On-line version ISSN 2309-9585
Print version ISSN 0259-0190
Kronos vol.39 n.1 Cape Town Jan. 2013
Lingundumbwe: feminist masquerades and women's liberation, Nangade, Mueda, Muidumbe, 1950s-20051
History Department, University of the Western Cape
In the aftermath of the war for national liberation, a group of Makonde women guerillas invented a flamboyant feminist mapiko mask, venturing in a terrain traditionally reserved for men. The invention spread throughout the plateau, becoming the signature dance of a generation of women empowered by the revolution. This essay reconstructs the history and fortunes of lingundumbwe, locating its roots in the experiences of the liberation struggle as well as in the late colonial era, and discussing the conflicts surrounding the invention and its eventual demise in favour of an apparently less provocative genre. The history of lingundumbwe offers a vista into the affective and aesthetic dimensions of the Mozambican gendered revolution - one which escapes the linear temporality and neat binaries established in the scholarship that has so far addressed the theme. Methodologically, the essay argues for a holistic interpretation of song-and-dance that is based on vernacular concepts and that privileges performance and the interplay between its various facets, in order to render expressive and affective complexity.
The participation of women in the anti-colonial guerrilla war through which Mozambique achieved Independence has the status of a myth within the great myth of Frelimo's Armed Struggle for National Liberation. The academic version of the narrative was laid out in two publications penned by the two main agents of historical production in socialist Mozambique, the History Workshop at the Centre for African Studies and the Isaacmans.2 The two articles follow a similar line of argument. The narrative begins with Portuguese colonialism, which is held responsible for intensifying women's subjection through a variety of interventions: forced labour, which reduced women's capacity for working on subsistence crops and therefore threatened household livelihoods; migrant labour, which disrupted social structures by separating husbands from wives for long periods of time; proletarianization, especially in the urban areas; sexual violence, objectification and prostitution; the use of women's bodies as tokens for the payment of the hut-tax; the exclusion of women from the system of education and more generally from social promotion; and the imposition of patriarchal universalistic religions, especially Catholicism. Pre-colonial social structures were however not found faultless: the Oficina paper especially portrays colonialism as an 'intensification' of relations of gendered exploitation embedded within traditional kinship systems, themselves 'finding their roots in slavery'.3 This interpretation echoes the slogan, 'five hundred years of Portuguese colonialism' that was a cornerstone of Frelimo's rhetoric.4 In the papers, Frelimo comes to rupture this long history of exploitation and subjection in two phases. In the first years of political mobilisation and struggle (1962-1967), the emancipation of women was promoted first through the involvement in underground political mobilisation and then in supporting the guerrilla; through an explicit pronunciation in favour of the 'social and cultural development of the Mozambican woman' in the first Frelimo congress; and through the creation of Lifemo, the Feminine League of Mozambique, in 1966. But this initial phase was still ripe with contradictions; women were still confined to the old roles of 'producer, re-producer and source of sexual satisfaction'.5 The real breakthrough would occur with the foundation of a female military corps, the Destacamento Feminino (Female Detachment, henceforth DF) and with the training of women guerrillas abroad in 1967. The rest of the story is blurred within the ideological mists shrouding the 1968-1969 crisis, which in Frelimo's own account saw the victory of the revolutionary line championed by Samora Machel against the petty-bourgeois reactionaries rallied around Uria Simango.6 The vanquished were accused of withholding gender transformations and fostering a patriarchal ethos within the movement. Under Machel's leadership, Frelimo came to understand that the problematics of the liberation of women was 'inextricably linked to the Revolution'7 and that 'the struggle for women's emancipation could not be carried out autonomously from the broader struggle to destroy the colonial capitalist system'.8 The two articles accordingly portray the DF as a vanguard of the Mozambican revolution - perhaps the vanguard - pioneering gender transformation in the Liberated Zones against male chauvinism and entrenched cultural mores and spearheading the Mozambican Women's Organisation, founded after Independence in 1976. The end of the story is tinged with sadness. Both articles consider the Liberation Struggle as the peak of a descending trajectory, the 'recoil from the conquest of the Armed Struggle' as part of the demise of Mozambican socialism itself.9
The narrative established in these two inaugural publications was enriched, but not substantially challenged, by the few works that broached the topic in the years to follow. An unpublished report from the ARPAC details the process of enrolment in the struggle, and the importance of traditional ideas of sexuality in shaping the role of female guerrillas.10 Harry West demonstrated the role of culture in mediating the subjectivity of women guerrillas, liminal child-soldiers empowered by their military status.11 A recent book produced by the State-sponsored Centre for the Research of the History of the National Liberation Struggle (CPHLLN) contains a number of interviews and life-stories with prominent guerrillas that fundamentally reinstates the official narrative, stripped however of all tragedy and imbued with a renewed triumphalism.12 A project that collected struggle life-stories was carried out by the Historical Archive of Mozambique (AHM) and finds its first outlet in the present volume, in Bonate's article, which demonstrates the role of Muslim women and their religious convictions in the struggle.13
All these projects were fundamentally bound to the limits of oral history. They had to deal with the monopoly of lived experience held by a few authorised oral sources, often the wives of prominent politicians;14 with interviewees' fears of contradicting the official narrative, in a situation in which Frelimo is fiercely reclaiming it;15 with the tendency of official tropes to slip into life-stories and personal accounts; and with the difficulty of accessing the domains of the intimate and the affective, which tend to be revealed only through glimpses, flashes and symptoms.16 West's endeavour of tapping the discourse of sorcery as a source of historical consciousness, which yielded fine fruits in Kupilikula, was less productive in excavating gendered frictions, as if the collective Muedan cultural subject that he constructed would not suffer to be fractured along gendered lines.17
Song-and-dance were productively tapped to discuss women's resistance to colonial oppression in the work of Jeanne Penvenne, and Leroy Vail and Landeg White,18 but not to reconstruct the history of gender revolution in Mozambique. This is part of a broader silence in Mozambican historiography, which remained fundamentally indifferent to song-and-dance in the post-Independence era, at the very same time in which the State elevated it as the foremost expression of national culture. Song-and-dance are also forgotten in discussions of Mozambican political memory, which focus on oral or written testimony at the expense of performance, at best touching on public commemorations.19 Silences are symptoms of epistemological embarrassment.20 Rather than dwell on the reasons that might have constituted such silence, this essay endeavours to lift it, by reconstructing the trajectory of a specific genre of female mapiko masquerade and its relationship to the history of gender revolution in Mozambique.21
In order to attune our senses to song and dance, it is important to discuss the methodological legacy of the seminal scholarship on Mozambican colonial song-and-dance. The theoretical framework was established in Vail and White's Power and the Praise Poem, which analyses a range of Southern African expressive forms as conveying collective 'maps of historical experience', transmitted from generation to generation and made relevant to the times and the conjuncture. The work was premised on the concept of poetic license: 'the convention that poetic expression is privileged expression, the performer being free to express opinions that would otherwise be in breach of other social conventions.22 Song lyrics were held to be the locus where such license would be articulated; henceforth, a variety of Southern African performative expressions were ranged in the category of 'praise poetry'. The intervention of political power to suppress poetic license would result in the ossification of expressive forms. Such was the case, for instance, of the Mozambican Chopi migodo orchestras in the socialist period:
Frelimo, it is true, did decide to preserve some of the orchestras through state patronage, in order to coopt them into the new, official 'national culture'. But more importantly, the Frelimo government also pressurised its chosen orchestras to produce patriotic anthems and songs in praise of state policies and the Frelimo party, in effect revoking the poetic license that had long lain at the core of the music's existence.23
Vail and White's work had the huge merit of subtracting African performative traditions from the domain of anthropology's timeless traditions and assigning it to historical consciousness.24 It also presented two important shortcomings, namely a binary logic and literate logocentrism.25
The binary 'licensed versus unlicensed' fails to convey a key characteristic of African orature: ambiguity. It has been widely argued that African performative traditions rely on concealment, rather than revelation; that they seek obscurity and polysemy, rather than clarity and univocality; that they convey ambivalence, rather than outright discontent.26 The functions of critique and praise are never divorced: they are two faces of the same coin and it is never possible to gauge precisely whether an utterance or performative gesture belongs to the one or the other.27 If one follows theorists such as Mikhail Bakhtin and Johana Huizinga, such playful ambivalence is at the core of ritual art across times and spaces.28
The binary 'licensed versus unlicensed' is especially inadequate for analysing the entanglement of performative forms within the politics of post-colonial states. I argue elsewhere that in Makonde revolutionary songs it is impossible to draw a clear line between what is produced 'from above' and what stems 'from below', between 'official' and 'popular', because party watchwords and formulas deployed in official anthems were widely appropriated in popular singing from the late 1960s onwards.29 Tropification did not result in the death of poetic license, but in new forms of creativity: formulas and watchwords were put to good local usage and brandished as competitive weapons in dance festivals to shape specific identities.30 It has also been shown that in a variety of African contexts acclamation and praise for political power are a way through which the ruled make the rulers accountable31 and through which the nation is imaged in a dialogue between the two, however unbalanced it might be.32
The interpretive privileging of song lyrics at the expense of performance is also a perspective error, the scholastic fallacy of the lector who analyses embodied cultural forms as if they were a text.33 Poetry is not the central category according to which song-and-dance are conceptualised in the South-Eastern Africa region.34 The drum, not the voice, is the metonym for the dance and the song, the drum being itself a metonym for rituals of passage. The widespread root *ngoma signifies this original unity of ritual, drumming and dance, of which the song is but a part and possibly not the foremost. The various drum-dances themselves should be understood as genres, obeying formal rules that reflect shared meaning.35 Songs, rhythms and choreographies can only be interpreted in relation to the historicity of the genres in which they are performed - each singular performance being an intervention that simultaneously reinforces and transforms the system of dance genres.
Even when it comes to songs and their lyrics, one should be wary of considering them as crafted messages addressed to an imaginary audience, following a communication model heavily influenced by literacy and mass media. This describes poorly the phase of composition - whether individual or semi-collective - in which verses are put together in impressionistic ways, often to confound or baffle rather than educate or enlighten; in which melody and euphony dictate the choice of words; in which borrowing and exchange between songs lead to fragmentation and recomposition; in which ironic surrealism is more valued than realism; in which dance groups sometimes sing songs whose meanings they have forgotten and that get reinterpreted idiosyncratically along they way.36 Reception can be no less haphazard. Listening to songs on the dance-field is nothing like reading their lyrics. Amongst the rumble of drums and the cheering of crowds, they are perceived in waves, as lines or at best as verses. Attention is paid fragmentarily, as spectators move from one spot to another, following now a dance gimmick and then a song, now this group and then the other. The fortune of a song rests on the evocativeness of its lines, rather than on the hidden narrative, that ultimately might not survive the author(s). While the original meaning might be known only to a few, songs generate alternative interpretations that should not be considered as false or meaningless. The songs' power lies precisely in their potential polysemy.
Our eyes and ears must heed the drum and the moving body first, then the tunes and the words; the genre first, then its individual elements; seeking not a collective voice critiquing power, but contradiction and entanglement, ambiguity and ambivalence. With this in mind, let us now move to the northerly Makonde Plateau, famously the last patch of land in Mozambique to be 'pacified' by the Portuguese (19171921); the cradle of MANU, one of the early nationalist movements; the theatre of a brutal colonial repression of organised protest (1960); and the main battlefield of the Liberation Struggle (1964-1974), which hosted Frelimo's Central Base and a large portion of its Liberated Zones.37
Independence in Mozambique was greeted with an explosion of enthusiasm channelled mostly through the expressive idioms of song, dance and theatre. Dancing crowds welcomed Samora Machel in his triumphal journey from the Rovuma river to the Maputo.38 The State responded by creating new national theatre and dance groups and by promoting in 1978 a grand National Festival of Popular Dance. A theatrical re-enactment of the Mueda Massacre - which became the object of Mozambique's first feature film - inscribed the event as the foundation of the national narrative.39 Throughout the country, performative expression blossomed. In the Makonde Plateau, the centre-stage was occupied by new dances rooted in the experience of the Struggle, such as limbondo - a circular dance of men in tatters representing the suffering of guerilla - and nnonje - a military dance with wooden weapons and revolutionary songs - as well as by old dances that the Struggle had deeply refashioned, such as the ancestral mapiko masquerading.
In these early moments of performative exhilaration, a group of Makonde DF guerrillas launched a daring dancing challenge to male chauvinism. It was during the uncertain times between September 1974 and June 1975, after the Lusaka Accord and before the declaration of Independence - the year of authority (madalaka) as it is called in Shimakonde.40 The madepe (DF)41 were stationed in the Beira Base in the northerly district of Nangade, close to the Rovuma and to today's village of Litingina, in the area known as the Fourth Sector during the war. As soldiers in a time of transition they were busy with everyday chores, full of expectations and slightly bored. With a view toward the celebration of Heroes' Day, the third of February, they started rehearsing something unheard of. They prepared a mapiko mask almost identical to the men's: dressed with cloths kept together by small wooden bits (vigwali, wearers') with a pelvis gown (ingbnda), two cloths crossed across the chest, and heavy iron bells (dinjuga) around the waist. Instead of using a wooden mask, they rolled a cloth around the dancer's face. The group was led by one Filomena Ntonya, who danced the unorthodox faceless mask. The invention was baptised lingundumbwe, a name whose meaning remains obscure.42
The extemporaneous creation directly confronted the rules of secrecy connected to the practice of Makonde ancestral masks.43 Such rules wanted the men and the women to dance two different kinds of masks in two asymmetrical settings. The men's wooden mapiko were to be danced publicly on ritual or festive occasions and were covered by a stringent 'public secret' according to which the women should pretend to believe that the masks were fearful ghosts of dead ancestors that the men dug up from a termite nest (lyoka lyuki' ndaimba kushushulu). The women's clay shitengamatu masks would dance only once a year in the final coming-out ceremonies (nkamangu) of feminine puberty rites (dingoma) held in the thick of the bush and anxiously guarded from men's intrusion and scooping.44 Such rules of secrecy were established and respected by common agreement between the men and women, the latter never really ignored that the lipiko conceals a man.45 They were a form of 'mutual respect (kwishimyana): to keep a great secret together, them the elderly men and us the elderly women; and our secret to be like this: one and the same'.46 Such 'mutual respect' was part of the complex educational performance of initiation rituals staged to turn children into adult and sexually-differentiated beings, as well as a way for men and women to define through ritual their respective gendered identities and roles.47 This ritual division of labour had the effect of promoting the men's public performances at the expense of the women's. From the early twentieth-century at least, the aesthetics of the men's mapiko developed and diversified, propelled by a thriving culture of cosmopolitan dance competition. This wasn't the case with the women's clay vitengamatu, which remained confined to a single moment of the ritual cycle and had little opportunity to develop a rich aesthetics.48
Lingundumbwe challenged this state of things by appropriating the men's costume, henceforth exposing gendered secrecy as a fiction and claiming a public performative space for masked women in the new liberated country. We do not know how the men stationed at Beira Base reacted to the provocation. Some even claim that it never happened in this form and that lingundumbwe was invented in the town of Mueda a few months later by a soldier named Josina Rafael on the occasion of Samora Machel's visit during the Rovuma to Maputo journey. Even in this version, the gist of the story remains the same: a group of women secretly prepared a lipiko dressed like the men's but with a cloth on the face and presented it as a bold surprise on the occasion of a grand state ceremonial.49
Wherever it might have been invented, the invention was quickly taken up in various villages, especially in Nangade and Muidumbe, spreading like new genres of masculine mapiko had since the 1920s. In the years after Independence, lingundumbwe became the genre of the New Woman empowered by the revolution, a defiant counterpart to its masculine equivalent:
We started dancing in the year of authority. During colonialism, only the men danced mapiko. [...] When we had authority, we tied up lingundumbwe. Because we were to have Independence. When we brought it out to the mango trees, people saw: 'The women are rejoicing their Independence'.50
Lingundumbwe ensembles were composed exclusively of women who also took up drumming, a role generally reserved for men even in feminine dances. The rhythmic basis was assured by the drilling sound of three or four makuti drums, boosted by can rattles (mashanjo) - an instrument associated with the times of war when loud instruments could not be played. The dance replicated the two movements in which the mapiko masculine performance is divided. It began with a slow introductory movement (lishesho); toward the end the pace accelerated dramatically and the dancer performed an inverted nshakasha - the signature choreography of masculine mapiko - moving forward and in a circle, rather than backward and in a straight line.
The aesthetics of lingundumbwe was consonant with the principles of socialist realism and political relevance promoted by Frelimo during the Struggle.51 In the slower part, mimetic choreographies (vikuvo) were performed. Some of them were realistic, mimicking cultivation, soccer playing, festivities and drunkenness; others were based on parent dances; and some even hinted at the secrets of masculine initiation. Lingundumbwe songs drew on the idiom of Makonde revolutionary singing, presenting a characteristic mixture of date recitation, invocations to the leaders and formulaic remembrances of struggle deeds:
Wetu twimbanga maimyo We sing the History
Vashitenda vamama kala That mothers did in times of old
Vandiinjilanga mapondo They slid into ditches
Vantukuta ashatu Fleeing attacks (assalto, Pt.)
Au mwedi wasita This sixth month
Italee ishilini na tanu On the twenty-fifth
Mu liduva lya-Uhuru kwetu ku-Moshambiki Is the day of Independence in our Mozambique52
The invention of lingundumbwe was indeed enabled by the experience of the Liberation Struggle. In the first years of war, both men and women's dances were curtailed in order to prevent the enemy from discovering the position of the guerrillas and of the populations that were supporting them by living in remote lowland areas. The new 'unplugged' dance genres introduced in the first years of war - ngoda, animated by the sound of can-rattles and magalantoni, by home-made guitars - were both led by men.53 The main dances practiced by women during colonialism, nshaila and nkala, faded out. 'It was the time of war, and we were running mbule mbule mbule. The dance of the time was nkala. We meet, e-e-e; we run away and go.54 Ultimately, women were left to handclap to the men's ngoda and magalantoni; at best to sing, to the rhythm of wooden logs, a music called dinguni ('wood').
The establishment of a female corps did not have an immediate empowering effect in the domain of song-and-dance. Frelimo was only then beginning to reflect on the importance of performative culture as a tool of political mobilisation and the men took the leading role.55 While never driving the process actively, women guerrillas were however exposed to the creative effervescence taking place in Frelimo's military camps, especially Nachingwea, where new political anthems were being composed and where dances from various parts of the country were being mixed with the objective of promoting a new national and revolutionary culture. With the creation of Frelimo's pilot schools in the Liberated Zones women pupils lived in the proximity of soldiers and assimilated a new political vocabulary through the daily performance of party anthems.56
The new revolutionary rhetoric that emerged after the 1968-1969 crisis emphasised the fight against obscurantism, 'the uncovering of the dark pall which 450 years of Portuguese colonial obscurantism cast over the eyes of our people'.57 In Frelimo's own account of the transformations of mapiko, gendered secrecy began to wane already during the Liberation Struggle.58 Despite such rhetorical claims, gendered secrecy connected to mapiko was enforced even in the most progressive settings of Frelimo's pilot centres:
There was secrecy. The women couldn't know - even here in the centres. We danced the mapiko but the women only saw the dance. They didn't know who danced. On the day of the concert, the professor disappeared from the centre. When we had to dress the lipiko, we had to leave for an area that was seven kilometres away; we prepared the lipiko there and no one knew. He comes back without the girls knowing, without anything.59
The fight against gendered secrecy was amongst the political objectives of Josina Muthemba herself, heroine and symbol of the DF and first wife of Samora Machel.60 According to those who have listened to her passionate speeches at the war front, her interventions recurrently addressed three questions. Firstly, Josina prompted women to challenge the men's dominance and seek gender equality. Secondly, she mobilised women to contribute to education and the caretaking of orphans, which materialised specifically in the precious work that she performed in the Frelimo orphanages. Thirdly, in line with Frelimo's anti-obscurantist rhetoric, she championed the radical transformation of some of the most backward ritual institutions such as puberty rituals and their attendant gendered secrecy - institutions that, despite some positive aspects, contributed to keeping women in a state of ignorant subjugation. While the first two points were embraced with enthusiasm by the crowds that Josina addressed, the third was met with embarrassed silences and resistance.61
As the war proceeded Josina's health deteriorated quickly. In 1971 she addressed a large crowd in Itanda, close to Nangade, touching again on the question of rituals and secrecy. During the speech her voice faltered and she bade farewell saying that she was not feeling well. She then crossed the border and went on to die in a hospital bed in Dar es Salaam. Her name became shrouded in legend and her memory was kept alive in the liberated zones. The seventh of April, the day of her death, became National Women's Day, as celebrated in this DF military anthem:
Nelo ni liduva Today is the day
Lyatulila In which we cry
Lyatunkumbukila Josina Machel In which we remember Josina Machel
Nae ashindava nanangaledi She was the leader
Wavakongwe va-Moshambikano Of the Mozambican women
Nkongwe waku-Kabu Women from Cabo (Delgado)
Josina Machel Josina Machel
Nkongwe Manyambane Inhambane women
Josina Machel Josina Machel
Nkongwe waingondo Woman of war
Josina Machel... Josina Machel 62
If Josina's attempts to undermine gendered secrecy did not achieve their intended effect, the war contributed to its erosion with other means. DF guerrillas had an uncertain status, somewhere in-between man and woman; although they underwent puberty rituals, they were sometimes called namaako, 'uninitiated'. The hardships and perils of war pushed men and women guerrillas to moments of intimacy and disclosure: 'What do you guys do at the masculine initiation?' 'This and that... and what about your initiation?' These moments were rare: soldiers had other things on their mind and military control on speech was otherwise strong. 'People spoke with great secrecy. We spoke little. There were intimate relationships, but they were secret'.63
These relationships existed nonetheless and lingundumbwe could possibly have been born through such forms of confidence. Filomena Ntonya was apparently in an intimate relationship with Jerónimo Vintani, a soldier from Mocimboa who was a master of mapiko costuming and explained to her the technical secret through which a mask is tied up with small wooden bits in such a way as to withstand the tensions of dancing. Whether these two characters - Filomena and Jerónimo - are real or mythical is not particularly relevant: the story of two lovers who, in the midst of the fires of war, share the secrets of initiation and forego tradition iconically condenses an experience shared by many.
Enabled by the experiences of the Liberation Struggle, lingundumbwe's roots also sink deeper, into the late years of the colonial period. Anthropologists Jorge and Margot Dias - who carried out research on the Makonde Plateau between 1956 and 1960 -describe in some detail a feminine dance called shinyala (a vernacularisation of the Portuguese word for lady):
These girls represented the so-called vinyala (shinyala = senhora), that used to dance on the Monday following the first day of a ngoma. [...] Here [the dancers] were young girls who however learned to dance with the elder women. [... ] They advanced slowly with their gaze cast downwards. They gave little steps and stopped. Finally, they arrived close to the drums. A group of women formed around them, which danced with gentle movements of the shoulder blades and undulations of the loins, tapping the feet with the hands open towards the outside, like in some oriental dances. [... ] Generally, they danced in pairs while a third one sat in a chair.64
The movements described here - shoulder blades and loin undulations - are typical of clay vitengamatu masks.65 The Diases did not make that connection, but remarked that 'the dance had a prominent ritual character'. Like mapiko and the initiation dance shinalombo, the women's vinyala was punctuated by comical carni-valesque interludes:
Two women began a dialogue that was very funny, not only because of what they were saying, but also for the expressions and mimicry. One of them, while talking, smacked her lips so that the compressed air pushed her ndona [lip plug], which came out a bit from the lip and went back in when she opened the mouth; this was truly grotesque. Then, they spoke as if they had a diction defect, pretending to be very angry. The public was delirious.66
The dance, which at the beginning of the ceremony had young girls as the protagonists, eventually captured older women with a magnetism reminiscent of trance possession:
Little by little, the rhythm of the drums took hold of the women; the older ones experimented [with] some dance steps and finally could not resist any longer. They undid the upper cloth that covered their breasts, tied it in a special way around their waist and began to dance. [...] Even a pregnant woman, towards the end of the dance, could not resist. One could see how there was a fight inside her between the desire to dance and the feeling that the state in which she was [in] could not have helped. But in the end the enthusiasm for the dance had the upper hand.67
Themselves captives of their ritualist bias, the Diases did not observe or report that vinyala enjoyed a great popularity outside of the context of initiation. In the late colonial period, the two main feminine dances were nkala and nshaila, both stylish circular dances drummed by men and with a strong singing component. At the same time, shinyala was taken out of its strictly ritual context and brought to the public field of play and competition. Shinyala groups formed almost as timid equivalents to the men's mapiko. They were called to dance for initiation rituals, but performed also for weekends and holidays. 'Shinyala comes from afar. In the beginning - because we used to put [the girls in] the ritual - we danced shinyala as they came out of the house. We rubbed white paint all over the face. Then we took it to dance on the field.68
Although the vinyala were not the object of ritual prohibitions, the process of preparation of the dancers was referred to with the same verb used for the men's masks, kupanga (to dress, to prepare). The process of transformation was deemed more important than the actual headpiece; though faceless (bila ingope), vinyala can thus be considered as masks embodying an ancestral spirit.69 The body of the vinyala was adorned with one cloth tied around the hips and two crossing over the chest. She wore on her head a long sisal wig, straightened and braided, and her face was painted in white. This faceless mask was a grotesque depiction of a white lady who has just shampooed her hair. For this reason, some called it shampunga, a wordplay with 'shampoo'. The song associated with the character describes it as a decayed lady:
Shampunga e-e Shampoo, e-e
Avé shinyala kala Once she was a lady
Maduvano nshenji Nowadays she's a savage
The dance of shinyala can be located within the regional history of feminine possession dances in which women are seized by the spirits of foreign people.70 The early twentieth century saw an intensification of such dances, possibly as a consequence of the erosion of the matrilineal systems that undermined the symbolic prestige of women.71 Meanwhile, majini healing practices of coastal origin came to the interior together with Islam, mingling with older forms of spirit possession.72 While possession by spirits per se does not exist amongst the Makonde - possibly because the practice of masks fulfils some of the same psychological functions - the appearance of the infectious drum of the white lady is related to this broad historical dynamic.
At the same time, shinyala was an embryonic form of feminine public mapiko, which began as the representation of a specific character anchored in ritual practice, but was slowly becoming a genre with a broader expressive range. This dynamic, however, was interrupted by a curtain of silence imposed by guerrilla war. Shinyala then reverted to its ritual function: a dance of celebration of feminine rituals, less and less performed and eventually abandoned as the war intensified. As the war ended and lingundumbwe appeared, it could present itself as something new and revolutionary. But many elderly women drew the connection with the old shinyala. They pointed out a pattern following which the dance movements of 'inside the house' (nngande) and 'out in the bush' (kumwito) - the secret spaces of feminine initiation - were brought into the open. Filoména Ntonya herself had been a shinyala dancer. These commonalities appeared more important than the changes and the new names:
These names... we call it vinyala, vinyala, vinyala. One year everything turns upside down and we leave it. And we call it another name. Don't be mistaken: lingundumbwe is one and the same with vinyala. During colonialism, it was another place; under Frelimo, it's something else. [...] As we saw that we are going to be free, we abandoned that name, and we picked lingundumbe, and now we go to utamaduni.73
Despite these continuities, lingundumbwe introduced a fundamental element of rupture: the usage of wooden 'wearers' (vigwali) to tie the cloths around the body and the ingonda around the pelvis, in a way that closely resembled the men's mapiko. Both vigwali and ingonda are considered the deepest technical secrets in the dressing of a lipiko, difficult to master in such a way as to make the cloths adhere to the body and the knots endure the dance's abrupt movements. Otherwise, lingundumbwe respected a few fundamental thresholds: it did not wear a corset charged with dinjuga iron-bells, which were instead tied around the waist, emphasising the hips and evoking femininity; it performed the nshakasha, but not as a separate movement and in an inverted form; and it substituted a cloth for the mask.
These courtesies were not sufficient and the dance soon awoke the men's wrath. The men considered the invention encroaching and insulting, and fought it with all means: in the domestic space, by using marital authority and sometimes even violence to forbid the women to dance; and in public by demanding that the government ban the dance and threatening to drop mapiko if the ban was not enforced. The men saw lingundumbwe as a disorderly and dangerous provocation, and the women gave them reasons to argue that it was so:
Mapiko letu lo' Josina Look at our mapiko Josina
Iyanje It's bad speech
Kudonga kwetu kwamashanja viva We clap hands with rattles, viva 74
The responses to the masculine backlash were varied. Mostly in the area of Nangade, women settled on a watered-down version dressed without vigwali. The cloths around the body were instead secured with vegetable ropes disposed horizontally across - a costume typical of facial masks practiced by the Makonde's neighbours and by Makonde youth themselves and considered more playful and less secretive. Nangade is a district located at the margins of the plateau and exposed to many influences, where the rules of secrecy and respect are not as stringent. By eliminating the most offensive element, the women could strike a truce with the men and lingun-dumbwe coexisted peacefully with the men:
Mapiko letu lamangundumbwe Our mapiko mangundumbwe-style, let's
tupange shana Josina dress it well, Josina
Amunavajedye valume Don't imitate the men
Baba Kumposha andyunga nkono Baba Kumposha shook hands
Amunavagwadye vigwali Don't wear 'wearers'
Amunavajedye valume Don't imitate the men75
Others withstood the aggression and struck back. In the village of Shidwadwa, also in Nangade, the women responded by crossing another threshold, dancing lingundumbwe in two movements with a distinctive nshakasha performed backwards, precisely like the men. As the government did not intervene, the men abandoned their own mapiko and the women won the battle. In Mueda, the ancient group of lingundumbwe sought the government's protection against spousal violence and other forms of pressure. The government intervened half-heartedly, on the one hand protecting lingundumbwe and its dancers, on the other asking the women to be accommodating and to leave space for the men, who refused to perform alongside them on the occasion of festivals.
In the district of Muidumbe, where the government is weaker and the rules of ritual secrecy stronger, the women gave up. In the early and mid-1980s, as the civil war raged in the country, lingundumbwe was replaced with a new dance called utamaduni. Musically and choreographically the two were similar: songs and dancing styles were carried over wholesale from one to the other. Only the offensive elements of costuming were dropped. Three or four girls - dressed stylishly with bandanas, sunglasses and cloths around the loins - took the lead dancing role, cheered, encouraged and hand-clapped by the elder women; significantly, they were still called mapiko. Utamaduni was a Kiswahili translation of the name of a new educational dance introduced in the feminine initiation rituals after the revolution called kurtura (culture, from the Pt. cultura).76 Like its ancestor shinyala, it was an initiation dance taken out to the public. Utamaduni groups began to form in all the villages of the Muidumbe region, sometimes more than one in each. The groups clustered around feminine masters of initiation (vanalombo) and functioned as support structures for the carrying out of puberty rituals. The women embraced the new dance enthusiastically without apparent nostalgia for the quasi-mask that they had left behind.77
Discussing the reasons for the change, many women critiqued lingundumbwe not because it was too provocative, but because it lacked originality and was a mere imitation of men. In other words, lingundumbwe broke a fundamental unwritten rule of the game of secrecy and dance rivalry: competition between men and women must always be complementary, leading to progressive gender differentiation and not escalating into gender warfare.78 Only dance competition within each sex - men against men, women against women - can be symmetrical, with all the risks involved. But now, lingundumbwe groups did not compete with each other: they openly competed with the men, disrupting their performance and seeking to draw crowds from them. Partly because of the shock of the novelty, partly because of the talent and artistry, they often succeeded in their mission:
When we come out to the open and we meet them as they dance, all people leave them and fill up our side. And that annoys them.
They see you dominating (kutawala)?
Eeeh. And we truly dominate. We dominate, truly. In that moment? We dominate. If you are ill, your illness will heal, right there!79
Utamaduni re-established the balance. It maintained the fiction that the dancers were not masks, although the dance was of a ritual nature and sometimes referred to as 'the ladies' mapiko' The groups competed amongst themselves, creating broad networks of exchange between villages, reinforced on the occasion of festival and puberty rituals. Within the villages, utamaduni groups twinned with mapiko groups, traveling together to festivals, the men drumming in the women's ensembles and the women singing in the men's. As it gained ground in the villages, utamaduni cut itself a prominent space in the provincial radio's broadcasts beside its masculine twin - an honour never bestowed on the contested lingundumbwe.
As elder women 'resuscitated' (kutakatuwa) the colonial dances nshaila and nkala,80 they did not lose the opportunity of mocking the feminist masquerade introduced by their younger peers as a failed challenge, a half-baked attempt comparable to the disreputable face-masks (makomba) danced by money-thirsty youth in times of initiation rituals:
Shiasa shavamama The mamas' politics
Vamama vandipanga lipiko lyao The mamas have arranged their mask
Vashema lina lingundumbwe They gave it the name lingundumbwe
Ingonda andigwadya sana It wears a gown alright
Vigwadyo andigwadya sana It wears 'wearers' alright
Kanji kulivaliva shinu shimu, ingope But they forgot one thing, the face
Lipiko shani likave ingope? What kind of mask doesn't have a face?
Kudyanga makomba lamashove It's worse than the youth face-masks81
The shift from the defiant lingundumbwe to the more accommodative utamaduni can be read as a poignant allegory for the trajectory of the Mozambican gender revolution, which ground to a halt after the first years of Independence. The rise and decline of lingundumbwe echoes the sense of betrayal and shattered hopes expressed by many DF, the melancholic feeling that the empowering years of the Liberation Struggle were the high point of a downward parabola.82
This melancholia lurks in both lingundumbwe and utamaduni songs, which are not cast in the bold mould of mapiko choirs, but unfold in long and tormented descending scales.83 The death and legacy of Josina Machel crystallised feelings of forlornness, expressed through the metaphor of orphanhood:
Malala kwanja nintangashidye Lie still, so I can tell you
Malaililo amwe Josina The farewell of Josina
Mwaka wapadile In the year when she died
Malaililo amwe Josina Josina's farewell
'Utamaduni amunaleke 'Don't leave culture
Mwenu iyanje leka uwavi leka You, leave envious speech, leave sorcery
Lishunge mene mananshiva Take care of yourselves, you orphans'
Ata mwiu tuvananshinva And truly we are orphans
Nanangaledi napagwa shinu We don't have a leader any longer
Ata kwatalau napagwa shinu We don't even have quarters
Lishunge mene mananshiva Take care of yourselves, you orphans84
These feelings resurface on the occasion of the death of feminine leaders such as Maria, the wife of Alberto Chipande:
Malala kwanja nintangashidye Lay still so I can tell you
Adi dimu dyatwimba wetu dyakulilanga These songs that we sing are tearful
Tuninnangila Malia mama andituleka We cry for mama Maria, she left us
Chipande baba andipakanyanga Baba Chipande insisted
'Ushwang'a madada Maria 'This is a lie, sisters, my wife Maria
ndyangu napela shinu didn't die
Ninkunnindila kungande kwangu I am waiting for her at my house
Anangukodya She will find me there'85
These expressions of mournfulness are common to lingundumbwe and utamaduni songs, but with an important difference: in the former they constitute the large majority of repertoires, while in the latter they stand side-by-side with other songs, poignantly - sometimes even aggressively - topical. The 'tropification' of Makonde singing expression in the aftermath of the Struggle had left little space for topicality. Realism was a dictate of Frelimo's aesthetics; in practice, this injunction translated into trite compilations of official watchwords and formulas, in invocations to the leaders and recitation of historical dates. Utamaduni choirs managed to break out of this aesthetic conundrum, addressing Frelimo leaders with expressions of dissatisfaction and bleak depictions of poverty:
Pamwalela petu pakumene Mwalela, our home is a big place
Katikati Mweda na Nangade Halfway between Mueda and Nangade
Shivalela vana vananshiva It is full of orphans
Dishikola napagwa shinu There are no schools
Ata medi napagwa shinu There is not even water86
Outright rage (shitundwe, 'nerves') was vented against masculine figures of authority accused of abandoning the rural areas despite the political support received from the peasantry:
Shitundwe wetu shipali Rage [lit. 'nerves'], we have it
Akunu kwetu ku-Moshambiki Here in our Mozambique
Vakulima vankuvenao shitundwe shoe The peasants are filled with rage
Chissano baba wetu tundakudya Chissano, father, we ask you
Ingondo kwetu indimalilika The war now is over
Dibei mene kupungula The prices won't go down
Pacheco baba wetu tundakudya Father Pacheco we ask you
Ushagushi undipita shana The elections went well
Dibei mene kupungula The prices won't go down
Aunatwone kujanda wetu Can't you see us becoming skinny?
Atujanda shitundwe We become skinny because of rage
Kwetu kumoshambiki In our Mozambique
A-tujanda shitundiwe We become skinny because of rage
Wetu kujanda kwoe Us, there is a lot of becoming-skinny87
The government's plan of war pensions was designed to stifle discontent amongst the Makonde, who had the sense of entitlement of having 'liberated' (kwambola is used in Shimakonde intransitively, as a mystic signifier for the victory in the Liberation Struggle and revolutionary upheaval). Even such generous measures fell under the critical swerve of utamaduni songs. The same leader that was tenderly consoled for the loss of this wife is here scolded for falling short of his duties as a provider:
Chipande wetu kulila Chipande, we cry
Lijimbo lyetu nalyaku-Kabu Our province of Cabo
Dingondo ditendeka twambola wetu The wars that took place, we liberated
Yamyaka kumi twambele wetu The one of ten years, we liberated
Ya-Mashanga twambwele wetu The one of the rebels, we liberated
Ding'ondo ditendeka twambwele wetu The wars that took place, we liberated
Wetu mabati nalala shinu We don't sleep under corrugated iron
Wetu magali nakwela shinu We don't get into cars
Tukaida nakukuandika And when you come to write our names
Munkuvalota majeshi gweka You want only the soldiers
Mwaa shani? Why?
Tuvaakulima mutujaa kwashi? We, the peasants, where do you chuck us?
Tukaida nakukwandika If you come to write down our names
Munkuvalota madepe gweka You want only the DF
Mwaa shani? Why?
Vamateriale mutujaa kwashi? The carriers of material, where do you chuck us?88
These expressions of social critique, which largely remained confined to the genre of utamaduni, took men as their favourite object. Utamaduni thus regained in the terrain of sung words the ground that had been lost by surrendering the faceless mask and its performative defiance. Utamaduni songs targeted especially the self-inflated ego of pension-receiving war veterans:
Munkulyambila mavetu You are repenting, you my friends
mulambanenge navantigo you who married the war veterans
Mulo' wakati wapake' njuluku Look, when he receives the money
Antigo kungande kwangu nawika shinu The veteran doesn't show up at home
Antigo ankulyona komandante mwene He thinks he's a commander himself
Antigo ankulyona nshadolo mwene He thinks he's an administrator himself
Mulo' wakati wamali' njuluku Look, when the money is over
Antigo ankwida kwetu nangupambadyanga The veteran comes home to a dying me
Aningupa nangu dielupu mbili He gave me ten thousand
Wena kushagoni ukashume udagala Go to the shop to buy some dried manioc
Aningupa nangu dielupu nsheshe He gave me twenty thousand
Wena kubazari ukashume ikanga Go to the market and buy a cloth
'Ushitenda shalumo ingondo 'We used to do the war together, with me,
namu nangu niliputa mwao I am your old man'89
The critique of men sometimes exceeded topicality to take a broad existential tone:
Muduniani ukalola In the world, if you see
Vatubyaidya wetu valume Who kills us, it's the men
Nkati mudunia Within the world
Vanyaku' mapinde valume Who take the bows, it's the men
Vanyaku' mashoka valume Who take the guns, it's the men90
In her inaugural text on the liberation of women, Josina Machel remarked that women should perform an important role in the Department of Security, being 'constantly on the look-out for enemy infiltration'.91 It is perhaps not by chance that Frelimo haina mwisho, the revolutionary song that singled out traitors and reactionaries - and which served as the secret soundtrack for their punishment - was inscribed in the national imagination through a recording by the DF choir:
Frelimo aina mwisho Frelimo is endless
Simango reaccionário Simango is a reactionary
Nkavandame anditukuta Nkavandame fled
Nkavandame tundinkamula Nkavandame, we caught him
Joana reaccionária Joana is a reactionary
Verónica reaccionária Verónica is a reactionary
Cambeu reaccionário Cambeu is a reactionary
Unhai reaccionário Unhai is a reactionary
Gomane reaccionário Gomane is a reactionary
DD haina mwisho The Defence Department is endless92
As Mozambique rapidly moved from Independence to civil war - initially fostered by the neighbouring white supremacist governments, then taking on a life of its own - the phantasmic Enemy of the People that was at the heart of Frelimo propaganda received a very concrete face: the one of the contra movement Renamo (Resistência Nacional Moçambicana).93 Initially, the Makonde Plateau was scarcely affected by the war, which unfolded further south. In 1984 Renamo extended its operation to Cabo Delgado province.94 In 1990, Renamo bands occupied the Messalo lowlands, where they subjected the villages built on the edge of the plateau ravines to bazooka fire. In 1991 the rebels hit at the heart of Frelimo, climbing the plateau areas of Muidumbe and occupying the historical Mission of Nangololo for three days until they were fended off by organised groups of popular militias. Shortly thereafter, peace came (1992), and then the ballot box (1994). The end of socialism brought, amongst other things, a new discourse of reconciliation that left no place for figures such as enemies, traitors, spies, scoundrels, bandits and xiconhocas. The denunciation of the political Enemy was at the heart of Frelimo's aesthetics and Makonde dances struggled to give it up. Reconciliation was experienced by many as betrayal or surrender. But dance groups had to adapt to the new rules, especially considering that the radio would not air explicitly anti-Renamo songs.
Women refused to give up the denunciation of enemies in general and of Renamo in particular. Using Samora Machel's baptismal name of 'hyenas' (matunu) or other euphemisms drawn from the animal domain, scathing attacks on Renamo continued to be at the centre of utamaduni songs well into the period of Independence.95 Defying the precepts of reconciliation, painful memories were brought out, which sometimes did not correspond to lived experience but relayed propagandistic films projected by local officers on the occasion of elections:
Nalivalive naukalau nawashitenda baba Jakama Let's not forget Dhlakhama's crimes
Amunalive Let's not forget
Renamo ukatukamula When Renamo captured us
Ata medi akakutika They would even refuse to give us water
Ata inyota ikakukola kukwaulila If we were thirsty, they told us:
tuke yadao carry on walking
Amunalive Let's not forget 96
Women remained attached to socialist aesthetics even more than men did. They were the keenest and steadiest participants in State festivals. Following the old ritual practice of 'telling the lion' (kwaula ntumi), they went out announcing the visit of leaders and national days, circulating at the break of dawn around the villages, singing and dancing. They volunteered for collective cleaning and defended the symbols of the Party against the possible disrespect brought by masculine drunkenness. Representatives of the Mozambican Women's Organisation (OMM, Organização da Mulher Moçambicana) were the most vocally supportive and the most visually homogeneous, all wearing cloths with Josina's face or the OMM emblem. They heeded the call of the Party, justifying the social critique and the refusal to let go of the anger against the Enemy as deep-seated fidelity.
Feminine cultural groups called likulutu (military training) enacted the feats of women guerrillas during the liberation struggle. Dressed in military uniforms, they demonstrated marching, carrying of materials (materiali), assaults (ashatu), anti-aerial combat (anterea) as well as more elaborate theatrical sketches. One of these sketches, which I saw in Chai in 2004, represented the popular stoning (kuntannola) of a mother who had stolen a blanket for her sick child. It was bereft of any compassion for the culprit and celebrated the punishment as a righteous feat and without any sense of tragedy. These old deeds of vigilance and punishment were directly plugged into the democratic present. Marching in national ceremonials, all dressed in uniforms and with wooden weapons, likulutu groups intoned their stern signature tune:
Ngupita nguwenawena ndyaludeya I roam the villages
Kuvalambela vanambili Looking for partridges [Renamo supporters]
Kwavapulike navona avó I can't see where they crowd
Lude Lutete nindiwika I arrived in the village of Lutete
Kuvalambela vanambili Looking for partridges
Kwavapulike navona avó I can't find where they crowd
Tunkande tunkande tunkande Jakama Let's stomp stomp Dhlakama
Tunkande-e-e Let's stomp him97
On the occasion of elections, women organised symbolic puberty rituals to propitiate the victory of Frelimo candidates.98 The women's attachment to the old socialist order - and the punishment of counterrevolutionaries that was at its core - extended beyond the domain of aesthetics and ritual. During the elections in 1994, 1999 and 2005, women's groups taunted Renamo supporters; forced them to lie on their beds for hours holding Frelimo pamphlets; and went around in triumphal processions in the villages after an electoral victory 'looking for partridges, sometimes even lynching them or burning their homes.99 Far from being the prerogative of men, political violence in the democratic period was symbolically channeled and physically carried out largely through the agency of women.100
The story of lingundumbe echoes the narrative of empowerment and disavowal established by the two academic interventions discussed at the outset, but it also enriches and displaces it in significant ways. The temporality of lingundumbwe is more complex than the linear Marxist narrative of feminine oppression. Seen from the angle of dance, colonialism doesn't figure as an absolute negative - as a time of chains, humiliation and despair - but as a moment of performative vibrancy and cosmopolitan curiosity. Shinyala, the feminine parody of a colonial master depicted as a slave, is the icon of this truncated explosion and the root from which lingundumbwe blossomed and utamaduni developed. This element of continuity is as important as the flaunted revolutionary rupture. The visible decline of lingundumbwe in the post-Independence era can be seen as an allegory for the standstill of gendered revolution; this is, however, tempered by the emergence of new critical saliency in utamaduni songs and by the women's cathectic investment in the punishment of 'scoundrels' and 'hyenas'. Rather than defeat, we have displacement in the realm of verbal bravado and of repressed rage. Rather than a linear narrative in which oppression is followed by liberation in turn followed by disavowal (a narrative that undergirds histories of liberation struggle in Southern Africa at large),101 the story of lingundumbwe is one of ambivalence and embodied contradiction, in which the present is entangled with the past and liberation with defeat.
Traditional institutions also figure differently than in the Marxist narrative: not as hindrances to change, but as generative of change. Women challenged the symbolic hegemony of men in the domain of ritual and through the expressive means of dance and song. The rules of ritual secrecy became a terrain for empowerment, contestation and reflection. Unlike the discourse of sorcery discussed by West, such domains of the traditional were neither completely unconscious nor ascribable to an indistinct collective subject. Dance-and-song generated and mediated specific local responses: some women dance groups, after all, flamboyantly held on to lingundumbwe; some gave up the mask only to focus their energies on the poetic slandering of men; some mourned Josina; some enacted the feats of the Liberation Struggle; and some vented out their rage on Renamo supporters. The potential to generate such social difference is inherent in the competitive nature of ritual song-and-dance.102
Methodologically, this translates into a warning against totalising interpretations of song-and-dance, focusing on one of the elements at the expense of the others, or ascribing definitive meaning to specific expressive forms. Songs lyrics, choreographies, costuming and music only take on meaning when considered as a whole - when each element is contrasted to the other and all are considered in relation to genres and their historicity. The result is essentially ambiguous and complex. For dance - and music - always exceed words; this excess is what makes them powerful expressive tools. As dancer Isadora Duncan put it, 'If I could tell you what it means, there would be no point in dancing it'.103
1 A previous version of this paper was presented at the Anthropology seminar of the University of the Western Cape in May 2012. The research is part of a broader project carried out between 2001 and 2009, the results of which are forthcoming in P. Israel, In Steps with the Times: Mapiko Masquerades of Mozambique (Ohio University Press, 2014). I thank Jimi Adesina, Edward Alpers, Ruchi Chaturvedi, Colin Darch, Jung Ran Forte, Allen Isaacman, Sanjay Krishnan, Harry West and the anonymous reviewers for their comments.
2 Oficina de História, 'La Lutte de Libération Nationale au Mozambique et l'Emancipation de la Femme' (paper presented by Isabel Casimiro at the UNESCO Meeting of Experts on 'The History of Women's Contribution to National Liberation Struggles and their Roles and Needs During Reconstruction in Newly Independent Countries of Africa', Guiné Bissau, 3-7 September 1983); A. and B. Isaacman, 'The Role of Women in the Liberation of Mozambique' Ufahamu, 13 (1984), originally presented as 'National Liberation and Women's Liberation: Mozambican Women in the Armed Struggle, 1962-1975' in the same UNESCO meeting. The History Workshop paper is presented as the fruit of 'systematic research on the woman's role [...] based on the lives of combatants, peasants and other women', Oficina, 'La Lutte', 1. The paper announces the publication of an article in a forthcoming issue of the Não Vamos Esquecer magazine, which would, however, never be published, Ibid, 29.
3 Oficina, 'La Lutte', 6, 7.
4 A Marxist interpretation of traditional kinship as a system of exploitation is what united French anthropologist Christian Geffray with the researchers from the Centro de Estudos Africanos who would later become the targets of his spiteful critical intervention. Geffray's book on Makua traditional kinship is a demystification of matriliny as a gerontocratic system of exploitation in which cadets are akin to slaves, Ni Père ni Mère: Critique de la Parenté: le Cas Makhuwa (Paris: Le Seuil, 1990).
5 Oficina, 'La Lutte', 13. See also Isaacman and Isaacman, 'The Role', 156.
6 For an interpretation of that crisis in terms of world-system theory - which however tends to replicate Frelimo's own narrative -see G. Derluigian, 'The Social Origins of Good and Bad Governance: Reinterpreting the 1968 Schism in Frelimo', in Sure Road? Nationalism in Angola, Guinea-Bissau and Mozambique, ed. E. Morier-Genoud (Leide: Brill, 2012), 79-100.
7 Oficina, 'La Lutte', 11.
8 Isaacman and Isaacman, 'The Role', 175.
9 The Isaacmans speak of the OMM's experience as 'disastrous', 'The Role', 176.
10 M.-J. Arthur, E. Mpalume, J. Aquimo and V. Labés, 'O estatuto da mulher na luta armada' (ARPAC report, mimeo, Maputo 1992).
11 H. G. West, 'Girls with Guns: Narrating the Experience of War of Frelimo's "Female Detachment'", Anthropological Quarterly, 73, 4 (2000), 180-194.
12 OMM and B. Zimba, A Mulher Moçambicana na Luta de Libertação Nacional: Memórias do Destacamento Feminino (Maputo: CPHLLN, 2013) See also the interviews in A. Bouene Mussanhane, Protagonistas da Luta de Libertação Nacional (Maputo: Marimbique, 2012) and Ike Bertels, Guerrilla Grannies: How to Live in this World (documentary, Mozambique/Denmark, 2012), 90'.
13 L. Bonate, 'Muslim Memories of the Liberation War in Cabo Delgado', this volume.
14 Gleaning the footnotes of the said publications, one finds recurring statements by the usual suspects - Paulina Mateus Nkunda, Marina Pachinuapa, Maria da Luz Guebuza, etc.
15 For two recent readings of this phenomenon see V. Igreja, 'Memories as Weapons: The Politics of Peace and Silence in Post-Civil War Mozambique', Journal of Southern African Studies, 34, 3 (2008), 539-56 and J.-P. Borges Coelho, Abrir a Fábula. Questões da política do passado em Moçambique', unpublished paper presented at Coimbra University, 1 June 2011. The former provides a criticism of Frelimo's symbolic hegemony aligned with the revisionist positions of Cahen, Cabrita and Ncomo; the latter is more optimistic and sees the dominant narrative as gradually opening up.
16 See D. W. Cohen's allegoric comb in The Combing of History (Chicago: Chicago University Press, 1994), 1-23.
17 H. G. West, Kupilikula: Governance and the Invisible Realm in Mozambique (Chicago: Chicago University Press, 2005). [ Links ]
18 J. Penvenne, African Workers and Colonial Racism: Mozambican Strategies for Survival in Lourenço Marques, Mozambique 1877-1962 (Portsmouth: Heinemann, 1995); L. Vail and L. White, Colonialism and Capitalism in Mozambique: A Study of Quelimane District (University of Minnesota Press, 1981), Power and the Praise Poem: Southern African Voices in History (Charlottesville: Virginia University Press, 1991). Susan Geiger demonstrates the role of women's ngoma as catalyser of Tanzanian nationalism, TANU women: Gender and Culture in the Making of Tanganyikan Nationalism, 1955-1965 (Portsmouth: Heinemann, 1997).
19 See A. M. Santos, 'Performing the Past: Celebrating Women's Day in Northern Mozambique, Cahiers d'Etudes Africaines, special issue Jeux de mémoires, 197 (2010), 217-234.
20 J. Depelchin, Silences in African History (Dar Es Salaam: Mkuki Wa Nyota, 2005).
21 Part of the reason for the silence might lie in the economism of Oficina historians and their deep suspicion of anthropological approaches. Why scholars who produced notable work on Mozambican colonial song-and-dance never expanded their research on the post-Independence period remains a matter of speculation.
22 Vail & White, Power and the Praise Poem, 319.
23 Ibid., 147.
24 This terrain had been opened by T. Ranger, Dance and Society in Eastern Africa: The Beni Ngoma (London: Heinemann, 1975).
25 For a neo-functionalist critique of Vail and White's work see A. Apter, Beyond Words: Discourse and Critical Agency in Africa (Chicago: University of Chicago Press, 2007), 48-50.
26 See amongst others K. Barber, The Anthropology of Texts, Persons and Publics: Oral and Written Culture in Africa and Beyond (Cambridge: Cambridge University Press, 2007), 79-86; K. Chinyowa, 'Towards an Aesthetic Theory for African Popular Theatre,' SATJ, 21, 1 (2007); F. Kramer, The Red Fez: Art and Spirit Possession in Africa (London: Verso, 1993); J. Bonhomme, 'Masque Chirac et danse de Gaulle: Images rituelles du Blanc au Gabon, Gradhiva, 11 (2010), 81-98.
27 This is the enduring lesson of A. Mafeje, 'The Role of the Bard in a Contemporary African Community', Journal of Modern African Languages, 6, 3 (1967), 194-223. As Achille Mbembe elegantly puts it, 'The colonial masters almost never knew if simple imitation was in fact opposition; if apparent opposition was in fact only simple inversion; or still if that which appeared as a clear-cut revolt did not in fact belong to a simple logics of desire', Sortir de la Grande Nuit: Essai sur l'Afrique Décolonisée (Paris: Payot, 2010), 89, my translation.
28 M. Bakhtin, Rabelais and His World (Bloomington: Indiana University Press, 1984); D. LaCapra, 'Bakhtin, Marxism and the Carnivalesque', Rethinking Intellectual History (New York: Cornell University Press, 1983), 291-324; J. Huizinga, Homo Ludens: A Study of the Play-Element in Culture (London: Francis and Taylor, 2003).
29 P. Israel, 'The Formulaic Revolution: Song and the "Popular Memory" of the Mozambican Liberation Struggle', Cahiers d'Etudes Africaines, L (1), 197 (2010), 181-216.
30 Israel, In Step With the Times, Ch. VIII.
31 M. Karlström, 'On the Aesthetics and Dialogics of Power in the Postcolony', Africa, 73, 1 (2003); L. Gillman, The Dance of Politics: Gender, Performance and Democratization in Malawi (Philadelphia: Temple University Press, 2009).
32 K. Askew, Performing the Nation: Swahili Music and Cultural Politics in Tanzania (Chicago: University of Chicago Press, 2002); L. Edmonson, Performance and Politics in Tanzania: The Nation on Stage (Bloomington: Indiana University Press, 2007).
33 P. Bourdieu, Méditations Pascaliennes (Paris, Le Seuil, 2003).
34 F. Gunderson and G. Barz, eds., Mashindano! Competitive Music Performance in East Africa (Dar Es Salaam: Mkuki Na Nyota, 2005).
35 Genre analysis is a fundamental achievement of Marxist criticism, especially in the work of Mikhail Bakhtin, Georgi Lukacs, Raymond Williams and Fredrick Jameson. For a poignant discussion in the African context see D. Coplan, In the Times of Cannibals. The Word Music of South Africa's Basotho Migrants (Chicago: University of Chicago Press, 1995).
36 I draw here on my own field research on Mozambican dance, 2002-2009. For more detail, see Israel, In Step, chapter VI.
37 See amongst others Y. Adam, 'Mueda 1917-1990: Resistência, Colonialismo, Libertação e Desenvolvimento', Arquivo, special issue Cabo Delgado, 14 (1993), 4-102 and the rest of the issue; West, Kupilikula; M. Cahen, 'The Mueda Case and Maconde Political Ethnicity: Some Notes on a Work in Progress, Africana Studia, 2 (2000), 29-46.
38 See the propagandistic R. Pachinuapa, Do Rovuma ao Maputo: A Marcha Triunfal de Samora Machel (Maputo, 2005) and Darch and Hedges's contribution to this volume.
39 R. Guerra, Mueda, Memoria e Massácre, Film (INAC, Mozambique, 1979) and R. Schefer, 'Re-Constitutions: On Mueda, Memória e Massacre, by Ruy Guerra', in 'You Do Not Stand in One Place to Watch a Masquerade', Le Journal de la Triennale, 4, ed. M. Bouteloup (Paris: CNAP, 2013), 38-40.
40 Madalaka: authority or power, as opposed to Uhuru, independence.
41 This is a vernacularization of the word 'DF' with a Makonde noun-class five prefix (ma-).
42 Bernadeta Mateus Shamwilanga, interview, Pemba, July 2004. Ernestina João Vingambudi, Regina Nkavandame, Bendita João, interview, Mbwidi (Nangade), September 2004, also believed that the dance originated in Nangade. Lingundumbwe is indeed widespread in the northern part of the district.
43 On Makonde masks see amongst others J. and M. Dias, Os Macondes de Moçambique, Vol. III: Vita Social e Ritual (Lisboa: JIU, 1970); R. Texeira Duarte, Escultura Maconde (Maputo: UEM, 1987); Art Makondé: Tradition et modernité (Paris, MAAO, 1989); E. Alpers, 'Representation and Historical Consciousness in the Art of Modern Mozambique, Canadian Journal of African Studies, 22, 1 (1989); K. Fenzl (ed.) Makonde: Mapiko (Linz: Neue Galerie der Stadt, 1997); P. Israel, 'Masques en Transformation: Performances mapiko des Makonde (Mozambique): Historicité, Création et Révolution' (Ph.D. diss, EHESS, Paris, 2008); A. Y. Bortolot, 'A Language for Change: Creativity and Power in Mozambican Makonde Masquerades', (Ph.D diss., University of Columbia, 2008).
44 According to a popular etymology, the name of the mask, shitengamatu, means 'open your ears, a provocation addressed to the men that can only listen to their drums.
45 Jorge and Margot Dias mistakenly argued that mapikos secrets were a way for men to maintain gendered supremacy over the women, Os Macondes III, 393. This interpretation was inspired by the functionalist theory that wanted masks to be agents of social control. For a rebuttal, see B. Bellman, The Language of Secrecy: Symbols and Metaphors in Poro Ritual (New Brunswick: Rudgers University Press, 1984) and P. McNaughton, 'Social Control and the Elephants We Scholars Make,' African Arts, 24, 1 (1991), 10, 12, 14, 16, 18. J. Wembah-Rashid also stated that 'initiated women know what is a lipico, independently from the official version. At least amongst the Makonde, we know this has been the case since the beginning, 'Le Masque et la Tradition de Danse Masquée', in Art Makondé, 41.
46 Interview with Rashidi et. al. 2004. Kwishimyana is a verb that comes from the word ishima (respect), itself a vernacularisation of the Kiswahili word heshima. The concept of ishima has a central place in the practice of mapiko; the fact that it is a Swahili loanword demonstrates the importance of cross-border cultural contact in the history of Makonde masquerading.
47 Kenji Yoshida comes to a similar conclusion for the Chewa's Nyau, which he sees as 'a means of creating differences rather than a mere representation or arbitrator of existing gender differences', 'Masks and Secrecy among the Chewa', African Arts 26, 2 (1993), 45. In my forthcoming book I conceptualise the mutual definition of gender roles through ritual secrecy by recurring to Gregory Bateson's concept of schismogenesis, Naven: A Survey of the Problems Suggested by a Composite Picture of the Culture of a New Guinea Tribe Drawn from Three Points of View (Cambridge: Cambridge University Press, 1936) and Steps to an Ecology of Mind (Chicago: University of Chicago Press, 2000), 64-71.
48 The main transformation was to appropriate the men's wooden masks in the secret nkamangu ceremony, partly out of desire for the object and partly because clay masks tend to break easily. A few women carvers emerged in the late colonial period, when it was difficult for the women to acquire a lipiko, see the case of Rufina Nioyo documented by Bortolot, A Language, 181-183.
49 For this version, Marieta Navina, Faustina Paulo Manjeda and Dorotea, interview, Mueda, April 2009. The women dismissed as a lie the idea that the dance originated in Nangade. That people in Nangade remember dancing 'in the year of authority' (19745) contradicts the version which has the dance as originating in Mueda in 1975. Also, lingundumbwe today is widespread in Nangade. Both Ntonya and Rafael had passed away at the time of my research and so I could not verify these claims. I have not come across any archival reference to lindungumbwe, except for a photograph in A. Fresu and M. de Oliveira, Pesquisaspara um Teatro Popular em Mozambique (Maputo: Cadernos Tempo, 1982), referenced as 'Woman lipiko (mask introduced during the Armed Struggle)', 39, 49. Bortolot discusses lindungumbwe in A Language, 184-192 without touching on the question of origins.
50 Navina et al 2009.
51 See the six-part series on Tempo, 356-361 (1977), Basto, A Guerra, 112-115.
52 Lingundumbwe song, Mwalela, recorded in June 2004. For a discussion of the tropes, see Israel, 'Formulaic Revolution, 208-209.
53 For more detail on these dances and their songs, see P. Israel, 'Utopia Live: Singing the Mozambican Struggle for National Liberation, Kronos 35 (2009),106-107, 111-114.
54 Collective interview with the lingundumbwe group, Mwalela, July 2004.
55 For instance, all the main composers of Frelimo anthems were men.
56 Filomena Simão Nshusha, interview, Mwambula, Dec. 2004.
57 'Africa Day', Mozambique Revolution, 34 (May 1968), 11. The fight against obscurantism would be emblazoned in the 15th Article of the Constitution of the People's Republic of Mozambique.
58 'The longstanding experiment of the creation of the New Man in the liberated zones of Cabo Delgado put an end to obscurantism, in which many traditional dances were, and still are, enshrouded', Frelimo, Programa do Primeiro Festival de Dança Popular (Maputo: GPFDP, 1978), 28.
59 Jorge Zaaqueu Nhassemu, interview, Pemba, February 2005.
60 On Josina see the celebratory R. Matusse and J. Malique, Josina Machel: Ícone da Mulher Moçambicana (Maputo: ARPAC, 2007) and Colin Darch's website, http://www.mozambiquehistory.net/josina.html, accessed on 7th of April 2013.
61 Vingambudi et al 2004. The women referred specifically to her last speech at Itanda.
62 'Nelo ni liduva', in Coral das FPLM - Destacamento Feminino, Canções da Luta Armada de Libertação Nacional (Maputo: Radio Mozambique, 1975), CD, track 2; and Frelimo Hinos da Revolução (Maputo, 1975), 62.
63 Bernadeta Mateus Shamwilanga, interview, Pemba, September 2004
64 Dias and Dias, Os Macondes, III, 261. The vinyala danced on a Sunday, not Monday.
65 This much, that vitengamato 'undulate the loins' (myongo dukuduka) is known also to outsiders.
66 Dias and Dias, Os Macondes III, 261.
68 Interview with Bernadeta Shamwilanga, Pemba, September 2004.
69 Working in the context of Zambian feminine puberty ritual, Elisabeth Cameron observed that women consider themselves makishi (masks) and that this prompts a re-consideration of masquerading beyond the European fixation with the mask-as-object, 'Women = Masks. Initiation Arts in North-Western Province, Zambia,' African Arts, 31, 2 (1998), 50-61, 93.
70 There is a broad literature on Swahili spirit possession, see esp. L. Giles, 'Sociocultural Change and Spirit Possession on the Swahili Coast,' Anthropological Quarterly, 68 (2), 89-106. Feminine spirit possession cults exist also in central Mozambique, see Bettina Holzhausen 'Youth culture in rural Mozambique A study of the significance of culture for young people in rural areas, based on fieldwork in the districts of Nangade (Cabo Delgado), Mossurize (Manica) and Chokwe (Gaza),' research report, Zurich, 16 January 2007 http://www.nestcepas.ch/_pdf/Youth_culture_in_rural_Moz.pdf
71 Vail & White, Power and the Praise Poem, 231-271.
72 For the Makua, see L. Bonate, 'Traditions and Transitions: Islam and Chiefship in Northern Mozambique, ca. 1850 - 1974' (PhD diss., SOAS, 2007), 63 and D. Trentini, 'On the Threshold of a Healer's Mosque: Spiritual Healing, Hazard and Power in Northern Mozambique' (Ph.D. diss., SOAS, 2012), 115-117.
73 Interview with Rashidi et. al. 2004. The same connection between lingundumbwe and shinyala was pointed out by Maria Mmaka, Tereza Mashaka, Consolata Lucas, interview, Mwambula, January 2002.
74 Lingundumbwe song, Mwalela, recorded July 2004.
75 Ibid. Kumposha is the name of the village president. The injunctions ('don't imitate', 'don't wear') can be read as self-restrictions or as quoting the village president.
76 It was the equivalent of the masculine makwaela, introduced in the boys' initiations, see P. Israel Kummwangalela Guebuza: The Mozambican General Elections of 2004 in Muidumbe and the Roots of the Loyalty of Makonde People to Frelimo, Lusotopie, 13, 2 (2006), 113.
77 For the case of Mapate, Rosalina Anapambula and Tomasina Mmembe, interview, July 2004; of Malangonya, Ernestina Liteka, Filomena Joana, Augusta Jaime, Samanina Mwalimu, Rosa Valerio, interview, August 2004.
78 For the distinction between complementary and symmetrical competition, see Bateson, Naven, 176-177.
79 Interview with Marieta Navina, Mueda, April 2009.
80 Two groups of nshaila were resuscitated: one in Mbwidi, which operated in the 1980s and 1990s, the other in Mwambula, which, under the leadership of the drummer Shuliki, operated from the 1980s through to the death of the drummer in 2009. One nkala group was resuscitated in Mwatidi, under the leadership of war veteran Sabina Bissali; its songs and dance-style were however radically re-imagined and strongly influenced by the Frelimo vocabulary.
81 Nshaila song, Namwembe, recorded in July 2004. Lipiko is the singular of mapiko and identifies a masked dancer; ingope (lit. 'face') is an euphemism for the mask itself. Makomba are facial masks originally danced by the Makonde's neighbours and then appropriated as youth genres within Makonde dance themselves. They are sometimes called 'slave masks' because the Makonde used the term slave (nshagwa) to refer to all their neighbours living in the lowlands.
82 'It is with a certain sadness that they remember the time of the struggle, when, despite all difficulties, their work and their contribution was regarded, appreciated and respected,' Oficina, 'Le Role,' 24. The same feelings were conveyed by the DF that I interviewed in Pemba; less so in Mueda or Muidumbe.
83 The following observations are based on analysis of a corpus of about 300 lingundumbwe and utamaduni songs recorded between 2002 and 2008, and mostly in 2004 and 2005.
84 Lingundumbwe song, Mwalela, recorded in July 2004.
85 These two songs appear in various versions, of which I give the most complete, from the Mwalela lingundumbwe group, recorded in July 2004. Maria Chipande died in the fatal Antonov crash of 31st of March 1986.
86 Utamaduni song, Mwalela, recorded in July 2004.
87 Utamaduni song, Mbwidi, recorded in July 2004.
88 Utamaduni song, Nshinga, recorded in August 2004. Kwandika ('to write') is a shorthand for the registering of war veterans' names.
89 Utamaduni song, Mbwidi, recorded in July 2004.
90 Utamaduni song, Mapate, recorded in September 2004.
91 Machel, 'The Role,' 24.
92 Coral das FPLM, Canções da Luta Armada, track 6. See also Frelimo, Hinos, 6-7. The spelling of this song varies according to whether one follows Kiswahili (haina mwisho), or the Shimakonde (aina mwisho or ayina mwisho). Artist Adelino Timoteo remembers that the song was often played on the radio and at national ceremonials. 'It was a song intoned by people whose voices seemed to me strident, sombre and at the same time in unison', in 'Porquê os cidadãos do centro e norte são reaccionários e os do Sul revolucionários?' Canalmoz, 29th of July 2009, http://debatesedevaneios.blogspot.com/2009/08/porque-os-cidadaos-do-centro-e-norte.html, accessed on the 7th of April 2013.
93 On the Enemy of the People in Frelimo see amongst others L. Buur, 'Xiconhoca: Mozambique's Ubiquitous Post-Independence Traitor', Sharika Thirangama and Thobias Kelly, eds., Traitors: Suspicion, Intimacy and the Ethics of State-Building (Philadelphia: University of Pennsylvania Press, 2010).
94 D. Robinson, 'Curse on the Land: A History of the Mozambican Civil War' (Ph.D. diss, University of Western Australia, 2006), 178.
95 Samora called 'hyenas' the Enemies of the People, see H. G. West, 'Voices Twice Silenced: Betrayal and Mourning at Colonialism's end in Mozambique', Anthropological Theory, 3, 3 (2003), 363.
96 Utamaduni song, Nshinga, recorded in August 2004. An officer confessed to me in 2005 that he had projected such films. Renamo war crimes were heinous, but they did not affect the Makonde Plateau as they did other parts of the country.
97 Likulutu song, Nampanya, recorded in Dec. 2004. The partridge is Renamo's symbol.
98 See Israel, 'Kummwangalela Guebuza,' 111-112.
99 Pedro Justino Seguro, interview, Dec. 2004 and André Shikumene, interview, July 2008, respectively district administrator and village leader of Mwambula at the time. I partook in one of these processions in January 2005.
100 My narrative ends in 2005, when the election of Guebuza to the presidency ushered in a new phase of Mozambican history, marked by a combination of unbridled liberalism and authoritarianism.
101 B. Raftopolous, intervention at the workshop Não Vamos Esquecer? Dialogues on Mozambican History, University of the Western Cape, 15th of October 2012.
102 This is an argument that I develop elsewhere but that forms the backdrop of the present piece; see Israel, In Step with the Times.
103 Cited in Bateson, Steps to an Ecology of Mind, 137-8.