‘In heavy rotation’: uncovering the phonographic industry and the ‘NGOMA national label’ in socialist Mozambique (1978–1990)

ABSTRACT This article outlines the structure and editorial practices of the phonographic industry in postcolonial Mozambique during the so-called ‘socialist period’. It details the production phases, the associated companies and delves into the material conditions and aesthetic values that guided the phonograms published by NGOMA—dubbed as ‘the Mozambican national label’—and their relationship with state-defined cultural policy between 1978 (when production on this series commenced) and 1990 (when vinyl production officially ceased in the country). Several themes are explored, such as predominant topics of song lyrics, repertoires and artists, copyright, women artists, and the restrictions on music production during the civil war. After analysing the main musical trends and acknowledging noteworthy absences, I reflect on NGOMA’s efficiency in the nation-building process.

sound technology companies, mass media (radio, television, music journalism), and entertainment industries (Losa 2010: 632-42). These can also be defined as 'cultural industries', referring to 'those institutions in society which employ the characteristic modes of production and organisation of industrial corporations to produce and disseminate symbols in the form of cultural goods and services, generally, although not exclusively, as commodities' (Garnham 1984: 156-57). 7 Some authors, such as Steve Jones, have noted that popular music studies tend to focus on reception, completely ignoring the different industrial processes that are inseparable from this category, such as recording and editing techniques, as well as machines used for mass reproduction (Jones 1992); others argue for the need to build models adapted to local realities, thus making a distinction between the industrialised countries where multinational companies operate, and the countries or regions that have been impacted by their commercial expansion. Through this distinction, the concepts 'big sounds' and 'small peoples/countries' were used by Krister Malm and Roger Wallis (1984). Although their operationalisation can nowadays be seen as partially misleading and overly paternalistic, most of the issues discussed in their classic outputs remain highly relevant for the Mozambican case. 8 Peter Manuel (2013), just like Richard Middleton (1990), situates the development of mass media and their audio formats-including vinyl records, cassettes, etc.-in the context of urbanisation processes, the emergence of modern social classes and the general context of late modernity. Nevertheless, Manuel makes a convincing effort to distinguish the western experience-the basis for most of Middleton's analysis-from nonwestern cultures, taking into account their individual socioeconomic development and intrinsic cultural values (Manuel 2013). For example, Manuel mentions that countries which adopted communist-type political systems tended to fiercely oppose the penetration of the big five multinationals, opting instead for often relatively undeveloped local and national organisations.
Do we find echoes of these patterns in the Mozambican case? The short answer is yes. However, there is a catch: as an integral part of his argument, Manuel also states that the 'democratisation of production was further stimulated by the advent of cassette technology, which, in contrast to the capital-intensive, one-way nature of the "old media" like records and film, lent itself to small-scale, grass-roots production', further adding that 'by the late 1970s, cassettes were the dominant medium of music marketing in Africa, the Middle East, and elsewhere, precipitating the flowering of innumerable local genres that would not have been represented on the corporate old media' (Manuel 2013: 62). Since cassette players and cassettes were virtually non-existent in the country during the 1970s and 80s, this well-established pattern does not apply to the Mozambican case. Hence, until 1990, vinyl discs remained in heavy rotation as the only phonographic medium available in Mozambique, raising many interesting questions regarding the Mozambican case's specificity, some of which I hope to clarify by the end of this article.
Much has been written about FRELIMO's cultural and musical policy during late colonial and post-colonial Mozambique and its eventual relation to nation-building processes. 9 For instance, I have been very active in clarifying the significance of many expressive modes and media for what I labelled as 'the sonorous construction of Mozambique' through the analysis of 4-part choral singing and traditional music (de Freitas and de Carvalho 2022), radio broadcasting before and after independence (de Freitas 2021a, 2021b, 2022a), and even some alternative political projects that never materialised (de Freitas 2022b). However, aside from some notable exceptions such as Schwalbach (2022), Filipe (2012), Sopa (2013), Laranjeira (2014) and Mendes (2021), 10 most studies bypass entirely the significance of música ligeira in the nation-building process. Most importantly, the phonographic industry of the socialist period has not previously been the focus of serious study. Consequently, I have needed to identify and reconstruct every part of the industrial process by analysing newspaper articles, reviewing official documentation, and conducting semi-structured interviews with key participants, including musicians, producers, and cultural brokers. Therefore, this article aims to fill an essential gap while providing a new and deeper focus on the phonographic industry and the production of música ligeira during the so-called 'Mozambican Socialist Experience'.

The phonographic industry in Mozambique after 1975: an overview
In contrast to South Africa, which took its first steps into the phonographic domain as early as 1912, in Mozambique such developments did not occur until the 1960s. Nevertheless, from the early twentieth century, citizens of Lourenço Marques (now Maputo) could purchase phonograms and turntables at specialist stores, including Casa Bayly, representing His Master Voice brand; J. Martins Marques Lda., representing Columbia; and Casa Spanos, where Polydor records were sold (Sopa 2013: 40). Although the first Mozambican vinyl pressing plant was installed in Lourenço Marques on 6 October 1967 (in association with the film company Cináfrica), significant growth in the number of active recording companies did not occur until the 1970s, these included Lourenço Marques Discos, Companhia de Discos de Moçambique, Fonoplay, Teal Discos and Somodiscos-Sociedade Moçambicana de Discos (in turn connected to the Portuguese company Rádio Triunfo) (Sopa 2013: 196). Simultaneously, labels such as 'Bula Bula' or 'Bayete' began to bring out vinyl discs recorded in the studios of local radio advertising companies such as Produções 1001 and Delta Publicidade (de Freitas 2021a: 450).
After independence, FRELIMO's government assumed direct control of pre-existing phonographic infrastructures, except for two notable companies that preserved their private status, although remaining artistically dependent on state-defined cultural policy. Hence, production became centralised in four companies: the state-owned Instituto Nacional do Livro e do Disco (INLD) and Radio Moçambique (RM); and the private companies Teal Discos (in partnership with Companhia de Discos de Moçambique), and Somodiscos (in collaboration with Minerva Central and Casa Bayly). The phonographic production stages, from the initial decision to produce a given phonogram to the moment consumers could hear the result, are as follow (see Figure 1): 1. Production decision and planning: the initial decision to produce a specific artist and repertoire was defined by the publishers (in most cases, INLD and Rádio Moçambique). 2. Recording and post-production: This phase encompasses the entire recording and audio production process. Rádio Moçambique and radio advertising companies (Produções 1001, Produções Golo and Delta Produções, later nationalised and integrated into the national radio station) were the only companies with studios equipped with TEAC 4-track and 8-track reel-to-reel recorders (16-track would only be available in the 1990s). After recording, technicians (most notably Fernando Azevedo and António Cuná) applied many post-production effects, generally without considering the musicians' opinions and intentions. The individual songs would later be grouped in the desired order for the phonogram, thus creating the reel-to-reel 'master' tape. 3. Authorisation and censorship: aiming to centralise the import and export of books and other publications (including vinyl discs) and to control their production, 11 the government created the Instituto Nacional do Livro e do Disco (INLD) under the auspices of the Ministry of Information. One of its main activities was to provide an approval seal that would enable the retailing of a given cultural product. This institution had a censorship committee that screened the musical candidates to be marketed in Mozambique. According to its coordinator, Domingos Macamo: The censorship committee had two main activities: listening and controlling everything that came from outside the country and deciding whether or not it should enter Mozambique, considering, for example, if it was subversive or not; the other was to control the songs from inside. Music was a weapon, and they [state authorities] knew it; that's why Radio Moçambique answered to the Ministry of Information. We would listen to records and conjuntos [music groups] in our office, and then we would decide if its publication was possible or not. If we didn't understand the language, for example, if it was a language from the North or Central Mozambique, we had to get someone to translate the lyrics for us to make an informed decision […] we had to make sure that the songs didn't insult the government. Afterwards, we would decide whether or not to publish them. (Interview with author, Maputo 26-03-2018) After 1982, the screening and authorisation process changed from the INLD (which had deteriorated financially) to Rádio Moçambique. 4. Matrix production (South Africa): Usually, this phase would be finished in the country where the recording was produced; however, 'due to the complicated techniques it required', Mozambican publishers had to send the reel-to-reel tape 'masters' to South Africa to create the matrix-a metal mould with grooves from which vinyl records could be pressed and mass-produced. 12 The South-African companies providing these services were Gallo 13 and RPM Record Company. 14 Until the matrix was returned to Maputo (it could take an extended period, from three weeks to many months), the production process would be on hold. 5. Release date decision: Since the matrix production could take some time, the launch date could only be defined upon its arrival in Mozambique. From that moment on, the remaining production stages would again be under the control of Mozambican companies. 6. Mass reproduction: when the matrix arrived, production would resume in the tworemaining private-owned record pressing plants in Maputo: Teal Discos, which operated in the Machava industrial area, and Somodiscos, located in the City Centre, at Mohamed Siad Barre Avenue. Teal Discos had a production capacity of 360 discs per hour (120 for each of the three pressing machines available) and 70,000 discs per month. 15 There is no information on Somodiscos' production capacity; however, we know that 'in 1980 it released 461,480 albums: 26 singles (total of 416,129 copies) and 45 long-plays (total of 45,351)', 16 most of which refer to imported albums. Producer Domingos Macamo also drew attention to the fact that each of the companies operated in association with other distribution and printing companies, many of which belonged to the same owner: Teal Discos worked with Companhia de Discos de Moçambique, which also operated as a distributor, while Somodiscos worked with Minerva Central for the production of liner notes, and Casa Bayly as a distributor (see Figure 2). 17 Figure 2. Inside Teal Discos factory. Left: one of three available duplicating machines. On top, one can observe the matrix, and below, the material that would produce the vinyl disc once pressed. Right: activity unclear. May involve worker cutting excess vinyl from pressed discs. 18

7.
Marketing: this phase refers to the strategies applied to promote the final product. Promotion was done through radio programmes and the press, notably in sections dedicated to films, records, books, and recreational shows. 19 Exceptionally, other promotional materials could be printed by the companies mentioned above (see Figure 3). 8. Distribution and sale: The discs could be purchased at INLD stores, at the so-called 'lojas do povo' ('people's shops') or stores directly related to publishing companies, such as Casa Bayly. Later, they could also be acquired informally (and illegally) through 'candonga' (itinerant street selling). According to Macamo, the records published by the INLD's NGOMA label 'sold well. We did an initial print run of 1000 discs that could be later republished depending on the success of the first print run'. 20 9. Reception and use: the final objective was for consumers to purchase the vinyl disc.
However, its reproduction presupposed the existence of specialised hardware, such as a turntable. A small proportion of the population had access to such costly equipment, and after independence ownership was largely restricted to wealthier families living in the cities and mine workers who purchased them while working in South Africa. 21 As shown in the appendix Table A1, none of the companies could undertake the entire production process alone. For example, Rádio Moçambique was the only company with recording studios, INLD controlled and approved phonographic publication (among other functions such as distribution), and private companies were the only ones that had the means for mass production. It is also relevant to note that multinational companies such as Polygram, Warner, CBS, RCA, and EMI did not penetrate the Mozambican market until the first decade of the twenty-first century. Nevertheless, some Mozambican pressings of international artists or groups such as Michael Jackson, Pink Floyd and Led Zeppelin were mass-produced and commercialised on Mozambican soil. The importation of discs could happen in two ways: firstly, via the purchase of the final product, paying import tax on each of the units acquired; or secondly, and perhaps the most common solution adopted in Mozambique, through the acquisition of matrix discs that would later be used for serial reproduction in Mozambican pressing plants (Malm and Wallis 1984: 76). In these latter cases, the production process would omit phases 1, 2 and 4.

Publishing and copyright policy
Given the inter-dependency between the various companies involved, a mutual aid agreement was negotiated, which guaranteed the exclusivity of some music markets. For example, the two private companies were mainly dedicated to editing foreign phonograms, while INLD was responsible for producing phonograms of national origin. There was, however, one exception: INLD oversaw the editorial exclusivity of the South African artist Miriam Makeba, at the time exiled in Guinea-Conakry, and with whom the label allegedly had signed a contract for the edition and distribution of some of her records, not only in Mozambique but also in Angola, Botswana, Cape Verde, Guinea-Bissau, Lesotho, Swaziland, Zambia, Zimbabwe, and Tanzania. 22 After 1982, most INLD activities would be transferred to Rádio Moçambique, 23 except for distribution. António Alves da Fonseca, one of the radio producers involved, explained this change: Instituto Nacional do Livro e do Disco took care of many things: it imported toilet paper, printer paper, paper for books, and managed the phonogram industries in Mozambique. At one point, I said: how do you want the Radio to solve your problems? INLD had a much bigger budget than the national radio station, where the recording process would actually take place. I told the director that I disagreed with that situation … They [INLD] didn't even have recording studios. So, from that moment on, they passed all recording duties to Radio Moçambique, and that's when we started to publish all national records. (Interviews with author, Maputo 11-10-2016, 14-10-2016, 18-10-2016, 10-04-2017) As far as private companies are concerned, there was a well-defined editing policy for foreign content marketed in Mozambique, which was in line with the music policy adopted for radio broadcasting. 24 Thus, according to the owner of Teal, Gomes Leitão, in an interview published at Tempo magazine in December 1976, 'the vast majority of the records we produce, close to 80%, is of African music', while the remaining 20% was Anglophone popular music. 25 Regarding export policies, there was an attempt to capture the foreign market, not only by publishing Mozambican music in countries such as Zimbabwe, Angola, and Cape Verde, but also to promote future collaborations with neighbouring countries that did not have record factories at the time, such as Tanzania. 26 These procedures reveal private companies' willingness to promote and maintain relations with international enterprises, especially since they knew that domestic production would not provide adequate profits to survive. However, besides selling Mozambican pressings of selected international hits approved by the censorship committee, the proposed export policies never materialised.
What about copyright? How did it work? The answer is simple and brisk: it didn't. After independence, the Portuguese Authors Society (SPA) ceased to function in former Portuguese colonies, leaving a legal void that, in Mozambique's case, lasted until the creation of the Mozambican Society of Authors (SOMAS) in 1998. However, the first royalty payments were not made until 2002. Consequently, artists working in music, literature, theatre, dance, and other audiovisual productions within Mozambique did not receive any 'official' copyright payment from the sale and use of their works between 1975 and 2002, even if the disc labels and inserts indicated otherwise. Within traditional music, the situation would be even more dubious since, a priori, authorship was assumed as collective-and, from a Frelimist logic, it would be 'the Mozambican people'. 27 Several factors explain this delay, such as artists' lack of knowledge that these rights could exist in the first place. According to radio producer António Alves da Fonseca, 'when we initially appealed for musicians to come and record on the radio, they would leave without any payment. In fact, they were happy simply because their songs were recorded'. 28 Later, musicians began to be paid only at the moment of recording. According to Fonseca, the fee varied from artist to artist, which caused disagreements between musicians and producers: 'for example, if you paid 500 to a famous artist such as Fanny Pfumo, a kid who didn't have any experience also wanted 500'. 29 Famous musicians such as António Marcos, received 2500 escudos, while Hortêncio Langa received about 'três contos' [or 3000 escudos] for the recording of his LP within the NGOMA series. According to producer Domingos Macamo, groups received around 13,500 escudos that would be given to the leader, who in turn distributed to all members. After distributing the money, each musician was left with very little. 30 It didn't matter if the record sold one or thousands of copies; it didn't matter if the song was successful or not; for musicians, the process would be finished when they received the monetary compensation for recording, and thus all sales profits would go entirely to the record companies and Rádio Moçambique or be redistributed to other government administration sectors, as was common practice in most socialist-based states.
Gradually, artists became aware that they were being exploited, and that there were numerous payment frauds. According to producer Domingos Macamo in an interview in Tempo magazine, in Rádio Moçambique, 'a book measuring approximately 60 × 30 centimetres is used to register the recordings published. It is a rudimentary record, but it fulfils the function of identifying any song that has been edited on the NGOMA label, and thus bars the way to musical plagiarism'. 31 However, in the absence of copyright law, that book could not protect the musician's intellectual property, serving only for internal review purposes.
Comparing the Mozambican case with the twelve countries analysed in the MISC project by Malm and Wallis, Mozambique may be seen to be on an equal footing with Tanzania, Jamaica, and Trinidad. In 1982, when the project published its findings, none of these countries had yet signed an international copyright agreement, such as Rome/Geneva or Berne (Malm and Wallis 1984: 209-10).
4. 'NGOMA: the Mozambican national label': musical trends and notable absences INLD created the 'NGOMA' label in late 1978, intending to produce and publish Mozambican music. The term Ngoma is intelligible in most Mozambican languages, meaning 'drum' and 'song' (Sitoe 2011: 214). It published more than 160 albums from 1979 until 1990 (including 45 rpm singles and 33 rpm LPs) mainly related to música ligeira, with occasional recordings of traditional music and revolutionary anthems. The disc labels contained little to no information on the song's lyrics or the producers involved. The singles did not have a cardboard sleeve, being packaged inside a fragile paper envelope with a circle cut in the centre so that the disc information could be read directly from the label (see Figure 4). With the catalogue number 'NGOMA 0001', the first publication was a 45 rpm single by Alexandre Langa entitled Aku hluliwa kava kolonyi ('The colonialists' defeat') (see Figure 5).
The lyrical themes revolved around FRELIMO's moralising values and sought to contribute to the 'nation-building' process. The songs Hinkwavo Vanu ('Everybody') by Tomás Ubisse 32 and Utlhassile Nkama ('The time has come') by Azarias 33 make direct references to the Liberation Front and its 'socialist project', while the song Zimbabwe by Eusébio Tamele 34 reflects on the independence of the neighbouring countries, while promoting the internationalist values of the movement. Entreaties against school disinterest and delinquency are present in the songs Akudonza ('To study') by Pedro Ben with Grupo RM 35 and Aku Famba Vusiku ('Walking at night') by Vicente Soto 36 , respectively. The texts always emphasise the consequences of such acts by employing a moralising tone: 'not knowing how to write a letter' or 'don't cry if you end up in prison'. On the subject of alcoholism, we find several songs that highlight the different consequences of this addiction: Abyala Byitxela Vusiwana ('Drinking brings sadness') by Rafael Cossa with Grupo RM 37 tells the story of a man who spends his days drinking instead of 'fixing up' his own house, and, for that reason, he would get sick, leading to his death; in Eusébio Tamele's Mingaholovissane 38 there is a plea not to drink alcohol, explaining that it sows discord and conflict; while in . The disc is enclosed in a paper envelope with the centre cut away so that the information can be directly read from the label.
Xinwanana ('Child'), 39 Azarias tells the story of a child that was taken from his mother because she was always drunk.
Nonetheless, by far the most common theme is marital problems. For example, Aniyengile Sati Wamina ('My wife's insistence') by Rafael Cossa with Grupo RM 40 speaks pejoratively about getting married and continuing to live in the parents' house; in José Matiranze, 41 Tomás Ubisse tells the story of a man who seduced a woman with just five hundred escudos; Nikuine 42 and A Nsati Lwé 43 by Grupo Revelação, and Mingatsikete Va Sati Veno ('Do not abandon your wives') by Arnaldo Silva with Xiwora Mati group, 44 focus on abandoning the home, which was very common in southern Mozambique. However, some of the most interesting examples emphasise the woman's perspective, notoriously sung by a man. In Nuna Wa Mina ('My husband'), 45 Vasco Mulhanga speaks in the first person as if he was a woman who is waiting for her husband to return home, while the song Uniyengile Wena José ('Insist José') by Vicente Soto 46 criticises men who promise women everything, but never fulfil those promises: 'You don't know how to conquer a woman. You promise everything, but then there is nothing, neither shoes nor headscarves'.
Most songs are characterised by direct interrogation, as if the musicians were reprimanding an action, be it a thief trying to steal someone or a teenager smoking illegal substances. There is also a didactic character evident in the lyrics of some songs, elevating them to a teleological nation-building cause or to the project of invigorating the 'new Mozambican man'. Accompanying this are incitements to counter the values of the socalled 'vicious man', related to colonialism, including, among others, ' … bribery, corruption, immorality, robbery, nepotism, individualism, prostitution, banditry, marginality, drunkenness, drugs, and insecurity' (Cabaço 2001: 112). As I have demonstrated elsewhere, música ligeira (including all national and international popular music genres), initially an outlier given its association with the large colonial cities and foreign influences, was now being instrumentalised by the state to the service of a national teleology (de Freitas and de Carvalho 2022). Therefore, every song should have a precise political angle, to the extent that singing about something trivial such as 'love' could be, for a time, considered 'reactionary' or 'imperialist', as made evident by radio producer Américo Xavier in the following statement: One day, I remember saying: from morning till evening, we're always singing about the revolution, but life is more than that. I asked producer Domingos Macamo if he felt that the songs were somehow 'Frelimist', given the fact that the messages came mainly from the current political discourse. Macamo confirmed this idea, adding that 'more than Frelimist, they were, above all, an exaltation of Mozambiqueness which, in my opinion, is much more profound and important than a political party'. 47 The analysis of these vinyl discs also reveals two predominant performing styles: the first has an intimate character, including voice and guitar accompaniment with a threechord harmonic structure. The melodies are simple and repetitive, highlighting the importance of lyrics in songs which were primarily sung in southern languages. Recordings by Vicente Soto, Tomás Ubisse and Eusébio Tamele are representative of this style. The second performative style is analogous to a 'group' or band typology, including one or two electric guitars (rhythm and solo), a bass guitar, drums, percussions, and sometimes wind instruments, such as saxophone and trumpet. All songs accompanied by Grupo RM-Radio Moçambique's official orchestra-are representative of this trend.
Coinciding with the creation of the NGOMA national label, the formation of a radio orchestra in 1979 was intended to increase the production of new music for radio broadcasting, phonogram publishing and, eventually, live music concerts throughout the country. Although inspired by the radio orchestras of the colonial period, this group differed substantially from its colonial counterparts, which mostly catered for art and Portuguese variety music (de Freitas 2021a: 450-54). The new radio orchestra, by contrast, focused on three main areas: 1. performing 'old' music, that was accepted as 'Mozambican'; 2. creating new, original songs inspired by traditional music; and 3. accompanying musicians who recorded at Rádio Moçambique, notably those who published with the NGOMA label (de Freitas 2022b: 9-10).
Since many groups shared the same instruments, their 'sound' became standardised, even regarding their technical problems. For example, every song with piano accompaniment sounds out of tune due to the use of the radio auditorium's (very) old piano that needed maintenance. In the process, even with its intricacies, a new 'national' sound emerged, and, aside from being commodified as vinyl discs, it became widely accessible through radio broadcasting. With refrains of songs in their heads with specific calls for action, Mozambicans felt they were creating a new Mozambique, transforming these songs into effective and exceptionally influential nation-building tropes. Moreover, the existence of a national recording industry capable of producing and forging its own 'national sound' also legitimised perceptions of the new country's cultural sovereignty, independence, and self-sufficiency.
Another notable aspect is the total absence of female solo artists on NGOMA discs, which stems from musicians' weak social status and stigmas related to women's status in Mozambican society. FRELIMO's first organisation dedicated to women-LIFEMO (Liga Feminina de Moçambique)created in Dar Es Salaam (Tanzania) during the liberation war, aimed to provide support to families that joined FRELIMO and integrate women into the combat ranks (Pachinuapa 2012). In 1973, as the liberation struggle unfolded, a new broader organisation was created-OMM (Organização da Mulher Moçambicana). According to sociologist Isabel Maria Casimiro, OMM aimed to counter so-called 'obscurantist values', including premature and forced marriages, lobolo (marital compensation), polygamy, initiation rites, and prostitution, while promoting the integration of women in domains that, until then, were traditionally consigned to men. However, as noted by Casimiro, women's integration into paid work was not accompanied by changes within the family sphere, as they continued to manage domestic duties alone (Casimiro 2001: 100). As such, one can assume that there were no conditions (or time) for developing an artistic career, more so since being a musician was not even considered a profession during the first years after independence.
I had the opportunity to ask Salomé Moiane, former General Secretary of OMM (from 1977 until the mid-1980s), for possible reasoning behind women's absence in NGOMA. She seemed very surprised by my inquiry, confirming that this wasn't felt to be an issue back then: 'I never thought about that! We almost didn't have female singers, that is the truth … now why was that? […]. Back then, our main objective was to integrate women into the productive areas of society such as farming. Perhaps the way we [OMM] handled things turned women away from cultural practices. Probably it was perceived by them as not being a priority'. 48 Furthermore, producers assured me that women refrained from recording because they adopted a self-censorship posture, preventing them from being misinterpreted as negligent housewives and mothers. Producer Domingos Macamo, who was once responsible for the music production at Rádio, expands this idea: Author: Why didn't they publish music by women? Domingos Macamo: Don't you know why? If a woman wanted to sing, she would be seen as a prostitute, someone with a wrong way of life. So, parents didn't let their daughters sing música ligeira, even if they had a good voice. There were exceptions, such as Elvira Viegas or Zena Bacar … but singing was like football: 'it's something for men'. Have you ever heard of any female footballers out there? A: But did you reflect on this? Did you realise there were only men? Or was it so natural that it didn't even cross your mind? DM: It's one of those stigmas that society imposes and stays in our minds. A: Who was the first woman recorded for publication on disc (not for broadcasting)?
DM: Elsa Mangue [1986], I think. But that's how things were; it was the mental system. No father would let his daughter sing. A: But here [Maputo], Miriam Makeba was heard a lot, and apparently, INLD even had exclusive publishing rights for her records in many African countries. Didn't that inspire young girls to sing? DM: It wasn't a question of inspiration; it was the system. If a woman dared to sing … she would be marginalised'. (Interview with author, Maputo 26-03-2018) Producer Américo Xavier echoed Macamo's perspective, providing a similar explanation for women's notable absence from música ligeira, as well its musicians' weak social status more generally: 49 'Our society is essentially sexist. In Mozambique, we have areas where matriarchy reigns, but in the end, it is the man who makes the decisions […] It's that idea when they introduce a boyfriend to their parents: "So what's your boyfriend's job?", "he's a musician", "yes, but what's his job?", as if being a musician was not a proper job. So, for a woman, that was even worse'. 50 Another obvious problem concerns the prevalence of repertoires from the three southern provinces of Mozambique: Maputo, Gaza and Inhambane. For an artist from the Centre and North to be published on the NGOMA label, it was necessary to travel to the capital, where the studios were located. For many, this journey was unfeasible, especially during the 16-year war between FRELIMO andRENAMO (1977-1992). There were, however, some exceptions, such as Chandú Lacá from Tete, who published songs in the Nyugunwe language. 51 Since NGOMA was advertised as the 'Mozambican national label', what about the other seven provinces of the country?
Another interesting aspect to highlight is that despite being adopted as the national language after independence, NGOMA albums sung in the Portuguese language are rare. This situation seems to contradict the official discourse of that period, which presented Portuguese language as a kind of unifying lingua franca for the country (Magaia 2010: 127-29). According to former Minister of Culture Luís Bernardo Honwana, 'the Mozambican languages did not enjoy any special status […] one can even speak of disdain', further adding that 'in the cultural area, production in Mozambican languages is scarcely used by cultural industries, except in the case of urban music' (Honwana 2017: 81). The question one might ask is: why was popular music an exception? It might be supposed that the prevalence of southern languages can be explained in terms of convenience, namely proximity to the recording studios. In that case, there remains an unanswered question: since the Portuguese language was promoted as the 'language of national unity', why was most music recorded in local languages? According to my interlocutors, several factors underlie this tendency: insufficient knowledge of the Portuguese language; preference for speaking and singing in the language used in everyday life since it embodies many linguistic metaphors that, when translated, would lose their intended impact and meaning; and the 'musicality' of local languages, whose characteristics, namely syllabic relationships, would lose its effect when translated into Portuguese.
However, the most frequently cited reason was the nature of the themes explored in the lyrics that, as previously mentioned, mainly concerned social problems: interpersonal relationships, family dramas, home abandonment, alcoholism, and drug use. If a song employs a moralistic or ironic tone to comment on a community's problems, it presupposes that the receiver belongs to that same community. Therefore, it would be only natural for the songs to be sung in the language used in the contexts described, constituting an indispensable element of that characterisation. This idea was expanded by former Minister of Culture for Mozambique, Luís Bernardo Honwana: [There is a prevalence of songs in local languages] because urban popular music is not a filtered expression, even though there have been attempts to police its use. When it occurs, it emerges spontaneously and from the community itself and expresses the things that happen there in a humorous, critical, or moralistic way. It is an artistic production in the form of dialogue, returning that comment to the community […]. The only vehicle through which this can happen are the languages that the community itself speaks Given these various factors, the incorporation of national languages into música ligeira in Mozambique may in part reflect opposition to frelimist 'unity' values. More importantly, this represents a striking contradiction since radio broadcasting announcements had to be given first and foremost in Portuguese. By favouring repertoires from the south of the country, which would later be amplified through the National Broadcast of Rádio Moçambique, the practice of música ligeira recorded in Mozambique was deemed as 'regionalist', which, in turn, would come to represent a major challenge for FRELIMO and its efforts to unify all Mozambicans under the same flag in an exceedingly arduous war context.

Que Venham!: phonographic production during the war
The civil war brought dire consequences to Mozambique, with a million people killed and five million displaced. In addition to carrying out merciless massacres, RENAMO's warfare strategy resorted to acts of sabotage that heavily impacted the already frail Mozambican economy, as it struggled to supply stores with the most basic products (Mosca 1999: 54).
Eventually, the 'war' became a standard theme in lyric songs, initially supporting a pro-government FRELIMO perspective; later this transformed to direct appeals to end the war, regardless of support for any of the factions. A notable example is the song 'Que venham' ('Let them come') by Yana, which was inspired by traumatic events on 30 January 1981, when South African apartheid commandos killed 16 African National Congress (ANC) fighters living in the suburbs of Maputo. In the aftermath, President Samora Machel made a direct appeal to the population to prepare for war: 'They want to come here and commit murder. So, we say: Let them come! Let all the racists come! … Let the South Africans come but let them be clear that the war will end in Pretoria!'. 52 This speech caused a great stir among the population. The expression 'let them come' became one of Machel's most famous slogans, and its potency was reinforced through the song by Yana, a young Mozambican musician who had just arrived from North Korea, where he had undertaken studies in music conducting and composition. 53 This is among the most striking examples of how music was able to raise awareness of the problems plaguing the country and of FRELIMO's internationalist cause. In addition to being broadcast incessantly on Rádio Moçambique, the President's musicalised speech could also be purchased on vinyl. 54 Simultaneously, the lack of resources started to impact the production of recordings in Mozambique. The scarcity of specialised technicians, the frequent 'economic sabotage' of production machinery, the lack of foreign exchange, and the so-called 'bureaucratic labyrinth', effected various industrial sectors. 'The result was almost always the same', according to João Mosca, a specialist on the Mozambican economy, 'after a few years or even months, the so-called machine cemeteries expanded' (Mosca 1999: 103), as much equipment ceased being used due to a lack of technicians, maintenance, or replacement parts. Naturally, the phonographic industry also suffered from this situation: over the years, many newspaper reports alluded to the poor quality of record pressings and inserts. 55 However, a fundamental issue was the difficulty acquiring reproduction hardware such as turntables. Before independence, these were easily obtained in South Africa, since migrant mine workers commonly brought them in their luggage when returning to Mozambique. After independence, the free movement of goods and people from and to South Africa was greatly restricted. Stores would only sell essential products, while a lack of imports meant that 'luxury goods' disappeared from the market (Mosca 1999: 98). Although the Mozambican State subsidised the production of low price 'Xirico' radio receivers, 56 turntables were not a priority and thus disappeared from stores (de Freitas 2022b: 4-5). Regarding the recording process, in 1981 INLD admitted that the radio recording studios were not of 'sufficient quality to satisfy producers' and confirmed that most artists published by INLD/RM used the Grupo RM instruments, thus guaranteeing the musical uniformity of some of the repertoire published under the 'NGOMA' label. 57 Matrix production in South Africa was disrupted by 'the lack of royalties' payments by Somodiscos and Teal Discos, essentially due to 'bureaucratic embarrassments, which caused the withdrawal of South African producers'. 58 A representative of the INLD explained the matter as follows: We tried to overcome these problems, but the situation is not yet normalised. We looked for other international companies to produce our matrices, namely in Portugal and Cuba. However, the results were not as good as we had hoped in terms of prices and delivery times and, in Cuba's case, the quality of the matrices. Due to these difficulties, we had to turn to South Africa again. 59 The matrix problem disrupted the timely release of several important publications, namely the disc from the 1st National Festival of Song and Traditional Music, published a year later than expected. 60 Other situations were also reported, such as 'unfinished covers without captions and images', 61 or the lack of raw materials such as paper and cardboard, a situation that held up the publication of a disc promoted by the Ministry of Health and UNICEF. According to a newspaper report, the disc's 'matrix has already been made, with traditional music intended for children […]. It was being prepared to be released on 1 June 1984, and did not happen due to a shortage of paper for the cover at Minerva Central and a lack of material to produce the record at Somodiscos'. 62 The flight of workers in the music business, the closure of musical instrument stores, the lack of turntables, and the scarcity of essential materials for phonographic production, such as vinyl and paper, contributed to the collapse of the record industry in 1985. Producer Domingos Macamo summarised the situation: 'at a certain point, we had to choose between music and health; which one is more important? Everything that was considered leisure was abandoned'. 63 Activity in the music business would only resume from 1989 onwards, boosted by economic aid from the Bretton Woods institutions, including the International Monetary Fund, which enabled the re-establishment of a supply of raw materials essential for record production. This resurgence brought new market trends and the (late) introduction of another reproduction format-the cassette. 64 Several factors supported the transition from vinyl to cassette: firstly, the low production and duplication costs, allowing a reduction in prices for the consumer; secondly, the portability of the cassette compared to vinyl discs; thirdly, the ease of the duplication process, which could be undertaken entirely at Rádio Moçambique, thus eliminating dependence on external factories, such as Somodiscos or Teal Discos, and South African companies for matrix production. This process happened during the transition from a single party to a multiparty regime (1990)(1991)(1992)(1993)(1994) and, as such, falls outside the period defined for this article. This transitional period will be the focus of a future publication.

Postlude: the tale of a fledgling industry
The performance of socialist popular music industries under state ownership has been generally mixed. On the one hand, popular musics under socialism avoid most of the negative features of commercialism, including the link to corporate sponsorship and consumerism, the fetishism of stars and fashions and the deforming pressures exerted on musicians by the market. At the same time, most communist countrieswhich were underdeveloped to begin withhave been unable to devote adequate financial resources to entertainment industries and related sectors such as consumer electronics. Bureaucratic inefficiency and authoritarian cultural policies have exacerbated problems of creative innovation and material production. (Manuel 2001) When juxtaposed with the Mozambican case, Peter Manuel's assessments of the phonographic industries in socialist countries could not have been more accurate. After the country's independence, the companies responsible for phonographic production underwent a complex adaptation process, aiming to conciliate the available repertoires with what was expected by the Mozambican State without disregarding the main objective of any industry: to make a profit. However, the process was not simple since production was heavily dependent on different organisations/companies that, in turn, were forced to work collaboratively. The fragmentary process was further complicated by the lack of essential materials for disc manufacture and dependency on matrix production companies in South Africa-at that time, widely regarded as FRELIMO's enemy.
Given the complexities of the production process, why would the government put so much effort into producing/publishing vinyl discs and creating the phonographic label 'NGOMA' (also known as 'the Mozambican record label')? As I have demonstrated elsewhere, 'sound' and 'performance' were primary vehicles for promoting 'the new Mozambican man': 'Indeed, sound-be it humanly organised as a "musical performance", or technologically mediated through radio broadcasting-played a major part in Mozambique's nation-building process, with news, encouragement words and choral singing' (de Freitas and de Carvalho 2022: 2). Phonography-mainly through vinyl discs-also played a fundamental part in that process as 'material vehicles of communication and culture' that could 'unite people much more than a book', as affirmed in the quote that opens this article. 65 However, the fragility of these creative and industrial processes leads to the question: are we dealing with a mature or with a fledgling industry? The following six factors elucidate why I am more inclined towards the latter: first, the idea of 'commercialising' music in a socialist country was highly devalued; second, the lack of financial resources for production and reproduction machines (including matrices), musical instruments, adequate studio material, raw materials such as PVC (vinyl) and turntables; third, the lack of qualified technicians to guarantee production and pressing quality; fourth, the imposition of an authoritarian cultural policy that, in many ways, constrained creative processes and the social themes addressed in song lyrics; fifth, the de-prioritisation of vinyl disc production, relative to state investment in the provision of low-cost radio receivers; sixth, the lack of copyright legislation, leading to benefits for the producer at the expense of musicians. This is why in the latter half of the 1980s, few artists were motivated to publish discs in Mozambique, instead seeking out alternative markets, especially in response to the growing international interest in what was subsequently called 'World Music'.
Why did Mozambique fail to adopt the cassette sooner? No producer was able to give me a good explanation. Besides material considerations, we might speculate that FRELIMO was anxious that the more democratic and flexible medium of the cassette could open the door to the circulation of music and audio recordings outside the party's control. Hence, the unitary and fixed nature of vinyl discs, without the option to erase and record again, would keep the phonographic process under the strict supervision of the state. Such control was even more critical in the context of the violent civil war engulfing the country.
Aside from developing music production in Mozambique and, in the process, promoting the publication of Mozambican música ligeira, NGOMA's original premise was to 'give voice' to different ethnic groups from all over the country, while controlling and standardising its primary message around moralising themes that reflected on social problems, such as home abandonment, delinquency, and disinterest in school. Slogans and political speeches inspired most song lyrics, with Yana's 'Que venham' offering a paradigm example. However, material considerations and the war meant that the published records would be mainly restricted to groups from the South of the country, including repertoire sung in Xironga and Xichangana, which in turn was unintelligible in other provinces. Another problem concerned the absence of female singers, a situation that stemmed from musicians' low social status and stigma related to the role of women in Mozambican society.
Ultimately, the absence of music from other provinces and the virtual exclusion of women's voices worked against FRELIMO's desired objective of 'unity' among all Mozambicans. Indeed, many did not feel included in this project, especially when amplified through broadcasts via national radio (where certain artists became widely known and promoted). In the end, the 'NGOMA national label' might have backfired as it was not so 'national' after all. One might even question its eventual contribution to the demise of the so-called 'socialist experience' in Mozambique from 1983 to 1986, and its role in igniting the ongoing feud between the country's northern and southern provinces. Nevertheless, even with its limitations and paradoxes, many of my interviewees still acknowledge the 'NGOMA years' as the best ten years in the history of Mozambican music. Notes 1. 'Fábrica de discos Teal. Rotação acelerada de produção em série', Tempo, 323, 12-12-1976, pp. 42-47. This and all subsequent cited documentation were translated into English by the author.
2. Mozambique Liberation Front (FRELIMO) was created on 25 June 1962, in Dar Es Salam, Tanzania. Eduardo Mondlane served as its first president until his assassination on 3 February 1969. As a United Nations official, Mondlane first tried to negotiate independence through diplomacy; however, after the continuous obstinacy of the Portuguese authorities, he decided to start an armed guerrilla struggle. On 25 September 1964, the first attacks took place in the provinces of Cabo Delgado and Niassa, starting a liberation war that lasted for almost ten years until the signing of the Lusaka Accords on 7 September 1974. After a transitional government, independence was formalised on 25 June 1975, with Samora Moisés Machel (1933Machel ( -1986 as the first president of the People's Republic of Mozambique. The armed struggle and the communal experiences in the so-called 'liberated areas' served as inspiration for creating a new society that would control its means of production and its expressive practices. 3. According to Ngunga (1999), 93% of the Mozambican population was illiterate at the time of independence in 1975. 4. In this article, the term 'post-colonial' is mainly used in a more literal connotation, to characterise the period initiated with formal decolonisation as part of the 'nation-building' process. This does not neglect the approach of 'postcolonial studies', where the term serves as a privileged space for the contestation and deconstruction of power structures and discourses produced from a privileged colonial perspective. It also aims to understand the persistence of exploitation processes-frequently labelled 'neo-colonialists'-beyond formal independence, as confirmed by the continuity of social, economic, and political inequalities that prevailed in previous forms of governance (Cooper 2005 in turn, applied here as defined by Richard Middleton, as being directly correlated with the goals of the mass media and music industries. The author also relates popular music to a social class in anglophone contexts (usually medium/low) (Middleton 2001). 6. I define 'cultural policy' as a set of institutional initiatives outlined by public or private organisations to stimulate and orient the cultural practices of a particular group or society that, among other objectives, can be used to obtain consensus in the process of social transformation, as in the case explored in this article (Côrte-Real 2002; Castelo-Branco 2008; Nery 2010). Through the analysis of this policy, one can understand how a particular group intends to represent itself, not only for internal coherence but also for external or international projection. Consequently, the study of 'cultural policy' should consider the social conditions that allowed the construction of the cultural field, its history, and the respective agents and institutions involved in its development, reproduction, and self-legitimation (Bourdieu 1989: 289). 7. For more information on the relationship between Ethnomusicology and Music industries see volume 19 (1) of the journal Ethnomusicology Forum (Cottrell 2010). 8. Big Sounds from Small Peoplesthe Music Industry in Small Countries, a book by Krister Malm and Roger Wallis, published by Pendragon Press in 1984, is a pioneering contribution to the study of the music industries in 'small' countries. In total, 12 countries were chosen for a comparative project entitled 'Music industry in small countries' (MISC): Jamaica, Trinidad, Tunisia, Tanzania, Kenya, Sri Lanka, Sweden, Finland, Denmark, Norway, Chile and Wales. According to the authors, the concept 'small countries' is defined by three factors: reduced population, reduced material resources and small geographical area. In 1992, the authors revaluated the project's content in Media Policy and Music Activity, focusing this time on only six countries (Jamaica, Trinidad, Kenya, Tanzania, Wales, and Sweden). For more information see: Malm and Wallis (1984), and Malm and Wallis (1992). 9. Such as de Carvalho (1999), focusing on the appropriation of Makwayela as a 'national dance'; and Israel (2014), focusing on Mapiko practices between 1917 and 2009, from the height of the colonial period, through the liberation war and subsequent socialist and post-socialist periods, among others. 10. Except for Mendes (2021), these studies focus on Marrabenta, a south-based popular music genre.