University of Birmingham Feminine futures: female initiation and aspiration in matrilineal Malawi

A vibrant recent literature documents the challenges of achieving social adulthood in the context of economic decline and high youth unemployment in Africa and beyond. Those who have already made this transition, however, tend to reside in the margins of these texts, casting disapproving shadows. This article seeks to redress the balance by focusing on the signiﬁcant efforts of social adults to shape young girls’ transitions to adult womanhood. In rural Malawi, this takes the form of female initiation rites, which have been adapted to include messages about the importance of formal education and the potential beneﬁts of ‘waithood’. Rather than constituting a problem, delayed marriage is lauded as an alternative to the real social limbo entailed in early marriage and childbearing in a context of widespread poverty. Subtle alterations to initiation rites are thus embraced in an attempt to guide young women towards a more desirable future, at the same time as they are prepared, in moral and practical terms, for lives much like their elders.

facilitate waithood: economic decline and high unemployment (for a recent exception, see Gilbert 2018). According to this view, young women are affected only indirectly as their potential spouses fail to establish themselves in the role of provider, such that they must remain on the proverbial shelf (e.g. Sommers 2012: 6-7). Such residual treatment of young women's hopes and ambitions, and those held on their behalf, is perhaps nowhere less appropriate than in the matrilineal, uxorilocal areas of southern Malawi in which I conducted this research. There, bridewealth is not paid, and marriages are easily contracted, without great expense, and just as readily dissolved. Indeed, on the whole, marital instability is, and long has been, somewhat ordinary (Kaler 2001). Unlike elsewhere in the region (Hunter 2010), the available statistics for Malawi, and especially for matrilineal areas of the country, do not suggest significantly declining marriage rates but do indicate a remarkably high incidence of divorce and remarriage, something my ethnographic work confirms (Johnson 2018;Reniers 2003). The difficulty for young people in this context, then, is not so much the business of contracting a marriage in the first place, but rather that of maintaining a marital relationship in the long term. This is reflected in the concern of elders that their daughters and heirs should be well equipped to establish lasting marriages, and thus that they should actively seek both to delay marriage and to improve the terms on which their eventual entry into marriage will be made. The perceived danger is not that girls might get 'stuck' in a pre-marital zone of waithood, but rather that early marriage and associated childbearing in the context of pervasive poverty would leave them in an even less desirable form of 'social limbo' .
By placing young women at the centre of this study of youth transitions, I question the implications of the waithood literature that marriage is the key to the achievement of social maturity and that 'waiting' is a social problem. On the contrary, a prolonged period of pre-marital youth might rather be understood as a strategy imbued with hope for an alternative future of relative social and economic security. The role of initiation as a forum for shaping the aspirations of female youth also suggests the need to rethink the relationship between initiation rites and social reproduction. Initiation might instead be considered a potent resource for those seeking social transformation since the adult lives elders hope to guide their heirs towards may be quite different from their own.
The adults whose concern for young women's futures is described in this article are at once hopeful and pragmatic, placing their faith in development narratives and yet maintaining a critical distance from the promises of politicians and NGO workers, which have gone largely unfulfilled in the decades since 'democratization' in 1994. Twenty years on from the advent of multi-party democracy and economic liberalization, Malawi languished in 174th position in the United Nations Human Development Index, with 62 per cent of the population living below the international poverty line of $1.25/day (United Nations Development Programme 2014). A total of 86 per cent of the Malawian population reside in rural areas, where smallholder agriculture is the principal economic activity for most people of working age. Small-scale farming is considered 'employment' for the purposes of national statistics, which helps to explain why 80 per cent of the population are counted as employed, of which 64 per cent work in the agricultural sector, and 89 per cent are in informal employment (National Statistical Office 2014). Secure employment in the formal sector remains the preserve of very few Malawian citizens.
While it tended to go unsaid in public discourse celebrating the advancements of the democratic era, privately villagers remarked on the apparent deterioration in the quality of education available to their children following the roll-out of universal primary education in 1994. This was reflected in the poor standards of English obtained even by those who did manage to secure a pass in their final primary school examinations. When universal primary education was launched, the numbers of students enrolled in primary school rocketed by a million in a single year, and they have continued to rise. This unprecedented increase in student numbers was not matched by investment in infrastructure, resources, or teacher training. By 2010, the pupil to trained-teacher ratio had reached 91:1 (Government of Malawi 2011: 21), and pass rates in the primary school leaving certificate were dismal: only 69 per cent of those who sat the examinations achieved a pass in 2009, down from 74 per cent in 2005 (Government of Malawi 2011: 24). Try as they might, very few students in the Chiradzulu villages in which I resided, educated in overcrowded classrooms alongside more than a hundred of their peers, achieved the grades required for admission to the more competitive and better-funded government secondary schools. Those who did were held up as role models, serving to prove that diligence and hard work just might pay off. For students who passed their examinations with lower grades, poorer-quality state or private secondary schools provided what everyone acknowledged to be a second-rate (not to mention more expensive) education (Chimombo 2009;Rose 2005;World Bank 2010: 68). 6 In this context, initiation rites tread the rough terrain between aspiration and pragmatism, celebrating women's fertility and counselling girls in the secret knowledge and skills required for successful marital relationships at the same time as they forcefully convey the message that deferred marriage and further schooling are key to a more satisfying and prosperous future. The degree of hope expressed in developmental futures is remarkable in a context in which local schools and development initiatives are so obviously failing. While villagers and traditional authorities share a faith in the developmental visions of their elected leaders and local NGO officers, they are also aware that such promises are elusive, and likely to remain unrealized for the majority.
It must be stressed that this is not so because their Malawian village homes are somehow untouched by global forces. On the contrary, rural areas of Malawi have long been integral to the broader regional economy, not least as a result of widespread rural-urban and intra-regional migration (Englund 2002;Groves 2012;Johnson 2017). These days, of course, global connections are also fostered by the ubiquitous mobile phone. Nevertheless, far from the West Africa that Charles Piot describes, marked by 'the death of tradition, [and] the end of chiefship ' (2010: 168), this is a setting in which traditional practices continue to offer potent means of influencing the future (James 1999;Kratz 1993).
In what follows, I introduce an ethnographic vignette illustrating the degree to which initiation rites are reflexively debated in Malawian village settings and their enduring importance. This is followed by an ethnographic account of the rites themselves, as observed in 2009, 2010, and 2015. Women's reflections on the shifting emphases of initiation rituals inform a discussion of the felt tensions between the preparation of young girls for marriage and the encouragement of marital delay: the desire to bring about a transition to adulthood, on the one hand, and to prolong youth as a means of striving for a future that might fulfil the promises of the democratic transitioneducation, employment, 'development' , and 'self-reliance' -on the other. At issue is the question of whether feminine 'waithood' ought to be overcome or embraced, and what kind of 'social limbo' ought to be feared.

Initiation under fire
By January 2010, the maize that had been planted just two months earlier was wilting in the fields of southern Malawi. The rains were long overdue and the prospects for a decent harvest looked bleak. In Chiradzulu district, the Group Village Headman (GVH) encountered above called a meeting of the seventeen villages under his jurisdiction. Urging the men and women gathered to acknowledge that things did not look good, he asked: 'What shall we do?' Many of their suggestions were infeasible. As the GVH pointed out, they did not have the necessary seeds to replant their maize, nor did they have alternative staple crops that could be planted, and the majority of villagers did not have access to wetland gardens where maize could be expected to fare better. The GVH suggested that villagers modify their meal sizes, cook smaller portions, and ensure that they did not waste nsima (the staple maize porridge) while others went hungry. He proposed planting peas as a cash crop to ease family finances, so that money might be raised for the purchase of maize later in the year. Then he gestured towards traditional celebrations: 'What about initiations, memorial feasts, and so on?' he asked, implying that these annual events were a cause of profligacy: 'What can we do to take care of our supplies?' Sensing that their initiation ceremonies were under fire, female villagers spoke up: 'It's not just women who waste food; men also sell maize to buy beer' , they pointed out. For their part, male villagers accused women of disguising maize in bundles of vegetable leaves so that they could sell it unbeknownst to their husbands. A male voice suggested that 'there should be no initiations this year' . Speaking for many, he complained that when men become angry with their wives for selling maize in order to buy the cloth wraps they give as gifts on these occasions, or for taking maize to the initiation hosts, their wives respond, Bola lithe! 'It would be better if the marriage ended!' A woman who had a daughter she hoped to initiate that year got to her feet to protest. 'Some of what is being said is true' , she conceded, 'but some is lies. Wives discuss things with their husbands, and some of us buy the maize that we take to initiations' . At this point she was waved down by the GVH, who insisted that since he had been born and raised in this village, nobody could tell him that women buy the maize they take to initiations! A male voice called out: 'Initiations should not be put on hold' , and he was applauded for his contribution by the women. Encouraged, he asked whether it would be possible to initiate girls without cooking nsima. A cacophony of voices insisted that it could not happen, but slowly women began to venture that the initiators might simply advise without nsima being prepared, as they do for the litiwo initiations conducted for women in the late stages of their first pregnancies.
The GVH made clear that he did not want to enforce a rule from on high. 'But' , he added, 'although some couples do discuss the management of their maize supplies, you women have the greater power. Perhaps we should just pause for one year and resume again next year' . Otherwise, he suggested, he could rule that all those who wanted to hold an initiation must pay K1,000, as opposed to the standard payment of K300 to the Village Head. 7 There was disquiet at this idea, and a woman challenged: 'In that case, will you pay us a goat if our daughters become pregnant before they are initiated?' , in reference to a rule already in place imposing such a fine on the parents of uninitiated girls who fell pregnant. Riled, the GVH retorted that anyone who planned to hold an initiation that year must have plenty of maize: 'In which case, I will rally the Village Heads and convene large numbers of villagers to go and eat at the host's homestead!' Female villagers were not prepared to allow the discussion to focus solely on girls' initiations. They began to point out that while their sons undergo initiation seclusion for two long weeks, they must cook much more food to feed them at the seclusion camp than they ever would at home. The GVH expressed some sympathy for this argument, and then he asked: 'What is initiation?' The unanimous reply: Miyambo, 'Traditions' . 'So what is all this carrying-plates-on-the-head business, then?' he asked, performing a comical mime of a woman rushing to an initiation ceremony with a stack of plates of dried maize balanced on her head. The women's response: Kucheza, 'Chatting/sociability' . The GVH pounced: 'Then we can cut that part out' .
Villagers continued to offer alternative suggestions until the GVH drew matters to a close, saying that he would give the women some time to think and, in a month or so, they should present him with a unified position. 'The problem with hunger' , he added, 'is that it will cause us to fail in the work of development' .
Moving on, but only slightly, he turned to the issue of young girls marrying. He stressed that he did not want women to hurry to give their daughters gardens and have them marry too young. 'If girls get married in Standard Seven [of primary school], will we prosper?' In the event, the rains returned just in time to revive much of the crop and the GVH did not enforce a ban. Nevertheless, as we shall see, the uncertainty generated by the meeting invited reflexive contemplation of initiation practices, and the articulation of changing aspirations for young girls' futures. It also highlighted the strength of feeling of the majority of villagers, who held not only that female initiation remained an essential preparation for adult life, but also that the social importance of initiation extended beyond the serious business of passing on traditions to their daughters and matrilineal heirs. The ambivalence expressed by male villagers can be read as a barometer of the powerful female sociality that is augmented, expressed, and celebrated through initiation practices (Werbner 2009: 452), thus indexing the 'greater social and political space' (Peters 1997: 133) accorded to women in this contemporary matrilineal setting, and challenging the assumption that women's association with 'culture' or 'tradition' is necessarily a source of their subordination and exclusion (Ribohn 2002).
Despite the recent emphasis on the so-called 'crisis' of 'youthful masculinity' (Weiss 2004: 11) in contemporary Africa, and a corresponding focus on male maturation and rituals of manhood (Ngwani 2001), the privileging of female initiation in this article mirrors Malawian villagers' own emphasis on the paramount importance of female rites. 8 Whereas it was tacitly acknowledged that boys' non-initiation could be overlooked, for girls, failure to be initiated would threaten marriage prospects and diminish their chances of establishing successful and long-lasting relationships. It would also leave them dangerously unprepared for the ordeals of pregnancy and childbirth. In addition, I was told by men and women alike that the traditions passed on to girls are more imperative to the well-being of the wider community than are the lessons of male rites. Women bear greater responsibility, for example, for ensuring compliance at times of enforced sexual abstinence (such as during the illness of a child), which, if neglected, put the health of those around them at risk. It is girls' future responsibilities as wives and mothers, and the centrality of their roles in wider social reproduction, that give their initiation training particular urgency, even if there is considerable ambivalence about the desirability of those roles for young women who might instead, through marital delay, redefine norms of adult femininity.

Unamwali wa ndakula
In the area of this study, three initiation (unamwali) ceremonies are conducted at different stages of a woman's life -a fact that serves as a welcome reminder that puberty rites do not mark a singular or straightforward transition to social adulthood (Johnson-Hanks 2002;Lutkehaus & Roscoe 1995). 9 The first, unamwali wa ndakula, consists of puberty rites conducted for girls who have recently begun menstruation. The second and third rites are referred to as litiwo, subdivided into that 'of pregnancy' (ya mimba) and that 'of childbearing' (yotulutsa mwana). 10 These are held in the final stages of a woman's first pregnancy, and some years later when she has one or several young children. 11 With the exception of the pregnancy rites, during which a woman's husband may also be instructed, men play no role in these female-only ceremonies. This article focuses on the first of these rites, the ndakula ceremonies, which are both the most prominent and the most contested, dealing as they do with the formation of young women prior to marriage.
In discussing these rites, I draw upon Audrey Richards' (1982Richards' ( [1956) classic study of the Bemba chisungu ceremony performed in neighbouring Zambia. In testament to the ethnographic richness of Richards' text, it has afforded a number of reinterpretations since its publication. Different readings have sought to place chisungu in the context of a wider body of ritual (Ruel 1997), and to question the notion of initiation as a largely educative process, concerned with socialization (Strathern 1993). Marilyn Strathern thus argues against the idea that initiation constitutes the person as somehow more 'complete' , suggesting instead that by producing an image of a marriageable person, initiation rites could be understood as rites of 'decomposition ' (1993: 48) that bring to the fore certain aspects of a person's social make-up (in the case of the Daulo people of Highland Papua New Guinea, their agnatic relationships), while occluding other constitutive relations that are less relevant to the girls as brides. While Strathern's analysis can be seen as pertinent to the Malawian rites, my concern here is with the explicit ways in which adults in contemporary Chiradzulu approach initiation rites as a means of influencing young women's futures by simultaneously preparing them for marriage and urging marital delay. The anticipated effects of initiation in this case are ambiguous: making girls visible as marriageable young women is only part of an equation that also emphasizes the potential desirability of 'waithood' or the deferral of the transition to social adulthood.
During my fieldwork, ndakula initiations took place during the dry season, between July and September each year. Generally speaking, a pair of initiation instructors (alangizi) were called to 'dance' for between one and three initiates (anamwali), who would be accompanied by at least one tutor (phungu), selected by the host mother. The other participants were initiated women of all ages whose energy and dynamism in leading much of the singing and dancing, and in assisting the girls' mothers with the chores associated with hosting the ceremonies, were essential to the rites' success. An emphasis on respect for elderhood in initiation teachings did not entail a concern for seniority and rank during the rites (cf. Werbner 2009). Rather, the ceremonies were marked by a relaxed, intergenerational sociality and a great deal of fun for all but the initiates themselves. For the most part, the latter were required to remain still, their legs outstretched, backs straight, and eyes lowered. 12 As with female initiation rites in many South-Central African societies, ndakula is associated with the onset of menstruation, and many of the teachings relate to the implications of the girls' periods. However, the rites are performed for girls at the age of about 15 or 16, ideally those who have reached the final year of primary school (Standard Eight), who may well be older. Indeed, the GVH had long since made clear that girls under the age of 15 should not be initiated in the area of his jurisdiction. For perhaps the majority of girls, then, initiation occurred some years after menarche (Munthali & Zulu 2007), but the fact that the girls already knew much of what they were instructed was not held to diminish the importance of either the more practical teachings or the more allusive aspects of the rites.
Initiates were advised and instructed in a variety of modes and through sensory engagement with what Corinne Kratz (1990) has called 'the secrets of sight and sound' , including song, dance, 'gymnastically demanding' (Werbner 2009: 446) physical routines, and bodily contortions. Some of what they underwent might best be described as 'ordeals' in the terminology of Richards, but Signe Arnfred could just as well be describing large sections of the Chiradzulu rites when she brands Mozambican practices 'participatory theatre' (2011: 147; see also Moore 1999: 12-13). Through shared bodily performance, song, dance, laughter, role-play, and games, female identity, female 'community' , and the womanhood of the initiates were produced and celebrated alongside an embodied sense of female sexuality, fertility, and collective responsibility for the maintenance of tradition. Particular stress was placed on teaching the girls to duka, to perform a dance in which they rotate their hips and pelvis in the manner in which they are told they must during intercourse (Arnfred 2011). Indeed, the women's emphasis on the joys of mutually pleasurable and socially generative sexual relations served to undermine dominant narratives about sex in Africa as 'a social problem . . . devoid of meaning' (Spronk 2014: 5).
The rites began with a ceremony at which maize for the large meal of nsima and beans to be shared on the final day was pounded communally and the first advice was given to the initiates. The initiates were assisted in pounding the first mortar of maize in a ritualized manner, symbolizing, as one initiator explained to me, that they were making a public promise to leave their childish behaviour behind them. Several of their early actions and ordeals were said to serve such a purpose, simultaneously marking off the time, space, and events of initiation as distinct from the ordinary routines of daily life.
The first advice imparted by the alangizi centred largely on menstrual hygiene and taboos, as well as the need to behave in a respectful manner in the presence of elders and traditional authorities. Other women joined in by placing coins on the initiates' laps, commenting on their character or behaviour as they did so, and thereby providing fuel for the alangizi's praise and admonition. They might also branch out to comment upon the comportment of the girls' mothers and close female kin, supplying specific examples of praiseworthy or reprehensible behaviour, and urging their continuation or reform.
The initiates then entered seclusion in a house in the compound of the host mother. The house was known as tsimba during this time, and the girls could be visited there by their initiated friends, relatives, and neighbours, who, along with their tutor(s), would school them in the secret knowledge, songs, dances, and challenging routines of initiation. In line with what readers of Richards' classic text would expect, much of what was learned during seclusion was not readily assimilable, coming as it did in forms that might at first appear to be incomprehensible to the initiates (Richards 1982(Richards [1956: 127). Girls were taught miyambo (literally 'traditions' , 'customs') and zisimosimo (riddles, or allusive linguistic formulae). The former are lyrical stanzas with a consistent form and structure, and pertain to menstruation. The latter are pithy phrases, posed as questions for the initiates to answer, and conveying cryptic lessons about marital life or agricultural work. 13 Linguistically challenging and requiring decoding, miyambo and zisimosimo are highly valued by women, and their familiarity with these varied and evolving genres serves as proof that they have been initiated. It was from their tutors and peers who visited during seclusion that the initiates learned the majority of the miyambo and zisimosimo. 14 Given their somewhat obscure nature, and their sheer quantity, girls were unlikely to grasp them all the first time they encountered them. As is also the case with initiation songs, role-plays, and physical routines, it is through their subsequent participation in the rites of their friends, relatives, and neighbours that women acquire more rounded knowledge, which they will go on to help future initiates master (McNeill 2011). Visiting friends and kinswomen initiated elsewhere in Malawi were also welcomed, indeed encouraged, to participate, as Chiradzulu women were made welcome when they travelled to attend the ceremonies of their own kin and associates. 15 This is only one of the more obvious routes through which new lyrics, routines, and miyambo become incorporated into the flexible repertoire available to participants, enabling reflexive shifts, responsive to and transformative of changing social conditions (James 1999).
Seclusion could last anywhere from one to five days. Its end was heralded by a night-time ceremony characterized by tireless singing, dancing, and advice-giving by the alangizi and young initiated girls. During the night, the initiates' mothers were also summoned indoors and subjected to the alangizi's advice while the initiates and younger women were occupied outside. Early the next morning, the initiates were taken to the river to bathe, as they had each morning during their seclusion. On this final day, however, they would remain at the riverside, where the most revered advice concerning married life would be imparted, and much of what had already been demonstrated, taught, and explained would be repeated. Activities introduced by the alangizi towards the end of the ceremony served explicitly to mark the initiates' acceptance of their changed status and their abandonment of their former ways.
By the middle of the afternoon, the riverside party would begin the shift back to the home compound, where the host mother and a large number of her close kin, friends, neighbours, and guests had finished cooking and awaited their return. First, though, the initiates were required to bathe one last time in the river, following which they were dressed in new clothes and rubbed with cooking oil, before finally being paraded home, arriving on the shoulders of their initiated friends. Back at the homestead, great quantities of nsima and beans were distributed to guests using the same plates and bowls that had earlier been used to bring gifts of dried maize and small sums of cash to the host. The final spectacle constituted the gifting of the initiates, their tutors, and close matrilineal relatives. The most common gifts included cash, cloth wraps, plastic plates, soap, and sugar. Proceedings would begin with the host mother's public gifts to the initiates and tutors, and were quickly transformed into a prolonged and bustling celebration as women reciprocated the gifts they themselves had received from the host and her kin at previous initiation ceremonies. 16 Catholic Church-endorsed ceremonies in the area were remarkably similar to their village equivalents. 17 These rites were conducted by recognized church alangizi, who were lay members of the local congregation. The similarity of the church and village rites was reflected in the fact that mothers showed little concern as to whether their daughters attended the village or church ceremonies: 'It doesn't matter, they are the same' , one woman replied when I asked her which kind of initiation she would prefer for her adolescent daughters. My argument as regards the ongoing importance of female rites, and the historical shifts they have undergone, is as relevant to church-endorsed initiations as it is to those held in the villages. Indeed, the ethnographic distinction between 'church' and 'village' initiations implies a false dichotomy because all the ceremonies discussed here are local and rural, as are the participants. 'Church' alangizi could be, and were, recruited to lead 'village' ceremonies and the 'village' alangizi I observed were Christians; it was not considered inappropriate for them to advise their charges to be good Christians and to pray to God throughout their lives.
A moral education 'Counselling' (kulangiza) lies at the heart of initiation, and is the job not only of alangizi, but also of the other initiated women in attendance who speak as they place coins in the initiates' laps, praising them for commendable conduct (such as fetching water for their elders, or behaving with decorum in the presence of their elder brothers), and reprimanding disrespectful behaviour, laziness, or the chasing of boyfriends. The tailoring of this advice to the particular initiates accounts for the degree of variation across ndakula ceremonies. Should the women's contributions portray an initiate as disrespectful or foolish, for example, she would be handled more roughly, and subjected to additional trials and verbal abuse (see also Longwe 2006: 61).
At stake here was the girls' moral personhood: as adult members of society, they must be respectful, hard-working, kind, and generous; they must not entertain greed or gossip; and they certainly must not steal. Emphasis was placed on 'respect' (ulemu), which entailed the girls living well with others: treating their parents, elders, traditional leaders, future husbands, and parents-in-law in the appropriate manner; abiding by rules or taboos that would prevent harm coming to those around them; and observing shared norms in relation to the property and privacy of others, thereby ensuring at once their own moral personhood and the harmonious co-existence of kin (cf. Rice 2017). This kind of advice, praise, and admonition was also directed at the girls' mothers and close kin and would be repeated at the girls' subsequent initiation ceremonies and yet again some years later at their own daughters' rites. Such moral commentary constitutes a powerful means of establishing, maintaining, and regulating a (loosely defined) moral community of women.
Folded into these moral teachings were messages about the importance of educational achievement and the need to avoid HIV and early pregnancy. These could be as simple and direct as the command: 'Continue at school' , or more elaborate: 'There are no doctors at Chiradzulu hospital who come from this district, but you, if you work hard, could secure such good jobs. Marriage and pregnancy are not what you should be thinking about at this time' . Alangizi also encouraged girls to consider training as nurses: 'You'll have no need of a husband until you've accomplished that, and then you can marry a man who is also well educated' . Education and employment were frequently lauded as strategies for delaying marriage and ensuring a satisfactory future partnership with a like-minded man.
Moral education bridged seemingly incompatible messages about educational achievement and sexual and married life. Whatever the future may hold, it seemed, the women's commentary on the girls' characters and conduct, and their tailored recommendations for reform, would stand them in good stead. Here we see the pragmatism of social adults as they guide their matrilineal heirs towards uncertain futures: on the one hand, they encourage the ambition to realize the promises of education and 'development' , while, on the other, they prepare young girls for lives much like their own.

Shifting aspirations for feminine adulthood
Despite the above descriptions of enduring practices, I do not mean to imply that unamwali is a timeless tradition: far from it. Most of my informants, however, acknowledged shifts in their practices only after some prompting, and maintained that those changes that had occurred were merely superficial. A former initiation instructor provided an interesting exception. Mayi (i.e. Mrs) Chisale had retired from her role through illness and, by the time we met in 2010, it had been several years since she had been able to attend an initiation ceremony. Mayi Chisale recognized the same shifts in initiation practices that were reluctantly acknowledged by other alangizi, but attached a strong moral judgement, regarding certain elements of the rites (now largely abandoned, see below) with a level of explicit scorn that was not widely shared. Aware of this, she explained to me that those other alangizi had not attended the same Salvation Army training courses that she had undertaken in 2004, ten years after the coming to power of Malawi's first democratically elected President. While this was true, she failed to mention that a good number had attended similar sessions organized by other NGOs. These trainings were offered at a time in which the ravages of the HIV/AIDS epidemic had made themselves felt throughout the country and projects of civic education and participatory development were in full swing, attracting considerable resources and attention (Englund 2006;Johnson 2012).
I was surprised by Mayi Chisale's vehemence when she told me that 'from 1968' , when she had first begun assisting her aunt in leading initiations, 'up to now, initiation has changed, we have abandoned everything from before because those traditions were bad' . She went on to specify a number of traditions that had been rejected, laying particular emphasis on the practice of 'removing the dust' (kuchotsa fumbi), in which young initiates were told that in order to shed the dust of initiation and be beautiful once more they must engage in sexual intercourse after their ceremonies. Indeed, alangizi of all stripes had ceased advising girls in this regard, as they had also sought to discourage promiscuity in the wake of the HIV epidemic. For the same reason, the practice of a girl's mother and father engaging in intercourse following their daughter's initiation, so that the mother would be in a 'hot' state and could prepare a protective meal for their daughter early the next morning, was also downplayed because it presented a conundrum for single mothers and the wives of absent men, who might feel obliged to engage a fisi (literally a hyena, but here meaning a man who will perform ritual intercourse). 18 Mayi Chisale informed me that, in addition to promoting awareness of the risks of HIV, the aim of the Salvation Army training she had received was that 'we should take the traditions of today, those that encourage children with their schooling' . While she was rare in her reflex to highlight the newness of the emphasis on formal education, other alangizi were happy to impart the same advice. As a church initiator put it: 'Indeed [girls] should continue at school, and with praying, because school is their future' . She Many of the recent adaptations of the advice given to young girls at ndakula ceremonies, subtle as they are, correspond to shifting aspirations for the girls' future lives as adult women. As we have already seen, these aspirations centre on a durable faith in formal education as a means of achieving economic security and hence of transforming gender relations by providing women with reliable cash incomes. However, while women tirelessly impart messages about schooling throughout the long ceremonies, there remains for many an uneasy sense of tension between initiation and their hopes for 'development' (chitukuko).
I lost count of the number of times men and women told me that girls leave school and 'chase' (thamangira) marriage soon after initiation because they have been shown the secrets of married life and are impatient to put into practice all that they have learned. It was acknowledged that this was nothing new, but there was a sense that it was more worrying nowadays than it had been for previous generations. For one thing, HIV complicated relationships between men and women and, in addition, in the words of one woman: 'It's like the world has turned upside down (dziko latembenuka) and these days girls should stay at school' . In contrast to the scenario outlined in the literature on waithood (e.g. Jeffrey 2010; Mains 2012; Masquelier 2013), for many Malawian villagers it seemed that the surest means of getting 'stuck' was not the inability to marry and consequent social limbo, but rather early marriage and childbearing resulting in fragile relationships and inescapable household poverty. Waithood, then, the drawn-out life stage between adolescence and full adulthood, was the potential key to the hoped-for future of adult prosperity and marital success.
For his part, the GVH's concerns linked initiation with the broader challenges of governance. He made his position clear at a meeting of women from across his seventeen villages during which he berated the assembled women for singing songs related to sexual intercourse at girls' initiations. (This they denied, saying that they simply sang because they were enjoying themselves.) At the close of the meeting, the GVH took to the floor to tell the women he had some messages for them to take back to their respective Village Heads; the most important of these concerned the fact that, as he saw it, too many children were avoiding school. He thus declared that the parents of those who failed to attend would be fined three goats. 'Going to school is important' , he insisted, and his reasoning could not have been more serious: 'the gardens are insufficient' . Gesturing towards the absence of uncultivated land, the GVH reminded villagers that the subdivision of plots was becoming less straightforward. While children who succeed at school will one day buy their own plot, he said, 'girls who concentrate on the activities of the bedroom just get pregnant and then you have to give them a garden' . What he wanted to avoid, he explained, were conflicts over land that would lead to serious accusations of witchcraft (see, e.g., Peters 2002). Hunger and conflict could only undermine his efforts to co-ordinate development work. He wanted his villagers to be strong, able and willing to labour on projects to bring bridges, schools, and health centres to the area. He also wanted young villagers to pursue educational advancement, so that they might generate prosperity for themselves and their kin.
As we saw above, the GVH counselled his own young female heirs in much the same way. 'These are the days of school' , he told his sisters' daughters: If you finish school, there are colleges; do you think I can't find the money for your fees? In just two years you can train to be an agricultural adviser . . . or you could do teaching . . . You will be self-reliant. Then you can marry and have good marriages, marriages characterized by development. Otherwise, your marriages will be marked by poverty: we will give you gardens and you will grow maize. 19

Journal of the
Like their traditional leaders, villagers hoped that their daughters would enter marriage from a position of strength, with the means of providing themselves with a cash income to supplement their agricultural resources, and the ability to choose a spouse who would prove industrious, resourceful, and supportive. Their fear was that girls would be bounced into marriage by early pregnancy and economic dependence and that this would mean a future of poverty, marital discord, and perhaps even HIV. As we have seen, at various points during initiation ceremonies alangizi expressed similar aspirations for the girls in their charge. Such concerns go beyond initiation rites and lie at the heart of villagers' aspirations for the future of their matrilineal society.
Young girls themselves differed in their reactions to these messages, as illustrated by the experiences of two sisters aged 19 and 21. 20 The younger sister delighted her matrilineal kin with her, frankly scandalous, response to her first marriage proposal: roughly speaking, 'Not even if the earth farted!' (Olo pansi pataphwisa!). Retelling the story, she waited for her audience to regain their composure, and then explained her decision by way of the rhetorical question: 'Am I supposed to go to school and think about marriage?!' Thus, she made clear where her priorities lay. Indeed, at the time, she was in the process of repeating the final year of primary school in the hope of securing a place at a more prestigious state secondary school. Her sister, on the other hand, despite continuing her education at a less competitive local state secondary school, frustrated her relatives with her vocal and defiant loyalty to her long-term boyfriend, who she maintained was her intended spouse. Try as they might, nobody could persuade her that her time would be better spent on her studies than with her boyfriend, who, to their dismay, had dropped out of school and was earning money by means of agricultural piece-work (ganyu). The diversity of voices and differing lifechoices of young women in these Chiradzulu villages are indicative of the difficulties of reconciling conflicting aspirations and the challenges involved in achieving adult womanhood in the contemporary era.
The faith that Malawians express in development is not blind. They are well aware that the promises of education, employment, and self-reliance are elusive; that the conditions of poverty undermine their efforts; and that local schools are woefully over-stretched and under-resourced. They might thus be said to hedge their bets, urging their daughters to strive for a 'modern' future at the same time as they equip them for a life much like their own, enriched by the playful sociality and potent sexuality that is renewed annually during the initiation season, and protected by the transmission of traditional knowledge that will keep them and their kin safe from harm. Channelling their hopes for modern futures through the rites, they worked to undermine the sense that initiation threatened to hasten the end of their daughters' education and open the doors to marriage. This compromise was also evident in the GVH's edicts: no girls were to be initiated until at least the age of 15; but just as he would fine the parents of children who did not attend school, he would also charge those whose daughters embarked upon motherhood before they had undergone unamwali rites. Ambivalence runs through the practices and aspirations described here. The future for today's young women is uncertain, and while prolonging 'waithood' by delaying marriage and childbearing is encouraged in the explicit hope that girls will thereby avoid the kind of 'social limbo' that young wives and mothers face as they struggle to establish households in conditions of severe poverty, people are well aware that, despite their best efforts, dreams of development are likely to remain just that.

Conclusion
While a vibrant recent literature has documented 'the anxiety permeating young people's attempts to imagine what lies ahead' (Masquelier 2013: 471) as they struggle to attain recognized social adulthood, I have sought to draw parents and elders into the picture, listening beyond their disparaging remarks to explore their considerable efforts to shepherd young women towards desirable adult lives. Parents and elders in Malawi could be scathing of young people who were seen as lazy and wasteful of the free primary education denied to previous generations; in this mode they resembled the adults glimpsed at the edges of recent anthropological studies of the predicaments of youth. But the same people also invested a great deal of energy in advising young women about the kinds of future lives to which they might aspire, and initiation ceremonies constituted a vital forum for the dissemination of such advice. Traditional rites, valued for the continuity they represent with past generations, provide opportunities to adapt the ways in which young people are prepared for the future in light of new visions of what that future might hold. Widening access to education, and intimate knowledge of the risks of poverty and HIV, feed into the advice they impart, and imbue adults' hopes and dreams for younger generations with a sense of 'ambivalent expectation' (Masquelier 2013: 481), since they are at once hopeful and all too aware that the odds are stacked against them.
By focusing on aspirations for young girls' futures, I have eschewed the dominant emphasis on 'masculine waiting' (Jeffrey 2010: 33), refusing to see young women's fortunes as merely residual to those of their potential spouses (cf. Sommers 2012). As this article demonstrates, shifting the focus from male to female youth decentres marriage as the key to the successful achievement of social adulthood. Rather than striving for marriage in the short term, young women are encouraged to further their education as a means of marital delay and in the hope that they might thereby make 'good' marriages in the more distant future. If anything, for young women in this context, 'waithood'the prolongation of the life stage between adolescence and marriage -was to be desired over the kind of 'social limbo' that early marriage and childbearing was likely to entail in a context of widespread poverty and reliance on subsistence agriculture.
Kuti munthu akhale munthu, ndi unamwali, 'What makes a person a person is initiation' . These were the words of Mayi Chisale, despite her condemnation of the 'old traditions' . Her words echoed Carolina Nordstrom's informants in Mozambique who described initiations as 'the ceremonies that make us human ' ( : 180, cited in Arnfred 2011. When I brought up the possibility that the GVH might ban initiation, she told me she thought that would be a mistake: 'Initiation should continue so that children can know the ways of the ancestors . . . Without them, people will just be like sheep! They'll just be like wild animals!' It was the rites' emphasis on moral education that Mayi Chisale emphasized here, signalling their importance for the maintenance and renewal of rural sociality. As they come together to advise their heirs, women augment, express, and celebrate the vital importance of their bodies and their traditions for the reproduction of their society. In so doing, they renew a reflexive sense of moral community that informs changing expectations for women's social roles and gender relations, at the same time as they articulate a desire for their daughters to reap the rewards of 'development' by pursuing formal education and employment, and thereby eschewing dependence within marriage. Rather than watching on, helpless, from the sidelines, adults strive to play an active role in shaping youths' futures. They do so in the knowledge of the tensions embedded in the contradictory messages they impart (tensions between 'tradition' and 'development'; preparation for marriage and the urging of marital delay; emphasis on formal schooling despite awareness that local schools are barely fit for purpose), for these are the vicissitudes of life in the twenty-first century that their daughters must navigate in the months and years to come.

NOTES
Versions of this article were presented at the University of Birmingham, the University of Cambridge, the University of Sussex, SOAS, and the European Association of Social Anthropologists Conference, Tallinn. I am grateful to the respective audiences, as well as to Maxim Bolt, Harri Englund, and Alice Wilson for their comments on earlier drafts. The bulk of the fieldwork on which this article is based was funded by the ESRC.
1 All names are pseudonyms. This article is based on fieldwork conducted from January 2009 to September 2010 and from April to October 2015.
2 Khat (Catha edulis) is a leafy plant that can be chewed as a mild stimulant. It is particularly popular in the Horn of Africa.
3 The term 'social adults' refers to those who are recognized as having made the transition to full adult personhood, generally through marriage and the assumption of responsibilities for the welfare of others.
4 Comparative reading indicates a degree of variation in the extent to which particular rites have been adapted to changing circumstances (see, e.g., Bledsoe 2005Bledsoe [1990 5 There is considerable scope for future research to incorporate a stronger focus on youths' perspectives on initiation practices, alongside their commentaries on their elders' efforts on their behalf. Such work would complement studies emphasizing young people's aspirations for the future. 6 Nevertheless, I did not encounter much ambivalence towards formal schooling of the kind recorded elsewhere in the region (see, e.g., James 1999: 127-31; Sommers 2012). 7 At the time, £1 was worth approximately K240 (Malawi Kwacha), and US$1 roughly K150. 8 There are two key ideas entailed in the idea of a crisis of youthful masculinity: that male youth have become trapped in a state of perpetual economic dependence; and that young men constitute a potentially violent and disruptive force that increasingly escapes social control.
9 Female initiation rites in southern Malawi are generally referred to in the literature by the Chichewa term chinamwali. My informants tended to interchange chinamwali with unamwali, a word with the same stem but a prefix that signals greater abstraction. There is a significant degree of local variation in initiation practices. The account given here should not be taken as a general description of initiation in Malawi as a whole.
10 See Dicks (2012: 152-4) for an account of a litiwo pregnancy rite. 11 I have attended ten ndakula ceremonies (including one Catholic Church-endorsed initiation), three sessions of litiwo pregnancy rites, and three of the third type of ceremony. I recorded interviews with three female initiators and engaged in innumerable informal discussions about initiation rites with villagers and initiators, male and female alike.
14 In at least one of the initiations I attended, the initiates were provided with a hand-written list of miyambo by an initiated friend and instructed to study them while in seclusion.
15 Irrespective of the 'style' of their own ceremonies: whether or not they were endorsed by religious institutions (see below) or associated with different regional or ethnic traditions.
16 These returns were made at approximately double the value of the previous gifts. 17 Religious authorities in Malawi have oscillated between attempts to encourage 'Christianized' rites and to enforce initiation bans (Chakanza 1998;Chingota 1998;Mair 1951a: 106-7;K. Phiri 1983: 268). Irrespective of denomination, church influence over initiation practices has been minimal and attempts to transform rites in line with Christian conceptions of 'decency' have been largely ineffectual (Fiedler 2005;Longwe 2006;I.A. Phiri 1998). That said, church influence has been greater in some parts of the country than others. This