Introduction
Ruth First came to Mozambique twice to work at the Centro de Estudos Africanos (CEA): first, in 1977, to lead a research project on the export of Mozambican migrant labour to the South African mines, and, second, from 1979 until her death in 1982, as research director of the CEA in charge of setting up and running its research-based development course. Before this, she spent a term in 1975 as visiting lecturer in development studies at the University of Dar es Salaam (where I was then working in the economics department). During her stay in Dar, Ruth went on a short visit to Maputo where she met Aquino de Bragança, director of the CEA, and Fernando Ganhão, Rector of the Eduardo Mondlane University. Aquino de Bragança was a long-standing confidant of the Frelimo leadership and, like Ruth, had been a journalist for many years, writing mainly for the Revue Tiers Monde. Fernando Ganhão, a historian, had joined Frelimo during the 1960s, was a teacher at the Frelimo school in Dar es Salaam and was appointed Rector of the University after independence. They agreed that Ruth would come to work during 1977 to teach a course on the history of southern Africa and to undertake a research project on migrant labour.1
At its inception in early 1976, the CEA was meant to be a research centre on history and anthropology, divided up into sections representing different pre-colonial and colonial periods. Its Mozambican researchers were mainly BA graduates from the history department working on their individual research projects. There was a vibrant series of seminar presentations given by these young graduates and by various visiting scholars (historians/archaeologists/anthropologists), among whom were a number of famous French scholars who worked on Africa. Little work was done, however, on contemporary issues and none on the challenges of transition after independence.
Although Ruth was not in Mozambique during 1976 she was nevertheless intensively involved with the CEA since she was busy preparing her teaching and research for 1977. This preparation did not go without some hiccups, not the least because of the confusing messages Ruth kept receiving from Maputo about the work she was supposed to do in 1977. At the time, the CEA was deeply involved in plans to set up a new faculty of social sciences. This engendered much talk but little action apart from the endless submission of drafts of course curricula, all of which were rejected by the Rector. At that time, this faculty never materialised. Conflicting messages about her teaching for 1977, however, led to concerns. The main problem was the lack of a clear development perspective – or, more accurately, a political economy of development perspective – in the various drafts of the course curriculum. In a letter to Dave Wield (21 September 1976), for example, Ruth commented: ‘I may say that if I did a SA course I would fashion it after the design of a development course anyway … less history than a political economy, but don't say it aloud to the wrong people … .’
Aquino de Bragança provided the initial impulse that would change the direction of research at the CEA when – at the end of 1976 – he initiated a small research project (or, more accurately, a desk study) on the Rhodesian Question in the run-up to the Geneva Conference on Zimbabwe. Aquino was then deeply involved with the process of decolonisation in Zimbabwe as advisor to the Frelimo leadership, which drew him into a multitude of discussions with the Zimbabwe African National Union (ZANU), then based in Mozambique. Samora Machel requested Aquino to prepare a background paper on the socioeconomic situation in Zimbabwe, since Frelimo wanted a better understanding of what kind of issues, tensions and contradictions were likely to emerge in the process of decolonisation.
The report itself was not a major piece of research since materials on Rhodesia/Zimbabwe were thin on the ground in Mozambique and none of the team members was an expert on Zimbabwe. In terms of content, therefore, it was a rather modest undertaking, although it was well received and subsequently published as a book in Portuguese and in English (CEA 1977a, 1980) and as an article in the Revue Tiers Monde in January 1979. But the experience itself – an intense process of team-based researching, writing and editing, spiced with fascinating group discussions chaired by Aquino, all at breakneck speed within about five weeks – was highly significant. It had an effect akin to giving the CEA an electric jolt.
It was the research project on the Mozambican miner, however, that turned this jolt into a novel research practice. Here I focus on how this project acted as a catalyst in changing the practice of research at the CEA by introducing three innovations: (1) a focus on the ‘actual’ (while taking account of its historical roots) rather than on history; (2) a switch from individual-based towards predominantly team-based research rooted in fieldwork; and (3) the introduction of a sense of (political) urgency in research to respond to immediate challenges of transition.
Fighting for space: contesting the nature of applied research
For Ruth, this research project was not about history, but about researching the challenges that the inherited colonial dependence on earning foreign exchange (more accurately, gold) through the export of mine labour raised for the transition to socialism in Mozambique. The first obstacle to overcome, however, was to bargain for space to do this type of research. This was not a minor issue. In fact, when Ruth went to see the Rector with the request to do the research over a seven-month period (including one month of fieldwork), his immediate reply was: ‘Why seven months? Can it not be done much quicker?’ Ruth was completely taken aback by this reply. She was well aware that a research project of this nature could not afford a lengthy period of gestation since the pace of transition might render its subsequent output obsolete. This is why she envisaged doing the research within as short a period of time as possible to produce an output that would be relevant to policy making (at a time when major changes were taking place all at once), yet with sufficient analytical depth to arrive at meaningful insights and conclusions. The Rector's reply, however, not only suggested disagreement about the time involved, but also reflected deeper conflicting conceptions regarding the point of research in the context of Mozambique.
There could hardly have been a greater contrast between the University of Dar es Salaam and Eduardo Mondlane University at the time. The University of Dar es Salaam was a vibrant place of innovative teaching and field-based research with lively wide-ranging debates involving local and expatriate scholars, staff and students alike, which acted as a magnet attracting left-wing intellectuals from all over the world, including Ruth First. In contrast, Eduardo Mondlane University had barely begun to confront the process of transformation from a racially imbalanced colonial institution imprinted with the marks of the long history of fascist rule in Portugal (ended by the coup d’état in April 1974). Teaching and research in social analysis was particularly limited, since, while the University had a faculty of economics (and management) and a faculty of languages, literature and history, it did not have a faculty of social sciences (de Brito 2011). After independence, most of the academic staff left as part of the massive exodus of the Portuguese settler population. Those who stayed, however, tended to be mainly young academics, many of whom had been linked to various underground left political movements, the most prominent of which was the Portuguese Communist Party. The gap in teaching staff was gradually filled through the recruitment of teachers from almost every corner of leftdom, East and West.
This, then, was a time of intense struggle over defining the role of the University in a period of transition and, more specifically, over how to transform teaching and research practices. It also involved intense contestation about whether Marxism should be taught as a given doctrine or as a method of enquiry and about what its role should be in research. It is naïve to say, however, that these struggles only took place between expatriate staff. On the contrary, they were equally alive in the then existing Mozambican academic community with its diversity of past political affiliations. A distinctive feature of these struggles, however, was that they tended to operate more through ‘positioning’ – that is, by restricting the space for alternative views – than through open debate (on which the Rector kept a rather tight lid).
It was quite common at the time – and Rector Ganhão was a major exponent of this view – first to make a rigid distinction between ‘pure’ and ‘applied’ research and then, second, to argue that the former required deep (theoretical) reflection and lots of time, while the latter mainly consisted of gathering data within predetermined frameworks. Pure research was considered to be a luxury Mozambique could not afford at that juncture. Priority was given to applied research, which, in this view, was not concerned with discovering novel insights. Instead, it concerned the ‘mere’ application of knowledge rather than its production, a task that could be done quite quickly and routinely. This view was particularly dominant at the Faculty of Economics, where applied economic research consisted mainly of gathering data for planning and implementation of policies using standard planning techniques. With some notable exceptions – like, for example, an unpublished paper by Joaquim Brum on the periodisation of industrial development in Mozambique and the joint work of Pereira de Moura and Amaral (1978) on the estimation of Mozambique's gross national product – the focus in economics was on planning, not on analysis.
The dominant view, therefore, was that applied research concerned filling in the details of an otherwise ‘known’ problem firmly grounded in pre-established premises, assumptions and blueprints. The opposite notion, which Ruth defended, was that applied research inevitably involved an act of discovery leading to inferences and conclusions that might challenge established assumptions and ideas and, hence, might not always be expected or welcome. Applied research must enter into the domain of contested views about how to define a problem or look for its solution. The fight over space, therefore, was not just an argument about the length of time involved, but also about the role of research in a process of transition: whether it involved passive execution or implementation of policy or instead active and critical engagement with policy.
It took Ruth all her powers of persuasion to overcome this first hurdle. She staked her reputation on getting the Rector to agree to this project in terms of its proposed scope and duration, although the Rector clearly viewed it as rather a luxury. To his credit, the Rector also took quite a risk in letting the project go ahead because the space for engaged but critical academic social and economic research on the challenges of transition was virtually non-existent or frowned upon within the party and state structures. For Ruth, therefore, the completion of this project in time became an absolute necessity, not just in its own right, but also to open up the intellectual space for a critical engagement with policy through research.
Developing a novel CEA approach and method to collective research
The research on the Mozambican miner was an initiative agreed upon between Aquino and Ruth when she visited Mozambique in December 1975. Research was to be carried out by Ruth during her first planned stay in Maputo during 1977. This was the year in which Frelimo held its Third Congress in February, which initiated a second wave of radical change in Mozambique within a short span of time.
After the setting up of the transitional government in 1974 followed by independence in 1975, Mozambique was in considerable turmoil, not the least because of the exodus of a large part of the Portuguese settler community, which accounted for most skilled labour and owned most medium-sized enterprises, while large-scale investments were owned by Portuguese or foreign capital. The state apparatus lost most of its qualified staff and its capacity for effective administration more or less collapsed. Frelimo's immediate response was to rely on a broad-based mobilisation of people through the formation of grupos dinamizadores (dynamising groups): popularly elected groups of 8 to 10 members set up in townships, rural villages and workplaces across the whole country. Their tasks were extremely complex: they were expected not only to explain Frelimo directives to the population, but also to administer the practical organisation of local-level society and work. This involved resolving issues such as marriage and divorce, settling local conflicts, organising vigilance against economic sabotage, initiating literacy campaigns and holding mass meetings to discuss local issues. These grupos dinamizadores were undoubtedly the principal vehicles through which social change was forged in this early period. They were heterogeneous in composition, a gathering of ‘patriotic’ forces in the struggle against the remnants of colonial society, but with diverging class positions, which often resulted in heated conflicts among the members or between some of the members and the population about what to do and how to go about it. These struggles also reflected real contradictions and gave this initial process of change remarkable vibrancy, fuelled by local-level debate and experimentation.
But another radical wave of change was imminent. When Frelimo held its Third Congress (the first Congress after independence) in February 1977, it decided, quite unexpectedly, to turn the liberation movement into a Marxist–Leninist (vanguard) party rooted in a worker–peasant alliance. In its wake, grupos dinamizadores were gradually phased out and replaced by party cells. Thereafter, the scope of local-level initiative and vibrancy was considerably dampened. Economic policy became centralised under the command of the newly established Planning Commission, which adopted Soviet-style central planning as its principal tool to effect economic change. It focused on the expansion of the state sector and the accelerated integration of the peasantry in aldeias communais (communal villages) and production cooperatives. Despite the various layers of rhetoric, it struck me at the time that, particularly where economic policy was concerned, the process of change was essentially viewed as the radical (i.e. tabula rasa) construction of something new, rather than as a gradual process of transforming the old into something new. This again reinforced the focus on planning rather than on analysis. Moreover, this was a time of great hope – particularly within the Planning Commission – that Mozambique would be able to rely on massive support from the Soviet Union to construct a new socialist economy. All this pointed to blueprinting a radically transformed future, rather than coming to grips with understanding past and present structures and processes.
It was in this (rather euphoric) context that Ruth arrived to undertake a down-to-earth study on the export of mine labour – an inherited colonial structure par excellence. Initially, she was minded to do this project jointly with only a few researchers – three or four at most – but she was open to suggestions. Based on our prior experience of the study on the Rhodesian Question, David Wield and I suggested a collective endeavour involving all researchers affiliated with the CEA. Ruth accepted this, although hesitantly at first because she well aware of the risks involved: on the one hand, the risky trade-off between potentially richer output versus heightened uncertainty of outcome and, on the other, an inevitable increase in organisational work. She was also particularly aware that if we did not produce a finished report at the end of it, the Rector's luxury theory of research would score very highly and, consequently, that the space for this type of research would be severely restricted or closed down altogether. Failure was not an option she was willing to risk, but the decisive factor for her was that this process would benefit Mozambican researchers through a collective process of learning by doing research.
The study on mine labour thus became the first collective field-based CEA research project. The problem was that none of us had experience with organising a collective research project of this nature, particularly not on this time scale. The core team consisted of a heterogeneous group of 12 CEA researchers with different disciplinary backgrounds – history, anthropology, sociology, politics, law, engineering and economics – and varied levels of research experience. Ruth had no readily available model so that the process followed could only be described as thoughtful trial-and-error improvisation. The first problem she had to confront was how to mould this group of researchers into a reasonably coherent team. Her answer to this was to set up a weekly seminar. This initially functioned as a study group, subsequently turning into a discussion seminar with presentations by members of the research team and, finally, ending up as a practical workshop to develop questionnaires, organise the collection of secondary data and discuss preliminary findings prior to rural fieldwork (which took place in July 1977).
The study group was to forge a common theoretical understanding of migrant labour in southern Africa and of the specificity of Mozambican inherited colonial structures and processes, particularly (but not exclusively) in relation to the export of labour power from southern Mozambique. We started by discussing the broader literature on the development of migrant labour systems in southern Africa, but focused in particular on the study of the theoretical contributions by South African Marxist scholars like Harold Wolpe (1972) and Martin Legassick (1974) amongst others. The study group became a seminar when we turned to the Mozambican context with a series of presentations given by team members. António Nogueira da Costa and Luis de Brito gave a seminar on the periodisation of the colonial economy and its impact on the regional diversification of rural colonial structures in Mozambique. Luis de Brito and António Pacheco talked about how the state-to-state agreements between Portugal and South Africa limited the export of labour to southern Mozambique and institutionalised a system in which payment of part of the miners' wages was deferred until their return to Mozambique, and paid out in gold (at the prevailing official price of gold) to the colonial government. David Wield presented a seminar comparing his own work on the periodisation of industrial development in Mozambique with similar work done by Joaquim Brum in the economics department. The seminar subsequently turned into a practical workshop when we focused our attention on the collection and preliminary analysis of secondary data on the flow of migrant labour to South Africa, and formulated questionnaires – on work histories and on rural household composition, agricultural production and employment. In this context, David Wield gave a very useful presentation on the minimum we needed to know about mining as a production process and about the terminology used in the South African mines to denote different categories of workers. The preliminary analysis of data collected was fed back into the seminar, thus gradually turning it into a forum to discuss work in progress, which rendered our discussions more focused and concrete. Through this process, the group developed into a more cohesive research team.
Initially, data collection focused on coming to grips with the system of recruitment and employment of Mozambican mine labour. Recruitment of migrant (mine) labour in southern Africa was monopolised by WENELA (Witwatersrand Native Labour Association), the labour recruitment organisation of the Chamber of Mines, representing the collective interests of its affiliate members (which included most of the major mining companies in South Africa). WENELA had a head office in Maputo and a number of compounds across southern Mozambique where recruited workers were assembled and subsequently dispatched to the mines. The head office was the place to get historical data on mine labour recruitment and to find out about the strategies and tactics of the Chamber of Mines in relation to the recruitment of mine labour from Mozambique. The compounds provided us with venues to collect work histories from workers going to or returning from the mines.
From the first trips to the labour compounds in Maputo and near the border with South Africa it quickly became clear to us that we were in the midst of a very volatile situation. Workers were nervous and uncertain about their future since there had been serious cutbacks in the recruitment of labour from 1976 onwards. Confusion ran high among mine workers about who was behind these cutbacks: the South African Chamber of Mines or the Mozambican government? Most believed that it was the Mozambican government that was clamping down on mine labour. That workers held this view was not surprising since the attitude of most local state and party cadres to miners was at best ambiguous, if not hostile. Going to the mines was treated as a moral vice or reflective of a lack of patriotism.
Ruth was very scrupulous when it came to dealing with mine workers at the WENELA compounds and with local Mozambican cadres. She insisted that during each visit a clear explanation should be given as to who we were and what we wanted to know from the miners. For this purpose, António Pacheco prepared a standard speech – carefully crafted, no promises, no agitation – which he would deliver in a public gathering before any interviews were held. When it came to dealing with the South African management of the WENELA office, the approach was very different. Here Ruth fell back on her skills as an investigative journalist. The manager of the WENELA office was enchanted with Ruth's frequent visits and enjoyed talking to her, which made him much less cautious than he should have been. Ruth was exceptionally good at prying confidential information out of people. She would usually do these interviews jointly with David Wield, who could read text upside-down with remarkable ease. While Ruth kept the manager occupied, interviewing or chatting, Dave would quietly read the messages on his desk top. Alternatively, Ruth would ask whether she could have another look at a fairly innocent telex the manager had shown her before and, while perusing the file with telexes in search of this particular telex, she would quickly scan the contents of other much more important texts. From these visits, and from parallel visits to the Mozambican Ministry of Labour, it became clear that, contrary to what mine workers believed, the initiative to cut back labour came from the South African Chamber of Mines. Numbers were reduced dramatically from 1976 onwards and recruitment was restricted to experienced workers (those with a valid re-employment certificate). It became clear that this was not a temporary measure, but a definite rupture with past trends. In the seminar we speculated on the reasons behind this new trend: whether the Chamber of Mines aimed to internalise labour recruitment within South Africa in the wake of the rise in mine wages and the rising levels of unemployment in South Africa during the early 1970s, or whether it was a deliberate move against Mozambique to pre-empt the threat of an abrupt withdrawal of labour instigated by the Mozambican authorities or to deprive Mozambique of an important source of foreign exchange and rural income.
This question of cutbacks in recruitment imbued our research with an increased sense of urgency. The situation was changing rapidly, which, for Ruth, reinforced the imperative to focus the attention of our research on the direction in which the migrant labour system was moving and not just on how it worked (or, more accurately, used to work). It also placed the research squarely within the domain of policy, particularly since there appeared to be little awareness within the Mozambican party and state structures of the potential economic and social consequences of this dramatic fall in labour recruitment. The prevailing position on mine labour tended to be passive but ambivalent: mine labour was tolerated as a remnant of the past without much consideration of the depth of its roots in existing economic structures and processes. The question of how to transform this deep-rooted system of migrant labour and what this implied for the nature of economic policies to be pursued was never seriously considered.
The month of July was a period of recess during which students (and staff) were required to do extra-curricular activities, usually involving fieldwork across Mozambique. In early 1977, floods in the Limpopo valley had caused major disruptions in Gaza province of southern Mozambique. The government responded by accelerating the process of resettling rural communities into communal villages. The University was then expected to devote its July activities to conducting a massive survey on behalf of the National Directorate of Habitation (which was in charge of resettlement). It took quite some effort on the part of Aquino de Bragança and Ruth First to convince the Rector not only to exempt the CEA research team from this huge commitment, but also to allocate a group of students (drawn from different faculties) to join the rural research project. We could not carry out research in Gaza province, however, but had to confine ourselves to Inhambane province. Ruth's original idea had been to cover both provinces as a challenging contrast between the more fertile lands in the Limpopo valley and the dryer lands of Inhambane.
The logistics of rural field research were daunting. After a brief preparatory workshop to train the students for fieldwork, the team was dispatched to Inhambane and divided into five brigades located at different sites, and a mobile brigade (which consisted of Ruth First, myself and Salomão Zandamela, the CEA driver). At least two members of the CEA core team provided direction to each brigade. By the time the dust settled, the total size of the research team had grown to slightly more than 40, with the addition of cadres from the local labour and agricultural offices.
A major problem we encountered was the widespread ambivalence towards migrant labour among government and party cadres. For example, in a public village meeting Ruth and I attended (along with the members of one of our brigades), various men stood up to voice concern about the drop in recruitment or their inability to get another contract, to which the chair of the grupo dinamizador responded by depicting mine labour in direct opposition to constructing socialism. After the meeting, the chairperson invited us to his home for tea. He was very open and honest in his discussions with us, highly committed to Frelimo's policies and clearly puzzled why we were doing research on mine labour since this was not exactly a political priority. Yet, later on in our discussion, he framed his own dilemma quite vividly when he said that ‘others can go to the mines and earn money, but I have to stay behind and build socialism.’
The brief of the brigades was to explore the interdependencies between the inflow of remittances from wage labour, the large-scale withdrawal of male labour from agriculture, and the dynamics of rural production and accumulation. The household questionnaire was designed to collect data on household composition, its dependence on wage labour, the level of its agricultural production and other local off-farm activities, and its sources of money income. The fieldwork was not organised around any inflexible prior hypothesis about the nature of these interrelations. We were, of course, familiar with the analytical literature on migrant labour by South African Marxist scholars who argued that – as Ruth later would phrase it – ‘the access of the migrant labourer and his family to domestic production provides part of the means of subsistence from which the capitalist sector benefits, and the means, thus, by which capitalism derives cheap labour power’ (First 1983, 7). This thesis implied that the ‘means of subsistence acquired by a worker are thus divided into two parts: the direct wages paid to him in and during employment, and the indirect wages which he receives in the form of social support derived from family agriculture – that is, care for the women, children and the aged, and care for himself during sickness and in between spells of employment’ (Ibid.). We took this as a guiding premise but not as a rigid hypothesis. For example, we did not take it for granted that this workforce, in the periods of its dispersal to the rural areas from which it had been recruited, was merely or entirely a reserve army of labour (Ibid., 8). Our approach was distinctly more eclectic and, hence, considerable space was left for further hypothesis searching though case studies, thus allowing for diversity in emphasis and approach of research across brigades. This was, of course, a risky strategy, but it also had a greater potential to get richer insights.
The mobile brigade experienced a fascinating adventure: days of travel on rural roads and tracks to visit and discuss with the different brigades about the progress of their work, give suggestions and transmit ideas between brigades, participate in meetings, bring food, carry letters to and fro and, occasionally, transport those who fell ill to town and back. One of the brigades got lost, having stayed at its assigned location (Sitila) for only four or five days and then moved on far into the sparsely populated interior of Inhambane province, stopping at each village for a few days before moving on to the next. Since they moved faster than the mobile brigade, we failed to catch up with them, each time being told that they had moved further on. Only after two weeks did a desperate continuous 10-hour search manage to locate them. By pure accident we happened to stop at the same roadside restaurant (one of very few in the area). They looked happy enough, but were a terrible sight – they had never stopped long enough to wash their clothes, shave or even cook properly. They saw research as covering the widest area in the shortest time and definitely established a record of sorts (some kind of early variant of rapid rural appraisal). As a result, they had not made much use of the questionnaires, but relied mainly on participating in village meetings, holding group discussions and, interestingly, visiting primary schools and asking children what their aspirations were for the future. Notwithstanding the major changes that were taking place in Mozambique at time, the boys almost invariably answered that they would go to the mines.
Fortunately, the other brigades stayed put and carried out their research in their assigned locations. In our earlier seminar discussions, we had thought that a good starting point would be for brigades to distinguish between peasant households that had a history of mine labour and those that had not, and then find out what differentiated them. In practice, it took most brigades only a day or two to discard this idea. It quickly became clear that all able-bodied men (with the exception of teachers, shopkeepers or state functionaries) had a history of mine labour. From then onwards, the focus and approach adopted by each of these brigades diverged quite markedly. For example, the brigade in Pembe zoomed in on the question of wage labour and peasant differentiation in agricultural production. They noted that, while most able-bodied men had a history of mine labour, some did so throughout the whole of their working lives while others relied on mine labour only at the beginning of their working life, using wage remittances to invest in agriculture, trade or local crafts (carpentry, tailoring etc.). In contrast, the brigade in Homoine took a more historical-anthropological approach and focused in particular on the changing historical relation between lobolo (bridewealth), the colonial hut tax and the level of money wages in the mines. Other brigades pursued different angles, but a common theme that emerged across all the brigades was the wage-dependent nature of agricultural production and accumulation. For a young man to establish himself in agriculture he first needed to pay lobolo, financed from mine wages, to get married. Remittances from mine wages were not used just for consumption, but also to acquire land, to pay for bricks or roofing to build a house or water reservoir, to buy a plough and oxen (for own use and for sale as a service to plough the land of others) or to buy tools for various crafts like tailoring, carpentry, brickmaking and so on.
The end phase of the project combined data analysis, writing up and editing, done in a mad rush towards the finishing line because Ruth was due to return to England in early November. The task of bringing together the disparate contributions of different authors with different traditions and very varied levels of expertise and experience into a coherent and quality piece of work would have been an enormous challenge, even under more favourable time constraints. Ruth did all the coordination and the bulk of the editing, as well as writing a considerable part of the report herself. Her flat turned into an editing house: drafts, books and papers scattered all over the table or floor, half-written chapters piled up next to the typewriter, and a huge map with the names of all involved with specifications of who had what to be handed in when to whom. In between spells of writing or editing, Ruth would rush all over town to collect drafts from authors and take them for translation (when required) and collect scripts from typists (mostly volunteers). This phase left a few bruises on the way because the looming deadline allowed for little reference back to authors. Moreover, Ruth knew that quality output requires firm editing and, undoubtedly, she was an excellent, but also ruthless, editor. It would not take long before a draft she was reading would be rendered unrecognisable, cut up into multiple snippets of paper scattered all over the table, which she then managed to assemble back into a coherent collage with handwritten text in between. The original intention had been that the weekly seminars would continue after fieldwork, allowing for transparent feedback on edited texts and work in progress. But the rush to produce the output in time went at the expense of the process of learning by writing research. It was only a few days before Ruth left that the final draft materialised. Printing took a bit longer and, hence, Ruth received the first English edition of The Mozambican Miner only after she had returned to the University of Durham in November.
Critical engagement with policy
The final product was a rich mosaic of arguments, data tables, case studies, life stories, detailed interviews and songs (CEA 1977b; see also First 1983). This was very much a reflection of Ruth's style of writing. It also reflected her attitude to dealing with theory in writing up concrete research. She read widely and was deeply interested in theory, always questioning, never taking anything at face value. In the seminar, during rural fieldwork, when looking at data and during the writing up, she would always start from or come back to theory – about the nature of migrant labour systems, about rural differentiation or about socialist transition. In her writing, however, she never put theory up front, but it always played in the background. As an investigative journalist she had a nose for getting at stories and insisted that we should keep our writing as concrete as possible. For her, theory was not for display, but a vehicle to bring a story to life. In South Africa, she had done this by putting concrete analysis to work in opposition to the apartheid regime. Mozambique was different. She was committed to Frelimo's project of socialist transformation, but this did not mean that she lost her critical perspective. On the contrary, she firmly believed that engaged but critical academic research – grounded in theory, but concrete in its analysis – was a necessity, although she knew that the space for this type of research was virtually non-existent when she arrived in Mozambique. For her, The Mozambican Miner was the vehicle to open up this space, cautiously but firmly. At the time, I do not think any of us, including Ruth, were fully aware of how critical The Mozambican Miner would turn out to be in terms of questioning Frelimo's policies, not just on matters of employment and migrant labour, but also on agriculture and on macroeconomic development. The different strands of this critique were clearly visible in the final output, but all the knots were not yet tied together into a cohesive argument.
In terms of employment, we could foresee major problems emerging in the rural areas of southern Mozambique because of the severe cut in recruitment of mine labour and its restriction to experienced workers with a valid re-employment certificate (issued by the last mine they worked in and valid for a limited period of time). The immediate effect was that young men and older workers without valid certificates no longer had access to mine labour, but the full impact of this would only be felt when ongoing contracts on the mines ran out. The average length of a mine contract was about 15 months, which meant that southern Mozambique was confronted with an imminent unemployment crisis. Moreover, our research had shown the extent to which the rural economy of the South depended on inflow from mine labour, not only for consumption but also for peasant production, whether for own consumption or for sale. This led us to discard the hypothesis that the social reproduction of the labourer and his family depended simply on income from wage labour supplemented by subsistence production in agriculture. Instead, we argued that the inflow of cash from remittances fuelled agricultural production and accumulation, resulting also in wage-dependent processes of peasant differentiation. In other words, we viewed the relation between wage labour and agricultural production as interactive, not additive.
The implication was that changes in the dynamics of mine labour recruitment deeply affected the viability and vulnerability of peasant agriculture in the South. In the early post-independence years this vulnerability was hidden since recruitment was high and wages on the mines had been increasing since 1973. The inflow of money into the rural economy rose substantially during this period, which also led to an increase in household investment in agriculture. The peasant economy boomed, but this dynamic was halted and reversed in 1976 when the Chamber of Mines in South Africa dramatically cut the recruitment of mine labour from Mozambique. The result was not only growing unemployment, but also the gradual weakening of rural production and accumulation, thus fuelling the need to look for cash income from other wage employment, compounding the problem of unemployment. Obviously, these processes took time to work themselves out and their consequences were not immediately visible. Agrarian policies did little to respond to this problem, in part because of the strongly held assumption that the peasantry in Mozambique mainly relied on subsistence production and, hence, could fend for itself until peasant labour was gradually absorbed within the expanding socialist sector. Implicit in our argument, however, was that agrarian policies should address these vulnerabilities of the peasantry rather than ignore them, by spreading investible resources (including channelling them to peasant agriculture) rather than concentrating them nearly exclusively on state farms.
The implications of our research related not only to employment and agrarian development in southern Mozambique. There were also wider macroeconomic consequences. In 1977, the South African government abandoned its policy of keeping the official gold price fixed (at the time, at a level well below the world market price) and allowed it to fluctuate in line with the world market price. As Ruth was quick to recognise, this was not a simple monetary measure, but had grave consequences for Mozambique. She asked me to investigate how this change was likely to impact on Mozambique. By triangulating different sets of data on levels of employment and the estimated average length of contracts, along with information on the evolution of the official and market prices of gold derived from the Financial Mail, I made an estimate of Mozambican foreign exchange earnings from mine labour in the immediate post-independence years, which turned out to be about one and a half times the level of all other exports taken together. In other words, foreign exchange earnings from deferred mine labour payments had kept Mozambique afloat during the immediate post-independence period. This explained the strange anomaly that, while Mozambique witnessed a major production crisis in this period, it did not have a balance of payments crisis. On the contrary, Mozambique was building up foreign exchange reserves during this period. The rapid decline in mine employment and the loss of the premium on gold (the difference between its market and official price) were about to reverse this condition. Projecting foreign earnings forwards showed Mozambique heading for a catastrophic foreign exchange crisis that was likely to destabilise its already fragile economy.
According to our own projections, the financial crisis was likely to take effect from 1979 onwards. I raised this issue with Joaquim Brum, the National Director in the Planning Commission, who replied that income from migrant labour was no more than ‘a drop in the bucket’ and, hence, insignificant and irrelevant to Mozambique's planning concerns. It is possible that Brum was genuinely unaware how much foreign exchange Mozambique earned from migrant labour. Indeed, Mozambique's gold reserves from migrant labour were stored and managed in Pretoria on instruction from the Governor of the Banco de Moçambique (Mozambique's national bank). Alternatively, he may have been counting on large-scale development assistance from the Soviet Union to fund Mozambique's development efforts. For most of the period of central planning, the interrelations between real and financial accumulation were poorly understood, and the importance of money and finance largely ignored, with money and credit made to adjust passively and endogenously to the dictates of planning. Aquino and I (Ruth had left Mozambique by then) went to see the Governor. He was far more receptive to our argument and confirmed that our guesstimates were surprisingly accurate (to the point that he was worried that he might be accused of having passed on this secret information to the CEA).
The notion that, by putting together and triangulating different pieces of data, you can actually arrive at a reasonable estimate of some unknown ‘third’ quantity was not something readily accepted in Mozambique at the time, nor was it part of the usual practices of data analysis, which amounted to discussing bits of data one by one. The Governor was genuinely worried about the inferences we had made about an impending financial crisis, but failed to convince the Planning Commission of its importance. As it turned out, the foreign exchange gap widened considerably in 1979 when the government embarked on a 10-year plan for an ambitious state-centred investment drive. This led to a severe squeeze in aggregate consumption as the financial crisis deepened during the early 1980s (Wuyts 1981a, 1981b, 1989).
Conclusion: Ruth First and the development course
The idea of a CEA development course emerged during the writing-up of the miners' research. A request came from the Department of History to set up a licentiatura (roughly equivalent to an MA degree) in history, as part of the plans to set up a fully fledged faculty of social sciences. Without Ruth's knowledge, António Nogueira da Costa, one of our team members, had suggested to the history department that Ruth be asked to develop a proposal for such a course and, possibly, direct it afterwards. The fact that they turned to Ruth was not surprising since, by then, she had established quite a reputation for effective organisation and producing results. Ruth stalled, however, not because she considered the request unimportant, but because her own interests were clearly in the political economy of development, not history.
Ruth discussed the issue with Aquino de Bragança, David Wield and me, leading us to submit a counter-proposal to develop a postgraduate course in development. The inspiration for this proposal derived directly from our experience with The Mozambican Miner. The course would be interdisciplinary in scope and focus on issues of socialist transition in Mozambique within the context of southern Africa. It would build its curriculum around the teaching of research by doing collective work. The course would therefore both teach as well as produce research output. This sounded ambitious but it was also realistic, given the circumstances. We realised that the teaching materials on Mozambique presently available were extremely thin due to the absence of a genuine research culture and practice at the University.
These were the ingredients upon which the development course was subsequently constructed. Ruth was appointed as research director of the CEA to start the course in 1979 with an enlarged team of researchers specifically recruited for that purpose. The development course had a unified structure without options or sub-specialisations and involved part-time study (about 12 hours/week during its two semesters and one month full-time fieldwork) aimed at students who had regular jobs in the University or in the different structures of government. Each course was built around a well-targeted collective research project, thus providing a training ground where students and staff could acquire and develop research skills. We then envisaged that the course would grow by building upon itself: each research project would add to the teaching materials for the next cohort, thus enriching the scope and content of the course. In other words, the underlying philosophy was to start small and rely on the compounding effects of successive research projects clustered around a common theme (the transformation of production and the conditions of labour) to deepen the content of the course. This perspective inevitably included an economic dimension, but it was not concerned with research in economics as would be typical within that discipline. In fact, while much attention was given to empirical methods, these were distinctly different from what was normally then taught in standard economics training.
When we started planning for the first intake and began looking for potential students, it soon became clear that the pool of university-trained BA-level candidates was extremely small. Conversely, cadres who might occupy quite important positions in government ministries, banks and so on often had no formal university training but were nevertheless deeply involved in practical policy making, including thinking about policy. The reality of the colonial inheritance meant that access to education had been extremely limited, particularly for black Mozambicans. In response, Ruth decided to open up the course – which was designed as postgraduate level – to people without formal university training but whose day-to-day work confronted them with the need to analyse problems as they emerged. This decision was risky and highly controversial and, not surprisingly, created a great deal of tension around the status of the diploma to be given to students. The Rector agreed reluctantly but wanted a two-level awarding system: awarding a postgraduate degree to those students in the course who had a BA-level degree, while others would only be given a certificate of attendance. Ruth vehemently opposed this solution, arguing that, if students did the same course, they should be treated equally. As experience subsequently showed, it was by no means the case that students with prior BA-level education did better than those without. The issue remained unresolved but all students received a diploma, the status of which, however, was left undefined within the Mozambican educational system. It was thus imperative for the course to prove its worth by establishing a reputation – both locally and internationally – that could serve as a substitute for the lack of formal certification.
For Ruth, full engagement in training Mozambican cadres in research through a process of learning by doing was not a withdrawal from her strong commitment to the struggle for liberation in South Africa, but sprang from her deeply held conviction of the unity of these struggles in the context of southern Africa. More specifically, she saw the development course as a vehicle to consolidate space for training Mozambican researchers to actively engage in concrete analysis and to explore alternatives about questions of transition, even if it meant raising awkward questions. Educating researchers to think independently and critically was, I believe, Ruth's main legacy in Mozambique.
Note on contributor
Marc Wuyts is (Emeritus) Professor in Quantitative Applied Economics of the International Institute of Social Studies, Erasmus University Rotterdam. He worked at the Economics Department of the University of Dar es Salaam from 1972 to 1976, and at the Faculty of Economics and the Centre of African Studies of Eduardo Mondlane University in Maputo. His academic interests concern the macroeconomics of accumulation, growth and economic transformation, and statistical methods of data exploration.
