sudanese village context she studied, boys and girls before puberty shared an 'extensive knowledge of the local environment', developed in the course of their work and play around the village. Her research focuses on spatial range rather than practices of surveillance per se, but the findings suggest that young children of both genders had considerable freedom and limited surveillance was in place (though she refers to children moving together to collect firewood and perform other tasks away from adult supervision). Once girls reached puberty and marry, Katz observed how their access to the environment narrowed considerably. From the onset of puberty in Malawi, we also find that concern for girls' safety rises substantially, with reduced mobility and increased efforts at surveillance a likely consequence, closely bound up with issues of sexuality. In Malawi sexual activity starts early, around the age of 10 or 11, with sexual intercourse in earnest at about 12-13 years for girls and a little older for boys (Bryceson and Fonseca 2006, citing Shah 2002). Fear of a daughter's early pregnancy (loss of labour and possibly another mouth to feed), is also now compounded by the threat of HIV/AIDS: older men are targeting adolescent girls for sex because of their fear of contracting AIDS from older partners (ibid).

Women's need for assistance at home may also contribute to girls' reduced independent mobility as they grow older and become more skilled in domestic tasks, especially in rural areas. Indeed, Andvig (2001, citing Elster 1989) argues that, because boys tend to be more unruly and harder to discipline (a view expressed by parents in our study as 'boys will be boys'), this encourages African societies to make girls work more from an early age, since doing so reduces the monitoring costs for women. Thus, the opportunity costs of surveillance are lower in the case of girls! However, it can also be argued that while girls' daily pedestrian mobility (especially in distant locations where surveillance is difficult) threatens household survival through the risk of sexual liaison, STDs and (once they reach puberty) pregnancy, it may have to be maintained if it is essential to household production and reproduction. Given the transport gap so common in sub-Saharan Africa, pedestrian load carrying is particularly crucial (Porter 2008; Porter, Blaufuss and Acheampong, 2011 in press). In Malawi (as in much of sub-Saharan Africa), load carrying for domestic purposes is culturally a task assigned to females and can take girls far beyond the village or urban neighbourhood limits. As we show below, the role of group travel as a route to surveillance in this context is important.

Mobility surveillance may become even more difficult under cover of darkness, and night time is the period when the attractions of (unsupervised) mobility for young people are likely to be at their strongest: bars, video houses and other entertainment places offer exciting diversions from humdrum daily life. Across Africa, bars in particular are associated with casual sex: in Malawi 'sexual liaisons are seen as a natural accompaniment to heavy drinking' (Bryceson 2006, 185). Consequently, we consider temporal variations in young people's mobility and associated surveillance practices, where these are in evidence.

The issue of witchcraftand supernatural forces was raised by respondents of all ages as a potential threat to mobile young people. This is particularly evident in the interviews in the Central Region of Malawi where the Nyau cult is prevalent and Nyau traditional dancers, who wear masks and costumes, are viewed by non-members as objects of fear. The potentially significant role of witchcraftin shaping mobility resonates with Punch's work in a rural Bolivian village (Punch 2000). Within the (extensive) village area, Punch (2000) found apparently few limits on the mobility of either gender in the 8-14 year old group which were the focus of her research. However, she observes that adults use the well-known local superstition of a dwarf figure to encourage children to travel straight home from school to help with household tasks and interprets this as a means of controlling children's time to persuade them to work, rather than wasting their time in seemingly unproductive play. The supernatural may thus be brought into operation as a tool for inculcating self-surveillance: we have found apparently similar practices operating in diverse contexts in the course of our mobilities research in three sub-Saharan countries (Porter et al. 2010b).

The research sites

In this section we briefly describe our diverse research sites because their locational characteristics may help shape the patterns of mobility and surveillance we observe in operation. Our eight4 Malawi research sites comprise poor high density urban, peri-urban, rural and remote rural locations in each of two different zones of Malawi: the very densely populated Shire Highlands region in and around the city of Blantyre in southern Malawi, and the (slightly less) densely populated plains region in and around the country's capital, Lilongwe, in central Malawi. HIV prevalence is high across both regions (especially the Blantyre region) and there are many orphans living with grandparents and other family members, a few on their own in child-headed households.

The two remote rural sites (henceforth RR) in each zone are characterized by poverty and lack of services. The nearest primary school is 3-5km distance. In the case of the Blantyre remote rural site, the nearest secondary school is over 5km away; it is 10km away in case of the Lilongwe zone remote rural site. There is no public transport serving these areas and few private vehicles reach either settlement: walking is usually the only option. People live in highly dispersed farmsteads and engage in smallholder maize production, sometimes also growing a few vegetables and keeping small livestock or cattle. These remote rural sites are each linked by poor laterite track to a settlement with services, 3-5km away, which we also studied. The rural with services settlements (henceforth RS) both have a primary school (also attended by children from the surrounding remote rural settlements). Access consists of dirt roads (linking onward to our peri-urban study sites) which are (seasonally) accessible by motor transport but lack public transport services. People generally walk or cycle to reach the main tarmac road 5-9km away.

The peri-urban research sites (henceforth PU) are better serviced, with government primary and community day secondary schools, private schools, a health centre, post office, some electricity, some boreholes and public transport access by paved road to their respective urban centres. The urban research sites (henceforth U), one in Blantyre, Malawi's commercial centre, and one in Lilongwe, Malawi's administrative capital, are both poor, high density neighbourhoods with local primary and secondary schools (including some private schools) and various other facilities within a radius of 3km. At the time of data collection, Blantyre urban site had a main paved road running through the area which is well served by public transport (predominantly minibuses), but other roads were mostly poor and unsurfaced. Lilongwe urban site was less well served by public transport since the bridge accessing the neighbourhood had partly collapsed and motor vehicles could not reach the area. Both urban sites are a focus of in- migration and consequently more ethnically mixed than other sites.

From a surveillance perspective, we can hypothesise that the rural sites present particular challenges associated with keeping watch over young people as they traverse remote routes to distant goods and services such as schools, water sources and maize mills. By contrast, the urban sites tend to have closer service access, but their diverse populations, with many recent migrants, militate against community cohesiveness and can make adult carers' potential to build surveillance networks more difficult.

Daytime mobility and surveillance in rural Malawi

The journey to school offers a useful introduction to daytime mobility and surveillance issues in rural Malawi. Many children live in dispersed farmsteads: they may walk up to 4km or even further to the nearest primary school. Typically, children join each other as they walk, in many cases travelling across isolated areas along narrow footpaths across fields, woodland and uncultivated bush. The few secondary school children in our rural and remote rural sites make even longer journeys, again mostly on foot, to larger centres where these schools are located. Such journeys can be dangerous, a point which was reiterated by many children and their parents (Porter et al. 2010a). They refer to potential risks associated with the physical environment, notably rapidly rising streams in the wet season, but also expressed fears concerning human attack. Fathers talked in particular about daughters, who may be 'cheated on or get raped' on the way to school [Blantyre RR]. In the Lilongwe zone, the Nyau traditional dancers were mentioned by many children and their parents as of considerable concern when children walk beyond the village area. For older girls on the way to school the 'threats from men with rape [including] Gule wamkulu dancers' are considered particularly great [15 year old boy, Lilongwe RR], especially in the maize-growing season when paths are overgrown and visibility reduced.

Rural mothers strongly emphasized their fear of older daughters becoming pregnant. The following quotations give a clear indication of the level of concern, but also associated views about the need to control girls' nascent sexuality, and the role of very strong restrictions on mobility in achieving successful surveillance:

Because as parents we are always afraid that girls can end up getting pregnant and thus bring other responsibilities to us. Those above 13 yrs of age have [more] restrictions than those below 13 years...because we fear pregnancies...we feel boys are more responsible and less susceptible to bad company...And also girls can get pregnant if not carefully monitored, boys don't get pregnant...girls from 13 to 16 years usually don't just go outside the compound...because parents fear they could bring pregnancies home. It's very easy for a 17 and 18 year old to get married when pregnant but for those between 13 and 16... These...cannot get married even when pregnant. [mothers' FGD, Blantyre RR]

some girls stay on the way and enjoy themselves with boys and get pregnant. They use herbs from the bush to abort-you know these kids- ... They remove the pregnancy, so they get sick and then there's a lot of days at home [away from school]. [mother of girl 13 years, Lilongwe RR]

Although parents and guardians are concerned about the potential risks their children face as they travel across isolated areas, they rarely accompany children to or from school or other destinations: we were told they are too busy with work to do so. Only 0.4% of the 749 school-going children 9-18 years old in our Malawi survey had been accompanied to school by parents or other adults on their most recent school day, and none whatsoever were adult-accompanied in rural areas. However, in this as in other contexts, 'accompaniment is only one particular aspect of surveillance' (Fotel and Thomsen 2004, 536). In-depth discussions with rural children and their parents show the extent to which surveillance is achieved instead by instilling the fear which will promote travel in (mostly single-sex or sibling) groups: 'for girls the fear [of travelling alone] is being raped. Therefore we wish them to be in the company of somebody who is old[er]' [fathers' FGD, Lilongwe RS]. This theme of safety through group travel emerged regularly in different contexts, as we show below.

Apart from school, rural children must make many other regular journeys on foot during the day: to collect water and firewood for the household, to the maize mill to grind corn, to the market to buy or sell goods, to the farm, and to meet their friends whenever there is an opportunity to socialize. Again, the importance of travel in groups was emphasized, whenever (as for fuelwood collection), travel beyond the main village area was required. For instance, a fathers' discussion group observed the particular dangers of travel for children when the maize has grown tall: because the maize mill is far away they said that their children are required to travel in groups [Blantyre RS]. Similarly a mothers' focus group emphasised that although children go to market to buy goods, 'but younger ones don't go alone. They are always in the company of older ones'.[Blantyre RR]

Usually these requirements to travel in groups relate more specifically to girls, and we have considerable evidence that most rural girls take on parental concerns for their safety to the extent that they do not question them. Thus when we interviewed a group of out-of-school girls [aged 8-17 years] they themselves commented on how they choose to walk in a group on leaving the village environs [Blantyre RR]. As a set of school boys aged between 8 and 14 years stressed, however, such groups are single-sex. Boys are forbidden to walk with girls: 'we have not yet reached that stage...because we are in school' [Lilongwe RS].

Rural boys are allowed much more freedom to move around than girls, on the basis of usually lower work loads, less danger of rape, absence of fears of impregnation and an implicit acknowledgment (as discussed earlier) that boys are less easily controlled. A father of four girls and two boys was very clear about the differences:

Boys have more freedom than girls [why?] they are just boys...girls work hard more than boys, hence boys have more time to move around...Girls ought to stay home and look after the house and cook... There is little for the boys to do [just] harvesting and cultivating.[Blantyre RR]

Another father with a boy of 14 and two girls of 13 and 7 similarly observed,

as for the boy I don't know [where he goes to play]-as you know boys they just go wherever they want...my girls just play and chat around the houses close to our home. [The children are not allowed to go to distant social events-especially the girls because] they can get raped.[Lilongwe RR]

However, even older boys are not immune from fear of attack as they travel around the village area:

During the rains it's the dambo (wetland) we have to go through [to school]-there are some gangsters hiding in the bushes and when they see us they chase us and when you're grabbed they take your private parts. [primary boy, 14 years, Blantyre RR]

The actual risk of harm may be small, but this fear has a significance which should not be underestimated. Stories and rumours abound which encourage self-surveillance such that children take care to travel at speed along major paths, avoiding route deviation or dalliance; this resonates strongly with Punch's (2000) findings in rural Bolivia.

So far we have focused on journeys made on foot. Public transport is only rarely accessed by most rural children even in the few places it is available. Cost is a primary restraint, but there are also issues around reduced opportunities for parental and community surveillance. Again, in this context there is strong gender differentiation: boys are more likely than girls to be considered able to travel alone safely on public transport without accompaniment. Most adults and children of both genders suggested that girls should only travel alone on public transport when they are considerably older than boys (though the precise age varies): interestingly, their argument relates at least in part to literacy and ability to find one's way. A mothers' group in Blantyre RR, for instance, suggested that a boy over 13 could travel independently on public transport, but a girl would have to be over 16 -'boys here mature faster than girls...if the child knows how to read, chances of him getting lost are very slim'. A girl of 18 put the ages lower, at 14 for girls, 10 for boys: 'boys are more brave than girls' [Blantyre RR].

Bicycles are a more widespread form of transport among poor village communities, but like motorized transport, are more accessible to boys than girls, for whom independent travel of any sort is clearly perceived as potentially dangerous, since it limits the possibilities for surveillance. The pressures on girls' not to cycle range from ridicule to sexual innuendo and are applied by males of all ages:

Girls are afraid to learn [to cycle] because they feel shy if they fall down during cycling lessons...people laugh if they see a girl riding a bicycle...[out-of-school boys' FGD, Blantyre RS]

a lot of girls feel ashamed to ride bicycles because it makes them expose their legs when cycling. This also makes some men to start admiring them when they come in contact. A lot of men are full of [chilakolako] desire to have sex with these young girls hence they avoid cycling. [father's FGD, Lilongwe RR].

We asked children in our questionnaire survey about permissions required for visits to the market, neighbours' homes, play and church/mosque-all potentially important foci of regular daytime travel- and whether in such visits the child needed to be accompanied by others. In all rural settlements, only a relatively small proportion of children always needed permission to visit neighbours' homes or to go out to play, but a majority of both boys and girls always need permission to travel to church/mosque, while permissions are even more essential for travel to market (Table 1).

Gender differences in permissions appear relatively small in rural areas, except in the case of visits to neighbours' homes and to go out to play in remote rural settlements where permission is needed substantially more often by girls than by boys. Failure to obtain required permissions can result in strong sanctions as our qualitative data emphasizes, including being refused food at home [FGD schoolboys 8-14 years, Lilongwe RR]. A 14-year old girl who is no longer in school observed: 'I have never travelled without seeking permission from my parents...they would be angry with me' [Lilongwe RR]. She has to be accompanied by children of the same sex and age.

The proportion of rural children who always need to be accompanied (Table 2) was lower than the proportion always requiring permission, but in the case of visits to markets and to a lesser extent church, is still a requisite for some. However, there was more variation between individual settlements. In terms of market visits, moreover, gender differentiation was strongly in evidence in rural locations. In the case of market visits from remote rural settlements, for instance, 49% of girls and 21% of boys said they were always required to be accompanied on visits to the market from the remote rural Blantyre research site compared to 31% of girls and 20% of boys in the Lilongwe remote rural research site. In rural with service settlements, the comparative figures were 33% of girls, 13% of boys in the Blantyre zone and 29% of girls, 22% of boys in the Lilongwe zone. Since there is a lack of thriving markets in the remote rural settlements, it is perhaps unsurprising that many children are required to be accompanied on their walk to markets of at least 5km and that, given fears for girls' safety, they, in particular, are required to travel with others. Figures for requiring accompaniment for visits to church/mosque are lower, but still show slightly stronger requirements for girls to be accompanied than for boys.

To conclude this section, girls of all ages in the rural study sites are commonly far more constrained in their independent travel than boys. This can be related to widespread perceptions of girls' vulnerability to sexual assault but also, once puberty is reached, concerns around potential promiscuity. Community surveillance is considered necessary to ensure their safety and prevent sexual (mis)adventures. By contrast, surveillance of boys, especially older boys, is mostly limited to self-surveillance, a practice encouraged by stories and rumour. There is little effort at direct surveillance because boys are perceived as less controllable and also, perhaps, because constraints on their independent mobility are less crucial to family and community welfare. Similar themes, emphasizing the particular significance of gender to surveillance practices, emerge again in the following two sections.

Daytime mobility and surveillance in peri-urban and urban Malawi

Although urban and peri-urban areas differ from rural areas in population density, range of services and livelihood repertoires, there are many parallels between Malawi's rural, peri-urban and urban areas in terms of patterns of child mobility and adult surveillance. The journey to school in peri-urban and urban areas is often perceived by parents and their children to be as hazardous as in rural locations, though the emphasis on purely physical hazards such as rivers in spate is rather lower: the emphasis is mostly on the potential for human attack. In Lilongwe urban and peri-urban sites, for instance, a number of informants emphasized the danger of overgrown areas where people can hide and girls, in particular, reported the potential for attack and harassment:

girls are raped during the rainy season [when there is high grass] on the way to and from school...two girls were raped on the way last year and the culprits are in jail right now.[girl 16 years, Lilongwe PU]

The higher density of bottle stores and bars selling alcohol in these highly populated areas appears to increase the potential for harassment, especially of girls. In our Lilongwe urban site, school girls aged 12- 15 years in a FGD told of being mocked by men who say they should marry and brew beer, as they cross the bridge to school. There are similar problems in Lilongwe peri-urban locations:

we as girls meet drunken people on the way and they shout ill words at us like swearing and some even propose love to us...we just walk silently past them and we understand that it's the alcohol. [secondary school girl 19 years]

Urban mobility patterns outside the journey to school are as distinctively gendered as in our rural sites and similarly shaped by felt needs for surveillance. One father of ten children in Blantyre succinctly observed, 'To raise a girl child properly, you need to have a keen eye on her always'. As in rural areas, girls are regularly warned of the dangers of walking alone from an early age and are widely expected to be home and engaged in domestic chores by around 4pm, while boys, by contrast, are allowed much greater freedom.

Continued repetition of stories, whether real or merely rumour, also encourage the same form of self- regulation apparent in rural areas. Thus, girls are not only required to travel in all-female groups but indicate that they themselves choose to do so (as we saw above in the rural context):

when we go to fetch firewood, parents always insist that we should go with someone...when it is a young girl, parents always insist they should go with older friends and should not go with boys. [why?] boys can rape you...I know of a girl who was raped when she was coming from firewood and the culprit was apprehended and taken to police. [13 year old in schoolgirls' 13-18 years FGD, Blantyre PU]

There is also the implication, in some interviews, that a girl who does not take the precaution of travelling with others may herself be to blame for unfortunate consequences. Consequently, even a younger sibling may be enlisted to accompany her, if no one else is available.

sometimes you are given a young child to accompany you on your journey [prompt: why?] To make sure that you are not going to be involved in risky behaviours because the kid will give you a burden if you attempt to do that [all laugh in agreement]. [17-year old in an out-of-school girls' 12-17 yrs, FGD, Blantyre PU]

I always seek permission [to travel] even...going to the market...I am always in the company of my aunt or if she is tied up I make sure I get a child from home I have to move with. [fostered out-of-school girl, 16 years old, Lilongwe U]

Urban boys, by contrast, often seem to push for freedom, for instance disregarding parental refusal of permission. If their parents refuse a request to travel, they may plead in the expectation that parents will eventually give in:

if we really want to go, [we] continue to ask for permission, we reason with them and explain how important the journey is. In other words we force them to allow us to go, and then they say "aaah go!" [but] girls are not allowed to go anywhere anytime...parents fear a girl would get pregnant or catch diseases or any problems while a boy aah can refuse responsibility for pregnancy and would be fine.[schoolboy FGD, Blantyre PU].

As in rural areas, urban boys are in a stronger position to negotiate their use of space than girls: adults tend to see boys as being able to take care of themselves, and fight back, unlike 'weak' girls and thus less in need of restriction or surveillance. However, our Blantyre young researcher diaries, in particular, indicate that urban boys by no means perceive themselves free of dangers. Although allowed more freedom than their sisters, they recount stories which suggest they may be almost equally open to attack: by bigger boys who steal their money and other possessions on the way to school or the market and by men who are high on cannabis. A group travel strategy is consequently adopted even by older boys:

we wait for each other so that we can walk in groups. This prevents attacks from bad people...sometimes they beat us and tell us not to let anybody know about it...[or] they will skin us to death. [schoolboy, 16 years, Blantyre U]

we still regard walking in a group safer...one pupil can walk to school unaccompanied comfortably during the morning because the road is busier in the morning unlike in the afternoon when we are knocking off. [schoolboy, 16 years, Lilongwe PU]

Public transport is widely available in urban and peri-urban areas, but remains mostly inaccessible to children travelling alone, due to cost, to fears of children getting lost (as in rural areas) and also sometimes because of concerns about potential abduction and its consequences for girls. Just as in rural areas, boys are given more freedom to travel on public transport, not least 'because they can't get pregnant' (orphan girl, 15 years, Blantyre PU). While bicycles might seem a relatively inexpensive alternative, girls have little access to them, again as in rural areas. As two mothers in our Lilongwe urban site separately expressed the issue: 'girls are told that they were not made to cycle'; and 'it is a boy's business to cycle'.

Our survey data on permissions required for (principally daytime) travel in urban and peri-urban areas (Table 1) indicates that permissions to go to market do not differ very substantially from those required in rural areas. By contrast, the requirement for permission to go out to play/socialize within the settlement is higher in urban and peri-urban areas, for boys and especially for girls (reaching as high as 40% for girls in urban areas). The need for permission to visit a neighbour's home is particularly high in urban areas, both for boys and girls. This is probably because in rural communities neighbours know one another, there is strong community cohesion and more social/ethnic homogenity, whereas in town even the people living next door may be strangers.

The proportion of children who always needed to be accompanied (Table 2) was much lower than the proportion always requiring permission, as in rural areas. Rather fewer boys and girls than in rural areas always needed to be accompanied on visits to market, presumably because of the proximity of local markets. In peri-urban areas more girls (around one-fifth) always require permission to travel to market but this figure is still substantially below the 31% for girls in rural-with-service settlements and again can be linked to market proximity. Accompaniment on going out to play/socialise or visit neighbours is rarely required in urban and peri-urban contexts (as in rural contexts), presumably because initial permission to visit neighbours or play (as noted above) is only granted when the environment to be visited has been assessed as safe.

Night travel and surveillance

Night travel is an issue which many rural and urban children raised in qualitative interviews. There is a widespread view in all sites that girls of all ages should be home by 4-5pm at latest. This is commonly expressed by parents as a requirement based on their domestic tasks: 'girls are the ones who perform most of the household chores so they have to come back home in good time if people are to eat food at the right time' [fathers' FGD, Lilongwe RS], but discussions with girls suggest it may be more to do with parental concerns to have daughters safely at home before night falls (between 6 and 7pm), when surveillance becomes more difficult. They report widespread restrictions on night movements:

my mother forbids me to travel at night...she spares us from the trouble which boys bring on us...Boys are always wanting us to be in relationships with them...boys are more active during the night, that's why my mother forbids us being out at night. [primary girl 16 yrs, Blantyre RS]

you can't see girls in this village walking late in the evening...girls just don't walk anyhow... parents want to protect their lives [prompt] from getting pregnant or catching HIV/AIDS. Boys can go out late in the evening. [out-of-school 13-18 years girls' FGD, Blantyre RS]

One mother of five who had allowed her 16 year old daughter to travel at night regretted her leniency:

Mary is pregnant. Sometimes I tend to think that I was very loose at times in letting her travel during late hours; [it] is giving them too much freedom. [Lilongwe RS]

In urban areas respondents focused particularly on issues concerning (risky) venues operating at night. The many urban bars open at night are frequently associated with prostitution both by parents and daughters, while places where videos are shown are considered by some parents to promote pornography and violence. Wedding celebrations and dances held at night raise similar concerns, while in the Lilongwe area, fear of Nyau and associated witchcraftis perceived to add to the dangers. Girls who arrive home late say that they are sometimes suspected (and accused) by parents and carers of prostitution: beatings and food deprivation appear to be standard sanctions. The following comment provides clear illustration of the perceived issues and the way girls may self-regulate their behaviour accordingly:

Baghdad [a local bar], that is a place which no one in his rightful senses would go to...almost all girls who go there are commercial sex workers so what would you do there if you are not one of them? [FGDgirls 13-18 years not in school, Blantyre U]

The tightly regulated lives of girls from puberty and the contrasting relative freedom experienced by their male peers sometimes give rise to inter-generational friction, especially in city contexts where night life attractions are widespread (Porter et al. 2010b):

Boys can even come home very late at night but not us...Parents always doubt our movements. They fear we might get pregnant...[if we are late] parents shout at us, we are not allowed to enter the house, we are whipped by our guardians. [FGDout-of-school girls, 16-18 years, Blantyre U].

Sometimes, perhaps inevitably, this leads to subterfuge 'We are not allowed to go and watch films at video halls...they also refuse to let us go to places like 'Baghdad'...it's a bar and lots of girls go there for promiscuous acts...a neighbour sent me [there] to buy beer for him...I didn't tell [my parents]'.[FGD, secondary school girls 13-18 years, Blantyre U]

Although concern is less for sons' than daughters' mobility at night (as at all times), anxieties for boys' safety are sometimes evident, both among adults and boys themselves. An out-of-school, fostered boy of 15, for instance, always seeks permission to travel, especially if he spends the night away. Normally, he must be home by 8pm at the latest: 'it is dangerous to travel around the village especially during the night [prompt] people may mistake you for a thief and beat you up' [Lilongwe RR]. There is no electricity in the village, so it is difficult to see or be seen, unless the moon is full. A schoolboy of the same age who lives with his grandparents made a similar point:

when walking during the day there are no fears but at night it is dangerous [why?] there are some places where bandits waylay people...at the junction...and at the graveyard. These places are outside the village...Around 7 and 8 at night...There have been stories of people being waylaid and ambushed. [Lilongwe RS]

He has been told that he should not travel around at night because of bandits: 'it's not like the grandparents bar us from walking in the evening but the stories about the bandits threaten them...[I can go] everywhere I want so long [as] there are my friends there'.

He has not argued with his grandparents about this, he says, because he knows himself those places are dangerous at night. The way gossip and myth helps inculcate self-surveillance, in a mode similar to that reported by Punch (2000), is particularly in evidence in this case.

Discussion

The data presented above suggests the ways material social practices of security help shape the lives of millions of Malawian young people in the everyday. The surveillance they experience is presented by both parents/carers and their children principally as a reflection of safety concerns rather than an issue of control. It is only in the urban context, where night life temptations are particularly great, that we see any clearly expressed indications of resentment among young people. Elsewhere there is substantial evidence of internalized self-surveillance and associated regulation of movement to perceived safe spaces and daylight hours.

Early childhood surveillance and mobility regulation

In rural areas, adult surveillance of a child's movements is in place from the time it leaves the chitenje (sling) on its mother's back. In their early years, prior to school, rural children play in the compound or village street close to home, watched by older siblings, parents, aunts, uncles, grandparents and neighbours. If the primary school is any distance, as in our remote study villages, they will be kept at home until considered sufficiently strong to walk the 4-5km journey to and from school. Once enrolled in school, children travel without parental accompaniment, but within a group of siblings and/or friends, which offers protection and co-surveillance: collective safety through peers, in some respects comparable to the strategy that girls use in UK (Brown et al. 2008). Similar group travel is expected as children undertake family-directed farming, marketing and domestic tasks (notably collecting water and firewood). Adverse impacts of chance encounters with danger (including the many cases involving physical hazards, which probably greatly outnumber dangers from people), can thus be reduced, at minimal cost to individual parents and households.

In poor urban neighbourhoods and peri-urban areas, where children are also regularly required to undertake tasks such as marketing, water and firewood collection, a similar pattern of group travel is also seen as key to safe journeys. However, more integral surveillance, of the sort feasible in a village context, where an ensemble of mechanisms can be brought into play (Foucault 1980: 71) is commonly difficult, because community members who know the child may be fewer and regular in- and out-migrations from the neighbourhood make local surveillance networks harder to establish and maintain. As one father of 10 in Blantyre city explained:

What is happening here is different from what would happen if we were in the village. Here are many places that he can sneak to...and nobody cares. While in the village...a child's misdemeanours are easily reported to parents because people take up the responsibility of raising up each other's child.

This reflects a common notion of the rural idyll: that villages are safe places for children, towns are full of lurking evils and malevolent stranger-danger-a perspective also implicit in the urban survey data on permissions required for visits to neighbours reported above.

Care and control in the early childhood years are inextricably entwined. Efforts at adult surveillance, whether in rural or urban areas, appear principally concerned with ensuring that children of both genders are safe from abduction and physical threat (especially dangerous rivers and hazardous traffic), but also (particularly in the Lilongwe zone, outside the city), from expressed fears of Nyau cult and other witchcraft. However, the dangers of specific places and a broader fear of travelling alone are persistently inculcated over time, directed towards a deeper behavioural control: 'yes they listen because we started telling them since they were young and this is just in their minds now' [mothers' FGD, Lilongwe PU]. At the same time, there is a more relaxed attitude to boys' movements, such that a wider area of play and later return home at night than among girls is commonly accepted. Much stronger efforts are made to socialize girls than boys into mobility compliance (in all contexts) from an early age.

Older children's surveillance and mobility regulation

As boys grow older they seem to be perceived to gain strength and invulnerability, such that their increasing freedom to travel is often implicitly accepted by adults, despite the reality of dangers which older boys may face. However, a careful examination of in-depth interviews indicates that the situation is probably more complex. On the one hand we have seen older boys discursively constructed as 'brave', 'responsible' and 'strong', on the other hand as uncontrollable young people who 'just go wherever they want', unheeding of parental strictures. Arguably both constructions are necessary components of masculinity as it is perceived in this cultural context. Consequently, direct parental surveillance of older boys is likely to be viewed as culturally inappropriate and extremely difficult to put into place. Instead, adults provide warnings through stories and rumours which encourage boys towards self-surveillance and group travel.

By contrast, surveillance of older girls, once they approach puberty, is viewed as not only appropriate but essential, whether in urban or rural areas. This is closely bound up with issues of sexuality and the widespread association, across Africa, of female mobility with promiscuity (Porter, 2011; see also Hampshire et al., 2011). Parental fear of early pregnancy is fairly ubiquitous and is presumably compounded by the fear of HIV/AIDS, although specific reference to that threat is relatively uncommon in our interviews (being limited to comments, such as that by a boys group cited above, that girls may 'catch diseases'). This is probably because AIDS is such a sensitive, stigmatized and frightening issue in Malawi. However, although girls' daily pedestrian mobility threatens household survival through the risk of sexual liaison, pregnancy and STDs, it has to be maintained, because it is also essential to household production and reproduction, particularly in the context of the need to carry water, fuelwood and other essential domestic requirements. Since carrying is conventionally a female task, route security has to be maintained to waterpoints, maize mills etc.; remoter routes are frequented by girls/women only in (all- female) groups. Avoidance of local sites of insecurity (cemeteries, high grass, tall maize) is often secured by stories and rumour. Rumours of rapists, robbers and witchcraftstrongly encourage compliance (as in Punch 2000). But to what extent are fears of rape and the level of regulation of girls' movements justified by actual incidents and to what extent are girls' fears purposely exacerbated by expressed parental concerns and neighbourhood rumour? Although there are widespread stories of abduction and rape, including plentiful reports in the national press, relatively few specific recent incidents were reported in our Malawi study sites.5

Distant travel is particularly restricted for girls. Any long-distance journey which entails public transport is likely to be undertaken rarely and, again, only with at least one female or family companion. Cycle riding is seen as unsuitable for girls (and a provocation to men), although just less than half of the girls surveyed (44% of girls in our Malawi survey, compared to 72% of boys) say they know how to ride a bicycle. Like public transport, the bicycle offers the potential for independent unsupervised travel which adults-women included-view as inimical to the proper and safe comportment of girls: as described above by a women's discussion group: 'We fear that the girls can meet somebody'.

Vigilance is ever-present and temporal aspects of mobility gain even stronger significance as girls reach puberty. Girls' carefully monitored daylight movement is essential to family livelihood maintenance; darkness by contrast presents a blanket of cover for mobile bodies which requires much firmer control. As we have seen, late return home from school or delayed domestic tasks in the afternoon may result in severe punishment, including beatings and food deprivation. Girls are expected to be in the home preparing the evening meal well before night falls, when surveillance will become more difficult. Girls' time-spaces are thus shaped, ostensibly, by the routine demands of household reproduction in a similar way to that observed in rural Sudan by Katz (1991; 1993; 2004, 146) and in northern Nigeria by Robson (2004), but surveillance objectives are intricately intertwined in those requirements. There is little indication in this context of the parental dilemma between wanting to protect children from dangers while also wanting to develop their independence that is so widely evident in Western-focused literature on children's use of space (for instance, Valentine 1999).

In urban areas night time surveillance of older girls is particularly challenging, since surveillance even of daylight pedestrian travel is difficult. At night the threat of female mobility appears to induce moral panic. This is not necessarily surprising, given the extent to which, as we have seen, bars are associated with casual sex. Young girls are discursively constructed (and governed) by their parents and guardians not only as weak and helpless victims (as in daytime contexts) but also as potentially promiscuous sirens who, under cover of dark and lacking self-control, may submit to the temptations of bars, nightclubs, discos and dances. Interestingly, this view of girls' sexual agency came from women as well as men: the argument that 'boys are more responsible and less susceptible to bad company' was made by a woman in a group of women and was not contested!

For the most part, girls seem to accept this highly gendered regulatory environment in which they grow up and even, in some cases, to take comfort from the constraints imposed: '[my parents] need to monitor every step that I make. My movements are strictly monitored at home...Perhaps because I am the last born and so dear to them' [out-of-school girl, 18 years, Blantyre U]. In a Foucauldian sense, girls may thus become prisoners of their own making. Many girls stressed the importance of girls' travel in (single-sex) groups to keeping safe from human and supernatural attack; some suggested that boys were cleverer, stronger and more capable mentally of making long journeys. Parental hypervigilance of the sort commonly experienced by children in the US and UK is arguably replaced in Malawi by a seemingly classic case of vigilance through internalized control and surveillance by the travelling group.

Despite this evidence of internalized control, we also pick up significant hints of defiance and transgression, the subversive invasion of forbidden and potentially dangerous spaces: the girl who went to buy beer at Baghdad bar in Blantyre, despite her father's rule, for instance, or the children who arrive home late, despite the threat of beatings and food deprivation. Since many girls become pregnant at a very early age, one must assume that family and community surveillance has ultimately often been inadequate to prevent illicit sexual liaisons. Indeed, pregnancy may even present itself as a perceived escape route from surveillance, if it offers the potential to move out of family control to an independent partner. In urban areas, especially under cover of darkness, there is more opportunity for children to resist, albeit discreetly, taking advantage of the temporal and spatial interstices which mobility in a less regulated environment provides. Indeed, reduced surveillance and the liberation it offers in urban contexts is, not surprisingly, a key attraction of the city for many young rural migrants:

there are lots of things happening here and everybody is busy...I am a free person here, nobody watches the moves I make here. In Thyolo6 [my birthplace] people are always noticing what you are doing while here everybody is busy doing their things. [young man, 22 years, life history, Lilongwe U]7

Our interviews with younger people in Malawi were mostly less revealing. In a society where adults frown on children's independent agency, acts of resistance among young people are widely deemed unacceptable. Even in the context of sympathetic exchanges between the young people and our young researchers, we suspect there was a reluctance to disclose incidents where parents and elders were overtly disobeyed. Resistances are thus more evident in the outcome (notably pregnancy) than the act.

Conclusion

A Western observer visiting our Malawi research sites for the first time (or, for that matter, many other parts of sub-Saharan Africa) would probably be extremely surprised by the apparent freedom with which groups of girls and boys seem to travel around without direct parental supervision. That freedom is more apparent than real. Young people are usually the least powerful in their communities in Malawi, as elsewhere in Africa, and experience sustained efforts by elders to control their mobility. In the context of poverty, livelihood insecurity and high HIV/AIDS prevalence, surveillance of young people-especially girls-is considered essential. However, parental hyper-vigilance of the sort common in Western contexts would be impossible, given the restricted means available to most parents. Instead, especially in rural locations, vigilance is maintained through a collective effort of group and community surveillance, to which children are themselves recruited at an early age, in particular through adult strictures to travel in groups and a mobility discourse which emphasises the dangers prevalent beyond essential routes: supernatural threats may be invoked to ensure compliance. The fact that the vast majority of young people's journeys take place on foot (in the absence of funds to pay transport fares), and consequently within a limited spatial range, enables a level of surveillance which would be difficult to achieve were other means of transport widely available to them.

To conclude, a brief reflection on broader developmental aspects of the surveillance practices we have observed operating seems necessary, given the depth of poverty prevalent in our study sites. On the one hand, the surveillance offered by group travel appears to give some protection to children and young people, especially girls, from dangers such as human attack (rape, robbery etc.). Regular travel in the same group may also encourage development of strong social networks that can provide support to individuals as they grow into adulthood. On the other hand, the utilization of supernatural threats, which encourage self-surveillance, are arguably damaging to the potential for independent thinking and action in the emphasis they place on the extreme vulnerability of mortal individuals in a wider spirit world. Moreover, particular limitations on girls' mobility-imposed because of the restricted opportunities for surveillance during distant travel, travel by bicycle or public transport and at night-may have very far-reaching implications for their livelihood trajectories and for broader community development. Restricted mobility limits girls' education (especially) beyond primary school and will substantially reduce their potential to build the more extensive social networks that are often so important in an Africa context for developing new livelihood opportunities for themselves and their families (Porter, 2011; Porter et al. 2010 c.; 2011b in press). We observed at the outset that it would be naïve to argue that mobility ensures freedom and opportunity, but the control over girls' mobility obtained through diverse surveillance practices reflects and reinforces established patterns of power relations and is thus a significant development issue. The mobility theme is not commonly central to Surveillance Studies (Fonio 2008), but as this paper has emphasized, surveillance in a mobility context has enormous significance for young people's lives and life chances: it is worthy of close attention.

Acknowledgements

The research was funded by ESRC/DFID (RES-167-25-008), however, the views expressed are those of the authors' alone. The research was made possible by the generous participation of the young people, their teachers, parents and communities. We acknowledge the contributions of the young researchers: Manes Banda, Alie Bwanali, Tendai Chiwawula, Lawrence Godfrey, Mary Kamphangwe, Dalitso Kaunda, GiftKawanga, Bernadetta Kuchonde, Christopher Lyon, Ludovicco Magola, Esther Malimusi, Christopher Mbeza, Anthony Merrick, Brasho Moffart, Towera Mwaungula, Smart Ng'oma, Alinafy Ntewa, Tionge Phiri, Georgina Pwele, Thokozani Tembo, Nenani Thindo, Micklina Welesani, Monica William, Tisunge Zuwaki and all the adult research assistants: Bernie Zakeyo, Brian Mkandawire, Linny Kachama, Matthews Nkosi, Faith Hakim, Meya Mkandawire, Lucia Desire, Maurice Zulu, Kissinger Chima, Chiwawula Joseph and others. We also wish to acknowledge the helpful comments received from two anonymous reviewers.

Footnote

1 We use the terms children and young people interchangeably in this paper to refer to the age-group with which we worked (principally 9-18 year olds). Definitions of childhood and youth are difficult, especially in an African context (Porter et al. 2010).

2 The wider project includes parallel comparable research in Ghana (Albert Abane et al.) and South Africa (Mac Mashiri et al.). Details of the wider project can be found at www.dur.ac.uk/child.mobility/

3 We use the term 'mobile ethnography' here as specific acknowledgment of the physical mobility of the subjects involved in making and recording the narrative, and the centrality of mobility to that narrative.

4 In Malawi we had qualitative data from an additional (less poor) urban site in Lilongwe zone and an additional subsequent quantitative survey of northern Malawi sites: data from these sites did not form part of the main study and are not reported in this paper.

5 However, as much Western experience suggests, lack of actual reportage of rape to the police, authorities, or other outsiders, does not imply lack of widespread attacks.

6 Thyolo is a predominantly rural district.

7 Interestingly, while youth are chastised by adults for their loose ways and unrulyness, Bryceson and Fonseca (2006) find youth in rural Malawi also blaming adults 'for not setting a good example'. This encompasses men's drinking and sexual liaisons, women's increasing engagement in transactional sex in the context of recent famine and household impoverishment.

References

References

Andvig, J.C. 2001. Family-controlled child labour in sub-Saharan Africa. A survey of research. Washington: World Bank, Social Protection Unit, October 2001.

Brown, B., Mackett, R., Yi, G., Kitazawa, K. and Paskins, J. 2008.Gender differences in children's pathways to independent mobility. Children's Geographies 6(4): 385-401.

Bryceson, D. 2006. Ganyu casual labour, famine and HIV/AIDS in rural Malawi: causality and casualty. Journal of Modern African Studies 44(2): 173-202.

Bryceson, D. and Fonseca, J. 2006. Risking death for survival: peasant responses to hunger and HIV/AIDS in Malawi. World Development 34(8): 1654-1666.

Cresswell, T. and Uteng, T.P. 2008. Gendered mobilities: towards an holistic understanding. In Gendered mobilities, edited by T.P. Uteng and T. Cresswell, pp.1-14. Alsdershot: Ashgate.

de Certeau, M. 1984.The practice of everyday life. Los Angeles: University of California Press.

Elster, J. 1989. The cement of society. Cambridge: Cambridge University Press.

Fonio, C. 2008. Review of T. Monahan (ed) Surveillance and Security. Surveillance & Society 5(2): 205-207.

Fotel, T. and Thomsen, T.U. 2004. The surveillance of children's mobility. Surveillance & Society 1(14): 535-554.

Foucault, M. 1980. Questions on Geography. In Power/knowledge: selected interviews and other writings 1972-1977 by Michel Foucault, edited by C. Gordon. New York: Harvester Press.

Gough, K. 2008. 'Moving around': the social and spatial mobility of youth in Lusaka. Geografiska Annaler 90(3): 243-257.

Hampshire, K., Porter, G., Mashiri, M., Maponya, M., and Dube, S. 2011 Proposing love on the way to school: mobility, sexuality and youth transitions in South Africa. Culture, Health & Sexuality 13(2): 217-231.

Kallio, K. P. 2007. Performative bodies, tactical agents and political selves: rethinking the political geographies of childhood. Space and Polity 11(2): 121-136.

Kathewera-Banda, M., Gomile-Chidyaonga, F., Hendriks, S., Kachika, T., Mitole, Z. and White, S. 2005. Sexual violence and women's vulnerability to HIV transmission in Malawi: a rights issue. International Social Science Journal 57: 649-660.

Katz, C. 1991. Sow what you know: the struggle for social reproduction in rural Sudan. Annals of the Association of American Geographers 81(3): 488-514.

Katz, C. 1993. Growing girls/closing circles: limits on the spaces of knowing in rural Sudan and US cities. In Geographies of women over the life course, edited by V. Kinnaird and J. Monk. London: Routledge.

Katz, C. 1994. Textures of global change: eroding ecologies of childhood in New York and Sudan. Childhood 2(1-2): 103-110.

Katz, C. 2004.Growing up global: economic restructuring and children's everyday lives. Minneapolis: University of Minnesota Press.

Langevang, T. and Gough, K. 2009. Surviving through movement: the mobility of urban youth in Ghana. Social and Cultural Geography 10(7): 741-756.

Lyon, D. 2001.Surveillance society: monitoring everyday life. Buckingham: Open University Press.

Lyon D. 2003.Surveillance after September 11th. Themes for the 21st Century. Cambridge: Polity Press.

Peters, P. 2006. Rural income and poverty in Malawi. Journal of Development Studies 42(2): 322-345.

Porter, G. 2008. Transport planning in sub-Saharan Africa. Progress report 2. Putting gender into mobility and transport planning in Africa.Progress in Development Studies 8(3): 281-289.

Porter, G. 2011. 'I think a woman who travels a lot is befriending other men and that's why she travels': Mobility constraints and their implications for rural women and girl children in sub-Saharan Africa. Gender, Place and Culture 18 (1): 65-81.

Porter, G., Hampshire, K., Abane, A., Munthali, A., Robson, E. and Mashiri, M. 2010a.Where dogs, ghosts and lions roam: learning from mobile ethnographies on the journey from school. Children's Geographies 8(2): 91-105.

Porter, G., Hampshire, K., Abane, A., Robson, E., Munthali, A., Mashiri, M. and Tanle, A. 2010b. Moving young lives: mobility, immobility and inter-generational tensions in urban Africa. Geoforum 41: 796-804.

Porter, G.,Hampshire, K.,Mashiri, M., Dube, S., and Maponya, G. 2010c. 'Youthscapes' and escapes in rural Africa: education, mobility and livelihood trajectories for young people in Eastern Cape, South Africa. Journal of International Development 22(8): 1090-1101.

Porter, G., Blaufuss, K., Owusu Acheampong, F. 2011a, in press. Filling the family's transport gap in sub-Saharan Africa: young people and load carrying in Ghana. In Geographies of children, youth and families: an international perspective, edited by L. Holt. London: Routledge.

Porter, G., Hampshire, K., Abane, A., Tanle, A., Esia-Donkoh, K., Amoako Sekyi, R., Agblorti, S., Owusu, S.A. 2011b, in press. Mobility, education and livelihood trajectories for young people in rural Ghana: a gender perspective. Children's Geographies.

Punch, S. 2000. Children's strategies for creating playspaces: negotiating independence in rural Bolivia. In Children's geographies: living, playing, learning and transforming everyday worlds, edited by S. Holloway and G. Valentine, pp. 48-62. London: Routledge.

Republic of Malawi. 2008.2008 Population and Housing Census: Preliminary Report. Zomba: National Statistical Office.

Robson, E.2004. Children at work in rural Northern Nigeria: Patterns of age, space and gender. Journal of Rural Studies 20: 193- 210.

Robson, E., Porter, G., Hampshire, K. and Bourdillon, M. 2009. 'Doing it right?': working with young researchers in Malawi to investigate children, transport and mobility. Children's Geographies 7(4): 467-480.

Shah M. 2002. Buying sex for three sweet potatoes: participatory assessment of adolescent sexual and reproductive health in Makala village, Lilongwe. Lilongwe: CARE international, August 2002.

Sheller, M. and Urry, J. 2006. The new mobilities paradigm. Environment and PlanningA 38: 207-226.

UNAIDS. 2010. Global report: UNAIDS report on the global AIDS epidemic.

Valentine, G. 1999. 'Oh please Mum. Oh please Dad'. Negotiating children's spatial boundaries. In Gender, power and the household, edited by L. McKie et al. Basingstoke: MacMillan, 137-154.

van Dijk, R. 1998. Pentecostalism, cultural memory and the state: contested representations of time in postcolonial Malawi. In Memory and the postcolony: African anthropology and the critique of power, edited by R. Werbner. London: Zed Books.

Vaz, P. and Bruno, F. 2003. Types of self-surveillance: from abnormality to individuals 'at risk'. Surveillance & Society 1(3): 272-291.