Cabora Bassa : engineering and politics in Southern Africa

http://www.aluka.org/action/showMetadata?doi=10.5555/AL.SFF.DOCUMENT.crp3b10006

Use of the Aluka digital library is subject to Aluka’s Terms and Conditions, available at http://www.aluka.org/page/about/termsConditions.jsp. By using Aluka, you agree that you have read and will abide by the Terms and Conditions. Among other things, the Terms and Conditions provide that the content in the Aluka digital library is only for personal, non-commercial use by authorized users of Aluka in connection with research, scholarship, and education.

The content in the Aluka digital library is subject to copyright, with the exception of certain governmental works and very old materials that may be in the public domain under applicable law. Permission must be sought from Aluka and/or the applicable copyright holder in connection with any duplication or distribution of these materials where required by applicable law.

Aluka is a not-for-profit initiative dedicated to creating and preserving a digital archive of materials about and from the developing world. For more information about Aluka, please see http://www.aluka.org Cabora Bassa : engineering and politics in Southern Africa

Author/Creator Middlemas, Keith Publisher Weidenfeld and Nicolson (London) Date 1975 Resource type Books Language English Subject Coverage (spatial) Mozambique Source Northwestern University Libraries, Melville J. Herskovits Library of African Studies, 627.80967 M627c Rights By kind permission of Keith Middlemas and Weidenfeld & Nicolson, a division of The Orion Publishing Group. Description This book looks at Mozambique's Cabora Bassa dam "as a lens through which to see southern Africa's political and economic developments as a whole." This dam, in particular, was chosen because it "offered a marvellous range of questions: its status as a development project; the difficulties of engineering, climate, transport, geography, and the problems of men working in harsh conditions in an inhospitable land; its political as weel as economic significance for Mozambique and for southern Africa; and its strategic importance in Portugal's colonial wars." Format extent 379 pages (length/size)

http://www.aluka.org/action/showMetadata?doi=10.5555/AL.SFF.DOCUMENT.crp3b10006

http://www.aluka.org P f rica na

P f rica na 627ý?0967 f.'ý627c LLS

Cabora Bassa

Other books by KEITH MIDDLEMAS: The Master Builder The Clydesiders Baldwin (with A.J.L. Barnes) Diplomacy of Illusion Editor of Thomas Jones' Whitehall Diary 3 vols FRONTISPIECE The gorge of the Zambezi in 1965 before work began.

Cabora Bassa Engineering and politics in Southern Africa Keith Middlemas WEIDENFELD AND NICOLSON LONDON

)/ ,2, Keith Middlemas 1975 Weidenfeld and Nicolson 11 St John's Hill London SW 11 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the copyright owner. ISBN 0 297 76994 4 Printed by Tinling (1973) Limited, Prescot, Merseyside (a member of the Oxley Printing Group Ltd)

No Mar tanta tormenta e tanto dano Tantos zvezes a morte apercebida! N a terra tanta guerra, tanto engano Tanta necessidade avorrecida! By sea such storms and such disaster So many times about to die! On land such wars and such betrayal And such malevolent fate! LUIS DE CAMOES, Os Lusiades, Canto I, 106. 'Thus there are innumerable intersecting forces, an infinite series of parallelograms of forces which give rise to one result - the historical event. This may again itself be viewed as the product of a power which works as a whole unconsciously and without volition. For what each individual wills is obstructed by everyone else, and what emerges is something that no one willed.' ENGELS to J. Bloch, in K6nigsberg, 21 September 1890. To our son, Hugo

Contents Introduction 1 Part 1 1 Mais cedo o mais tarde 9 2 Building a consortium 41 3 Contractors' politics 66 Part 2 4 Preparation 89 5 Building 107 6 Revolt in Mozambique 131 7 The propaganda war 160 8 Completion 186 Part 3 9 Development or exploitation? 209 10 Towards economic autonomy 235 11 Que d Moqambique? 259 Part 4 12 Mozambique and Southern Africa 281 13 The Portuguese crisis 308 14 Sequel 335 Index 359

Illustrations FRONTISPIECE The gorge of the Zambezi in 1965 before work began (Ministerio do Ultramar) 1 & 2 Architect's impressions of the dam at various stages (G. R. Campbell, ARIBA) 3 The gorge looking upstream with the coffer dams and outlet of south diversion tunnel (Senhor Carlos Alberto Vieira, Lourenco Marques) 4 Intakes for the pen-stocks, leading down to the powerhouse, partly completed (Senhor Carlos Alberto Vieira) 5 Main wall of the dam looking downstream. Summer 1973 (Senhor Carlos Alberto Vieira) 6 Aerial view of the site, October 1973 (Senhor Carlos Alberto Vieira) 7 The wall and the sluice gates half completed (Senhor Carlos Alberto Vieira) 8 Heavy plant in use in the surge chambers (Senhor Carlos Alberto Vieira) 9 Main chamber of the powerhouse with the first generating sets in place (Senhor Carlos Alberto Vieira) 10 Beam crane lowering one of the sets into position (Senhor Carlos Alberto Vieira) 11 Converter sub-station on the plateau above the gorge with banks of thyristors on insulated platforms (Senhor Carlos Alberto Vieira) 12 Upstream view of the highest flood, April 1974 (Senhor Carlos Alberto Vieira) 13 Looking downstream, August 1974 (Senhor Carlos Alberto Vieira) x List of Illustrations 14 Aerial view of the site, August 1974 (Senhor Carlos Alberto Vieira) 15 Samora Machel, President of Frelimo, with other Frelimo members (Popperfoto) Maps and Diagrams Province of TFete 88 Mozambique 92 Cabora Bassa, dam and reservoir 121 Cabora Bassa plant 122 Southern Africa 226

Introduction Great public works are rarely conceived in simple terms. Underlying them are layers of political and strategic motives, compromise, and international horse- trading. The dams of Africa are no exception - Aswan, Kariba, Volta, Inge. This book began as an attempt to describe the origins and construction of one major project, treating it not simply as a matter of engineering, human endeavour, or economic benefit, but rather outlining its total impact. The Cabora Bassa dam offered a marvellous range of questions: its status as a development project; the difficulties of engineering, climate, transport, geography, and the problems of men working in harsh conditions in an inhospitable land; its political as well as economic significance for Mozambique and for southern Africa; and its strategic importance in Portugal's colonial wars. Thus it was possible to ask not only why the dam was built, and whether it was profitable, but for whom, and in whose real interest. Was it a form of development, aid, or exploitation? Was it, as the chairman of the construction company claimed, a perfect example of capitalist enterprise in a technically free situation, or was it undertaken in artificial circumstances similar to those governing construction of the Tanzam Railway or the Aswan High Dam? Cabora Bassa also afforded an opportunity to study the operations of at least one multinational company in close detail, and a means to assess the weight of its leverage in developing African countries. In foresaking the one-dimensional approach, characteristic of cost benefit analysis or of the meticulous series of sociological and anthropological Kariba Studies, an attempt is

CABORA BASSA made here to use Cabora Bassa as a lens through which to see southern Africa's political and economic developments as a whole. This way it is hoped to avoid the distortion inevitable when starting with generalities. The method has its own obvious distortions, but may possibly be less constrained by ideological assumptions since it starts, literally, with the dam itself, and works outwards in an investigation of how the massive investment it entailed affected the affairs of Mozambique and its neighbouring states. The result may be of some value in testing existing models of development and political relationships. During the four years after this book was begun, in 1970, change in Mozambique cut across long-established patterns in southern Africa, helped to unpin the shaky structure of the Portuguese empire, and profoundly affected the cohesion of South Africa and Rhodesia as well as that of Zambia, Tanzania and Malawi. As Frelimo (Frente para a Liberaqdo da Mocambique) waged an increasingly successful war, the conflict in Mozambique highlighted Russian and Chinese involvement in African affairs, and the concern of the Western powers for their strategic interests in the Indian Ocean. Finally, in 1974, the Portuguese revolution, and Mozambique's subsequent independence, created a strange situation for the author - who some years earlier had watched the British empire in Africa decline - as the hypothesis of his text became established fact. Until recently, to observers outside the Portuguese-speaking community, Mozambique was an area of darkness. Much of this book, therefore, charts the course of the transition of Cabora Bassa and the imperial development scheme of which it was a part, from the dreams of a handful of civil servants in the Overseas Ministry to its completion in the face of war and final disintegration, as Portugal forcibly discovered its true twentieth-century dimensions. At the same time it was possible to analyze, for example, Portugal's relations with South Africa, which took the major share in financing the scheme, and the reasons behind the colonial policy of the Caetano Government. The history of Frelimo has been sketched, not only because of its fundamental importance to the story, but to refute the muddled thinking which so often 'identifies

Introduction 3 nationalism with socialism, the peasantry with the proletariat, anti-imperialism with anti-capitalism',, and to suggest the existence of factors common to many African liberation movements. Two caveats should be made: this is not a history of the last days of the Portuguese empire, and says little about Angola and Guinea-Bissau. Mozambique's development was singular and different from that of Angola (though sometimes lumped together with it in studies of the colonial period), and it had a quite specific effect on the Portuguese revolution of April 1974. Second, the speed of change during 1974 has been so great that the work of the contemporary historian becomes very hard. Inevitably, greater distortions, and greater chances of serious error, threaten than in the earlier period, for sources could not be checked and evaluated to the same degree. Nevertheless, it would have been absurd not to conclude with the signing of the Lusaka agreement on 7 September 1974. The book set out, deliberately, to conduct an experiment in the uses of oral history and contemporary documentation in order to describe a changing situation at several different levels, over ten years. It was necessary to obtain information before it disappeared irrevocably. Once the various civil and mechanical engineering groups had finished, and after the consortia had been dissolved, few records of the project or the prior negotiations would have been kept. Memory among those concerned could be distorted by drama, nostalgia, or plain exaggeration; and the sheer geographical difficulties of resuscitating the whole, after ten or twenty years, from remaining records in New York, Paris, London, Nuremberg, Milan, Lisbon, Johannesburg, and Mozambique, need no emphasis. In addition, government archives are often inaccessible or found to have been edited. During this period no Portuguese official records later than the 1920s were available, and the absence of a sound historiography of the Salazar period compounded the difficulty. Apart from official sources, which often presented a misleading or confused picture, a great deal of the available documentary material cited in the bibliography offers only secondary or politically biased information. Consequently it was necessary to proceed very largely by

CABORA BASSA interviews, building up a framework over a period to evaluate information. Gresham's Law operates vividly in oral history: dramatic news drives out sound analysis. Fortunately, people were keen to talk about a project as controversial as Cabora Bassa, either in defence or in criticism; and since the dictates of secrecy rarely covered events in the recent past, a continuous narrative could be constructed. This core of fact and sound interpretation served as a test of the flood of disingenuous propaganda which itself, at its worst, nevertheless contained valuable insights. The contemporary historian has two advantages over the journalist; time to watch processes, and analytical skills. At his best, he can start off the process by which events become history, construct, in Oakeshott's phrase, 'the practical past' and offer it as a contribution to future debate. More than three hundred interviews took place, of roughly a hundred separate, carefully chosen people, in the period 1970-4; and while the future will show up many inaccuracies and gaps, it may perhaps not undermine this thesis. A serious difficulty, however, arises over the use of sources. In some cases to obtain information undertakings of anonymity were given, and in others it would be a breach of faith and courtesy to name names. Some of the people interviewed would undoubtedly be endangered if they were cited, others - and their work - gravely compromised. Government officials and diplomats might have to repudiate their statements; any evaluation in public of their truthfulness would be open to the law of libel. Eventually it was decided that no statements should be cited, except in the case of public speeches. Beyond this, the author can only affirm that every effort to check them has been made. A list of sources, by general categories, is given in the bibliography. All documents, except where shown under a pledge of secrecy, are cited in full. I should like to express my deep gratitude to all those who talked to me and answered my exhaustive lists of questions. In particular I should like to acknowledge the assistance of the Portuguese Government, and the Overseas Ministry, in facilitating my visits to Cabora Bassa, and in affording me wide scope for my inquiries without seeking to make any condi-

Introduction 5 tions, other than that I tell the truth as I saw it. This I have tried to do. Without generous grants from the Nuffield Foundation and the University of Sussex towards the expenses of travel in Europe and Africa it would have been impossible to undertake the programme, and I am extremely grateful for their assistance, and for leave of absence given by the University. My thanks are due especially to Dr Guy Routh and Senhor Sergio Fernandes who read the manuscript and made many helpful criticisms; and also to Mrs Sylvia Burton who reduced my manuscript to order. Finally, lest it be thought that I approached the subject wholly in the abstract, I should make clear that my starting point was always the project itself. The Cabora Bassa Dam is a marvellous entity, beautiful, mechanically and scientifically, and set in a landscape of amazing grandeur. It has been an exciting, sometimes thrilling, experience to have witnessed it grow, to have known its builders, and those who will benefit from it.

Part 1

Chapter 1 Mais cedo o mais tarde David Livingstone was the first white man to encounter the gorge of Cabora Bassa, although it had long been known to Portuguese traders. There the Zambesi, diverted by a primeval shudder of the earth's crust from its placid course across the central African plateau, enters a tangled mass of mountains and bursts its way roughly through to the coastal plain. On his first, most famous, journey across Africa in 1853-6, Livingstone discovered the Victoria Falls and followed the great river to the coast. But at the edge of the plateau he crossed to the south bank and avoided the mountain ridges ahead, only rejoining the Zambesi at the small Portuguese fort of Tete. Thus he missed the rapids and the changed level of the land and heard nothing from his native bearers to diminish his faith in the Zambesi as the 'excellent highway' for trade with the salubrious highlands of the Barotse. Livingstone reported his findings to Sir Roderick Murchison, President of the Royal Geographical Society, Lord Shaftesbury and Queen Victoria. In May 1858, bearing the title of Her Majesty's Consul for Quelimane and financed and supplied largely at government expense, he arrived at the mouth of what he called 'God's highway into the interior'. Early optimism survived intractable mangrove belts and disasters to his ill-built launch, but vanished when, forty miles above Tete, he met the first signs of what lay between him and the highlands. Livingstone had hoped to be able to blast away the rocks which obstructed the passage but on 15 November he had to write instead:

CABORA BASSA This Kebrabasa* is what I never expected. The only person who ever saw the river above where we did, was Jose Santos Anna [their guide], and he described it as fearful when in flood. The scenery is quite remarkable and totally unlike anything that has ever been said of the rapids. A series of lofty hills, among which those of a conical shape prevail most, enclose a dale resembling a river bed. This dale is the bed of the river at high water and it is a strange mass of huge rocks containing a deep chasm winding from side to side along it. This fissure or cleft is at first not more than twenty feet deep but it soon becomes eighty or even one hundred and twenty feet.. . . The rocks where the water flows for any time are all covered with a black glaze. Everywhere they present evidence of wear. The granite is often fluted perpendicularly ... great pot holes are worn out in other parts. They are grooved in every direction besides and present evidence of tremendous wearing force being exerted on them. The rocks themselves are chiefly igneous and they have been heaved up, contorted, bent, jammed and fissured in every direction. Great round amorphic masses interchange with huge fragments presenting lines of doubtful cleavage and various stratification. Pieces of rock from mere pebbles to large boulders have been driven into chinks by the sheer force of the rushing waters so firmly that they break rather than move. The gorge narrows and heightens rapidly after this and it was perhaps as well that no one in London took seriously Livingstone's assertion that a good steamer capable of fifteen knots could surmount the cataracts when swamped by the full flood. Where the river passes what the explorer named Mount Stephanie, after the Portuguese Queen, it has a normal rise of sixty feet and in some flood years well over eighty. Then 10,000 cubic metres of water a second sweep by at a steady ten knots, bearing full-grown trees like straws and turning fifty-ton boulders into battering rams. This is not a mountain torrent but the flood water of Central Africa, brown and yellow, turbulent on a massive scale. * Meaning, in the local dialect, 'an end of work' - that is, the point at which river navigation became impossible.

Mais cedo o mais tarde 11 For ninety-eight years, until the building of Kariba, no white man tried to pass through 'Kebrabasa', and when the Portuguese engineer Abecassis Manzanares surveyed it in 1956, he echoed what Livingstone had said: It was in truth the worst tract I ever travelled over. Our strong, new English boots were worn through to the soles. The sun's rays were converted by the converging hills into a focus and the stones were so hot that the hand could not be held on them for a moment, though we were in danger of being dashed down into the gorge by letting go for an instant. The reflection from the rocks felt exactly like the breath of a furnace; I felt sure that if I had come down this way in 1856 I should have perished before reaching Tete: for now, with but a fortnight's experience and an examination of a mere thirty miles, we all returned as lean and haggard as if we had been recovering from a serious illness.' In that century the gorge had been by-passed. The district of Tete is one of the poorest and sparsest areas of Africa, arid in summer, inundated in the rainy season, inhabited by tribes whose agriculture is traditionally based on the cycle of burning scrub, scratch cultivating for a few years, and moving on. Five centuries of Portuguese administration had barely touched the country beyond the small district capital. The dusty unmetalled roads radiating west and north from Tete towards the broad upland valley of the Zambesi and the villages on the Rhodesian and Zambian borders avoided the mountain ranges. Tete itself, richer but not much larger than in the 1850s, sprawled along the river bank, a random collection of stone-built houses for administrators, with the old fort, the Church of St Joseph and the Governor's mansion breaking the low roofline. A ferry of small launches connected it with the north bank where the road ran straight to the settlement of Moatize, rail-head for the line down to Beira and site of a Belgian company which had been mining open-cast coal since the 1920s. From Moatize the only reasonable road ran to Zobue on the Nyasaland (Malawi) border and thus connected Salisbury to Blantyre. Otherwise, Tete province consisted of tiny administrative postos - Zumbo, Fingo6, or

12 CABORA BASSA Vila Vasco da Gama - set among native villages in the middle of an enormous empty hinterland. In 1958 Tete could be reached either by plane, from Beira to the narrow airstrip at Moatize, or by road via Vila Pery and thence north, hoping that the bridges were intact, to meet the main Salisbury-Tete road at Changara. If time were unimportant, the journey could be made by train over the mile-long bridge across the Zambesi, through the rich sugar plantations of Sena and Dona Ana - relics of the nineteenth-century feudal estates or prazos - past where the Shire River flows in from Lake Nyasa, and the curious domed mountains of Nyasaland. Beyond Tete there was a choice of light aircraft or a sixty-mile drive in a cloud of fine dust to Estima, where steep conical mountains force the road southwards towards the blue ridges along the Rhodesian border. In the dry season the plains are utterly brown and parched; from December through to May they are luxuriantly green. The bush is open, the soil sandy and full of shale. Here is the home of the baobab, immensely longlived trees which hoard water in their fibrous trunks and proliferate in barren soil. With short stubby branches stuck at odd angles to an enormous trunk, they seem like childrens' drawings, livelier giants for Don Quixote than Spanish windmills. Occasional beehive huts indicate a native smallholding or shamba. There are no villages, but every dozen miles a seedy cantina offers excellent Mozambique beer, a plate of dried fish and coarse local bread, harsh coffee and perhaps a bed for the night. To reach the Songo plateau from Tete, the road breaks off at Estima and ascends fifteen hundred feet, up a dozen hairpin bends, to a natural fortress of rocks with ramparts almost too precipitous for trees to grow on: a broad plateau, some three miles by six, enclosed by small rocky hills. Four or five centuries ago, when the early Portuguese traders were making treaties with chiefs of the great central African empire called Monomotapa, the Songo plateau was inhabited. At one end is a flat outcrop of granite, crudely but unmistakably walled in stone. The primitive Nyungwe tribesmen, whose huts are built of wattle and mud, call it simply 'the mound of the stone people' - a little Zimbabwe. Songo has no other history. Until the arrival of the bulldozers there was no way down

Mais cedo o mais tarde 13 to the gorge except on foot, following a stream which rises on the plateau through a long, densely wooded valley carved out by erosion. At its end, the stream drops over the edge of the gorge, opposite the summit of Mount Stephanie and the main line of the mountain ridge, a wall of precipitous cliffs of greyish-red granite. Between, there is a void: at the bottom, so far below that the whirlpools of the current cannot be seen, the river reflects the sun. The far side is unscalable except by mountaineers; on the southern wall, overhung- by sheer cliffs, a boulder- strewn scree, propped up by outcrops of granite, gives enough lodging for trees and a vertiginous slope down two thousand feet to the flood water-line. There the Zambesi has no bank, only the smooth curve of rocks, dark with algae. Three miles up-stream the gorge opens to the upland valley; down-stream, two headlands of naked rock, six hundred feet high, project from the highest mountains. No more perfect site for a dam could have been conceived. Plans had been prepared by Portuguese administrators even before the Second World War to exploit the flow of the river and to mine coal, iron and copper; but, lacking finance, they remained pigeonholed in the Overseas Ministry in Lisbon. Thus, in the early 1950s, to imagine an immense hydroelectric scheme in the Cabora Bassa gorge needed more than faith. Why design, in such a desolate, inhospitable and remote area of the continent, a dam not for irrigation but to produce energy, large enough to rival the greatest public works in Russia, Europe and America, when the fertile parts of Mozambique were already served by the Zambesi, the Limpopo and the Save? No consumer of so much electricity existed in the southern hemisphere. Mozambique, known, if at all, to the Western world only as a thousand-mile stretch of flat, fever-ridden coastline, broken by the ports of Lourenqo Marques and Beira, was a forgotten corner of Africa. Decolonization might be in the air south of the Sahara (chiefly in West Africa, where in 1951 Kwame Nkrumah's Convention People's Party gained a major part in the administration of the Gold Coast) but the only industrial nation, South Africa, was self-sufficient in electric power. Lines of later change and conflict were still hidden: Portugal had not needed the support

14 CABORA BASSA of the white South to defend her overseas territories since 1938 - when Neville Chamberlain's plan to recast the administration of the whole of Central Africa in order to appease Germany's re-awakened demand for colonies had been contemptuously ignored by Hitler - and under the rigid colonial economic regime Mozambique was closed to foreign investment. By the end of the Second World War, Antonio de Oliveira Salazar, President of the Council of Ministers in Lisbon, had been dictator of Portugal for seventeen years and the Estado Novo (New State) seemed at its apogee. Kept shrewdly neutral through the war, Portugal, like Spain, retained the lineaments of Fascism. Unlike Spain it was not a military dictatorship. While the cautious budgetary policy of Salazar, former professor of economics and finance minister, had restricted the growth of the African territories, it had stabilized the escudo and enhanced the economic strength of the Portuguese dominions as a whole. Thus the chaotic inflation of the 1920s which had created the conditions for Salazar's rise to autocratic power appeared no more than a memory - albeit a memory of political value, like the German inflation of 1923, a terrible example held out as a warning against the laxity of renewed democratic rule. The African territories, Angola and Mozambique, shared some of this post-war security and prosperity and there was little understanding in Lourenqo Marques or Luanda of how badly their performance compared to British colonies in southern and eastern Africa. In 1953, when the Overseas Ministry drafted the first six-year plan, they were secure and well-off in comparison with the standards of the 1930s. In contrast to the debilitated governments before Salazar, Portugal, as a member of Nato, could count on the acquiescence of her allies in her African rule, even if only the most sanguine imagined they would come to her aid against African insurgents. The results of Salazar's policy in Europe encouraged confidence about the future of Portugal in Africa. In the 1950s a modest social and economic revolution took place in Angola and, somewhat later and less effectively, in Mozambique (see Chapters 9 and 11). For the first time the Lisbon Government laid down comprehensive plans of development. A wave of

Mais cedo o mais tarde 15 emigration from Portugal took place, inspired partly by the appearance of opportunity abroad and partly by unemployment and poverty at home; and a recognizably Portuguese society began to develop in the main centres, Luanda, Lobito, Beira, and Lourenqo Marques. In these last years before the revolt of the African population the empire experienced a flowering of prosperity, all the more surprising for its belated appearance in what the British, Dutch, French, and Belgians, had already resigned themselves to seeing as the era of independence. Much of that expansion was directed from Lisbon in favour of a tiny minority of white settlers, even while it conformed to the paternalistic ideal expressed by Morais Cabral in 1939: 'Our whole policy has been, and continues to be, to raise the cultural, economic and social level of the Negro, to give him opportunities and to drag him from his ignorance and backwardness to try to make of him a rational and honourable individual worthy of the Lusitanian community.'" Salazar's Government had evolved a theory of development which, in Portuguese eyes, justified this attitude to an increasingly hostile world. Two years after the independence of Ghana, Salazar replied to criticisms that the colonies were not being prepared for self-government by saying: 'There is a work of human understanding and sympathy which from generation to generation builds up inter- racial contact that is invaluable. This is the basis for the solution of the problems of Africa, for without it they can have no lasting solution. This is more than our conviction; it is our way of being.'4 In November 1960, only four months before the first major uprising in Angola, he told the National Assembly in Lisbon: 'I do not see that we can rest in our labours nor can we have any other care than to hold with one hand our plough and with the other our sword ... Great sacrifice will be called for, as well as the most absolute devotion and, if necessary, also the blood from our veins.'5 Cabora Bassa, therefore, had its origins in Portugal's unique colonial theory. But it did not at first figure as a hydro-electric scheme. The drafting of the six-year plan indicated a change towards expansion, but the document itself was cautious and conservative, a schedule of limited priorities rather than a

CABORA BASSA blueprint for development. It allocated a mere $85 million to Mozambique and $100 million to Angola, and these funds were restricted to the provincial infra- structure - railways and road building, port development and the facilities needed for settling a limited number of white farmers on the land. Even the second plan, covering the years 1959-64, gave only $125 million to Mozambique. The sums were raised half from the current and future surplus in the provincial budget and half by a series of Treasury loans, backed in 1951 by a major credit of $455 million provided by the Economic Cooperation Administration. The Treasury remained reluctant to apportion more loans than could be serviced out of future revenue while supporting Salazar's policy of restricting the access of foreign capital (which, in former days, had financed all the big public works the Benguella Railway, the Delagoa Bay Railway, and the ports of Beira and Lourenqo Marques). Exclusive and nationalist in their nature, these overseas schemes were put into effect only after prolonged study of the colonies' assets and the risks involved. To outsiders in Europe or South Africa, seeking outlets for investment and sources of raw materials, the process seemed dilatory and the opportunities unattractive. Fearing to be hedged in by the restrictions of an alien bureaucracy, they took their capital elsewhere, to the mining areas of what was then Northern Rhodesia and the Belgian Congo. The view that the rapids of Cabora Bassa represented only a hindrance to the navigation of the Zambesi, as impenetrable as the cataracts of the Nile, was abandoned in the mid- 1950s when at Kariba, four hundred miles up-river, the Central African Federation projected a dam across the Zambesi to provide electricity for the newly formed Central African Power Corporation. The Rhodesias were, of course, infinitely richer and more developed than Mozambique and had easier access to World Bank money. But once the Italian consortium began to raise the coffer dams at Kariba, Portuguese attitudes to the river and the great mineral wealth of the gneiss and granite mountains of Tete changed. The primeval contortions of the earth's crust had dealt them a remarkable hand and in 1956 they began to play it. At Cabora Bassa the Zambesi,- carrying the waters of the

Mais cedo o mais tarde 17 Kafue, the Luangwa, and a dozen smaller rivers, is over twice its size at Kariba. The height of the gorge would have allowed for a dam a thousand feet high and a lake reaching far back into the I Rhodesias, to the foot of Kariba itself. So dramatic a plan was never seriously considered, if only because it would have flooded a vast area of fertile land beyond the Portuguese boundary. But from the moment in March 1956 when the Overseas linistrv instructed its engineers in Mozambique to inspect the site, they assumed that if a dam were built it would be as large as international obligations allowed: five hundred feet high, with a lake stretching up-river one hundred and sixty miles to the point at Zumbo where Mozambique and the two Rhodesias met. In May, after the annual flood had subsided, Professor A.A. Manzanares was sent out from Lisbon to investigate the gorge, which could then only be reached by helicopter or by the most arduous climb. He found what he described enthusiastically as / one of the world's finest sites for a hydro-electric project. The Ministry's report put it in visionary language: The basin of the Zambesi in Portuguese territory contains more economic possibilities for the future than any other river in Africa or even in the rest of the world. The study of a scheme for its development is not a matter of years, nor even of decades; but we must appreciate that in the Mozambique basin the potential energy of the river is roughly 50 billion KWH [50,000 megawatts] of which more than half can be achieved in a relatively short space.6 Energy formed only part of the cornucopia. The navigation of the river would be opened up to the Central African Federation and, if the gates at Kariba were operated in collaboration, the floods which for centuries had prevented efficient exploitation of the sugar estates of the lower Zambesi basin could be regulated. The floods, when the works are built, will become a memory, a spectre from past nightmares; and the lowlands formed over billions of years by the alluvial silt from Central Africa, product of primeval erosion, will be turned to productive use by the patience and tenacity of men.7

CABORA BASSA Unfortunately, Portugal's resources were wholly inadequate for it to emulate the Kariba dam, which had been built to supply power to a well established and rapidly growing market including the Ndola copper belt. The total foreseeable demands of Mozambique would not amount to a twentieth of the capacity of Cabora Bassa. On 1956 estimates, the cost of transmission would make its energy unattractive to other neighbouring consumers. Irrigation, flood relief, and navigation, however desirable, could not be expected to amortize the costs of such a scheme. Instead of pigeonholing the report for decades,' however, the Overseas Ministry, following the example of the Tennessee Valley Authority in the United States, decided to set up a department to study the overall development of the .Zambesi basin, relating the resources of Tete province to the economic and social organization of the whole of Mozambique. How seriously this was taken may be gauged by the generous allocation of staff and equipment to the new Missio do Fomento e Povoamento do Zambesi (MFPZ) which was set up in March 1957.° In effect, the upper Zambesi basin, a quarter of all Mozambique, was to be taken out of the sphere of the administration in Lourenqo Marques and run directly from Lisbon; and its development was intended in the long run to symbolize the newly affirmed Portuguese mission in Central Africa. At a time when all other powers were preparing to withdraw from their colonies, leaving to national leaders with World Bank finance the blueprints of vast schemes like the Volta barrage in Ghana, the MFPZ was given a brief like that of a Roman governor to reorganize his colonia in the interests at once of the colonato and the imperial revenue. It is as important to see what was omitted as what was planned. The obvious money-spinners, the fluorite deposits and the copper, titanium, and vanadium concessions, were not immediately rented off as, for example, the diamonds of Angola had been conceded in 1920 to Diamang (de Beers and Union Mini~re) in return for a revenue which, however profitable to Lisbon, has benefited Angola itself very little. From the beginning, it was clear in the Overseas Ministry that the greater part of the cost of development, as of the surveys themselves, must fall on the Portuguese taxpayer who would

Mais cedo o mais tarde 19 benefit, even in the long run, very little. The creation of the MFPZ thus resembles in many ways the unwilling British take-over of Kenya from the British East Africa Company's jurisdiction in 1895, though without the aim of large-scale white settlement. (There is no trace in the early planning stages of plans for colonizing Zambesia or Tete from Portugal; the million white settlers theory derives from a much later incident, see p. 145.) The basin of the Zambesi in Portuguese territory covers an area of 185,000 square kilometres, roughly twice the size of the whole of Portugal, equal to an expanse stretching from London to Edinburgh and from Exeter to Cambridge. From March 1958 the MFPZ was given the help of the engineers and equipment of the Hidro- Technica Portuguesa, a government advisory body with long experience in hydro- electric work in Portugal, Spain, and Africa; and during the next five years the two unrolled a series of surveys, reports, and preliminary plans, charting the characteristics of the region, its social aspects, tribal customs, demographic and religious patterns, as well as tabulating its rainfall, soil structure, and mineral wealth. Eighteen months' intensive work enabled them to publish a Relatorio Preliminar in 1958. This was followed, in 1961, by a massive compilation of all the reports in the Esquema Geral, summarized for the government services by one of the MFPZ staff, in a report which was at once an attempt to win general political support and a statement of policy as it might be expected to develop in other under-developed areas of Portuguese Africa.l° The theme of this study (which received the imprimatur of the Minister, Adriao Moreira, in 1963), and of the final Plan Geral in 1965, was that it appeared possible, by selecting a number of the most beneficial and profitable schemes, to set in train a form of cumulative development covering the whole area - that is, to achieve economic 'take-off not on the basis of industrialization alone, but by a combination of mining, processing ores, agriculture, livestock, fisheries and forestry, at 'an inevitable slow rhythm'." Sustained by a programme of health, education, and aldeamento (collection of the population into villages), and the building of a road and rail system, these schemes would ensure balanced development. The Plan Geral

CABORA BASSA declared the certainty 'that, sooner or later [mais cedo o mais tarde] the economic and social panorama will be transformed'. Briefly, out of 5.5 million hectares in the first survey, 1.5 million were selected for irrigation development and 300,000 for normal agriculture. Along the border with Zambia 2.5 / million hectares were set aside for forestry, mainly hardwoods, and an area was delineated to support at least 250,000 head of cattle. The river was seen as the fundamental axis of transport and was expected to sustain new industries - natural fibres, sugar refining, vegetable oils, timber processing, and cellulose. The principal minerals were found to be coal, iron, copper, fluorite and manganese, vanadium, titanium, magnesite, with smaller deposits of bauxite and chrome. Some would be exported as ore, others processed by hydro-electric power from Cabora Bassa and other dams at Mpanda-Uncua and Boroma, whose total theoretical capacity was envisaged as being over fifty thousand megawatts. As the Plan Geral eagerly pointed out, South Africa, generating 57 per cent of all the energy in Africa, had a theoretical capacity of only 35,000 megawatts. A dam at Cabora Bassa would have an actual capacity calculated at 4000 megawatts (3600 megawatts effective power, allowing for a safety margin of one generator set falling out of action at any time). The dam designed by the Hydro-Technica Portuguesa was a double arch, a curve of extreme elegance and apparent fragility, like a U-shaped section cut from the side of a sphere, whose simplicity and economy was made possible by the narrow dimensions of the gorge and the quality of the granite cliffs at either side. Ultimately, if consumers were found, there would be a power station at each side; in 1968, when this scheme was finalized, its dimensions put Cabora Bassa third in size among the world's hydro-electric dams. But the work of a group of dedicated civil servants, translated now into a millenial project, had to face two major difficulties when presented to the Government in Lisbon in the mid-1960s. The finance required for this type of development scheme was enormous, the return long-term, and based partly on the unproved assumption that agricultural products could become competitive in world markets. And the years since the formation of the MFPZ had seen a sharp

Mais cedo o mais tarde 21 regression of Portugal's standing in the outside world. Portugal was not poor. After nearly thirty years of the Gladstonian finance of the New State, the once despised escudo had become one of the world's more stable currencies. But while Portugal's credit was good among European governments, especially in Germany, that of the overseas provinces was not. Under the rigid system of control exercised over the 'escudo zone', Mozambique could not on its own raise any foreign loan. Investment had to come either from money loaned by Mozambique institutions or as money channelled through Lisbon which, for fear of impairing its own future credit-worthiness, the Portuguese Treasury opposed. Given the already high proportion of taxation spent overseas, and the traditional hostility of Portuguese entrepreneurs to invest in their own African territory*, there was little likelihood of attracting private metropolitan capital; and although an international loan might have been available at the beginning, in the political climate of 1965 there was far less chance that institutions such as the World Bank or Overseas Development Institute would entertain an application on behalf of what the United Nations now defined as a colony.t On the other hand, the great international companies - the Anglo-American Corporation of South Africa, Tanganyika Concessions and the Union Miniere, and governments like that of Japan, eager for raw materials, might be attracted into a reciprocal bargain: mining and other concessions in return for finance for public works. For at least five years, Portuguese industrialists had been trying to circumvent the restrictions on the import of foreign capital at home and overseas. German investment penetration in Angola and the early dominance of the Belgian company, Petrofina, offered grounds for the hope that Cabora Bassa might become a reality within the lifetime of those who planned it; but they also revived the * 'Only fools or careless people, or those who have a passionate love for the colony those whom we call "good colonists" because they bury there everything they make, frequently losing it - dare to use their wealth here in new undertakings. Everyone else takes from the province all that he makes and invests it where he may have the certainty of greater and surer gain without work or worry.' (Simoes Vaz, Problemas do Mozambique p. 25). t For further amplification see Notes at end of chapter.

CABORA BASSA original suspicions which had induced Salazar to close the colonial sphere in the 1930s. If Portugal lost economic control, the empire might dissolve. Against this, the authors of the Plan Geral stated bluntly that the development of all resources in the Zambesi basin, except those reserved in the national interest, should be handled by private enterprise. The Government's part should be restricted to a majority share holding in the enterprise and the creation of the necessary infrastructure. Overseas Ministerial policy took for granted the premise that 'at present it will only be possible to reach a level and rhythm of investment compatible with the needs and economic possibilities of the region if public investment is provided in a form favourable to internal and foreign private investment."2 Although the old shell of isolationism was being weakened by such arguments, a much more powerful case was needed before the Plan Geral could be taken seriously by the Council. Hence the emphasis given to the Cabora Bassa dam itself. For if the dam could be built at all, far from it being a white elephant, the 90 per cent of excess power could be sold to finance its own cost and the fulfilment of the general plan for the Zambesi. But sold to whom? Research in high voltage transmission provided one answer: by 1965 it had become possible to transmit the power 800 miles to Johannesburg by AC or DC current without excessive losses. Thus the dilemma for the Portuguese created when Kariba fulfilled the demands of the Rhodesias could be avoided. But would South Africa buy power from Mozambique? In the face of world disapproval after the Sharpeville massacre of 21 March 1960, the Nationalist Government turned in on itself. The 'laager mentality' of Verwoerd seemed to preclude economic and strategic involvement with countries outside the borders of the Republic, which had enough problems of its own in extending apartheid without propping up the insecure colonial empire of the Portuguese. Afrikaner papers like Die Burger pointed out that General Smuts' decision not to take over Mozambique in 1919 had been made with the same point in mind. On the other hand, a more open (verligte) opinion could be detected and the Overseas Ministry noted the view of H.J. van Eck, chairman of the South African Industrial De-

Mais cedo o mais tarde 23 velopment Corporation, that the Republic must become a major exporting nation if it was to survive the crisis of the 1960s. The most serious obstacle developed out of Portugal's African wars. Revolt in the provinces was scarcely foreseen before 1960,13 and even enlightened sceptics were amazed at the violence of the uprising in Angola in February 1961 and the hideous atrocities and reprisals which followed. Revolt spread rapidly to Guinea- Bissau and built up in Northern Mozambique after the Frelimo (Frente para a Liberaqo da Mozambique) uprising in 1964. At the time the Plan Geral was published, there were more than a hundred thousand troops fighting sporadically on three separate and extended fronts at a cost to Portugal of not less than 2,540 million escudos a year - 3 per cent of GNP, 20 per cent of all public expenditure and nearly half the projected cost of the Zambesi scheme. (A comprehensive breakdown of the military budget will be found in Chapter 6.) If nothing else, these wars would have ensured that Cabora Bassa could not be built by Portugal alone. Worse, they had changed the climate of world opinion. Of the former colonial powers, France and Britain continued to show some sympathy and usually voted appropriately or abstained on critical resolutions at the United Nations, but opinion in the United States grew, in Portuguese eyes, increasingly unfriendly. The MFPZ were not so insular as to ignore the external situation. In the manoeuvres which preceded the introduction of the scheme to the Council of Ministers, they ensured that very careful publicity cushioned the impact of the claim for what, in all, was expected to cost the vast sum of 6,400 million escudos. With Treasury arguments in mind it justified past expenditure: 'We can conclude that this ... represents an incontrovertible economic investment, for it will permit the State to determine the value of its assets, whose exploration is to be undertaken by private concessionaires.'14 The plan also contained more substantial temptations. First, the calculations of the Hidro-Technica Portuguesa now showed that 90 per cent of Cabora Bassa's energy could be offered for sale at a price which, even on delivery, would compete favourably in Rhodesia and South Africa, thus

CABORA BASSA offering the prospect of self-generating finance. Second, the lakes formed by Cabora Bassa and by the ancillary barrages lower down the Zambesi would make the river navigable from the Rhodesian border to the sea, where a new port was scheduled at Chimbe. Before Rhodesia's illegal declaration of independence (UDI) in November 1965 it was possible to imagine offering the services of a thousand-mile inland waterway both to Rhodesia and to the newly independent state of Zambia. River transport by 20,000-ton barges could be expected to cut the transport costs of agricultural produce or the copper of Ndola by at least 60 per cent - to say nothing of offering more rapid facilities than the existing railways. The third inducement was the exploitation of the mineral wealth of Tete whose frontiers, so casually drawn in the last twenty years of the nineteenth century, included an area of ancient igneous rocks containing one of the world's richer mineral deposits. As with the vast resources of Northern Canada, the prospect depended on how cheaply the metals could be extracted and delivered at a coastal port. Given cheap electrical power, however, it was possible to count on deposits of coal, admirable for coking, and iron ore wh-ich though only partly suitable for steel making, contained vanadium and titano-magnetite, rare metals both heavily in demand in Europe, America and Japan. In addition, there were marketable reserves of copper, fluorite, and manganese. The world commodity market was short of all these in the late 1960s arhd the Japanese had already begun their intercontinental search for the raw materials to fuel industrial expansion. As early as 1966 representatives of Sumitomo Mining investigated the value of coking coal delivered at a deep-water port, together with the rarer metals. The MFPZ and the Minister, Adriao Moreira, pressed its case: build the dam; sell the power; offer a package deal to other neighbouring countries like Zambia and Malawi; draw the revenue and apply it to amortizing the cost of the dam, local industry and the future of Zambesi agriculture; and wait for such side-benefits as tourism and the fisheries of the new lake. For the price of a little faith Portugal would then have created an area of self-generating prosperity without parallel in the rest of Africa.

Mais cedo o mais tarde Attractive though the theory was, and despite the weight of evidence contained in the fifty-six-volume Plan Geral, opponents of the scheme in Lisbon were easily able to recall other attempts to achieve that desirable 'rhythm of development' which had left thousands of hectares of scrubland dotted with rusting machinery - most recently the British plan to grow groundnuts in Tanganyika. It was by no means certain that Salazar would prefer expansion to continuing the status quo. The aged President was apparently rigidly committed to orthodox financial policy. Ever conscious of the formidable assault on Portuguese colonialism, he was mistrustful of committing what strength his small country possessed. Moreover, the mordant example of India's seizure of Goa in 1961 was still a matter of outrage. Admittedly, the Council of Ministers had been given a more liberal look after the failure of an officers' conspiracy against the head of state in April 1961 by the inclusion of Moreira as Overseas Minister and Franco Nogueira as Foreign Minister. But Nogueira, who made the main case against decolonization at the UN in the early 1960s, retained few signs of his former liberalism, and Moreira was transferred by Salazar to the Institute of Overseas Studies in 1964, because he was thought to be 'Africanizing' too fast.5 Nevertheless, the assimilado system in the provinces (see p.262) had been abolished and the war in Angola could be seen to have acted as a massive stimulus to development, as the administration strove to 'win the hearts' of the loyal population. Might not the same be true of the poorer province of Mozambique? In the autumn of 1965 Silva Cunha, the new Overseas Minister, put the Cabora Bassa section of the Plan Geral to the Council and appealed at the same time to Salazar's own pronouncements about Portugal's African mission. For nearly six months the issue remained undecided and, more than any other point of contention during that deceptively quiet period, split the Council into its inward- and outward-looking elements. The Ministers of Finance and Economics both deployed budgetary arguments designed to appeal to Salazar's preference for retrenchment: the price (over £100 million) was far too great for Portugal to bear; servicing the debt on top of the cost of the African wars would be a crippling burden and

CABORA BASSA would divert slender resources from the far more important work of raising the standard of living at home; the necessary loans could only be raised abroad and to do so, or to offer export credits, would impair the credit of the Government at a time when large development loans were necessary for Portuguese industry. t Above all, they questioned the value of any investment at all in Africa when the problems of poverty, housing, under-employment and unemployment at home threatened to overwhelm the regime itself. Why indulge in mere speculative fancy for monuments dear to failed dictators like Mussolini when the true mission of the Estado Novo was to safeguard Portugal itself? The Overseas Ministry could not prove its case without providing evidence that there was a consumer ready to buy the surplus energy at a price high enough to set off the self-generating process. Since the declaration of UDI on 11 November 1965 put both Rhodesia and Zambia, for differing reasons, out of the running, there was no alternative to South Africa, and Franco Nogueira was enlisted to help. After five years' experience as Foreign Minister, facing United Nations' hostility and tending Portugal's tenuous links with her friends, Nogueira showed himself far more sensitive to the balance of power in Africa than his insular colleagues. To reinforce the MFPZ experts he brought the professional backing of his own diplomats, especially the Director General for Economic Affairs, Calvet de Magalhaes, whose negotiations with South Africa ran parallel to the Council debates in Lisbon. The Foreign viinistry had its own view of Cabora Bassa. Projecting the Portuguese dilemma onto the map of the world, they argued that the project could actually resolve the fundamental question of what to do with the colonial empire. It would prove to the critics not only that Portugal meant to stay in Africa (the wars showed that) but that her rule was neither stagnant nor conservative. Committing such huge sums (which would be nearlv doubled if the Cunene barrage scheme in Angola went ahead) would demonstrate to the U.N. and the Organization of African Unity (OAU) that Portugal had not merely ceased to exploit her colonies but was developing them t For further amplification see the Notes at end of chapter.

Mais cedo o mais tarde as territorio nacional, altruistically and at her own expense. After four inconclusive years of expensive military operations it would convince opinion - especially among those due for conscription - that eventual military success would lead to a peaceful future for the provinces. And it would have a multiplier effect on the African standard of life throughout Mozambique with repercussions that would benefit metropolitan Portugal: the balance of trade across the escudo zone would improve; prosperity would undermine the guerilla offensive; all in all, Mozambique would be less, not more, of a drain on the Treasury. (To meet the Treasury criticisms, Nogueira also undertook to secure the titles of export credits solely for Cabora Bassa, and this in such a way that they could not be transferred between foreign companies and banks.) Finally, even if the world refused to be convinced, the good-neighbour policy, expressed through creation of a southern African electricity grid and economic cooperation, would reinforce the ring of interest and security which already linked Mozambique with Rhodesia, South Africa and Angola. The Foreign Ministry sought an agreement with South Africa which extended well beyond the limited aims of the MFPZ. Although a measure of cooperation dated back to the days of Kruger and the building of the Delagoa Bay Railway, it was clear that Mozambique had become in effect a client state, exploited for certain useful services. The provincial administration in Lourenqo Marques relied on South Africa and Rhodesia for most of its foreign currency earnings, for the prosperity of the railways and of the ports of Beira and Lourenqo Marques, for tourism and the living standards of the urban white population. Although heavily dependent on the supply of Shangaan labour under contract for the Transvaal mines - by the agreement of 1928, renewed in 1971 (see p. 131) - South Africa was not noticeably grateful for this advantage. In 1938 her Prime Minister, General Hertzog, had consented to Chamberlain's proposed disposal of the Portuguese empire. And the existence of a poor and underdeveloped country, run on multi-racial principles by European. Catholics continued to offend the heirs of Kruger on almost every count. By 1965 the situation had changed. Buffeted, if not influenced, by world condemnation of apartheid, fearful of CABORA BASSA direct political and guerilla attacks by an increasingly violent African National Congress, and deeply shaken by internal unrest, the Nationalist Government of Hendrik Verwoerd came to welcome the continued existence of two buffer states between the borders of the Republic and the black states of Tanzania, Zambia, and the Congo. So long as this frontier - sometimes referred to, even by officials, as 'our northern border' - was fixed a thousand miles away, South Africa's internal security problem could be controlled. The revolts in Mozambique and Angola, Rhodesia's UDI and Zambian independence threatened this buffer state theory; but, outside the narrow circle of government in Pretoria, the fear of what would happen if Portugal were to withdraw or be defeated in the African wars if a black President took office in Lourenqo Marques, only a hundred miles from the real border - was not widely felt. The Afrikaner electorate, while accepting the need for regular security conferences with the Portuguese and Rhodesians, would not yet tolerate direct intervention in these areas. To Nogueira and Silva Cunha an agreement to buy electric power from Cabora Bassa would engage the white South in a political commitment, and bring a valuable injection of moral strength to the three colonial wars. Troops were not needed: the Portuguese army was still much larger and probably better trained than the South African defence force. But the signature of the only great power in southern Africa would show the Portuguese public that they did not stand alone, and would substantiate to Western eyes the thesis that the wars were being fought on the West's behalf against a Communist threat which only Portugal, by virtue of her history and African mission, could fully understand. The initiative did not come wholly from Lisbon, but Salazar refused to conclude the prolonged debate in the Council of Ministers, until he knew whether or not South Africa would cooperate. During 1965-7 political decisions were made in Lisbon and Pretoria which, though linked, were also decided in purely national contexts. Wide-ranging ideas of economic cooperation in southern Africa had been evolved in the 1960s. Admittedly, these were usually discussed at academic level; but the involvement of

Mais cedo o mais tarde Angola, Mozambique, Rhodesia, Malawi, and Zambia, in some general grouping, as a common market or customs area, was beginning to have an intellectual respectability.6 Of greater relevance, the scheme for an electricity grid linking the Cunene, the Zambesi, and the South African system, became more attractive in direct proportion to technical improvements in high voltage transmission and the industrial growth of the Transvaal. The campaigners for Cabora Bassa were fortunate in their champion, Dr H.J. van Eck, one of the three members of the South African 'Scientific Cabinet' of the inter-war years. Of solid Afrikaner upbringing, a follower of Smuts who had become a trusted adviser of Malan and Verwoerd, van Eck was as acceptable to the English-speaking community as to his own. His career as chemist and scientist, as well as economic planner, had fitted him for a prominent position in the narrow elite of South African government. A country boy from Kimberley who had made good in the depressed years of the 1920s, he had worked for van der Bijl, the architect of the State Iron and Steel Corporation (Iscor) and he became vice-chairman of the newly formed Industrial Development Corporation (IDC) in 1940. Earlier he had drawn up the van Eck Report, blueprint for much of South Africa's state-directed capitalism in the 1940s and 1950s - 'a masterpiece of investigation which revealed South Africa as a country rich in minerals, wool and sunshine, but poor in most other respects, and inefficient in its use of available natural resources ... It was the first coordinated national planning programme ever undertaken in South Africa and had a profound effect on the economy.'7 To Calvet de Magalhaes, Nogueira's personal envoy to the South African Government, van Eck seemed a visionary, with such a grasp of world affairs that in talking of Cabora Bassa he would conclude with a survey of the balance of economic power between China and Japan. To engineers like Henry Olivier he was one of the handful of compatriots who understood that in order to survive, South Africa must break through the carapace of Nationalist isolation and become the industrial leader of the continent. Van Eck had been shaping IDC as a medium for export finance for some years, and after linking up with the Reserve

CABORA BASSA Bank to form the Export Credit scheme in 1960 IDC became a potential source of loans for outward-looking schemes. By then van der Bijl and van Eck had had the constitution amended so that it permitted, 'with the approval of the head of state', the establishment and conduct of any industrial undertaking ... 'whether within the Republic or elsewhere'.8 Given the ability to raise funds in excess of 150 million Rand and latitude under the personal mandate of the Prime Minister, IDC could offer some of the institutional support Cabora Bassa needed.19 Van Eck's advocacy of Cabora Bassa was limited, however, by two forms of entrenched conservatism. Despite its apparent freedom, IDC was bound by a primary aim of freeing the Republic from dependence on outside sources of fuel and raw materials - an offshoot of wartime emergency mentality, enhanced in the aftermath of Sharpeville - which worked against buying energy from Cabora Bassa. Moreover IDC could only lend on its own terms and, as van Eck admitted to representatives of the British consortium in 1966, this raised in Portuguese minds the same fear of losing control of their national assets that had induced Salazar to exclude foreign capital from Mozambique in the 1930s. The second South African source of interest in Cabora Bassa lay among a handful of engineers and businessmen. Chief among the former was Dr Henry Olivier, a graduate of the University of Cape Town who had worked for most of his life with the British firm of civil engineers, Alexander Gibbs. Olivier had acted as Chief Engineer of the Owen Falls Dam in Uganda, was Gibbs' resident director at Kariba, and in 1965 was acting as project manager for the World Bank's enormous scheme in the Indus Basin, West Pakistan. His experience in dams and hydro-electric projects was unrivalled in South Africa, where he acted as consultant for Water Affairs on the Hendrik Verwoerd and P.K. le Roux dams on the Orange River. Just as van Eck was able to draw together the differing strands of opinion in Johannesburg, so Olivier was to form a link between van Eck, the Portuguese and other likely participants. Speaking at Canning House in May 1967 as part of the campaign to win over outside opinion, he pointed out that South Africa's power needs for 1980 were likely to be 21,000

Mais cedo o mais tarde megawatts, an increase of 200 per cent. Even more important than the cost of this expansion, estimated at £1000 million, was the fact that 'South Africa, unlike Mozambique, was a water hungry country. The installation of efficient modern steam stations will involve not only a very large capital investment programme, but further losses of water for cooling purposes.'20 As chairman of the commission studying the Republic's increasing water shortages he could add with considerable authority: 'South Africa would be interested in large blocks of reliable hydro-electric energy at prices per unit which are competitive in terms of the most recent steam stations - around 0.3 cents per kilowatt' (at that time, exactly the price suggested for Cabora Bassa). Looking far ahead, he spoke also of a grid linking the Cunene, a thermal station at Wankie in Rhodesia, the Kafue dam in Zambia, Kariba north and south, the Shire river in Malawi and Cabora Bassa with the South African system and the Oxbow scheme in Lesotho. 'This will make possible the regional interchange of energy on an import-export basis along the same lines as now operating so successfully and profitably in the (UCPTE) grid in Western Europe which links more than thirteen countries.' Van Eck asked Olivier to report on the Cabora Bassa site in 1965. Both men, aware of the opposition in South Africa and of the danger of offending Portuguese sensibilities, found it wise to take the line that South Africa should buy as much power, and no more, as was necessary to provide the revenue to get Cabora Bassa built. Only one large South African company was interested in Cabora Bassa, but the scope of the Anglo-American Corporation is so vast that the merest hint was important. The chairman, H.F. Oppenheimer, was already well known and respected in Portugal, the original connection arising out of Diamang, a company set up in 1920 by de Beers and Union Minifre to develop the diamond concessions in Angola. The association was extended in the 1960s in three areas in Mozambique - by an abortive venture into the fishing industry of the southern coast, by the development of cashew nuts (chiefly for oil) in Zambesia, and by prospecting for oil and natural gas in an enormous off-shore concession, in conjunction with the French Soci~td des Ptroles d'Acquitaine.

CABORA BASSA Being an outspoken Progressive, Oppenheimer was not exactly persona grata to the Nationalist Government; but those interested in Cabora Bassa and the concomitant mineral concessions were well aware that on most occasions what was good for 'Anglo' was good for the Republic. Tactics were concerted with van Eck. Esmond Brown, one of Anglo-American's executive directors, was assigned to the scheme to prepare and if necessary, lead, a multi-national consortium to build the dam. Willingness to provide financial help and, perhaps more important, to offer its good offices as international brokers was expressed by the Anglo- American finance house in London and Johannesburg, Union Acceptances. Thus by discreet manoeuvres Anglo-American indicated that Mozambique, like Angola, was an area of unexplored potential, ripe for cooperative investment. In sharp contrast to Zambia, where the Corporation's empire in copper mining had already been threatened with nationalization, the Portuguese provinces offered scope for long-term development, undisturbed by the threat of expropriation - as long as Oppenheimer could put faith in the Portuguese will to resist demands for independence. There would be less need of the crude leverage exercised in Lusaka, and a mutually profitable relationship with Lisbon could be founded on the example of Diamang. Unfortunately, the one institution on which all this depended showed not the slightest interest in Cabora Bassa. The Electricity Supply Commission (Escom) did not want to buy power from outside South Africa. Escom had been set up in the 1920s, like Britain's Central Electricity Board, to take over and coordinate individual local suppliers and it now provides over 90 per cent of all electricity used in South Africa.2' Escom's record since 1945 showed clearly why its chairman, Dr H. Straszacker, disliked the plans for Cabora Bassa. Faced with the backlog of the war years and the geometric progression of demand by industry, Escom planned and built rapidly through the 1950s a series of relatively small power stations, in order to gain time to think about the technological implications of the future and to make choices between coal burning, waterpowered, and nuclear-powered large stations. The next, intermediate, but still conventional, series proved

Mais cedo o mais tarde 33 far more costly: two were completed in 1962 and 1963; and the next four, culminating in a 6 x 350 megawatt station, were built between 1966 and 1971. Capacity in South Africa had doubled every ten years since 1940; but, as two of Escom's senior engineers put it, 'due to war periods, when plant could not be obtained, the construction of power stations has been a story of bursts of frenzied activity, rather than orderly increase. This has strained the resources of Escom and the plant manufacturers... '22 Van Eck's campaign hit Escom in the middle of one of these crises. The planning cycle was not due to start again until 1970, following a firm decision on whether to go for nuclear stations in the late 1980s (fuelled by enriched uranium from South-West Africa). Yet van Eck wanted Straszacker to commit Escom to taking not less than a thousand megawatts from an outside source in 1972 (the original target date for Cabora Bassa), at a time when South Africa's total producing capacity would have just been raised to over 9000 megawatts by the introduction of the Arnot station - with the even larger 500 megawatt sets of Krial due for commissioning in 1975. This was not the only objection. Escom's spinning reserve was normally maintained at 15 per cent of total capacity. Striszacker did not welcome the idea of half of this reserve coming from a dam nearly a thousand miles outside South Africa, let alone from a Portuguese province torn for several years by guerilla warfare. The security risk enhanced two natural suspicions: Escom was simply not used to European habits of importing and exporting energy across national boundaries, and was offended by the bland suggestion that it should help to finance Cabora Bassa at a time when its own resources were overstrained. Van Eck had to use political leverage; and so in the winter of 1965 Verwoerd was caught in the cross-fire. The debate about participation in Cabora Bassa came before the South African Cabinet at the very moment when verligte (enlightened) and verkrampte (ultra- conservative) forces were ready to divide on the general issue of foreign policy vis a vis the rest of Africa. Tension between the two factions of the National Party was not yet as obvious or openly expressed as it became in 1968 when Albert Hertzog took the hard core of

CABORA BASSA verkramptes into opposition, but at cabinet level it arose over Cabora Bassa regardless of how the project was viewed, whether as an economic, political, or strategic matter. The arguments in favour were, first, that the predicted rate of South Africa's growth demanded more energy than Escom's calculations suggested; that power from Mozambique would by 1980 be cheaper than anything the Republic could produce; that it would conserve slender resources of water and oil and that involvement would provide an opportunity to exploit the mineral wealth of Mozambique. Secondly, every legitimate form of support South Africa could give to Portugal in holding the line of the 'northern border' would be repaid ten times in stability at home. As Prime Minister and Defence Minister, Verwoerd was amenable to both arguments and J.B. Vorster, then Minister of Justice, also came to see it as a question of what he called 'the band of security'. In so far as it mattered, Sir de Villiers Graaff and the United Party (though unlikely candidates to replace the Nationalists) were known to be in agreement. Against them Eric Louw, then Foreign Minister, always applauded by the Nationalists for his policy of exclusive self-reliance, allied with the verkramptes and appealed to the basic philosophy of Afrikanerdom. Direct involvement in Mozambique was only a stage in a policy of engagement with black Africa: 'you cannot do this and keep the Republic as it is.' The argument which had begun at least two or three years before the critical Cabora Bassa decision rumbled on into the era of 'dialogue' and Dr Banda's visit in 1971; it widened out, for example, into the long running battle over whether or not to allow television, that likely solvent of strict Afrikaner morality, and whether to grant Cape Coloureds the right to participate in white privilege or to attain their own equivalent of a Bantustan. The remorseless logic of self-sufficiency, given a temporary advantage by the remarkable economic boom of 1964-8, seemed to bear out arguments that, since it offered no overriding economic benefit, substantial investment in a Portuguese enterprise would be hard to justify to the rural Afrikaner electorate. Cabora Bassa might turn out a Trojan horse. Similar objections had already prevented Verwoerd's government from advancing loans to Lesotho, in many ways

Mais cedo o mnais tarde a more attractive case for economic cooperation. Moreover Louw could point to the optimistic military communiques about the war which were being published in" Lourenqo Marques and Lisbon, and suggest that Portugal was able to maintain control without any help at all. The views of the government's economists in thisperiod are not without importance. Cabora Bassa did not figure in any of the long-term plans drawn up before 1966, nor in the five-year economic development programme23 but the climate of opinion in Pretoria certainly did not exclude it. Even though ostensibly multi-racial, Mozambique was also almost exclusively controlled by white capitalists and landowners and there were far fewer ideological and practical objections to investment than in any independent African state.24 The historic link was as old as the Transvaal itself, and as a leading academic economist wrote at the time: 'South Africa has no designs on the independence and freedom of action of her neighbours but is vitally interested in seeing them economically prosperous and politically stable.'25 At official level, the advice offered to ministers seems to have been to take the chance if the bargain was good enough. P.J. Rieckert, economic adviser to the Prime Minister, told a symposium in Johannesburg in 1971 that the key to recent (post-1965) policy had been: ... that closer economic cooperation in Southern Africa should not take the shape of a common market [but] rather ought to lie in a further development of the growing network of bilateral arrangements linking the economies of the sub- continent. The Government is strongly opposed to any form of 'charity' towards its developing neighbours. South Africa's economic relations with them should be aimed at strengthening their self-reliance ... trade rather than aid.26 Given the ingrown and secretive nature of Nationalist politics, it is not surprising that no public debate took place at the time over Cabora Bassa, nor even before the first press release in 1969. It is doubtful if anyone outside the inner circle at Pretoria even knew what was being discussed, for Verwoerd had enjoined closest secrecy. The first official approach from

CABORA BASSA Lisbon, after all, was not made until November 1966, nearly a year after an agreement in principle had been reached. But it appears that the decision was made personally by Verwoerd and that the two sides in the Cabinet were well aware that it had more than symbolic importance. Van Eck was fortunate in having influence with the Prime Minister. According to two close associates, he had a private telephone line from IDC to Verwoerd's office. More important perhaps was his grasp of the geography of African economics, a quality not shared by verkrampte ministers, who were still blinded by the uniquely favourable conditions of the economic boom. Van Eck believed that the very future of the outwardlooking policy rested on Verwoerd's assent to Cabora Bassa participation. Yet Verwoerd's decision was made as much for strategic as for economic reasons, in his capacity as Minister of Defence. 'A realistic decision', H.F. Oppenheimer called it27 - South Africa could no longer remain indifferent to the rest of Africa. The division in the Government was revealed when Louw resigned in 1967 to be replaced by the obviously verligte Hilgard Muller. But since political reasons had inspired Verwoerd, Escom remained unconvinced. Straszacker was unwilling to accept even the minimum cover needed by Portugal and refused a load higher than 500 megawatts in 1972 rising to 1000 megawatts in 1978. Two years of hostile bargaining passed, followed by the collapse of the stock market boom in 1968, before the advantages of Cabora Bassa were accepted on the economic level. Consequently the Heads of Agreement between the two Foreign Ministries were arranged only in principle. The details, being intractable, were postponed; and there is some doubt as to what, if anything, was actually signed. If the South African Government had realized that the Portuguese Council of Ministers was still locked in discord waiting for Verwoerd to make up his mind, the judgement might well have gone the other way. The Heads of Agreement has been unkindly compared to the hiring of a Rolls Royce by a speculator to give him reputable cover, when arriving at the bank to raise a loan. Certainly it gave Cabora Bassa the credentials needed in Lisbon, where the

Mais cedo o mais tarde Hidro-Technica had completed feasibility studies of the transmission lines which, they claimed, showed that power could be delivered within the magic limit of 0.3 cents per kilowatt demanded by Escom from the beginning.* No one could be certain of what Salazar would decide. After hearing the news of the South African agreement he told Nogueira enough to show where his mind lay, but he was rarely dictatorial in the Council of Ministers and, speaking last, tended to collate the majority view. There was also the vital matter of security; not until the spring of 1966 did the Ministry of Defence feel able to accept responsibility for guaranteeing the site of Cabora Bassa against Frelimo guerilla attacks. During the final series of debates in March the Council divided roughly as before. When they had finished Salazar gave a lucid and brilliant resume, the best Nogueira had ever heard him make. Clearly in favour, he remained scrupulously fair to the Treasury view and did not deny that the objections were great. At the end he added an argument which was entirely novel: certainly, Mozambique was not in a position to absorb even 200 megawatts for several years and to that extent the depressing fact of the territory being a service economy to South Africa would continue. But 'if Cabora Bassa is not built now we will never build it.'2 If they waited for the province to be ready, they would wait twenty years and lose the only chance of generating a self-financing development of the Zambesi basin, simply because by the end of the century technological development of fast- breeder nuclear reactors would have surpassed all hydro-electric schemes. South Africa would be self-sufficient and largely nuclear in the 1990s. No more dams like Cabora Bassa would be built. If, on the other hand, they built it in the 1960s, in twenty years it would have paid for itself. Among a group accustomed to thinking in terms of decades, the President's verdict made a vivid impression. Salazar was then seventy-eight and it was hard to imagine that he would * The study envisaged a first phase of a thousand megawatts (contrary to Escom's expectations) delivered either by two AC lines (500 megawatts at 525 KV) or one bi-polar conductor DC plus or minus 400 KV. The final capacity would be 2300 megawatts and the line was planned to run entirely through Mozambique.

38 CABORA BASSA live to see Cabora Bassa completed; but what he had said, taken with the South African agreement, was sufficient. In the two and a half years left to him the Prime Minister never wavered, in spite of the accumulations of trouble at home and overseas when the second great question was put: could Portugal finance Cabora Bassa without losing control, either to South Africa or to the European banks?

Notes and references Introduction 1 G. Lichtheim, Inperialispn, (London 1971) p. 12. Chapter 1 ,lais cedo o mais tarde 1 David Livingstone, The Zambesi Expediton 1858-64, (London 1865) 1, pp. 60- 2. 2 Ibid, II, p. 296. 3 .\lorais Cabral, 'A vitoria do nosso espirito colonizador', 0 Alundo Portughes, (Lisbon, May 1939), 6 p. 216. 4 Speech to the National Union chairmen, 23 May 1959. 5 Speech on Portugal and the anti-colonialist campaign to the National Assembly, 30 November 1960. 6 Diarno das Sessoes do Assembleia Nacional, Supplement 184, 8 March 1958. 7 Ibid. 8 The Limpopo Valley scheme, where the Canicedo Dam now provides for the irrigation of 100,000 acres, took twenty-five years to get through the planning stage. 9 Porteria No. 16214, 16 March 1957. 10 A.F.A. Falcio, Urn caso sere paralelo no Ultramar Portugies (Lisbon 1961). 11 Plan Geral de Fomento e occupaclo de vale do Zambesi (Lisbon 1965), Introduction, p. 4. * Note page 21. The World Bank made five loans to Portugal, totalling $57.5 million, all between 1965 and 1966, two for hydro-electric and three for thermal electric power development. All were guaranteed by the Government of Portugal, There have been no World Bank loans for projects in any of the overseas possessions. In theory there was no reason why the Bank should not have lent against a Portuguese (as opposed to Mozambique) guarantee as it had, for example, to Kenya in the early 1960s; but officials of the Bank made it clear to the Portuguese in private conversations, that it would have been very difficult, so that when the Cabora Bassa project did emerge, Lisbon did not make a formal approach. In 1966, when the cycle of loans was complete, Portugal informed the Bank that no more were required. But in any case the Bank would have faced strong opposition in the United Nations, whose Secretary-General annually reminded the directors about UN policy on decolonization; while the original Bretton Woods agreement gives the World Bank the right to draw up its own resolutions, it was scarcely worthwhile to ignore specific UN requests. Loans to South Africa, for example, were curtailed in 1964. 12 Plan Geral, Introduction, p. 7. 13 James Duffy, Portuguese Africa, (Harvard 1961) p. 342. 14 Plan Geral, p. 7. 15 Hugh Kay, Salazar and Modern Portugal, (London 1972) p. 238. f Note page 26. Portugal's external debt was rising rapidly in the 1960s: 1960, 1,899 million escudos, 1961 - 1,839m, 1962 - 5,223m, 1963 - 6,323m, 1968 - 8,362m. (Eduardo Guerra, Evolucao da Economia Portuguesa, (Lisbon 1967) p. 28. 16 See, for example, J.A. Lombard, J.J. Stadler and P.J. van der Merwe, The Concept of Economic Cooperation in Southern Africa, (Pretoria 1968) published by the (Government) Bureau for Economic Policy and Analysis.

CABORA BASSA 17 A.P. Cartwright, The IDC, 1940-70, (Johannesburg 1971) p. 8. 18 Sections 3 and 4 of amended constitution, IDC, p. 51. 19 See, for example, Dr H.J. van Eck, 'A Central Scheme for the supply of Electric Power in Southern Africa', Tegnikon (31 August 1967). 20 'The Development Potential of Mozambique and Angola', address at Canning House, 5 Mav 1967. 21 The last major private company, the Victoria Falls Power Co., whose ambition back in 1Q06 - not unlike van Eck's view of Cabora Ba,,a - had been to harness the Victoria Falls to erve the Witwatersrand, fell to it in 1948. 22 N. 1roo,t & H.B. Norman, 'Electricity Supply in South Africa 1909-1969', Transactions of5.-. Institute of Electrical Engineers (September 1969). 23 Cf. P.J. Rieckert, 'Economic Cooperation in Southern Africa through the Seventies' (Paper for the Symposium on Southern Africa, Johannesburg, July 1971). 24 G.ME. Leistner, 'Aid to Africa', Communications of the Africa Institute, No. 3 (Pretoria 1966). 25 G.M.E. Leistner, 'Economic Development in Africa', Communications of the Africa Institute, No. 2 (Pretoria, November 1965). 26 Rieckert, 'Economic Cooperation'. See also the address given by the ,Minister of Mines and Planning (J.F.\V. Haak) to the Wanderer's Club, Johannesburg, 31 March 1965: 'South Africa would also be favourably disposed towards assisting these countries to improve their infrastructure.' 27 Interview with the author, November 1971. 28 Nogueira's account: interview with the author, 15 November 1971.

Chapter 2 Building a consortium Early in May 1966 the General Trade Company of Geneva wrote to the managing directors of AEI (Associated Electrical Industries) International in London and to General Electric in the United States, explaining that they were acting as agents for the Portuguese Government and that they had, in the past, introduced groups of foreign companies to particular projects in Portugal and the overseas territories. Following the agreement between Portugal and South Africa (which they were at some pains to emphasize) they had been asked by the Director of Public Works in the Overseas Ministry 'to approach three or four major firms in the UK, USA and one, or possibly two, on the Continent, to ask them to form consortia capable of undertaking the development of the Cabora Bassa hydro-electric scheme on the Zambesi, a project very similar to Kariba'. Included with this invitation was a copy of the MFPZ plans, the details set out in the Plan Geral, the Hidro- Technica's design for the dam wall and their survey of the transmission line to South Africa. AEI and GEC were expected to set up consortia capable not only of undertaking the civil and electro-mechanical sides, and the major and only partly explored aspects of conversion and transmission on an unprecedented scale, but also to find their own cover against the financial risks involved. To approach electrical rather than civil engineering companies was a reversion of the 'normal' procedure, but logical in the circumstances; and the rather roundabout form of invitation was designed to give time for the formation of several consortia while preserving the appearance of an open market when (as

CABORA BASSA the Government then intended) a tender was offered publicly later in 1966. General Trade's letter, however, did not mention that other consortia were being formed, at least one under the auspices of the Anglo-American Corporation. The form of invitation did indicate that, in spite of its lack of financial resources, the Portuguese Government intended to maintain control of the dam at every stage of construction and in its final operation. A system had to be evolved which allowed the use of Portuguese designs, specifications, continuous supervision by Portuguese engineers, and overall control, thus ruling out van Eck's original plan for South African financial dominance. Portugal wanted the advantages of both worlds - to use the skills of the great international companies and the services of European banking and yet not to relinquish more than a minority share. But was the Government in any stronger a position than those of Ghana or Egypt in the cases of the Volta and Aswan High dams, where Presidents Nkrumah and Nasser had been equally susceptible to influence exercised from outside?' The Portuguese had already avoided an application to the World Bank. Equally adamant was their refusal to be bound hand and foot, like Egypt with Russia, to a more powerful (South African) partner. But although they possessed the capacity to design Cabora Bassa and did not therefore have to buy the expertise of companies such as Impregilo (the Italian group which had built Kariba and won the Volta tender), what bargaining power they had against a situation such as that of Ghana, whose foreign debt had rapidly become a crippling restriction on Nkrumah's freedom of action, was questionable. If they wanted Cabora Bassa, had the future development of Mozambique to rest ultimately with foreign bond holders? Without the existence of the agreement with South Africa, it is unlikely that anyone would have been interested in the terms outlined by General Trade, excluding as they did export credits or Portuguese Government guarantees. The Treasury did not intend to pay for the works directly but over a period of twenty years and at rates of interest not above six per cent. Consequently, the risks either of failure to complete or of failure to pay appeared unreasonably high. European engineering firms were not so short of public works that they

Building a consortium 43 would easily venture into a part of Africa much more insecure and remote than Kariba. But a number of political factors perceived by European governments overrode the lack of export credits; and the technological challenge turned out to be irresistible. Direct current was widely used in the early days of electricity, but by 1900 it had been almost replaced by alternating current. Nevertheless, the losses involved in AC power lines over long distances were so great that in the 1930s research into direct current was taken up again in Germany, then unquestionably the leading nation in the electrical world. Transformation into DC of the alternating current produced by generators and its reconversion at the far end of the line before assimilation into a consumer network proved to be the crucial problems. All attempts failed in the days of conversion by rotating machines because the cost and losses of power were greater than in the AC system; but with the development of the grid-controlled mercury-arc rectifier, long-range DC transmission ceased to be a mere dream. Under the febrile stimulus of war, converter stations were evolved in Germany using the mercury-arc rectifier, and Siemens had just completed a DC link from the Elbe power station to Berlin when the Third Reich collapsed. The Russian army promptly dismantled the converters and took them home for study, together with as many of the Siemens staff as they could take prisoner. Research in Germany ceased for ten years. Then, in 1954, the state-run industrial research agency, Deutsche Forschungsgemeinschaft, set up a study group at Heidelberg University. With a government subsidy, a model high-voltage DC unit was built at Mannheim-Rheinau in 1960; three years later the commercial development and manufacture of mercury-arc valves was begun there by a partnership between the three main competitors in the electrical field, AEG, Siemens and the Swiss Brown-Boveri. Each company provided a subvention of ten million marks and the rest of the very high cost was covered by the Forschungsgemeinschaft. By 1966 this group had begun to investigate the totally new system envisaged in the use of the thyristor valve - an offshoot of space research which was potentially more

CABORA BASSA stable and effective than the mercury-arc valve. The gap of nearly twenty years had, however, lost the Germans their wartime lead. The Russians rebuilt the Elbe- Berlin system. The majority of their prisoners preferred their research to the alternative of prison and hard labour. Given this start, Russia was able to unroll a vast new system in the late 1950s and 1960s. When Cabora Bassa came on the market a 475-kilometre line was already in operation in Volgograd, carrying 750 megawatts at 400 KV; and a line of 2500 kilometres had been projected to carry 6000 megawatts at 750KV from the lignite field of Ekibastus to Moscow. Nothing so vast had been conceived of in the West. In Sweden the firm of ASEA, part of Marcus Wallenberg's industrial empire, pioneered the mercury-arc valve and laid the first high-voltage DC cable from Sweden to Gotland in 1953. In the ensuing decade ASEA monopolized the market for components and granted a licence to English Electric. Not until the mid-1960s were they threatened by the German research and by the work done in the United States by General Electric.t Although DC transmission works out cheaper than threephase AC over a given distance, converter stations are highly expensive. The point of overall advantage of DC over AC (on 1966 costs) lay between 500 and 1000 kilometres, but it could be foreseen that with a wider market for components the DC cost could fall much lower. Moreover for a submarine cable, or a line through heavily populated areas, the maximum effective distance of an AC system could fall to as little as 80 kilometres. DC thus became important not only for the vast spaces of Canada and Africa but for almost any new line in Western Europe. In addition there were opportunities for coupling asynchronous AC grids by DC links (thus allowing for inter-territorial exchanges) and for the elimination of the problems of stability in controlling the flow of energy. Since world consumption and generating capacity was expected to double between 1966 and 1980, and double again by 1990, the future of DC transmission seemed inexhaustible. The Cabora Bassa scheme was launched just when the research of the Germans, English, and Americans needed to be t For further amplification see Notes at end of chapter.

Building a consortium put to the test in order to capture this booming market. Millions of pounds had been sunk into research and development which could only be recouped in contracts like Cabora Bassa - and the number of foreseeable schemes outside 'protected' territory (such as North America and Russia) was not great. Immediate profit was, therefore, if not immaterial, at least a lesser priority. The Portuguese had chosen their companies shrewdly, for the other sections of the project were not especially attractive. On the civil engineering side Cabora Bassa, although a large dam by any standards, presented no unusual technical problems other than the time-table imposed by the annual flood of the Zambesi. There was prestige to be gained and hope of substantial profit, but no inducement to sacrifice profit to the proving of new skills. Recognizing this the Portuguese, while specifying the precise design of the dam wall, actually encouraged tendering companies to submit alternatives or incorporate their own ideas in the electro- mechanical sphere. At the same time they hinted that the lowest tender on the first bid would not necessarily be accepted, because the cost of money was equally important. Extremely complicated tactical manoeuvring followed. Apart from the tender specification, the only common factors were that the British-Italian group and the European group relied on and competed for the goodwill of South Africa's IDC and the banking facilities of Union Acceptances; and that all the groups utilized very influential associates in Lisbon and South Africa: Henry Olivier for Concassa (British-Italian), Nicolas Kagan of Compadec for Cabora Bassa Builders (United States), and Mario Fereira of Anglo-American for Zamco. Concassa was the child of AEI, married to English Electric (for the converter stations), British Insulated Callenders Cables (BICC) for the transmission cable and supporting towers, Boving (the turbine experts), and Costain (who had made the access roads at Kariba), with Sir Alexander Gibb and Company, as civil consultants, and Merz MacLellan, as electro-mechanical consultants. In June 1966 Concassa sent a letter of intent to Lisbon, offering a package deal for the entire project including the access roads and the township of Songo.

CABORA BASSA (To be built on the mountain plateau above the gorge, and to house not only the dam personnel but the local administration.) They had already approached the Export Credit Guarantee Department in London where enthusiasm was tempered by a certain mistrust that the enterprise would not be wholly under British direction. Nevertheless, ECGD's fears that Cabora Bassa might not be commercially viable appeared to be stilled by the major South African commitment. No-one yet suspected how vague that agreement with Portugal was. Olivier made a first inspection of the site at the end of June 1966. The MFPZ had made enough trial borings to convince him of the stability of the granite cliffs; the projected diversion tunnels appeared capable of taking even unusually high flood water; and his chief impression (shared with the emissaries of the other groups) was that the access road, winding down 3000 feet of unstable and precipitous mountain wall, was going to be much harder to build than at Kariba. He also warned against the entangling complexities of the provincial bureaucracy. By the autumn of 1966 Concassa had studied four types of scheme, allowing for stations of either 1200 or 2400 megawatt, AC or DC. The results showed clearly that the overall cost was lowest for the highest dam with the largest capacity. There was a margin of total cost in favour of DC transmission at the lower level and of AC at the higher, and very little difference in the price per unit delivered to South Africa. Lacking political leverage in Lisbon, Concassa approached the South African end. A discussion with van Eck in September 1966 elicited the fact that he personally considered it valuable, for political as well as economic reasons, to have a British consortium. One of the long-term advantages he hoped for would be 'a realistic and meaningful settlement in Zambia'. He strongly disapproved of allowing the Americans the chance of penetrating Mozambique. On the other hand, he could not directly influence the terms: IDC was ready to offer finance, but on most matters had to give way to Escom, the actual buyer. Van Eck hoped Escom was ready to act more boldly, 'especially in view of the studies initiated by Verwoerd on nuclear power which are now completed. Also with the integrated system, the threats of peak problems are mounting.'

Building a consortium 47 To find out whether to plan for the smaller or larger capacity at Cabora Bassa, the heads of Concassa visited Lisbon, only to discover that no detailed figures had been agreed. They asked whether Escom were not dragging their feet; and in November, having set up a special South African committee, sought to find out what additional finance could be raised there and what progress, if any, was being made. So much one-sided pressure may have offended the Portuguese. Although Concassa were not wrong to assume that Escom's attitude was hostile, their attempt to find out what was going on in Johannesburg was only partly successful. IDC, Union Acceptances, and the Standard Bank of South Africa offered them financial aid; but from Escom they learned only that Portugal had made no formal approach. The South African Government appeared unable to take basic political decisions; and Escom revealed such a profound mistrust of the security risk and such unwillingness to take more than 500 megawatts in the early years that the Concassa leaders concluded that their only choice was to return to Lisbon and suggest, tactfully, that the 'agreement' between the two countries did not suffice for a tender. By December 1966, however, the formal Portuguese request to tender had been sent and the heads of Concassa met Dr Castro Fontes, Director-General of MFPZ, now styled the Gabinete do Plano do Zambese (GPZ), and Pimental dos Santos, the Minister of Public Works. On 20 February 1967 Concassa entered into a formal agreement with General Trade. The fluctuations of the talks during 1967 made life difficult for all three consortia, but particularly for the British. Having been first to organize - they were the only group actually able to conclude a deal for completion of the dam in 1972-3 as originally planned - they were forced to stand by and watch the European group, Zamco, from its inception under the guidance of Anglo-American, work slowly through its enormous initial difficulties to become increasingly competitive and influential in Lisbon. At the same time Rhodesia's UDI and the Labour Government's growing aversion to involvement with white governments in southern Africa began to undermine their chances.

CABORA BASSA It was comforting for Concassa that Escom found English Electric competent to build converter stations; also to have van Eck's assurances that, despite the Wilson Government's gyrations over the issue of arms for South Africa, his own Government had no political worries: 'their sole concern is with safe construction, on time, within the cost estimates. This is a Portuguese job.' He even offered to finance Concassa's local costs in Mozambique, thereby discounting what he feared might be the Labour Government's main objection to the granting of export credits. By August 1967 there were firm enough grounds to assume that the Portuguese would build a south-bank station at Cabora Bassa with three 400 megawatt sets as a start, with two more sets to come. The Overseas Ministry was able at last to draw up designs for the power station and the electro-mechanical side at the end of September 1967; and Concassa, knowing exactly what to plan for, began to prepare their tender for the closing date, 12 March 1968. A year had been lost but their export credits were still safe - the Rhodesian crisis had not altered Foreign Office thinking about Mozambique. The official line was: there is no mystery about the Government's position towards Cabora Bassa. We have made it clear, throughout, that we should not wish to dissuade British firms from participation in this project if they felt that it was in their commercial interest to do so, except insofar as participation might involve contravention of our existing Rhodesian sanctions legislation.2 In spite of the ambiguous answer given by Lord Shepherd, junior Minister at the Foreign Office, to Lord Gifford3 this doctrine stood during the lifetime of the Labour Government. Of course, ECGD conditions excluded supply of Rhodesian cement or transport facilities. But so, under United Nations mandatory sanctions, should other governments' credits departments have done for Zamco and the Americans. During 1967 the composition of Concassa was enlarged. The Italian group, Impregilo, who had built the major part of the Kariba and Volta dams, had wished to tender. Although they allied with the Grupo Industria Elettromechanica they

Building a consortium 49 failed to get on to the original list because they had no experience of DC transmission. None the less, they approached General Trade and set out to form a consortium including the French electro-mechanical group, Creusot- JeumontSchneider, with Societa Anonima Elettrificazione (SAE) of Milan to build the actual line.t When the Italians realized that an AC solution would not be acceptable, they decided to merge with Concassa, and thus the British acquired the immense asset of Impregilo's African experience. CreusotJeumont-Schneider and SAE were shed, amicably enough, on the way. The American group, Cabora Bassa Builders, was formed differently. With the sponsorship of Compadec, the Paris finance house, Morrison Knudsen, the largest private civil engineering company in the world, teamed up'with General Electric to offer a complete package deal. Thus there was no need for the patient assembly of multi-national groups: consultancy and project management services were provided automatically from the resources of the twin companies and for finance Cabora Bassa Builders turned to the Export-Import Bank. But any American group suffered from severe political handicaps. The United States Government and, by implication, American private enterprise, was scarcely popular in Portugal in spite of the agreement over, and revenue from, the Azores and other NATO bases, which only partly nullified the effect of US votes at the United Nations against Portuguese colonialism. Portuguese opinion tended to associate Americans in general with the supporters of liberation movements in colonial Africa. Admittedly, prejudice did not exclude American aid on the right terms - US Steel and Morrison Knudsen finally built the Tagus Bridge - but if the terms were equal, say with the Franco-German offer, there could be some doubt as to whether a decision would not rest ultimately on political factors. Cabora Bassa Builders were at a worse disadvantage in respect of South African participation. The United States' t For further amplification see Notes at end of chapter.

CABORA BASSA prolonged hostility to apartheid and the supply of arms, and White House sensitivity to the power of the black vote in America, could not easily be offset when asking for credits from the Export-Import Bank. Nor did it make the American companies popular in South Africa, and they found it expedient to bring in Roberts Construction of Johannesburg, various French electrical engineers under the direction of Compagnie Electro-Mecanique, and the Portuguese engineering concern Mague. Most significant was the fact that their original sponsor, Compadec, was headed by Nicolas Kagan, who wielded substantial influence in Lisbon. But Cabora Bassa Builders relied principally on their world-wide reputation for package deals and filled in the tender in the conventional way, neither seeking to cut corners nor to offer alternative solutions. The result was possibly safer and more elaborate than the Portuguese required (although the Hidro-Technica evinced some suspicion about the testing of GEC converter valves) and suggested either a lack of communication with the client or a lesser need to win. Unlike the other consortia, whose structure grew naturally out of existing industrial links, Zamco was a novel and artificial construction. Normally, for very large international schemes the civil and mechanical sections - dam wall, power station, access roads - are awarded as separate contracts, or else responsibility for the whole is taken by a single consortium, such as Impregilo. At Cabora Bassa the Portuguese desired a single contract with one entity; yet Zamco, formed out of a heterogeneous mass of companies, failed to throw up a leader. There was no European equivalent of GEC or English Electric. ASEA, which had the knowledge, had never worked in harness with firms from Common Market countries; Siemens, though one of Germany's largest private companies, was also the major competitor of the even larger AEG, except in the one area of DC transmission. Apart from ordering electro-mechanical equipment from Alsthom and Neyrpic, suppliers of turbines and generators, neither had links with the French. After seven years of the Common Market it was remarkable how little cooperation in this field existed. On the other hand, the Portuguese wanted a minimum of

Building a consortium 51 three consortia to ensure genuine competition and thus strongly wished for French and German participation. South Africa, too, had an interest in an association of French and German companies, not least because of French arms sales and German loans and the favourable trading policies towards southern Africa of both governments. But the particular impetus came from the Anglo-American Corporation which had kept its interest in Mozambique discreetly quiet during the first round of talks between Portugal and South Africa. Anglo-American, for straightforward commercial reasons, hoped that Union Acceptances or the Banque de Paris et des Pays Bas, in which it had a 10 per cent stake, would act as brokers to whatever consortium won; and since each of the groups intended to incorporate a South African associate it was natural to introduce the Anglo subsidiary, LTA - one of the larger, but not strikingly profitable, South African civil construction firms. In the background lay the Mozambique mineral, oil, and natural gas concessions. How far the Corporation could exercise leverage in Lisbon is hard to determine. H.F. Oppenheimer himself was well-known and his policies acceptable to Salazar; in Angola, for example, Portugal was deeply committed to Diamang and other Anglo subsidiaries.4 In 1967 Diamang rendered to Lisbon 1204 million escudos worth of diamonds, a fifth of Angola's total exports. The subsequent history of the Mozambique concessions (Chapter 9) certainly suggests premeditated intervention. Two high-ranking members of Anglo-American put Zamco together - the Portuguese Mario Fereira, local director in Paris, and E.T.S. Brown in Johannesburg. Fereira was given to understand that firms with close links with or subsidiaries in Portugal (and by implication in South Africa) would be looked on favourably at the adjudication. British and American firms were tacitly excluded because they already dominated the other groups. Existing industrial commitments and national pride, then, dictated the allocation of the various aspects of the project within Zamco - the converter stations to the Germans, civil engineering mainly to the French and the South Africans, and electro-mechanical engineering between the Germans and the French. The civil engineering side was easiest to put together

CABORA BASSA because, if the Germans were to lead the electro-mechanical group, this had to fall to the French. The five largest firms in France were already associated in the Compagnie des Constructions Internationales (CCI) an organization with a distinctly monopolistic flavour, which tenders as a single entity and then allocates jobs to its components on a fair shares basis.t To lead the civil side, CCI nominated the massive Socidtd des Grands Travaux de Marseilles (GTM). Then, spreading the load for reasons of political balance rather than any inadequacy of GTM, were added the French company, Fougerolles de Limousin, and the giant German concern, Hochtief (which had, ten years earlier, planned the Aswan High Dam), together with LTA and the South African drilling specialist subsidiary of Anglo-American, Shaftsinkers. The value of the work was divided roughly into three equal parts, corresponding to the nationalities. The ubiquitous SAE of Milan, which stood to gain by the success of either Zamco or the Americans, became responsible automatically for the power line for the entire route. (SAE was related, industrially, to Alsthom and Brown-Boveri, and hence to the electro-mechanical side of Zamco.) Favoured by the Portuguese Government, for which it had built the line from Beira to Umtali in Rhodesia, it was also familiar to Escom who had been impressed with the work of its subsidiary in South Africa, Power Lines. Among several disharmonies, the worst disputes arose over the sharing of the electro-mechanical side. The French were neither content with German leadership nor willing to be subordinate in an area so profitable for the future. In addition to Siemens, AEG and Brown-Boveri (with Alsthom-Neyrpic as suppliers of generators and turbines), therefore, a place had to be found for the Compagnie Generale d'Entreprises Electriques (CGEE) - like CCI a monopolistic grouping of the main electrical engineering companies in France.tt It was partly to compensate Germany for this enforced sharing that Hochtief was brought in on the civil side. At some time early in 1967 these arrangements were upset by a firm demand from Escom that ASEA should be included. t For further amplification see Notes at end of chapter. tt For further amplification see Notes at end of chapter

Building a consortium 53 The Mannheim-Rheinau research did not, apparently, offer a sufficient degree of proven reliability. Because six months had already been lost Anglo-American were prepared to keep Escom happy at almost any price and Fereira extended an invitation to the Swedes. But the incorporation of ASEA introduced grit to the machinery of Zamco. They were not only keen competitors of the Germans; Swedish was incomprehensible to the others and ASEA had to use English, to the disgust of the French; moreover the Swedish Government, vociferously anti- colonial at the United Nations and pledged to support liberation movements in Africa, fell into violent disfavour at this time in Portugal - so much so that Volvo cars tended to be daubed with paint or have their windows smashed in the streets of Lisbon. The Overseas Ministry eventually accepted that ASEA was a necessary evil. In an odd way, this alien element resolved some of the Franco-German tensions which arose because the contract had to be made as a whole. Not only was national pride to be subordinated, but any one of the firms in Zamco might find itself saddled with the differing techniques, to say nothing of the errors or incompetence, of another - a risk as offensive to the South Africans as to the Europeans. But hostility towards the Swedes proved to be a unifying factor and when, after a dozen stormy meetings in the spring of 1967, ASEA was finally accepted, most other internal arguments dissolved from sheer exhaustion. ASEA itself seems to have been unconcerned at entering the arena of mutual mistrust. Marcus Wallenberg was on good terms with Franco Nogueira and discounted the political opposition in his own country; and ASEA's own financial risk of $15 million was largely covered by Union Acceptances and the Banque de Paris et des Pays Bas. Two problems remained: the creation of an organization which could negotiate as a single entity in Lisbon; and the covering of the disparate sections of Zamco with export credits. In contrast to the other consortia, these problems preoccupied Zamco for over a year, after June 1966, and were not wholly solved by the time the printed tenders were sent out. Indeed, if further time had not been lost in the arguments between Lisbon and Pretoria, Zamco might not have been able CABORA BASSA to submit a price by the closing date in March 1968; even then, its tender was in sowe ways more speculative than those of Concassa and Cabora Bassa Builders. In spite of seven years' experience of the Common Market, multi-national commercial organization remained in its infancy. The obligation to employ the French, German, Italian, English (for the Swedes), and Portuguese languages created difficulties, but the fundamental disagreements centred on control. Who should designate the 'speakers' for the three main sectors? Who should form the secretariat? Where should it meet and who should be chairman? National and local interests strove for advantage, while any pooling of information necessary for pricing and coordination was regarded with intense suspicion. It was as if a Council of Europe had to be formed in a matter of months to arbitrate on the most intimate national questions. Fortunately, the Anglo-American Corporation offered a framework and a chairman, E.T.S. Brown, who was not the spokesman for any particular group. Meeting the demands for representation for each company left a rather large negotiating team of six members. There remained one major defect. Partly to avoid the appointment of an uncontentious nonentity, and partly because nothing was certain until the final adjudication, Zamco at first had no project manager. Brown, though an excellent and impartial chairman, was based in South Africa and could not fill the gap. Detailed control at the centre lay therefore in the hands of the six, and later of their representatives in Lisbon. With such a cumbersome machine crises had to be overcome by individual action, usually by Fereira. The unease between French and Germans was only mitigated, not dispelled; and right up'to the opening of bids in March 1968 an element of hasty improvisation persisted. The problem of financial cover also derived from the nature of the contract.' European banks were interested, but not, they made clear, in order to accept the responsibility rejected by the Portuguese Treasury. They would provide initial risk-capital until Zamco could win the preliminary adjudication, but the long-term risk had to be carried elsewhere, by export credits from the French, German and Italian Governments and IDC in South Africa. Thus the consortium of banks formed by

Building a consortium 55 Union Acceptances and led by the Banque de Paris et des Pays Bas was assembled to act simply as insurance brokers. Because the individual companies could not yet apply to their respective governments, and because of doubt about whether in any case governments' export credits would suffice, Anglo-American tapped other sources, mainly Portuguese private banks. Anglo-American itself lent very little, the final commitment being a mere 3.6 million Rand. Yet in what can best be described as agency services, they spent 2 million Rand for which no reimbursement was sought. Interest rates on the money markets began to rise sharply in 1967 because of the world liquidity shortage, and cover offered by commercial banks soon cost much more than the 6 per cent on which all 1966 calculations had been made. By the middle of 1968, money was only to be had short-term at 10 per cent. Zamco found itself critically dependent on the answers given by its members' governments to the (as yet) hypothetical question of export credits. Unlike the British and Americans, who needed to apply to only one authority, Zamco could suffer from the whims or separate interest of three European governments whose answers, even in the preliminary stage, were based on widely different premises. Already the French Government had shown itself more aware of the strategic advantage of aid to Portugal and entry to Mozambique than either Germany or Italy - both of whom lacked the aggressiveness of France's policy towards Africa during the last years of de Gaulle's rule. Thus when CCI and Cogelex applied in the usual way to the Banque de Commerce Exterieure, giving an approximate idea of the scope and cost of French participation, the Ministry of Finance considered the application on commercial and political grounds - and decided that it deserved not merely consent but promotion. (For the purposes of such applications, the Ministry of Finance includes the Foreign Relations Department of the Treasury [DREE] which also receives the advice of the Foreign Ministry.) The reply, given before Zamco submitted its tender, indicated that the Ministry would give adequate credits for fifteen to twenty years. The fact that the Quai d'Orsay had concurred with the Treasury signified that the final steps would be a mere formality.

CABORA BASSA From the point of view of the French construction firms, an overseas contract is highly desirable. The tardy system by which government departments pay for public works built in France means that cash may be delayed anything from two to four years, causing a permanent liquidity crisis for even the largest members of CCI. Hence the propensity to form combines to share out the work; hence also the keenness of the Government to support them by guaranteeing foreign contracts on favourable terms. There is a close relationship between those who build most in France, and those who win overseas work. (Significantly, the French equivalent to the Confederation of British Industries, the CNPF or Patronat, is very largely staffed with ex-colonial civil servants.) Thus the firms grouped in CCI have built two-thirds of the dams in France since 1965 and have, in the same period, worked on the Queban Dam in Turkey, Tarbela in Pakistan (the largest in the world), and Cabora Bassa. This monopolistic situation leads to ineffective cost control inside France, but the effect outside is that French contractors receive a disguised subsidy. Siemens and AEG, though world-wide industrial empires, concerned themselves in the 1960s with 'good business', not with where their business lay, and largely ignored political repercussions. As it happened, the conjunctions were good: Siemens had historic connections with Portugal, for example, it built and still operates the Lisbon metro and had also done substantial work for Escom in providing generator sets too large for South Africa to manufacture, while the AEG factory in Johannesburg (telephones, electric motors, and transistors) is one of the largest in South Africa. The Christian Democratic Government of Dr Kiesinger looked with favour on an extension of exports, particularly since it was beginning to regard South Africa as a major potential market. Kiesinger did not seem to be aware of the colonialist overtones of Cabora Bassa - with some justification for, as late as 1969, the United Nations Economic Committee for Africa favoured the project, specifically because Malawi, one of the poorest countries in Africa, was expected to benefit. But above all, having sponsored years of expensive research, the German Government was concerned that their own electrical industry should capture the new market in DC transmission. Thus 400 million

Building a consortium 57 Deutschmarks in credits was offered through the Kredit Anstalt, a sum reckoned to be enough to cover all German participation in the electro-mechanical side. Italian companies were then building up substantial influence inside Zambia: Alitalia had just captured the supply contract for Zambian Airways and SNAM- Projette were constructing the oil pipeline from Dar es Salaam to Lusaka. President Kaunda visited Italy in 1967 at the invitation of Enzo Nazionale Impregilo (ENI) who were on the look-out for future investments. When SAE applied for credits there was no suspicion that Kaunda would one day denounce Cabora Bassa as an imperialist plot, but there was nevertheless a faint hesitation in Rome at supporting Portugal, a country potentially hostile to Zambia. On the other hand, what SAE regarded as the inveterate dilettantism of politicians and the waywardness of the Roman bureaucracy made it by no means certain that it could count on any Italian Government promises. Prudently, it inquired through its subsidiary, Power Lines Ltd., whether South Africa's IDC would cooperate; and also suggested to Zamco that the Portuguese might be willing to pay it direct. Meanwhile the application went to the Ministry of Foreign Trade which, early in 1968, issued the necessary affidavit, confirming that there were no major objections from the external economic viewpoint. The affidavit was only an indication of government attitudes, not a firm promise; and it did not guarantee credits. SAE continued to the next hurdle - the verdict of the Instituto Credito Estero, where investigations into what might be offered, and when, occupied the rest of 1968. Local costs, however, and wages. and supplies in Mozambique itself constituted a gap for the Portuguese Treasury of roughly $75-80 million. The Paris-Bas consortium offered this amount to the Overseas Ministry as a loan, but were refused, partly because of well-justified fears of rising world interest rates and partly because such an arrangement would have undermined Treasury principles. This problem was far from solved at the time of tendering. The year's delay in 1966-7 was in no way deliberate. Until

CABORA BASSA the specification was issued, the consortia could not price the job; and the Overseas Ministry could not issue it until they knew what size and type of sets the power station would hold. Since 90 per cent of the energy had to be sold, these questions depended entirely on the figure that South Africa could be induced to buy. Unfortunately, at the end of 1966 van Eck relinquished the negotiating role to Escom and the absence of his enthusiasm was felt severely. The situation was made worse by the assassination of Verwoerd in September 1966. B.J. Vorster, the next Prime Minister, continued to support Cabora Bassa, having been deeply impressed with Verwoerd's commitment. From the moment he assumed office he endorsed the outward-looking policy; in his first public statement he declared, '... we must realize that we are part of the African continent and must act accordingly,'6 but he lacked the prestige, accumulated experience and diplomatic skills of his predecessor. Escom was unwilling to negotiate with a foreign government without the assistance of its own and was particularly worried that it would be forced to accept, for political reasons, terms which it would not have countenanced in the open market. It opposed any suggestion that Portugal might provide taps in the line either for Lourenqo Marques or Rhodesia in order to get up to full capacity at Cabora Bassa as fast as possible. South Africa, as van Eck made clear, 'was not interested in a line which could be tapped'. The Portuguese Government feared, equally, that they might lose overall control. A line to Johannesburg would facilitate Cabora Bassa, but it would be unthinkable to subordinate the whole development of the Zambesi valley to the wishes of a single customer. The Overseas Ministry was also concerned that the preliminary agreement had been based on too small a capacity. Uneasy lest the final size of the power station should be determined by StrAszacker's caution, they tended to ignore the practical argument that the rate of commissioning such vast sets signified more than the final capacity. Consequently they insisted on a firm commitment from Escom, rather than trust in the growth of the South African economy over the 1970s. Differences in principles were enhanced by alien habits of

Building a consortium 59 negotiation. The Portuguese found the South Africans over-optimistic, vague, and brash; the South Africans thought the Portuguese devious, legalistic, and unpredictably Latin. Misled by the diplomatic accord during the talks between Escom and the Overseas Ministry in Lisbon, in March 1967, the Portuguese assumed that Escom and the South African Government were as closely linked by common interest as their own Foreign and Overseas Ministries. In fact, Escom's decisions depended on several related factors which could not easily be brought into alignment. After the March encounter, it decided tentatively to accept 1000 megawatts on completion of the dam and to use 90 per cent of this as base load for the national grid. But the corollary was that Escom needed a spinning reserve of at least 500 megawatts from South African sources, so that Cabora Bassa power should not at any time exceed half its total reserve. Maximum capacity would have to be achieved two years earlier, therefore, at the Hendrick Verwoerd and van. der Kloof dams on the Orange River, requiring not only the full cooperation of the Ministers of Finance (Haak) and Planning (Carol de Wet) but a substantial alteration of priorities in the National Development Plan. Before the next round of talks Nogueira appointed his Director of Economic Affairs, Dr Calvet de Magalhaes, as special envoy to South Africa with plenipotentiary powers, and in July 1967 Magalhaes was able to conclude a settlement with Escom and the South African ministers in Pretoria. Against their original figures (based on a commissioning date of 1972) Escom, under considerable pressure, undertook to accept 1000 megawatts in January 1974 and 1200 by January 1978, at 0.3 cents per kilowatt hour at 80 per cent load factor. Of this, 75 megawatts was to flow back from the South African grid to supply Lourenqo Marques; and the South African Government undertook to be responsible for whatever power was delivered over and above the agreed total. In return, Portugal was forced to abandon the idea of a tap-line to Rhodesia, and to accept the condition that South Africa, as priority customer, had to be consulted both on price and quantity of any electricity sold elsewhere. On this basis the Overseas Ministry and the HidroTechnica were able to draw up detailed plans for a south-bank

CABORA BASSA power station of three 400 megawatt units, leaving space for two more as a second stage to be commissioned before 1978. But the South Africans proposed taking a financial share of 25 per cent in the power stations and 50 per cent of the whole transmission line as a precondition of IDC loans, thus giving them a substantial say in the operation and maintenance of the system. Calvet de Magalhaes, on grounds of long-term overseas policy, was bound to refuse. At this point, the South African Foreign Ministry issued a demand that the Cunene scheme on the Angolan border should go ahead in conjuction with Cabora Bassa. This ddmarche, which threatened Portuguese plans in Angola with political exposure,* evoked an immediate reply. In the name of Salazar, General Trade told the South Africans that the two schemes must be kept separate; Cabora Bassa, however, would proceed at once. Yet another year passed before a compromise was reached, under which South Africa would own the converter station at Apollo near Johannesburg and all the line inside the Republic, and would provide loans and guarantees totalling 160 million Rand. The passage was long and arduous: according to Calvet de Magalhaes, two or three times the talks were almost broken off. Incompatible demands were only resolved because the negotiators themselves were overtaken by the deadline of March 1968. Despite political goodwill, and Vorster's wish to achieve good relations with Malawi, Escom only gave way reluctantly, finally accepting figures of 1040 megawatts for 1974-7, then 1280, rising thereafter. A period of six months had been allowed for between sending out the tender specifications and the submission of bids in March 1968. The Overseas Ministry had intended to make a preliminary adjudication in April and reach final signature in October of that year. The specifications, delayed by the South African negotiatons, were not sent out until November 1967, but the date for submission of 12 March was rigidly adhered to, and the loss of two months put an extra and * The Hidro-Technica's engineers had predicted the destruction of 50,000 hectares of fertile land and the necessary removal of 25,000 inhabitants. But the real danger was political, for the scheme involved widespread white settlement, in areas previously owned by African farmers, and the Government feared an international outcry.

Building a consortium 61 almost unbearable load on the resources of the consortia. Not surprisingly, given its federal and multilingual constitution, Zamco suffered most. There were, however, other difficulties. The openness of the electro-mechanical specifications allowed for more than one solution, and a considerable variety of prices. Cabora Bassa Builders chose to tender a single price, Concassa allowed for alternatives, and Zamco decided to contain several options within one figure. The sealed bids were opened in a meeting at the Overseas Ministry in Lisbon on 12 March. The result was: Escudos Zamco 7,033,048,345 Concassa 7,087,911,503 (standard solution) 7,512,640,703 (alternative) Cabora Bassa Builders 9,713,665,685 The Americans appeared to have written themselves out. Between the others the margin was not only small but extremely hard to assess because of the variations implicit in the Zamco price and because these figures did not include the cost of financing, which might add another 30 per cent. Much more had to be investigated than the Portuguese had imagined: there would be haggling and price cutting all round; and finally there was the question of what, if any, political weighting should be applied. For the first time the Portuguese were constrained to seek outside advice, choosing a British firm of consultants, Preece, Cardew and Ryder, to evaluate the electro-mechanical alternatives. But an April adjudication was ruled out. Even the lowest bid was much higher than had been expected in Lisbon; the Overseas Ministry had to obtain all the cuts it could, even if that involved prolonged bargaining. Moreover, the cost of financing, particularly the short-term cover and local costs in Mozambique, threatened to impose a burden cumulatively heavier than had been expected in the crucial years 1972-80, before the full volume of revenue would be recouped from South Africa. The calculations on which Cabora Bassa had been approved could be nullified by a 'bow wave' of interest charges which would fall directly on the Portuguese budget,

CABORA BASSA to the horror of the Ministry of Finance. Portugal began to feel the effects of the economic squeeze in South Africa as Dr Diederichs' counter-inflation campaign retrenched on public spending. Escom suggested cutting back its intake to 1000 megawatts in 1974 and to a mere 1070 megawatts in 1978. If those figures were accepted, the 'bow wave' would continue into the 1980s. The only resort was to squeeze the two leading consortia for economies. American offers of reductions did not stand up to investigation and Cabora Bassa Builders did not come back into the game until the following year. The other prices offered wide scope for bargaining. Although Concassa's B Scheme was ruled out by its cost, modifications could be made to the spillways, such as those it and Zamco had already sketched out. Some dispute also remained about the sum of $21 million, the cost of moving and resettling the population of the Zambesi valley, for which the Ministry had asked, in effect, for a loan. The Americans had refused; Zamco agreed, but made it contingent on the preliminary award; Concassa included the sum in its price - a fact which was not discovered for some time. Over the next few weeks, during which complete secrecy was imposed, an inordinate amount of haggling took place over the cost of financing and economies in construction. Portuguese expectations that the various European export credit departments might be brought to reduce their rates found confirmation in the diplomatic barrage which opened up in Lisbon. Nogueira received information from the British Foreign Office, where Michael Stewart had just been appointed after George Brown's resignation, that special consideration for Concassa would be welcome.7 Sanction for long-term loans at low cost had already been put before the Wilson Cabinet and the Foreign Office made it clear to Concassa that, with a future investment of more than £130 million in the Portuguese empire, the Government was serious in its intentions. Pimental dos Santos indicated that 'if some further small reductions could be made', and the rate of interest held just below 6 per cent, 'their tender could be very attractive indeed.' Similar instructions went to Zamco, of whose member

Building a consortium governments France at least had been active at government level. Delegations from both sides arrived in Lisbon at the end of March, offering cuts which were, in the nature of this sort of bargaining, used against the other. Concassa spent these days besieging Lazards Bank, ECGD, the Standard Bank of South Africa, and the South African Reserve Bank. Zamco's resources were greater: the Paris- Bas consortium was able to offer, within a fortnight, total cover at 6 per cent and what amounted to a moratorium on interest repayments during the critical 'bow- wave' period up to 1978. Zamco then added a flat cut in the price of 400 million escudos, achieved by economies in excavation and sluice design. Concassa counter-attacked on 18 April with a cut of 550 million escudos. Later, at the end of May, it returned, after a series of negotiations in which the Labour Government and the British Treasury showed remarkable goodwill, with a means of deferring payment and capitalizing interest which amounted to a seven-year moratorium. The offers were now indistinguishable. Both sides had made concessions amounting to roughly £ 10 million and clearly could make no more. Costs and techniques being equal, politics ruled the choice. The Portuguese Foreign Ministry told Concassa on 29 March that Nogueira's feelings towards it 'had taken a more favourable trend' because of the seriousness of its offer; and crude judgements that the offence caused by the Beira Patrol ruled the British out seem to be wide of the mark. (Two years had passed since the Joanna V had been blockaded in Beira harbour in April 1966.) Nogueira maintained that there was no political element in the choice, but certain evidence suggests otherwise. The weight of South African opinion was made clear after the bids were opened, not least because the shock of the high tenders showed the Overseas Ministry how essential that support was. A series of events had already prejudiced the Vorster Government against Britain, in particular the unedifying squabbles in the Labour Party after the Cabinet crisis over arms for South Africa in December 1967 and the developments in Rhodesia, where, however unwillingly, the British Government might face United Nations' demands for the use of force against the illegal regime, or for sanctions

CABORA BASSA against South Africa. In contrast, France and Germany began actively to support Rhodesia by buying much of the tobacco crop and by supplying cars and other scarce machinery. Portugal's lack of progress in her African wars may also have been a factor. In order to safeguard Cabora Bassa, Rhodesian military cooperation might become necessary. Beyond that strategic consideration lay the implications of the Portuguese 'good neighbour' policy of security - collaboration with South Africa and Rhodesia. Finally, with regard to her current and future trade in armaments, as well as commerce, and her hopes of eventual associate membership of the Common Market, Portugal was closer to France and Germany than she ever could be to England, her 'oldest ally'. These feelings were certainly reciprocated in France: when Wilson pressed de Gaulle to stop French oil companies supplying Rhodesia through Mozambique, de Gaulle replied: 'The Portuguese faced great difficulties ... and France tried, on the whole, not to make life too difficult for them.'9 Most observers in Lisbon and South Africa at the time were aware of a fundamental mistrust at dealing with the Labour Government, and the final decision in favour of Zamco was not surprising. The Daily Telegraph correspondent, Ian Colvin, had written from Lisbon in November 1967: 'The suspicion in Pretoria of British policy towards Rhodesia, and terrorist movements from neighbouring states, have made it nearly impossible for British firms to compete for contracts.' In the event, Concassa gained another month's extension, but was unable to improve on its offer. Zamco held to its price and the Portuguese, after consultation with South Africa, gave it the preliminary award on 10 July. Concassa disbanded, having spent more than £100,000 in tendering. Pimental dos Santos was able at last to order a start on the works; and in the hills beyond Tete, Zamco prepared to take over from the GPZ.

Notes and references Chapter 2 Building a consortium 1 Cf. R.W. Steel (on the Volta scheme) and D.C. Watt (Aswan High Dam), in Dams in Afri'ca, ed. N. Rubin & W.M. Warren (New York 1958). t Note page 44. ASEA and English Electric were responsible for the Sweden-Gotland, Denmark-Norway, France-England, and Sardinia-Italy undersea cables. General Electric laid the line from North to South Island in New Zealand (600 megawatts) and in 1966 was projecting three big lines in North America Canada-Vancouver Island, the United States West coast link and Nelson River- Winnipeg. 2 Written reply (by Maurice Foley) to Parliamentary Question, 27 April 1970. 3 House of Lords debate, 17 September 1969. Lord Gifford was Chairman of the British Committee for Freedom in Mozambique, Angola and Guinea. f Note page 49. SAE is probably the largest supplier of cable and structural steelwork for transmission lines outside Russia and the United States. When the rest of the Italians joined Concassa they agreed to work for Zamco and also offered the same terms and tender price to Cabora Bassa Builders, the American consortium. 4 Dr Americo Boavida, in Cinco Seculos de Exploitaco Portuguesa (Rio, 1967), suggested that the real exploiters of the colonial situation were Western capitalists who held Portugal itself in economic dependence. t Note page 52. CCI is made up of Grands Travaux de Marseilles; Socidt6 Franqaise d'Entreprises de Dragages et des Travaux Publiques; Compagnie Industrielle de Travaux; Socitd G~nerale d'Entreprises; and Entreprises Campenon-Bernard. tt Note page 52. A working federation of the production and distribution aspects of the French electrical industry, CGEE is run by private enterprise under a state mandate. It coordinates policy with state undertakings in other countries, such as the CEGB and LINAT in Britain, and Hermes in Germany. For Cabora Bassa, CGEE set up a subsidiary group, Cogelex, which drew export credits from the French Government. 5 In legal terms, the contract was signed between the Overseas Ministry and Zamco Limitada of Lisbon, a small company with minimal capital, which then signed subsidiary joint and several contracts with each firm. 6 Minister of African Affairs, interview, November 1971. 7 Interview with Nogueira, 15 November 1971. 8 See, Harold Wilson, The Labour Government 1964-70, (London 1971) pp. 470- 2. 9 Ibid., p. 407 (June 1967).

Chapter 3 Contractors' politics The demands of the time schedule imposed by the Zambesi's floods forced Zamco to begin work at Cabora Bassa before the final contract was signed, but its directors assumed that after a short period of haggling over the details and after confirmation of the financial arrangements by their own governments they could achieve this as a formality. In the event, eighteen months went by before signature, and the deal was not finally settled until the end of 1970. Time and again the hard-won understandings on which Zamco had based its tender were dissolved by the acid secretions of international politics. The pledges of financial cover made during the tender were redeemed in direct proportion to the long-term national interests of the governments concerned, although the reasons actually given for refusal or confirmation were couched in more exalted terms. When the leaders of the three sides of Zamco, secure in the psychological advantage of victory at the preliminary stage, made their firm applications, they received warm support from the French and South African Governments, punctilious service from the Germans, but equivocation in Rome. The French companies' request reached Coface, (Compagnie Franqaise Pour L'Assurance Du Commerce Extrieur), the French equivalent of ECGD, early in 1968, and after the preliminary award the application went on in the normal way to a commission consisting of members of Coface and the Ministries of Finance and Foreign Affairs. Although the commission did not give its avis de principe favorable until

Contractors' politics 67 August 1969, it gave private assurances of the utmost consideration. How favourable may be judged by the outcome: 560 million francs (E45 million), of which 459 million was a fourteen year credit at the excellent rate of 6 per cent interest, with 12 per cent of the total made over at once to help with local costs. The terms were unusually good and the local cost cover exceptional in French overseas contracts, a favour which corroborates the strong support from the Quai d'Orsav and the Bureau des Affaires Africaines et Malgaches. But although no shadow of public opposition on grounds of anti-colonialism clouded these talks (indeed, very few observers in France were even aware of French participation) the Government was careful to play down its part: the Banque de Paris et des Pays Bas was instructed to deny that it had played any part at all in financing Cabora Bassa. South Africa's contribution in mid-1968 was 126 million Rand - 45 per cent of the net total at that time. By the end of 1969, after final signature, this had risen to 165.2 million Rand ( 82 million), but in 1970 two further sums were added: an Escom loan to SAE of 17 million Rand to cover the transmission line in South Africa and a 7 million Rand credit from IDC to the German consortium. Thus the total South African investment reached 190 million Rand (still roughly 45 per cent of the initial cost) and, if the cost of financing were included, must by then have neared 300 million Rand (E150 million). The implications of the subscription of this enormous sum will be discussed later; already, however, South Africa had become a bank of last resort for Zamco and to a great extent the saviour of its part in Cabora Bassa. Lacking IDC cooperation, Zamco might well have been forced to withdraw in 1969-70, despite the preliminary award. Opposition to Cabora Bassa as a symbol of Portuguese colonialism began to stir in Germany as soon as the preliminary award was made public; and Manfred Eppler, the SPD Minister of Trade, questioned the credits already promised to the Siemens-AEG-Brown-Boveri group. Nevertheless, the Kiesinger Government (which included Willi Brandt as Foreign Minister and Karl Schiller, Finance Minister) confirmed the main credit of DM 400 million and duly guaranteed it through the Kredit-Anstalt ftir den Wiederaufbau (Bank for Reconstruction).

CABORA BASSA In Italy, however, the SAE application for 32 billion lire ($51 million) was caught by political storms. In spite of the certificate from the Ministry of Foreign Trade, serious objections were raised. The Central Credit Commission (IMCC), which handled the actual amount of money, as opposed to granting theoretical approval, refused the requested rate of 6 per cent over twenty years and reduced the total. Then came the election of 1969 and a lurch to the left. The new Foreign Minister, Signor Moro, was caught in a dilemma: respect for Portugal's position as a fellow member of Nato, and awareness that a credit for Cabora Bassa might affect Italy's rapidly growing interest in black Africa, notably in Zambia. In the end the Commission, whose Socialist chairman, Professor Paravicini, evinced a fierce dislike of Portugal and its colonial heritage, offered a derisory sum - 18 billion lire at 6.5 per cent over ten years - and only on condition that funds were available. SAE, which had not been paid credits for earlier contracts already completed,t feared that availability would depend on the political whim of an apparently intractable Commission and turned to the Portuguese Government for the remaining 14 billion lire, threatening to void its contract if the credit were not made up. Kenneth Kaunda, President of Zambia, returned to Europe in the summer of 1970, mainly to enlist support in the imposition of effective sanctions against Rhodesia but with the secondary motive of boycotting the Portuguese. By threatening to withdraw favour from Italian companies in Zambia he extracted from the Minister of Foreign Trade, Signor Zagari, a promise to refuse any credit for Cabora Bassa. Technically Zagari could not revoke the Government's existing offer, but he announced that IMCC funds were exhausted. For SAE the funds would remain exhausted and even old debts would remain unredeemed. Deterred by government action, and by hints that the Italian trades union movement could not ignore a company which bolstered white racialism in Africa, SAE set up a holding company in Switzerland which then took over the majority share of its South African subsidiary, Ferolimo (Pty) t For further amplification see Notes at end of chapter.

Contractors' politics 69 Ltd. Ferolimo in turn controlled Power Lines (Pty) Ltd., to which SAE had already assigned the South African end of the transmission line and to which it now added the Mozambique sector. Having covered its chain of control against adverse publicity, SAE continued as before, except that the structural steelwork for the towers was exported from Italy to South Africa rather than to Mozambique. Complete cover was now demanded from Lisbon. When the Portuguese Treasury, predictably, refused a guarantee, SAE suggested payment in cash - a request backed by the South Africans. Dr Diederichs was persuaded to allow IDC to put up another credit of 7 million Rand as an inducement, and to keep out the only alternative, the British company BICC. In the end, beset by both the South Africans and the Foreign Ministry in Lisbon, the Treasury gave way and in September 1970 agreed to pay the Italians in cash, direct. The detailed engineering designs had allowed for some modifications in the civil engineering side and a very great freedom on the electro-mechanical side. Since the discharge from Cabora Bassa would be two and a half times as great as that at Kariba, where difficulties had been experienced with erosion of the river-bed down-stream, Zamco feared that the eight huge spillway jets and the downflow from the turbines would set up uncontrollable currents and weaken the narrow bed behind the dam. Working with the Hidro-Technica Portuguesa engineers, the French produced a new design for deep sluices, intended to throw the water 600 metres down-stream. At the same time Zamco agreed to incorporate the Portuguese state-owned steel company, Sorefame, which had made the gates for Kariba, and to give it a contract for all the steel structures for the sluice gates and for the roller gates themselves, thus providing a much-needed injection of Portuguese manufacturing enterprise. On the electro-mechanical side, Siemens, spokesman for the German group, spent six months with the Hidro-Technica working out a modified design for the main underground chamber and discussing technical specifications of the most important single item, the converter stations. The preliminary award had been based on a mercury-arc valve converter, but

CABORA BASSA although generally satisfactory and proven over a number of years, this solution had operational disadvantages, including the fact that under certain conditions it ceased to behave as a valve. Research in several countries suggested that by the early 1970s most large schemes would be using thyristor valves instead. (The thyristor is a solid state valve basically similar to the transistor, a silicone- controlled rectifier. The Arbeitsgemeinschaft HGU had developed a whole series of banks of these connected valves in stacks, like multi-storey flats, in an air- cooled tank at the combined research project at Mannheim-Rheinau.) Well aware that the progressive delays in undertaking the Cabora Bassa contract were working in their favour, the German companies included a sketch of a possible thyristor converter station in the Zamco tender.' The research results at Mannheim-Rheinau were far enough advanced at the time of final signature for the Portuguese Government to reserve for one year the right to choose this system in preference to the mercury-arc valve. The Germans were not slow to emphasize the disadvantages of the latter solution at conferences of Zamco suppliers; but despite their own success, the new technique had not been proved outside the laboratory. It was clear, as the Mannheim-Rheinau working group reported in June 1968, that 'the testing and demonstration of the valves . . . under conditions similar to those encountered in practice ... is an essential factor for the procuration of orders'; and it required a great act of faith on the part of the Overseas Ministry to commit itself to a £50 million system which could only be judged experimentally. The builders of the Churchill Falls Dam on the Nelson River in Canada had chosen experimentally- tested thyristors from AEI, but conditions in North America diminished the risk. Moreover, even if the Germans could prove their case, ASEA would be excluded at a price, perhaps, of some South African goodwill. Once again, the heterogeneous composition of and conflicting motives within Zamco made for greater difficulties than would have been experienced, say, by the American consortium. At first the Portuguese found that to work with Zamco was like dealing with the sub-contractors building the Tower of Babel. In the absence of a single project manager, or a

Contractors' politics 71 chairman with overriding authority, the differences of language, method and, above all, of temperament and style were sharply exposed. The French would turn up in Lisbon with their three representatives; the Germans, on the other hand, brought not only their directors but a copious staff, most of whom were replaced during the next three years. The Germans complained of being faced with a Latin conspiracy between Portuguese and French; and the Italians tended to go their own way with as little reference to the Controlling Committee as possible. The difficulties of coordination might not have mattered greatly at this stage. A project manager was, after all, appointed in 1970 and his work, though begun late, became increasingly effective as the dam's construction progressed. But there were months of intense controversy, after the preliminary award in mid-1968 and twice again in the summer of 1969, when the diffuse mechanism of Zamco was alarmingly overstrained. At the same time Zamco faced a team of Portuguese civil servants and engineers who, while sympathetic to them as contractors, nevertheless exacted every concession and exploited every internal difference with a bland courtesy that led the Germans to curse their 'Arabian mentality'. Whereas the Zamco Committee met only monthly and its members had to commute between Nuremberg, Paris, and Lisbon, or rely on local representatives, the Portuguese ministerial team - Silva Cunha, the Minister, Arantes e Oliveira, chairman of the Conselho Superior de Fomento Ultramarino, and Castro Fontes, head of the GPZ, and the Commissao Financeira played on its home ground and consequently held great advantages. The government service, particularly in the Foreign and Overseas Ministries, resembled that of Whitehall at the end of the nineteenth century. Though the basis of entry was wider than before the 1939 war, most senior civil servants shared a common elitist background - a degree in engineering ranking graduates with Polytechnicians in France. In the Overseas Ministry career civil servants served long terms, often stretching into decades, in Timor or Macau, Guinea-Bissau or Angola, and their headquarters in Lisbon had a polyglot air, as if it had absorbed the colonial tradition from its transient

CABORA BASSA inhabitants. As the British did in India, or the Germans in Africa, the Portuguese prided themselves on a high tradition of adminstration, admittedly slow, but just and dedicated. Products of a system designed to provide rulers, they perpetuated it even when they spoke of change. Characteristically, Arantes e Oliviera, the central figure in the talks with Zamco, was chosen as Governor of Mozambique to see the great enterprise through. Thus the contractors faced men with common ties and shared assumptions, who could pool the knowledge of three Ministries, who spoke a language that only one man in Zamco understood, and who were used to planning over decades rather than months. As the contractors found, after discreetly testing the atmosphere, they were not open to corruption; a point worth making, given the fact that bribery of one sort or another is endemic in public works contracting - subtle in most parts of Europe and North America, increasingly blatant in the Middle East, and rampant in Africa and Latin America. Even the South Africans were surprised at the probity shown by officials in Lisbon. The Portuguese negotiators also believed that until a bargain had been made any strategy was fair. Irritated by the consequent manoeuvring, the South Africans and Germans chafed at having to haggle in an alien fashion; and they were not pleased to observe the French and Italians relying on a bond of Latin sympathy. What few of them understood was the problematic nature of the Portuguese position: the fear, on the one hand, of too large a commitment, of loss of control and of being shackled in the future; and on the other, the brake of a rigid bureaucratic tradition which demanded on every document the seal not only of a civil servant but of all the grades up to and including the Minister himself. Too often, hesitation was explained away as the stupid cunning of the peasant, unable to see the good deal offered by the openhanded, trusting men from Zamco. In fact, Zamco, a confederation often at war with itself, held out to the Portuguese an advantage which would not have been present in the British or American consortia offering a complete package deal. The technocrats in Lisbon bargained with the engineers from Italy, France, and Germany, whose qualifications lay in

Contractors' politics 73 experience rather than research and who lacked political cohesion. The agile minds of Castro Fontes and Arantes e Oliviera were able to win endless concessions against a body held together only by self-interest and the vision of a handful of individuals. Of the six Zamco members, a French civil engineer (Lemperire), a French electrical engineer (La Bossi~re), a Portuguese (Mario Fereira), a German (Wilhelm Forstmann of Siemens), an Italian (Roberto Pavolini), and a Swede (Bungt Unden of ASEA), only the first two were acquainted, and their only bond was a common dislike of the Swede. Later, despite the language barrier and wide differences of background and age, these six became friends whose allegiances were defined by their contention with the problems of multi-national organization. They struggled to communicate in English, the lowest common denominator; they commuted to Lisbon, talked endlessly on international telephones, wasted hours in referring back. Every point of detail took twice as long to assimilate as it would have done had there been a single main contractor. If a bargain was concluded at first in English and then failed to match up to the original understanding when translated into literal Portuguese, the French would help Fereira seek the spirit of the translation. Then the team would negotiate on the Portuguese text and return the result to English. Each man left Lisbon after the final contract with two vast folders of drafts and re-drafts. Yet when these six men could agree the Portuguese found them a formidable team. Long afterwards, when the hectic months before final signature were forgotten, La Bossi~re would call this cooperation a unique event - 'a great work for Europe'. Like the meetings of the EEC Commission in Brussels, their wrangles always hard up against a deadline, they would seek a compromise in the heat generated by Germans shouting 'ich wiinsche' and French countering coolly Ye voudrais bien'. Their tangible memorial was Decreto Lei No. 49225, passed by the Council of Ministers and translated into four languages, which embodied the contract in September 1969. Four months afterwards new offices were taken in the best floors of a modern block on the Avenida da Liberdade - appropriately

CABORA BASSA enough below the floor given over to the Anglo-American Corporation, and flanked by Air France and the Commercial Bank of Lisbon. Portuguese internal problems in these months added sombre overtones to the project. Salazar publicly affirmed his faith in the future of Cabora Bassa in August 1968, but a month later he collapsed, paralyzed after a stroke, leaving the Cabinet which he had dominated for forty-two years without a director. Marcello Caetano, a former Minister and professor of economics at Lisbon University, emerged as Salazar's successor, but he was a man almost unknown, believed to be vaguely liberal but in the shadow of his peers. Orthodox mistrust of Cabora Bassa at once revived. The project was caught up in an enormous agony of enquiry about the entire attitude of metropolitan Portugal to the overseas empire. To Silva Cunha and Nogueira, the architects of a system of provincial autonomy combined with integral membership of greater Portugal, Cabora Bassa symbolized the imperial mission; but it was attacked by two dissimilar groups, each of whom commanded substantial support in and outside the Council of Ministers. On the political and financial grounds of over-commitment, one group reverted to the doctrines of Salazar's middle years, of a tightly balanced budget and political repression. The other, younger and more radical than either Cunha or Nogueira, saw Portuguese destiny as lying solely in Europe, particularly in associate membership of the Common Market, and it put forward a programme of steadily increasing devolution, followed by self-government for all the overseas territories. The closest links that it wa-,ted were the sort of cultural and religious harmonies that were popularly believed to exist with Brazil. Portuguese investment in Cabora Bassa and South African domination of Mozambique were both undesirable and deeply harmful to a Eurocentric future. They feared, correctly, that only a Portugal which had dissociated itself creditably from its empire would be acceptable to the EEC. Caetano had never been closely associated with the past history of Cabora Bassa, and, in spite of his earlier writings on the colonial question, he was not yet politically equipped to

Contractors' politics 75 adjudicate on an argument of this scale. Understandably, but to the dismay of the contractors, he refused to make any statement about Cabora Bassa and even allowed to continue unchecked speculation that the scheme might lapse. In private, however, he conducted his own investigation of the political strength of the South African agreement. For some months before Salazar's stroke, relations between the two countries had been soured by the disputes with Escom, consequences of the higher costs of Cabora Bassa and of the deflation imposed by the South African Government. A meeting between H.F. Oppenheimer and Salazar to discuss the future of the AngloAmerican Corporation enterprise in Portuguese territory had therefore been arranged for November 1968, and it was thought so important that the old man's illness was not allowed to affect it. In discussion with Caetano, Oppenheimer confirmed his interest in the development of Mozambique, and Anglo's readiness to invest in mining, oil, and diamonds. They discussed the price of gold on the world market and the general political situation in South Africa and, despite Caetano's evident doubts about Cabora Bassa, reached what an Anglo-American executive called 'absolute accord' on all points. Soon after his accession as Prime Minister, Caetano arranged a visit to Angola and Mozambique for March 1969, thus gaining six months in which to make up his mind on the future relationship between Portugal and the empire. Nogueira found his way very different from Salazar's - more purely political, a sensing of prevailing attitudes and possibilities rather than the imposition of logic on recalcitrant facts. In these six months Caetano reviewed the chances of breaking out of the pattern set by Salazar in his last years and of rejecting the 'indissoluble bond' with the empire in favour of an opening towards Europe. Cabora Bassa was central to this decision and when, in Lourenqo Marques on 18 April 1969, he delivered the judgement, 'Cabora Bassa will soon be a reality,' he confirmed irrevocably for his tenure of office the imperial idea he had inherited from his predecessors. Several reasons for this decision have been suggested by members of Caetano's Cabinet and by critics of his policy. Creation of the dam would undeniably have been a triumph

CABORA BASSA for a colonial power never rich in trophies, and a sign to Africa that modern Portugal was no less mean in aid than Russia or China, sponsors of the Aswan High Dam and the Tanzam Railway. Cynics could have added that the decision came at just the right time for Caetano to acquire most of the credit and thus bolster his own position at home and also his popularity abroad, sufficiently to carry on the colonial policy against United Nations hostility and African rebellion. As a matter of internal politics, affirmation was a sign to the disciples of Salazar that Caetano had assumed the moral as well as the practical power of the old dictator. (Salazar, despite almost total incapacity, actually put his signature on the final contract.) But, more than these, Caetano's decision revealed that he accepted what had only been half hinted at before. The road towards an equal and multi-racial society in all Portuguese territories was to be shortened from the century and a half envisaged by Salazar to a mere ten or fifteen years. The visit to Mozambique, though played down by the Lisbon press, was a triumph greater than any accorded there before to a metropolitan prime minister. Caetano indicated that there would be a redress of the imbalance between metropolitan Portugal and the exploited provinces, and he promised to overcome and, if necessary, outlaw the opposition of the white settler minority to increasing multi-racial development. How far these promises were kept will be seen later (Chapter 11), but in the stagnant atmosphere of Lourenqo Marques, and by Portuguese Asians and the assimilados of the south, Caetano was seen as a liberator. In these six months Caetano also settled the renewed grumbling of the Treasury. There was now a clear gap of $80 million on Cabora Bassa, owing largely to local costs not covered by South African loans nor by the contractors. The 'bow-wave effect' would therefore fall in the years 1975-80, and there seemed no way to meet it without cutting back on development plans in Portugal itself. Not much of the burden could be imposed on the revenues of Mozambique (see p. 215) and it would have been prejudicial to existing bond issues to raise it abroad. Political reasons ruled out increased taxation. In the end the Treasury gave way and reluctantly guaranteed

Contractors' politics 77 the local costs - an obligation which it spent the next three years ingeniously evading. However, the price exacted for its concessions was significant. The Overseas Ministry undertook to prune the total cost, to persuade South Africa to take a greater percentage of energy, and to quicken the schedule for the hypothetical north-bank power station. Though somewhat lightly accepted, these were to become major determinants of the future. The Ministry began by employing the technique of the second bid, which Zamco regarded as a distinctly low form of horse-trading. The second bid is not a new phenomenon in public contracts. (In October 1971 one of the largest road contracts ever offered in Portugal was re-opened after the Government declared that none of the three original bids was suitable, to the fury of the bidders who complained bitterly that 'this must seriously affect the credibility of the Portuguese Government in making future offers'.2) Nevertheless, it was unusual to re-open a contract after the preliminary award. But Zamco had allowed itself to appear over confident in the latter half of 1968, and the Overseas Ministry might have been tempted to use the continued American interest against it, even without the stimulus of the need to economize. There is no doubt that Morrison Knudsen, in collaboration with US Steel, made an offer in 1969 substantially lower than the original American tender. Both firms were feeling the effects of recession in their immense overseas trade, particularly that resulting from the contraction in war expenditure in Vietnam. It is difficult to judge to what extent, if at all, the Portuguese were dissatisfied with Zamco's performance, as opposed to its costs, and to what extent the Overseas Ministry was using American interest as a lever. Certainly, the negotiating commission disliked Zamco's loose, almost anarchic, organization and complained that though Esmond Brown was an admirable chairman, he was rarely seen in Lisbon. In April 1969 Jean Gonon, the head of CCI, was so disillusioned that he offered to withdraw. The Americans announced (prematurely) that they had complete financial cover; and Fereira, accepting the need in that case to work with Morrison Knudsen, arranged a joint meeting between Silva Cunha and Nicolas Kagan of Compadec in order to replace

CABORA BASSA Grand Travaux de Marseilles and Hochtief. The French electro-mechanical group, Cogelex, agreed; only the Germans were left in the dark. At the last moment during the ministerial conference Kagan was unable to fulfil the American promises, thereby confirming deep Portuguese suspicions and losing significant credit for the future. Afterwards Zamco settled down more easily: everyone trusted the others a little less, but also knew a great deal better who was creditable and who credit-worthy. The Portuguese themselves gained little, and the Treasury noted gloomily that the German revaluation of the mark would add substantially to their long-term costs. Meanwhile, the effects of the South African recession threatened more fundamental calculations. Escom had never been enthusiastic about power but had been prepared to accept it on the basis of a cost per unit of 0.3 cents, falling after 1990 to 0.2 cents (against its own selling price of 0.55 cents). But, in an attempt to amortize the increased cost of the dam, the Portuguese asked for more revenue. South Africa's power graph now showed a rate of consumption rising faster than had been predicted at the time of the original agreement, but dependent on the extent of cutback in Government spending. The most that Escom could offer was a flat payment of 0. 1 cent for every megawatt over the agreed total, which was not enough to meet the Portuguese Treasury's conditions. The Portuguese had already lost a year's revenue by delays, and they had now also to assuage Escom's anger at the exclusion of South African engineers from consultation over the specification of the dam. Neither side would give way on its demands. At the last moment, when the time schedule of the whole scheme was threatened, the South African Government intervened following a direct appeal from Nogueira. The drop in price from 0.3 to 0.2 cents was postponed from 1990 to the year 2000 and, of great importance in principle, Escom agreed to own the converter station and transmission line on South African soil. No repayments for this £60 million section would fall on Lisbon. In addition, the South African Government would have to accept repayment after the year 2000, when Escom would still be paying at 0.3 cents until, in some distant future, the 0. 1 cent difference had amortized not

Contractors' politics 79 only the cost of the converter station, but the accrued interest over thirty years - in practice, never. These concessions, the importance of which can be gauged simply by the fact that almost all other government expenditure in South Africa came under the axe in the period 1968-72, were made after two personal visits to Johannesburg by Calvet de Magalhaes and a great deal of lobbying by van Eck and Hilgard Muller. Throughout, the Prime Minister, B.J. Vorster, showed himself a disciple of Verwoerd, more open to arguments about diplomacy than about finance, and once again the Portuguese took advantage of South Africa's long search for political support. But they failed to persuade Escom to take more than 1920 megawatts in 1979 - a figure which included 75 megawatts for recirculation to Lourenqo Marques - and in the penalty clauses they conceded a highly important veto to South Africa on the supply of energy to any other potential consumer. Escom viewed with the utmost disfavour Portuguese attempts to offset vet more of the cost by offers of separate lines to Rhodesia or Malawi. A Portuguese subsidy to a black republic in which South Africa had a growing interest, out of a scheme made possible by South African money, was intolerable. And though Rhodesia expressed only the mildest interest in buying power, the South African Foreign Office had no wish to be further embroiled with sanctions-busting at a time when the Smith Government was holding its referendum on a republic. By conceding a South African veto, and by consenting to the severe penalty clauses,t the Portuguese finally admitted that South Africa had a major, if not necessarily an equal, interest in Cabora Bassa and that they had to accept on trust that Escom would eventually take a greater share of the total energy - just as they had already accepted that if South Africa made Cabora Bassa possible, Lisbon had to keep faith with the future of the Republic. The South African press, fed by judicious leaks, reflected some of this story. The first notice, given by Esmond Brown in August 1968, was clearly designed to take the edge off t For further amplification see Notes at end of chapter.

CABORA BASSA prejudice against external involvement: 'From the Portuguese point of view ... they derive very little benefit from the project until 1990 and yet they will bear the entire responsibility of guaranteeing the finances . . .' Many South African papers congratulated their Government on cornering so large a share at a small cost.4 But, as disagreements developed between Lisbon and Escom, there grew a rancorous suspicion, headlined in the Financial Mail: 'Who is going to pay for all that expensive electricity?'5 Through 1968-9 the well-informed Mail continued to suggest that Zamco would drop out, and scrutinized closely what little was known of the political talks. The main theme of attack, however, was the cost to South Africa, about which Pretoria stubbornly kept silent. The Mail contended, accurately enough, that the cost had risen above the level foreseen by Escom. 'So somewhere along the line the difference must be made up .... Right now Pretoria is saying nothing. Neither is Escom, except to stress that "It will not be Escom that is out of pocket." It begins to look like Jan Publiek who will be! '6 Secrecy was maintained in spite of these questions. Brown announced that South Africa's total loans were 165.2 million Rand, repayable over thirteen and a half years at 6.5 per cent and he explained that this much had been undertaken 'so that South Africa's skills and materials will be able to participate'.7 No mention was made of financing costs or of direct cover for Zamco, nor of the fact that the Government would pay for all installations within the Republic. Further to allay suspicion, Hilgard Muller unveiled a plan for future economic cooperation: Consideration has been given to the best means by which South Africa on its part can endeavour to stimulate the economic growth of southern Africa. Although, of course, South Africa cannot or will not compete with the big states in the field of capital investment, it is in fact better equipped to stimulate this development than almost any other country in the world.8 In September 1969 Zamco Limitada signed the final contract with the Portuguese Government for the first stage of Cabora Bassa - the dam and the south-bank power station - at

Contractors' politics 81 a fixed price (based on 1968 figures) of 8.8 million contos (E126 million, excluding the cost of financing and the installations in South Africa). During the ceremony Silva Cunha called the scheme 'the greatest Portuguese undertaking of all time', and was appluded by Nogueira, Esmond Brown, and Straszacker of Escom, who, finally converted to the grand design, cited the contract as . . . just another proof of how intimate the cooperation between Portugal and South Africa will be .... Cooperation with other neighbouring states, in fact good neighbourly relations, will provide the basis on which all countries of Southern Africa can participate in this essential commodity. .. I can foresee that all countries south of the Zambesi will soon have an inter- connected system which must give advantage to all concerned.9 Pious homilies which might have been drafted in the Foreign Office in Pretoria had barely left the government press in Lisbon when the Swedish partner, ASEA, announced its withdrawal from the contract - ostensibly because of fear that Zamco might infringe sanctions against Rhodesia and embroil it with the United Nations. There was some truth in the statement: Tage Erlander's Government refused to give ASEA an assurance of indemnity from prosecution. Of greater importance, the Wallenberg industrial and banking empire, of which ASEA was only part, had come under attack for its monopoly position in Sweden in much the same way as Axel Springer's publishing group was being criticized in Germany. When Erlander was replaced by the Socialist, Olaf Palme, in the 1969 election Wallenberg's political base suffered, and the campaign against him as a monopolist took on the added charge that 'ASEA supports Cabora Bassa and apartheid.' Since liberation movements in Africa, and Frelimo in particular, had become popular causes among students in Sweden, the case against Cabora Bassa was made with precision and force. Miguel Mrupa, Frelimo representative in New York, arrived to inspire the campaign, and pamphlets with stories of guerilla attacks on the site at Cabora Bassa were sedulously posted to ASEA employees. These were sensitive issues, and it is clear that ASEA's withdrawal was designed for home

CABORA BASSA consumption. Wallenberg was depicted in the Swedish press as the good citizen, respecter of international obligations. Other partners in Zamco were quick to suggest that he had preserved his monopoly by a single concession, and that he had been suitably rewarded for his behaviour by further government contracts in Sweden. According to Zamco directors, however, ASEA had already been ordered out, with South African approval, because of its inability to sort out these political problems. The French and Germans were simply not prepared to concede the Swedish Government's demand for a promise not to deal in any capacity with Rhodesia. Foreseeing the loss of ASEA, Fereira had approached General Electric, in case the German group was unable to perfect the thyristor system within the allotted year.t GEC's answer showed that its research into thyristors was at least as far advanced at that of the Siemens-AEG-Brown-Boveri group, with the advantage that it had constructed a pilot scheme in New Brunswick. Lisbon swarmed with vice- presidents and Kagan once more issued a glowing prospectus. The Germans fought back aggressively, seeking to undermine confidence in the American experiments and especially in the political will of the Nixon administration to provide export credits. Although GEC offered to undertake the converter station contract with a $55 million loan from the Export-Import Bank there were genuine doubts as to whether any such loan had been sanctioned. In any case, Congressman Bingham announced his intention to block it in Washington. Political and technical arguments mingled inextricably. The Germans had a right to remain in Zamco which they defended furiously. Escom showed great unwillingness either to forgo the German converter station at Apollo or to allow a situation in which German equipment would be installed at one end and American at the other. The technical decision, aided, ironically, by the consultants of English Electric, was made in favour of Germany in December 1970; but that did not end the matter. The HGU group had spent DM30 million on research and development for the contract. Though they had t For further amplification see Notes at end of chapter.

Contractors' politics 83 expected to be given priority after ASEA's withdrawal, they waited until completion of their tests before asking the German Government for credits to cover what was expected to cost at least DM140 million more. (ASEA's costs had been covered by the consortium of French banks and Union Acceptances, an arrangement which was not transferable.) When this request was made, after the 1969 election which broke the Christian Democrat monopoly in Germany and installed Willi Brandt as a Social Democrat Chancellor, it was received in a climate of opinion changed by the Lusaka Conference and Kaunda's European campaign. Brandt's Finance Minister, Karl Schiller, honoured the existing pledges but refused more. The three German private companies, lacking state assistance, were faced with a monstrous obligation. After defending themselves against the initial GEC bid, by making economies, their best price to the Portuguese Ministry allowed only a narrow margin of profit; now they saw this converted into a heavy loss - if indeed, they could hang on at all. True, the Paris-Bas consortium of banks was willing to lend the money, but at a 14 per cent interest rate; and the Portuguese seemed favourably inclined towards the alternative American offer. Determined not to lose the chance of competing in future high voltage transmission markets, but setting the cost of finance against that already spent on development, the three German companies finally turned to South Africa - where IDC had placed on ice the 7 million Rand credit originally intended for the Italians. It was suggested that if Escom really disliked the idea of GEC machinery at one end of the South African grid, and if it was satisfied by the German test programme, then this credit should be used to fill the gap. Thus, in effect, the South African authorities lent to a German company over the head of the German Government; and the gesture carried such weight that a consortium, headed by the Deutsche Bank, filled in the remaining $20 million. The Mannheim-Rheinau project was saved, but at a cost to the participants which cannot yet be evaluated, beyond saying that they were forced to borrow $30 million at 10 per cent, having already effectively lent it to the Portuguese Government at 61 per cent. In May 1971 GEC announced in the American press that

CABORA BASSA they had 'withdrawn' because of hostility from the US State Department t - a neat excuse when decrying the German achievement to future clients. More than five years after signing the original Heads of Agreement, the contractual basis of the Cabora Bassa scheme was settled; of these five years at least two were wholly wasted in political and financial altercation. The terminal date for supply to South Africa was held to 1 January 1975, at the cost of savage strain on the capacity of the Zamco consortium. Rarely in the history of public works contracting has the winning team had to survive such a passage. t For further amplification see Notes at end of chapter

Notes and references Chapter 3 Contractors politics t Note page 68. At the end of 1969, SAE had held $12 million of promissarv notes which should have been discounted in 1967, the result of accumulated delays on the part of IMCC. SAE had, in fact, completed $240 million worth of overseas work since 1950. Well aware that Portuguese notes for Cabora Bassa would not be discounted until this backlog had been cleared, SAE not unreasonably resented making, in effect, a loan to both governments. I For the technical scope of this project, see HVDC Transmission Report, HGU Schriften-reihe, ETZ-A, 1968, 1970. 2 The Times, 13 October 1971, reporting on results of a tender by British, Spanish, and Italian consortia for new toll highways between Lisbon, Oporto, and other centres. ± Note page 79. The agreement, signed in August 1969, provided (inter alia) that two monopolar lines would be erected, and that if one fell out of commission 80 per cent of the current had to be directed through the other, with earth return. If power fell below the 80 per cent load factor there was a penalty of three times the cost per megawatt hour if in peak load time, and of less if at any other time. Destruction by force majeure would have provided temporary relief from this penalty. 3 Johannesburg Financial Mail, 8 August 1968. 4 See Inside Industry, Johannesburg, 26 July 1968. 5 Financial Alail 15 August 1968. 6 Ibid. 7 Financial Ma l 26 September 1969. 8 Financial Mail. 26 September 1969. 9 Portuguese official record of the ceremony. f Note page 82. It would have been still possible, in theory, for the choice to have fallen on English Electric; but, in spite of the option held open by the Foreign Office the year before, a motion was signed by seventy Labour MPs early in December and Lord Shepherd, a junior Foreign Office minister, answering questions in the House of Lords, issued a stringent warning against participation. t Note page 84. According to the South African press (Financial Mail, 11 May 1971) their financial cover had been under consideration by the National Advisory Council on International Monetary and Financial Policy. It had not, then, been approved, as GEC alleged, by the Eximp Bank; GEC simply withdrew the application on 6 May. Their announcement of export credits seems to have been a purely speculative activity to give them cover in their attack on the Germans.

0

Part 2

4P, Jo 0 0 0 . _ LL tolo -, 0><0 0 -~ (D co * * 0> ~ g E c 9 u 10 cp .0 N .0 %EI N f 1 -M L co e E '0 c

Chapter 4 Preparation At Cabora Bassa almost every factor conspired to thwart construction. The arduous approach, the unpredictable nature of the floods, and the hostility of the climate, dictated a massive programme of preliminary work; while the inaccessibility of Songo, the absence of skilled local labour, the extent of underground excavation, and the dangers of guerilla attack, indicated, from the start, an abnormal level of risk. Some of the natural hazards could be forecast on the basis of geological surveys and past records of rainfall and flood. The climate was worse than in the uplands where Kariba had been sited, but bilharzia in the water, malaria, yellow fever, and typhoid could all be avoided by medicine and careful sanitation. The mean temperature at Songo in the dry season is 30-320, no worse than the Mediterranean in July, but down in the gorge it can rise another 10'C. Humidity is always high and the sun's rays, reflected off the polished granite of the cliffs, are vibrant and destructive. In the excavated chambers, half a mile into the rock, the air is heavy and oppressive, even on the clear days after the rains are over. Then the landscape is green and small clouds dot the sky like grapeshot fired from a cannon on the horizon. Later in the year, as a thin film of haze blots out the contours, the heat becomes intense. On certain days the temperature can rise to 60'C on the river, and all outdoor work ceases. October and November were known as the suicide months in Tete. The haze thickens in December, a few showers slanting out of the mist. Heavier storms follow, becoming steadier and finally almost continuous. High winds and sharp, cold nights

CABORA BASSA in January and February made the camp inhospitable, and cataracts of mud poured off the edge of the Songo plateau. Water courses, no more than dark smears on the rock in May, overflowed and the access road became a morass, liable to give way at the edges. Lorries slid and slithered down the slope and churned their way upwards with the aid of chains. Sometimes it rained so much that breathing became difficult. The tunnels filled three feet deep in spite of pumps, and the crash of rain on corrugated iron drowned even the noise of drilling. But the worst rains last only two months and there were greater problems. At the time of the preliminary adjudication the GPZ established a small cantonment of tin huts by the airstrip at Songo, and rather grander offices in Tete that began to rival those of the provincial administration. The GPZ executive was made up of three small work forces, one (GTZ) concerned with the construction of roads, the others with agricultural improvement and the resettlement of the people of the valley whose land was to be flooded by the dam. The GTZ had been given priority in the use of earth-moving machinery in order to carve out the first access road in time for the preliminary adjudication in May 1968. The gorge had to be opened up to bring in the largest machinery, bulldozers, diggers, rock drills and lorries, some weighing up to thirty tons. In six months the GTZ dug and banked a sort of road down the vertiginous scree to the 230 metre level - thirty feet above the flood-water mark. But it was a road in name only, continually menaced by falling boulders and landslides; and for months after completion the journey down by lorry took nearly five hours. Landrovers and the tough Peugeot wagons made it in half the time, but in the rainy season no-one could guarantee that cracks and wash-outs would not appear between descent and return. Gradually the contractors themselves widened the track and built it out by tipping rubble so that the bends became easier to negotiate; and they carted away some of the monstrous overhang of outcropping rocks to prevent them sliding through the repair sheds and fuel dumps below. But as late as the end of 1971, when five hundred lorries a day were using it, the road slipped again and endangered the excavations below. Whatever the engineers said, the mountainside

Preparation 91 remained unstable, a hazard greater than all the risks of skidding or backing over the edge. No-one then or later cared to speculate on what would have happened if guerillas had mined the base of the long line of granite cliffs which overlook the whole of the works at a height of three thousand feet. Giant Caterpillar excavators could navigate the hairpin bends, but it was useless to imagine that the turbines, generators, and transformers - loads ranging up to 190 tons gross, on multi-wheeled articulated bogies - or the steel gates for the sluices and penstocks, could by brought down that way. A second road, six metres wide, by-passing the Songo plateau, was projected from a point on the new Tete- Songo highway round the western end of the massif and down the much longer, well-wooded Mukangadzi valley to break through to the gorge at a point five hundred feet lower than the first. It could then run straight down, at a gradient of roughly one in three, to the tunnels leading to the main chambers inside the rock. But this road had to be cut into solid granite, and in some places, where buttresses jutted out, bulldozers would work on the sheer edge 2500 feet above the river. When Zamco took over the works in August 1968 it was no more than a line on the map. The Ministry had promised only a supply line to the river. Under its contract, Zamco was responsible not only for the new access road but also for remaking the 135 kilometres between Tete and Songo by the end of 1971, with a tarmac surface and sixteen new bridges to carry the fleet of bulk cement lorries for the next three years. The Mozambique Government, now headed by Arantes e Oliveira, had already doubled the existing schedule of road and rail building. Many of the priorities of the Plan Geral were re-ordered, concentrating effort on tarring the road from Beira to Umtali and Tete and on completing the rail link between Vila Cabral and Blantyre, capital of Malawi, in order to build up the trade of the deep- water port of Nacala and tap the bauxite trade which they expected to follow from the supply of energy from Cabora Bassa. But the old-fashioned Public Works Department, geared to piecemeal improvement over decades, ran out of contracting capacity, and the work was finally sub-contracted to a hybrid

TANZANIA 0 Palma *Mocimboa Mueda da Praia ZAMBIA Macomia Maniamba Porto V. Cabral V*Ameljia MarrupaMontepuez uNova Fre~x Nacala Furancungo v. Coutinho * Entre-Rios Fingoe 1 9CABORA apl oaniu Zumnbo * BASSA NmuaMcmuu zarnezi ilane /A*t.Molocue~ Nametil Tete u. Antonio Enes Z *Mutarara Mungari Donama Pebane ______V Fontes~ ______RHODESIA ~'Gouveia M Mpeia Quelmane V Paivade Andrada V. de Manica'9 Canxi 0 V Perya i Gorongosa G R. Nova Luzitania BEIRA______Nova Sotala INDIAN__save ______OCEAN ______* Inhassoro Massangena Malvernia Mabote 3 ~ Funhalouro GAZA Homoine ... Massingir * Inhambane Canicado Panda Inharrime ______7kp Magude V.d ooBi .LOURENCO MARQUES ______SWZILAND ______* Bela Vista--~ NATAL______

Preparation group, Enmoc, led by the French. Delays, arguments over the price, and over what currency it was to be paid in - since it came under the Mozambique budget - all held up construction, so that the road was still not tarred by the critical date of May 1971, when the first guerilla mines exploded. Of three separate routes from the sea to Tete one at least had to be capable of taking 190-ton loads. The railway from Beira harbour passed the cement works at Dondo, but the company had no special rolling stock, and had to limit its schedules of delivery because its facilities also served the Rhodesian and Zambian railways. The most serious problem, however, was the absence of a bridge across the Zambesi at Tete. It was easy to build a road from the Moatize rail-head to the river, but the existing pontoons could not take loads of more than sixty tons gross; and, quite apart from the heavy machinery, a bottle-neck would develop with the cement traffic. Hence the importance of the work of Professor Cardoso, a civil engineer then constructing a five-span suspension bridge at Tete. Unfortunately, at the time of the preliminary adjudication completion of the bridge had just been postponed by a year, and it soon became embarrassingly clear that the local administration would be fortunate to finish it before the end of 1971, when bulk cement traffic would already have begun. Moreover, Cardoso had designed the. bridge for a maximum load of a hundred tons. When pressed by the Germans, who were responsible for the transport of the machinery, he conceded that greater loads could pass on thirty-two-wheeled bogies - but that he reserved the right to inspect the bridge after each passage and, if necessary, to forbid other loads. Given the tight schedule imposed by the flood, the Germans found his assurances unattractive. Alternatively, they could have taken the Vila Pery-Changara road; but, as the German survey team discovered, its bridges were limited to fifty-ton loads. Most were due to be rebuilt under the provincial road programme, but that across the wide alluvial bed of the Luenha would be a major undertaking. As a possible third choice, permission was obtained from the Rhodesian Minister of Transport to deliver via Salisbury, but since this could be held to infringe sanctions it was kept strictly secret.

CABORA BASSA The contractors had also to cope with the low capacity of the Beira harbour cranes and with the fact that the docks had been designed to unload to the railways and not to a road transport system. The port authorities, unused to demands for precise schedules and speedy transit, spoke vaguely of buying a floating crane, but nothing materialized. Afraid that the railway monopoly and the inadequacies of the port would dictate its priorities, Zamco tried to negotiate for a free choice, but all it gained was experience in the workings of old colonial bureaucracy. The clash between Lisbon, concerned for the time-schedule of Cabora Bassa, and the provincial administration of Lourenqo Marques, avid for taxes on foreign traffic and suspicious of customs exemptions, forced Zamco to rely on ad hoc decisions and on the facilities offered by South African transport concerns. In spite of administrative and geographical constraints, Portuguese insistence on the commissioning date of January 1975 dictated that the Zambesi should be diverted during the dry seasons of 1970 and 1971, so that the twin coffer dams could be built before the 1971-72 flood, and the main wall raised in the dry seasons of 1972 and 1973, and finished, and the lake filled up, in 1974. Since diversion was critical to the whole programme, one tunnel large enough to take the normal flow of the river had to be completed during 1970 and the second, for most of the storm water, in 1971. Work had to start in March 1969 immediately after the floods subsided. Yet the access road was scarcely in a condition to survive the rains at the end of 1968 - and the main contract had still not been signed. The annual flood of the Zambesi thus controlled every stage of the building of Cabora Bassa. For nine months of the year the river runs through the gorge at 200-207 metres above sea level and varies in depth from 30-37 metres. Because the bed is V-shaped in section, a small rise in level produces a disproportionate increase in flow. The average is 1000-3000 cubic metres a second. In January, following the onset of the rains in the vast hinterland of the Angolan and Congolese highlands, and swollen by the waters of the Luangwa from Zambia, the river starts rising, reaching a peak level of around 220 metres in February, when the flow is 10,000 cubic metres

Preparation 95 a second. A second, shorter, flood usually comes in early March, after which the river subsides rapidly. In recent years this second peak has reached a level similar to the first because of the discharge from Kariba, the effects of which take six days to reach Cabora Bassa. These are 'normal' calculations. Some idea of the forces involved can be got from comparisons with European rivers: Cabora Bassa xvill eventually discharge at peak flood through eight sluices and five penstocks, and the volume through each aperture will be equal to that of the Danube where it leaves Germany at Passau, or of the Thames at Teddington Lock. In 1858 Livingstone thought that the high- water mark was eighty feet above the low; in 'catastrophic years' the river may rise to 225 metres. With a basin of 1,200,000 square kilometres, when the average increase in flow is ten times that of the dry season, it does not take an exceptional difference in level to produce catastrophic results. Accurate records only go back to 1935; but there are photographs of 1926 which show nearly the whole town of Tete under water. Engineers talk of extraordinary floods which statistically are likely to come once in a century, yet which seem to strike dams half completed, as happened at Kariba, nearly breaking the main wall in 1958. The architects of Cabora Bassa had planned against such risks, but the French found it wise to run through every imaginable condition on model tests under the supervision of Coyne-Bellier, civil engineering consultants in Paris. Given the narrow margin of time allotted for construction of the dam wall, it was never possible to consider diverting the whole flood. The Hidro-Technica's designs provided for a south-bank tunnel, 500 metres long, to take the river in the dry season, complemented by a smaller northern tunnel to take part of the flood. At peak periods the rest of the flood water would flow over the coffer dams in the first two seasons after they were built, and also through gaps in the main wall in the third. The slightest error in the designer's calculations could, in such conditions, cause the underground works to be flooded or let the river sweep away the coffer dams or the main wall itself. No-one had previously taken the deliberate risk of overtopping in such a situation. A certain measure of security was introduced by the regu-

CABORA BASSA lation of the Zambesi at the Kariba dam, the sluice gates of which could be operated with an eye to the effect of the discharge on works down-river. If the level of the Kariba lake rose dangerously, the Central African Power Corporation would have to spill water, but if the lake were lowered to the minimum in January, then the expected peaks and the worst of a 'catastrophic flood' could be avoided. Cooperation with Rhodesia was very close: in the contract the Portuguese Government had promised to 'take steps to ensure that the spilling programme at Kariba will meet, in part, the demands of the working plans'.' Zambia, second partner in the CAPC, might not be so helpful, but since all the spillways were then controlled from the south bank, her only means of threatening Cabora Bassa was through a new dam on the Kafue, wholly in Zambian territory. This dam, then being built by the Yugoslav consortium ZECCO, could, at worst, discharge a tidal wave of 6000 cumecs for a brief time. Prudently, Zamco tests allowed for this unlikely political contingency. If any of these stages failed to meet its deadline, or if the preliminary work broke down, a year would certainly be lost. Tunnelling therefore began on the south bank as soon as the access road was finished. The quality of the gneiss and granite, though better than that of similar schemes such as the Churchill Falls, was poorer than the early drillings had led all to expect. Movements of the earth's crust had left the rock flawed and fissured; and throughout 1969 the drillers pushed outwards from the centre shaft towards the channel of the river, knowing that the slightest hitch could be multiplied in a geometric progression of costs and delays. The foundations of the dam would not be dug until the dry season of 1972, more than half the way through. At Cabora Bassa, unlike most dams, the simplest and quickest part of the contract was the building of the main wall. Preparation took up three quarters of all the work involved and the contractors had to find the skilled labour, black and white, house and supply them, bring in the cement and steel for construction, and yet keep within costs which had been worked out in a European framework and with limited knowledge of African conditions. The dam wall is 160 metres high and only 303 metres wide, compared with Kariba which is 128 metres high and 617

Preparation 97 metres wide. Cabora Bassa lake is roughly the same length as Kariba, 250 kilometres, and only half the volume. But the extra height and the greater volume of the Zambesi at Cabora Bassa allow for a potential energy supply from two power stations which is nearIy four times that of the potential supply from the two stations at Kariba. The generating sets are among the half-dozen largest in existence and required a prodigious amount of space within the underground power station. Consequently, the amount of excavation was colossal. Digging the foundations for the dam wall necessitated the removal of 210,000 cubic metres from the river-bed. The main underground chamber, which was to house five turbines and generators, measured 60 metres high, 28 metres wide and 210 metres long - three times the size of the Siemens multi-storey headquarters in Erlangen, or twice the size of St Paul's cathedral - and needed 300,000 cubic metres of rock cut out, the same as for the main chamber at Churchill Falls, which houses eleven sets, not five. Excavation of the transformer and surge chambers added nearly as much again, to say nothing of the diversion tunnels; and all this 21, million tons of rock had to be shifted out through three tunnels converging onto the single access road down the mountain side. (A million tons was crushed for mixing into concrete.) The sheer volume of excavation on so narrow a site necessitated a complex schedule of work priorities laid down and operated by experts in industrial processing. Among the engineers of Zamco, the absence of a project manager soon appeared a grave lack of foresight. Three different white nationalities, French, German, and South African, had to work together, while cooperating under the direction of Portuguese engineers. Portuguese remained the official language of business; Fanagalo, the lingua franca of the South African mines, was the only means of communication with the largely African labour force. Moreover, the constricted site made the work capital intensive.Whereas at Kariba it had been possible to employ 6000 Africans at the peak, here at Cabora Bassa 1300 was the maximum (750 whites were employed at both). Thus the demand was for highly skilled men, capable of drilling, blasting, driving, pouring concrete, and fixing steel reinforcements, to the standards normally expected in Europe. There

CABORA BASSA was little time to train men in new skills and no room for mistakes; and this inevitably led Zamco to put South Africans in charge who had worked with black labour before. The original intention of the Overseas Ministry was that the Portuguese Directorate would ensure conformity with the plans and standards laid down by the Hidro-Technica and that any modifications would be submitted by the contractors for approval- a model laid down in the Plan Geral for the relationship between private enterprise and the state. Practice was not so simple, for the Directorate itself lacked authority. All decisions had to be referred first to the GPZ in Tete and then to Lisbon. Caught frequently in a scramble for resources with other sections of the GPZ, the Directorate only gradually established the rule that Cabora Bassa took priority over the remaining developments of the Zambesi area; and its ill-defined relationship with the provincial authorities in Lourenqo Marques, who remained suspicious about the influx of so many foreigners, added to the contractors' mistrust of an apparently all-entangling local bureaucracy. Some months passed before the contractors realized that the Directorate, and especially its head, Braz d'Oliveira, was in fact their closest ally. The problem appeared at first to be one of language barriers, but underneath the differences were of temperament and personality. The Portuguese had worked in isolation for decades (with the exception of their joint hydro-electric schemes in Spain) and there were substantive doubts as to whether they could operate in the same rhythm as South Africans. Nor were their engineers' assessments necessarily equivalent to what Zamco had understood at the time of the contract. Complaints, for example, that the GPZ had overrated the quality of the rock for lack of sufficient trial borings, were rebutted by arguments that all information had been supplied 'errors and omissions excepted' - which was no more calculated to endear the Directorate's methods to Zamco than the advice that, if bridges needed reinforcing, Zamco should do its own work. Such uncertainties went far beyond administrative argument. Normally in great public works the Ministry states quite precisely the limits of authority and responsibility. Cabora Bassa, however, mirrored some of the ambiguous relationship

Preparation 99 between Portugal and her political partners. The attempt to preserve complete authority over companies which were doing what Portuguese resources could not (and in some cases, such as that of the Tete-Songo road, what Portuguese companies had promised but failed to do) could never be wholly successful. Mutual tolerance on both sides was the only way to avoid disputes. The siting of the converter station, to take one case, was changed during the course of negotiations in 1969. Who then was to build the road to the new site - a matter of £250,000 worth of highway engineering? The French declined responsibility; the Germans declared that, being on the electro-mechanical side, they never built access roads; the Portuguese replied that the scheme was a 'Turnkey' project and if Zamco altered it, they must accept the consequences. ASEA proposed that they sort it out later; but by the time the road came to be built they had withdrawn. Extra works and alterations of this sort tend to make for expensive litigation in the end, and though accurate, it was not a complete defence to say that Zamco should have sent out better survey teams before submitting their tender. Wide changes were made during 1969 before the contract was actually signed, and as late as 1970. Because of the ramifications of even minor alterations once the schedules were in operation, Zamco consistently sought a free hand which Portuguese engineers were most unwilling to concede. Over Zamco's internal organization the Directorate discreetly avoided confrontation. Practical considerations dictated the division of functions: since the South Africans were responsible for the camp and all questions of African labour, and the Germans for the electro-mechanical side, the French naturally ran the civil engineering and the vast bulk of preparatory work. The Italians isolated themselves quite deliberately and did not begin to survey the route for the transmission line until October 1970. Zamco installed itself at Songo in a camp two miles north of the airstrip and the Directorate's cantonment at the end of 1968. The main township was not begun for another year, but the work force had to be assembled and the supply lines opened up immediately. White labour was relatively easy to recruit. The companies brought in their senior staff - Bill Chambers, an experienced Anglo-American executive as chief administrator, and Alexandre Chappelle of GTM to run the civil engineering side. LTA, Hochtief, Shaftsinkers, and the other French companies, sent in the remainder of the executive staff. Local recruiting in Europe and South Africa supplied the rest. The companies deliberately chose married men where they could, promising houses for families later, reckoning that a three to four year stint in the wilds would drive most bachelors to drink or worse. Like the mysterious signals which draw vultures out of an empty sky towards a carcase, messages circulated among the ranks of roving engineers and mechanics and a polyglot assembly arrived in Tete - Rhodesians, Belgians, Australians, Brazilians, as well as 'Zamco nationals'. A diesel engineer drove up from Beira after twenty-five years in the Argentine, without thinking of visiting his native Trieste on the way; Franz Jordan of Hochtief transferred direct from Malaya, and many Frenchmen, including the second-incommand, Roger Br~bant, arrived from the dams of West Pakistan. The transport manager turned out to be an Englishman living in Rhodesia, who had spent ten years piloting paddlesteamers up the Niger and the Benue; and there arrived probably the only German engineer who had ever played Rugby football for Cardiff. The staff of the Directorate was almost as used to remote corners of the globe. Hidro-Technica engineers found it as natural to work on the Zambesi as on the Tagus or in Macau, Timor, or Guinea. Since the contract imposed a certain ratio of Portuguese to other whites (which increased each year in order to provide local training and employment), the Zamco administration set up recruiting offices in Lisbon and Lourenqo Marques. At first these Portuguese were given subordinate jobs, but by 1970 they numbered 600 as opposed to 150 other white nationals; the latter figure fell to 100 in 1972, and to only 60 in 1974. Zamco had to pay well to get anyone out to Cabora Bassa - thirteen or fourteen months' pay for the year as bonus, plus help with the families' fares. Part of the salaries were paid in Europe in currency of the men's own choice. Good money could be made here; the danger was that gambling and substitutes for good living would drain it first. 100 CABORA BASSA

Preparation 101 For political and financial reasons, Zamco had been put together out of disparate nationalities, only one of which habitually worked with Africans or spoke African dialects. So African workmen were chosen and managed entirely by Chambers - a pragmatic decision which took for granted also that it would be easier for apartheid-conscious South Africans to acclimatize to multi-racial society in Mozambique if a South African was in charge. The scheme worked well: Chambers brought in Bill Smith, an associate of many years' experience in the Rand mines, and insisted on hiring for the skilled work only men who had previous mining experience. The majority were Shangaan from the province of Inhambane, southern Mozambique, which annually supplied 125,000 men to South Africa under the contract labour agreement of 1928. The Shangaan are, in South African terminology, 'Good Bantu - clean, proud and well motivated.' Later on a few Malawians came in, rather than make a long journey to the Rand, and some labourers came from Zambia until prohibited for security reasons. For unskilled work, Smith turned to the local tribesmen, Angoni and Nyungwe, and the Songo locals, called Mademe. In spite of the accusations made by Frelimo guerillas, it is clear that only voluntary labour was employed. In the nature of the work, coercion would have been impossible, quite apart from ethical and legal considerations.* The Shangaan earned good wages in South Africa, even though part was paid in Mozambique on their return, and Zamco had not only to compete in wages but to contend with the traditional pattern of migration. (For a description of the contract labour system, see page 131.) Zamco needed 1200 men a year, either fresh recruits or on re- engagement, and it had to offer a better deal than Chambers and Smith had done when they worked on the Rand. Two recruiting offices were set up in Inhambane and the word was sent round to tribal chiefs in the villages. Good use was made of local civil servants and the offices of shipping agents to encourage men to go north; and money was spread out liberally as an enticement. Yet, in the early days, the response was not good. Pride controlled the response as much as piece rates. Work in the deep mines was hazardous and * This is not to deny that forced labour did exist in Mozambique, as late as 1955, before abolition of the assimilado system. could serve as an industrial substitute for the ancient tribal manhood tests. The fact that many Shangaan did not marry until they had completed at least one contract, owed as much to social pressure as to economic necessity. The man who wields a rock drill or sets the dynamite stands at the head of a hierarchy, just as the coal face cutter does in South Wales. It was more than a macabre coincidence that after a premature explosion early in 1970, in which seven Africans were killed, the standards of recruiting improved. Word went round the kraals: Cabora Bassa was a test of maturity and experience by which in old age men could measure their lives. So they came,a thousand kilometres in the recruiting bus, or by boat, a fortnight's journey from Lourenqo Marques. Chambers set out to recreate exactly the style and atmosphere of a South African mining camp, within the rules laid down by the Mozambique Institute of Labour. Galvanized huts, six men to a room in bunk beds, were set beside a stream two miles below Songo camp on the way down to the gorge - segregated in practice from the life of the white community. The men could use the facilities, the shop, and the canteen there, but they had their own stores and films and a ration of Bantu beer, the sweet, mildly alcoholic drink fermented from ground millet. Families had to remain behind; the contract labour system enforced the same migratory habits as in the Bantustans. Only the locals could set up wooden huts for wives in the hillside scrub and introduce them, shyly and awkwardly at first, to the self-service stores at Songo. The Shangaan's contracts ended after twelve months, allowing either for a six months' extension or a passage home. Each was given a card providing for re-engagement at the same wage, with a bonus if he returned within three months, and the card was marked to show how much, if at all, Zamco wanted him back. Rotation to preserve family life was part of Portuguese labour law, as paternalistic but much less harsh than South African restrictions. The labour law's safety regulations were written into the contract; and Zamco covered all men with insurance. Wages were good by the standards of southern Africa, and for Africans who rose to work at the same level as whites, the rates were the same. In practice, again, the two 102 CABORA BASSA

Preparation 103 were segregated economically. The majority of blacks began, in 1969, at a basic rate of 600 escudos a month, slightly more than the South African equivalent. (The South African basic rate was them 8.06 Rand a month, but the twenty-six day month there made no allowance for sickness or injury, so that the Mozambique rate could be substantially higher.) Most, however, would reckon to get 900 escudos with overtime (£11 a month based on a 1969 exchange rate of 80 escudos = £1). Skilled men worked on hourly rates - carpenters at 10 escudos, drivers at 20 escudos - and, given at least an hour of overtime a day, some of these could earn 6,000 escudos a month. The highest-paid drillers or drivers of heavy plant might collect 12,000 escudos or £150 per month - a good wage, since it included food, housing and equipment. Of course black Africans did the heaviest work and, since no man works simply for love or pride, it was the shrewdly judged attraction of relative comfort and good pay which brought them in. No black engineer applied, no native African held a supervisory post, and European journalists visiting the camp asked cynically why South Africans needed apartheid when the Portuguese achieved it painlessly under the cloak of multi-racialism. Nevertheless, all this work in a Western country would have been done by white men; mechanization at Cabora Bassa was extended to the limit and native labour was not maltreated. South African companies dominated the supply line, just as Afrikaans managers ran the labour force. All of the 'heavies', the drills, bulldozers, and diggers, had to come in from Europe or America, from Caterpillar (through the Portuguese agents Steia), or Ingersoll Rand. But the supply of cement and steel, constituent elements of the dam wall and the linings of every part of the underground works, posed a choice between economy and politics, the results of which almost crippled Zamco's calculations. (661,000 cubic yards of cement were needed for the dam wall and 7,900 tons of reinforcing steel.) Zamco had worked out a tender on prices submitted by the giant transport firm, United Transport (which covers most of central and southern Africa), the bulk of whose cement and steel would have come from Rhodesia. (Concassa's price had been higher, partly because its port of supply was Mombasa, in accordance with the British Government's sanctions against Rhodesia. From the table of pricest it can be seen that the savings in transport costs, given a demand for half a million tons of cement, would be over £ 1 million; this economy might well have been doubled by Rhodesia's willingness to cut the cost of materials in order to earn scarce foreign exchange. Rhodesia's eagerness did not stem only from the difficulties of the Smith regime after three years of fragile independence. The main supplier, United Portland Cement of Salisbury, was owned by United Transport which, in turn, was 45 per cent owned by Anglo-American, which also had a stake in Riscor (Rhodesian Iron and Steel Corporation), the main supplier of steel. Supplies could be delivered in bulk by road at Tete or Songo, on the right side of the river, saving at least two of the handling operations of the sea-rail-road route. In 1968 the Rhodesian Government put up five million Rhodesian dollars as a subsidy to suppliers and began to tar the road north-east from Salisbury. According to a Financial Mail report in July, they were looking forward to picking fat plums in Mozambique, including exploitation of the coal and mineral deposits. They were defeated by Portugal's awareness of the United Nations' sanctions lobby, and by a new and powerful group of Portuguese industrialists in Mozambique. At the end of 1968, the Overseas Ministry began its prolonged flirtation with the contractors of the American group and this, together with the withdrawal of ASEA and the swell of hostile opinion in Europe and at the UN, made them unwilling to risk giving gratuitous offence by buying so extensively from Rhodesia. But the influence of the Champalimaud group, and the patent advantage in foreign exchange for Mozambique if local suppliers won the contract, seems to have been more important. Antonio Champalimaud was already known as one of the foremost overseas entrepreneurs in Portugal. He controlled the first conglomerate in Mozambique, with £15 million worth of investment in cement, chemicals, fertilizers, insurance, banking, and steel, by 1968. The only serious rival of the great Compania Unijo Fabril (CUF) complex in Lisbon, Cham- f For further amplification see Notes at the end of chapter. CABORA BASSA 104

Preparation 105 palimaud also wielded substantial political influence through the Banco Pinto e Sotto Mayor, a fast expanding and profitable commercial undertaking which ranked second in the overseas territories. He had also established a bridge-head in Malawi through the Commercial Bank, and was closely associated with the shadowy figure of Jorge Jardim, eminencegrise of Portuguese diplomacy in black Africa. Who better, then, to provide cement from his new factory at Dondo, on the Beira railway, and steel from CIFEL, the only large foundry in the province? In March 1969 the Portuguese Consul in Salisbury, Dr Freitas Cruz, warned the Rhodesian Government that local suppliers would be preferred,t and that the Mozambique administration would not be prepared to waive the stiff tariff barrier. In the end, Rhodesia lost very little; as liquidity built up in the middle years of UDI, investors turned inwards and produced a monstrous burst of speculative building in Salisbury and the large towns. This soon swallowed up the surplus of cement, while renewal of the railways took the steel. But the French contractors at Cabora Bassa were hard hit, having reckoned on the original price quotations. They complained, forcefully, in Lisbon and extracted from Champalimaud a guarantee both of price and equivalent quality - the source of much trouble later. Thus Portuguese industrialists benefited, and Mozambicans gained employment. Only foodstuffs came in from Rhodesia in any quantity. But United Transport took all the main supply contracts: lorries of their Cabora Bassa Transport carried the cement in bags, and twelve special conveyers took the bulk cement in twenty- three-ton loads from Moatize rail-head to the storage silos above the gorge; another subsidiary, Thornton Transportation, brought South African goods 1,770 kilometres from Johannesburg via Salisbury; and a third group, United Transport of Tete, covered the run up from Beira. Cory, Mann George, its shipping agents, handled everything for Zamco at Beira and Lourenqo Marques, and Republic Framework provided the structural steel, including the framework for the ceiling of the caverns, the largest ever built in South Africa. The Anglo-American Corporation held Cabora Bassa in both its wide hands. t For further amplification see Notes at the end of chapter.

106 CABORABASSA Notes and references Chapter 4 Preparation I Contract specifications Vol. Al (Civil). Kariba maintained radio contact with Songo. Note page 104 United Transport (1968) Prices in escudos per ton through Beira through Salisbury Cement in bulk 606 450 Cement in bags 564 385 Steel reinforcing 728 672 Constructional steel 938 672 Timber 691 672 Refrigerated goods 1,496 1,000 t Note page 105. A curious insight into the politics of African commerce was given in mid- 1969 when the Zambian Government-run monopoly, Chilange Cement Ltd, blandly put in a tender for supplying Cabora Bassa. Being a commercial bid, it was taken seriously, but in September the Zambian Government realized what its agents had done and blocked the offer immediately.

Chapter 5 Building Few men within the contractors' organization were able to see Cabora Bassa as a whole. Had it been built by a single firm, planning would have been in the hands of a team led by a project manager. Instead, only the Portuguese had a synoptic vision of the contract. Differences of language, technique, methods of management, and variations in currencies and technical specifications, handicapped Zamco, while the managers of its component companies became increasingly aware of unrelated and sometimes conflicting responsibilities to the client, the consortium, their own governments, and the shareholders. Cabora Bassa was only a single contract, albeit an important one. The efforts, even sacrifices, made to win it could not necessarily be justified later in the light of further compromises made to avoid, say, competition between Siemens and AEG or Cogelex in other engineering fields, particularly as inflation of the cost of materials and wages entrenched deeply into the profit margins originally calculated in 1968. Control of the project could neither be maintained effectively by the boards of member companies in Europe, nor by the negotiating group or its successors in Lisbon. Satisfactory as an instrument for resolving major crises, or for confronting the Overseas Ministry, the group of representatives lacked continuity. Company policies and personnel changed radically during the six years. After the signature of the contract, negotiation was reduced to matters of horse-trading within Zamco, and claims for extra work and inflation against the client. Regulated strictly by the annual flood of the Zambesi, progress had to be an exercise in close planning of schedules of work and supply of materials, in which cooperation rated more highly than individual advantage. Despite this overriding necessity local managers found it hard to transcend national and company interests. Consequently, the French, who had to start first on the site, asked for a project manager. Even in the context of the delays of 1969 and the brief incursion of Cabora Bassa Builders, some jealousy had still to be overcome, and the successful candidate, like the chairman E.T.S. Brown, turned out to be an outsider. Bernard Bendixen was born a German, but his engineering experience had been largely in Latin America, and at the time of appointment he was an American citizen, fluent in Portuguese, and the author of several specialized articles on project management. Bendixen was appointed early in 1970. Months were to pass before he could establish his authority and he never wholly overcame the handicap of late arrival, although he did eventually introduce sound management techniques, while retaining the trust of fifteen separate companies and the Portuguese. Most important, he was able at a high level to bridge the gap between the groups at Cabora Bassa, where the problem of nationalities intruded every time work or costings were delimited. The original allocations had been made simply on the basis of fair shares and bore no relation to competence or to local difficulties. Hochtief, a civil construction company even larger than GTM, had a smaller part than the French and was regarded by them as a sleeping partner. LTA, a then unprofitable and relatively small subsidiary of Anglo-American, and the expert rock-drilling concern, Shaftsinkers, held one third of the work - but also under the jurisdiction of the French. The assignment of jobs within the Zamco federation, which had begun a year before Bendixen's arrival, proved no guarantee of the integration and teamwork necessary to fulfil Cabora Bassa's restrictive deadlines. Perfect planning of a dam in a normal situation should leave nothing to chance, be it weather or Act of God; but up in Tete both these and mundane things like transport and supply were rarely predictable, to say nothing of the disruption caused by war. However well the team leaders, Chapelle, Chambers, and later Horst Lange (of Siemens) on the electro-mechanical side, collaborated, CABORA BASSA 108

Buihing 109 there was a tendency at first to apportion the blame for breakdowns on Mozambique cement, South African transport, Portuguese bureaucracy, French deviousness, or German inexperience in overseas work. South Africans ran down French rock-drilling, Europeans complained of South African arrogance and short temper. Both blamed the Portuguese and the only concern of which all spoke well was SAE of Milan, which did not start work until 1970 and with which scarcely anyone came into contact. On the site itself there was a separate problem of discipline. In the early days of isolation and poor facilities, when the easiest relief was heavy drinking, fighting frequently broke out, and usually on national lines. After a particularly violent fracas in which the local police intervened, two or three engineers were deported, and the rest concluded that it was unwise to assault the Portuguese. Chambers made it clear that jobs depended on strict observance of the law, and with the arrival of wives and the building of a club wildness diminished; occasionally, however, accumulated tensions erupted into blows, as in the case of the Austrian engineer who turned on his Italian superior with a broken bottle, after weeks of suppressed anger in the dry season. Chapelle deliberately shifted his men around from job to job to avoid excessive routine, but clashes of temperament and methods of work could only be diminished over time and with the experience of common endeavour. By 1972 the French were prepared to acknowledge Shaftsinkers' expertise at rock-drilling, whilst the most arrogant Afrikaner had come to accept the virtues of French management direction. The Germans who arrived afterwards to install the electro- mechanical machinery benefited from four years of enforced learning. But one. administrative observer concluded that cooperation only became possible because both sides had to work with African labour and under Portuguese command. In civil engineering projects the client always becomes something of a scapegoat, or at least a means by which to rationalize cumulative errors. After the original negotiations Zamco representatives did not penetrate far beyond the fringe of officialdom in Lisbon, and 'Lisbon' became a stereotype for a bureaucracy insensitive to their needs, yet insistent on prompt delivery. Insofar as Zamco needed assistance from the provincial government in Lourenqo Marques, its chances inproved with the arrival as Governor-General of Arantes e Oliveira, formerly chief negotiator in Lisbon - a sign of the increasing sensitivity of metropolitan authority to the impact of Cabora Bassa on Mozambique. But in Tete there were, for a long time, conflicts of jurisdiction. In practice, Zamco looked to the GPZ, representative in Tete of the Overseas Ministry, which, until the civil governor was replaced by a brigadier in mid- 1971, remained well-integrated with the local administration. Yet even within the confines of the provincial town diversity made for slow communications. The GPZ was divided into SREP (Serviqos Regionais de Estudo y Planeamento), covering the Zambesi basin as a whole, under Crispin de Sousa, and the Directorate at Songo, responsible for Cabora Bassa. For many local officials Zamco constituted a disturbance of accepted methods, and the task of Braz d'Oliveira was made doubly difficult by the need to coerce his own colleagues in Tete, or his superiors in Lisbon, as well as Zamco, in order to achieve the greater aim of finishing the dam on time. At Songo the Portuguese appeared aloof: their encampment, a rash of small, unpretentious tin huts near the airstrip (only much later expanded into the nucleus of the future town), was built nearly two miles away from Zamco headquarters. Their officials and engineers took care to deal correctly but formally with the foreigners; they rarely visited socially, and only slowly 'unbent' in personal relations. Their role was ambiguous, for in addition to the function of supervising the engineering of the dam, the Directorate dealt with all questions of personnel, supplies, and administration, pertinent to the Portuguese and Mozambique Governments. In a more open system of government this would scarcely have mattered, but in a system where all commerce and communications were rigidly controlled the burden could be serious. If Zamco needed light aircraft, extra labour or tinned beef, the request would be sent to departments whose policies on charter flights, the native labour code, or imports, might be in direct opposition. Such matters would then be referred automatically to Lisbon, causing long delays. Gradually Braz d'Oliveira 110 CABORA BASSA

Building 111 succeeded in charting a degree of autonomy and what can best be described as customary licence to reinterpret instructions, so that, while safeguarding the interests of the state, he endeared himself to Zamco. To this extent the Cabora Bassa project was, as its orthodox opponents had always feared, a solvent of the rigid colonial system developed during the years of Salazar and indeed of the attitude of mind rooted in centuries of fitful planning on inadequate resources - the period of 'uneconomic imperialism'. Each time a point was conceded to the urgent requirements of Zamco, a customs preference waived, or a forty-year-old precedent reinterpreted more loosely, the 'alien organism' disrupted the local administration. Surprisingly, the process evolved without very great disharmony. Zamco's leaders acquired a minimum of Portuguese, accustomed themselves to the procrastination of the Beira harbour board and the Companha do Ferocaril do Mozambique, and learnt the meaning of papel selado,t and the stamp taxes which suggested that letters were objects of suspicion rather than means of communication. They learnt to unravel the system of ddpechants, essential intermediaries between client and customs officer, like eunuchs in eighteenth-century Istanbul. (These semi-officials, appointed by government, are paid by the line for their communications. Rich in corruption, the system was actually designed to prevent it, by removing the grosser temptations from the customs officers themselves.) Nevertheless, the civil contractors sometimes found it intolerable to wait for decisions which in their own countries would have been taken under delegated power, decisions about sources of supply (when the cost of importing from Europe was prohibitive), or the installation of telex facilities, or permits for imports and traffic on the railways. These irritants mingled with lesser ones, like the inadequate supply of French wine or German wuirst, because the conditions of the Zambesi valley did not encourage a sense of proportion. Again, they were dispelled pragmatically, as Chambers solved the matter of delayed deliveries and pilfering by a system of tags which showed where bottle-necks occurred and with whom responsibility lay. tFor further amplification see Notes at end of chapter.

Thus, by roundabout means, Zamco's uncomfortable jockeying assisted a little in the long process of Mozambique's emancipation from bureaucracy, and in the growth of awareness of foreign methods among its officials. But the contractors also acknowledged the capacity of those Portuguese they actually met. Without pretending to understand either their imperial sense of mission, or the decision-making process in Lisbon, men like Lemperi~re and La Bossire and the Italians, Mangani and Patanella, felt a certain link of Latin temperament with the PortugueSe. The paradox of their elitism, allied with sudden flights of emotion or romanticism - as when, at a dinner for the administrators of Trete province, a leading banker suddenly burst out, 'But he is my brother!', embraced a black waiter, and set him down at the head of the table, to the accompaniment of clapping and tears - amused but also attracted sympathy from the engineers of Zamco. They themselves formed an elite. The great contractors of the nineteenth and early twentieth centuries who had laid railways across Africa, India, and Latin America, or dammed the Nile, were no longer at the centre of construction as entrepreneurs, welding together political purpose, finance, engineering skill, men and materials. But they did not disappear, no more than did the groups of navvies who learned new skills, drove bulldozers, laid explosives, and mixed concrete. Sometime in the inter-war years, a degree in engineering became a requirement; but the qualities of management and enterprise remained, and the engineers now at work in Brazil, Pakistan, or Cabora Bassa, are different only in the scale of their activities from men like Thomas Brassey, Henry Meiggs, or Ferdinand de Lesseps. If a generalization may be made, contracting engineers tend to be conservative - aware but distrustful of the political dimensions of public works, regarding development as a benefit in its own right - aiid respectful towards sound administration as the precondition of commercial security. The Italian contractors, especially, thought of their own Government as alien and frivolous, run by demagogues (urlatori), or amateurs (improvisatort). Politics was regarded as being indistinguishable from self-interest and corruption - a view held also by many middle-aged Germans, CABORA BASSA 112 Buihling 113 deeply depressed by the lack of purpose in what they thought of as the socialist wilderness of the last fifteen years. In this way, they tended to look back to an idealized early twentiethentury society, finding in the Portuguese dictatorship many of the values and conditions they so missed - sentiments with which the French, subscribing to the precepts of Gaullism, could agree. This is not to say that those belonging to Zamco were simply national stereotypes; they shared, however, to an unusual extent, authoritarian habits of mind and a contempt for the results of democratic politics. Like the South Africans, the Europeans found it hard to see any virtue in the socialist foreign policy of Lusaka and Dar es Salaam; sympathy with what Portugal was attempting in the colonies came far more easily. Hostile to foreign journalists, they tended to deprecate public opinion in their own countries, as ill-informed or vacillating, and their attitudes became especially important when Cabora Bassa came under physical and political attack. A sense of support from the business community - 'serious people' - reinforced the conclusion, put most blatantly by Germans in response to the outcry of 1971-2, that business was good in itself and not to be confused with political morality by the urlator" at home. Cooperation between the Portuguese and Zamco could be founded therefore on a consensus of opinion about the aims of Cabora Bassa; and this was not the least element in what E.T.S. Brown called the 'finest exercise in international cooperation in public works contracting since the war.' In the enforced social mix of nationalities at Songo, however, the majority of whites were not permanent employees of the main companies or of the Overseas Ministry but specialists recruited in Europe, an average of 600 as against less than 200 administrators. These constituted a fluctuating population in which, over the years, Portuguese nationals came to predominate. They came on contract for at least a year, the majority for three or even four years, and their horizons were bounded less by an awareness of the dam's grand design than their own particular sector, local conditions, and the amount they could earn. Lines of prefabricated houses soon altered the look of the plateau which had once carried only a tiny Catholic mission and a few native shambas and banana plantations. The township, at its fullest extent, held nearly 4,000 people, and the airstrip, too small and hazardous for anything but light aircraft, was superseded in 1973 by another capable of taking DC3s. The old clusters of metal rondavels of the Directorate gave way to streets with banks and two supermarkets, while a fine clubhouse and restaurant replaced the can tina which had catered, in something near squalor, for upwards of 500 bachelors. A swimming pool, tennis courts, the lavish bar, and lines of guest houses, like the characteristic safari lodge of the African game parks, enticed newly arrived wives and helped to give Songo the lineaments and social ambitions of the prosperous suburbs of Johannesburg. After the initial tour of duty, the companies gave preference to married men, hoping that family life would stabilize the volatile mood of the camp. Their fares paid, with promises of schools for the children and shopping facilities, the wives came out, and the illusion of civilization settled over Songo. Its box-like houses sprouted gardens and flowering shrubs, in spite of the dust from the roads; dedicated amateurs grew roses and under the trees they built grass huts, rustic arbours for barbecues at night. Songo held all the attributes and most of the problems of a town: the need for incinerators, refuse dumps, a laundry, water - metered because of the scarcity of pure supply and drainage. As well as directing the main labour force, Chambers had to act as town clerk and borough surveyor, and deal with protests, in several languages, of wives furious at being balked of furniture lost on the journey up. But by the end of 1971 Songo was a lively place, boasting a twice-weekly film show and half a dozen sports. Its inhabitants bought cars duty-free and ran down the dirt road to 'Tete at weekends, or, if they could afford it, flew to Salisbury or the beaches of Beira. Memories of the cantina and the Austrian chef who drove the French mad with endless rows of steak puddings receded, becoming the stock- in-trade of a few, now rather lonely bachelors, whose sole amusements were to gamble at Crown and Anchor, and prey on colleagues' wives. The schoolchildren produced a ballet, and the South African 114 CABORA BASSA

Building 115 headmaster was able at last to order his books from the Cape. It was a tranquil existence, disturbed only by recurrent bickering among the women over the status of their houses, and by a few spectacular losses - up to 30,000 escudos - in gambling. (Even that scarcely compared with Kariba, where the Mafia had moved in among Italian labourers and the contractors were forced to pay two-thirds of the men's wages in Italy, leaving only a third for them to lose to each other.) The enterprising and ambitious found it all damned dull. Few animals remained in the bush to be shot and the army discouraged hunting safaris. Though Tete grew rapidly, its single hotel adding air-conditioning and four floors in a year, it remained unattractive. The flashy veneer of shop fronts disguised nothing. Wealth rubbed shoulders with squalor, expensive refrigerators lay outside in the dusty streets for lack of storage space, and lounging soldiers gave the streets the air of a boom town in the American West. Goods proved far more expensive and were shoddier than in Europe, and tastes cultivated in Marseilles and Hamburg were scarcely appeased by Maxims, a red plush, second-rate nightclub boasting a handful of cosmopolitan whores. Money brought the Europeans to Cabora Bassa, just as it brought the Shangaan from the south, and money gave the mix of nationalities its common purpose. Since few Frenchmen leave France voluntarily, CCI salaries were raised 15 per cent above those current at home, with the extra inducement of thirteen months' pay (in francs) for a year's work, and the customary exemptions from French taxes for families living overseas. Germans earned little more than they were used to, but few of them fell below the administrative grade, in which overseas experience is a useful aid to promotion. The Portuguese employed by Zamco earned sums considerably greater than they could have hoped for inside Portugal. Money even equalized some of the differences of nationality and grading, enabling men who would have been artisans at home to fraternize with directing engineers, thus undoing some of the harm caused by social differentiation in graded accommodation. On the whole, Songo turned out to be a harmonious place, and many things which could have caused bitterness - like the building of a separate social hut by the

Germans, who used it to show from video-tape the only television in Mozambique - created amusement instead. Boredom remained the worst enemy; and, after restrictions following the guerilla attacks of 1971, became a serious contingency. What had been feared most, however, came to very little: Europeans and even South Africans reacted to the black community with equanimity. A few South African wives organized a petition against the Goanese schoolmistress, but were countered by an engineer with the good Boer name of van Rensburg: 'My kids were taught in Zambia by a Kaffir and better than in the Republic.' In the absence of apartheid guidelines, some found it hard not to object to sharing the hospital, and its instruments, but they were satisfied by the underlying economic apartheid characteristic of Mozambique - different wards being allotted to different grades. No black engineer was employed and so the town stayed white, and the situation of supervision of whites by blacks did not arise. Chalked up above one of the main tunnels was the graffiti, 'Free State of Cabora Bassa', in French and Afrikaans. It signified a considerable human triumph, but it omitted the same words in Shangaan. Down on the river, in heat or rain, in ninehour shifts round the clock, at the butt end of a compressed-air drill or the seat of a thirty-ton dumper truck, driving down a rock tunnel with only a foot to spare, there was a certain equality. Accidents strike indiscriminately: the roof falls in, a bulldozer slips. Some places, like the shaft from the power house to the surface, were always unsafe, being scarred with faults in the brittle rock, continually scattering stones down two or three hundred feet. The accident record was better than average, apart from the seven deaths from one premature explosion, and the terrible rock fall in November 1973, and a helicopter service to the hospital saved many lives. Knowledge of what had happened at Kariba enabled heat-stroke to be avoided and few Europeans caught malaria and bilharzia. Notwithstanding, at least half the local Africans working at Songo suffered from both. The physique of the mineworkers from the south was better, but tuberculosis and venereal disease were common. The hospital with its three doctors turned out to be a blessing for the community, CABORA BASSA 116

Building 117 its range widening to take in casualties from the whole district and its facilities serving as a base for the efforts of Dr Maurizio to immunize the entire 150,000 inhabitants of the upper Zambesi basin. Fanagalo, the lingua franca, links all underground labourers in southern Africa as much as the work itself or the loyalty to each other in emergencies which transcends colour. Part Xhosa, part Zulu, part Afrikaans, it includes words from Portuguese and English., and a man who has worked in the mines on the Rand may use it in Mozambique, Rhodesia, or Malawi, wherever the contract labourers have come from. It was the only means at Cabora Bassa by which Europeans could reach the Africans, or Shangaan talk to Pondo, Yao, Shona, or Nyungwe. No-one could claim that equality of black and white existed at Cabora Bassa; but in contrast to South Africa, the statutes and rules of the Mozambique Institute of Labour on hours and conditions of work were at least binding on both and were enforced by the inspectorate from Lisbon. In a negative sense, too, the state enforced multi-racialism. Neither black nor white then possessed the freedom to organize trades unions. But the power of the Sindicatos under the corporative scheme of the Estado Novo gave scant representation to black workers. At its best, the colonial system protected minimum rights and wages; at its worst, it perpetuated a low level of both and led to growing economic disparity. Hence the appearance of the African camp presented a problem for visitors from Europe. More than two miles down the valley from Zamco headquarters, but served by frequent buses, it was set under great baobab trees by the Songo River. Journalists from Johannesburg recognized it at once as a model of the camps on the Rand. To critical visitors, however, the native labour camp seemed a clear case of practical apartheid, if not actually of the forced labour occasionally alleged by Frelimo propaganda. The justification usually given by Zamco managers, that Africans do not like to live in close proximity to Europeans, tended to be dismissed as special pleading. A television camera crew from Germany preferred to film the old kraals of the Mademe on the edge of the plateau, for the camp presented too uncomfortably the challenge of the moral issues inherent in the South African claim that, mani- festly, it worked. The Shangaan are not docile and object vigorously if expectations are not met: yet here, where they were given one meat meal a day in the huge communal dining room, plenty of sweet pombe or Bantu beer, maize, and green vegetables, they rarely complained. Less than fifty broke their contracts during the whole five years and many returned for a second tour of duty. The huts were small, but not actually cramped. The men were given a film show a week in their own surroundings and free access to the open-air cinema up at Songo, and leisure to dance and play the guitar. They filled in the rest of their time during the week mending clothes, repairing shoes, listening to the radio, and shopping in the supermarkets, their preferences reflecting a classic pattern - first a bicycle, then a radio, then a sewing machine. Separation from wives and families appeared no worse than in South Africa and, on the evidence of the headmen or indunas, was partly compensated for by an awareness of the virtue of working in Mozambique, for a wage paid in escudos which freed them from the traditional rapacity of money-changers at the frontier. Zamco managers quickly learned that Africans could be a profitable alternative to importing semi-skilled white labourers from Portugal or Lourenqo Marques, as well as a means of by-passing the restrictions on maximum wages laid down by the Mozambique administration. Soon, in steel erection for the concrete linings of the main chambers and the dam wall itself, whites were almost entirely replaced. Macheke Makwakwa, short and energetic doyen of the Shangaan instructors, was brought up from the Rand and installed in three huts at the back of the camp, where, depending on intelligence and aptitude, he reckoned to turn a man into a rigger in four months, a carpenter in three.t No other training, however, was provided, and many of the men regretted the failure of the company to provide opportunities for adult education, other than an occasional night school which few had time or physical energy to attend. Few of the managers or engineers complained of the quality of the work once they had learned how to operate the South African 'bossboy' system and to speak Fanagalo. Discipline on the site and underground was an iron rule for black and t For further amplification see Notes at end of chapter. 118 CABORA BASSA

Building 119 white; but up in the camp it dissolved. Fighting, usually on tribal lines, broke out frequently, though never with the shattering violence of the Rand mines; and drunkenness, to Western eves the curse of Africa, was endemic, a factor of long custom as well as a reflection of the strain of the work and isolation. Thus Zamco paid its wages monthly, with subs in between, reckoning that one prolonged alcoholic weekend a month was better than four absentee Mondays. In all of this, there was a close parallel with the customs of the navvies who built the canals and railways in Britain in the nineteenth century, drinking and brawling their way across the countryside to the horror of evangelical clergy, while claiming to be an elite amongst a downtrodden working class. What the Shangaan thought of Cabora Bassa, and how pride in the work served to develop an embryonic political consciousness, is discussed later. In a way, the spirit of the camp was more deeply subversive of obedience to the colonial system than the overt infiltration by Frelimo feared (and largely prevented) by the DGS, the secret police. The new quality of their life and the ability to earn a relatively good wage in Mozambique currency soon set the majority of the Shangaan against a return to South Africa. By the standards they knew, chiefly by earning capacity, they were well off, t and aspired to even better. Separation from their families for eleven months at a time seems not to have rankled, nor did the relatively mild economic separation between black and white at Songo. But in retrospect, when the constraints over industrial and political action were removed, it became clear that the methods of the South African managers were resented, and the strike which followed the revolution in Portugal in April 1974 was accompanied by demands for their removal. Political aspirations had remained hidden for four years, partly from fear of the DGS, partly because it was simply never the habit of South Africans or Portuguese civil servants to discuss such matters with natives. No political meetings were allowed and the language barrier militated against discussion, even at the indabas (meetings) of the indunas. How little the thoughts of the labour force mattered could be seen from the fact that the t For further amplification see Notes at end of chapter. films made about Cabora Bassa were never shown in the native camp. An unadulterated diet of violent Westerns instead confirmed the judgement of the Bantu as a child embodied in the native labour code, the paraphernalia of colonial legislation, and implicit in the slow steps of the Mozambique system of education. Until the summer of 1971 work on the site kept fairly closely to the time- schedule. Among the ancillary works, which still occupied a major part of the effort, occurred a small, but at the time symbolic, act of vandalism. On the very edge of the buttress of cliffs on the south bank, where the crest of the dam wall was due to end, stood an ancient baobab, sometimes called Livingstone's tree on the rather tenuous assumption that the explorer had passed the gorge that way. Some in Zamco for reasons of sentiment or publicity hoped to spare it, but the tree lay in the path of cable cars carrying cement and was bulldozed down. From where it had stood, it was possible to survey the whole sweep of the site: in the foreground, 400 feet above the river, the narrow suspension bridge, only access to the north bank, and, high above it, the massive cables of the transporter system, 1340 metres long, strung across the river by helicopter and anchored high on the opposite mountain. Above and behind the tree rose the rock-crushing plant, built jointly by the French company, Babbitless, and South African steelriggers, the row of thousand-ton cement storage silos, and the immense winding machinery for the cable cars, contained in two huge sheds. The cars ran laterally on rails, so that the cement could be poured directly on the top of the dam wall, at no matter what point in its extended arc. The access road now ran straight down into the main tunnel, leading to the underground chambers, while on the western end the new and final road was being cut into the precipitous side of the gorge. Here, in perhaps the most spectacular aspect of the project, bulldozers scrabbled at heaps of rock blasted out of overhanging buttresses and nosed along on the very edge of the gorge, hundreds of feet above the river, until, reaching a deep cut in the summit wall, they broke through into the long, heavily wooded valley which skirts the 120 CABORA BASSA

Cabora Bassa, darn and reservoir Songo plateau, to meet the graded road from Tete. Up this road, in and after the dry season of 1973, came the heavy electro-mechanical machinery, generators and transformers, turbines and the banks of solid-state thyristor valves. All over the site, in heaps of debris, lay tyres, shredded by the rough edges of granite chippings, worn-out machinery, and buckets from diggers, bent and ripped out by the rock - practical evidence of the cost of transport in an untamed environment. The cement silos and stacks of hundreds of steel rods showed the need to stockpile raw materials against the risks of flooded roads or late deliveries in Beira docks. Cement proved to be a continuing and serious problem. Zamco had no control over the manoeuvring which initially excluded Rhodesian supply in favour of the Champalimaud factory at Dondo, and soon regretted it. Twenty-eight rail tanks were ordered, each of forty tons, a thousand-ton silo was then built at the rail-head at Moatize, and through United Transport a fleet of bulk cement lorries was hired, each capable of carrying twenty-three and a half tons, to bring the cement up to Cabora Bassa. Theoretically, this should have

Cabora Bassa plant been sufficient to meet the peak demand, in 1972-4, for just under a quarter of a million tons. But, in addition to the usual complexities of dealing with the railway company's monopoly of freight, the delays on the Zambesi bridge imposed a crippling restraint. Only one load could be taken at a time and as the bridge schedule ran later and later into 1972 Zamco was forced to look for alternatives. The erratic capacity of Dondo and, far more seriously, the fact that whole batches sometimes contained too much chalk or gypsum led the civil contractors to threaten to buy from Europe or Japan. On at least one occasion before 1972 they reverted to Rhodesian sources, albeit in great secrecy for fear of being labelled as sanctionsbusters. A brief description of the workings of a hydro-electric barrage may be useful (see accompanying diagram). The purpose of the dam is simply to give a sufficient head of water to turn the turbines at the necessary speed for generating current, with no risk of interruption since the maximum wear on the blades occurs at speeds between 10 per cent and 60 per cent of full capacity. Hence the intakes to the power house must be set high and be large enough to allow the right volume and speed of water to flow down the penstocks, while being far enough below the minimum normal level of the lake to ensure that they are always covered. At the foot of the penstocks is a narrowing tube, curled like a snail, directing the flow of water into the turbines. As they revolve, the turbines turn the core of the generator in the power house whence electric current passes into transformers in the transformer house. The current then passes up the cable shaft to the surface and along the primary line into the sub-station, where it is turned from alternating current into direct current. Meanwhile the water, discharging from the turbines, flows out down the tail-race and back into the river, interrupted by two surge chambers, cut out higher even than the power house, which allow for variations in pressure and keep a roughly constant flow. The final layout of tunnels, however, is superimposed on the original network of access and diversion tunnels, so that for a time two overlapping systems exist. The schedule demanded that the tunnels to divert the river (and thus allow excavation of the bed for the dam wall) should fit precisely into the programme of building the power house, so that in 1969-71 all energies were directed to this excavation. The primary diversion tunnel measured 30 metres high by 20 metres wide, but seemed small in comparison with the main chamber, 60 metres high, 210 metres long - 300,000 cubic metres of rock. Even that was only a quarter of the total excavation underground. Moving inwards from trial tunnels, the miners drilled out sufficient space at the top of each projected tunnel or chamber, and then worked steadily down, using 'jumbos' - machines driving drills in pairs, fours, sixes, or even eights. Section by section they worked along the perimeter, drilling holes every six inches. These holes were then filled with dynamite and detonated in rows, so that the rock peeled away, ready for carting to the surface. The more accurate the work, the less finishing there was afterwards, and the less concrete needed to coat the wall against seepage or rock falls. After the floor of the transformer chamber had been sunk to its final level, the regular scars of drill- holes on the wall, like giant graffiti, showed how high were the standards set by the South Africans of Shaftsinkers. In the main hall, Building 123 partly owing to the poorer quality of the rock, the French were less successful, and the immense curve of the vaulted roof needed weeks of trimming before it could be lined. The roof had to be completed before the full depth was reached, since the concrete lining of the span could only be cast from a steel frame, like a gigantic template, inched forward on rail tracks. Behind, as this structure was moved on, were set the supports for two beam cranes, each able to lift 500 tons and move the complete generator sets for maintenance or repair. At the same time, work began at the base of the chamber, on the transformer link, the cable shaft, and the lower end of the penstocks. By 1972 the ceiling was almost completed over its entire length, the transformer chamber finished, bar its concrete lining, and the cable shaft cut, through dangerously unstable rock, to the surface. No-one looking at the 350-foot shaft today, smoothly finished in reinforced concrete, could know what it was like before, a death trap of loose stones. Outside, the face of the southern mountain was cut away to a flat surface, ready for the intakes, and as each shaft opened up, the buttresses in between grew to look like monumental Nubian statues overlooking the Nile Valley. To excavate each intake, a pilot shaft was cut from the bottom, upwards, at an angle of forty- five degrees, by Pondos specially brought in from the Rand, working in conditions of extreme hazard. Later the whole of each tube, 6 metres in diameter, was to be lined with steel to resist the erosive thrust of water. The surge chambers and the tail-race came later, growing outwards from pilot tunnels as the more urgent work permitted; and from mid-1971 the drillers began to pierce the surface of the underground chambers with yet more holes to take steel fixing rods for the lining, and to plug leaks and fissures which inevitably appeared. On the whole the rock inside the mountain was better than the dam abuttments, where fissures led to a vast increase in the quantity of excavation, but there were numbers of rock falls (the most serious in November 1973) and at least one access tunnel had to be abandoned. An extra 100,000 cubic metres was excavated gratuitously, in order to give freedom of movement to the thirty-five- ton dump trucks extracting all the debris. 124 CABORA BASSA

The main chamber looked like a cathedral, but no cathedral ever sustained the noise of jumbo drills and explosions, or such damp heat under arc lights. Yet the sense of ordered team-work amidst the chaos was almost tangible, and the cathedral simile has value. The itinerant masons who worked the stone of the soaring structures of the middle ages had their own lingua franca, like Fanaglo, and worked, within the same parameters, for architects less technically skilled but bound by similar constraints. Only time was different. In place of the centuries available to the Church, the demands of the contract and the annual flood set precise and absolute limits. Because there was no access to the north bank, other than by the footbridge, the diversion tunnel on the south bank had to be cut large enough to take the whole of the river at low water. A temporary dam could then be built in the summer of 1972, to allow excavation of the smaller northern tunnel at the same time as the building of the coffer dams above and below the site of the dam wall. So much depended on this south tunnel that elaborate precautions were taken and its size increased. (Originally, the specification allowed for a flow of 3,000 cumecs in the dry season. Partly from fear of what could happen to the coffer dams if the bed of the river, with its enormous boulders, were disturbed or if the dams overtopped, and partly because of the results of model tests made by Coyne-Bellier in Paris, the French suggested a new design allowing for 6,000 cumecs. In the end, they compromised on 4,500, and redesigned the coffer dams to allow for a limited degree of overtopping in the flood.) Since it would be the last gap to be plugged before the lake was filled, the south tunnel was also fitted with elaborate headwork and roller gates. Protected by a coffer dam of its own, the tunnel was dug in 1969-70 and opened as soon as the water level fell in April 1971. In this matter Zamco was fortunate. The north diversion tunnel was completed before the flood of 1971-2, and the coffer dams, constructed of enormous boulders, grouted with smaller rock, and held by steel piles driven deep into the river bed, stood quite undisturbed by the unusually light floods of that year and of 1972-3. The precaution of coating the down-stream coffer dam with a thick layer of concrete against Building 125 the battering effect if the river did overtop, proved almost unnecessary; but the extent of the risk which the French accepted, may be gauged by the effects of the extraordinary flood of 1973-4 (see Chapter 8). It was not an excessive fear to have envisaged and allowed for the effect of perhaps 5000 cumecs flowing over the top, hurling fifty-ton boulders like shot from a Roman ballista. 'Money doesn't matter, time does,' was a commonly used phrase in the days before inflation of costs and wages turned the Cabora Bassa budget into a nightmare. Zamco gained time considerably from the close cooperation with the Central African Power Corporation at Kariba. Following Portugal's signature of the contract with Zamco, the CAPC had agreed to complete all spilling for 1969 by May, and to spill no more until January 1970; it was also agreed to finish all heavy spilling in 1970 by the unusually early date of March, unless essential to Kariba's safety; and to control the water from then on so that the total discharge in the dry season never exceeded 2800 cumecs. Yet more time was gained - and most of the delays of 1969-70 made up - by the speed of setting out the coffer darns, by an agreement with the Portuguese to start concreting closer to the excavations, and by acceptance of a possible 8000 cumec flow over and through the dam wall in the 1973-4 season; this latter was a farsighted, almost prophetic, decision. By mid-1972 the ancillary works had been finished, the bulk of the underground chambers excavated. Concreting had begun and the abuttments of the dam, nearly twice the planned size because of the poor condition of the rock in the gorge, carved deep into both the cliffs' faces. The quality of coordination on the civil engineering side could now be gauged by Zamco's capacity to take daring decisions; and Lemperifre's paper on Cabora Bassa, given to an assembly of international engineers in February 1973, exemplified the assurance of control.' On the electro-mechanical side, very little occurred at Cabora Bassa before 1973. All the machinery of the five 415 megawatt generating sets was built in Europe, the turbines and generators in the factories of Alsthom and Neyrpic, the transformers by AEG in Germany and CGEE in France. Each CABORA BASSA 126 set was to be identical in order to obviate the need for large stocks of spares and to make for easier maintenance, and the only problem at that time was to organize the shipping and transport up-country in Mozambique. The final form of the converter assemblies remained undecided, buo the banks of thyristors, heart of the system, were ready for assembly in Germany. The chief concern, as among the civil contractors, had to do not with site difficulties but with inflation in Europe. Both French and German salaries rose by nearly 80 per cent between 1968 and the end of 1972, as against the 16 per cent allowed for inflation by Article 48 of the contract. Similar rises in prices of materials far overreached the fixed prices of stage one, that is the first three generators. Even worse, the original terms allowed nothing for modifications and alterations on the civil side. A great volume of argument and potential litigation between Zamco and Lisbon was building up, long before the military events of 1972-3 altered their relationship and gave both an overriding interest in completion at any cost. Hundreds of miles away the Italians, SAE of Milan, had begun work on the transmission line in October 1970. Despite the political retreat, described in Chapter 2, SAE continued to ship out steel from Italy, though much of the structure of the 7000 towers came from South Africa. Nothing in the work either in South Africa, on the link from Johannesburg to the frontier, or inside Mozambique - presented anything but logistic difficulties, principally that of transporting 65,000 tons of steel, several thousand miles of aluminium cable, and a quarter of a million insulators. Most of the country is flat bush or scrubland and the chief obstacles were the great flood-plains of the rivers Limpopo and Save, and the escarpment which rises in the south through dense bush four thousand feet to the highveld of the 'Transvaal. SAE cleared two wide strips, a kilometre apart, using bulldozers and mobile cranes, and then laid the foundations and raised the hundred-foot-high towers themselves. Maintenance depots were spread along the line, with a continuous road and access from the few towns within range. In the empty countryside, the danger of sabotage was obvious, but rarely discussed; SAE, conscious still of unfavourable publi- Buidting 127 city at home, worked discreetly, even secretly, and left strategic considerations to the Portuguese. Despite the loss of a year during the political storms of 1969, they were six months ahead of schedule by mid-1972. During these first three years the war in the north threw no shadows. Black- jacketed Africans patrolled the camp at night, carrying rifles, giving a measure of reassurance about the security of the dam. Senior men in Zamco knew of the Portuguese Ministry of Defence guarantee of protection, and had a vested interest in keeping down speculation about the efficacy of the ring of army posts, in company strength, in the villages north of the dam and in the upland river valley. The hills around Songo itself were manned, rather desultorily, by the DGS, and, apart from the occasional alarms such as the discovery of two German students claiming to be employees of Hochtief, but suspected of distributing Frelimo propaganda, the calm lasted well into 1971. The army was assumed to control the wasteland to the north, which stretched, a hundred miles of uninhabited bush, to the Zambian border. The war, so it seemed from the Portuguese press and army communiques, was confined to Cabo Delgado and Niassa, hundreds of miles away, and to the eastern side of Malawi. Established wisdom indicated that the opposition to Frelimo of the Macua, largest tribe in Zambesia, barred the way to any further advances. Adverse publicity in Europe, like the opposition of President Kaunda to Cabora Bassa, counted as merely another exercise of the urlatori, and was in any case offset by the appointment of General Kaulza d'Arriaga and by his well-advertised 'Gordian Knot' campaign to drive the terrorists back across the Ruvuma into Tanzania. In September 1970 Caetano pledged the resources of metropolitan Portugal to defend the empire at all costs, and the war in Angola at least began to show success. In March 1971 d'Arriaga told a reporter from the Johannesburg Star that, despite their boasts, Frelimo could never touch Cabora Bassa. But this interview was given after a lorry had been attacked in broad daylight on the road from Tete to Songo. Despite censorship of the incident, Zamco officials began to consider the implications of a hostile 300-mile-long frontier with Zambia. For two months fears that the site was exposed were 128 CABORA BASSA kept under control - indeed news of one raid by the Frelimo splinter group, Coremo, was suppressed altogether inside Mozambique - but Coremo's subsequent capture of five surveyors, working for the GPZ near Fingo , could not be hidden because its implications were argued noisily in the European press. Attacks on transport on some of the isolated roads in the Fete province occurred, and on 19 April a communique from Nampula, d'Arriaga's headquarters, admitted that Cabora Bassa had become a Frelimo target. In spite of a slight uneasiness, no-one at Songo gave credit to the idea of a terrorist sabotage group hitting the dam itselx. But outside the perimeter, army and DGS activity increased sharply. Two more battalions of troops were brought in, despite renewal of the onslaught in the north, and 'Operation Minerva' was launched in Tete Province. Accounts of lorries being mined or machine-gunned began to appear in South African papers, often with derogatory remarks about Portuguese military performance, and perceptive listeners in Tete heard stories about local villagers fleeing from some new and unspecified menace. In May, a military governor was appointed for the province, and the historic lines of development of the war and the dam were finally linked together. Buildting 129

Notes and references Chapter 5 Building t Note page 111. The use of stamped paper served - like the formalities needed to register a letter and the speed of its subsequent transmission - as an inverse indicator of administrative efficiency. In Portugal it is still required for all transactions with government departments, and in the provinces, until the reforms of the 1950s, it was sold in two categories: the cheaper for handwritten submissions, and the more expensive tor typed letters, since the latter were held, defacto, to be less respectful of authority. + Note page 118. On the whole he found the Shangaan suited for steel erection, the Nyungwe for carpentry. Some idea of the impact of this training can be gauged from the following figures. Wages were, of course, markedly lower than for whites: the lowest-paid Portuguese labourer earned 34 escudos an hour (6500 escudos a month) in January 1974. (But see page 103 for rates for skilled men.) Total unskilled Semi-skilled as Basic wages: and semi-skilled proportion of total Jan. 1972 1600 25 per cent (422) Unskilled -600-900 escudos a month Jan. 1973 2284 22 per cent (694) Semi-skilled -1250-2300 escudos a month Jan. 1974 3178 48 per cent (1466) *i- Note page 119. In South Africa, between 1969 and 1973, the starting wage rose from 31 cents to 80 cents a shift (R8.06 a month to R20.80), with Anglo- American workers usually earning 15 per cent more, in conformity with Oppenheimer's policy. In the same period, the statutory minimum wage in Mozambique rose only 6 per cent; but Zamco paid roughly 25 per cent more than in South Africa, and evaded the regulations by training and upgrading. I 'Methodes de Derivation pendant la Construction', Madrid 1973, French working party report to the Commission Internationale des Grands Barrages.

Chapter 6 Revolt in Mozambique The Front for the Liberation of Mozambique (Frelimo) had its origins far back in the colonial period, and some explanation of its history is needed to understand the reasons for the military eruption in Tete in 1971, and for the propaganda assault on Cabora Bassal. Frelimo claims that the movement incorporates the anti- colonialist struggle of all Mozambicans are examined later, and there is no need to refer to the better-documented period before the Second World War in detail. However, the contrast in Portugal's colonial policy before and after Salazar's rise to power should be noted. In the period 1900-20, partly as a result of' the administration of Antonio Enes, Mozambique, like Angola, experienced relative prosperity. Under a series of able proconsuls, not unlike the members of Milner's Kindergarten in South Africa, civil administration was restructured, and the old tribal chiefs largely replaced by Portuguese-speaking nominees. The chartered companies and commercial enterprises in sugar and sisal profited greatly, controlling, in Zambesia especially, virtually autonomous fiefs. Unlike Angola, few Portuguese actually settled on the land, but the vast latifundias or prazos were managed by a handful of whites and assimilados.(The white population in 1930 was a mere 18,000, rising to 27,500 in 1940, 48,000 in 1950, 85,000 in 1960 and 225,000 in 1970.) Labour for the mines in South Africa became a vital export, regulated under the laws of 1903 and 1928 by which half the wages were paid in gold to the Government of Mozambique and returned to the workmen in local currency, less taxes, only at the end of his contract. Slavery had disappeared in the late nineteenth century, to be replaced by various categories of forced labour, several of which were still in operation long after Salazar's 1930 Colonial Statute; and a complete legal separation had been achieved between those subject to the Portuguese code (whites and assimilados) and the vast majority of Africans living under the regime do indigenato. Salazar's legislation heralded a period of stagnation, in which colonial policy concentrated on balanced budgets, stringent economic control (with heavy trade discrimination in favour of the metropolis), and a theoretical preoccupation with what the colonial philosopher, Jodo Ameal, called 'the task of civilizing ... a transcendent campaign, a sharing of values'.2 The next fifteen years saw both the peak of the assimilado system and the beginnings of opposition among a small group of black intellectuals, educated in Lisbon, who found the new constraints frustrating and so founded the Liga Africana, embarking on tentatively Marxist analyses of the colonial situation. A certain cultural and poetic revolt took place, but these aspirations centred on claims for participation in white privilege rather than for Pan-Africanism or black nationalism. Political movements in Beira and Lourenqo Marques after the Second World War drew strength from the apparent liberalization of the time. The colonies, in name at least, were promoted to the status of provinces under the liberal administration of Adriao Moreira, who abolished the regime do indigenato in 1961 and introduced a new Overseas Organic Law two years later. But this Portuguese- speaking, intellectual opposition, confined to the cities, looked for leadership to the predominantly white liberals associated with General Delgado, and their coherence dissolved during the repression which followed Delgado's exile in 1959. The grievances of politically educated Africans differed greatly from those at tribal level. The last tribal revolt ended in 1918 with the suppression of the Angoni in north-east Tete, but undercurrents of hostility to Portuguese rule survived, especially in the north among the Makonde and Nianja. One aspect of the drafting of native labour, the use of Makondes on public works in Lourenqo Marques in conditions very close to slavery, continued into the 1950s, leaving a tradition of resentment and intense deprivation in the poor northern plateau, 132 CABORA BASSA

Revolt in Alozambique long distinguished economically from the rich plains of Zambesia and Quelimane. Even after the Second World War, Portuguese administration there was confined to a handful of postos, dotted across the vast areas of Niassa and Cabo Delgado. Beyond the small towns such as Vila Cabral and Mueda, Portuguese itself was hardly understood. The Makonde suffered from geographical as well as economic disadvantages: Carlos Reis, the anthropologist, noted in the 1950s an abnormally low physical development - which he attributed to their isolated life - and a separate culture, with strong Islamic affiliations. They claimed not to be related to other surrounding ethnic groups and showed particular hostility towards the \lacua of Zambesia.3 Peasant revolt in the north sprang from resistance to increased taxation, imposed to pay for the piping of water to villages on the arid plateau, and to Portuguese attempts to impose cooperative agriculture on traditionally individualist smallholders. The Makonde's attempts first to raise the price of cotton and then to organize themselves, were ignored by the white opponents of colonial rule and severely repressed by the Portuguese administration. The movement culminated in an attempt at mass mobilization, copying the methods used by Julius Nyerere in Tanganyika. The Portuguese agreed to a meeting at Mueda in June 1960 to discuss Makonde grievances, but the local commissioner, appalled at the size and intransigence of the crowds, lost his head (like General Dyer at Amritsar in 1919) and ordered his troops to fire. Probably five hundred were killed in what is officially labelled a revolt, but on the available evidence must be called a massacre. Attempts to arrest Makonde leaders followed, and a four-year reign of terror ensued, almost wholly ignored by the outside world; hundreds of peasants fled into Tanganyika and settled in the dense bush country north of the Ruvuma, sometimes forming groups of raiders, sometimes merging with the local, ethnically similar, population. The danger of divisions inside Mozambique seemed very great. At the meeting of liberation movements held in Accra in July 1958, in the heyday of Nkrumah's inspired Pan-Africanism, there had been suggestions that the incipient political groupings - Manu, Udenamo, Unami - were trying 133 to revise the boundaries of Mozambique in order to align with similar groups in Malawi and Tanganyika (the Makonde, Nianja, Angoni, Nyungwe, and Shangaan, overlap into neighbouring countries). But Tanganyika's achievement of independence in 1961 and the disruption of the Central African Federation emphasized the importance of nineteenth century colonial frontiers, however ill- drawn. The Pan-African dream faded rapidly, as did Nkrumah's influence, and its only institutional expression, Pafmeca, gave way in 1963 to the OAU Liberation Committee and the policy of encouraging liberation movements to combine on a purely national scale. Opposition to Portuguese colonialism among Africans began to rally round the figure of Eduardo Mondlane, a Mozambican from the south. Educated in South Africa and the United States, Mondlane had worked for five years as a research officer at the United Nations, married an American girl, and joined the faculty of anthropology at Syracuse University. Like Amilcar Cabral, later leader of the PAIGC in GuineaBissau, and Marcellino dos Santos, second-in-command of Frelimo, he owed his rise to the tiny margin of flexibility in the assinilado system, while campaigning vigorously for its destruction. Mondlane's dilemma was that sporadic action against the administration, and the clandestine activity amongst dock and railway workers, had broken against the power of censorship and the surveillance of PIDE (Policia Internacionale Defesa do Estado) in a virtually closed society. Yet the means of conducting an alternative struggle were lacking, unless he resorted to the examples of the Mau Mau in Kenya or the revolt led by Holden Roberto in Angola in 1961 - an affair of the utmost horror, with uncontrollable massacres carried out by both sides, intent on reprisals. It was the attempt of PIDE finally to crush incipient rebellion in Mozambique before it grew too fast which forced the political opposition into exile and alliance with the peasant movement in the north. In 1962, in Dar es Salaam, Mondlane presided over the union of the three groups, later to be joined by the student movement, Unamo, into the Frente para a Liberaqio da Mozambique. Uria Simango, a Protestant pastor from Beira became VicePresident, and the Central Committee constituted itself of 134 CABORA BASSA

Revolt in Alozambique representatives from each region of the country. Alienated completely by the failure of attempts at a reformist solution, the members of Frelimo announced an armed struggle, on the pattern of the MIPLA in Angola, and appealed to the OAU and the United Nations for support. They had a cause, an organization, a reasonably secure base in Tanganyika, and a sympathetic population in the northern sections of their homeland. But they gravely underrated both the residual power of Portugal and the logistic problems of starting a war six hundred miles from Beira, a thousand miles from Lourenqo Marques, the seat of government; and they overrated the efficacy of their allies and the United Nations, just as PAIGC did in Guinea, and as the soon dis-united movements did in the back-end of Angola. Two years in Dar es Salaam passed in preparation for guerilla war, and in political education, and saw the start of diplomatic activity at the United Nations and in sympathetic capitals like Algiers and Cairo. Recruiting missions were sent south, but had greater success among the Makonde already living in Tanzania, for Mozambique, as Frelimo was forced to admit, was not then ripe for armed insurrection. Numbers grew slowly at first; regular cadres in groups of between fifty and seventy underwent military training in Algeria in 1963-4, while a further training centre was set up fifty miles north of Dar es Salaam at Bagamoyo. In Dar es Salaam itself, Mondlane was able to build the Mozambique Institute, with money raised largely from well-wishers in the United States, as a school for the movement, which by 1968 was capable of taking 120 pupils. Frelimo began to issue occasional propaganda sheets in Portuguese, and soon was publishing a regular magazine in English, designed for European and American readers, called Mozainbique Liberation. Shortage of weapons and finance determined the date of launching of the armed revolt. Seriously to hamper the Portuguese, Frelimo needed to penetrate beyond Makonde and Nianja territory into the rich heartland of Zambesia, to disrupt the economy of the province, and threaten the white strongholds in the cities. Given the goodwill of independent Tanzania, and especially of Oscar Kambona, then Foreign Minister, Frelimo was soon assured of recognition by the 135

OAU Liberation Committee t and of secure bases in the southern provinces bordering the Ruvuma River. Arms came in slowly at first, some from Algeria, relics of the war against France, some paid for by the OAU; then later a more steady supply from Russia, and finally, after 1967, from China. Other aid, medical supplies, school equipment, was sent or paid for by money donated from European committees of support, evangelical churches in the United States, and official sources, such as Scandinavian governments, unwilling to contribute directly to military expenditure. Inevitably, ideological divisions opened up, though not so disastrously as in Angola. Inherent in the compromise of Frelimo's origins lay diverse interpretations of aims, even of the concept of Mozambique. Was it fighting for the independence of a nation-state or a federation of ethnic groups? If the former, on what language and political system should it settle? What elements of the Portuguese heritage could be used and what had to be rejected? These questions divided and embittered Frelimo many times before they were resolved in the early 1970s. Usually the currents ran along lines of tribal affiliation and personality, and Mondlane (jokingly referring to himself as the 'umbrella') deliberately set out to overcome the divisive tendencies of the different dialects, religions, and cultures, of his supporters. The first actual split appears to have arisen from a clash of personalities, thinly covering a dispute about fundamental aims, between Mondlane and Paulo Gumane, former Secretary of Udenamo. Born in 1918, Gumane, a teacher, was deputy National Secretary of Frelimo in 1962. In 1963 Gumane and others broke away and set up their own splinter organization in Cairo. Two years later he became President of Coremo (Comite Revoluqionario de Mozambique), based in Lusaka. Other smaller groups, Manu, Papomo, Ulipomo, divided and disappeared, and with them several of the original leaders, many of whom died in circumstances suggesting elimination by erstwhile colleagues. Infiltration by agents of the PIDE accentuated fears of heresy and betrayal, culminating in the murders and arrests of 1968 associated with the defection of the Makonde leader, Lazaro t For further amplification see the Note at the end of this chapter. 136 CABORA BASSA

Revolt in Mozambique Kavandame. (Gwenjere, from Manica, and Sigante were both wounded in Lusaka in 1968, and Kankombe, second-incommand of the army, was murdered at Mtwara in the same year.) The most serious dispute occurred in March 1968, when the Institute boiled over in a violent argument about the question of teaching in Portuguese. Many of the younger men in Frelimo saw the Portuguese language as a colonial relic and espoused English, a language which would link them - and a future independent Mozambique - more closely with Tanzania and Zambia. The leaders, who realized that Portuguese was the only means of guaranteeing a unified Mozambique, while retaining at least the benefits of a cultural link with Portugal, and perhaps Brazil, resisted. The struggle lasted for months; several instructors had to be expelled, and Mondlane's leadership weakened in the face of further troubles. In retrospect it can be seen that this was the beginning of the division between those who looked to Western democratic models of political organization and those who saw in Marxism and in African one-party structures, especially the cell system of the Tanzania African National Union (Tanu), a more appropriate form for the advancement of nationalism and socialism. But the pattern was still confused, for Mondlane, already to many in Frelimo the epitome of Western orientation, took the dispute as an attack on his leadership and sacrificed the Institute's independence in order to align Frelimo more closely with the Tanu precepts set out in the Arusha Declaration in 1967. Four years of growing success in the front line sustained Frelimo through these controversies which might otherwise have broken it into competing factions like the Angolan and Rhodesian movements. At the same time that it launched what the regular broadcasts on Radio Tanzania described as 'the general armed insurrection of the Mozambique people against Portuguese colonialism', Frelimo possessed 500 trained men under arms; three years later their numbers had risen to nearly 8000. The insurrection, launched in September 1964, was carefully planned to avoid the horrors of Angola or the Mau Mau in Kenya and relatively few atrocities were committed against white civilians, although the Makonde, keen to avenge 137 decades of exploitation, slaughtered numbers of Portuguese settlers on the cotton and tea estates of the north. What was intended as a combined attack after several weeks of infiltration actually ended in sporadic attacks on postos and farms across Cabo Delgado; but Frelimo followed these up with a surprisingly rapid penetration to the edge of Zambesia and Quelimane. At first, the guerilla tactics threw the Portuguese into confusion, and during 1965 most of the settlers withdrew from the north - never to return. Although the colonial administration had expected a revolt, it had counted on holding a series of small garrisons and cutting off Frelimo's lines of communication. Instead, it discovered a well-armed and trained enemy, able to live off the land, independent for months of bases across the border. Within the first year, ambushes and attacks flared up across the north as far as Vila Cabral on the border with Malawi; along the Zambesi, in the lands of the Macua, however, they failed. Zambesia was not ready for the people's war, and Frelimo regrouped for a longer, harsher struggle, more intensive training and campaigns for recruitment and arms. Extensively supplied by the OAU Liberation Committee, Frelimo sent in more columns, numbering roughly 5000 men by 1967, against an army of 30,000 Portuguese. (In the aftermath of the Angolan revolt there had been no lack of volunteers in Portugal, and the effect of Salazar's call to the defence of the empire, and the new-found status of the army, sustained recruiting until 1967, when conscription was extended to eighteen-year-olds and raised from two to three years.) The Portuguese attempted at first to restrict the war to the northern plateau, which was regarded in Lisbon as an inhospitable region of little commercial value, and gradually to break up the insurgents in the rural areas. They were only partly successful. Encounters with Frelimo were usually bloody and brief. Increased fire-power gave the guerillas a killing capacity far greater than in earlier colonial wars such as that of the Mau Mau in Kenya. Armed columns found themselves ambushed and their vehicles mined. Losing mobility on the few roads, the Portuguese army retained all its important garrisons at the price of control outside them, thus 138 CABORA BASSA

Revolt in Mozambique allowing Frelimo to claim large but sparsely inhabited areas as 'liberated'. Concluding quite early on that the effort to expel the terrorists completely was futile, as much on the grounds of the army's inability to run search and destroy missions with the vigour of the Rhodesian or South African armies as on those of cost in men and materials, the administration sought easier means of controlling the population. In an attempt to capitalize on Macua dislike of the Makonde, and following the examples of French resettlement in Algeria and the so-called strategic hamlets in Vietnam, the Portuguese began slowly to build fortified villages along the Msalu River and the northern edge of Zambesia, without realizing how far in fact Frelimo had come to control certain parts of the country further north around Mueda. Despite 5000 casualties in three years, the war here looked less significant than in Angola, by far the richer of the territories. Both sides reached something of a stalemate; as Frelimo grew slowly it attracted deserters from black companies in the Mozambique army, but the army itself was stepped up to 40,000 men. Working on extended supply lines, involving journeys on foot of up to a fortnight, and heavily dependent on local food and support, Frelimo, though it could not be defeated, could not easily expand. In 1967 Frelimo attempted to step up its attacks, though not to the level of major engagements with Portuguese troops. By then it had benefited from the return of members trained in Russia, the first shipments of weapons from Eastern Europe, and the established support of Tanzania. Half a million Makondes in southern Tanzania, including at least 15,000 refugees from Mozambique, offered a fertile ground for recruiting. But although it overran several small posts in Niassa, ringed Mueda airport, and cut all the coastal roads in the 1967 offensive, Frelimo's aims remained modest.* The attacks, however, brought a sharp reaction. The Portuguese, now commanded by General Adriao Pires, counter-attacked, for the first time on foot, using mine-detectors, helicopters, and *Thus in Mozambique Liberation of 9 September 1967 it argued that there were three possibilities for the outcome of the war: perpetuated colonial rule; white- settler-dominated independence; or neo-colonialism with political power in the hands of African puppets. At that stage complete independence seemed to lie in the distant future, following decades of insurgency and political education. 139 light bombers, in support; the population - albeit illiterate - was bombarded with leaflets and submitted to sky-shouting propaganda. Further south the DGS (the former PIDE) busied itself arresting Frelimo sympathizers like the lawyer, Domingos Arouca, and the poet, Jos6 Craveirinha. Reprisals on both sides became habitual, raising the level of suspicion and hatred in areas far removed from the actual fighting. The level of Portugal's effort in Mozambique could be gauged by the increase in military expenditure, at a time when the Angolan campaign had stabilized.t Until 1971 the effort was concentrated exclusively in the north. For several years Portugal had enjoyed good relations with Zambia, largely because of Zambia's dependence on the Beira and Benguella railways, and, apart from isolated raids in the early days, Frelimo had had little sucess in the province of Tete. These realitities were emphasized when Coremo essayed a long raid from Zambia, with members of the South African Pan-African Congress, as far south as the river Save in March-May 1968. Pressed vigorously by the South African Government, the Portuguese retaliated against Zambia by sending a patrol boat up the Zambesi and landing a Scottish mining engineer who blew up the main bridge across the Luangwa River, destroying the oil supply line to Lusaka. Frelimo's first efforts in the west we-re confined to opening an office in Lusaka, stockpiling arms, and raiding isolated postos across the border. These raids proved, by the end of 1968, that there was little chance of gaining a permanent foothold south of the Zambesi with such scant forces, or of sabotaging Cabora Bassa - although in The Struggle for Mozambique Mondlane set both aims high in his list of priorities. Meanwhile, in Dar es Salaam, voices were raised against Mondlane, accusing him of Western affinities. For a time the movement threatened to disintegrate. Sections of the Makonde led by their traditional chief, Lazaro Kavandame, attempted to throw off Mondlane's authority, attacked the Institute in May 1968, and prepared to defect. On 3 February 1969, while staying in a house at Oyster Bay, Mondlane opened a parcel tFor further amplification see Notes at end of chapter. 140 CABORA BASSA

Revolt in Mozambique 141 containing a French translation of the works of Plekhanov - and the subsequent explosion of the parcel bomb killed him instantly. The assassination was immediately blamed on the PIDE, and later sometimes on Uria Simango, one of Mondlane's successors. It may never be possible to say with certainty, but the most likely candidate, if not the PIDE, was Kavandame, who fled to Mozambique immediately after, and who seems thus to have evaded arrest by the Tanzanian police. In the event, the Portuguese used his defection shrewdly to spread mistrust among the remaining Makonde. Frelimo suffered severely from Mondlane's death and the tribally-based, separatist dissensions which followed. Mondlane had been an international figure - perhaps the only liberation leader of the period to be acceptable to large numbers of Western sympathizers. His links with America and the UN ensured supplies and money from church groups and private individuals, and in the propaganda war balanced the weapons from the East. In the lean years up to 1968 Frelimo disposed of more income than the Guinea-Bissau and Angolan movements put together. Most important, Mondlane had been able to impress on the movement a vision of national unity, incorporating the experience of fighting the war with the necessity of political education, principles which gave Frelimo much of its strength. But since both depended on an increasing political literacy inside Mozambique, as well as a loss of Portuguese will to continue the struggle, the time-scale of independence appeared lengthy. Even before his death voices were raised in favour of capitalizing on what had been won; in particular of turning Cabo Delgado into a breakaway state like Biafra. Later, the form of political education came under question. Following a familiar pattern, personality clashes at first concealed political differences. In the increasingly parochial atmosphere, funds diminished, but not the supply of arms. The soldiers of the movement, particularly Samora Machel, became prominent and, despite the 'continuing testament' of Mondlane's memory, its remaining Western orientation diminished, so that the Committees for Freedom in Europe tFor further amplification see Notes at end of chapter. and America lost political significance and reverted to fund-raising bodies and the Institute, run by Mondlane's wife, lost its political influence. For a time Frelimo looked like a purely nationalist, military movement; and the triumvirate of leaders that succeeded, Uria Simango,t Samora Machel,tt and Marcellino dos Santos,ttt reflected the uneasy balance between rival groups. The triumvirate lasted only nine months before Simango broke away, accusing the others of Western deviationism and blaming them for continuing tribalism and the indolence of the political wing in a pamphlet, The Gloomy Situation in Mozambique. Despite OAU and Tanzanian Government favour, at the Frelimo Central Committee meeting in May 1970, against the brilliant advocacy of dos Santos, Simango lost his majority. Tanzania at once expelled him, to prevent subversion; and within a year Mozambique Liberation, rewriting history, had linked him, as a traitor, to the earlier revolt of Kavandame. Machel became the leader, with dos Santos as his deputy. Frelimo's 1968 Congress had ruled in favour of increasing the armed struggle, and drafting not only more recruits but the political cadres who had been accused of avoiding the hard business of fighting. Given a free flow of arms from Russia, China, and Czechoslovakia, increasingly effective training under Chinese instructors in the camps at Mtwara and in Nachingwea province - and the return of specialists trained in Russia, these resolutions could now be fulfilled. During the OAU Conference at Addis Ababa in 1970 the Liberation Committee renewed its mandate with acclaim. Dar es Salaam now regarded Frelimo as virtually a government-in-exile, albeit with reservations on the supply of arms.tttt For two years Frelimo grew rapidly, until it claimed a membership of 10,000. It had an efficiency and professionalism unmatched even by the PAIGC; and for a while its troops enjoyed freedom of movement over the Ruvuma and the greater part of Cabo Delgado and Niassa. Mueda was ringed by large camps with dugouts and artillery, and early in 1970 access to 1For further amplification see Notes at end of chapter. f- For further amplification see Notes at end of chapter. ;-; For amplification see Note, at end of chapter. tt t+For further amplification see Notes at end of chapter. 142 CABORA BASSA

Revolt in Mozambique the town was cut off except by air. All whites had fled the countryside and the roads north of Montepuez were impassable. The Portuguese Government could not stomach these developments. Afflicted by the mounting cost of the war, by President Kaunda's open opposition to Cabora Bassa, and by the success of the propaganda campaign in Europe, Caetano decided to make a sustained effort at final victory and set a tacit limit of five years on a military-psychological campaign to pacify the province and prepare it for greater autonomy. The war budget was increased yet again, including huge but hidden items for the building of tarred roads (better proof against mines) and the movement of the whole population of the north into aildeanentos. General Kaulza d'Arriaga was appointed, first as commander in the north, and then as Commander-in-Chief, Mozambique, in March 1970. Politically ambitious, paternalistic, and imperialistic in the tradition of Lord Milner in South Africa, highly conscious of the value of publicity and looking uncommonly like a late- Roman legate, d'Arriaga accepted a mandate from Caetano to gear up Mozambique to the point of sustaining her own self-defence. D'Arriaga's command was marked by three innovations: a direct offensive, using guerilla tactics, intended to break Frelimo's hold in the north; a pacification programme, derived from his lengthy study and observation of American tactics in Vietnam; and a huge increase in the black elements of the army. In retrospect, it can be seen as a last throw of the dice successful at first, but condemned by the logic of the Portuguese position to failure. The offensive, styled 'Gordian Knot', began in September 1970 and, unlike previous campaigns, continued through the rainy season of early 1971. To relieve Mueda and destroy Frelimo's permanent bases, the army advanced with a series of artillery bombardments and helicopter-borne assaults, followed up by painstaking mine-sweeping of the roads and consolidation by troops on foot. At best, the assault achieved four kilometres a day, but by the middle of 1971 a degree of mobility had been restored in Cabo Delgado. Despite severe Portuguese casualties, the bases had been destroyed and Frelimo effectively driven back across the Ruvuma. (In the second half of 1970 the Portuguese army lost 200 killed, 1000 143 seriously wounded; Frelimo lost roughly 700 killed, 1800 captured, with wounded unknown. Portuguese casualties were relatively high for guerilla warfare, given the totals involved.) To a great extent, the 'liberated areas' ceased to exist. 'Psychological warfare' - d'Arriaga's favourite phrase in his many press interviews - followed, with little of the earlier concern for economy. In two years 250 villages were built not, as before, to the south of the Makonde plateau, but behind the Ruvuma escarpment, and in the Mueda- Montepuez area, incorporating more than 300,000 people. Towns of up to 5000 people added a genuinely strategic element. With this impetus, the aldearnento programme expanded, so that by early 1974 one million Mozambicans in the north and in 'Fete had been rehoused - a social and economic transformation of enormous significance for the future (see Chapter 9). As a defensive measure, however, its efficacy was questionable. While these villages formed an effective barrier to Frelimo penetration in Cabo Delgado, they failed in the face of guerilla attacks in Tete and Niassa and do not seem to have been a stabilizing factor when Portuguese authority crumbled in 1974. Africanization of the army similarly evoked a variety of responses. To d'Arriaga, and, for different reasons, to the Mozambique businessman and politician, Jorge Jardim, a trained black army could be a potentially loyal element - a body above tribalism, a genuinely Mozambican force. Through the GEs and GEPs,* d'Arriaga consciously tried to create an African military elite - in place of the political leadership lost years before to Frelimo. Undoubtedly the army benefited, and sincewagesof indigenous troopsweremetbytheprovincialgovernment, Lisbon approved. Others, however, deplored this militarization of the province; and Frelimo found it hard to adjust to the existence of 30,000 black troops, relatively highly paid and united, who represented in war a *The GEs (Grupos Especiais) were infantry under black officers, qualified after twenty-three weeks training; they were recruited, from 1970, on a homogeneous tribal basis. GEPs were paratroops, trained in groups of 250 for five months on lines similar to Portuguese commandos. Thev were recruited on a cross-tribal basis, but chiefly from the north. Both were army units, quite distinct from theflechas, employed by the DGS. The great mass of the army, of course, was composed of infantry; between 1968-74 the proportion of blacks to whites rose from 40 per cent to 60 per cent. Few African officers rose above the rank of lieutenant, but the NCO cadres took a substantial share. 144 CABORA BASSA

Revolt in Alozam bique resolute enemy, in peace a dangerous alternative to its own hopes of winning sole legitimate authority. Despite early victories, the Portuguese counter-attack failed to defeat Frelimo, just as in Guinea the gifted General Spinola had failed to defeat PAIGC. D'Arriaga, however, did not follow Spinola's example in talking to his opponents, nor did he seek an alternative solution in Lisbon (where his ambitions lay rather in standing as candidate for the presidency, were Admiral I'omis to step down in the 1972 election). He tended to agree with at least one of Spinola's conclusions, that the war could not be won by either side, except in political terms, and therefore concentrated on his 'hearts and minds' campaign. Frelimo might have been content to pin down upwards of 40,000 troops in the north, but its leadership was too conscious of the wasting effect on opinion inside Mozambique of the pacification programme and of the diminishing returns in support from Europe or OAU countries if the armed struggles were seen to be contained. After the serious reverses of 1970-1, dramatic action was required. Tete province offered the chance of a second front. Frelimo's attitude to Cabora Bassa had not changed since Mondlane's day. Unwary phrases about 'one million settlers' - used in 1969 by the Chief of Staff, General Deslandes, to refer to the black farmers of Tete, but usually interpreted as meaning white Portuguese settlers - and Zamco's use of Rhodesian supplies, served as useful propaganda; Cabora Basssa remained 'a hostile act against the Mozambican people'.5 It is doubtful whether Frelimo privately envisaged achieving more than impeding the progress of the dam and raising its cost to an insupportable level. But it drew support from President Kaunda's break with Portugal and the results of his European tour in 1970, as well as the increasing belligerency of the OAU. At the same time, the dialogue with South Africa pursued by President Banda threatened to cut off access via Malawi, while Tanzania, afraid of further incursions by the Portuguese in 'hot pursuit' advised alternative action on the TFete border. (Several bombing and strafing missions took place across the Ruvuma in 1970-1, and clandestine talks at the UN led the Portuguese to conclude that Tanzanian support of Frelimo might diminish.) 145

The logistic difficulties which had impeded Frelimo in 1968-9 had largely disappeared as a result of the apparent breach between Zambia and Portugal, caused, strangely enough, by the small splinter group, Coremo.t The status of liberation movements inside Zambia was thus regularized. Since Zambian law forbade the passage of armed guerillas, an arrangement was worked out for the delivery of arms by the Tanzanian authorities, and for their conveyance through Zambia, under Government auspicey, to the Mozambique border. Liberation movements were permitted only refuge or 'rest camps' and an office in the Liberation Centre in Lusaka (under the benign auspices of Nkolosi, Kaunda's personal representative). Shielded by this quasi-legal faqade, Zambia was able to disclaim sponsorship of the guerillas while affording a good deal of practical help. Nevertheless the lines of communication from Zambia presented greater difficulties than those to northern Mozambique from Tanzania. Owing to lack of transport or wireless, the local command was constrained to give each of the guerilla groups a specific area of operations, in case they overlapped and hit each other; and the resulting lack of manoeuvre often weakened their effectiveness. Operations in Fete, however, benefited materially from a tacit understanding with the Malawi authorities that, provided no arms were carried, guerillas could cross Malawian territory from Tanzania. By the middle of 1971 Frelimo had built up a force of 1500-2000 men inside Fete, consisting of a number of raiding bands. Considering the high standard of their equipment, t t and the vast desolation of the area, they were not strikingly successful by comparison with operations in the north; but in the very different climate of opinion in 'Fete, mine-laying on roads and rocket attacks on postos and cantinas achieved disproportionate results. In its basic aim of forcing the Portuguese to fall back on fortified positions, Frelimo succeeded without making a single frontal assault. Movement into Fete province had begun in the rainy season, early in 1971. Creeping down the wooded river valleys from Zambia, Frelimo mounted the first of a long series of ;-For further amplification see Notes at end of chapter. ftFor further amplification see Notes at end of chapter. 146 CABORA BASSA

Revolt in Alozambique ambushes. Some patrols crossed the Zambesi, so that the circle of villages around Cabora Bassa - Mague, Chicoa, Estima - was ringed in turn. In one of the earlier engagements, they killed three Rhodesian soldiers at Mukumburu near the southern border - an omen for the future; but at that stage, their main strategy was to cut the roads radiating out from "Fete, in an attempt to make the province ungovernable, and to disrupt the railway from Beira to Moatize and the road carrying supplies up to Cabora Bassa. By the middle of the year they were mining the main highway (at that stage still untarred) from Salisbury via Tete to Blantyre, killing indiscriminately Rhodesians and ,\lalawians as well as Portuguese civilians. Censorship in Mozambique suppressed this news for a while, and in June General d'Arriaga could still issue a communiqud, aimed at reassuring Zamco engineers, 'small guerilla actions could do some psychological damage at Cabora Bassa .... however it is unlikely that they will be less safe in any great city of Europe or America.' It became impossible, however, to hush up the degree of sabotage after September: the Moatize railway was cut only days before President Banda's well- publicized visit to the dam, and large areas of the strategic districts of Mutarara and Changara were admitted to have been subverted. In the most spectacular action, Frelimo destroyed a radio beacon at Tete airport and ambushed the relief operation, achieving notoriety with ten mines where two hundred had failed in the north. The Portuguese reacted with a mixture of haste and brutality, tempered only later by realization of the political significance of this new theatre of war. As the civil government and police had been caught out, the army was given authority ancl Brigadier Rocha Simoes appointed as Military Governor. Five battalions of troops were added, making a total of 10,000 in the province. The army's writ now counted for more than that of the Governor-General in Lourenqo Marques, for d'Arriaga owed responsibility directly to the General Staff in Lisbon. Thus the change can be seen as part of Caetano's strategy of gaining total military victory while still preparing for economic autonomy; unfortunately for his Government, it imposed a role on the army which it could not sustain. 147

To protect Cabora Bassa, Simoes instituted a convoy system for all civilian traffic and restricted all movement other than -by air. These measures were received with some hostility at first, but Zamco eventually accepted isolation as the necessary corollary of security; few who had seen the wrecks of vehicles blown up by mines cared to test the efficacy of the army's mine-detecting. Over the next two years, in proportion to Frelimo's sabotage, precautions were increased: the village garrisons were upgraded in strength, the hills around Songo fortified with artillery; finally, the whole complex was ringed with a double wire fence and a minefield, more than forty miles in length. As soon as the population had been moved from the river basin, into villages around the perimeter of the future lake, the Zambesi was declared a free-fire zone; posters in every village warned the population not to cross, and patrol boats coursed between Tete and Zumbo on the western frontier. Meanwhile the DGS extended its role of internal surveillance. In the north uncovering intelligence about Frelimo had been the army's prerogative, but in Tete the secret police network was expanded from an existing base of informers. At the Centre for Rehabilitation of prisoners at Machara the DGS converted 'selected' prisoners, after their terms of sentence had finished, into guards for the ahteanentos, at the same time retaining them as informers. There is no doubt that DGS information, then and subsequently, was superior to that provided by army sources; the DGS also had the advantage of working closely with the Rhodesian Special Branch and the South African Bureau of State Security (BOSS), both of which had a vital interest in preventing Frelimo penetration of central Mozambique. But this new function weighted the DGS - already an imperium in imperio responsible only to Lisbon, and not to the Government but to the Minister of the Interior - against the army. Where possible, the DGS worked separately, using its own forces on 'search and destroy' missions and gradually evolving the structure of the flecha, a form of paramilitary commando. In the confused months after May 1971, when responsibility for counter-terrorist activity was, in practice, contested between army, DGS, and the civilian police, and when aware- 148 CABORA BASSA

Revolt in Mozambique 149 ness of the extent of support for Frelimo even in the suburbs of Trete induced something close to panic, reprisals occurred. The genesis of the Wiriyamu massacre lies in the nature of the war, of the enemy, and in this confusion and panic, just such conditions as led to the behaviour of US troops at Mai Lai, or French parachutists in Bizerta during the Algerian war. Although the army was not deliberately conditioned, as in Vietnam, reprisals seem to have been institutionalized at platoon level; they had occurred in the north, unnoticed, ever since 1964. (Commando officers admitted to the author that they had ordered the 'wasting' of Makonde villages, though not the slaughter of their inhabitants.) Very few Frelimo prisoners were taken alive in the field before the DGS rehabilitation programme began in 1969, and it is fair to conclude that both Portuguese and African soldiers of the army took revenge for the mutilation by mines of their friends. As one stated: 'I said that I would never kill a man. Everyone says that. But when it comes to it, one is capable of anything.' For its part, Frelimo returned the majority of Portuguese prisoners through the Red Cross, but, on its own admission, usually shot commandos and GEs; and, according to Portuguese statistics, killed more than forty regedores, or headmen, in Tete alone in the twelve months after May 1971. The Portuguese reprisals of 1971-2, however, were directed against villages suspected of harbouring Frelimo, often on only scanty evidence. Terror was met with counter-terror, and the most senior white official in Tete, at least in private conversation, encouraged the practice of selective executions in villages known to have supported Frelimo. The facts of Wiriyamu may never be known,t but beyond doubt attacks on defenceless villages did take place. At least 3000 Nyungwe and Angoni fled to Malawi at the end of 1971 and settled in the Chikwawa valley, where they formed a supportive base for Frelimo. Doctors in Tete and even at Songo that year noticed a surprising incidence, not of kwashiorkor, but of actual starvation, among villagers found wandering, totally disorientated, in districts later alleged to have been the site of atrocities. The consequences widened, as the war developed, +For further amplification see Notes at end of chapter. so that isolated families, fearing to die at the hands of one side or another, began to sell off the goats and cattle on which they depended. Terror in the open countryside and later the cholera epidemic of 1973-4 proved to be powerful arguments in support of the regrouping of population into aldeamentos. Army discipline was sharply tightened up after the storm over Wiriyamu broke in July 1973. But the laxity of authority which had allowed reprisals to continue unchecked rebounded badly on subsequent attempts to win 'the psychological war'. The aldeamento programme of the GPZ, for example, had been subordinated to the needs of the military, to the detriment of the dream of prosperous agricultural development. GPZ methods were too slow and strategically haphazard for the authorities, who watched Frelimo operating against the army from isolated shambas close to Tete and Cabora Bassa. The GPZ was therefore instructed, after transferring the population of the lake basin, to clear first the area between Estima, Mazoe, and Changara, and then Mutarara, on top of the existing programme; by 1973 the time-schedule for creating each village had been cut down so much that the population in subverted areas such as Vila Gouveia was being dumped on half-cleared sites with only minimal provision. In theory, the strategic villages were easy to defend. Built on the grid system, ringed with a barbed-wire fence, guarded by a handful of militia, and under surveillance of the army, they represented a stronger barrier than isolated huts, hundreds of yards apart. But without the goodwill of the population they could not be held at all; and although the majority in Tete, from loyalty or fear of reprisals, proved hostile to Frelimo, several aldeamentos in Changara and Mutarara - regions where force had been used to round up the people - did not. These (such as Cadmera, Cova, and Chinhando), in sharp contrast to the show villages of Estima and Mazoe, were not seen by outsiders, except local missionaries, whose complaints were ill-received by the Church authorities. The aldeamentos facilitated the use of political propaganda, and so long as the Portuguese army exercised authority firmly, 'mentalization' of the population appeared successful. But, just as the long-term aims of the GPZ were altered by the war, so the role of the army itself was jeopardized. It became the agent CABORA BASSA 150

Revolt in Mozambique 151 of pacification. Street posters in towns in Mozambique and in Lisbon showed soldiers associated with medicine, education, welfare. D'Arriaga announced boldly in one of his many discourses to the troops: 'Counter-terrorism is, first, telling the truth to the population, and to the enemy, convincing the minds and conquering the hearts. It is primarily the development, in all aspects, of the people, and only when it is absolutely necessary, capturing or destroying the enemy.' However much - or little - this may have endeared the army to local Africans, it did not appeal to the white population when Frelimo eventually broke through into the settler farming regions of Manica and Sofala. Absurd over- optimism about the application of American methods of psychological warfare to this undeveloped, desperately poor backwater of Mozambique, exposed the army to complaints from all sides in 1973 - from civilians, for inability to provide security, from the DGS, for incompetence in pursuing terrorists, and eventuallv from its own middle-ranking officers who saw the Service foundering in ambiguities, and blamed for failure on both fronts of an impossible war. In this sense, the army revolution of 25 April 1974 grew out of the dilemma of Caetano's Government, as much as out of the actual frustrations of the Mozambique campaign. In strategic terms, the defence of Cabora Bassa encouraged the fortress mentality characteristic of Portuguese colonialism, frequently evinced earlier in the ten-year war. Constant exposure to mines and ambushes had already given rise to a form of war psychosis in the north, despite d'Arriaga's attempts to raise morale. In Tete it showed up clearly in a reluctance to patrol far from the main roads along tracks possibly impregnated with anti-personnel mines. Although the army was numerically large, with 50 per cent of its total effective in the front line, quantity counted less than quality in counterguerilla warfare. Hence the complaints from Rhodesia, heard increasingly after 1972. D'Arriaga's strategy might have been justified if Frelimo had accepted the implicit challenge and ordered a frontal attack on the dam, but it did not. Meanwhile, the decline in morale of the army had two serious consequences: the political will to defend Portugal overseas was lost, owing, at least partly, to the conditions of service of the regular elements which had been grossly neglected; and large areas in the west and north of Tete province were virtually abandoned by troops on the ground. The land around Zumbo on the Zambian border was, in the words of one Portuguese official, 'reduced to a desert'. Every cantina along the once busy road to Malawi had been destroyed, leaving a region like no-man's land. Air defence proved no substitute, and the tacit agreement with Rhodesian forces to fill in the gaps brought its unwelcome results in Frelimo's pact with ZANU guerillas in 1972. At the same time, air raids along the north-eastern border led to clashes with Malawian troops in June 1972 and to something like a breach in the relationship with President Banda. Events in 1972-3 lent temporary credence to the claim that the war in Tete was aimed at Cabora Bassa. Attacks on transport to Songo occurred frequently, leading to sour comments in the South African press about the inaction of the army, and indicating how deeply entrenched was Frelimo support. The railway line to Moatize was disrupted repeatedly. Despite counter-measures, such as the use of mine detectors, and the clearing of a cordon sanitaire either side of the main provincial roads, the authorities could not provide protection everywhere. The Companha d'Uraneo mine, for example, could be reached only by helicopter, while lorries for Vila Coutinho had to be redirected a hundred and twenty miles through Malawi to avoid a perilous stretch of road. In November 1972 Tete airport was hit in a rocket and mortar attack, coinciding with the attendance of Frelimo delegates at the UN Committee on Decolonization, and six months later the model farm at Estima, only nine miles from Cabora Bassa, was shelled with 120 mm rockets. Nevertheless, the convoys of cement lorries running up twice daily from Fete to Songo got through almost unscathed, and plans to supply cement by air-freight proved alarmist. The crises of supply at Cabora Bassa were not caused by sabotage. Frelimo did not hit the vital loads on the railways the generators and turbines, which would have had to be returned to Europe for repair - but only empty trucks; and no long-range attacks on Cabora Bassa, or attempts to subvert the workmen, took place. In private, Frelimo leaders never CABORA BASSA 152

Revolt in Mo:a.ambique claimed more than propaganda value for Cabora Bassa,t and documents captured by the Portuguese in June 1972 show Machel as admitting that 'Cabora Bassa has become impregnable.' As in the case of the Mau Mau, the liberation movement seemed to be set against destroying future assets. From the declining figures of losses in 1973tt the Portuguese concluded that counter-terrorism in Tete had been successful. In September civil government was restored and the army began to concentrate on how best to protect the admittedly vulnerable transmission line through the less secure area of Changara. There were now 15,000 troops (a quarter of the Mozambique total) covering the province. Frelimo had lost quantities of weapons, hundreds killed (ninety-eight in December 1972 alone), and a number of defectors, the most important being Jose Magonga, formerly Frelimo commander, in Dar es Salaam, of the TFete region. But beyond this apparently pacified area an entire new theatre of war has been opened up. It is not clear whether the change in guerilla tactics took place as a response to the lessons of the 'Fete campaign, and failure to delay the dam, or because of the urgency expressed at the Frelimo Congress held in Dar es Salaam in December 1972. Penetration of Manica and Sofala, and the rich white-settler-dominated farmlands around Vila Pery, had been a strategic aim long before it was singled out by Machel in his presidential message in September 1973. Operating south of Changara and Mutarara, Frelimo had the chance slowly and secretly to build up new bases here, and in the dense forests around Gorongosa game reserve, thus bypassing the former barrier of Zambesia and threatening at once the white heartland of Mozambique and the town of Beira. But while the Tete campaign made this advance possible, the new tactics of 1974 - of attacks on white farms, the aldeamentos, and the railway to Malawi - seem to have arisen from deliberate political decisions following the general aims of the Second Congress.6 Two main channels of communication led south from Tete: along the border with Rhodesia to the high forests of Inyanga t For further amplification see Notes at end of chapter. i tFor further amplification see Notes at end of chapter. 153 and the escarpment above Vila de Manica and Vila Gouveia; and through Mutarara and the hills of the Malawi border, across the Zambesi, to Gorongosa. Cooperation with ZANU in north-east Rhodesia (discussed in Chapter 12) helped to divert Rhodesian patrols operating inside Tete and pushed Ian Smith into protesting to Lisbon - an action which drew a magisterial retort from Caetano but did nothing to ease the strain on Rhodesian farmers. In spite of preliminary ambushes near Vila Pery in April and at Gorongosa game park in August 1973, and attacks on the Italians building the transmission line, neither the Portuguese nor the Rhodesian military authorities would admit the existence of more than a few raiding bands so far south. Frelimo's store of equipment improved steadily; rumour soon credited them with portable SAM ground-to-air missiles, although these did not actually come into service until early in 1974, having proved their value in the war in Guinea-Bissau. More important, Frelimo had already attracted widespread support in the south. The Johannesburg Financial Mail pointed out early in 1973 the possibility that urban guerillas might soon operate from the shanty towns around Lourenqo Marques; and a DGS map, captured by Frelimo at this time, shows a heavy concentration of suspects in the Gaza district. For a long time, however, the white population ignored such warnings. Tete was a long way off, geographically and socially, and few of the boulevardiers of the south had ever seen an aldeatnento. Nevertheless, the attacks on the new villages indicated that the pace of the war had altered. From the middle of 1973 Frelimo began to launch regular assaults, rising to five or six in a single night by 1974. In the worst districts, Changara and Vila Gouveia, despite army reinforcement of the militia, many villages were demoralized or even, like Nhacambo (where a massacre took place, later well- publicized by the Portuguese supporters), destroyed. As Colonel Sousa Teles told the Beira District Council, Frelimo's aim was to force tribesmen back into the bush where they could be used to transport supplies and provide intelligence about Portuguese troop movements. Similarities with Mau Mau attacks on local Kikuyu in Kenya, and with guerilla penetration in Malaya, encouraged CABORA BASSA 154

Revolt in lozamibique 155 the view. The same pattern occurred in Niassa and Cabo Delgado, and the ferocity of the attacks suggest a deliberate aim to destroy by violence what Frelimo propaganda called the 'concentration camps', before their inhabitants could appreciate the real benefits of collectivization. In "Fete, Frelimo frequently behaved like conquerors, forcing the awe-struck locals into carrying their heavy equipment further south. Communications overrode other considerations. As the Central Committee representative in Lusaka declared, the war came first, political conversion second. Communications were now grossly over-extended; the bases in central Mozambique were established at enormous risk and after months of clandestine effort, while Frelimo propagandists covered up by extolling the virtues of being able to march long days almost without food. The gamble succeeded. In January 1974 Frelimo attacked the Beira-Umtali railway, severing it nineteen times in a month; it then turned to attacks in the Vila Pery district. In one of these attacks the wife and son of a farmer named Diaz were killed. Frelimo's justification for an act which contravened all its publicity about the need to cooperate with white civilians against the Lisbon Government was that Diaz had been a DGS agent; but the murder roused all the latent fears of the white community. In Beira, Vila Pery, and Vila de Manica, riots broke out, civilians blaming the army for its failure to protect them. The Governor-General, Pimental dos Santos, feared to jeopardize his authority and it was left to the army to restore order. To pacify opinion, General Costa Gomes, Chief of Staff, came out from Lisbon, ostensibly for the funeral, and was constrained to make concessions to white settler opinion, notably by closing down the extravagant army rest centre in Beira and assenting to the supply of automatic rifles to three hundred farmers in Vila Pery. It was a severe blow to army morale, made no easier by the fact that the DGS now extended its operations, and increased the size of its flechas, with barely concealed support from the civil administration in Lourenqo Marques. February brought a different strategic threat, as Frelimo mined the Beira-Blantyre line at Inhaminga, and repeatedly disrupted supplies to neighbouring Malawi. In succeeding weeks the Portuguese failed to safeguard the line and their prestige declined abruptly in the only sympathetic state in black Africa, while Beira, less than a hundred miles away, simmered in the new sensation of insecurity. Meanwhile in the north, after two years of stalemate, the war revived. General d'Arriaga, after serving an extra term of duty, had been replaced by the less flamboyant General Basto Machado. D'Arriaga's policy lived on, however, but with progressively diminishing returns. Comparisons with the Vietnam war, though inaccurate, reflected a sense of hopelessness, despite the completion of one thousand aldeamentos and the building of six hundred more. On Portuguese figures, Frelimo now maintained between 11-12,000 guerillas inside Mozambique. This was the position when revolution broke out in Lisbon on 25 April. Breaking the stalemate of the unknown, unpublicized war in the north, Frelimo had found a way to by-pass the Portuguese army and the inhospitable barrier of the Macua. Partly by chance, partly by design, it had unpinned the defence structure, frightened the white population, and embroiled Portugal with her neighbours in Africa. In the process, Frelimo had developed its political organization, and won the acclaim of countries previously unsympathetic to liberation movements. The period assigned by Caetano for preparation of Mozambique for autonomy was coming to an end, with huge gains in the economic sphere quite unmatched by political cohesion. Above all, the army, primary source of Lisbon's authority, had been demoralized and downgraded by its failure, not in defence (for the Manica and Sofala raids were not intrinsically damaging) but in the impossible task of winning the hearts and minds of an already divided population. CABORA BASSA 156

Notes and references Chapter 6 Revolt in Mozambique I Printed sources for such a historv are limited and inevitably partial. The main book for the early period remains Eduardo Mondlane's The Struggle fir Alozambique (London 1969). In addition Colonialism in Africa 1870-1960, ed. L.H. Gann & P. Duignan (Oxford 1Q70) pp. 301, 332; The Black Alan in Search of Power (collected Times reports, London 1968); and J.F. Kahl's Pro und Kontra Portugal (Stuttgart 1972); further information in Essays on the Liberation of Southern Africa, ed. N.M. Shamuyarira (University of Dar es Salaam 1971). For detailed evidence on the military and political situation: the runs of Portuguese military communiques, published in Nampula, 1967-74; the Frelimo paper, Mozambique Liberation, Nos 1-58 (1965-74); various Frelimo documents captured by the Portuguese; and an article by Jorge Rebelo in Africa Quarterly, No. 7, January 1968. 2 Jo~o Ameal, 'Mostuario do Imperio', O Mundo Portugues (March 1934), J ; 100. 3 Carlos Reis, 'Contribuicgo para a estudo do robustez da raqa Makonde', Boletim Sociologico da Colonia do Alozambique, 24 July 1954. + Note page 136. The first conference of the OAU in 1963 set up a liberation committee, composed of Tanzania, Uganda, Senegal, Guinea, Algeria, Zambia, Ethiopia, the UAR., Congo-Kinshasa, and Somalia, to supply and deliver aid to liberation groups. The Committee was intended to concert planning, prevent corruption and mis-use of arms supply, and to extract regular contributions from its members. Over the next seven years few of its members actually paid subscriptions in full, and its coordinating work amounted to little more than giving pious advice; but at least until the arrival of Russian deliveries, its arms supply, and the recognition on which that was based, remained vital to Frelimo (cf Liberation Essa vs, Dares Salaam, 1971). + Note page 140. Portuguese military budget estimates, total of overseas categories, including extra-ordinary expenditure (all African territories) in millions of escudos 1963 1964 1965 1966 1967 1968 1969 1970 1971 1755m 1758m 2012m 2500m 3500m 4000m 4000m 4000m 4000m plus excesses (actual) - - - 3000m4940m6197m6200m6500m6800m (By 1971, Angola was paying 50 per cent of the cost of her war, Mozambique less than 20 per cent. The total for Mozambique's war was probably at least 5,500 million escudos, equal to the overall estimate of 1966.) t Note page 141. According to investigations (Observer, 6 February 1972) by the Tanzanian police, the letter bomb was posted in Dar es Salaam, but was made of foreign materials. It was one of four; two others were addressed to dos Santos and Simango, but discovered in time. The battery was traced by Interpol to a consignment sold in Lourenqo Marques. Simango's parcel was addressed to Nachingwea camp, at an address which only a handful of Frelimo leaders would have known at that time. + Note page 142. Rev. Uria Simango, a former Protestant pastor, had worked among Mozambique community in Rhodesia in 1957-62. He became coordinator of Frelimo, and was Vice-President in 1964-9, Acting President in 1969-70; after his expulsion he fled to Cairo, and in 1972 joined Coremo as Foreign Secretary.

158 CABORA BASSA f Note page 142. Samora \iachel, born in 1933, is a former male nurse. Trained in Algeria, he organized the 1964 insurrection. He was commander-in-chief of the Frelimo forces from 1166, a member of triumvirate in 1968, and has been Acting President since 1970; he received the Lenin Centenary Medal in 1971. f Note page 142. Marcellino dos Santos, the mulatto poet, studied engineering at Lisbon University. He has been Political Secretary of Frelimo, and Secretary for Foreign Affairs, and acting Vice-President since 1970, as well as Secretary-General of CONCP, the coordinating body for liberation movements in the Portuguese colonies. it+ Note page 142. The disorders of 1968 caused a certain anxiety about the direction of arms supplies, which increased sharply after an abortive rebellion by a group of Tanzanian officers in 1970. The danger of misappropriation of weapons, either by elements favourable to Oscar Kambona, the exiled Foreign Minister, or bv a renewed army revolt on the lines of that of 1964 - a posibility made closely evident in the rise of Idi Amin - forced the Tanzanian Government to insist on strict security and control of deliveries at the port of .\ltwara. 4 Mozambique Liberation, No. 40, September 1969. 5 Mozambique Liberation, No. 43, December 1970. f Note page 146. Coremo captured five Portuguese GPZ employees, on a survey mission near Fingoe, and, copying Frelimo's practice of handing over prisoners to the Red Cross in Tanzania, took them to Lusaka in February 1971. Unfortunately, the Zambian Government, already embarrassed by proofs that Rhodesian guerillas were operating from Zambian soil, refused either to accept the hostages or permit legal entry. Coremo were told to remove them, and they were later murdered, somewhere inside Mozambique. Portuguese anger stemmed as much from Zambian mishandling of the affair as from its outcome, and official protests were backed up by a highly effective blockade of the railway line from Beira. Following a bad maize crop in 1969 and 1970, and a fraternal but ill-judged sale of reserves to China, Zambia experienced a desperate shortage of its staple food. When deliveries in Beira were rescheduled by dockers, so that steel reached Lusaka but maize did not, Zambia was reduced to buying from Rhodesia at four times the normal price. The blockade lasted until July and the dispute was 'finally settled through unofficial channels in Malawi: during the five months, Zambia's vulnerability was driven home upon her, but it left her government determined to break out by completing the Tanzam railway and by working more overtly towards Mozambique independence. f-t Note page 146. Each group carried thirty pound mines (in the later stages, of Czech manufacture, in heavy plastic cases to avoid detection), Kalashnikov and Semenov machine-guns and automatic rifles, bazookas, mortars and, by 1973, recoilless cannon and 60 mm rockets. t Note page 149. Information about the massacre of \'irivamu is alread% copious. In addition to the evidence of the Burgos Fathers, collated in Wir Vamu, by Father Adrian Hastings, there are corroborations by Frelimo, (Mozatnbique Liberation, Nos. 45, 48 (pp. 17-18) and 55 (pp. 23-6), and a counter by the Portuguese: Wlrivamu a Mare's .VSt. Reports appeared in Le Monde as early as 3 November 1972. A UN subcommission has recently gone over the case and adduced new evidence about the other massacres. The most coherent account given to the author is as follows: Following up information that Frelimo was there, a platoon of the army reached Chuwala village, some sixteen miles from 'Fete (the site corresponds, even if not the name, to Wiriyamu) and saw men rtnning away. The women in the village were being held hostage by Frelimo, but the soldiers opened fire, killing about sixty in the huts and the majority of the fugitives. They left at once without burying the bodies and reported nothing. Of the survivors, one woman reached 'Fete, and reported the massacre. The authorities sent out an ambulance, which brought back seventeen survivors, who were treated. Nothing else was done until the scandal broke.

Revolt in Mozambique 159 While journalists scoured the region in July 1973, Caetano asked Jorge Jardim to institute a secret inquiry; and using the female survivor (sister of the boy spirited away by the Burgos Fathers), Jardim and two journalists found the village, exactly as it had been left, with fifty seven skeletons lying around the huts. Jardim's report was never published, but the story appeared in the South African press in substantially this form. In view of the lack of first-hand evidence, further speculation would be unwise; but it does seem that some of the circumstantial detail of rape and mutilation cited by Father Hastings was fabricated later. This is at least partly borne out by Frelimo's own evidence; but it in no way mitigates the circumstances. Shortly after Jardim's report, the local commander in °Fete, Colonel Armindo Vieira, was withdrawn and his troops disciplined. f Note page 153. Cf. Machel's interview with Newsweek, November 1972: 'It is not a question of preventing the building of Cabora Bassa. We are trying to increase the price of it four or five times.' ±t Note page 153. On the Moatize railway, for example, there were in 1972 forty- eight attacks, hitting fourteen trains; 1973, ninety-four attacks hitting only four. 6 See Mozambique Liberation, No. 53, January-March 1973.

Chapter 7 The propaganda war Reverses and leadership struggles during 1968-9 were as much a spur to Frelimo's propaganda campaign outside Africa as was the search for a second front in the war. Conscious of its need for legitimacy, Frelimo, like all African liberation movements, sought world recognition, first by the OAU, then by the United Nations. Yet before 1970 it achieved less success than smaller but better publicized groups like the South African ANC and the Zimbabwean movements, and concluded rightly that northern Mozambique had proved an unfashionable area to fight in. Given Mondlane's awareness of the need for publicity, Frelimo's organization had been geared both to the war and to a wider audience, so that, just as subscriptions went to pay for the Institute or for weapons, depending on their assignment, so the Department of Information had a dual role - combatting Portuguese propaganda and joining with External Affairs in reaching a world audience. Nevertheless, at the 1968 Congress, when delegates from all Mozambique met in one of the liberated areas in Niassa, only a handful attended from outside the movement, none of these from outside southern Africa. Conscious perhaps of the parochial nature of the assembly, Frelimo passed resolutions conveying solidarity not only with fellow movements in the Portuguese colonies - movements in Angola, Mozambique, and Guinea-Bissau, were linked by the loose organization CONCP, an offshoot of the pre-war Liga Africana in the Portuguese colonies - but all liberation movements in southern Africa, with all socialist states and 'progressive countries of the West'. Yet while

The propaganda war 161 Mondlane gave thanks for foreign aid, he made no appeal for external political assistance other than to call upon all countries to continue to isolate Portugal at the United Nations. At the Congress, and in his own writings, Mondlane tried to maintain equilibrium between East and West, careful to offend neither side by affiliation with the other. During the confused period following his assassination, actual subscriptions from Western countries declined, and Russia made it clear that future arms deliveries would be contingent on Frelimo regaining political stability.' The Portuguese decision to build Cabora Bassa gave Frelimo a chance to reverse this trend, just as it assisted them to transform the strategic situation. The theoretical objections to the project were set out some time before the decision to engage in Tete, and may have been designed to test the degree of international sympathy. In September 1969 Cabora Bassa was stigmatized as a fascist-colonialist plot, 'Based on the international cooperation of capitalism . . . it is not to be built for Mozambique or Mozambicans. It will bring profits to the capitalists who actually build it, to the Government of Portugal, to the white settlers in Mozambique, and to the white racists in South Africa and Rhodesia.'2 There was truth enough here for an appeal on issues troubling the consciences of groups in Europe and America which had barely heard of Frelimo's earlier struggles. However, Western governments could hardly be expected, at that stage, to accept that a fellow member of Nato was engaged in a 'hostile act' in her own territory, and Frelimo began by publicizing the very clear disagreements inside Zamco, and between Portugal and South Africa, revealing, then and later, the high standard of its intelligence service. In December 1969 Mozambique Liberation cast a delicate fly on the European water: 'Although the task of freeing our beloved country from foreign domination is the responsibility of we Mozambicans, we must not become too introspective, looking solely inwards to our fight.'3 Material assistance would be needed from abroad, and the Department of External Affairs had to convince supporters that it would be well used. Declining the idea of representation abroad, through the (exclu- sively white) Committees for Freedom, the magazine advocated using Frelimo's own existing representatives in Lusaka, Cairo, Algiers, Stockholm, and New York. Already, in August 1969, a long memorandum had been submitted to the UN Sub-Committee on Apartheid, linking Cabora Bassa closely with South Africa. Frelimo never took the idea of permanent external representatives seriously, possibly because of a fear that 'ambassadors' would involve themselves in compromises unwelcome to the movement - a well-justified suspicion in the case of Miguel Mrupa, once representative at the UN, then delegate in Stockholm, who defected to the Portuguese in 1970 - but it took full advantage of a series of conferences on apartheid and colonialism organized in its favour by European opposition movements. Preparing for the Rome Conference of CONCP in June 1970, Mozambique Liberation declared: 'It becomes imperative for us to develop international solidarity with the peoples of those countries whose Governments support the Portuguese colonial regime. These are, in the main, those countries which are members of Nato.'" In this case, the appeal was directed to a group of trades-unionists, the Communist Party, and CP youth organizations, hostile to SAE's membership of Zamco (p. 68). Sixty- four delegates attended, including representatives of the World Council of Churches, and the resolutions condemned successively not only Cabora Bassa and Portugal, but Nato, the Lisbon-Salisbury-Pretoria military arrangements, and the operations of multi-national corporations in southern Africa. These were fashionable causes for all European socialist groups, and it was perhaps not mere hyperbole to say: 'the Rome Conference opened a new and vast prospective for the victorious development of the revolutionary armed struggle.'5 The most important influence came from Zambia, in particular from the stance adopted by President Kaunda after publication of the Lusaka Manifesto in April 1969. Five years after the break-up of the Central African Federation, Zambia's relations with her neighbours were ambiguous. It was necessary for her to redefine her attitudes towards the rest of black Africa in the aftermath of the injudicious decision to support Biafra in the Nigerian civil war (in which, with Tanzania, 162 CABORA BASSA

The propaganda war Zambia had formed a tiny minority of OAU countries); but even more urgent it was necessary to show how Kaunda's humanistic policy and the demands of OAU membership could be made compatible with economic realities. Zambia's heavy dependence on trade with South Africa (increasing after her deliberate boycott of Rhodesia, following UDI), her reliance on Rhodesia for coal and the free transit of goods by rail, and on Portugal for transport of nearly all imports and the export of copper, enjoined a policy of trimming if not actual cooperation. Although the oil pipeline from Dar es Salaam had been completed, the contract for the Tanzam Railway was only signed at the end of 1969, and the immediate future offered no alternatives to dependence on the white South. Moreover, the state of the Zambian economy demanded reappraisal: owing partly to the fall in the world price of copper, and partly to an over-ambitious national development plan, 1968 revealed the first deficit for six years. South African economists asked if Zambia could survive after Rhodesia's L'DI.6 Rising prices induced political discontent and Nkumbula's opposition subjected Kaunda's party to severe defeats in the 1968 election. These troubles were smoothed over rather than solved by nationalizing the copper industry and replacing Britain by Italy as a source of investment and technology. In this sense, the Lusaka Manifesto was shrewdly judged, because, although originating in weakness, it achieved something like a common Central and East African foreign policy, not unwelcome to the majority of the OAU, and acclaimed by the UN General Assembly. Only the liberation movements, upset at being ignored, demurred. In conciliatory language the Manifesto appraised the antagonisms of southern Africa and offered the white states cooperation so long as they gave evidence of progress towards majority rule. It enhanced Kaunda's fame and its reception (in sharp contrast to that of the Arusha Declaration) seemed to accord him the Pan-African status formerly held by Nkrumah. The fact that it was ignored by Portugal and rejected by South Africa gave deep offence, and, with or without Kaunda's assent, the language of violence supervened. He had been prepared to go some way with Portugal, recognizing that 163 the regime was colonialist rather than racialist,7 and he had hoped for amelioration after Caetano came to power. But though Caetano was confirmed in office by the Portuguese election of 1969, nothing changed. In spite of a long series of unofficial contacts - via Jorge Jardim, a friend of Salazar and of Caetano, and the Ambassador in Malawi, Futscher Pereira Lisbon refused to concede even the principle of eventual self-determination in Mozambique.8 Kaunda was not given the chance to induce Frelimo to cooperate, and, frustrated by Portuguese inaction and spurred on by the activity of Frelimo and Coremo in Zambia, he let it be known that he saw 'no moral responsibility to stop the freedom fighters from continuing the armed struggle'.9 It would have served Kaunda's policy and prestige badly, in any case, to have had the Lusaka initiative pre-empted by the guerillas, to say nothing of their potential leverage in a possible link-up with Nkumbula and the opposition. 1970 was to be the tear of Kaunda's chairmanship of the OAU, and Cabora Bassa offered a unique chance to denounce both colonialism and South Africa's apartheid. In the early part of the year Zambia protested against participation in the finance by Barclays Bank DCO, and the part taken by SAE of Milan; in March the OAU council issued a general condemnation; and campaigns began in London, Rome, and Germany. On 20 May, Kaunda asked the French and German ambassadors in Lusaka if they were going to support apartheid by spreading it further north, or if they were going to invest their money elsewhere. To show that he was serious, he approached the Yugoslav Government, asking for technical assistance in the copper mines, thus threatening Italy's economic preponderance. This first round ended with bland replies from Germany and France, and evidence of Italy's willingness to bargain.t At its seventh conference, in September, the OAU agreed to send a delegation of protest to all countries concerned, under Kaunda's leadership. As he told a French journalist, it was clear that 'the West must be for us or against us.'1o The group visited Bonn, where they claimed that German export credits t For further amplification see Notes at end of chapter. 164 CABORA BASSA

The propaganda war implicated the Government in the moral issues of Cabora Bassa; they visited Rome; London, where a dinner between Edward Heath and Kaunda ended in a quarrel over whether British firms were participating or not; and Paris, where President Pompidou received them with enigmatic courtesy. Finally, in New York, Kaunda appealed to the United Nations to destroy the hated monument; by 71 votes to 10, with 12 abstentions, the Trusteeship Committee gave his request its blessing. (Australia, Belgium, Canada, France, New Zealand, the Netherlands, Portugal, South Africa, Britain, and the United States, opposed.) Kaunda may have meant what he said: in November, the United Transport share in the Zambian Bus Company was taken over by the state. The reaction in Europe was at first defensive. The Italian Government obstructed the award of export credits to SAE; the Germans announced cuts over the period 1971-2 in the supply of corvettes and aircraft for Portugal's own defence forces; and Pompidou promised a 'review' of South Africa's order for helicopters and armoured cars." None of this, however, affected Zamco. SAE found its devious way to build the transmission line. Willi Brandt, the German Chancellor, defended Cabora Bassa as a purely commercial venture, of economic and social benefit to Mozambique, and Pompidou disclaimed any power to affect the investment decisions of French private companies. Kaunda's campaign evoked more opposition than he had imagined. The Portuguese Government, according to Franco Nogueira, was surprised, not at the publicity, but that the previously compliant Kaunda should involve himself. A vigorous propaganda battle followed, Portugal using publicity agencies, sympathetic MPs, deputies, and journalists, in all the countries concerned. B.J. Vorster, the South African Prime Minister, spent some days in Lisbon in June 1970, as part of his European tour, countering the 'arms for South Africa' movement, and he was closely followed by the French Foreign Minister, Maurice Schumann, making the first official visit there for nearly ten years. In spite of the temporary breach caused by Portugal's support for the attack on Guinea in November 1970 (showing an alarming insensitivity to France's commitments in west Africa), Franco-Portuguese solidarity 165 was made even more manifest in January 1971, following a return visit by Rui Patricio, when Schumann consented to support Portugal's application for associate membership of the EEC. Sales of Alouette helicopters for the Portuguese air force were renewed; and at the same time, Leopold Senghor, President of Senegal, welcomed the chance of dialogue with Portugal. Pompidou's specious assurances, and the fact that the student movement, still exhausted after the upsurge of 1968, refused to interest itself in causes outside metropolitan France, ensured a remarkable immunity for the French companies in Zamco, which, themselves, took great care to play down the facts of their participation. Unlike their German partners, the French elements in Zamco issued no statements, hired no public-relations men. It was ironic, for example, that Barclays Bank, involved in financing only at several removes, should face the protestors, while the Banque de Paris et des Pays Bas, lynch-pin of the structure, successfully denied all responsibility. Both the continuing involvement of France in West Africa and the historic memory of Indo-China and Algeria induced caution. In Britain, in spite of sympathetic resolutions at the 1970 and 1971 Conferences, the official Labour Party offered no more than moral support for any of the liberation movements; while condemning British participation in Cabora Bassa, Harold Wilson was careful not to commit his party against a European ally. Members of the Committee for Freedom, however, formed the 'Dambusters Committee', disrupted shareholders' meetings of Barclays Bank, General Electric, GKN, and Charter Consolidated, and helped to persuade the Joseph Rowntree Social Services trust to make its controversial grant of £30,000 to the Mozambique Institute. Even larger sums were raised in Sweden; in addition to widespread private fund-raising operations in schools and youth centres, the Social Democrat Party subscribed, and in 1971 Olaf Palme's Government budgeted for a contribution of £125,000. Scandinavian governments, with few Portuguese links to consider, might take an outspoken line - Norway hosted an anti-colonialist conference in April 1973 and promised firm support to Frelimo - but support in those countries which habitually voted for Portugal in the UN had to be restricted 166 CABORA BASSA

The propaganda war to minority pressure groups. In Holland a vocal group of deputies and churchmen for several years harried the Foreign Minister, Dr Luns, for his support of Portugal in Nato, and in July 1973 the Minister for Development Aid, under pressure from the Social Democrats, announced grants 'to speed up the process of decolonization in the territories occupied by Portugal, and not for humanitarian reasons'. Across the Atlantic the pattern repeated itself: there were protests at share-holders' meetings, against Alcan's supply of aluminium for the Cabora Bassa transmission line, against Gulf Oil for its interest in Angolan oil. An outcry by Harvard students led to a visit to Angola by one of Nixon's special assistants, and Harvard later divested itself of shareholdings in companies associated with Cabinda. None of this pressure, however, was brought to bear with such publicity as attended the denunciation of the German electro-mechanical consortium. At the time of his meeting with Kaunda, Willi Brandt had offered to reply at length on the merits of the case; and as a matter of courtesy, when defending the 'purely commercial' role of German companies, he instructed Heinz Kifhne, the Chief Minister of North-Rhine Westphalia, then on a visit to Africa, to explain the position in person in Lusaka. Ktihne also visited Dar es Salaam, and returned convinced of the justice of Frelimo's cause. He told Kaunda where his sympathies lay and on his return announced at a press conference that Cabora Bassa could light a spark like Biafra or Algeria.12 Later, in his capacity as Chairman of the Friedrich Ebert Peace Foundation, he awarded DM 16 million to Frelimo. Kiihne was too important to dismiss, and the Brandt Government found it almost impossible to dissociate itself from him - not least because part of the Friedrich Ebert funds came from official sources. Its embarrassment was not the least reason for refusing Siemens and AEG their export credits at the second application. West Germany had already suffered a humiliating reverse in Tanzania earlier in the year, when offers of aid had been rejected because of her trading with Rhodesia; the issue was, therefore, very much alive and by the end of 1970 a combination of student groups, Protestant churches, and the left wing of the Social Democrats (including 167

Manfred Eppler, Minister of Economics), had latched on to the Cabora Bassa case, and directed their protests against Siemens and AEG. Students at the Free University of Berlin organized a series of teach-ins on 'West German imperialism'; Wolff Geisler and Eduardo Ferreira, a Portuguese exile, set up committees at Heidelberg and Bonn. The Frankfurt Information Office of Aktion Dritte Welt of Freiburg ensured publicity and conducted its own campaign; letters appeared even in the Siemens and AEG house journals, while the outside of the great Siemens building at Erlangen was daubed with 'Nieder mit dem Imperialismus'. The campaign extended to Hochtief and to Brown-Boveri in Switzerland, where a visit by Diederichs, the South African Finance Minister, in June 1970, was disrupted. At its peak, the campaign included twelve to fifteen deputies in the Bundestag and enlisted support of papers like the Frankfurter Rundschau. Well-organized, particularly at shareholders' meetings and in mobilizing the support of Lutheran churches, it forced AEG, Siemens and Hochtief to react, first by hiring public- relations firms, then by erecting their own intellectual justification3 - 'The dam is neutral: once it is built it is there for enriching whoever runs the country, black or Portuguese' - and finally in engaging in public disputation, like Reformation clergy. After a period of extremely crude counter-propaganda, full of suggestions that the student leaders were financed by East Germany, their tactics became increasingly subtle; for they stood to lose, not by boycotts or votes at annual general meetings but from the impact on their corporate image, particularly where the intake of young future executives was concerned. Above all, they feared that their ability to tender for future contracts in black Africa, such as the Inge Dam in the Congo, might be impaired. Hence the earnest debates with church groups in Catholic Bavaria and Protestant Berlin (where Siemens' main factories were situated) and the extreme sensitivity to press and television comments. In all this Frelimo took little part. The fight against Cabora Bassa was a response to a particular phenomenon in Germany, the confluence of distinct pressure groups on a single, fashionable cause. It had none of the roots in social discontent 168 CABORA BASSA

The propaganda war which had brought about the quasi-revolutionary situation of May 1968, and in mid-1972, as the churches looked elsewhere and the fire went out of the student movement, it died. The companies emerged victorious, regarding the money spent in defending their part as a prudent safeguard for the future. The only lasting effect in Germany had been definition of the political parameters of export credits. In July 1970 the Government had seriously debated whether or not to confirm the original Kredit Anstalt guarantee;4 in deciding to do so, they conceded three arguments: that if not German companies, others would take over; that the money spent on HVDC (high-voltage direct current) developments would be largely wasted; and that, whatever was said in Germany at the time, such projects were of long-term benefit to Africa. These considerations overrode the temporary risks to investment elsewhere (which could be seen, in retrospect, as insubstantial), and in reaching its decision the German Government, which had not possessed colonies since 1918, began to evolve a new theory, that Africa should be a free market, tied to none of the old colonial powers, France, Britain or even Portugal. Frelimo's own propaganda effort concentrated on the theatre of war and political education. Heated arguments in Europe seemed, if not irrelevant, at least inadequate to its situation. 'Seen in isolation, all these activities would not appear to have much effect on our liberation struggle. Questioning directors and officials at public meetings is unlikely to force them to withdraw from profitable contracts. Capitalism does not exist to be moved by moral arguments.'5 However useful the publicity, 'we have no illusions ... our freedom will only come through our own military and political efforts and possibly that of the oppressed Portuguese people themselves."'6 Thus the main propaganda effort went into regular broadcasts in Portuguese and five local languages on Radio Tanzania, beamed at Mozambique, which analyzed the nature of neo-colonialism and propounded Frelimo solutions, and into the publication of Mozambique Liberatilon, which conveyed the achievements and ideology of the movement to an outside audience. The Information Department in Dar es Salaam, however, benefited considerably from research done by the various Committees for Freedom, especially in Amsterdam, 169 published as a series of effective polemics: Portugal and Nato, British Financial Interests in Portugal, Portugal and the EEC, etc.; and from the generally supportive line taken by such periodicals as Jeune Afrique, Anti-Apartheid News, and the fact-sheets of the Africa Bureau. An increasing number of European journalists marched with Frelimo bands into the liberated areas to record their impressions, and the experience of Portuguese deserters, granted political asylum in Sweden, offered useful data for propaganda." The interests of Frelimo and outside supporters coincided most over the question of Nato. With a good deal of accuracy, Portugal was accused of using Nato supplies to lighten the load on her home defence forces. The degree to which these were used in Africa is hard to establisht and is less relevant here than the significance of the charge, which attracted the double enthusiasm of Communists and other left-wing groups. Seen in the context of Nato bases in Portugal and the United States base in the Azores, it contributed to the picture of a Western capitalist conspiracy to bolster up white racialism in southern Africa - a travesty for which the pronouncements of the Portuguese General Staff were at least partly responsible. The real interests of Nato (particularly of Britain and the United States) in offsetting the Russian threat to navigation in the Indian Ocean and in the South Atlantic afforded some hard evidence in support (see Chapter 12). Opposition to Cabora Bassa had been conflated in a short space of time by political attacks on Nato and on white supremacy in southern Africa, thus apparently vindicating the stand taken by Kaunda and generalizing hostility towards Portugal. In this climate of opinion (rather different from the attitudes towards Portugal prevalent before 1970) the churches took over when the student campaign declined. Financial assistance had been given by such bodies as the United Presbyterian Churches in America from the early days of Frelimo for the work of the Mozambique Institute, but it was not until March 1971 that the World Council of Churches became actively involved in condemning both Cabora Bassa and the Cunene development. Later, the Council made a t For further amplification ,ce Notes at end of chapter. 170 CABORA BASSA

The propaganda war 171 highly controversial gift of $60,000 and offered safe conducts for deserters from the Portuguese army, to the detriment of its relations with the Vatican, already exacerbated by the disputes over Wiriyamu. Frelimo had not made much of the Wiriyamu story and was, indeed, caught unprepared by the storm of anti-Portuguese abuse which followed its publication in The Times in July 1973, shortly before Caetano's visit to London for the sixth centennial anniversary of the Alliance. Early accounts of the massacre drew on the evidence of missionaries and were transmitted to London (finally in book form) by Father Adrian Hastings; in a wider sense, too, the churches, both Catholic and Protestant, at last entered the moral debate about the war in Mozambique. For the first time in Britain political leaders spoke out; Harold Wilson stated that he would like to see Portugal expelled from Nato; and the usual ranks of Conservative apologists stayed silent. While the world repercussions died away in the wake of more important news, the Yom Kippur war and the oil crisis, Wiriyamu accentuated a vigorous dispute inside Portugal and the colonies between missionaries and the established hierarchy of the Catholic Church. For at least a century there had been tension in Mozambique over the duties towards Africans of foreign missionaries. Salazar's 1930 Colonial Act made it clear that the Church in all aspects 'shall be protected and aided by the State as institutions of instruction and assistance and instruments of civilization'.' However the role of ally to the government in time of civil war grew steadily more uncomfortable to Spanish and German priests who owed no specific loyalty to the Portuguese Government. In July 1971 the Order of the White Fathers, founded in 1869 to work against slavery in Africa, decided to withdraw, not, as the Superior-General, Theodore van Astern, declared, because of shortage of clergy or persecution, but because of 'on the one hand, the basic ambiguity of a situation where our presence ends by being a counterwitness; on the other, the sincerity of a mission which recoils from having two conflicting faces in Africa ... Too often, certain acts of apostolic ministry, especially those aimed at fpromoting real social justice, are considered as subversive.'" They had failed to obtain the support of the Mozambique hierarchy against police brutality towards Africans, and they criticized a regime 'which shrewdly uses the Church to consolidate and perpetuate in Africa an anachronistic situation which in the long run is a dead end.' Furious at the charge, the Government formally expelled the Order and accused some of its members of sympathizing with Frelimo. The Burgos Fathers also were expelled at the end of 1973; already the majority of African missionaries had been arrested and sent to Machara prison outside Lourenqo Marques, together with twenty Protestant pastors. The reasons were clear enough: as the DGS knew, they were in touch with Africans at local level in a way impossible for the Portuguese hierarchy to be, and the mere fact was held to be subversive, just as their past encouragement of local dialect and culture had been suppressed as harmful to imperial requirements. But by then the pattern was changing. Although the hierarchy was for months reluctant to comment on Wiriyamu - as late as October 1973 Archbishop Ribeira of Lisbon could only bring himself to condemn acts of violence by both sides - a certain pressure from the Vatican, where the missionaries' reports had been taken seriously, made itself felt. The Bishop of Oporto began to criticize the conduct of the war, and in Mozambique in direct opposition to Archbishop Alvim Pereira, who defined the Church's duty as 'help and collaboration in religious instruction with the State and private enterprise',2° the outspoken Bishop of Nampula signed a protest by the Combonian Fathers in his diocese, admitting the people's right to self-determination, and launched a stern pastoral letter to the Bishop's Conference in Quelimane early in 1974. On a Portugal still deeply entrenched in the association of Church and State, with a primitive and traditionally Catholic peasantry, these phenomena had a deeply disturbing effect probably more than anything else in the propaganda war outside Mozambique. Far from being a symbol of a new imperial future, Cabora Bassa had become a byword for higher taxation, military service, and now, apparently, imperial and ecclesiastical dishonour. Frelimo was scarcely known inside Portugal, indeed lacked 172 CABORA BASSA

The propaganda war 173 contact with the underground Communist Party, the revolutionary groups ARA and LUAR, or even with student movements, another sign of its purely Afrocentric concern. But the wave of international feeling in 1970-2 was felt in Portugal either with bitter resentment or, among the liberal intelligentsia, with a certain hope of change. The former opposition, despite a brief appearance at the 1969 election when Mario Soares returned from exile, knew little of the colonial situation or of Cabora Bassa.t Its leaders inside the Chamber of Deputies, Cunha Leal, Miller Guerra, and Sa Carneiro, after the failure of the election campaign, contented themselves with parliamentary tactics and appeared satisfied with progress towards what Caetano called 'autonomy'. But in the universities, where the threat of conscription hung over every student, the case for decolonization was hotly argued as one facet of a new debate about the future of Portugal itself. Coimbra was closed down for a while in 1969 and a year later the Defence Minister, General Sa Viano Rebelo, attacked the universities as 'veritable centres of subversion'. For the first time he mentioned the existence of deserters in Sweden and, in an unguarded phrase, admitted the unreliability of ex-students as officers in the army. This was the nub. The Chiefs of Staff feared worse than draft-dodging. Resorting to a propaganda campaign which emphasized their overseas responsibility - 'the army is not only an instrument of war, it is essential in the development of society' - they covered the public places of Lisbon with posters showing a map of Europe, with the colonies superimposed from Lisbon to Warsaw: 'Portugal in her true dimensions.' More trouble occurred at Coimbra during the trial of Father Andrade, and the law faculty at Lisbon was closed in March 1971 after a demonstration against the war. In the army intakes of 1969-72 can be found one of the strands leading to the Armed Forces Movement revolution. But in other aspects, and particularly in its response to the dictates of the United Nations, Portugal was stiffened, not weakened, by the war of words. Frelimo's final court of appeal lay at the United Nations. + For further amplification see Notes at the end of chapter.

Like other liberation movements in Angola and Guinea, it first sought recognition as a belligerent, and then as a legitimate government-in-exile. Until the 1970s, however, the UN was unsympathetic. In 1969 the UN Economic Committee on Africa had actually commended Cabora Bassa, especially for its likely service to Malawi.' Since Portugal's admittance as a member in 1955, she had claimed that the overseas territories were provinces, not colonies. Gradually, however, by lobbying the Committee of Decolonization, Frelimo was able to channel its protests; in 1970 it was represented before an ad hoc committee visiting Africa. The same year, Australia, Britain, the United States, and Italy, withdrew from the Decolonization Committee, leaving a vociferous but powerless caucus. Frelimo then turned to the specialist agencies, achieved observer status with Unesco and the Economic Committee for Africa, and in February 1972 appeared before the UN Security Council, meeting for the first time in Africa. This rapid progress could not have been won without the support of the Third World; but it benefited also from the changing attitudes of the Scandinavian countries, Canada, Holland, and even West Germany. The Trusteeship Committee invited representation; and in October 1972 the General Assembly voted 98 to 6 in favour of a resolution requesting Portugal to open talks with the liberation movements. (Britain, the United States, Portugal, Spain, Brazil, and South Africa, formed a minority - France voted in favour). For all the inadequacies of the United Nations and for all Portugal's stoical indifference, and her withdrawal from Unesco - 'a body lost in slow degradation' as Rui Patricio put it - these were real triumphs for a movement in Frelimo's position (although less dramatic than PAIGC's application for recognition as a government on the basis of de facto rule in conquered territory). Later the Security Council confirmed the resolution, with the added support of Britain and the United States. The final humiliation came in December 1973, when Portugal's four delegates representing the African territories were disqualified by a majority vote of the General Assembly. Whatever counter-propaganda the Portuguese responded with tended, in these circumstances, to be discounted in advance by the world press. On the admission of Foreign 174 CABORA BASSA

The propaganda war Office officials in Lisbon, Portugal had not been prepared for such condemnation; and a tendency, common to embattled regimes, to look inwards became a dominant characteristic of her political life. In one sense, Kaulza d'Arriaga in Mozambique and Spinola in Guinea were cast as exponents of Portuguese virtues in Africa, and thus success was staked even more nakedly than before on the dual campaign to win the war (following the apparent success in Angola) and the hearts and minds of the population. Former supporters meanwhile fell away; there was consolation in the votes of Brazil at the UN, but Britain's allegiance turned fickle, Australia refused to sell transport aircraft, and France witheld support, leaving only South Africa - never, at best, a welcome coadjutor. Introspection, reliance on inspiration from the past, and Sebastiacdo (the myth of miraculous survival), became endemic, giving rise to comparisions with the dismal conditions of the 1890s. To an audience no longer interested in claims of multi-racialism, Portuguese spokesmen stridently outlined the view of world strategy in which their army in Mozambique stood in the front line against the Communist advance. Kaulza d'Arriaga claimed: 'The neo-racialism of black versus white, organized international. banditry, and Communist neo-imperialism, are indeed the causes of modern world conflict, especially in the Portuguese provinces .... Communism must be defeated in its fight against Western civilization in southern Africa'.2 To Nato countries, preoccupied with the strength of the Russian fleet in the Indian Ocean, there seemed a strong enough case here to merit private acquiescence, whatever was said or done at the UN; and, to some extent, Portugal benefited from the way in which, in Western eyes, Frelimo had been stereotyped and linked to major Communist-bloc objectives, such as the weakening of Nato. Solidarity with China, condemnation of American policy in Vietnam, links with Somalia, and deepening identification with Tanzanian socialism, combined to make support for Frelimo an act of political commitment rather than a gesture of liberal sympathy. In the early 1970s the war of words had not seriously impaired the Portuguese will to fight, although it had affected a section of young national-service officers, and profoundly 175 disturbed the country's religious leaders. These were long-term factors, and their lack of immediate relevance was quite correctly assessed by Frelimo, which had already resigned itself to doing without physical support from the West. Just as the Portuguese gambled on psychological victory in Mozambique itself, so Frelimo concentrated on the ideological elements of the struggle. Within the context of African liberation movements Frelimo's status compared favourably even with that of PAIGC, whose gifted leader, Amilcar Cabral, was assassinated by PAIGC dissidents in 1973. At the OAU Liberation Committee meeting at Accra in January 1973, Frelimo delegates asked boldly for an increase in arms over and above the 50 per cent extra grant of the previous year, and this confidence reflected a new internal harmony, contrasting sharply with the confusion following Mondlane's assassination. In the late 1960s Frelimo's main problem had been to evolve a coherent ideology, in which the antagonisms between different affiliations, tribal or political, could be subsumed. Progress was not made easier by a profusion of advice and aid, offered disingenuously by countries interested in Frelimo primarily as a vehicle for their own policies, and eager to take advantage of the publicity campaign against Cabora Bassa. As an institution, the party evolved rapidly, drawing strength from the circumstantial need for collective leadership, and example from the structure of Tanu (Tanzanian African National Union), with its central committee, local cells, and political cadres. Organized on orthodox Communist lines, Frelimo's authority rested on a four-yearly Congress, which elected a Central Committee composed of forty representatives of all Mozambique districts. This in turn elected an Executive Committee for day to day business. So far the theory, but in 1968, because of the troubles, Congress did not meet, and the leadership crisis was resolved a year later in the Central Committee. From 1969 to 1974 the Executive Committee consisted of Machel, dos Santos, Chissano, Rebelo, Ribeira, and Matsinya (Lusaka representative). The ability to hold a second Congress in December 1972 was evidence of institutional maturity, borne out by the development of poli- 176 CABORA BASSA

The propaganda war tical cadres and the rapid execution of Congress's new policy of extending the war during 1973-4. Structural harmony could only be built on common principles. As Joaquim Chissano said: 'Tribes are a reality tribalism is a political question,'-' and tribalism concealed not only differences of birth and culture, but differing concepts of the country for which Frelimo was fighting. Equally, divergent political affiliations, with East or West, concealed fundamentally opposed views of the sort of society Frelimo wished to establish. Events such as the death of Mondlane helped to resolve some of the conflicts, as Frelimo had then to make a virtue of collective leadership, and defection or death among the dissidents removed the dangers of a complete division, such as that only too evident in Angola, where GRAE (later FNLA), MPLA, and Unitd, angrily disputed the crown of political legitimacy. But Frelimo's relative success was due to its own efforts and the example of its tutors, Tanu and the People's Republic of China. Tanzanian socialism, with Tanu as the sole party, imbued with the doctrines set out in the Arusha Declaration, proved to be far more congenial than Zambian 'humanism', with its acceptance of a mixed economy and inevitable contacts with the white South. Severe, puritan, even ascetic, Frelimo's leaders took over a doctrine based on the stoic acceptance of poverty, believing it appropriate to Mozambique's 8 million subsistence-farming peasants; and the more that they elevated the party and its cadres, the closer they came to accepting the Tanu theory of the one-party state. Tanu, in turn, began to see Frelimo as a kindred force, suitable for cooperation in the long-term future (see Chapter 11). This tutorial, though not paternalistic, influence could be detected in the fact that the first Tanzanian Government delegation to Cabo Delgado in November 1973 was led, not by a minister, but by Chedial Magonja, Southern Regional Secretary of Tanu. Many elements of Tanu policy such as ujamaa, or voluntary collectivization of agriculture, and its attitude towards foreign enterprise and investment, were absorbed into Frelimo thinking by a form of osmosis during the war decade, although this was not made clear until Frelimo had to answer these questions in the wake of the Portuguese revolution of April 1974. 177

What was useful, served: what was not, was rejected; and this pragmatism governed Frelimo's response to Chinese aid in all its forms. Nothing in the recent history of south-eastern Africa has been more contentious than the analysis of Chinese motives, and the opaque language used in Peking offers little clarification. But the accusation that Frelimo is a puppet of either Russia or China is hard to substantiate. After the hasty and inept diplomacy of the mid 1960s China resumed contact in Africa circumspectly, and nowhere more than in Tanzania. Aid to Frelimo, in the form of weapons, food, clothing, medical supplies, or books, began to flow soon after the start of the great Sino-Soviet dispute, and may well have been a function of the prolonged battle against 'Soviet revisionism'; but actual training started only after the arrival of the Chinese contingents building the Tanzam Railway. After 1970, Chinese instructors and engineers gave regular courses in military tactics in Frelimo camps in southern Tanzania. This military aspect can be exaggerated. Expert training, in the use of explosives or rockets for example, was still provided in Russia and the majority of Frelimo leaders benefited from Soviet skills in this way. Chinese training was directed principally to the ordinary soldiers and the NCO cadres, but the fact that it encompassed a very large number of Frelimo's 15,000 activists, that it always included a large element of political education, and that by 1973 almost all the junior military commanders had experienced it, tended cumulatively to outweigh Soviet influence. Moreover the Chinese way of life, on the railway as in the camps, contrasted markedly with that of other expatriate advisers. They demanded neither Western salaries nor Western conditions; they laboured manually, side by side with African -peasants, lived very simply, and generally won respect as missionaries for a faith which, in itself, could be made comprehensible and relevant to African needs. While some of the Frelimo commanders read Guevara, Debray, or analyses of the French Resistance, Maoist teaching on peasant warfare was better suited to Mozambique's circumstances. Above all, Chinese teaching influenced the movement's categories of thinking, inculcating a search for the 'correct line' which perpetuated, among the political cadres, the habit of critical self-analysis. The methods 178 CABORA BASSA

The propaganda war 179 of the Cultural Revolution, if not its moral, could be exported, and in the confused situation within Frelimo served to expose heresies while retaining the heretics themselves under the umbrella of the people's struggle. That struggle itself ramified beyond the war into questions of education, production, and the spheres of hospital work and the women's department. Embryonic and unsatisfactory as some elements may have been (training of nurses lapsed for three years after 1968, and the women's service, in the field at least, rarely developed beyond a system of camp followers), they none the less challenged older, particularly tribal, ways of behaviour. The search was not for equality of women, but for a new women's role; not for classical democracy, but for a form of popular-based authority which owed nothing to traditional chiefly rule. Frelimo acknowledged the debt. On his return from a long tour of China and South-East Asia in the autumn of 1971, Machel proclaimed solidarity with the Chinese and all other patriotic forces; emphasized the need for revolutionary ideology, to identify the true enemy and to inspire the cadres to work, and the need for the army 'the most conscious and political elements of our people' to study and work in the fields as well as to fight.-' Later, in a manifesto celebrating the 1964 rebellion, he extolled Mao Tse-tung's teaching on peasant warfare, and added: 'The glorious victories of the heroic peoples of Vietnam and Cambodia are a valuable contribution, both theoretical and practical, to our fight.'24 But the aim of the struggle was defined by Frelimo itself. Despite its five centuries of history, Mozambique, like most African states, is a wholly artificial creation of the late nineteenth century. While the overthrow of Portuguese colonialism constituted the immediate objective, the long-term policy, and the only way finally to eliminate tribalism, had to be that of creating from this unpromising geography a nation-state. Hence the crucial importance of the Portuguese heritage and Portuguese language, and the extreme delicacy of safeguarding this in the context of African socialism without perpetuating the taint of colonialism. The 1968 attempt to substitute English was never renewed, but advocates of a tribally-based structure emphasized the merits of local languages. Against them, the party leadership argued that Portuguese was no more colonialist than English, and that local languages would impose impossible barriers. According to Machel, It is necessary that we all learn and understand what Mozambique is. It is erroneous to think that it cannot exist independently of Portugal.... Those who do not believe this cannot understand the struggle. To wish to return to tribalism or regionalism is to fight against our revolution.... There are those who think of Mozambique in terms of peoples or the narrow area they live in, [but] when they are confronted with a lasting mission, they feel lost. They must study and learn about Mozambique itself, its size and population, as the true aim of our struggle." However slow to learn the political lesson (and this document frankly recognized the difficulties of such a programme), ordinary soldiers in Frelimo were forced into cross-contacts by the system of mixing tribes at platoon and cadre level. And in spite of backsliding, the unwillingness of the cadres to go to the front, to which Machel's exhortations point specifically, this training had a cumulative effect. Junior leaders as well as senior would avow a determination to fight for years.26 The December 1972 Congress seems to have marked a decisive point in political and strategic analysis.27 Without the military advances described earlier, this could all have been the small change of a minority group, whiling away the tedium of exile. But in the military context it filled certain basic requirements: the need for a vigorously defined revolutionary situation (colonialism in decay), and the need of a working population to be made aware of its fighting aims. This blend of nationalism and socialism was and is unique among African liberation movements. Assurance and pride in its own resources showed in Frelimo's relations with other potential sources of influence. An ingrained resistance to instruction by outsiders, as much as the incompatibility of Guevara's tactics, led to a rejection of aid proffered by a Cuban mission in 1967. Any East German influence vanished after the assassination of Sheik Karume in Zanzibar. Frelimo was also aware, after the European campaigns of 1970-1, that its allegiance was a valuable prize. Mistrust of being caught up in Cold-War polemics about 180 CABORA BASSA

The propaganda war 181 Nato's role in the southern hemisphere showed in declining returns from Russia's supposed leverage: the outbursts in Pravda against Western monopoly capitalism and Portugal's links with South Africa at the time of Wiriyamu may have been welcomed in Dar es Salaam, but Frelimo was cautious of political involvement in areas so far beyond Mozambique. Assurances that China regarded itself as a Third-World country (see p. 302) gained acceptance simply because, on past experience, Frelimo believed that an identity of interests existed in which Russia could not share. Western governments obtained no leverage and in many cases failed even to gain merit as donors. Even the OAU played no more than a minor coordinating role. Political advice would have been unacceptable, for, as the Assistant Secretary- General admitted in 1971, 'the OAU has remained what it was when set up nine years ago, a compromise.'28 The OAU 'committee of military experts' sent representatives to Mozambique in 1968, but learned from Frelimo rather than taught; and a meeting in Tanga, scheduled for February 1969, to set out a 'complete review of the armed strategy to be used for success in the territories under foreign administration,' was cancelled for lack of a quorum.29 If anything, CONCP exercised more influence, since at least common elements existed between Frelimo and PAIGC. This independence, almost isolation, was to be important when the Portuguese finally came to terms with Frelimo; but it did not exclude awareness of the value of balance in wider international disputes. Without bargaining with Russia and China, Frelimo drew benefits from both; after the Yom Kippur war it followed Tanzania's somewhat spurious claim to an Islamic heritage as Machel championed the right of Palestinians to regain their lands in the face of 'Zionist imperialism'. On the eve of the 25 April revolution in Lisbon, Frelimo's aims differed very little from the slogans of the sixties - fight, educate, produce. 'A Luta Continua' ('The Struggle Continues'), concluded all directives. In spite of a number of attempts at mediation, beginning in 1964 with an overture from Oscar Kambona, then Tanzanian Foreign Minister and Chairman of the OAU, to Nogueira, Portuguese Foreign Minister, Frelimo had not evolved any strategy for negotiation

CABORA BASSA 182 until Kaunda took up the role of mediator in 1973 (see Chapter 12). On military affairs and logistics, its intelligence system proved satisfactory, but the Central Committee knew very little of the policy of economic autonomy in Mozambique after 1971. Portuguese colonialism, by definition, appeared monolithic, and when General Spinola launched his devastating analysis, Portugal and the Future, its policy was assessed - not unfairly as reactionary.31 Yet Frelimo did not, and indeed could not, repudiate the Portuguese heritage, culture, or language. Its members mostly retained a desire for links with some form of Lusitanian community, especially, for practical economic reasons, with Brazil. At a more philosophic level, they came to see their mission as that of mediating between the past and the future, and, unlike the parties in Tanzania or Zambia, whose colonial heritage ran for a mere sixty years, found something of value in a transfer of power won by battle, which should confer on them something of the legitimacy of five centuries of Portuguese history. They expected Lisbon to concede the struggle out of exhaustion, not military defeat, on terms honourable for both sides; in retrospect, they believed that this could have happened, had Caetano been more imaginative, in 1971. Frelimo was better informed about the state of Mozambique than, say, the Zimbabwean movements about Rhodesia, and feared two possibilities - either the militarization of the country, following on from d'Arriaga's deliberate creation of a black elite, or a neo-colonial structure directed by white Mozambicans on the basis of the existing corporative National Action Party (ANP). Apprehension about UDI or the rise of an Idi Amin encouraged Frelimo to increase the level of hostilities, as proposed by the 1972 Congress. But the new strategy with its attacks on white settlers and the aldeamentos immediately unleashed forces beyond Frelimo's control. Reasons existed for both actions, but in stepping up the war, Frelimo betrayed the over-confidence borne of reliance on so much political dogma. While the events of early 1974 went far to discredit the army and sparked off the first wave of white emigration, they also lessened the chances of Frelimo inheriting a stable and prosperous country. The Diaz murders provoked considerable dissension in Dar es Salaam, and a long

Architect's impressions of the dam at various stages (i) showing the twin coffer dams, south diversion tunnel, access road and preliminary works. (2) intakes, underground works in the powerhouse, surge chambers, and tail race, together with coffer dams and both diversion tunnels in operation during building of the main wall.

3 The gorge looking upstream with the coffer dams and outlet of south diversion tunnel. 4 Intakes for the penstocks, leading down to the powerhouse, partly completed.

.* , ,,.. 5 Main wall of the dam looking downstream. Summer I973. 6 Aerial view of the site, October 1973, showing the hairpin bends of the access route, crushing plant, storage silos and the cable supply lines. The twin dams (bottom left) protect the tail race from flooding. zzw 7 The wall and the sluice gates half completed. 8 Opposite above Heavy plant in use in the surge chambers. The worst rock fall occurred on the right of this picture. 9 Opposite below Main chamber of the powerhouse with the first generating sets in place. i~~j r »1 II - s

1o Beam crane lowering one of the sets into position. i i Converter sub-station on the plateau above the gorge with banks of thyristors on insulated platforms.

12 Upstream view of the highest flood, April 1974, shortly before it flowed over the unfinished centre section. 13 Looking downstream, August 1974, shortly before closure of the south diversion tunnel (on the right). Lake began to fill in November before the dam wall had actually been completed.

14 Aerial view of the site, August 1974 (compare with Plate 6). 15 Samora Machel, President of Frelimo, with other Frelimo members, addressing a press conference in Dar-es-Salaam on his return from Lusaka after signing the agreement with Portugal for Mozambique independence September 1974.

The propaganda war article in Mozambique Liberation on 'defining the correct enemy' - the soldier, not the white farmer or engineer, who were seen as vital to the new Mozambique. But the magazine was not read in the south, which talked wildly of UDI or of seeking support from South Africa. A more tortuous defence was provided for intimidating and destroying the aldeamentos. These raids, according to Frelimo spokesmen, were designed to liberate Africans from 'concentration camps'. (Alternatively, they were carried out by the black Mozambique army, in disguise.) But in fact they contributed to a wave of misery and disorientation in large parts of Tete and Manica and Sofala, to say nothing of the north, without turning the population further against the Portuguese, and they awakened a hostility among certain tribes, especially the Macua and Nyungwe, which was used to some effect against Frelimo after the Portuguese revolution. The logic of war led Frelimo to appear more ruthless than it had seemed in 1968. Pared down, purged of alien elements, masterful, the party had matured, survived the popularity and blandishments of the European left, and defined its aims. The future might be greyer and less prepossessing than in the heyday of Mondlane, but it was more certain; and to a degree it had been brought closer by the Portuguese themselves, in building Cabora Bassa. 183 Notes and References Chapter 7 The propaganda war I This was admitted tacitly, in Mozambique Liberation, No. 55, June 1973 - 'During the period before the popular line finally emerged clearly victorious, world support was drastically reduced.' 2 Mozambique Liberation, No. 40, September 1969. 3 Aloambique Liberation, No. 41, December 1969. 4 Mozambique Liberation, No. 42, March 1970. 5 Mozambique Liberation, No. 44, October 1970. 6 Financial lai l, 15 January 1968. 7 See D.G. Anglin, 'Confrontation in Southern Africa', in Essays on Problems relating to Zambia's Foreign Policy, ed. S.V. Mtshali: (Lusaka 1970). 8 The offer, however, was not lost on the exiled Socialist leader, Mario Soares, later to become Portugal's Foreign Minister. 9 Essays on Zambia's Foreign Policy, p. 23. f Note page 164. Italy was highly vulnerable: at that time she held the contract to train Zambian pilots and supply jet trainers. Italian firms had just built the pipeline from Dar es Salaam, and were constructing a refinery at Ndola, a car assembly plant, an air force base, and part of the highway accompanying the Tanzam Railway, among other projects. Some contracts, notably the Kafue Dam, were actually lost to Yugoslavia. 10 Le Monde, 10 September 1970. 11 Le .1f tnd., 23 October 1970. 12 Sundayv Times, 25 October 1970. 13 For example, the Hochtief report, Die Baubiude, October 1972, and the well- produced 230-page book answering for Cabora Bassa published by the German HVDC group in collaboration with the Portuguese Information Service, 1972. 14 The Hermes credit, being a purely commercial proposition, required less justification. Schiller, the Finance Minister, told the Frankfuirter .4llgemeine Zeitung: "There should be no political direction given to economic experts in making decisions.' (5 February 1971). 15 Mozambique Liberation, No. 47, June 1971. 16 Ibid. 17 For example, the interviews recorded by Lt. Rocha de Almeida and Captain Jaime Morais, recorded in Aftonblatt, March 1971. t Note page 170. While the ordinary equipment of the Portuguese army in Mozambique was of standard Nato pattern, the 7.62mm rifles and mortars were of Portuguese manufacture. Supplies from fellow Nato countries, such as frigates and submarines from France, corvettes from Germany were of no use in Africa; the Alouette 3, in constant use in Mozambique, supplied by France, has not been proved to have been used in combat. The combat aircraft, Fiat 91 jets, Harvards, and Dornier DO-27s (observation), were mainly bought from South Africa, or second-hand in Europe. On the other hand, the SSA 330 helicopter was used in Angola, and probably also Americanmanufactured napalm. In general, the tenor of the criticism was that Nato supplies freed Portugal from her own European commitments (one division and one armoured brigade), so that she could sustain the wars in Africa. The propaganda war 185 18 Quoted in James Duffy, 'Portuguese Africa', in Colonialism in Africa, II, p. 188. 19 The Times, 26 May 1971. 20 The Times, 20 December 1973. t Note page 173. One opposition member, writing to The Times, compared Cabora Bassa with Kariba and asked why the world was complaining now that Portugal was at last taking positive action to develop her territories. (The Times 12 March 1972.) 21 'A survey of Economic Conditions in Africa,' United Nations, 1969, Part I (E/CN 14 2480 /Rev. I). 22 Commander in Chiefs address to his troops, Nampula, October 1972. 22a Interviews with the author, May 1974. 23 Mozambique Liberation, No. 49, December 1971. 24 Mozambique Liberation, No. 55, September 1973. 25 Necessidade de Aprender, proclamation for the political cadres, based on a speech by Machel, 1 May 1971, issued by the Frelimo Political Commissariat, May 1971. 26 Barbara Cornwall, The Bush Rebels (London 1973). 27 Mozambique Liberation, No. 55 September 1973. 28 Times of Zambia, 11 April 1971. 29 On the inadequacy of the OAU as coordinator of liberation movements, see J. Mtuma in Essays on the Liberation of Southern Africa. 30 Mozambique Liberation, No. 58, March 1974.

Chapter 8 Completion Though shaken by these acts of war, the engineers at Cabora Bassa refused to be deflected. The delays and disturbances of the time-schedule of the dam owed less to Frelimo activities than to the extra work on the site; costs surged dramatically without military intervention. When the attacks moved south from Tete it became clear that, as a symbol and objective, Cabora Bassa had served its purposes; Frelimo leaders were even prepared to admit that to destroy it would have been a moral crime. Willi Brandt's argument, that the dam would one day support an independent Mozambique, had suddenly found contemporary relevance. Zamco engineers noticed that the skilfully-laid ambushes and mines were never directed against the vital electro-mechanical components. By the time that the turbines and generators were transported, Frelimo had turned to attacks on the Beira-Blantyre line rather than the link to Moatize, and Zamco concluded that it was free to finish work on a curious form of sufferance. Nevertheless, Zamco had to contend with the distinct possibility that some of its personnel might withdraw, to escape attacks on their wives and families, a move which could have started panic; and it had to rearrange time-schedules and organize the protection of convoys to ensure a regular supply of cement. In this, Frelimo, by its presence alone, entrenched deeply on the running of the project, and contributed to the geometric progression of rising costs. The minefield, a sinister brown band cutting through the green of the forest, gave a degree of protection, although concern about the possibility of rocket attacks remained. The army spent money on security as freely as it had been denied them in the past; and by 1973 all roads in the province used for supplies had been tarred. Armoured dummies ran on bogies in front of supply trains, to draw the force of explosions, and repair gangs stood by to fit temporary loop-lines where the tracks were cut. "Fete took on the appearance of a garrison town, or a fort in the American West, with lorry-loads of soldiers tearing up dust and buglers sounding 'last-post' at sundown. This reaction to intimidation was absent at Songo, where the Portuguese took infinite care not to alarm the expatriates by a blatant display of precautions. While the restrictions on movement of civilians remained, the feelings of claustrophobia and frustration were eased by multiplying the number of flights to Tete, and finally by putting the airport on a scheduled service. Leave for families was improved, enabling them to visit Beira and the coast - where they learned that, in the austerity conditions following Portugal's fiscal reforms in 1972, the rest of Mozambique envied Songo its ample range of foodstuffs and supplies. Behind the scenes, the DGS kept up its surveillance, carefully screening the Africans who arrived seeking employment at the camp gates. Movement, even for whites, was closely restricted and in fact, as early as 1972, contingency plans existed for flying out Zamco personnel. Such alarmism could only be justified by the hypothesis that Frelimo would shell the township, capture and hold hostage a family, or destroy something as irreplaceable as the cable car. In the event, the only expatriate killed by guerillas was a Siemens driver on the road. The worst period, when there was a distinct slump in morale, ended in 1973, and the companies noticed with some surprise that the news of ambushes had scarcely any effect on their recruiting from Europe. A handful of French and Germans refused to renew their contracts, but exaggeratedly high wages for the former, and the promise of promotion for the latter, generally served to keep numbers constant. Most wives returned, almost unconcernedly, from leave. In any case, expatriates numbered just over a hundred by 1974, against 1,150 Portuguese. Occasionally the war intruded. One Sunday the heavy Completion 187 artillery on the neighbouring hills opened up, giving fire-power to an ambushed post five miles away; but such events, once explained, became a sort of joke. Living with the problem brought a form of acclimatization, even for late-comers like the electro-mechanical engineers, who arrived to the sound of gunfire in the dry season, 'the suicide months' of OctoberNovember 1972. Long-term residents admitted that they had come to tolerate a degree of danger which would have deterred them in the beginning. Only the Italians were actually shot at. One steel- erecting team was machine-gunned while suspended fifty feet up a tower, and another hit mines on the transmission-line track near the black spot of Changara. The local office stopped work for a fortnight, threatening absolutely to withdraw; but as LTA prepared to take over the contract, the SAE board in Milan agreed to pay sufficient danger money to offset the risk. At Songo a sense of unity, of Zamco contra mundun, grew out of being physically beleaguered and morally condemned. In that mixture of nationalities, a certain social pressure operated against gloomy speculation or thoughts of flight. Instead, a great deal of energy went into giving the township an appearance of permanence; some brought in roses, others planted trees; most of the families imported furniture as if setting up house for years. The club began to look less of a safari lodge and more like the centrepiece of a South European holiday camp. But with surprising chauvinism each nationality concentrated on its own affairs, rather than those of Portugal. Few members of Zamco read Spinola's book, and the April revolution took them all unawares. On the far side of the plateau, the Directorate retained its sense of austere detachment. The air service now brought delegates from Mozambique and Angola, and even the metropolis, to see Cabora Bassa. Rhodesian and South African reporters arrived, to benefit from a public relations service which, after dealing with hostile German and British television crews, and critical American journalists, had perfected a bland presentation of the project. Braz d'Oliveira, as befitted the principal civil administrator, who had seen his work grow to maturity, granted interviews in which he emphasized the integrated nature of Zambesi development. But he had to CABORA BASSA 188 admit that little enough, apart from the dam, had actually been begun. All the preliminary works were finished by the end of 1971. Where lorries had once laboured down the vertiginous escarpment, they now ran smoothly down the long tarmac road, straight into the main chamber. On the civil-engineering side, the twin coffer dams, built in the summer of 1971, were reinforced. On the up- stream side, a crest was built, so that the flood should fall, as in a weir, on an inclined plane of massive boulders and break its force; down-stream, the dam was covered with a thick concrete apron, to armour it against trees or rocks shifted when the water overtopped them. In the event, the grouting was of such a standard that seepage in the middle area was reduced to less than 100 litres a second; and when the 1971-2 flood came it washed out no more than 18 inches of the up- stream crest. A year later, the season was so dry that the tunnels took the flood without any overtopping. Meanwhile, excavators dug deep into the great rock buttresses, seeking an impermeable base for the arched wall. The cleft was cut in fifty-foot steps, allowing for drills and bulldozers to be lifted on the cable cars and set down, higher and higher up the gorge. But the quality of the rock undid the original estimates of the surveyors. Fissured and faulted, it denied a firm foundation and the civil engineers were forced to hack out enormous extra quantities. In the river- bed, on the other hand, the estimates were more accurate. After pumping out the space between the coffer dams in the spring of 1972, the engineers completed this section rapidly, hindered only by a continuous fault between the layers of granite, of powdery porous rock, called the lamprophyre dyke. Having drilled and filled it, like a cavity in an old tooth, they began to fill in the foundation with concrete. Their chief worry - of finding vast, partly detached boulders, deposited in prehistoric times - proved unnecessary. By the rainy season, three years after Zamco had reached the site, the foundations were finished, and a tenth of the wall also, which was high enough, in two blocks, to permit a start on the spillways; meanwhile concreting could continue in the main chamber and the upper tunnels, unaffected by the storms outside. Completion 189

The abnormally low flood of 1972-3 enabled Zamco to excavate the surge chambers in the following year, complete the intakes and the greater part of the penstocks, and raise the wall, in several blocks, almost to its full height. For a time it looked as though the delays caused by extra works could easily be made up; but then in November 1973, for no ascertainable reason, the whole wall of the northern surge chamber collapsed, killing eight men and blocking for weeks the construction of that section underground. For a time it was simply walled off for safety; then the mass of rubble had to be cleared, the gaping hole filled up, and the entire wall restored and refaced in concrete. This was the worst disaster of Cabora Bassa, and it added two months' delay. The winter that year started as dry as its predecessor. But then, in the uplands of Angola, the rains fell heavier than usual and for far longer, into March and April, so that a colossal head of water built up, just as it had during the building of Kariba in 1958. There was no danger of Cabora Bassa being destroyed, as had seemed possible at Kariba, but the coffer dams were inundated until June, while down-river the sugar estates of Sena, and towns in Zambesia, suffered the worst floods for half a century. Kariba held back the worst, but because of the previous droughts the Central African Power Corporation had kept the level high, refusing to spill water until the volume of the rains could be seen. By March 1974 there was no choice but to spill from three gates, continuously, for three months, thus holding Zamco back by another six weeks. Fortunately, the decision had been taken to leave two triangular holes low down in the dam wall to supplement the tunnels against such a contingency. Even so, the water overtopped the wall itself several times, threatening to undermine the lower coffer dam; and to prevent it flooding back down the tail-race into the main chamber, two more coffer dams had to be thrown out hastily into the water. Throughout the rainy season, in spite of the weather, the main wall continued to rise as cement was poured in block after block. The spillways, intricate exercises in sophisticated reinforcing, arched out, ready for the gates to be fitted; and when the water fell, in June and July 1974, the holes were at CABORA BASSA 190

Completion 191 last filled, the centre blocks completed, and the down-stream coffer dam slowly removed, preparatory to its complete demolition, at the time that the south tunnel gates were shut in November. By then the changed political situation had removed all cost constraints, leaving Zamco free to pour concrete as fast as it could be shipped on site. In the end, the emergency completed the long exercise of managerial coordination. Bendixen had succeeded in drafting an acceptable Master Project schedule, but, with delays and rising costs, this had to be revised every three months in the later stages, at a cost of continually renewed negotiation inside Zamco and with the Government. The civil engineers, long since resigned to future bargaining over extras, accepted this and formed a common front; but when civil works impinged on electro-mechanical installation, some of the old antagonisms between the French and German companies revived. Faced with the political commitment to finish on time, at whatever cost, the Germans, whose contract did not allow for any inflation, not unnaturally objected. A further strain on the management machinery came with the need to align works at Cabora Bassa with the more advanced Apollo station in South Africa - more a matter for a politician to adjust than a project manager. In both cases the thing was done, not by dictation from boardrooms in Europe, but by cooperation on the site, born of a fundamental consensus between fellowengineers. Chambers used to encourage his colleagues' engineering standards by offering them land in the Zambesi valley - down-stream. In the event, Portuguese supervision, perhaps excessively meticulous on the civil side, gave way to German expertise on the electro-mechanical. Once the underground caverns and the penstocks and tail-races had been finished, it remained to install the machinery, the lines, and the converter sub-station. After long practice, and to ensure that any defects of manufacture should not be over-looked, the German firms had set up a mobile team, especially charged with the installation of machinery. This division of labour meant that, no matter where the contract, the team could work almost independently, relying on civil engineers only for the supporting works. They began to arrive in July 1972, and a year later, Horst Lange, director of the three sections (sub-station, central chamber, and transmission line) took over as site manager from Chapelle, though the latter remained to finish the civil works. In the main cavern each generator set was built up in turn, to avoid overlapping, starting from the southern end. The civil engineering work surrounding this assembly formed the most demanding part of the whole project, for where the turbulent energy of the water, backed by its colossal head of pressure, is directed into the blades of the turbine, there is no margin of error. Once the snail and draught tube are complete and lined with steel, there is little that can be done to repair faults; so the assembly was reinforced meticulously at each stage as the generator sets built up. The great height of the main cavern was slowly reduced until the floor, more than half-way up, hid the machinery underneath, leaving only to the imagination the interaction of water and steel below. Then, into the curved end of the penstocks, the giant cranes lowered the turbines. Forged by Neyrpic- Voight on the Francis pattern, these were the second largest in the world, after those at the Churchill Falls, yet each ninety-ton bulk was machined to a razor's edge and polished like the inside of a watch. The stresses this steel must take resemble those of jet aircraft: at full speed, the turbines' rim revolves at more than the speed of sound. Inside fits the drive shaft, and above, the generator, made up in several factories;t a total weight of 900 tons - needing both of the overhead cranes to take it out for maintenance or repair. The current leaves the generator sets via isolated phase buses at 18,000 Amps, 16 KV, and passes into the transformer chambers, where the voltage is stepped up to 220 KV by banks of transformers, three to each set (four manufactured by Siemens and AEG, five by Cogelex, which was responsible for the busbars and the line to the sub-station). From there it flows by single coil, oil-filled and cooled cables up the shaft to the surface and by overhead line to the AC switchyard and the converter sub-station. Meanwhile, below the sets, the civil engineers again took over, fitting the draught tubes, completing the surge chambers, and lifting the end of the tail-race f For further amplification see Notes at end of chapter. 192 CABORA BASSA in a sort of weir, intended to break the force of the torrent and prevent erosion where it met the river-bed. No deadlines troubled the German HVDC group, or Sorefame, the Portuguese manufacturer of steel gates. Uninterrupted by the floods, their only delay came from the rock collapse in the north surge chamber. And on the plateau, close to Songo township, on a more defensible and less humid site, there grew the sub- station. The intricate division of Zamco's labour obtained here also; Cogelex handled the AC switchyard, Brown-Boveri the DC yard; Siemens and AEG the DC transformers, and the HVDC group the thyristors, whose final design, vetted by the Portugueset and the British consultants, Preece, Cardew and Ryder, had been agreed two years before. In the intervening period, the banks of valves had been built up in Berlin and Erlangen and shipped out; and in an excess of enthusiasm the group added, free of charge, a computer monitoring system of unparalleled efficiency a form of enlightened self-interest, however, since the control desk panel was likely to impress every future visitor to the site. The sub-station existed to convert current into DC and reconvert it to 275 KV AC at Apollo; but it also contained the control room, cooling machinery, and all the ancillary electrical equipment. Cabora Bassa had to be self-sufficient, insured against any possible failure of the tiny Mozambique grid. Thus the 220 KV AC current flows in at 50 HZ through filters which remove the AC harmonics, induced by changes in the wave pattern during conversion. Thence it flows in busbars into the AC switchyard and through converters to the heart of the apparatus, the thyristor valves; emerging as DC, it builds up in bridges, 133 KV at a time, in series, to pass through a DC smoothing reactor (to eliminate ripple effect) and reach the transmission line at full power - 533 KV. But the system also provides for an ancillary supply, to the township and the power house at the dam; and to the sub-station itself, for valve control, emergency services, cooling, switchboard, and the diesel power house - the latter holding two diesel engines to start the whole cycle off and make it totally independent of the outside world. t For further amplification see Notes at end of chapter. Comipletion 193

Above this immense and complicated platform of wires, valves, insulators, and towers, stands the control building with its switchgear and monitoring systems, manned permanently by a crew of the most skilled electrical supervisors, who are responsible not only for normal operation, putting sets in and out of service, in liaison with the team down in the power house and with their colleagues in South Africa, but also for the intricate fail-safe mechanisms. DC is potentially more dangerous than AC if it should get out of control; for whereas the latter will pass successfully through zero conditions, DC will not, making it difficult to cut off. If, for example, one of the transmission lines were cut, there would be only twenty seconds to take a group out, redirect the current through the other line and the earth return, and put the group back, via the auto-synchro mechanism. Malfunctioning or incompetence on the part of the operator could blow the entire South African system, as happened to the East Coast network of America in the great blackout of 1969. Like the twin cranes in the power house, which exist only to lift generator sets for maintenance and repair, highly expensive machinery was imported for the sub- station - the largest fork-lift truck ever constructed (at a cost of a quarter of a million pounds), which may never be used more than a week a year, in order to lift the 132-ton valve banks onto their insulated platform seven metres high. But transmission can never be interrupted and Songo is too far from engineering resources to allow the possibility of breakdown. Similar resources were stockpiled along the transmission line, though with military rather than natural disasters in mind. SAE always refused to contend with questions of security. Despite their encounters with Frelimo, perhaps because of them, the Italians finished a year ahead of schedule, early in 1973. Long before they left, and before the attacks in Vila Pery and Vila Gouveia, the line had come to appear vulnerable. The choice of a route running for its entire length within twenty miles of the Rhodesian border had been deliberate; while on Mozambique soil, the line could, if necessary, be overlooked (literally from the top of the escarpment) by South African police commandos serving with the Rhodesian army. By 1973 that army operated far inside Tete province, in CABORA BASSA 194 conjunction with the Portuguese. But it had never been anticipated that guerillas would be able to base themselves south of the Zambesi for long periods, and when that happened Portuguese assurances that the new lake would prove an insuperable barrier began to be discounted. The DC system provided for a 90 per cent load factor, using the earth as a return, in the event of one line failing. Since the cables rode a hundred feet up, and were made of aluminium as thick as a man's wrist, it would not have been easy to sabotage even one of the two lines. But if the supply road were mined, and the towers sapped by explosives, delays could have risen to a total intolerable to Escom in South Africa. Hence, as Frelimo established itself in Changara and along the escarpment, the authorities switched the bulk of the aldeamento building south, towards the area of Vila Pery, and with imprudent haste rammed the local population into half-finished villages, devaluing the merits of the agricultural programme, while actually preparing the way for political disaffection. Since the army was already overstretched the policy appeared justifiable; but, lacking many of the attractions which had led d'Arriaga to believe the aldeamentos successful, these villages were rapidly subverted and formed the base for Frelimo's attacks on the white settler lands in 1974. A deluge of troubles blighted Zamco's hopes of quiet completion during the last two years. Although the delays of cement supplies resulting from ambushes did not mount sufficiently for the companies to claim force majeure bonuses from the Government, the quality of Dondo cement fluctuated dangerously. A giant new mill, sent up from Johannesburg, failed to prevent a recurrence of trouble at the end of 1972; and in February 1974, when supplies were already impaired by floods on the road to Vila Pery and sabotage of the railway, an excess of gypsum occurred, rendering the factory useless for five weeks. By March only 1600 tons remained in the silos at Cabora Bassa, instead of the planned 12,000 tons, jeopardizing the completion of the dam wall. Yet the French, fearful of sanctions- busting, refused this time to buy from Rhodesia; and after some anguished debate the Overseas Ministry agreed to handle the transaction itself and to ensure that the bags were re-labelled as being of Mozambique origin. Surprisingly, Contpletionl 195

Frelimo ignored the waves of cement lorries coming from Salisbury down the vulnerable escarpment road; and concrete pouring picked up again, until the Dondo factory had been restored. The German group suffered, predictably, from transport problems. Although the Zambesi bridge (designed by Prof. Cardoso and optimistically christened after Caetano) had been completed in 1973, it belied expectations. The transformers passed, but not the 170-ton generator assemblies. Zamco reassigned the contract to French and German transport groups (ATA Vallon, owned by the Paris-Bas Bank, and Kuihnenagel), which renewed and reinforced several critical bridges on the road to Fete and built a special platform on the pier at Beira to redistribute weight when unloading. But at least two loads were sent, in great secrecy, via Salisbury. So much extra work had been undertaken, mainly of rock excavation, that by February 1973 six months had been lost. Such delays Could only accumulate progressively, postponing completion by at least a year. The Portuguese therefore agreed to an 'acceleration' programme, intended actually to restore the original deadline. They paid $5 million for more men and housing and accepted responsibility for 25 per cent extra civil engineering work. The new terms made it clear that Zamco was not to blame. Nevertheless, they bore hardly on the electro- mechanical group, which failed to obtain more cash and was constrained, as part of the northbank scheme (see below), to save a month by combining trial operation of the sets with the formal acceptance tests. To the end, even when spending wildly to achieve its deadline, the Overseas Ministry retained what Zamco called the 'Arabic mentality' of its prime. The acceleration programme recovered part of the time lost, but was overtaken by a combination of further delays caused by the flood of 1974 and poor cement. Awareness of the need to complete on time was then so great that Sorefame, the gate manufacturers, accepted an increase of $7.5 million before obtaining a guarantee to cover it, and their partners agreed to absorb this amount if the Portuguese refused. Cooperation had now become a matter of overriding self- interest, for only by completing their work could the companies hope to restore 196 CABORA BASSA

Completion 197 their bargaining position. Inflation had made nonsense of all the calculations of 1968-9. Braz d'Oliveira admitted that in thirty years of experience he had never encountered rising costs on such a scale. But this could not comfort the signatories of an agreement which bound them to 1968 prices plus a mere 16 per cent to cover inflation in the first stage, up to 1975. At the time, and on past trends, the bargain had appeared hard but not unreasonable; in any case, the Germans had been desperate to win the contract and had not given consideration of costs as high a priority as had the civil engineers. The contract permitted price escalation for hourly labour, but not for overtime or monthly salaries. Even before the world oil crisis, and the spiral of the commodity market, steel costs had risen spectacularly: 15 per cent up by 1971, 65 per cent by 1973. Shortages in South Africa, the main supplier, and the extra costs of transport, added to the burden. Only cement, as a local product, fell within the price escalation clause, and this excluded the cost of buying from Rhodesia. Other local costs ran clear out of hand as price restrictions in Mozambique were lifted in the wake of currency 'reform' - flour jumped by 50 per cent in December 1973, meat by 32 per cent a month later. Local demand-inflation, superimposed on the world crisis, presented an impossible situation for the administration. Overall, the civil-engineering firms faced an increase of 75 per cent on works, only part of which could be reimbursed. But the 'acceleration' programme enabled them to restore the balance a little and the Portuguese admission of urgency gave further leverage for future negotiation. On wages they suffered less because, despite a rapid rise in European salaries, the number of expatriates was falling; and until the strike of May 1974 the wages of blacks were held down by Mozambique law. The electro-mechanical group, however, had neither a price escalation clause, nor the bargaining counter of extra works. Hit by the changed price of copper and aluminium as well as steel, they saw a contract with never more than a faint promise of profit lurch downhill into monumental loss. Up 25 per cent by mid- 1973, their final excess costs probably topped 40 per cent, and the 1972 devaluation of the Rand added to the debacle by reducing the value of South African credits. Only the Italians, who had stockpiled cable and towers before the worst of the inflation, escaped within the margin allowed by the contract. As the terms of the relations between client and consortium changed, giving both a vested interest in completion at any price, the long-debated question of the north-bank power station settled itself by a compromise typical of the last stage of Portuguese colonial rule. The Overseas Ministry had never been able to pin down Escom to a written agreement to buy power in excess of the 1969 figures, nor to hasten the schedule of stage two (sets numbers four and five). Stage three (three sets on the north bank) had been envisaged either as a DC station, waiting on eventual upsurge in the South African power graph, or as an AC station serving Mozambique (which was to draw a mere 75 megawatts from stages one and two, as a by-product from Apollo) and its neighbours. Negotiations with Malawi and Rhodesia proceeded desultorily for some years (Chapter 12), but neither was then willing to consume the power offered. Yet the advantages of building the second power house while the consortium was in being and achieving economies of scale were so tempting, both to the German HVDC group and Portugal, that the Portuguese sought alternatives. (The existence of a second station would also offer a technical advantage, as a form of safety-valve in the eventuality of having to shut down the system; in that case, without the safety-valve, the turbines would spin, producing energy without load, until the intake gates were shut, thus imposing a great strain on the filter system.) A substantial demand for power might have been created by the hoped-for industry in the Zambesi basin; but the strategic situation deteriorated so fast that these plans were shelved. The power excess from stages one and two was recycled for a projected Mozambique grid, and, on the presumption that at some future date the expense of the power house could be covered by agreement with a fresh consumer (rather like Kariba North), the Ministry decided to embark on sufficient preliminary work to obviate the need for lowering the lake level again. CCI and LTA signed a new contract (Hochtief was ex- CABORA BASSA 198 cluded for fear of a recurrence of German protests against export credit guarantees) to excavate the three intakes on the north bank and prepare them for their gates, so that work could later continue inside the mountain. In retrospect, these works must be seen as an act of historical generosity, an uncommitted legacy of colonial rule to emergent Mozambique. Moreover they imposed a further restraint on Zamco, because, although a coffer dam was built to protect them, the decision to excavate came too late for completion before the lake filled up. Its height could not be permitted to rise above the 295 metre level before early 1975. Consequently, even if the tunnels were shut on time on 15 October 1974, and the lake allowed to fill for thirty days, with help from Kariba, the first turbine would have to be tested with the intakes half uncovered, at a level of 286 metres. To fill the lake sufficiently to cover them needed three months more, depending on the rainy season. The final level, of 320 metres, could not be expected until late in 1975. Thus a new set of imponderables had been imposed on the already over-extended master project schedule. To a certain extent the rock fall disaster relieved the pressure, because number three set had to be deferred six months, by agreement with Escom. Nevertheless, the target date of 1 January 1975 allowed a margin of safety which had fallen to roughly four weeks when the Portuguese revolution broke out. Zamco's directors used to hold up conditions at Cabora Bassa as an example of sound labour management transposed from South Africa to an alien environment. There had been no discontent, in the obvious sense, in five years; but absence of trade-union organization, or even of collective bargaining, did not prevent the spread of unrest in the rest of Mozambique. Within a fortnight of the assumption of power by the Junta of National Salvation in Lisbon, strikes broke out in the docks and on the railways, where demands for revindicacbes, 100 per cent wage increases, matched those in the metropolis. The contagion (as Zamco -saw it) infected Cabora Bassa. On 20 May, the native construction camp refused to begin the day shift. Within hours, the black administrative staff and most of the houseboys came out in support of a demand for a doubled wage and for the dismissal of all the South African repre- Comnpletion1 199 sentatives in Zamco. Decades of inaction were replaced by organizational activity, and the self-styled United Group of Suppressed Africans, which was its first result, reflected a coherent political consciousness born of the experience of working on the project. Unsure of its ground, least of all with the military rulers in Lisbon, Zamco offered a 20 per cent increase, which was rejected out of hand by the indunas. Fear of violence from below, of a sort of subterranean rumbling, spread in a Songo now strikingly isolated in a hostile country. For four days the once-stable community appeared to disintegrate, to the point where contingency plans for evacuation were agreed with Ministry officials in Lisbon. Since the police could no longer control crowds in the cities, the army prepared to take over. Braz d'Oliveira was on holiday in Europe and his second-in-command flew to Lourenqo Marques to meet the new Overseas Minister, Almeida Santos, and impress on him that unless the strike were settled the dam itself would be lost. Santos had not looked after Anglo-American's legal interests in Mozambique for nothing; in a long evening of bargaining he put the new Government's authority behind the principle of completion at any price, and the strikers settled for their full wage demand, though not - then - for the political dismissals. A semblance of order was restored; the men and the servants returned to work. But the old hierarchical understandings had vanished, and the new situation confirmed for some members of Zamco the direst warnings of South African advocates of apartheid; for others it revealed that Mozambique might develop the lineaments of a nation. A rapid revaluation of Frelimo followed, a kaleidoscopic rearrangement of established wisdom, in which the only relief lay in Zamco's conviction that the Portuguese Government was committed to seeing its project through. Less than a week had been lost. The situation envisaged six years earlier by Morrison-Knudsen - that if the north-bank tunnel and the holes in the wall were closed, but not the south tunnel, the lake might fill during the rainy season to a point at which it would be impossible to empty it without losing more than a year, and risking damage to the tunnel gates - had not arisen. In spite of the unusually long rains, the dam wall 200 CABORA BASSA continued to rise through the summer of 1974 and Frelimo held back for much of that time, only intensifying its attacks on the railway in the crucial last week of July, while the Armed Forces Movement disputed with President Spinola the manner of independence. In frequent conferences in Lisbon and at Songo Zamco directors tried to forecast the sequence of events, and decided, against the will of the German group, to take the risk of shutting the tunnels to fill the lake, even if the middle blocks of the wall were not ready, and to rely on the spillways to siphon off the flood of 1974-5. In the event, this dramatic choice proved to be unnecessary. The main wall rose steadily in the second half of 1974, the building effort being spurred on, after July, by Frelimo itself. In October, Joaquim Chissano, by then Frelimo Prime Minister in the Provisional Government, came up to Songo, castigated some of the labourers for not working overtime, and exhorted the rest to surpass all previous efforts to finish the dam on schedule for the state of independent Mozambique. The last moment of real danger occurred in November when the gates of the south diversion tunnel were closed. Disaster at the Tarbela Dam in Pakistan, and on some projects in Brazil, where new tunnels had had to be cut at a cost of millions because the original diversions had been badly sealed, pointed only too clearly to the implications if the gates did not fit perfectly when lowered. Experts of all Zamco departments met Portuguese engineers on the site, to check and re-check every detail of procedure in advance. In the last, most exacting test of a project manager's skills. Bendixen's long experience was of immense value. Seepage past the gates ran at one cubic metre per second, but then slowed to a trickle, and the plug, twenty-five metres thick, was constructed behind to seal the tunnel forever. From that day, the newest lake in Africa began to fill up. In January 1975 the spillway gates were also shut, to allow the water level to rise to 286 metres. The first generator was ready for testing a month later, leaving the wall itself to be completed at leisure by June - the month of Mozambique's independence. 'Most of the losses can be made up in the last stages of bargaining', one of Zamco's directors claimed, privately, and Completionl 201 these stages could be expected to last beyond 1975, into a period of prolonged weakness on the part of the client. The German group had actually to maintain a physical presence of twelve months after the acceptance tests to check the working of electro-mechanical equipment during the guarantee period; and there remained stage two (number four set, raising output to 1440 MW on 1 January 1977, number five to 1920 MW on 1 January 1979). But it is already possible to give some answer to the question of whether or not Zamco has achieved its objectives. In the most straightforward sense, to have finished at all conferred great merit on the Directorate and on Zamco's constituent companies. A handful of project engineers, the elite of half a dozen nationalities, had made real the plans and aspirations of two decades of Portuguese planning. Given the hardships and difficulties, the confusion brought about by war and political change, this may rank it with the first Aswan Dam or the Berlin-Baghdad Railway, and its builders with the great nineteenth-century entrepreneurs. But the business of public-works contractors is first to make profits, and then to gain experience', and the publicity which ensures future work. By these criteria, and evidence so far, the Germans will lose heavily, the French slightly, the South Africans and Italians not at all. No balance sheet can ever show Zamco's consolidated profit-and-loss account, depending as it would on individual company circumstances, multiple renegotiation, the performance of different currencies, and the inclusion of items financially unquantifiable. While the French and Italians came in to further existing business, to keep their skilled teams in being, and their plant (in the widest sense) in use, the electro- mechanical firms joined for more complicated reasons. Cogelex needed, in addition, experience in DC work which could not otherwise have been gained; the HVDC team needed a proving ground and found it. Yet both failed to win the Inge Dam contract on the Congo, on which they had staked recovery of the losses sustained at Cabora Bassa, and their dual association broke up as a result. A true profit and loss account can only be taken over many years in the history of a company. The HVDC group was 202 CABORA BASSA committed, by its 1968 decisions, to continue development of the thyristor valve to compete in the export market. The loss of the Inge DC line (to a Belgian group licensing thyristors from General Electric) affected it on two counts, for it broke its potential monopoly of a future southern African DC grid, incorporating the Cunene scheme, Inge, Kariba, and Cabora Bassa, and it reduced the number of open' markets available. At the time that Siemens, AEG, and Brown-Boveri, committed themselves, their competitors were limited to General Electric and ASEA. Six years later they numbered ten, although not all these were likely operators in the same fields.t Real profits - a matter of the utmost concern for a private company such as Siemens depended on construction of at least five sets, after which the pronounced 'learning graph' would even out. Until that point, for each of the HVDC partners, the MannheimRheinau project would remain a continuous expense, with no return other than experience. Since the electro- mechanical side formed a relatively small part of company operations, it could not be justified to shareholders unless that experience, and the potential value of the learning graph, could be exploited. Yet the market itself had become less certain. Apart from Germany, and to some extent South Africa, which may be said to be private preserves of AEG and Siemens, the world offers a few very large projects, and a number of small ones, usually couplings of some hundreds of megawatts at 50-60 HZ. Since world consumption is predicted to double every ten years, the likely demand increase by 1990 is 30-40 GS (MW x 103); and over the next ten years DC transmission may be needed for east-west links in the United States, for joining AC systems in Europe or America (one possibility being that of single- feed cables to the centre of conurbations like Detroit and Chicago, saving greatly in land use taken up by installation), or for the possible east-west tie between four nuclear plants (to be supplied by West Germany) at Kaliningrad (Konigsberg) in Russia, and the West German grid. But however attractive DC transmission appeared when measured environmentally and in terms of t For further amplification see Notes at end of chapter. Comnpletion 203 saving land, it remains more costly than AC. Its future is bound up with political d6tente and the growth of environmental accounting; and while its development in Germany was prolonged by Cabora Bassa, it was not made easier to integrate into normal company strategy. Had the West German Government not intervened with subsidies in 1965, the HVDC group would have broken up then. The future of DC transmission is too bound up with the unstable balance of research and development on the one hand, and production resources on the other, to be compared with anything but the unresolved technology spiral of the aircraft industry. (Thus present research trends are towards small units for compact city- centre stations, while production plants are still turning out large units for projects such as Cabora Bassa.). In this changed world the enthusiasm and illusions of the sixties have been replaced by a wary admission of the need for state patronage, and a recognition that concerns like Howard Hughes Aircraft have benefited greatly from access to political favour. The circumstances of 1968 are unlikely to recur. For the civil engineers, however, Cabora Bassa fitted into the normal pattern and, if anything, reinforced existing trends and prejudices. CCI had always been less committed overseas than Hochtief, and among its constituents GTM regarded 30 per cent as the maximum tolerable percentage for overseas work, pointing to the grave results for Morrison-Knudsen of over-committment in Vietnam, or for UK construction firms like Mitchells in Zambia, thus reinforcing a natural antipathy to leaving France. 'The least exportable commodity in the world is the average Frenchman' - according to a CGEE official. The chairman of Alsthom recently declared that for exports, 'absolute priority still rests with our national customers: it is dangerous to sell too much of your output abroad." LTA, on the other hand, benefited greatly, and rose from being an unprofitable concern to become a leader in the South African field. Shaftsinkers added to an already distinguished reputation, as did SAE, which won the transmission line at Inge, with none of its earlier disappointments over export credits. Anglo-American intentions formed part of a larger whole and are so considered in Chapter 10. Otherwise, Cabora Bassa has become unremarkable. It may CABORA BASSA 204 be remembered as an occasion on which, in adverse conditions, multi-national cooperation became, for a short time, a reality. It will be remembered amongst engineers as a theoretically elegant scheme, where risks were taken within fine margins of calculation with almost complete success; and among public-works contractors it will reinforce caution over contractual arrangements with foreign clients. Although Zamco has been dismembered, many of the dreams associated with its inception survive, with able propagandists like Dr Olivier advocating water conservation and hydro-electric power as the twin buttresses of an apolitical southern African economic cooperation. But what of Portugal, who paid for the dam, and Mozambique, for whom, ultimately, it was built? Was Cabora Bassa a form of expiation for latter-day colonialism, a legacy, as Braz d'Oliveira claimed, 'On l'a construit, pour que les autres sachent profiter, or a last-ditch attempt to realize the 'dream of empire' in concrete? Comipletion 205

Notes and References Chapter ,8 Completion t Note page 192. Stator - Alsthom; rotor - Brown-Boveri; excitation equipment and voltage control - Siemens; the whole i. designed and built for almost any eventuality, including operation at runaway speeds. t Note page 193. An atmosphere of uncertainty, even of hazard, still hung about the performance of the thvristors; and the Portuguese insisted upon an unusual number of evaluations, to the annoyance of the self-confident group, which had already completed the Apollo converter station to the satisfaction of Escom. + Note page 203. These were the HVDC group; GEC (including Canadian affiliates); Westinghouse; Howard Hughes Aircraft Company (diversifying from space research with a small very high-powered research unit at Malibu, California - a late-comer, but developing a revolutionary plasma valve; possessing close political contacts with the Nixon administration, and familiarity with US bidding procedures); ASEA; Japanese firms (Hitachi, Mitsubishi, Toshiba; none in Western markets); the Russians, again not in Vestern markets. 1 Observer, 23 February 1974.

Part III

I Chapter 9 Development or Exploitation? In the political sense, Cabora Bassa ceased to be a controversial issue with the granting of independence to Mozambique. There remained for the new Government in Lourenqo Marques questions of great complexity, such as the apportioning of revenue from South Africa between Portugal and Mozambique, and the future, if any, of the north-bank station and its potential customers. But since Frelimo had decided not to try and prevent its completion in the last and most vulnerable stage, and since the Junta in Lisbon determined to hand it over as a functioning entity, it became possible again to judge the dam in its original conception. The Cabora Bassa scheme cannot be described as a normal development project in an underdeveloped country, at least not in the sense in which such projects are usually evaluated. Benefits for Mozambique were constrained within the limits and understandings of colonial policy directed from Lisbon, and although the project was intended to form the centrepiece of the Zambesi basin scheme, and act as a catalyst of future growth, these intentions were never defined with the interests of a politically independent Mozambique in mind. On the other hand, as time passed, and the Caetano Government decided to plan for greater economic autonomy, the original evaluation of the benefits altered. Collectivization of agriculture, spurred on by the emergency, gave some evidence of that 'self- generating prosperity' beloved of the authors of the 1965 Plan Geral; but the promised industry and mineral development took on a more speculative aspect.

Although the dam was a hydro-electric project, it was quite unrelated to Mozambique's demand for energy even in the distant future. There could never be any question that its cost-benefit analysis rested on the assumption that a very large revenue from South Africa was the desirable - indeed the only means to regenerate the Zambesi basin. Only 75 MW had been assigned for local use (transmitted to Lourenqo Marques from the Apollo station), although this was subsequently increased by the creation of a 220 MW network for the north. The issue of whether to build a series of small thermal or diesel stations in Tete and the surrounding districts did not arise. Thus, even without the political considerations, Portugal would have been unlikely to receive loans from the World Bank or the International Bank for Reconstruction and Development. Cabora Bassa fell into the relatively rare category of projects where the majority of benefits were unquantifiable, and may be compared with the first Aswan Dam, constructed in the 1890s, with the reconquest and rehabilitation of the Sudan as much in the British mind as the future of Egyptian agriculture. Secure in its faith of continued colonial rule, the Overseas Ministry did not seriously question the willingness of foreign investors to take up mineral and industrial concessions, and until the very end took for granted the inevitably long- term benefits to be derived from flood control and river navigation. The GPZ could hardly do otherwise than calculate within fixed political parameters, since these altered little even after the days of Salazar. But, given these assumptions, it undertook a prolonged cost-benefit analysis before signing the contract which continued to guide subsequent development; and it fulfilled adequately sound decision-making criteria: 'In public affairs, the final decision will always be political, and may in part be intuitive. But it will be a better decision if it is preceded by as much objective quantification as possible." Unlike the majority of schemes appraised by the World Bank, Portuguese control of the project, within the context of the overseas budget, obviated the need to ensure that the loans were properly applied and that the expected revenue ensured the service of the debt. There are six other elements of project 210 CABORA BASSA

Development or exploitation? appraisal, generally applicable," of which three - technical, managerial, and commercial - had been assured, respectively, by the work of the Hidro-Technica Portuguesa and the French and British consultants, by Zamco project management, and by the Portuguese Treasury and the export-credit guarantees granted in France, West Germany, and South Africa. The others - financial, organizational, and economic - require analysis, although the final question of whether Cabora Bassa was best suited to Mozambique's overall needs can only be answered within the wider political context. The original estimates, at 1969 prices, did not include the cost of the transmission line and the Apollo sub-station in South Africa, since these were paid for and owned by Escom. Nor did the total of 9,881 million escudost (£125 million) include either the estimated 30 per cent for financing, or the 16 per cent permitted in the contract for rising prices and wages. Traking E50 million as the cost of the South African installations, plus the same extras, the total cost was envisaged in 1969 as being £245 million. The result, as suggested earlier, turned out to be vastly more expensive. Financing costs were higher than 30 per cent, although little of the extra cost fell on Portugal; extra works added 10 per cent to the electro-mechanical, 50 per cent to the civil-engineering works; and inflation put on between 65-70 per cent for wages, 60-70 per cent for materials. Thus the final total cost over thirteen and a half years can be counted reasonably accurately, in the region of £340 million. Although Portugal accepted responsibility for only £290 million (discounting the South African share and a large part of the cost of inflation), this still represents an enormous increase compared, say, with the final cost at Kariba.tt When set against the hypothetical revenue from South Africa, calculated at 0.3 cents per unit, the calculations looked much less healthy than they had done in 1969. On the output of stages one, two, and three, by 1979 South Africa would be paying 6 million Rand a year. This could hardly cover the cost of the actual works within the financing time-schedule and it would certainly not suffice to amortize the South African t For further amplification see Notes at end of chapter. tt For further amplification see Notes at end of chapter. 211 loan within the fifty-year lifespan of the dam* - unless Escom agreed to take more power or unless new customers came forward for the north-bank station. Moreover, there would be nothing to cover the operating costs of the GPZ in the Zambesi area and, more important, nothing for Mozambique. On the original pre- revolution figures, it would have been twenty years before Mozambique benefited in cash from South African revenue - contrary to the sanguine expectations current in Lourenqo Marques in recent years. Moreover, Escom had long since ensured against being liable to be blackmailed into paying more per unit, by making certain that at any point up to 1985 the permanent loss of all Cabora Bassa output could do no more than delay its generating programme by six months.t There was no leverage available to the supplier. Portugal, however, decided to gamble on the growth of the South African power graph exceeding Escom's estimates; and on the increasing desirability of extra hydro-electric power at peak periods of demand at 0.3 cents when compared with other forms of energy during the 1970s and 1980s. Evidence began to accumulate showing this calculation to be correct, and helped to offset the failure to persuade first Zambia, then Malawi, and then Rhodesia, to take enough to make the north- bank station viable. In 1969 calculations had put the cost of Cabora Bassa power below that of the cheapest existing hydro-electric scheme, and at scarcely half the cost of coal or gas. It compared favourably with all nuclear processes except the hypothetical figures for the fast breeder reactor. Considering that the north-bank station would be relatively easy to build, it was reasonable to predict that another 2000 MW of peak power could be supplied to the South African grid, either by DC or AC transmission, at the same price as the first. In the event, the peace treaty with Frelimo made it possible to think again of Zambia, Malawi, and possibly even a transformed Rhodesia, as potential buyers. * A dam such as this has an engineering life of at least a century, but may be badly affected by silting in the annual flood. The machinery, on past experience, has a working life of thirty to fifty years, depending on the quality of maintenance and repairs. The actual lifespan is determined primarily by cost effectiveness - whether it pays, say, to renew worn-out turbines, or to build a wholly new, larger- capacity plant. t For further amplification see Notes at end of chapter. 212 CABORA BASSA

Development or exploitation? A new, political element was introduced with the transfer of power to Frelimo in 1974; and negotiations on the contentious question of who would pay, and whether the revenue from SouthAfrica wouldbedividedbetween Portugaland Mozambique, or assigned to the latter as a final gesture of colonial generosity, only began in January 1975. Nothing had been considered in detail in the Lusaka Agreement (see p. 332). Zamco's contract had been signed with the State of Portugal, not the Caetano Government; hence, even if it had wished to, Zamco could not have dealt with the Mozambique authorities after independence. At stake was the sum of $675 million spread over thirteen and a half years, the first repayments being due in 1975. By then, Portugal faced enormous liquidity problems and a balance of payments problem; yet, as an applicant for EEC membership, could not conceivably run the risk of alienating French, German, and Italian opinion by reneging on the debts. Although Frelimo's attitudes had become conciliatory, its leaders were bound to demand that Cabora Bassa should be nationalized and that Mozambique should take at least a percentage of the revenue from South Africa. The most hopeful solution appeared to be for the dam to be considered a Portuguese asset, liable like that of any foreign company to Mozambique taxes, and managed by Portuguese engineers within the context of Mozambique external politics. Another possibility of surmounting political requirements lay in the output of the north-bank station which was not already pledged to South Africa and was potentially more valuable than the power already contracted from the south bank. As for South Africa itself, the necessary political guarantees to and from Frelimo had already rendered Cabora Bassa energy acceptable; and there was no reason why the north-bank scheme, financed perhaps by the World Bank or by South Africa itself, could not go ahead. At present South Africa's weakest period - a shortfall on base load - is expected to fall in the period 1980-4, before the first nuclear station at the Cape comes into commission. But the nuclear programme has already been delayed by rising costs, to the detriment of projected supply. Escom has as yet no plans to buy north-bank power, but in spite of South 213

Africa's immense reserves of coal Escom's price per unit is already 0.55 cents. Total consumption is doubling every eight years. The dreams of engineers like Henry Olivier are therefore not dead, but dependent on South Africa's external relations. The financial estimates proved fallible, but the capacity of Portugal to organize the project was never in doubt. Critics from Zamco might complain of inadequacy in the test rock-borings, but they acknowledged with a certain bewilderment the expertise of Portuguese engineers, economists, and financial analysts. Beyond Zamco's limited field, the Overseas Ministry provided teams of specialists - agronomists, demographers, cartographers, seismologists, ecologists, sociologists - able and conscientious officials whose only deficiency was that they were too few for the demands of the enormous area covered by the GPZ. Their exiguous numbers were further depleted by war casualties: the GPZ in Tete lost eight men in one mine explosion, and a total of sixty in the years when the dam was being built. Portugal, last of the colonial powers, was for much of her history the poorest colonizer, but in the last decade, under the stimulus of war, she has paralleled the achievements of the British in Central Africa or the French in West Africa. Because of the language barrier, and the narrow circulation of Portuguese scholarly journals, this work is little known and has not been fully credited - a lack of reward resulting partly from a form of exclusive pride (which, for example, led the Portuguese delegation to the 1972 Stockholm conference on ecology to refuse all offers of assistance for research in the African provinces). But the unhurried planning of the Ministry and the slower activity of the provincial government married badly with urgent demands after 1971. Any large scheme such as a steel factory or a coal mine had to be referred to Lisbon. The slow transfer of economic control to Lourenqo Marques scarcely entrenched on the Zambesi area. Confined to annual budgets, the GPZ lacked the political capacity to carry schemes contingent on Cabora Bassa through the Council of Ministers, and was increasingly buffeted, on the one hand by the demands of the security forces, and on the other by the increasing diffidence shown by potential investors, alarmed by the lax progress of the war. CABORA BASSA 214

Development or exploitation? 215 The question, common to most developing countries, of the best use of scarce capital resources, hardly arose, since with the exception of local costs relatively little of the Cabora Bassa burden fell on Mozambique. Some of the province's reserves of foreign exchange were used to pay for heavy plant; and a large item of 1969 million escudos was assigned as a contribution to Cabora Bassa, in the fourth (1974-9) development plan. But even the third development plan figures for Mozambique were at best approximate, and as late as 1974 loans to meet them were still being raised. It is by no means certain what actually was paid out by the provincial revenue before 1974, or even whether the 1969 million escudos was intended to be more than a book transaction. It can be said that, as long as the colonial system lasted, Mozambique's lack of a cash reward was balanced by the absence of demands on her exchequer; and that as the direct benefits, in particular the 220 MW network, materialized, so the charges grew. On the credit side, Mozambique received at virtually no cost (since 80 per cent of the GPZ budget fell on Portugal) the benefits of technology and the experts from the metropolis. The drain on Portuguese resources, of which this formed only part, needs to be emphasized. Taking the Zambesi basin as an entity, there is no reason to doubt that, in theory, the planners were correct in defining Cabora Bassa's economic benefits as being the stimulation of agriculture, forestry, stockbreeding, mining, and industry. In its accumulated plans between the mid- 1950s and 1974, the Overseas Ministry came very close to the ideal of a national development plan - of laying down precisely priority sectors and production targets for project selection, taking into account the progressive discovery of natural resources, the growth of new markets overseas, the upsurge of domestic demand, and demographic changes. But the cost of the dam remained a separate matter, and so the real marginal rate of return on capital invested in it bore no relation to other capital projects in the Zambesi basin, or to their likely profitability. In a way this was an advantage: it allowed for the spending of enormous sums on agriculture and collective villages (for military reasons also) which could never have been justified in terms of reasonable return, and which largely account for the superiority of the aldeamento programme when compared, say, with ujarnaa in Tanzania, or agricultural schemes in Zambia. On the other hand, it encouraged the feeling that any project might one day be worthwhile if the administration waited long enough - an attitude only partly justified by the spiral in world commodity prices in 1973-4. The healthy scepticism of the World Bank or the Colonial Development Corporation was absent in Lisbon, though not, unfortunately, in Johannesburg, London, or Tokyo. In this, as so much else, the colonial habit of mind affected everything. Mozambique, like Ireland in the period 1880-1910, was never conceived of as an entity distinct from the economic sphere of the metropolis, even when autonomy - or Home Rule - came under consideration. It may not be possible for many years to calculate whether the 'spin-off' from Zambesi development has benefited more Mozambique, or the Portuguese and international companies which at present hold the mineral and industrial concessions, or indeed (depending on the re- arrangement of Mozambique's debts to Portugal) the original colonial power. The country had had no semblance of a free market economy since the early 1920s, and the reaction of Portuguese capitalists to the fiscal reforms of 1971-2 (discussed in the next chapter) suggests that the answer will include all these groups. It can be argued that of all the projects only two of the industrial schemes, the steel factory and aluminium processing plant, relate directly to Cabora Bassa; and that everything else could have been done without it. But this is to ignore the correlation between its construction and the war emergency, between the movement of population from the lake basin and wider strategic resettlement; between the search for prosperity and the 'hearts and minds' campaign itself. A qualitative approach must include all aspects - the creation of employment, the benefit in transition from subsistence agriculture, the indirect benefits to health, education, transport, exports, and, via import substitution, to the reserves of foreign exchange. The fundamental problem of African rural development is not so much the transition from subsistence farming to cash cropping, as the change of mentality needed to increase pro- CABORA BASSA 216

Development or exploitation? 217 duction beyond the needs of survival. In a vast area of eastern and southern Africa famine is rare, except in unusual conditions: it cannot be compared with parts of India or Sahelian Africa. Moreover, since land does not belong to individuals in the subsistence farming areas but to the community or tribe, and since there is nearly always more land available for the clearing, there is no compulsion to farm more intensively than before. Even in a country like Malawi, probably the most intensively cultivated in south-east Africa, productivity per acre is retarded by the availability of more land when increased production is required. In the enormous, sparsely populated rural parts of Mozambique a small and relatively primitive effort suffices for family needs; where more is required, the historic pattern has been to leave the land and find jobs in the towns. Various schemes have been tried out in order to improve African agricultural techniques, raise productivity, and earn cash crops, some with the benefit of the region in mind, others with the aim of national self-sufficiency or of preventing the drift to the towns. None have yet succeeded in raising standards on more than a few farms to the level of white-settler farming. The disparity is bitterly evident between, say, the Eastern Transvaal and the Bantustans, between the 'line of rail' and the North-Eastern province in Zambia. Tanzania and Malawi, however, being almost wholly dependent on agriculture, have gone a long way towards adapting subsistence farming within an African context, and consequently have less rural unemployment and squalor. In Mozambique the Overseas Ministry aimed to collectivize the rural population of Tete by slow stages. Until its acceleration under war circumstances, the schedule was as leisurely as the narrow limits of the budget permitted. (The GPZ actually moved 235,000 people between 1970-4 in Tete and Manica and Sofala; of whom only 25,000 came from the lake basin.) The first phase consisted of map-making, soil-testing, and hydrology, followed by decisions about the best sites for future villages, and a series of surveys to discover the numbers of families to be removed, their ethnic group, numbers of dependants, patterns of migration, and their needs over and above the basic amenities of the village. In phase two, wells were drilled, wood for houses cut ready, the population moved by lorry, the plots within the village allocated by the regedor, or headman, the cleared land parcelled out to families, seed issued, and machinery, if any (such as a water pump or plough) either sold or lent on credit by the GPZ. But when war overtook the planners in May 1971, and when the Government proceeded to site aldeamentos in the areas of maximum Frelimo subversion, the part of the GPZ was gradually whittled down and subordinated to the army and the provincial government, which, by the beginning of 1974, had already put nearly one million Africans into six hundred villages, chiefly north of the Zambesi, and was bent on collectivizing seven out of the nine and a half million black Mozambicans. In the process some of the original criteria, about plentiful water, good soil, and a hectare of land, cleared and ploughed (with four more hectares in reserve) for each family, were overriden. A typical village consisted of roughly 1500 people, with perhaps 250 huts grouped on a grid pattern within the square perimeter. Apart from the mud-and- wattle huts, each with its dependant huts for second and third wives, and for friends, there stood the head-man's hut, a concrete schoolroom, a small health centre, a social centre or hall, a well, latrines, and a rough football pitch. The local cantina-cum-shop was usually run by a white or mulatto trained by the GPZ. At their best, where the forest trees had been left for shade and gardens cultivated, the aldearnentos were spacious and apparently congenial places to live in. In some areas the fields lay two or three miles away, but most were nearer, and water was always plentiful. Preliminary studies suggested that when a family had cleared its full area of land it would earn, after subsistence, 12,000 escudos a year on 1970 prices (E200 a year). The programme for Sangara, a typical village in Mutarara district, began with the preliminary studies in 1973 under the auspices of the GPZ and Codam, a South African based construction firm, which cleared the site and dug the wells. In March and April the land was demarcated, the access roads and the village site prepared, and the fields cleared of bush and trees. In May, after the rains were over, the people were brought in - often with persuasion, sometimes by force, 218 CABORA BASSA

Development or exploitation? despite earlier propaganda,t and with the destruction of remaining huts. By June the land was ploughed and graded, the permanent buildings finished; in July it was parcelled out; seed was issued in August, and a campaign of instruction followed, concurrent with the autumn planting. By the year's end Sangara was complete, except for marketing arrangements which were due to follow in the spring, after the rainy season. Collectivization produced a double effect: on agricultural production and on the population. To date, the transition from subsistence farming has been only partial in the 'Fete area. Very few families have begun to clear the four hectares allotted them over and above the ploughed land; and the marketing organization has been delayed by the effects of the war and the consequent undermining of confidence among the inhabitants. The only crop offering a rapid return is tobacco and the evidence suggests that, as in the more developed parts of Zambesia, a good product suffers from lack of expertise in curing. On the other hand, the methods and type of cultivation showed a marked improvement. The seed issued by the GPZ came from strains improved at Mazoe, just as livestock breeding benefited from the work done at the research station at Estima. Taken with better soil and water supply, the increase in returns on time and labour has been dramatic, and the best of advertisements for the new system. A cash economy is certainly developing, with minimum prices fixed by the Instituto dos Ceriais, and a greater acreage of land put under crops such as cotton, millet, and maize, rather than other semi-arid strains. For many of the inhabitants the aldeamento had brought the first use of the plough, irrigation, and experience of livestock; for parts of Fete, when completely cleared of tsetse fly, could be a great ranching country, if the customary view of cattle as wealth by numbers, rather than wealth by quality, can be overcome. In the villages, the benefits to agriculture were widely recognized.tt Relative security from famine and diseases such as smallpox and cholera needed no pointing out. To families harassed by both sides in the war the village and its tiny militia offered a semblance of protection, even if in practice it t For further amplification see Notes at end chapter. tt For further amplification see Notes at end of chapter. 219 amounted to little. Both the stability of the village communities and the fact that where they were disrupted by Frelimo the majority of inhabitants returned, suggest a reasonable level of contentment. Change, however, is cumulative and derives not only from better medicine and education, but the interaction of population in the village. Collectivization involves a high degree of cultural shock, especially among the people of an area where the traditional method of cultivation has been migratory and the family - Nyungwe, Tauara, or Tonga - accustomed to living in some isolation. It is too soon for general conclusions: first impressions among officials and headmen suggest that those under forty adapted well, but that older men could not come to terms with the regular, close-knit character of the new village. A surprisingly high mortality rate from minor ailments such as diarrhoea and conjunctivitis suggests an impaired resistance to disease. But, again on impression, the effect of proximity, and a measure of education and social promotion, has been to create a youthful community, capable of rapid transition by a sort of multiplier effect. The cantina, the road, the schoolroom, open up demands which cannot easily be fulfilled by the traditional once-a-lifetime labour contract in Salisbury or Beira, and which, according to the Portuguese plans, should provide a labour force for future local industry. Whether after the immediate requirements of these consumers (radios, bicycles, sewing machines) have been met, the administration can prevent this culminating in a drift to the towns is another question. Tribal customs are not shed in a generation (women still do the bulk of heavy labour, despite the schemes of 'social promotion') and the Portuguese therefore showed care in delineating ethnic groups within villages, and in preserving, where possible, customs or objects of importance to a largely animist culture. But their overriding aims were to improve physical standards and education. After centuries in which the teaching of Africans had been confined to the religious sphere, on strong imperialist principles, the 'hearts and minds' programme of the administration called on the lay and tiny African elite to take over primary education in the villages. Some idea of the problem may be gained from the fact that in three typical CABORA BASSA 220

Development or exploitation? areas of Mutarara in 1971 only between 3 per cent and 12 per cent had been to school at all, and of those 84 per cent had never got beyond the first (pre-literate) grade. Only 6 per cent to 16 per cent could speak more than a word or two of Portuguese. But in those areas one-third of those of school age in 1971 were actually at school; by 1973, in the aldeamentos in Mutarara as a whole, 50 per cent were in school and on the verge of literacy.3 The gulf between young and old was thus enhanced, and the old regretted it. One elder told the author that he would have liked to join the primary school - 'for it will open my head, and help me to earn more.' The effect, spread across the whole of Mozambique, of this geometric progression towards literacy amongst teenagers was seized on not only by the Portuguese Government but by Frelimo, which saw its significance for the political system which could control the process. While the results are likely to fall short of the aims of 80 per cent literacy by 1980, they compare well with similar programmes in Tanzania and with primary education in rural Rhodesia. A similar improvement can be seen in the field of medicine. Beyond the limited aid given by a handful of missionaries, Tete province in the 1950s offered half a million Africans only a single hospital in the capital, and first aid posts in the administrative townships. By 1973, under the GPZ campaign executed by the indefatigable Dr Maurizio, over 200,000 had been given basic vaccinations; while in the aldeamentos, smallpox, cholera, and typhoid, had been brought under control. Malaria, bilharzia, and other debilitating diseases, however, continued to rack more than half all adults; and in the absence of continuing funds and continuing medical aid many of these gains may be lost in the next five years. The greatest change brought about by the aldeamentos is already clear. Children are living in far greater numbers, as the conditions of childbirth, early nutrition, and sanitation, improve. The birth pyramid of those regions which have been accurately surveyed shows that the mortality rate of children under five has long been around 50 per cent. This is not greatly different from rural areas in neighbouring countries Malawi ranges from 40 per cent to 60 per cent in some parts of the lower Shire River; Zambia around 45 per cent; Tanzania 40 per cent; the Bantustans in South Africa 40 per cent. 221

But adult mortality in northern Mozambique is also high;t and when the rural population is collectivized on such an immense scale, conditions exist for a population explosion. The age of marriage is already dropping and fertility increasing: it is possible that the numbers could double in ten years. Not only will it be difficult to house and educate this new age cohort, but those under fifteen will outnumber those above, imposing an intolerable strain on the ability of the mature population to feed them. Some GPZ officials welcomed the prospect of filling up a province where the average density per square kilometre was less than ten. But the reserves of good soil are few. The development of the Zambesi valley may lead, as the Nile dams have from 1890 to the present day, to a population spiral which wipes out most, if not all, of the gains. On the other hand, it may lead to what demographers call a 'modified Irish curve' - a decline proportionate to the diminishing yield of the land. Collectivization is a gamble, with every aspect - fertility, the age of marriage, size of family, the number of wives, the possibility of migration - variable. There is no easy answer to the question, 'was the aldeamento good or not for its inhabitants?' The Portuguese spent wildly in their last five years. Agriculture, from being the lowest priority in the Zambesi scheme, had by 1974 assumed the lead. The GPZ budget (excluding Cabora Bassa) rose from 131 million escudos to 511 million in 1973. In the north even larger sums went on agriculture and villages, often under military expenditure headings. There can be no clear evidence that this money brought any long-term return in loyalty, although many aldearnentos, especially along the northern fringe of Zambesia, resisted Frelimo infiltration. The collapse came too quickly to test the effect of strategic hamlets. But in terms of the basic needs of the population, it was probably well-directed. Inward-looking, in that the programme emphasized indigenous techniques and resources, it aimed at rural transformation. At the time there was a clear Portuguese expectation that agricultural exports, chiefly cotton, tobacco, and cashew nuts - the production of which rose from 7000 tons in 1963 to 240,000 tons in 1973 - would aid t For further amplification see Notes at end of chapter. 222 CABORA BASSA

Development or exploitation? the balance of payments. Now that the colonial link has ended, the world market will determine the future. Such evidence as exists suggests that the returns in the next decade will be modest but valuable: a higher standard of living, without endangering Mozambique's self-sufficiency in food. A full comparative study of rural transition in Mozambique, Tanzania, Zambia, Malawi, Rhodesia, and South Africa, might throw up surprising conclusions. All that can be suggested here is that of these only Tanzania, Malawi, and Mozambique, have come to terms with the problems of the transition of agriculture to a market economy while maintaining population stability. Only Tanzania aims at collectivizing such a high proportion of its people as Mozambique. The doctrine of ujamaa enunciated by Nyerere and canonized by the party, Tanu derives partly from Saint-Simon and partly from a somewhat idealized conception of traditional village life. Land for all, a community of extended families, voluntary organization, and socialist precepts, inspire the vision of villages which differ from the Mozambique aleamento chiefly in their lack of regimentation. The methods, the surveys, the grouping of population, even the progress towards a cash economy and state-run marketing organizations, are remarkably similar, as is the impact on the inhabitants. Only the military aspect is completely absent in ujamaa, as the political was in Mozambique. Tanzania, as Nyerere has never flinched from admitting, is a poor country and, rather than choose to industrialize or capitalize agriculture in the hope that a rapid rise in GNP would eventually filter down to the masses, its Government prepared to attack poverty directly, concentrating its modest resources on the basic needs of nutrition, shelter, education, and health. Lacking the capital and the mechanical resources of the Portuguese, and the strength of central dictation, ujamaa has proceeded haphazardly, often without coordination (as in the summer of 1971, when over 100,000 peasants were stranded in Dodoma, their villages uncompleted). The intention is eventually to apply it throughout the land, even in areas like Bukoba and Wagogo, where peasant proprietorship obtains, or Iringa, where it has been met by armed resistance. Roughly 10 per cent (one million) had been resettled by 1972. 223

Ujamaa's virtues are as obvious as are the defects of voluntary activity - inadequate preparation, soil surveys, well-digging, transport. Many villages have grown out of control, to 5000 or more; they are less well served by communications, or by the demands of the internal market. The standard of education is probably higher than in Mozambique and established villages, especially in the poorer, southern region, show a strong social and political awareness. On the other hand, there is no evidence yet of a transition from primitive communalism to a cooperative or collective system, and their institutions, even for collective decision-making, remain based on tribal patterns. Comparisons are hard to make. But in practical terms, the aldeamentos of Mozambique have been set up and maintained in a way which, though far from Portuguese intentions, could lead to the same end as ujamaa - that peculiarly African brand of socialism which has been Nyerere's lifetime vision. Frelimo, which has its own well-developed ideas about the relation between production and politics, may well choose to perpetuate the marketing arrangements, and the inducements of the Mozambique Institute of Labour against a drift to the cities, which it finds already in operation; and may yet thank the colonial administration for so soundly preparing the way. It is not likely to be interested in learning from Rhodesia (where the building of fortified villages, more like the 'concentration camps' of Frelimo propaganda, was a late reaction to the guerilla war) or the South African Department of Bantu Administration. But Frelimo may react against the example of Zambia, whose soil is as fertile as Rhodesia, yet whose agriculture in a bad year cannot feed its relatively small population of four million. Zambia is a rich country, with easy access to external loans, yet her Government has failed to stimulate either collective rural development or individual peasant proprietorship. Cooperatives, intended to enable family farms to generate sufficient income for the family to enter the cash economy, have rarely succeeded, despite large loans to the agricultural sector, including, since 1968, 45 million Kwacha (£22m) from the World Bank. High wages in the towns, unaccompanied by countervailing measures, and in sharp contrast to low agricultural returns, CABORA BASSA 224

Development or exploitation? lead to a drift to the towns. Lusaka, like Kampala, Nairobi, or Lourenqo Marques, is surrounded by sprawling and squalid suburbs and shanty towns. Its population has increased four fold since independence. A mere 10 per cent of Zambians live on the land, compared with 88 per cent in Malawi. The result is that, in a year of poor harvest like 1971, Zambia may have to import food, a habit permitted by her favourable balance of payments. Yet, as the OECD Development Centre suggested recently, the real cost of changes which lead to unemployment or under-employment of rural labour and to trapped, backward sections of the community, 'may be beyond economic calculation." The reasons range from a fundamental lack of stimulus to produce beyond daily needs, to the lack of coordinated administration; a preference - and ability - to pay for imports ratherthanreorganizeinternalsupply;atotalabsenceofwagespolicy; the absence of overriding political or ideological commitment to collectivize villages; and a deep political cleavage caused by the contrast between the efficient white- settler farms on the 'line of rail' and the mass of peasants elsewhere. A few middle-class black farmers can compete successfully, for example, in vegetable growing for the city market; but in the production of maize, the test case, the emergent sector has failed. Cabora Bassa's other hoped-for offshoots scarcely began to materialize before the colonial period ended. The attached map shows the rich variety of minerals lying in the ancient rocks of Tete province; iron, titanium, magnetite, copper, chrome, nickel, gold, beryl, fluorite, lead, manganese, coal, bauxite, graphite. Not all existed in commercial quantities, and for some years lack of resources inhibited the making of precise mineral surveys. By 1972, however, exploration had led to preliminary concessions in the most obvious areas.t These concessions were taken up primarily by South African firms in association with Portuguese, and since the Mozambique authorities, true to their almost nineteenth century doctrine of minimal state intervention, put nothing into t For further amplification see Notes at end of chapter. 225

\ubx ~Portugiia, Luluabourg ______Kapanga*~ig Lund mal ne H-enrique de*Kma Maa ______TANZANIA AZtZP rnAboim TGa eI ii vixMtara Reod Cea -da Silva' be9 i-q fos--ýo z r Mtvura ______Roert Silva Pot Luso L ' iLumbashi 4ý4~ ______Gbto Wi lams.Nova Lisboa Czmo SIz Begea ANGOLA Kitwe 4\ MOZAMBIQUE ~~arano *Chitembo dI1 caeså Bani,ä *Serpa Pinto ZAMBIA ' " yila Cabral .~ZPort 1'9i-_ CabrN ampula 10Ieanr Culto Cuanavale *LusW: Bassa Dam ,Zma Ncl Cunene * oaa ~"ete Blantyre River DaniP 'av Tsumeb SNorbrv ______Wanl, le RHODESIA Utl hn& Grootfontein X-- 'Geo'~Bia______SOUTH WEST AFRIlCA Franicstown Buaao Serowe* 13, 1hridge MassangenaZZZ Swakopmund . *in ______Walvisbay *Windhoek BOTSWANA ______I_ rAessrd Loui trichards GibonGaberones Petersburgå 51Limpopo ______Kany 'Pretoria lnha __r___me_KetmnhopKrugersdorp. / *WItbank ý d ilde JoaoBe o______Luderitz *JohannesburgwaýSpriflgS iLourenco Marques - Potchefstroom . c. *Benons ______t- SWAZILAND______Orange Rivel Irne~ .{. ý,hrsBa = = &- Parvs 19, Cch Bav BloemfonteiPietma tzbuP gt SOUTH AFRICA LESOOt Du ~ ~.Victoria West ______~ 2Beautort West *______Graaff Reinet EsL ndn - ~ *Mamesbur Worcest <~ ------.-idp ~ Paarl______-- -- aaL CapeAguhas Port El izabeth:5ý ---______---- ___ ------

Development or exploitation? the enterprise except the information from their own geological surveys (which turned out to be sketchy), the capital also came from South Africa. Johannesburg Consolidated Investments (JCI), partly owned and controlled by Anglo-American, took the largest single concession for exploration. Actual development followed existing work in the coalfields of Moatize, where the Belgian enterprise, Compania Carbonifera de Mozambique, linked to the Union Mini~re, had been producing a modest tonnage since the 1920s. The Moatize field is large, offering an output of more than 2 million tons a year, and Sumitomo Mining of Japan showed interest in taking half a million tons a year, provided it could be delivered at a deep-water port. Later, the South African Iron and Steel Corporation took a preliminary option on another million tons of output a year, to be delivered at the new east-coast industrial complex of Richard's Bay. An even larger coalfield lies under what will become the northern edge of Lake Cabora Bassa, and to handle the water-borne export of 2-3 million tons a year, a small harbour was constructed near the dam, ready for pipeline or railway connection with the river navigation system below the Zambesi rapids. Iron ore deposits had been found in the province half a century earlier. Traced in detail, they showed that reserves of high-grade ore of around 10 million tons existed, of which between 20,000 and 100,000 tons a year could be mined. Of equal interest, at Machedua deposits of titano-magnetite could be developed, using local coal and fluorite, to produce a quarter million tons a year of steel, and 100,000 tons of titanium and vanadium alloys. A ready market existed in Europe, and the project for a steel factory near the rail-head and river at Moatize was allocated to the Compania d'Uraneo, a subsidiary of the commercial empire of Martin Diaz da Cunha, richest and most politically powerful of Mozambique entrepreneurs. Fluorite is found in marketable quantities in three areas, all having a possible production of 60,000 tons a year: Canxixe, assigned to a Portuguese consortium, Interminas, drawing its capital from the Portuguese cotton combine and CUF; Chioco, Compania d'Uraneo, Compania Mineira de Lobito, and Bethlem Steel; and Changara, Anglo-American and JCI 227

(Comoxmin). Elsewhere lay substantial reserves of manganese near the Rhodesian border, and an enormous field of low-grade iron ore, containing perhaps 120 million tons, at Honde between Beira and Vila Fontes. JCI also held concessions to search for platinum and copper, while an Anglo-American subsidiary, Diamoc, prospected for diamonds. In the long term, the mining future depended on whether extraction and transport costs could be contained within limits dictated by world market prices. But more immediately, prospecting was cut short by Frelimo's extension of the war in 1971. In remote areas access was restricted to helicopter. First the Compania d'Uraneo and then JCI were forced to halt their work. One of JCI's largest drills was captured by Frelimo. In spite of military escorts, and reluctant to risk lives, the companies retired, claiming an extension of the contract under a force majeure clause, and retaining only an office in Lourenqo Marques. Dependent on the interest shown by South African enterprise, Portugal had to agree. By 1973 JCI had spent upwards of 680,000 Rand; a not inconsiderable opening bid, but small in comparison with their stake in Angola. To the question of whether they will return to Mozambique there is no immediate answer. Exploration demands peaceful conditions, political support, and the prospect of long-term profit. Neither of the first two yet obtains, and with present transport facilities the mines of Tete are marginal compared, say, with those of Zambia and Angola. The demands on capital in mining are so high - 800,000 Rand in the first three years of one mine, but then 1.5 million Rand for detailed drilling, and 25 million Rand to reach full production - that the return must be guaranteed. As it is, after two and a half years, work has been shelved and the dreams look tarnished. Unless the steel works is built, vanadium and titanium extraction will lapse. The iron deposits are small and remote by comparison with South Africa's Sishen field; the fluorite is enough to attract capital without being irresistible. 'Minerals do not make a mine,' and difficulties at Moatize coalfield, where it appears impossible to increase production beyond 2 million tons without heavy extra investment, have refuted the optimists. The Japanese require best quality coal, but to obtain it leaves a quantity of slag so vast as to make profit impossible. CABORA BASSA 228

Development or exploitation? So much depended on cheap transport that the Portuguese expanded their original plans for river navigation to include both mining and industry. They had originally imagined that a water route, starting on the Rhodesian-Zambian border, continuing down the new lake, bypassing by rail, canal, or pipeline, the dam and rapids, and then using the river from Mpanda-Uncua to a new port in the delta, would attract both Rhodesian and Zambian exports. Flat-bottomed barges of up to 50,000 tons could use the Zambesi if its floods were regulated and sandbanks and shallows dredged. The Rhodesian Ministry of Transport showed some interest: agricultural produce of the North-East and chrome from Umvukwes could be sent down the Mazoe River into the new lake; so, if political circumstances changed, could Zambian copper. Sumitomo's interest in coal pushed these plans to the point of serious study, and from 1971-3 a firm of German consultants, Intertrafik, analyzed the river route in relation to alternatives, such as improvements in roads and the Moatize railway. Their report was being considered by the Overseas Ministry when the Portuguese revolution occurred. The grandiose scheme may yet come to pass. The delta port was abandoned in favour of using river barges capable of making the shallow sea-passage from the Zambesi mouth to Nacala, the finest deep-water harbour on the east coast of Africa. (An alternative scheme, much welcomed in Malawi, was for Portugal to extend the railway from Blantyre to connect with Moatize, thus diverting the coal traffic from Beira to Nacala.) Already the terminal of the railway to Blantyre, and likely recipient of a Lisnave (CUF) scheme to spend £60 million on a repair yard for the largest ships using the Mozambique Channel and the Cape route, Nacala is capable of almost unlimited expansion, and was, before the hand-over to Frelimo, a key position in the game of naval supremacy in the Indian Ocean (Chapter 12). In the near future the Nacala Railway may be extended to connect with Zambia, since Dar es Salaam's congested harbour, at the end of the Chinese-built Tanzam Railway, cannot cope with the demands of the copper trade; and if the river route can be phased in with Nacala's port development, and with that of Richard's Bay in South Africa, the Zambesi may become, as 229

Livingstone once naively dreamed, 'God's highway into the interior,' like the Rhine, the Volga, or the St Lawrence. In the period of rising world commodity prices and in an attempt to offset some of the capital costs, the Overseas Ministry evolved a new scheme, still far from detailed in 1974, to process Malawi bauxite on the river at Vila Fontes, using Zambesi water and Cabora Bassa power. Converting bauxite into alumina, but especially converting alumina into aluminium by conventional means, requires a very great supply of electricity, and when the Portuguese took a share in the Aluminium Corporation of Malawi they had envisaged selling power to a plant on the Shire River. The bauxite, discovered in the crater of Mount Mlanje, would have been piped down. But the cost of the plant in 1970 would have been £140 million, and Malawi, having seen the effect of changing world mineral prices on Zambia, was unwilling to be so tied to a single export. Later, studies by Lonhro showed that the deposits contained only 30 million tons, not the 50 million originally estimated. Alternatively, the Portuguese could buy bauxite, increasing Malawi's small reserves of foreign exchange, pipe it down to Vila Fontes, process it, using 250 MW from Cabora Bassa at cost price, and ship out aluminium, either by the railway to Beira or by river to Nacala. The Overseas Ministry was not wrong to calculate that rises in the price of minerals (aluminium went from £250 to £410 a ton on the London metal market in six months after December 1973) and the world oil crisis would enhance the attraction of the coal and iron of Tete, and of hydro-electric power. But very little had materialized before the revolution; and the problems of capital and labour supply remained. Da Cunha's steel factory, for example, was planned as early as 1965. Nine years later he had beaten off a challenge by the Champalimaud group, financed a survey by General Mining, and planned an output of a quarter million tons of steel, 130,000 tons of titanium slag, and 9000 tons of vanadium; yet only a third of the £29 million capital had been subscribed. Construction should have begun in 1972 but was postponed until the war was over. It remains a viable proposition: at current world steel prices it must be profitable. Whether the World Bank would finance it, whether Mozambique could 230 CABORA BASSA

Development or exploitation? 231 find the skilled work force, and whether a Frelimo Government will wish to industrialize for the Western market, are open questions. Some positive industrial benefits to Mozambique did accrue from Cabora Bassa under Portuguese direction. To meet an old deficiency, the North Mozambique Electric Company sketched out a grid, linking Nacala, Montepuez, and Nampula, to be tied in with Nova Freixo and Vila Cabral, to a new 220 KV line from Cabora Bassa to Tete, Chibata (near Vila Pery), and Beira. This had the triple benefit of creating a grid for the whole Zambesi basin, of supplying the prosperous settler farmlands, and of bringing Cabora Bassa power within a short distance of the Rhodesian grid at Umtali - in case it should be needed. The line was to be opened in 1977; and if the assumptions. are correct, it may yet be decisive in the development of the north-bank power station.t Even if no tap-line to Rhodesia is built, the line will supply such industry as is capable of expanding (cement, textiles, and perhaps steel) at less than half the cost of existing electricity. In an even more roundabout way, via military expenditure, northern Mozambique acquired a network of tarred roads, a communications system, and a great deal of miscellaneous capital investment in public buildings and town development. If the colonial era had ended in 1964, when the Frelimo revolt began, Portugal would have had hardly any claim to have developed her colony to the standard, say, of Northern Rhodesia or Algeria. But ten years of war, and the distant results of war, of which Cabora Bassa and the economic development of Tete were part, have left Mozambique in an economic position similar to that of Ghana at independence in 1956. An infrastructure exists, together with Cabora Bassa; the rest has been explored and planned. If the future lies with a mixed economy, then development has not been wasted. But that is a political question. State planning had been subordinated to events long before the transfer of power to Frelimo. Once it became clear that the contingent benefits of the dam, and the economic mobilization of the Zambesi basin, could not be gained without winning the war, Portugal succumbed to t For further amplification see Notes at end of chapter. hysteria in executing the process begun in 1968, when Caetano's Government had decided to make the colonies economically autonomous. What had been a carefully coordinated plan, undertaken by a paternalistic but far-sighted authority, disintegrated steadily after the middle of 1973 as military demands took priority, and as the resettlement programme absorbed more and more energy and money. The GPZ, rather forlornly, concentrated on finishing Cabora Bassa, hoping desperately for a political miracle which did not come. For a brief period, it had known a sort of euphoria, as the state of emergency drew unprecendented funds from Lisbon for its long-delayed projects; but, paradoxically, the emergency then forced the GPZ into a role supporting the armed forces, and robbed it of its brief freedom of action, before, in the last stage, the armed forces themselves succumbed to impossible demands. CABORA BASSA 232

Notes and References Chapter 9 Development or exploitation? I Trevor Newton, Cost Benefit .lnalysis in PublicAdministration, (London 1972) p. 19. 2 See John A. King Jnr, Economic Development Projects and their Appraisal (Baltimore 1967). t Note page 211. According to the contract, Decreto Lei No. 49225-(3 September 69), the first stage comprised: escudos Civil engineering work 3,187,762,198 Electro-mechanical - generation 2,002,333,865 substation 739,343,250 Transmission line 1,501,583,970 Local and other costs 2,450,000,000 Total 9,881,023,283 tt Note page 211. At Kariba the final cost was roughly £10 million less than the last revised estimate of 1955 - although the project changed considerably during the process - and it is fair to add that the 1955 total of £107.5 million was 30 per cent up on the 1950 estimates. (King, pp. 255-67). t Note page 212. Escom's spinning reserve will cover any temporary loss of the largest set at Cabora Bassa at all times. There remains the cold reserve and the overhaul reserve to cover the rest, and although in the years 1977-84 Cabora Bassa's input will exceed the 8 per cent limit of South African output originally envisaged, this too has been covered in the programme for extra generation. + Note page 219. Another casualty of the haste of war. In 1971, when the lake basin was cleared, the GPZ was able to conduct a series of public meetings, or banjes, to explain what was happening, but this experiment in consent was later abandoned in favour of working through Portuguese-appointed headmen. t+ Note page 219. An accurate survey of opinion was impossible, because of linguistic problems, to say nothing of time. But the author's interviews, conducted with assemblies of elders and headmen, suggest a keen appreciation of improved standards, and a desire to move into a market-orientated economy. 3 Eduardo da Serra, 'Survev of Charre, Sede and A ncuaze' July 1971. 4 Ibid. t Note page 222. Thus, re. Eduardo da Serra, Survey of Charre, Sede andAncuaze, July 1971: Charre Charre Sede Sede Ancuaze Ancuaze MFMFMF 0-1 34 27 48 45 9 14 1-4 73 80 103 83 98 108 5-9 135 114 132 138 138 130 10-14 103 86 59 70 86 74 15-19 52 62 40 65 34 43 20-24 42 44 41 48 31 34 Population density per square kilometre 38 10.3 77.7

234 CABORA BASSA 5 Technological Change in Agriculture and Employvnent in Developing Countries (OECD 1072). t Note page 225. The normal Portuguese mining concession was for exploration for two vear , renewable for a further two v'ears; followed by twenty-five or thirty years to mine with a further extension. JCI, for example, do not pay for their concession, but meet all exploration expenses. If they eventually decide to mine, the province gets a royalty on gross revenue, which may be as high a, 7 per cent; tax i' also payable. A Portuguese associate is required to hold at least 25 per cent of the equity - hence the link with the Compania Mineira de Lobito, and Compania Mineira de Mozambique (Comoxmin). Only minerals of 'national importance', such as uranium, are reserved wholly for the Portuguese sphere. t Note page 231. The contract was signed on 26 March 1974, just before the old regime fell, by LTA and the French civil and electro-mechanical group already at work in the north-bank intakes. The cost, including AC sub-stations, was 1000 million francs; preliminary studies were paid for by IDC, Anglo-American, and the French Government.

Chapter 10 Towards Economic Autonomy After Salazar's stroke in 1968 and the accession of Caetano, some among the long-suppressed opposition in Portugal, and many liberals outside the country, hoped for a relaxation of the dictatorship and for a gradual move towards decolonization. After a brief period in which an optimistic interpretation was placed on Caetano's introductory and often conciliatory speeches, they were disappointed; and, like rejected suitors, added their condemnation to the strident voices of Third World countries at the United Nations. From 1969 onwards nothing Caetano's Government did could be right: few attempts were made (except in some French newspapers) to analyze its economic, as distinct from its political, attitude towards the overseas territories, and even these were largely stultified by the Government's reaction to criticism; for secrecy and the use of stringent censorship, repression of independent thinking, and the manipulations of the DGS, combined to disguise the fact that Caetano did have a plan for the colonies, grafted onto Salazar's but nevertheless quite distinct in its time-scale. Although it did not succeed, it contributed, like the building of Cabora Bassa, to the dramatic events of 1974. In 1968 Caetano deliberately bought time to review Salazar's colonial theory and policy - of which the decision to go ahead with Cabora Bassa was then the most urgent single outstanding issue. While the contractors waited uneasily, Caetano brooded on the possible choices: a continuation of the status quo, implicit in all the legislation flowing from the Colonial Statute of 1930, leading eventually to full integration with Portugal and regarded as the only course by right-wingers such as Silva Cunha, Overseas Minister, and Antonio Rapazote, Minister of the Interior; independence, which was not seriously considered except by two or three of the junior ministers, keen on associate membership of the EEC; or some middle way. In the last category, Caetano rejected the federalist solution (later to be advocated by General Spinola) which he had himself put forward to Salazar, when he was Rector of Lisbon University in 1962, on the ground that it would not permit a flexible enough development of the different territories. By the time of his overseas tour in 1969, Caetano had made up his mind that Portugal had no choice but to speed up the evolutionary pattern, at whatever cost, in order to achieve harmony inside the colonies, so that they should be capable of conducting their own defence, on the basis of developed political institutions and a sound economy. This 'autonomy', however, was not to be confused with independence; it would only grow from Portuguese roots, within a Portuguese framework, into a form pre-defined by Portugal. However broadly, and often ambiguously, the terms were defined, this doctrine did not differ essentially from Salazar's: on the other hand, it seemed to provide a means for achieving Salazar's aims in ten rather than two hundred years. There are certain similarities with French aims in Algeria before de Gaulle came to power, and also with British policy towards the Central African Federation, but whereas these were politically orientated, Caetano's policy was couched in terms of economic change. Strangely Marxist for so ardent an anti-Communist, he perceived that political stability in Africa could only grow out of a sound balance of trade, prosperous agriculture and industry, and a stable currency. The plan altered as circumstances changed, and even in its original form showed marked differences in treatment of Mozambique and the much richer territory of Angola, to say nothing of commercially worthless Guinea-Bissau, an albatross of historic proportions hung round the neck of successive Overseas Ministers. But its implications revealed a profound revulsion against the mercantilist colonial system of the previous fifty years. The point can be made by a brief look at the CABORA BASSA 236

Towards economic autonomyV well-charted period of 'uneconomic imperialism',' and that of Salazar.2 Until 1910, while Portugal hung on to Mozambique in only nominal control and in the constant fear that the colony might be mortgaged against her foreign debt, or partitioned between its chartered companies, colonial officials regarded the results of five centuries' rule with the utmost gloom: 'The annals of our colonial administration are a web of misery and disgrace,' wrote Oliveira Martins in 1880.' Emigration flowed to Brazil, not to Africa, and a senior official of the Banco Nacional Ultamarino averred in 1893 that 'colonies serve no useful purpose and will always bring us into discredit'; the Governor-General of Mozambique added, in 1907, what was still to be a complaint sixty years later: 'The Montepio Geral is replete with Portuguese funds, yet of all the contos deposited therein, not one is drawn out to be invested in Mozambique.' If there was a profit at that time, it accrued to South Africa via the use of ports and the Delagoa Bay Railway, or to Portuguese companies which imported sugar and cotton on exploitative terms for processing in the metropolis. During the period of settled administration which followed the reforms of Antonio Enes, roads and railways were extended and the economy tied even more closely to servicing neighbouring countries' requirements. Mozambique's principal source of foreign exchange, paid in gold, came from the supply of contract labour to South African mines. By any calculation, the administration of the country was a drain on Portuguese Government resources. For a few years, during the world commodity shortage in and after the First World War, private Portuguese and foreign investors drew a large return, and paid absurdly low prices for raw materials; but by the time of the slump of the 1930s even this advantage had disappeared. Salazar's imposition of tight budgetary control forced Mozambique into economic stagnation. The colony's economic institutions faltered; the chartered companies declined; the prazos - vast semi-feudal estates in Zambesia which had passed into local, often mulatto, ownership - were taken over by Portuguese companies; and such foreign investment as had previously come in was excluded in favour of injections of Portuguese capital, on grossly partial terms, with dividends 237 largely repatriated to Lisbon. The embryonic banking system, like the currency, was subordinated to the political aims of an undivided escudo zone - though the Mozambique and Angolan escudos were not, and could not be, permitted free convertibility. Up to the 1950s, budgets were balanced at the cost of natural development. There followed a time of expansion, of increased Portuguese immigration, of lowered tariffs, further investment in transport, public buildings, and education, leading into the era of successive development plans. The white population rose rapidly, from 48,000 in 1950 to 205,000 by 1970.t In spite of the cautious reopening to foreign capital, notably for Cabora Bassa, control still rested with the metropolitan government. Portuguese investment grew steadily, so that by the 1960s Mozambique was more, not less, the victim of exploitation than at the end of the Second World War. Until the Frelimo revolt broke out, it appeared that the underlying intention of the Colonial Statute was being fulfilled: outside Beira and Lourenqo Marques white Mozambique had come to resemble rural Portugal. The creation of a 'Portugal in Africa' seemed to be assured by Decreto Lei 44016 in 1961, which enjoined full integration into the escudo zone by the early 1970s. At the time of Caetano's review of colonial policy, after five years of war in the north, Mozambique had no financial institutions of its own, no currency convertibility, no stock exchange, and its local entrepreneurs depended on the ubiquitous Banco Nacional Ultramarino for permission to raise loans for more than ninety days. Under the provisions of the 1965 legislation, foreign concerns had benefited more than local. South African influence dominated mining, prospecting,tt tourism, and private banking. Gold from the earnings of contract labour was shipped directly to Portugal, to be offset against the escudo debt (as with Angolan diamonds, Portugal benefited from the difference between the official price of gold and the free market price - an anomaly not challenged until after the revolution); and Mozambique's foreign earnings went to pay for imported consumer goods, tFor further amplification see Notes at end of chapter. tt For further amplification see Noteshat end of chapter. 238 CABORA BASSA

Toards economic autonomylv rather than into productive investment. Progress towards economic autonomy was, therefore, necessarily slow. But the commitment to industrialize the Zambesi valley, in theory at least, reversed the ancient mercantilist orthodoxy under which Portugal had processed Mozambique's raw materials and exported to the colony cheap industrial goods. Following the Salazar thesis that the role of the state was to support Portuguese private enterprise, rather than to dictate to it, Caetano's Government restricted its efforts in enforcing economic stability to fiscal measures, and with an almost Gladstonian fervour set about reforming the abuses of the escudo zone which had grown up in the late 1960s. The scheme for a zone, not unlike the post-war sterling area, had been conceived by officials in the Bank of Portugal a decade earlier: all overseas provinces were to conform to a system of monthly inter- territorial payments, supported by a contributory reserve fund. Debts were to be settled in Portuguese escudos; and if the provinces lacked these, they could in certain circumstances buy them with foreign exchange. Unfortunately the provinces, especially Mozambique, needed to buy from Portugal more than they could sell. It was no use complaining that they were ill-paid for raw materials; a system of import licences was imposed by Lisbon. In the late 1960s, however, during the period of expansion, and to counter the depressive effects of the war, strict budgeting was relaxed; Mozambique was allowed to run up a debt, which helped stimulate a boom in consumer expenditure, of which the most notorious sign was the lavishing of 2695 million escudos (8 per cent of the total revenue of the third five-year plan) on housing and speculative building in Lourengo Marques in the three years after 1968. A backlog of trading settlements built up, so that by 1971 it took fifteen months for a Lisbon exporter to receive payment. Consequently, he added on the cost of financing these loans, driving up the price of imports in Mozambique and worsening the balance of payments. The effect was compounded by evasion of exchange control on a colossal scale by Portuguese companies which repatriated their dividends in metropolitan escudos by transfers through subsidiaries. The reserve fund was by now hopelessly overdrawn, 239 and Lisbon banks grew reluctant to sink further funds into propping up such instability. Four million contos (1 conto equals 1,000 escudos), 30 per cent of the Mozambique money supply, lay frozen in the vaults of the BNU, and however much the monetarists applauded the anti-inflationary effects, there was no alternative but to fund the debt and start again. Thus the legislation of 1971-2, which accomplished the transition from Province to State - the Organic Law, and the Ways and Means Law - grew out of the immediate need to resolve an intolerable economic muddle; it contained, however, the lineaments of self-sufficiency in economic affairs. Once and for all, the imbalance was to be corrected, by import licences, not imposed from Lisbon, but from Lourenqo Marques. Paternalism dictated that solutions lay in the development of Mozambique industry and agriculture, and local financial institutions. These reforms, of course, owed a great deal to the parlous situation of Portugal's own economy: the growing rate of inflation; the drain of labour to Common Market countries, as Portuguese fled their homeland to earn higher wages, or to escape conscription for the African wars; and the cost of the war itself (see p. 157). But they could be represented as a coherent plan, not only for resolving the imbalance of payments and stimulating import substitution, but for transferring the cost of the war, t for diverting emigrants from Europe to Africa, tt and for easing Portugal's own application for associate membership of the EEC.* The reforms tightened import control, as predicted, by imposing the need for prior cover. All future transactions had to be made from reserves; the backlog was left as a debt to be liquidated by metropolitan loans. The full effect was not felt until 1972 - a year in which, despite inflation, Mozambique's t For further amplification see Notes at end of chapter. tt For further amplification see Notes at end of chapter. * The Portuguese never attempted to apply to the EEC on behalf of the colonies (on the pattern of the Yaounde agreement) well aware of the opposition this would arouse and relying, in any case, on the doctrine that, as an 'integral part' of the metropolis, they were automatically excluded. In the event, Portugal applied to the EEC herself, with the remaining EFTA countries, and received similar agreements to them in July 1972. The colonies had already been included, somewhat surreptitiously, and with French diplomatic assistance, in the General Preference system for developing countries, in July 1971 (EEC Reg. No. 1308/71 Appendix B). CABORA BASSA 240

Towards economic autonomy imports fell in value by 8 per cent, in real terms hy 20 per cent. Portuguese exports also suffered, falling by 22 per cent from 1970 figures. Sporadic shortages of food and consumer goods were followed by the granting of licences: at times matches, meat, or soap, would be unobtainable (though not at Cabora Bassa). Prices climbed, reflecting the interest burden on trade settlements over a five-year period; and 30 per cent inflation of single items occurred as soon as controls were lifted. Some idea of the financial predicament of the provincial government can be seen from taxation figures (taxation, 1974, as a percentage of 1972: direct 272 per cent; indirect 377 per cent; industry 187 per cent; services 170 per cent); and from the fact that revenue from consumption tax (VAT) was screwed up from 102 million escudos to 1,159 million escudos between 1972 and 1974. The bewildered public was encouraged to believe that revenue from Cabora Bassa would in a short time ameliorate these demands: a novel form of Sebastiacao, or the myth of miraculous survival. Exposing the provinces to the full rigours of inflation could be defended on grounds of financial probity. But there remained the problem of the accumulated trading debt, or atrassos (amounting to 6.4 million contos in 1971). Metropolitan banks were coerced into funding a 1.5 million contos loan at a mere 4 per cent by the Lisbon Government; this was used up in two years. An approach to the IMF being out of the question, despite the technical freedom of the new Mozambique Legislative Assembly to contract external loans, Portugal underwrote a final loan (at crippling cost to national investment) to eliminate the trading debt - 500 million escudos in October 1973 and 3,500 million escudos in January 1974). Thus the entire imbalance was paid off by the metropolitan government, in return for state loans guaranteed by Mozambique; the 6.4 million contos had become the National Debt of the new state. Intellectual rigour in Lisbon forced Mozambique at the same time into a desperate flurry of import substitution. There could be neither an appeal against the system, nor yet, for political reasons, wholehearted suppression of the spending power of a new generation of African consumers, growing, like Mozambique's GNP, at a rate of 10 per cent a year. To 241 economists in the Overseas Ministry, import substitution seemed valuable in itself; but in two important respects it was ill-planned, and the fault must lie not with the beleaguered Mozambicans, but with the Portuguese who assumed that fiscal measures would be enough to determine the proper direction. Habituated over forty years to the doctrine of non-interference, government departments in Lisbon refused to offer advantages or to bring pressure to bear on Portuguese companies to reinvest their profits in Mozambique; and, having funded the atrassos debt, ignored the need to stimulate the local internal market. It is true that gradually the colonial protective system was run down; but the benefits of free trade in sugar and cotton came too late.t Even where suitable local industry could be funded, the Portuguese Government safeguarded existing interests: thus the Texlom textile factories, planned originally by Mozambique interests for Lourenqo Marques and Nacala, were taken over by the Melo family (CUF) and the Societa Financeira Portuguesa, with loans from the French Socidtd Generale. In the end, import substitution emerged at a low level of competence in consumer goods rather than in the necessary fields of agricultural machinery, or processing of raw materials. Factories for making razor blades, soap, or toothpaste, competed extravagantly with South African or Rhodesian imports; at the nadir, a match factory imported boxes, the paper for labels, and the matches, ready coated with phosphorous, supplying only local labour for printing and counting. It would have been better had the effort gone into finishing raw materials for export, or concentrated on the labour-intensive field; either would have led to future growth, whereas these enterprises foundered as soon as the supply of external capital was cut off. On the other hand, these years witnessed a remarkable growth in Mozambique's GNP, which reached 10 per cent per annum after 1971. (Exports, even to Portugal, benefited also from the reform period, increasing by 22 per cent from 1970-2.) Since exchange control prevented private investment abroad, or even in local industry, the growth in white and t For further amplification see Notes at end of chapter. CABORA BASSA 242

Towards economic autonomy African wages produced a savings boom which enabled the newly enfranchised financial institutions to survive. Before 1971 banking had been dominated by the Banco Nacional Ultramarino: Standard and Barclays, through their Portuguese associates, serviced the through trade; and the Portuguese private banks served local needs and repatriated the profits of local trading. Only the BNU was permitted to lend money long-term, in order to reserve to Lisbon decisions about the pattern of investment. (The Mozambique Chamber of Commerce raised 60 million escudos in 1963 to found a national bank, but were refused permission.) To add to the frustration of local enterprise, the BNU, by law, exacted deposits of 10 per cent of the capital of private banks, railways, etc., on which it paid no interest, and which it then loaned to the provincial government at commercial rates. The BNU, in fact, played a dual role, as reserve bank and as leading commercial bank. But in the 1960s an element of competition appeared, from the Banco de Fomento (Development Bank); then in 1970 the Instituto do Credito do Mozambique was founded, specifically to compete with the BNU and to provide long-term credit for industry and agriculture. In the hectic period preceding the revolution, the ICM superseded the BNU so that, though technically retaining the function of reserve and issuing bank, it was largely reduced to commercial operations. The Banco do Fomento and ICM, and the local Montepios, or savings banks, made the loans which enabled the third development plan to be executed and the fourth sketched out; and which completed the transition to economic autonomy. t In the preamble to the third plan in 1968, it was clearly laid down that, 'it is the duty of the Institutions in the Overseas Provinces to mobilize local means of finance for the Plan.' Inconceivable as this would have been in previous decades the third plan was so financed, and at rates rarely exceeding 6 per cent, due almost entirely to the savings boom.tt The experience turned out to be so invigorating that the Finance Department in Lourenqo Marques and the Governor General envisaged raising, in the fourth plan, four times that amount t For further amplification see Notes at end of chapter. tt For further amplification see Notes at end of chapter. 243 in the years 1974-9. That exercise may now be irrelevant: certainly its figures were grossly inflated - essential expenditure amounted to only 12 million contos out of a projected 55 million, and the bulk of that was scheduled for the years 1977-9. But as with the other developments, a successful outcome turned out to be less important than the political significance of what had been done. The 'Alice in Wonderland' aspect of claiming Mozambique as an integral part of Portugal was quietly forgotten; colonial trade preferences were gradually abolished; and, at very substantial cost, the financial infrastructure was built up to serve the domestic market.t After November 1972 budgetary control rested with the Mozambique Legislative Assembly. The Mozambique Government acquired a National Debt, but also the power to raise loans at home and abroad through its own institutions, the ICM and the Banco de Fomento; and although in practice this freedom was limited by the need for prior sanction from Lisbon and, in particular, from the Societa Financeira Portuguesa, its guarantee had the force of law. The ICM itself, owned wholly by Mozambique and with a board of national directors, had taken over the BNU's rights to deposits from municipal funds and parastatal institutions, and lacked only the power of issuing currency to become the State Reserve Bank. The Mozambique Treasury had become independent, so far as the operations of exchange control in the escudo zone permitted (convertability would have led to an immediate flight of capital, as it would for the kwacha or shilling of neighbouring Zambia, Malawi, and Tanzania). The Mozambique Government could control imports, by volume and country of origin, exports, the direction of investment, and industrial and mining developments, with the sole exception of commodities of 'national importance' - petrol and uranium. The absolute Portuguese control of 1970 had been transferred three years later as an act of deliberate policy 'deeply planned' in the critical year 1968, according to Jorge Jardim, a close associate of Salazar and Caetano. Before the revolution Mozambique was on the verge of economic autonomy, which Jardim, among others, hoped would lead to political independence. t For further amplification see Notes at end of chapter. 244 CABORA BASSA

Towards economic autonomv Caetano had intended in this devious and secretive way to create the conditions for a form of wider political integration: a 'free association' of mature states with the metropolis. No details were given at the time. His speeches, even in February 1974, contained nothing but conventional phrases; perhaps, after the revolution, it no longer matters. Yet it is worth recalling that, from his first speeches in Mozambique in 1969, he sought to avoid a situation where the white population, either legally or by UDI, would lead Mozambique into the ambit of the white South. The secret five-year plan adopted in 1968 envisaged a multi-racial, politically mature community, to be evolved by means of enormous expenditure on rural and regional development, transport, education, and health; and the fact that this goal was contingent upon a dream of Lusitania which did not survive should not detract from the courage and vision which inspired it. But Portugal could not afford the price. The great corporations twisted the plan into a form of neo-colonialism; and Caetano eventually succumbed to the force of metropolitan orthodoxy. If the cost of the war had been contained, if inflation in Portugal had not been compounded by the oil crisis and world commodity prices, if the strain on manpower had not been too great, there might have been a chance. But by 1973 Portugal was carrying 50 per cent of the cost of the war; unlike that of Mozambique, the metropolitan budget had shifted decisively away from development and civil expenditure in the previous decade.t Funding the atrassos debt for Angola also imposed intolerable strains on Portugal's capacity to raise domestic loans in the future, and on the banking system to fund industrial development at home, and Caetano had to confess to 'continuing economic difficulties' in his speech announcing the 'final settlement' of atrassos in February 1974. All this must be seen as one of the largest proportionate injections of capital aid ever undertaken by a colonial power.tt Cabora Bassa was the most obvious legacy, but, because of its self-financing provision, perhaps the least costly element. It is not possible, using the aggregate figures for t For further amplification see Notes at end of chapter. t t For further amplification see Notes at end of chapter. 245 overseas expenditure published in Lisbon,t to construct a balance sheet or profit- and-loss account for these years; but calculations for 1968 show an excess of Government funds, in favour of Mozambique, on the balance of payments, of 90 million escudos, and a probable total cost to Portugal for the year of 2150 million escudos. But beyond that, who profited? Mozambique clearly gained capital investment; but the private movement of funds from Mozambique to Portugal in the same year reached approximately 2000 million escudos. Thus the Portuguese taxpayer financed the Portuguese capitalist via a war, costing thousands of Portuguese lives, and a development programme which only served to prepare Mozambique for independence.* The overall cost to Portugal of maintaining her presence in Mozambique in the war decade after 1964 is beyond calculation. The private sector in Portugal flourished in the last days of empire. Annual returns of 2000 million escudos encouraged a new interest in Mozambique. But for large companies and private banks the continuing war presented the risk that at some point Caetano's Government, despite its pledges, would recant and grant independence. Expropriation and nationalization being, evidently, the rule in black independent states, it became necessary for them to insure a measure of white control in any future political structure; or, failing that, so to invest their money that it should be an inalienable charge on any future Government. Naturally they would have preferred the old system to survive: but as the economic measures of 1971-2 unfolded, as Portuguese exports suffered and mercantilism was finally repudiated, and as political demands swelled in 1973, they sought a measure of accommodation. A handful of families controlled the majority of Portuguese enterprises outside the government sector, and they formed an oligarchy with great political power. The wealth of the Melos, the Espirito Santos, the Champalimauds, the Quinas, is not only colossal in relation to the national economy,5 but its power has conditioned the economic policy of successive f For further amplification see Notes at end of chapter. * Calculations made by the usually well-informed Le Monde (15 November, 1968) show a similar pattern for Angola, although, since they do not differentiate between the state and the private sector when considering repatriation of funds, the conclusion that 'Portugal' made a profit, at a high military cost, is suspect. CABORA BASSA 246

Towards economic autonomy governments in the century since the industrial and commercial interest succeeded to the position once filled by the great landowners. The same monopolistic pattern was repeated overseas; the Melo family, for example, owned or controlled 75 per cent of all newpapers in the Portuguese empire, and their principal enterprise, the Compania Unido Fabril, virtually ran the economy of Guinea-Bissau. In Mozambique, the largest Portuguese interest is controlled by Antonio Champalimaud: Mozambique Cement; Quimica Geral (fertilizers); CIFEL (iron and steel); two insurance companies, and the Bank Pinto e Sotto Mayor, ranking second in the Portuguese commercial zone. His Mozambique assets amount to over E25 million, yet form only part ot a whole which includes the National Steel Company in Portugal, huge interests in Brazil, Angola, and Swaziland, the Commercial Bank of Malawi, and the largest brewery in Mexico. Chainpalimaud is probably one of the twenty richest men in the world; it is said that his credit- rating puts him in the top ten. When he and the Melos fell out, Lisbon witnessed a battle of Olympians, whose five-year feud in family lawsuits masked a territorial dispute about demarcation between their respective empires. As early as 1971, while still in self-imposed exile in Paris (because of a law-suit against the Government, which had legislated retrospectively against his attempt to buy a majority share in the Banco Portugus do Atlantico), Chainpalimaud began to invest more heavily in Mozambique; he was followed by others, and supported by the advice of the Societa Financeira Portuguesa and its chairman, Professor Texeira Pinto. Being so close to Caetano, the oligarchs understood the meaning of the 1971 reforms and the need to rearrange the pattern of their existing investments, so that they were linked with the Mozambique Government, or dressed up as local entities; and to ensure that future expansion was financed locally, by the State of Mozambique, out of loans provided from their local profits through the agency of their own associated private banks. Thus profits, instead of being directly repatriated, would be turned into loans against state guarantees and drawn, in lesser amounts, at fixed interest but in hard currency, and secured against expropriation by any future nationalist Government. 247

The only Mozambican entrepreneur with political weight in Lisbon was Martin Diaz da Cunha, who owned the Mozambique Company, the Compania d'Uraneo, Eatreposto (commercial vehicles), Societa de Turismo, the National Cotton Company, Mozambique Coal, and a share with Lonhro in the Rhodesian pipeline, an investment, even in 1971, of $70 million. Da Cunha showed himself keen to invest in Mozambique, especially in the Tete region, but his venture in steel and mining rested on large state loans (two-thirds of the total for the Tete steel mill) and corporate policy appears to have followed the guidelines tacitly adopted by the Caetano regime. Too much weight should not be placed on the assumption of an explicit contract between Caetano's Government and Portuguese capital. But it is significant that, in the first week after the revolution, President Spinola called two meetings of bankers and industrialists, and reassured them that the army would safeguard economic stability at home and overseas. Close associates of the former regime and leading private bankers confirm the existence of an understanding that private enterprise would support Caetano's reforms and the third and fourth development plans in return for political concessions designed to perpetuate their influence. By lending to the State of Mozambique, and submitting almost without protest to the extortionate 4 per cent loans raised from their private banks by the Government to pay off the atrassos debt, they became primary bondholders of the new National Debt, doubly assured of the future by the provisions of the Organic Law and the continuing control exercised over Mozambique loan facilities by the Societa Financeira. Hence the free flow of capital into the state coffers; hence the downgrading of the BNU, the state institution long hated by the private banks; and the bland assurance of Mozambique ministers in 1973-4 that the bulk of the fourth plan's 'non-essential expenditure' in the profitable fields of building, mining, and import substitution industry, would be funded by private enterprise. Insofar as it was possible, Mozambique had been shaped for a period of neo-colonialism even before its economy had become 'autonomous'. At the time, these plans envisaged only qualified political change and long-term white CABORA BASSA 248

Towards economic autonomy dominion, but they allowed for the possibility of black majority rule, and for the sort of influence by bondholders as obtained after Nkrumah's fall in Ghana. And on that assumption the oligarchy did not object too strongly to the erosion of authority overseas, or - at first - to the programme of the Armed Forces Movement. Although the acceptance by Portuguese industrialists and bankers of the need for change in Mozambique was a prerequisite of success for Caetano's policy, the policy was also governed by the reactions of South Africa and of the multi- national companies involved. As an important vehicle for services to South African transport, shipping, and labour requirements, and a potential source of strategic commodities, Mozambique affected Pretoria's economic thinking almost as much as it did the political calculations of the Nationalist Government (Chapter 12). At an academic level, schemes had flourished in the 1960s for some sort of customs union or federation of southern Africa - a case usually argued in economic terms and sometimes sponsored by South African official resources,6 but which overlapped naturally into the political sphere.7 By the early 1970s, however, such thinking was seriously out of date, since the assumption on which it had been based, that South Africa must trade with her black neighbours and become in Diederich's phrase 'the workshop of Africa', had been revised. While these ideas circulated as part of the intellectual matrix in South Africa, they had not evoked much political or public support; and there is very little evidence that they were taken into account in actual policy- making. The Prime Minister's Economic Advisory Council,* for example, never presented recommendations on such a broad scale. In any case, Mozambique was not then seen as part of black Africa; its links were so clear and well-established that it was assumed already to be part of the South African system. The five-year plan did not spell out the need to supplement * An influential group of businessmen, bankers, and economists, whose advice filtered through to the machinery of government via the Prime Minister's Economic Advisor, P.J. Rieckert. 249 the Republic's deficiencies, but the Cabinet Economic Development Committee showed itself aware of the importance of strategic economic planning. Certain Mozambique resources could fill gaps, especially in the sort of emergency which began to be envisaged as the Russian navy came to dominate the Indian Ocean: bauxite, or semi-processed alumina, from Milange and Malawi; oil, if the Mozambique Channel concessions struck lucky; and coal, for the state oil agency's conversion process, if not. A large part of the case for Cabora Bassa had been built on South Africa's chronic shortage of water; electricity from outside could be justified if it permitted more of the Orange and other rivers to be used for irrigation.8 By 1972 the once visionary suggestion of van Eck, Olivier, and Kuschke, that a southern African grid could link the Cunene barrages in Angola, the Inge Dam on the Congo, Kariba, Cabora Bassa, the dams in Lesotho, and on the Tugela River scheme with South Africa, had acquired practical corroboration. The shortage of water was then so great that the Republic would have to rely by 1990 on sources in Lesotho and the Transkei, or turn to immense and costly desalination plants.9 Escom soberly admitted: 'There is already a marked trend towards expansion of the South African system into neighbouring countries and it seems likely that within the next one or two decades the whole of southern Africa will be electrically interconnected."0 Continuing stability in Mozambique seemed essential, as a safeguard also for the supply of the quarter million contract labourers in the mines, and for continuing use of the railway to the port of Lourenqo Marques as a relief for the overloaded South African network. So, quite apart from political considerations, South Africa sought ways of increasing its economic participation in Mozambique. No loans were provided other than for Cabora Bassa, but the Anglo-American Corporation and JCI were encouraged to develop their stake by easy exchange control facilities during the Diederich squeeze. (A substantial concession, because Anglo then possessed large resources of cash in South Africa, yet few openings for investment, whereas abroad the position was reversed. These facilities were noticeably absent in the case of Anglo's ventures in Australia.) CABORA BASSA 250

Towards economic autonomy During the period 1968-72, the time of dialogue with Malawi and the Ivory Coast and loans to Madagascar, when the verligte 'outward-looking' policy triumphed over 'selfsufficiency', the concept of being the 'workshop of Africa' seemed to be an answer to the loss of the British market following Britain's entry into the EEC and to the problems which followed - devaluation of the Rand, uncertainty about the price of gold, inflation, and lack of jobs for Africans. As Mozambique grew into economic independence great possibilities of freer trade emerged, only to be spoiled by escudo zone reform and renewed import restrictions. The South African Government soon concluded that Mozambique offered less in reciprocal trade than could be gained, say, by cooperation with Zambia.t On the other hand, there was a chance of investment in mining and the export of South African technology which for the first time put Mozambique in the same category as the High Commission territories, Lesotho, Swaziland,and Botswana. At the same time, South Africa had taken the largest share of new investment in Portugal itself. (In 1970 South Africa's share was $17 million; that of Britain $7.6 million; the US $7.5 million; West Germany $5.3 million.) As a recipient of investment and technological aid rather than as a trading partner, Mozambique fitted the new pattern of South African interests which emerged in the 1970s. The export market in Africa, like the policy of dialogue, was, if not an illusion, at least evanescent. As G.J. Steyn, the Minister for Commerce, admitted in 1971, South Africa could not offer enough credit to underdeveloped countries and had therefore to rely on trading in the world market. Drawing this conclusion, and responding to the crisis of 1969-71, when growth rates declined, South Africa imposed restrictions at home and sought expansion overseas, not only by exporting raw materials but also processed materials - such as bauxite, imported from Australia and smelted at Richard's Bay. Japan became South Africa's largest trading partner, importing ore from Sishen, Witbank coal, and manufactured goods; and on Latin America the Republic focussed energies which had once been directed towards black Africa. The Government stimulated t For further amplification see Notes at end of chapter. 251 trade with Brazil and the Argentine and frted the unpopular dictator, President Stroessner of Paraguay. Brand Fourie, Secretary for Foreign Affairs, made clear not only the economic importance to Pretoria of Latin American countries' trade, but the political value of their bloc voting power at the United Nations, on issues of common interest such as the law of the sea and coastal economic zones.'' This transition began smoothly - the proportion of exports to Africa had fallen to 15 per cent by 1973. Meanwhile the volume of investment and technical aid rose rapidly in those countries prepared to accept it - Botswana, Lesotho, Swaziland; and in the Bantustans, where it had reached over 300 million Rand by early 1974. Mozambique and Angola formed vital links in the chain, for operations of South African companies - in mining, civil construction, and industrial development - could be transferred, later, within the Portuguese-speaking zone to Brazil, on the basis of proven experience. In this sense, Cabora Bassa was an object lesson, an illustration of what Hilgard Muller declared in September 1969: 'Although of course South Africa cannot compete with the big states in the field of capital investment, it is in fact better equipped to stimulate development than almost any other country in the world."2 An indication of what was going on can be seen from the volume of IDC lending in these years, and the mobilization of Afrikaner capital in the Republic. But while money could be poured into neighbouring and docile black states, or the debatable 'border' factories of the Bantustans, or under certain conditions (such as the South African Breweries' association with CUF in tourist development) into Mozambique, a strong political resistance developed further north. Mining companies could, perhaps, reinvest profits in Zambia to the satisfaction of the Zambian Government, or reach discree4 private agreements, such as that covering the refining of Mauritanian copper in South-West Africa; but direct offers of capital and technical aid, especially on the sort of commmercial terms laid down by IDC (which still preferred to finance exports rather than to make long-term, low-interest loans), could not easily be made. The recent founding of a development bank, EDESA, by Anton Rupert, leader among Afrikaner entrepre- CABORA BASSA 252

Towards economic autonomy neurs, has led to speculation that South Africa is searching for new ways of 'washing' capital for investment.t This revaluation made it possible to accept changes in the economic, and later the political, status of Mozambique with a readiness which would have been lacking in Verwoerd's day; yet, confined as it was to the political caucus in Pretoria, it highlighted the problem of public ignorance about developments outside the Republic, and National Party fears about loans to African states. Thus it was not accidental that the press, Afrikaans as well as English, was encouraged by Government to welcome the export of capital as a means of ensuring 'stability'; and in the same way, the Portuguese came to be depicted as heroic defenders of the white South. Academic bodies, sometimes financed by the corporations, embarked on research into Portuguese Africa and the real nature of the wars: shifts in phraseology can be revealing, and the description 'terrorist' was gradually replaced, in Frelimo's case, by 'liberation movement'.'3 Once that conceptual leap was taken, it offered a base for further accommodation, even of black majority rule. As a function of its own internal economic development, South Africa was ready, before the Portuguese revolution, to compromise with any future Mozambique Government, whatever its complexion. The major South African companies, without in any way acting as agents for their Government, accepted this situation; and in the case of Anglo-American went some way to bring it about. According to H.F. Oppenheimer, Mozambique in the Caetano era, 'was not an area in which Anglo-American is frightened of investment'. Yet he found the restrictions of the escudo zone and the bureaucratic control from Lisbon frustrating and time- wasting, and he welcomed the development of autonomy. Pursuit of the Corporation's aims, succinctly defined as, 'making money, making the world receptive to Anglo, and making the world a better place', could benefit Mozambique, particularly the offer of its financial services (as in the case of Cabora Bassa) and mining technology. Although, in retrospect, Anglo's interests in Mozambique may appear to have been put together in pursuance of cor- t For further amplification see Notes at end of chapter. 253 porate policy, it would be wrong to imagine a decision comparable to a General Motors' directive, say, to open up a new market in Cambodia. Successive analysts have failed to find a central 'power house' in Anglo, other than the chairman himself. Its decision-making is an opportunist, even haphazard, process of evaluating opportunities and taking up bright ideas, giving a wide measure of devolution to regional headquarters and subsidiary companies. It is the task of the centre, in Johannesburg, to provide the services, finance, and skills, and to generate the climate for new investment. Thus Anglo's expertise in mining technology, and the facilities of Union Acceptances and the Banque de Paris et des Pays Bas (in which Anglo holds 10 per cent) both led into the Cabora Bassa venture; the contracting side, LTA and Shaftsinkers, provided a large share of the civil-engineering capacity, and the chairman of Zamco himself. The Mozambique interest, of course, owed its origin and political strength to Oppenheimer's personal and corporate links with Portugal; the privileged position was reinforced and extended when Caetano came to power. Yet it rested not so much on crude 'leverage' (though many Portuguese in the empire worked for and were beholden to the Corporation, and doubtless exerted themselves, in the civil service or in politics, on its behalf) as on a mutual balance of self-interest. It would have been impossible, simply by influence or bribery, to have suborned a Portugal which retained the powers to grant concessions, to insist on Portuguese participation, to impose taxation, or to control profits. A very high degree of compromise was required of Anglo, t particularly on the issue of ownership (in the long run, to its advantage, when the transfer of power to Frelimo took place). The only hold, as JCI showed, is not to invest - an adequate weapon only in a monopoly situation. The power of multi-national corporations against the state has been grossly exaggerated, as far as Africa south of the equator is concerned. Anglo's other ventures in Mozambique relied also on the input of capital and technology: a concession for fisheries proved unprofitable but it led to a new type of undertaking, in the collecting and processing of cashew nuts in the north-east. t For further amplification see Notes at end of chapter. CABORA BASSA 254

Towards economic autonomy Anglo's civil works contracting side has benefited materially, quite apart from involvement in Cabora Bassa; and after sinking $30 million in unproductive oil drilling the concessions were renewed in 1973, in conjunction with the South African state corporation, Soekor. Mining offered Anglo excellent opportunities, once the war was over - more profitable than its normal cooperation with JCI. (Anglo does not control or manage JCI, but owns 50 per cent of the shares. In the case of Mozambique, JCI allotted, in addition, 40 per cent of its stake to Anglo.) Although the Anglo-American stake remained small in comparison to its investment in Angola, it showed a common pattern, being based on the Corporation's skills and capacity for financial management. On the political level, like the South African Government - if not the South African public - Anglo was prepared to deal with, even to welcome, a black republic. Other foreign companies and governments'4 lagged behind in their understanding of what was happening; perhaps the Belgian and British concerns had been so long established in Mozambique that they had accustomed themselves to perpetual colonial rule. British merchant banks proceeded very cautiously when approached by Lisbon on the fourth development plan.t West Germany, keen to increase a very small stake in the province, was not entirely deterred by the political furore over Cabora Bassa; Egon Schiller, then Minister for Economic Affairs, seemed to be offering Portugal a quid pro quo when he advised the EEC Council of Ministers to facilitate Portugal's associate membership by making concessions on the overseas territories.'" The French Government, building on the political affinity between de Gaulle and Salazar, inherited by Pompidou and Caetano, encouraged prospecting and such investments as the Texlom factories. Neither Government, however, seems to have been aware of how economic enfranchisement in Mozambique was changing the nature of Portuguese authority. As for Mozambique's other neighbours, Rhodesia cared little and remained, as before, profoundly ignorant; Malawi, and to a lesser extent Zambia, privately accepted the inevitability of a neo-colonialist solution; only Tanzania upheld the faith of Frelimo that an independent socialist state could emerge. t For further amplification see Notes at end of chapter. 255

Notes and references Chapter 10 Towards economic autonomy I Richard J. Hammond, 'Uneconomic Imperialism', in Colonialism in Africa 1870- 1960, ed. Gann and Duignan, (Oxford 1970) 1. 2 James Duffy, 'Portuguese Africa 1930-60' in Colonialism in Africa II. 3 Ibid, I, p. 353. 4 A. Freire de Andrade, Relatorios de Mozambique, (Lourenqo Marques 1908) Part III, p. 5. t Note page 238. 1930 - 18,000; 1940 - 27,500; 1950 - 48,000; 1960 - 85,000; 1970 - 205,000; 1973 - 225,000. ft Note page 238. The oil concessions in the Mozambique Channel, and in the alluvial coastal plain, were allotted to five groups In 1968 - Gulf Oil, Sunray- Clark-Kelly; Hunt; Texaco; and the consortium Soci~t des Petroles d'Acquitaine and Anglo-American. f Note page 240. The budget for 1974 envisaged a central government contribution of 3075 million escudos, a Mozambique contribution of 1243 million escudos (9 per cent upon 1973). t t Note page 240. The third development plan, 1968-73, envisaged immigration to Mozambique of 6000 settlers a year; the Overseas Ministry sent out regular recruiting circulars to Portuguese working in West Germany and France (cf. Jornal Portuguis, Cologne, January 1970). f Note page 242. In 1972, Portugal was still buying raw cotton at 5 escudos a kilo, processing it, and selling it at 14 escudos; and semi-refined sugar at 3.7 contos a ton, to resell, refined, at a London market price of 11.8 contos. In this way, the largest share of profit was transferred from Mozambique to Portuguese industry, in metropolitan escudos, by the agency of the Portuguese state. Similar practices governed the marketing of sisal and copra. t Note page 243. Even in 1972 the province's long-term debt was held as follows: Ministry of Finance 3009 million escudos BNU 715 1CM 260 Banco do Fomento (bonds) 783 Banco do Fomento (promissory notes) 40 Savings Banks 21 Treasury Bonds 879 Total 5707 million tt Note page 243. Third Plan, 1967- 73. Public bonds 1972 200m escudos 1973 300 escudos Loans- Banks 424 100 ]CM 424 689 Bank- Lisbon 826 634 Other 21 Fourth Plan, 1974-79: Total 55 million contos, of which 18.5 million were given by credit institutions and Government; the rest came from the private sector and overseas

Towards economic autonomy 257 (of which only 12 million contos was defined as imperative expenditure). Local banks 7.5m. Direct local loans 7.2m. Public bonds .6m. Metropolitan bonds .765m. Note page 244. Railways & Ports Roads Budgets for 1969 1.692m escudos 83.7 1970 1.830 150.9 1971 2.000 354.1 1972 2.300 395.4 1973 2.700 526.0 1974 3.300 540.3 T Note page 245. 1961-9, distribution in the Portuguese budget: 1961 1969 civil expenditure 52.8 per cent 44.1 per cent military 25.6 40.7 development 22.6 15.2 ft Note page 245. To take the single area of the GPZ, only 35 million escudos out of a total expenditure of 511 million escudos came from Mozambique, although during the 1960s the provincial share had been fixed at 15 per cent. t Note page 246. Total outflow, for all three African colonies for 1968: 50 per cent of military budget - 5m contos Net outflow of Government expenditure - 2.4m Military expenditure in Portugal for overseas - 3m Transfers for overseas development - 2m Total 12,400 million escudos. 5 For an analysis, see Maria Belmeira Martins, Sociedados e Grupos em Portugal (Lisbon 1970). 6 As for example, J.A. Lombard, J.J. Stadler & P.J. van der Merwe, The Concept of Economic Cooperation in Southern A./ica, (I retoria 1968). 7 See Leo Marquard, A Federation of Southern Africa (Oxford/Cape Town 1971)and from a critical standpoint, Robert Molteno, Africa and South Africa (Africa Bureau 1971). 8 H. Olivier, l"ater Resources in South Africa (Pretoria 1968). 9 Financial Mai, 27 October 1972. 10 H. Norman, 'Power Supply in South Africa,' a paper for Escom (July 1970). The author envisaged South Africa taking a large part of the ultimate 20,000 MW production of the Congo and 4000 MW from Cabora Bassa. t Note page 251. South African enthusiasm tended to drive the Portuguese onto the defensive. Thus in 1970 a proposed association with Angola and Mozambique in a Planning Committee, sponsored by the Afrika Handels Institute, was abandoned after long preliminary negotiations. 11 South African Institute of International Affairs, Newsletter, 6, No. 2, 1974, p. 26. 12 See also G.M.E. Leistner, 'The Economic Problems and Policies of South Africa's Neighbouring Black African States'. (SAIIA, May 1973). t Note page 253. EDESA, founded in Zurich in 1972, 'to foster the economic development of black countries of Africa, south of the equator', originated in Anton Rupert's Rembrandt Corporation, but the $11 million subscribed in its first year came from a broad band of interests, including the Anglo-American Corporation, Roberts Construction (South Africa), Marubeni Corporation (Japan), and US, German, Swiss,

258 CABORA BASSA and Canadian banks. It is now chaired by Dr Karl Schiller, former West German Finance Minister. Although no undertakings have yet (1974) been sponsored, it is intended to supplement existing sources of finance in Lesotho (Government Information Bulletin, 2. February 1972) and it appears to offer facilities similar to the Commonwealth Development Finance Corporation - but with more obvious political implications. According to its brochure, EDESA aims to overcome the threat of nationalization in Third World countrie,, by underwriting and placing enterprises in such a \% av as to allow local capital to participate in the equity; presumably by setting up investment corporations, financed partly by the state, partly by institutions with power to isue trust bonds for local sale. 13 South African Institute of International Affair,, Ne-usletter, 6, No. 2, 1974, pp. 7-9. f Note page 254. The rearguard action of Anglo-American in Zambia i, instructive. Preparations for decentralization to a regional centre in Lusaka were made well before independence; and as soon as Anglo realized that the Central African Federation was doomed, its companies, a huge holding in copper mining and ancillary processes, were registered in Lusaka. Planning continued, to forestall the expected attack from militant nationalists, and Anglo's survival benefitted (perhaps more than its great rival in copper mining, RST) from Oppenheimer's friendship with President Kaunda. The mines were nationalized in 1970, but the Government agreed to pay, on a generous valuation, for a 51 per cent holding, over twelve years. Then, short of foreign exchange and hamstrung for lack of tax revenue, in the wake of the Rhodesian border clashes, and the Mufulira mining disaster, and a fall in world copper prices, the Government raised a S222m loan, largely on the Eurodollar market, to buy out the bonds in August 1973. At a cost of borrowing at 13 per cent to pay off a debt carrying 6 per cent, they gained control over the repatriation of dividends, and increased their tax revenue; but Anglo was able to retain management control. As a new era in the struggle began, Anglo diversified; and, it i, rumoured, have been planning to refine Zambian copper in the demilitarized zone of Suez, using Arab finance. The mining companies can spin out the process for a decade or more, without losing so much ground as to become unprofitable. But in the end, survival depends on value, as perceived by the state in whose territory the multi-national company operates. 14 The largest holders of investment in the Portuguese colonies in 1970 were: Belgium, 775m escudos; US, 158m escudos; South Africa, 105m escudos; Britain, 53m ecudos; West Germany, 19m escudos. J. Martins Angola, Part V (Rome 1970). t Note page 255. A BNEC (Southern Africa) survey of the Mozambique market, 1971, affirmed a belief in dramatic growth, when minerals became commercially viable and oil was found; 'In the meantime, however, the structure of the economy is not strong and the present position of the balance of payments must cause anxiety to the authorities and to the entrepreneurs who feel it directly in the lack of money and tight credit conditions.' 15 Die Europaer, 103/4, 1969.

Chapter 11 Que e Mozambique? The corporations, banks, and individual industrialists who sought ways of insuring their investment in the future of Mozambique scarcely envisaged a hand- over to Frelimo; most of them appear to have expected a 'revolution from above', such as was proposed by General Spinola to Caetano early in 1974, accompanied by a gradual transfer of political power to a mixed white and black Government. In this they relied on Caetano's optimistic assumption that stable political institutions could be created within the framework of the new Organic Law. But outsiders, even in South Africa, knew very little of the real situation, and relied almost wholly on the evidence presented by the Portuguese civilian and military authorities about the success of the 1971 reforms and the 'hearts and minds' campaign. They saw no need to question the Portuguese definition of Mozambique as a coherent state. Yet it might well have been asked: 'What was Mozambique?' In his directive to the troops about the use of the Portuguese language, in May 1970, Samora Machel had empha3ized the need to teach 'que Mozambique'. Territorially diffuse, like most of the late nineteenth-century states in Africa, ethnically and tribally divided, it possessed resources and population, but was it in any sense a nation, as distinct from an administrative province? Mozambique fitted generally with the description given in a survey of African states in 1968: 'On the whole, African countries are distinguished from other Third World clusters by extremely weak national centres, a periphery which consists of societies until recently self-contained, and levels of economic and social development approaching the lowest limits of international statistical distribution." Her black population was growing at 2.2 per cent a year and, although the white sector expanded faster by immigration, by 1973 there were still only 250,000 whites to 9.3 million blacks, and a small group of roughly 15,000 Asians. Of the whites, less than half had actually been born in Mozambique, or had their residence and wealth wholly in the country; the rest comprised administrators, civil servants, bank officials, and so on, whose houses and bank deposits rested in Lisbon, or businessmen with a substantial stake either in the metropolis or South Africa. For most of them service overseas might last half their working lives, but it remained, essentially, a career and a vehicle of promotion. They acknowledged a crude truth in the slogan: 'Without Africa, Portugal is nothing.' A deep and traditional resentment of the payment of their salaries in the metropolis, and their long periodic 'leaves', helped to define the white Mozambicans - a disparate group, often of mixed racial origin, for whom no other homeland existed, and for whom the tombs of the early navigators in the castle of Ilha da Mozambique provided the cornerstone of history. The Asians formed a series of coteries within the general commercial community. Among the 15,000, a large contingent of Portuguese Goans, and another of Ismailis, owed a traditional commitment to the country; many Indians and Pakistanis, on the other hand, watched for trouble, safeguarded their rights to external, sometimes British, nationality, and transferred money where possible to South Africa. By 1973 they were restless; and following a wave of bankruptcies, and takeovers by Portuguese shopkeepers of cantinas in the aldeamentos, they began a cautious exodus. Ethnic and tribal divisions, and migrations, precluded any historic claim that the 9.3 million Africans formed a nation, as could be maintained, say, for the heirs of Monomotapa in Rhodesia. The eight main ethnic groups, Nianja, Makonde, Macua, Marari, Shona, Chapi, Swazi, and Shangaan, comprised more than fifty tribes. Five principal languages Swahili, Nianja, Yao, Swazi, and Shangaan, - subdivided into twenty-three dialects. Geographical frontiers meant little CABORA BASSA 260

Que Mozambique? when Makonde, Nianja, Angoni, Yao, and Shangaan, felt greater affinity with their compatriots in Tanzania, Malawi, Zambia, Rhodesia, or South Africa, than with their Mozambique neighbours. On a demographic basis, also, the country lacked unity: vast areas contained less than one inhabitant per square kilometre, and the mass of the population was concentrated in Zambesia, Beira district, and the far south 400,000 alone in the sprawling suburbs of Canice outside Lourenqo Marques. Nor did religion offer a substitute unity, for, in spite of a century of missionary work, Catholicism had put down few roots: the bulk of Africans remained animists, and in the north-east there existed a large Islamic contingent. White Mozambique's history up to the mid-nineteenth century was written on the water: treaties with native rulers, with the sovereigns of Monomotapa, a few postos in the interior, and agreements over the gold of Manica, or tributes, enforced by sporadic wars, comprised the sum of politics beyond the coastal region. A colonial system scarcely existed until the end of the last tribal wars in Gaza in 1895, and the suppression of the Angoni revolt in Tete in 1918. The prazeiros ruled in their own domains, the chartered companies engrossed more than a third of Mozambique. Effective military rule was only completed in the period 1905-14; and civil administration, product of the reform of the Escola Colonial, dates from the early 1920s. The visionary and impractical Colonial Statute, though passed by Salazar in 1930, actually completed the first stage of Portuguese rule, bringing the country to the level achieved in the Rhodesias, Nyasaland, and German Tanganyika, some thirty years before. Mozambique's subsequent political development was shaped by the Portuguese dream: the incorporation of an imperial tradition, rooted in Portuguese culture and Christianity, into a 'spiritual unity' of provinces with the metropolis. The New State insisted absolutely on advancement for the African down the single path of Portuguese language, history, and education, and Portuguese ideals: the State offering mediation as the Church offered salvation to the devout. Needless to say, the white community benefited, as it did from the post Second World War progression from colony to province, while the overwhelming majority of Africans remained in the 'primitive 261 condition' for which the regime do indigenato was prescribed. Pace the accusations of Amilcar Cabral and Mondlane, there was probably no conscious attempt to destroy tribal culture and language, only to emphasize that advancement on that basis was impossible. Indeed, paternalism provided some safeguard for what was defined as the child-like nature of the indigena, to which a leading Portuguese sociologist ascribed the relative tranquility of rural life in the 1950s.3 On the other hand, African rural life was bounded by strict liabilities for forced labour (only ameliorated after 1950) and payment of taxes; and the political authority of traditional chiefs was abolished in favour of a hierarchical system, descending to the white chefe do posto and to his nominees, the regedores or headmen. Benefits could be obtained only through acquiring assimilado status, via an educational system so fragmentary and so inadequate for the provision of mass literacy that a mere 5000 assimidados existed in 1950. Yet the aim stood; as Morais Cabral wrote in 1939: 'Our whole policy has been and continues to be to raise the cultural, economic and social level of the Negro, to give him opportunities and to drag him from his ignorance and backwardness, to try to make him a rational and honourable individual, worthy of the Lusitanian community.'4 The growing disparity of wealth and opportunity between blacks and whites, however, weakened an earlier basis of inter-racial mixing and enforced, except in rural areas, an increasing degree of economic apartheid. When African demands for a reversal of this process erupted in the late 1950s, the apparatus of coercion was used to undo what little strength had developed in associations such as the Liga Africana and the embryonic African trade-unions: the issue of passbooks, the operation of the Labour Laws, and the manoeuvres of the PIDE, accomplished either their suppression or exile, so that, in contrast to the experience of neighbouring British colonies, Mozambique possessed no indigenous African political structures up to the moment of decolonization. Those few who had passed through enseno rudimentar to secondary and finally university education faced a choice of total assimilation or rebellion; hence the predominance of intellectuals with experience of Portuguese universities among Frelimo and its splinter groups. CABORA BASSA 262

Que Mozambique? The wars accomplished a certain modification. Under the 1963 Organic Law more functions accrued to local government and to the regedores. Having pruned the abuses of the contract labour system, Adriao Moreira, then Overseas Minister, abolished the regime do indigenato. Mozambique and Angola acquired representation in Lisbon and councils to advise the Governors-General. Characteristically, the white constituency elected twenty-seven members to the Legislative Council, the 'indigenous authorities' only three. The war, however, had other effects: although it scarcely showed in the 1962 provincial elections, when most of the few African voters supported opposition candidates associated with General Delgado, a new division opened up, as the whites, of whatever political complexion, took an apparently irrevocable stand on racial lines. In this sense, Frelimo's revolt exposed the fundamental contradiction in colonial policy; but in practice, over the following decade, it deprived the vast African majority of liberal white support. Only 20 per cent of qualified Africans voted for white candidates in 1968. During the 1960s the white opposition itself foundered in ambiguity and persecution. Delgado himself was able to retreat for a time from Portugal to a political base in Brazil; but his followers in Mozambique, led by Almeida Santos and Soares de Melo, failed to sustain the party in the 1968 election in the face of rigged ballots, legal manipulation of electoral qualifications, and a fraudulent register, and held only a few seats in Beira and Quelimane. The Government offensive continued, with the aim of forestalling a link with Frelimo, and eventually, unwilling or unable to take the path of underground activity against a highly efficient PIDE, the old liberals abandoned politics altogether. The Delgadists' programme had not offered a very clear alternative to colonial policy, only amelioration of its harsher aspects. They advocated self-government, to be acquired over an unspecified period, education for the African, and a long, slow transfer of power from Portugal, first in home affairs, finally in defence and foreign policy. Much preoccupied with the external position, they looked to membership of the United Nations, and hoped that the loyalty of the African population could be held by their multi-racialism, against the 263 strident nationalism of Frelimo. Their policy was most clearly articulated in a secret letter to Caetano in 1969, signed by thirteen leading citizens of Mozambique and Angola, which protested against the closed fascist system and demanded progress towards independence, but within a federal scheme of the sort which Caetano had already rejected. Development of the resources of the population, the ending of the war, reconciliation with the rebels, and representative institutions, were aims common to almost all whites, and in Caetano's early days the white centre held fast to the ideal of integration with the metropolis. Caetano himself received the greatest applause during his 1969 tour that Mozambique had meted out since President Carmona's visit thirty years before. Few foreigners realized how sincere this homage was; and it seemed to have been repaid by the unrolling of the reform programme. Only two small groups at this time favoured independence under a white Government: the settlers of Manica and Sofala, including a number of Germans and South Africans; and a group of industrialists and technocrats, including Champalimaud, da Cunha, Manuel Bullosa, and Jorge Jardim, who were associated with Castro Fernandez, then chairman of the BNU. They controlled two of the main daily papers, and looked towards Rhodesia as a model; but some altered course during the next three years. In his speech announcing the revision of the Constitution, in December 1970, Caetano rested his case securely on the precepts of Salazar, yet, in a characteristic manner, shifted his ground: protesting, nevertheless, that autonomy was no more than fulfillment of the old Colonial Act. He equated 'spiritual assimilation' (as defined by the 1933 Constitution) with legislative as well as economic autonomy, and announced the Government's intention of meeting 'national desires'. The ambiguous phrases promised what his hearers wanted; later it could be seen that the old certainties had been abandoned. Under the Constitutional Law, in 1971, Mozambique became a state. After two years of preparation by administrative statutes, elections were fixed for a new Legislative Assembly. The Government intended to develop white and black political cadres over a period of ten years, on the basis of a sound and prosperous economy; but since autonomy meant CABORA BASSA 264

Que Mozambique? that the country inherited the corporate-state system of Portugal no party except National Action (ANP) contested the 1973 election. Only 2 per cent of the adult population voted; yet this 111,000 numbered 91 per cent of registered voters (as against 71 per cent in Angola). They returned eleven black or coloured members and nine whites for the elected seats. Most of the remaining thirty places were filled by the 'estates' business, banking, public service, sindicatos, etc. - but the urge to present a multi-racial aspect was now so strong that the Governor-General could be found filling his own nominated seats with names until recently on the proscribed list of the secret police. In spite of the appearance of urgency in Mozambique, movement in Lisbon was as slow as state elephants in procession, and the Legislative Assembly, scheduled to meet in May 1974, never actually convened. Instead, other forms of political organization supervened, despite the denial to Africans of any political activity outside the ANP and the sindicatos. The black elite having either defected or assimilated, sub-elites developed - not on specifically political lines, but with a certain degree of political consciousness - among the professionally qualified: schoolteachers, doctors, lawyers, officials, and clergymen. Many were arrested by the PIDE/DGS, and almost all African priests and clergy suffered in the grim Machara detention camp outside Lourenqo Marques. Domingos Arouca, the lawyer, stood out, both for his defence at his trial for subversion and for his courage during eight years' imprisonment. Groups lower in the social scale attracted less attention; but suspects among African clerks, white-collar workers, and artisans, probably suffered more, since the PIDE rarely brought them to trial. Not all chose to express their beliefs in opposition: the ANP attracted numbers of Africans, much like the Bantustan leaders in South Africa, prepared to work within, but not to endorse the aims of, the corporate state. Men like Nazard and Max Fernandez made their voices heard as Mozambique representatives in Lisbon, and attracted substantial support at home among their own people. But they encountered the same problems as did Frelimo - tribalism, and the growing significance of the army. 265

Recruitment for the army among the African population had found favour with the Portuguese authorities for many reasons: it offset some of the cost of the war, since their wages came from Mozambique funds; it reduced the drain by conscription on Portuguese manpower; it involved the local population directly in the anti-guerilla war and anticipated a future self-defence force; and it could even be seen as a perverse form of social promotion, since it reduced unemployment and added a little (though in marginal fields) to the narrow base of African skills. As the war went on, ever larger drafts became necessary, which the Commander- in-Chief, General Kaulza d'Arriaga, turned into a virtue in his pacification programme, by instituting the special corps of GEs and GEPs (p. 144). The black army never grew into a homogeneous political organization: most of the 35,000 soldiers, at its peak, had served too short a time, and were too conscious of their new status to be affected by the war-weariness of the old regular army and the Portuguese conscripts. But even these became a demanding and potentially volatile force - 'he who puts on shoes, does not wish to take them off.' They aspired to the rank of an elite and took pride in the fact that they were often pushed into the fighting, in the latter years, where white troops were held back for garrison duty. The GEs, recruited from single tribes, never lacked for volunteers; and the NCOs generally, on the evidence of a number of Portuguese commanders, came to regard themselves as an important professional cadre within the body of the state. Unlike the handful of black officers who tended to adopt Portuguese attitudes, these looked warily at Frelimo, understanding something of their appeal, and of the situation which could involve them personally if the Portuguese withdrew. D'Arriaga believed in the army as a means of creating a truly Mozambique elite; for different and more devious ends, the DGS recruited for its flechas, widened its network of collaborators and informers, and tried to establish an imperium in imperio against the expected time of failure of the policy for autonomy. Aided by white adventurers like Daniel Roxo, who raised a private commando against Frelimo, the DGS deliberately inculcated tactics of violence in training theflechas, and took whatever 266 CABORA BASSA

Que t Mozambique? chances were offered of arming white settlers as authority crumbled in 1973-4. There is some evidence that a transfer of power, not to Frelimo, but to a white group, in association with the Mozambique DGS, would have been welcomed by right-wing politicians in Lisbon. The black army certainly fostered a form of nationalism; but for Mozambique's neighbours it raised uncomfortable comparisons with militarism elsewhere in Africa: coups in Nigeria, Ethiopia, Somalia; nearer home, attempted coups in Kenya, Uganda, Tanganyika in 1964, and with the rise of Idi Amin in Uganda. Concern about the possible 'Congolization' of Mozambique - between two armies far larger, better trained, and better armed, than those of Tshombe and Lumumba - forced Presidents Kaunda, Banda, and Nyerere, to say nothing of Prime Ministers Smith and Vorster, to take preventive action (Chapter 12). A pervasive element of nationalism, hard to define, and largely unnoticed by the administration and the secret police in their preoccupation with Frelimo infiltration, developed among the tribes, though not necessarily in conformity with traditional modes of organization. Tribal elites had grown up among the Makonde (as a result of implementation of the pacification programme through local chiefs), the Macua, and the Shangaan, which, without actually evolving into political groups, took up competitive positions vis-A-vis the Portuguese Government, with an eye to their advantage once the introduction of autonomy began. In some cases, such as that of Kavandame, traditional chiefs resumed a leadership forfeited half a century earlier; in others, a new group emerged out of the changed socio-economic conditions of the tribe. Thus the efforts to win Macua support, to buttress Zambesia against Frelimo entailing not only the building of villages, towns, and roads, but also the provision of large farms, with working capital, and the decentralization of local government - produced a vigorous regional association among educated and skilled Macuas. Later, as Portuguese control disintegrated, attempts were made to capitalize on Macua ambitions and regional coherence, in order to forestall the Frelimo advance. The Shangaan experience was different, for it owed very little to politics. The Shangaan inside southern Mozambique 267 number about two million; the tribe laps over into the Lowveldt of South Africa (750,000, 4 per cent of the total African population), and into Rhodesia (600,000). Over the seventy odd years of the contract-labour system, a very high proportion probably 80 per cent of the male working population - had found work in the mines, and the entire social life of the tribe had been conditioned by the contract cycle.t This is not the place to evaluate the system (whose benefits were provision of employment, at wages higher than could be found in Mozambique, at the cost of disruption of family life, perpetuation of the payment of lobola for wives, and neglect of agriculture); but it is worth observing that, as a result of working within the 'boss-boy' hierarchy of the South African labour camps, an elite had been formed on a basis of labour skills. Those who became boss-boys were given responsibility and command; they served ten or even fifteen annual contracts and retired relatively wealthy and respected, to farm on tribal lands. South African managers at Cabora Bassa faithfully copied the Rand model; but the results were surprising. Shangaan indunas had found no meaning in nationalist struggles in South Africa. Working at Cabora Bassa, however, they perceived, first, how they had suffered financially from the exchange control provisions of the contract system,tt and second, how their skills and organization could be used in their own country. Conversation with the indunas there5 suggested that, while proud to work on so great a project for Portugal, they were prouder still of being able to prove themselves Mozambicans. 'The work we do here shows others what we can do; it makes our country better.' They looked to the dam to provide jobs for their children, and to help their neighbours in Malawi. In their appreciation of the link between farmers, water, and electricity, they unconsciously reflected the Plan Geral (though the Portuguese had neglected to provide them with any explanation of the purpose of the Zambesi scheme) and, no more naively than the whites in Lourenqo Marques, they pointed to the revenue it would shortly bring in. Most important, politically, they now desired t For further amplification see Notes at end of chapter. + For further amplification see Notes at end of chapter. CABORA BASSA 268

Que 1o.zmlbique? to work in Mozambique, never to return to South Africa. Despite their continuing loyalty to Portugal, they, and a large number of their men, had discovered a national consciousness; in due course it found tongue, during the strike of May 1974, in the demand that all South African managers be withdrawn. True to the tradition of the indivisible Portugueseness of things, the authorities built on generalities, not tribal institutions. Primary education had been expanded more than any other service, for a growing teenage population.t Out of 2 million of school age in 1973, 800,000 were actually in schools of various sorts (418,000 in 1960; 520,000 in 1970; 800,000 in 1973). At the same time, under the liberal categories of the new constitution, voting qualifications were lowered to admit anyone who could read or write simple Portuguese. Although only 111,000 actually registered for that election, 1973 figures suggested that more than a million could have qualified - a black majority of more than 80 per cent. This predominantly youthful electorate, culturally more homogeneous than before, raised hopes that it might serve as a power-base either for the ANP or for the protagonists of a democratic party system. Meanwhile, accelerated economic autonomy, the uncertainties of the war, and the febrile atmosphere of 1973, encouraged lurid fantasies among the white population. The Portuguese Government may be blamed for excessive optimism about the growth of sound institutions, because the precondition of economic stability was never achieved. Instead, bewilderment and lack of direction followed the tacit confession of inadequacy by the authorities. Inflation eroded faith in existing institutions and encouraged irresponsible outbursts. Four attitudes of mind could be discerned. Separatism flourished now among white farmers and small businessmen, many of whom, bankrupted by import restrictions and lack of credit, looked to a future in South Africa. There was talk of partition, abandoning the north to Frelimo, Tanzania, and Malawi, and holding the Zambesi in conjunction with Rhodesia. This solution attracted the gilded youth of Lourenqo Marques, the fileus de papa' who had scarcely known that the f For further amplification see Notes at end of chapter. 269 north existed until conscription touched them; and, later, the DGS - both the generally low-calibre assistants and the senior officers, hard, ultramontane counterparts of the South African Bureau of State Security - whose mandate from the Ministry of the Interior (over the head of the Governor-General) seems to have been to salvage white supremacy at any cost. Second, those based in Portugal began quietly to emigrate or repatriate their funds. According to the Planning Department in Lourenqo Marques, the white population fell by 20,000 in the eighteen months to December 1973. The third grouping combined political activists in the armed forces with democrats born in or tied to Mozambique. Young conscripted officers, whether local or Portuguese, generally reacted to the war by becoming more radical, and the corrosive influence of their criticism rubbed off onto their war-weary regular army colleagues. The democrats included old Delgadists, businessmen, and members of the professions: the first stirring of what later became CDE could be found in the decision of men like Jorge Abreu and Adriao Rodriguez to appeal to the African electorate, and acknowledge the justice of Frelimo claims. Such a leap from the political affiliations of the previous decade required great courage: it was paralleled by a movement among whites in remoter regions who, probably in order to survive, allowed Frelimo to know their hidden sympathies. But this movement was still in the nature of individual treaties with the unknown. Although the so-called Third Force emerged at the end of 1973, and met with notoriety and harassment even in the National Assembly in Lisbon, it avoided contact with the African working class or peasantry (possibly to avert suppression by the DGS) and waited, passively, for Government recognition on the lines of Sedes - that is, not as a political party, but as an association (for anything, sport, or social promotion) with the right to hold public meetings. Its leaders, and a motley mixture of whites and defectors from Frelimo and Coremo, inclined to challenge the ANP through the machinery of the Legislative Assembly. Jorge Jardim now emerged as leader of a fourth group, as a builder of bridges between Frelimo and the Mozambique Government. He believed that modern Portugal had failed to replace the harmony between authority and people which, as CABORA BASSA 270

Que t Mozambique? a monarchist, he ascribed to the past; once a supporter of Salazar and an associate of the commercial-industrial oligarchy, he now advocated a 'new Lusitania', incorporating Brazil as well as Angola and Mozambique, on the basis not of federation but of an association of democratic states - not unlike the 'free external association' put forward by de Valera for Ireland in 1921, and later the model of post-war British Commonwealth development. While he acknowledged Frelimo as a force, he condemned any direct negotiations between it and Lisbon, on the ground that Frelimo represented only part of the Mozambique nation. He attempted a tactic of great complexity: using his friendship with Presidents Kaunda and Banda to bring home to Frelimo the need to acknowledge the rights of the black majority and white minority; using his favoured position with Caetano to urge the rapid granting of independence and free elections; and, inside Mozambique, working to enrol the million qualified voters, in order to break the monopoly of ANP. Unlike the democrats, he had financial resources; he had support from well-disposed African sections of the army, powerful influence in black Africa, and a coterie of defectors from Frelimo, like Miguel Mrupa, or martyrs of DGS persecution, like Domingos Arouca. Jardim owned and used Noticias da Beira to further his cause, and Mrupa's popular weekly, Voz Africana; yet he too, like the Third Force, depended on licence from the authorities. He needed status like Sedes; and time, for the electoral strategy could not be ready for at least two years. Unlike the democrats, Jardim's platform could not be compatible with Frelimo's claims, for he challenged Frelimo's right to legitimacy: 'We Mozambicans cannot accept a group outside Mozambique as its sole representatives.' Jardim moved in the shadows, suspected by the democrats and reviled in Frelimo propaganda. He depended too heavily on influence in Lisbon, rather than Lourenqo Marques, and after the failure of Kaunda's peace overture of September 1973 (Chapter 12) he lost ground. But the strength which he perceived among African sub-elites undoubtedly existed. In many regions, even in Manica and Sofala, local government by regedores had begun to break down; there was a rash of requests for higher wages and for the power to organize unions, 271 at the end of 1973, an overture to the unrehearsed debate which followed publication of General Spinola's Portugal and the Future, and a warning that political demands might shatter the carapace of confidence in Portuguese reforms. The DGS became aware of incipient unionism outside the framework of official sindicatos, in the docks and on the Mozambique railways. The degree of sympathy with Frelimo in 1973 is hard to assess, and cannot be read back from the post-revolutionary wave of euphoria. A DGS map, captured by Frelimo, shows concentration of suspects all over the country, even in Gaza and Inhambane in the south; heavy in the suburbs of Lourenqo Marques, despite the total absence of urban warfare, and patchy only among the Macua and Shangaan. According to Frelimo, a number of traditional chiefs had secretly pledged allegiance; and regular reports on public opinion kept Dar es Salaam in touch with 'Fete, Nampula, or Quelimane. Yet, for the majority of Africans, Frelimo was still a mythical being, aggregating all sorts of wild hopes and rumours. Total censorship had not prevented them from appreciating that in ten years the guerillas had progressed from being moleches (corresponding to the US army view of the Vietcong as 'gooks') to inimigos (enemies); and Caetano's striking offer of reconciliation, in March 1974, was widely published in the African press: 'We are ready for dialogue, to discuss the return of the terrorists to their homeland, their reintegration in Portuguese society, and even to study their participation in administration and local government.'6 But actual information about, say, the Frelimo 1972 Congress, filtered in only by word of mouth. Early in 1974, placid Lourenqo Marques woke up to find some of its leading citizens advocating compromise with terrorists and ex-guerillas, or like the flamboyant Coremo defector, Joanna Simiao, marching from foreign embassy to foreign embassy, trading promises of political support. As the Governor-General and the DGS held back, suspicion grew that licence had now been issued to find ways towards independence: which found confirmation in the publication of Spinola's outspoken critique. But these disparate groups had no firm base. Instead of attracting support of the manifestly weary among the civil adminstration and the army, they 272 CABORA BASSA

Que Mozambique? seemed likely to lead uncontrollably to faction, perhaps civil war. While hysteria coloured expectations in Mozambique, cooler heads in neighbouring countries and in Portugal assessed the weight which two institutions, the army and the state itself, could bear. Except for the right wing in Lisbon, an army coup had no attractions; and was, in any case, inherently untimely, in view of the surreptitious growth of the Armed Forces Movement. On the other hand, the framework of the corporative state, faithfully copied from Salazar's 1933 model, corresponded remarkably closely to the lineaments of the one-party state as developed in much of black Africa. If Mozambique had no institutions capable of sustaining democracy, why waste the years which Tanzania and Zambia had lost before arriving at presidential and one- party rule? For some years after independence in most of the former British and French colonies it seemed that the nationalist parties had spent themselves in the struggle against the colonial power, and that their function had been to create a governing elite, dominated by one man.7 In no case did the party appear strong enough to withstand either the fall of the individual ruler (the CPP disintegrated in Ghana, when Nkrumah was deposed) or a military coup (such as Amin's take-over in Uganda). But recently in East and Central Africa there have been signs that the ruling party has become a vehicle of substantial power, supporting and, to a great extent, guiding the actions of the leader. These cases are of interest, because of their close influence on Frelimo and on developments in Mozambique. The Tanganyika African National Union came into existence seven years before independence, in 1954, as a means of mobilizing support for Julius Nyerere; originally intended to be an intellectual group, it widened out into a mass party and achieved a united front after independence, not by compromise with the other parties, but by eliminating them. Then Tanu appeared to atrophy; its structure was subordinated to Government authority, especially after the army revolt of 1964, and in the 1965 elections it was further weakened when Nyerere allowed a diversity of candidates under its auspices. But later, when the Government suffered the defection of its 273 ablest minister, Oscar Kambona, and encountered increasing opposition to its extension of socialism, the need for mass support, transmitted through the party, became urgent. Nyerere was vulnerable, and after the deposition of his kindred head of state, Milton Obote, felt himself threatened without a party shield. Hence the Arusha Declaration of 1967, and the acceptance of 'Tanu Guidelines' which strengthened internal security, extended nationalization, and achieved much wider public consent, at the expense of Nyerere's own power. In 1972, for example, seventy Tanu MPs overturned a Government bill to permit MPs to hold the office of regional commissioner; and as Nyerere retreated more into the style of mwalimu, or teacher, the Government itself came to be seen as an extension of the Party. In Zambia, too, the party sometimes provided unity over tribal dissension, security against foreign intervention (or intervention by exiles), and legitimacy of succession. UNIP (United National Independence Party) channelled tribal energies and united the differing leaders, Kapepwe and Kenneth Kaunda, at the time of independence. But Kaunda allowed democratic powers which, as in Kenya, enhanced tribal and economic divisions. For a time the party foundered, battered by opposition at elections and excluded from political power by the technocrats who held the majority of Government offices. Party jealousy, and fears of an unholy league between these 'Young Turks' and the white South, emerged periodically in backbiting and skirmishes, and eventually led to the 1974 'leadership code', introduced supposedly to root out corruption but actually to consolidate party authority. But UNIP triumphed only because the opposition parties nearly brought Kaunda down. Kapepwe's UPP, backed by the powerful Bemba tribe of the copper-belt region, launched a populist campaign for socialism, as opposed to the mixed economy of Zambia, and aligned itself with the student movement. Fearing a return of the disastrous 1968 election, Kaunda successively closed Lusaka university, imprisoned Kapepwe, and formally instituted a one-party state in the autumn of 1971. Kaunda later declared that he regretted not eliminating the parties earlier; and both he and Nyerere have claimed that CABORA BASSA 274

Que t Mozambique? greater popular consensus derives from the single-party system. UNIP now incorporates primary elections and a vigorous youth movement, like Tanu; and it has assumed a large share of power in Government via the Central Committee in Lusaka. In contrast, the Congress Party in Malawi, creation of Dr Hastings Banda before independence, remains the vehicle of his personal rule. But there are certain similarities: the party served to restore stability after Banda's expulsion of Harry Chipembere, for challenging his power in the Cabinet in 1964; it serves still to control traditional tribal chiefs and to mediate between the Life President and the Malawi people. Given that Malawi is nearer to a dictatorship than its neighbours, it is not surprising that the Malawi Young Pioneers are closer to the archetypal fascist youth movements; and the press is not permitted even the limited criticism of Lusaka or Dar es Salaam, since it is largely owned by Banda himself - the Beaverbrook of middle Africa. But the Congress Party is popularly based in a way which the others have only recently developed, and it has given Banda - a populist himself - a framework for an independent and often courageous policy. Now, in his old age, it represents security, for the presidential succession is like the search for a legitimate male heir in Tudor England. Cautiously and secretively the three organizations, UNIP, Tanu, and MCP, have also developed common links and affinities, in spite of conflicting state interests; and some African political analysts believe that the era of presidential leaders is being replaced by another, of party caucuses. After a decade of independence, stereotypes of democracy or fascism have little meaning when the need is for structures capable of conferring popular consent on governments, without the countervailing discords of ethnic, religious, or economic rivalries. It is also arguable that the one-party state is more closely akin to traditional tribal forms of government than any other developed in modern Africa. The similarity of the course of Frelimo's development from the period of one-man domination to its present collective leadership, and its claims to hold power on a popular mandate expressed through the party, scarcely need emphasizing. While Tanu's importance as a model cannot be ignored, however, the 275

Portuguese state itself, against which Frelimo revolted, offers an equally valuable example. As a foreign diplomat in Lourenqo Marques observed cynically: 'All that is necessary for Frelimo to take power, is for the DGS to hand over their files and replace the Governor-General by a black President.' A few perceptive observers in Mozambique realized, when Portuguese dominion collapsed in 1974, that it was too late to search for democratic institutions. Failing a quiet transfer of power, civil war and 'Congolization' threatened. Kaunda, Banda, Nyerere, and several South African leaders, could all have agreed on the proposition that Mozambique was too vitally placed to be allowed to disintegrate. She had become her neighbours' concern, and the closer some form of independence, the greater the possible danger to them. There seemed to be only a choice: rule by a multi-racial group, or by Frelimo using existing forms of government; failing which, a military coup led by an Idi Amin, or white generals seeking UDI. The only distinction between black and white neighbours, therefore, lay in their preference of brand of one-party government. CABORA BASSA 276

Notes and references Chapter 11 Que e Afozambique? 1 A.R. Zolberg, The Structure of Political Conflict in the New States of Tropical Africa', in Ainerican Political Science Review (March 1968). 2 See Amilcar Cabral, 'Culture and Colonialism', published (posthumously) in The Ecologist. 4, No. 7, 1974. 3 R. Ferreira, 'The concept of the tribe in Mozambique' Bol. Soc. Est. Moz, 27 January 1958. 4 M. Cabral, 'A vitoria do nosso espirito colonizador', in 0 Alundo Portugu s, 6, (Lisbon, May 1939), p. 216. + Note page 268. The numbers of Mozambique Shangaan employed in South Africa in any one year, from 1964 to 1972, averaged 110,000 (South African Chamber of Mines). These figures did not include men employed by non- members of the Mining Labour Organization, or illegal immigrants: the true total might be 15 per cent or 20 per cent more. Since few men take on more than two contracts, the total over nine years may be estimated at 400,000, that is, 80 per cent of the male working population. +-+ Note page 268. Under the agreement with South Africa, renegotiated in 1971, a part of the labourer's wages was paid in cash in Rands, which he was supposed to use for living expenses, and to convert at the frontier on his return. Too often, in practice, he was cheated by unofficial exchanges, or at the cantinas or whore-houses along the border. The other part was paid in gold to the Government of Mozambique, and handed over, in Mozambique escudos, less taxes, to the man on his return. As taxes rose, and as the Rand was devalued, the old equation 'One pound (2 Rand) equals 100 escudos' ceased to be accurate. By 1974, 2 Rand equalled 80 escudos; and a deep, though partly unjustified grievance, was felt. 5 Held in English and Shangaan, without official supervision, with the chief indunas, Salvador and Lousinho Mauss6, Constantino Ezimba, Mario Mutumbane, and Machote Makwakwa. t Note page 269. Infant and child mortality rates fluctuated violently in earlier years, owing to the haphazard incidence of disease, flood, or famine. Thus, according to the 1940 cen-sus, 21 per cent of the population was under five, but in 1950 only 15.3 per cent. Due to improved health and living conditions this rose rapidly, so that in 1970 it was 22 per cent, with 58 per cent under twenty. 6 Speech to the National Assembly, 5 March 1974. 7 Cf. Gwendolin Carter, African One Party States (Cornell 1962); and Ken Gladdish, 'Systems of Government in Africa', Journal of Contemporar, Histor,, 4, No. 1, 1969.

Il

Part 4 Chapter 12 Mozambique and Southern Africa From the 1890s until 1969 Mozambique remained a backwater, a part of Africa ignored except by the few who took up the case against Portuguese colonialism, and actually unknown save to those who had experience of the lazy hedonism of Lourenqo Marques. Cabora Bassa and the dramatic extension of the war changed that; but it was not widely known, before the Portuguese revolution, how anxiously her neighbours watched, seeking to discover from obscure pointers the balance of power, and tempted increasingly to intervene in their own interests. The significance of their interest varied according to their relative power and leverage in Lisbon and on Frelimo; and a comparison will serve not only to show the multiple dimensions of the 1974 crisis, but also the devious play of factors concerned. Rhodesia had to contend with the sharpest outcome of Cabora Bassa, the substantial weakening of her position by the final transfer of power from Portugal to Frelimo. At first, in the years of bland confidence in the security of the Zambesi border, the Government in Salisbury concerned itself chiefly with Cabora Bassa as a source of energy. As a trading partner, Mozambique had never offered scope, because of its tariff barriers and preferences for Portuguese goods; these constraints increased during the period of attempted import substitution. But the offer of energy from the dam tempted the Ministry of 'Transport and Power, because, in the years after UDI, Rhodesia's demand was rising faster than the world average and doubling every seven or eight years. Rhodesia's entire import substitution business - essential to avoid a crisis of foreign exchange - had been built on electric power, a dependence only enhanced by the rise in oil prices and the difficulty in obtaining oil after the Yom Kippur war. Mining and chemicals, the two fastest growing areas of the economy, relied on the supply of electricity more and more heavily. Without Zambian cooperation, no new dam could be built on the Zambesi; nor could the Rhodesian Government rely on the output of Kariba North, transmitted through the CAPC. As relations with Zambia deteriorated, Rhodesia was left with the choice of taking a Portuguese supply for the 250 MW needed by 1977, or building a thermal station at Wankie. The latter had great advantages - it could be expanded indefinitely, it was secure from sabotage of the transmission line, and it would burn the low-grade surface coal of the Anglo-American Corporation's Wankie colliery. But the cost of the machinery had to be met from declining reserves of foreign exchange - so serious a contingency that for three years the Rhodesians flirted with the idea of buying power from Cabora Bassa. South Africa would not permit a tap-line; and, since the surplus of Cabora Bassa South was reserved for the Mozambique grid, Rhodesia looked to an AC supply from the north-bank station. Negotiations dragged on, in a rather formal fashion, while the Portuguese waited to discover Malawi's intentions, and whether South Africa needed enough extra power to warrant the capital cost. But Rhodesia's power crisis overtook this leisurely planning; by 1976 the country would be running out of base load, and it would be facing a shortfall in 1977. Since four years were needed to build at Wankie, a decision had to be made before 1974; and, fearful of the exposure of the Cabora Bassa link to Frelimo, and that a Portuguese contract would encourage Zambia to break up the CAPC, the Ministry chose Wankie. Given the circumstances, they were probably right: a Cabora Bassa line might well have proved indefensible, and all foreign exchange reserves had to be concentrated on renewing the run-down railway system, without which Rhodesia could not survive at all. But Wankie suffered the serious drawback that the machinery had to be imported from Europe, South Africa being unwilling to falsify the documentation for so obvious a transaction. The £130 million could be raised, partly from reserves, and partly by CABORA BASSA 282

Mozambique and Southern Africa contractors' finance arranged through Geneva banks. (Even in early 1974, Rhodesia's credit was good enough, in spite of sanctions, for the Government to raise large external loans in Europe at only slightly over world commercial rates, on normal state guarantees.) But at least two European suppliers, including Brown-Boveri, were forced by United Nations sanctions investigators to back out of the contract. By the end of 1974 Wankie's future - and that of Rhodesian industry - seemed insecure. The power crisis merged with the railway crisis as Frelimo threatened to cut the Mozambique outlets, and, running on worn-out capital equipment, unable to extend import substitution further, and bearing the third highest per capita internal debt in the world, Rhodesia's long-term economic capacity began to be undermined. Despite the confidence of Rhodesians, the guerilla attacks at the end of 1972 set a term to the future of white supremacy. The previous year had seen Zapu and Zanu incursions from Zambia reduced to a minimum, and the shock of this new assault was correspondingly great. Rhodesian security forces had not ignored the north- east border, with its mountainous escarpment, densely wooded and honeycombed with caves, falling towards the Mozambique plain; but they had imagined it guarded by their defence arrangements with the Portuguese. Under the provisions for unofficial co-operation arranged at regular meetings of the so-called Council of Three - attended, for example, in February 1971 in Salisbury by Assistant Commissioner Robinson, Rhodesian Special Branch, Major Silva Pais, head of the DGS in Mozambique, and General van den Bergh, head of South Africa's Bureau of State Security Rhodesian forces were permitted to extend the limits of 'hot pursuit' far into Mozambique territory. Helicopter strikes had been made along the area between the frontier and the Zambesi; foot patrols covered the roads running south, particularly near Zumbo; the Rhodesian Special Branch maintained liaison with Portuguese headquarters in Tete. But at some point after the Frelimo incursions into Tete in 1971, fearing that the Portuguese army would retreat into a handful of strongholds, the Rhodesians and South Africans pressed for a more specific agreement. Unofficial cooperation was replaced by a new agreement called Alcora - the Alcora Commission, composed 283 of military representatives as well as security police, had its permanent offices in South Africa - and Rhodesian troops effectively took over as a second line of defence in the threatened areas. Salisbury was not yet afraid of Portuguese withdrawal, and looked forward to establishing a position where it could sustain, if necessary, a Mozambique white settler movement on the lines of the Rhodesian Front, or dominate, with South African aid, a 'sensible' black regime. Rhodesians did not anticipate an alliance between Frelimo and the Zimbabwean movements. Zanu and Zapu's long fratricidal conflicts, the despair of the OAU and of President Kaunda, reinforced long-established conclusions about the inherent weakness of guerilla movements operating from outside their country of origin. Portuguese liberation movements had seemed no better: MPLA, FNLA, and Unita, wrangled for recognition by the OAU and, in spite of some cooperation through CONCP, Frelimo had never joined them in a common front - indeed, it had carefully stood aloof from doctrinaire disputes, fearing a revival of the internal clashes of 1969. But the Rhodesians' assertive policy in Tete, and their use of the border areas as a free-fire zone, forced Frelimo to consider anew the general strategic aims of cooperation advocated by the OAU and President Nyerere. Faced with a determined white army, whose ruthless 'search and destroy' missions contrasted sharply with the less incisive patrolling of the Portuguese, Frelimo turned to Zanu in order to stave off the Rhodesians and keep open its routes through the escarpment hills to Manica and Sofala. Zanu's and Zapu's earlier tactics of frontal assaults by large raiding bands had failed totally to meet the fire-power of the Rhodesian army, and the bombing and strafing capacity of the air force. By 1971 they numbered only 600 in Zambia (500 Zanu, 100 Zapu) compared with a total of 12-15,000 Frelimo. But, given safe conduct through Tete and launched into the tribal lands of the north-east around Mount Darwin and Centenary, and the forests of the Inyanga national park, Zanu were able to regroup and to stockpile arms, following Frelimo's own example, in the six months before December 1972. Whether Frelimo also instructed them in the campaign which followed is not clear: waves of attacks on white farms, CABORA BASSA 284

Mozambique and Southern Africa ambushes, and the mining of roads, all copied Frelimo's tactics; but the spontaneous combustion which occurred in Rhodesia revived instead memories of the terrible uprising of 1895 - the 'spirit war', now emulated by the 'prophets' of Inyanga.* Comparable with the Mau Mau, the new movement was engendered in the tribal lands, so that the security forces had to fight it inside Rhodesia. As a result, Frelimo's supply routes were left open, until their increasing numbers killed on or inside the border alerted the Rhodesians to the nature of the double manoeuvre. (In the twelve months, December 1972-December 1973, 168 were killed in the north-east: 75 Frelimo, 93 Zanu. The total had risen to 450 a year later.) By then it was too late: large areas of the north-east had become deeply subverted, and the brunt of the fighting was being carried not so much by Zanu members as by fresh, usually teenage, recruits. In spite of repression and the arming of every farmhouse, the movement had taken root and even threatened Salisbury with urban terrorism. Attempts at coercion by collective fines, the confiscation of cattle, and increases in the arbitrary power of district authorities to arrest and sentence to forced labour, alienated the population further. Shotguns were issued, but only for loyal chiefs to distribute, while the campaign of counter- terror raised protests even in Salisbury.' Strategic villages in the north-east bore little relationship to the experiments of the GPZ and frequently destroyed the pastoral life of the Shona. At the peak 60,000 were moved (against the advice of the South African Government) under conditions which were bitterly resented; and most of these villages became the seed-bed of further insurrection. Rhodesia's swift response suggested a deep uneasiness at discovering the guerilla's entry, in Ian Smith's words, 'by a door we didn't know was open'. The Government increased defence spending by 12 per cent, doubled the intake for national service, and raised a second regular battalion. Only new immigrants escaped the net; and great moral pressure was put on those who had already completed their service to * In 1895, the Shona and Matabele tribes, linked after years of hostility by a religious revival based on the holy places of Zimbabwe and the Matopos, rebelled under the leadership of priests of the Mwami cult. 285 volunteer again. To supervise the tightening up of internal security, Smith appointed a new minister, the Afrikaner, Wickus de Kock. Less predictably, and in what later seemed unreasonable haste, Smith announced the closure of the border with Zambia. By cutting Zambia's railway to the south, he hoped to enforce the expulsion of the guerillas; instead, Kaunda accepted the challenge and capitalized on Rhodesian folly. Aid flowed in to sustain Zambia through the crisis; her exports were re-routed, via Tanzania, and by the Benguella Railway to Lobito on the Atlantic. The closure failed even in its immediate object of sealing out Zanu, for the wild north-east border terrain could not be guarded effectively; and it inflicted on the already overstretched Rhodesian railways a loss of £ 12 million a year. Frantically casting round for ways to deflect the attacks, the Rhodesians patrolled even further inside Mozambique, securing the Umtali area by attacking Frelimo's routes in Vila Gouveia and Changara. They collaborated closely and secretly with the DGS and the flechas, rather than with the Portuguese army, for which Rhodesians evinced a growing contempt, and for a time in 1973 they harassed Frelimo so effectively that Zanu, in turn, suffered.2 While they failed to prevent the stockpiling of explosives, which were used to sabotage the Beira-Umtali Railway early in 1974, these manoeuvres suggested a new means of exercising influence in Mozambique, if and when the Portuguese army weakened; and they received a certain recognition when Frelimo was forced to transfer its attacks from that line, in February 1974, to the Malawi line at Inhaminga. The Smith Government had never, according to Sir Roy Welehsky, shown proper appreciation of what it owed to Portugal's willpower to continue the long struggle; when the attacks began, in November 1972, Smith visited Lisbon, and treated Caetano to a tactless harangue on the subject of joint defence preparations. Portuguese pride reasserted itself. Caetano delivered a well-publicized rebuke against 'some less experienced neighbours who do not conceal their fears, thus playing into the enemy's hands'.3 Lisbon was particularly angry at the implied advice to adopt reprisal techniques, which 286 CABORA BASSA

Mozambique and Southern Africa struck directly at the 'hearts and minds' campaign. But Rhodesian fears were real enough; and South Africa, still loyal to her, sent Defence Minister Botha and Admiral Biermann to Lisbon to mollify the Portuguese. Carefully judged leaks to the South African press hinted at a greater degree of support for Rhodesia. But South African goodwill suffered a mortal blow at the border closure, that 'watershed in South African politics'.4 Smith had failed to consult Vorster, suspecting perhaps (as Eden had the United States at the time of Suez) that the reply would have been a veto. By abruptly forcing Zambia further into the Central-East African communications system, South African long-term interests had been damaged, for the possibility of a diplomatic rapprochement had ranked high with the Foreign Ministry, in spite of surface currents of hostility. Smith should have seen a tacit rebuke in South Africa's re-routing of its own traffic with Zambia via Dar es Salaam. The South African press now began to veer away from doctrines of white solidarity, even suggesting that Rhodesia had become a liability. Disenchantment, and South Africa's own changing interests, led to harsher attitudes. As Welensky foresaw: 'South Africa will do nothing, unless there is a direct threat to her integrity'.5 Vorster and Muller had pressed Smith to settle with Britain in 1971; when Smith returned to Pretoria two years later for reassurance, it seems that he was given the blunt reply by the Minister of Foreign Affairs: 'Rhodesia cannot always be white.' The Frelimo-Zanu offensive exposed Rhodesia's fundamental insecurity. Smith's long negotiations with Bishop Muzarewa and the ANC - a manoeuvre to buy the support of moderate Africans - came too late, for when the Rhodesian offer was at last unveiled in June 1974 the ANC rejected it as inadequate. After an unparalleled effort, the Rhodesian security situation had improved: but with Zanu bidding against them, Portuguese authority ebbing, and Frelimo in the ascendant, the majority of ANC leaders saw no need to accept such modest, even marginal, concessions. Conscious that the vaunted 'Settler 74' campaign had brought in few of the predicted immigrants, and that in certain months in 1973 there had been a net drain of whites, Smith called for and won a 287 general election, without revealing whether it presaged concession or a last ditch defence. South Africa reacted differently, because her interests were not directly threatened. Pretoria had already accepted the economic implications of Portuguese withdrawal, and sought in Mozambique only stability and an attitude of mind no less flexible than that of the Zambian or Botswana Governments. If necessary, Escom could do without Cabora Bassa, just as the mines, at considerable cost, could survive without Shangaan contract labour; and the new Richard's Bay railway and port complex could by 1977 replace Lourenqo Marques. Security chiefs had already accepted the logic of an independent Mozambique. Angola was not yet in question; but Admiral Biermann acknowledged privately in 1972 that Mozambique was ceasing to act as a buffer state. Alcora had been a Rhodesian, not a South African, necessity; and while South Africa encouraged Portugal by every means, it was clear that the measures of 1973-4 - increases in the Defence Force, and an exercise in the Tshonga homeland (on the north-east Transvaal border with Mozambique) - did not take Rhodesia into consideration. If the Zulu, Xhosa, and other homelands were seriously to be considered as separate states, together with Ovamboland, then, in default of the Portuguese army, South Africa's strategic border would unquestionably contract to the Limpopo rather than the Zambesi. Rhodesia was not wholly excluded: South African police reinforcements actually increased during this period, because of fears of attack through the Lowveldt; but the old justification for their presence, that they were defending South African interests, had ceased to be true. By contracting South Africa's strategic frontier, Frelimo and Cabora Bassa had changed the face of southern Africa. But South African acquiescence depended on Mozambique's 'stability' - a favourite South African phrase - and non-alignment in the Cold War. The twin horrors of Chinese hordes and Russian naval dominance haunt all South African defence planners; the thesis that, in the context of the Sino-Soviet dispute, they are mutually exclusive, was not then widely accepted. The Russian threat was real enough, and a shared tenet in British and American naval policy-making in the South Atlantic and the Indian Ocean. This gave substance CABORA BASSA 288

Mozambique and Southern Africa to the Simonstown naval base agreement, and a respectable justification for buying Mirage fighters, Daphne-class submarines, and helicopters, from France, or for the request for corvettes from Portugal, to replace British supplies in 1971. The possibility of a Mozambique Government granting naval facilities at Nacala and Lourenqo Marques to the Soviet Navy had been foreshadowed in 1973 by the sudden pro-Russian alignment of Malagasy, formerly a docile recipient of French and South African aid. China's menace, however, had a more immediate and lurid impact, for it linked with the perennial nightmare of black revolt. A curious but popular propaganda film, Captain Caprivi, with an introduction by the former Foreign Minister, Eric Louw, depicted recently the stand, on a nameless but familiar border, of a group of heroic whites against black terrorists, sinuous and well-armed, led always by Chinese. Evaluation of the actual danger, even at diplomatic level, seems to have been retarded by persisting popular myths; as late as May 1972 Botha reaffirmed, in the most ultramontane terms, South Africa's duty to halt Communist penetration.6 Three stages in South African policy towards Mozambique can be observed. During the first, roughly 1968-71, military aid was offered in a variety of forms, and it seems that some pilots were seconded to the Portuguese air force, if only to gain experience of low-level attacks. This was the period of Vorster's aggressive claims for the legality of hot pursuit, and his declaration: 'We shall fight terrorism ... in any country where our interests are involved.'7 Overtly critical attitudes showed in the press: 'The Mozambique high command have often had as much trouble keeping tabs on its own units as it has finding and engaging the enemy .... Back in their bureaucratic glasshouses, most urban Mozambique whites seem to be content to be ignorant about the state of affairs up north - just as long as their side isn't actually losing .... 8 A time of questioning followed, of analysis of the likely future of Portuguese Africa, and of the real nature of the co-called 'band of stability'. Confined to the Nationalist Cabinet and its immediate advisers, it was characterized by a rapid increase in knowledge of the area, and a flexibility quite different from the response of the Rhodesian Government. 289

Because foreign affairs had never been the subject of wide debate in Parliament, or between the Nationalists and the United Party, the electorate remained uninformed and ignorant of change. The press and the South African Broadcasting Corporation, however, began to reflect an appreciation of the Portuguese: Die Burger spoke of 'a much misunderstood civilization bearing up under adverse conditions'; a view not unconnected with advice received from Pretoria. Admiral Biermann apparently asked it to play down the Smith-Caetano quarrel, and the press reflected a general sense that Portugal deserved all the moral support South Africans could give. This mood ended, roughly at the same time as the policy of dialogue with black Africa was seen to be dead, and was replaced by an urgent adaptation to the Mozambique crisis. Botha and Biermann had probably seen what was coming, even before General d'Arriaga came south to brief them, after his retirement in July 1973. Then, in November 1973, at the time of President Kaunda's attempted mediation between Frelimo and the Portuguese Government, Botha saw and apparently approved the terms of the Zambian proposal; by implication, South Africa accepted it as a means of furthering cooperation with Zambia, even at the risk of undermining Rhodesia. Meanwhile, the Foreign Ministry expanded its information services, sought to find out about Frelimo and its leaders, and began, very cautiously, to allow the public to see them, not as Communist terrorists, but as nationalists, fighting for liberty in their own country. Surprising though it was that the Portuguese should talk to black rebels, Vorster himself was now sitting down to talk to 'homeland' leaders such as Matanzima and Buthelezi; and the elections of 1974, which returned a solid Nationalist majority, could be seen as giving Vorster the strength to do what no other South African could, and grant the homelands full independence. Naturally, South Africa used its influence to delay Mozambique independence, and used its lobby in Lisbon first against Spinola, at the time of publication of Portugal and the Future, and then against the Armed Forces Movement. But a rearguard action did not preclude accommodation, and South African diplomats appeared satisfied at the thought of Portugal creating another Malawi or Lesotho in Mozambique. CABORA BASSA 290

MAozambique and Southern Africa At no time did South Africa suggest or welcome the idea of white separatism, on the Rhodesian pattern; and the interest shown in alternatives was restricted to the multi-racial solutions propounded either by d'Arriaga - who retired again to Johannesburg after his abortive 'coup' in December 1973 (p 316) - or Jorge Jardim. Concern might have been greater if there had then been any question of immediate independence for Angola, recipient of far greater South African investment. After the Portuguese revolution, when the preliminagy talks about a cease-fire failed, and as unrest spread across the province, the South African Government showed itself meticulous in its dealings with Mozambique. Keen only for a peaceful solution, Vorster rebutted all requests for intervention, and, while suppressing pro-Frelimo demonstrations at home, allowed free rein to the 'legitimizing process'.9 Such a transition was only possible within the South African context of a government and civil service wholly dominated by the Nationalist elite, with considerable control over the Afrikaans press, and a high disregard for opposition from the English-speaking sector. The contrast, and tension, between the policies of this small group and the views of the electorate - the archetypal Afrikaner of the platteland - explains some of the Government's antagonism towards its critics and its obsessive secrecy. Large questions also stayed unanswered; at the least, reliance on disintegrating Portuguese authority to bind a future independent state to genuine non-alignment, contained an element of gambling. No Western power owed such a hostage to fortune. The British Government's concern was almost wholly with Rhodesia. For the Conservative Government, Lourenqo Marques had served as a useful listening post, in lieu of a mission in Salisbury, to advise on the correct balance between black and white Africa, or, as one official put it, on how to maintain trade and investment in South Africa - worth £ 1300m in 1973, one-tenth of Britain's overseas investment - without too greatly offending the susceptibilities of Nigeria, the fastest growing African market. This leisurely process of waiting for the Rhodesian economy to run downt without exercising too t For further amplification see Notes at end of chapter. 291 great a pressure, altered a little, in response to the appreciating value of Nigerian oil after the 1973 autumn crisis, and was replaced, under the Wilson Government, by a renewed attempt to coerce Rhodesia into a settlement using Mozambique independence as a lever. Important as this was, not only for the future of Britain's severest African problem, but for continued good relations with China and opposition to Russian dominion over the Indian Ocean, it scarcely touched the future of Mozambique. French policy was directed, fairly straightforwardly, to fulfilling de Gaulle's long- term aim of preventing southern Africa deteriorating into another Congo. The search for influence and commercial penetration, beyond Francophone West Africa and Madagascar, continues in any developing country and there is no need to derive conspiracy theories from the activities of Monsieur Foccart, once- powerful head of the Bureau des Affaires Africaines et Malgaches, to explain French persistence and success in a trade offensive backed by aid, civil or military, in Malawi, Rhodesia, and South Africa. The same phenomena can be found in India and South-East Asia, as successive French Governments since the early 1960s have sought export-led trade backed by favourable credits. Economic penetration of Mozambique awaited the expected lowering of tariff barriers, contingent on Portugal's association with the EEC, but in Portugal itself rested on the solid advantage of arms supplyt combined with political favour. Support for Portugal's EEC application from Maurice Schumann and Giscard d'Estaing in 1972-3 related to expectations in the Portuguese-speaking world, especially in Brazil.tt D'Estaing's crusade in March 1974 to offset the French balance of payments deficit was directed especially to parts of Africa 'where our exports are inadequate'.0 Insofar as the Quai d'Orsay considered Mozambique in isolation, French interests were best served by rapid progress towards autonomy and stable development as a buffer state between black and white Africa. The continued existence of Rhodesia, and security of the debts accumulated in the evasion of UN sanctions, concerned For further amplification see Note\, at end of chapter. + For further amplification see Notes at end of chapter. CABORA BASSA 292

Mozambique and Southern Africa 293 both French and German commercial interests; but the German Government considered Mozambique only within the context of a general trade policy for southern Africa, in which, without much state direction, the emphasis was laid on export promotion. Early, jejune essays in dealing with black socialist countries like Tanzania had reinforced a natural caution about Third World diplomacy. But overseas aid bulked larger after being incorporated in a new ministry in January 1973. Hermes, the export credit agency, took a more aggressive part in promoting industry abroad, especially in South Africat and the German Government even began to criticize the monopolistic nature of French aid in West Africa. These pressures extended to Portugal, and contributed to the feeling in Lisbon that all Portugal's foreign investors, Britain, South Africa, France, and Germany, had a vested interest in a peaceful transition for the former empire. The United States shared these concerns, both commercially and strategically. During the latter half of Nixon's presidency, American policy in Africa was concentrated on building up sound relationships with black countries, in order to offset the impact commercial relations with South Africa had on black America. Dissent at home about wages paid by American companies, especially in the Polaroid case, induced a form of appeasement; aid, which had fallen substantially from a high peak in the Kennedy years, was partly restored, as a means of financing the economic independence of countries such as Botswana and Lesotho. These gestures accorded uneasily with the involvement of Gulf Oil in Angola, and the extent of official equivocation could be seen, for example, in the gyrations of the Export-Import Bank over Morrison-Knudsen's application for export credits for Cabora Bassa - credits which were made freely available for the politically more respectable Kasson barrage on the Ivory Coast. In practice, the major share of aid went in tied form to countries with strategic implications - Nigeria, Morocco, Kenya, Zaire, and, in direct competition with Russia or China, to Guinea and Tanzania. Portugal's primary importance to the United States is as a t For further amplification see Notes at end of chapter. member of Nato; and Guinea-Bissau, Angola, and Mozambique, remain key points in South Atlantic and Indian Ocean strategy. Anxiety about possible Russian naval arrangements with Malagasy and Guinea, as well as the experience of the Nigerian civil war, lent point to plans by Britain and the United States to extend Nato's sphere into the South Atlantic, at least as far as the Portuguese islands, Cape Verde and Sdo Tomd. In 1972, President Nixon agreed to pay the vast sum of $430 million - delivered as Eximbank loan finance for projects delayed by heavy expenditure on the colonial wars - to extend America's lease of the Lajes base in the Azores until 1974, and the base was used to refuel aircraft sent to reinforce Israel during the 1973 war. Lisnave's plans to build dock facilities for trans-continental shipping at Nacala were given the same substantial encouragement which had already been shown to the gigantic new developments on the Portuguese coast, at Setenave and Sines. A major share of American and European trade uses the Cape route: the disputes over development of the base at Diego Garcia in the Indian Ocean relate to the same strategic thinking. The Russian naval threat, and South Africa's inability to counter it alone, bulked large in publications of the Institute of Strategic Studies in Washington, to say nothing of supportive lobbying by the American South African Foundation. Tangible support, beyond the Azores agreement, derived from Nato itself. As early as 1970, sections of the Conservative Party in Britain began to point out the significance of the Cape route, as part of a dual campaign to increase arms sales to South Africa and to invoke the ancient alliance with Portugal in the South Atlantic. Two years later, a sub-committee of the North Atlantic Military Committee of Nato began to examine the 'Soviet maritime threat', and its report was adopted by the Nato assembly in October 1972" despite some opposition from Germany and the Netherlands. Nato Atlantic Command (SACLANT) then prepared a detailed study for contingency planning against a possible Soviet interruption of oil or other supplies; and this gave rise to a very close monitoring of Portugal's overseas policy, which continued into the months after the Armed Forces revolution. Alarmist rumours about American attempts to intervene CABORA BASSA 294

Mozambique and Southern Africa circulated in Lisbon during and after the revolution. The emergence, in the Junta of National Salvation, of Generals Spinola and Costa Gomes, who had served Nato loyally, was evidently welcomed. In the final stages of the scramble for the bodies of Mozambique and Angola, Portugal's freedom of action was seriously limited, and constrained by that framework of Western strategy which defenders of her role in Africa had always claimed existed. On the other side of the hill, black African countries sought a similar containment, though within a narrower range of interests, and with less rigidity than was suggested by their public statements and the positions taken at OAU summit conferences. Only three countries exercised. much influence, for the OAU remained virtually powerless to direct Frelimo, and Nigeria and Ethiopia, however intent on decolonization, lacked the means to intervene. Portuguese links with the Malawi Congress Party atid Dr Banda date from a time even before independence; based on gratitude for financial assistance, and later on the commercial assets of Champalimaud and da Cunha in banking and mineral exploration, these were reinforced by extension of the Nacala Railway, thus giving Malawi two outlets to the sea, and by promises of a new road or railway to link Tete with Blantyre. Diplomatic contacts operated on two levels; officially, through the Portuguese ambassador, and unofficially via Jardim, a long-time friend of Dr Banda, originally Salazar's emissary and then personal representative of Caetano, and Malawi's honorary consul in Lisbon. Cooperation reached its peak in the period when Banda took on the advocacy of dialogue with the white South: his visit to South Africa was followed by a tour of Mozambique, and at Ilha da Moqambique and Cabora Bassa, the alpha and omega of Portuguese imperial history, he admitted a certain value in the force of Portugal's historical claims. "2 But Malawi was vulnerable to external pressure. A working arrangement, quite unacknowledged but effective, regulated the passage of Frelimo up to 1971. Under this guerillas could pass through Malawi so long as they did not carry arms. To keep the balance, Portugal was given tacit rights of pursuit 295 along the border; and two armed launches, staffed by the Portuguese navy, patrolled Lake Malawi and prevented Frelimo occupation of the disputed islands near the Tanzanian border. But this understanding, itself a product of the dilemma of a poor country surrounded by wealthier neighbours, was weakened by the Frelimo campaign in Tete. Banda's delicate trimming, his practical if unpopular links with Rhodesia, and the growing through-trade with Zambia, were all threatened; and the absolution given so freely by the OAU to Zambia's trading with the white South was persistently withheld.3 The greatest danger from Frelimo lay not in armed entry, but in association with Malawian exiles: the faction of Harry Chipembere, expelled after challenging Banda's authority in the Government in 1964, and Chiume, who launched an attack on Malawi three years later. As Banda grew older, with the succession question unresolved, the exiles, or 'Capricorns', in Lusaka and Dar es Salaam, grew in importance; and Frelimo, hoping for bases first of all in the south-west (Tete frontier) and then Mlanje (to raid into Zambesia) profited from Malawi's fears and isolation in the period after the dialogue policy decayed. An apparent breach of the entente with Portugal over a succession of frontier incidents in 1971-2t showed that Banda was less free to respond: the unofficial links survived, but the ambassador was withdrawn. Portugal had very little leverage after Malawi declined the offer of Cabora Bassa power, and the 'goodwill missions' which arrived from Zambia and Tanzania indicated a shift in Malawian internal politics, as well as the existence of party tendencies (p. 275) to favour a middle African grouping. Banda kept his options open as far as possible, and it seems that Kaunda and Nyerere, careful not to force him into an impossible position, withdrew their protection from the leading Capricorns. But Frelimo, unaware of, or rejecting, the need to compromise, used blackmail tactics in succeeding months, attacking the Beira-Blantyre Railway, and sabotage of the line at Inhamingha cut Malawi's export-import trade most seriously in the first half of 1974. Faced with threats to the economy, to internal security from possible pro-Frelimo movements among two of the three principal tribes t For further amplification see Notes at end of chapter. CABORA BASSA 296

Mozambique and Southern Africa in Malawi, the Angoni and Nyanja, and to future links with the sea if Mozambique collapsed in anarchy, Banda looked desperately for means to mediate between Frelimo and Portugal. Using the well-tried line of unofficial communication, he joined with Kaunda in an appeal to Caetano at the end of 1973, and renewed it to the Armed Forces Movement after the revolution. More than his coadjutor, he had long experience of the Portuguese, and was uniquely fitted for the role of interpreter of one side to the other. Zambia owed few favours to the Portuguese, and had experienced mixed results from dependence on the Beira and Benguella Railways; it appeared that the Portuguese alternately cajoled and bludgeoned to gain their ends. Relations in the period after Kaunda's chairmanship of the OAU, and during the campaign against Cabora Bassa, touched a nadir in 1970-1, with the blockade of Zambian supplies on one side, and the supply of a British-made Rapier ground-to-air missile system (required as much to counter Portuguese infringement of the border as a hypothetical Rhodesian or South African air-strike) on the other. A bond of interest survived. Kaunda never denied the theoretical importance of the Portuguese claim to have created a multi-racial society, and in public distinguished between the evils of colonialism and the far greater scourge of apartheid. The difference can be seen in the Lusaka Manifesto: 'The basic problem is not racialism, but the pretence that Portugal exists in Africa."4 Diplomatic interchanges, either at Blantyre or at the United Nations, resolved the blockade, and reduced the tension arising from a series of border infringements during the Tete campaign. The trade agreement between Zambia and Malawi in May 1971 brought Zambian traffic into both Beira and Nacala; and when Rhodesia closed the border in December 1972, Portuguese cooperation on the Benguella Railway, and on the building of a new cut-off to by-pass the mountain section, enabled Zambia to re-route the bulk of the copper trade via Lobito, thus avoiding congestion on the Tanzam rail-head in Dar es Salaam and defeating Smith's attempted blackmail.t Cabora Bassa had ceased to be t For further amplification see Notes at the end of chapter. 297 a contentious issue; by 1973 it was clear in Lusaka that it would be finished, and that it would serve the needs of the future state of Mozambique. But Kaunda walked a very narrow path in foreign policy. He had deliberately tied his country to a socialist foreign policy, with clear, trenchantly declared aims, in conformity with OAU doctrines; and he was, in a real sense, heir to Nkrumah as spokesman for the nations of middle Africa. Zambian economic dependence on the mining industry and trade with the white South, might, therefore, be condoned by the OAU states, but the Government was not in a position, internally or externally, to make a virtue out of such necessary co-operation. Kaunda could not easily respond, for example, either to Vorster's secret overtures, or to his impatient reaction to trouble in the Caprivi Strip. Zambia's need to cooperate in the future with Mozambique - partly because of the importance to the copper trade of a future Zambesi waterway, and partly because, once Kariba North had been completed, she would have a surplus of energy which could not, for political reasons, be channelled through the CAPC to prop up Rhodesia's continued economic independence - could not be announced in public, any more than could the reasons of state which had forced her to expel Unitd and restrain MPLA from attacks on her lifeline through Angola. Contacts with Portugal took the same discreet route as those of Banda: Mark Chona, Kaunda's political adviser, journeyed when necessary to Lisbon or discussed with Jardim in Malawi. In this secret diplomacy, more weight was put on the connection than it would actually bear, for, as the principals discovered later, Caetano reserved the right not to be bound by offers when these conflicted with interest groups at home. The fall of Milton Obote and the rise of Idi Amin acted as catalysts of the transition from presidential to party rule in Tanzania and Zambia, and raised acute fears of an extension of caudillismo elsewhere. Mozambique appeared only too likely a candidate: Spinola's deduction, in Portugal and the Future, that the war had become a way of life, applied to Frelimo as well as to the Portuguese army; and it seemed possible that Portuguese rule could be replaced either by the Mozambique army or by white settlers, with Rhodesian and CABORA BASSA 298

Mozambique and Southern Africa South African military backing. In any of these cases, greater pressure might be put on Zambia to expel Frelimo or curtail its activities. Progress towards autonomy and reforms of the Legislative Assembly, although slow, did appear to offer a way out: if Frelimo could be persuaded to define its war aims, and to participate in the sort of political activity inside Mozambique advocated by some white democrats, the danger might be avoided; and if Zambia could act as mediator, she would gain merit at home and elsewhere in Africa. Beyond that lay the greatest prize - the undermining of Rhodesia's will and capacity to carry on. Earlier attempts at mediation had failed, either because Portugal was hostile (as with Oscar Kambona's overture in 1964) or because they were made in public. One Zambian official regretted the lack of debate in Lisbon, which had precluded public debate on the implications of the Lusaka Manifesto. But in 1973, Jardim was able to give assurances that his Government found Kaunda's recent speeches 'more satisfactory'; by the summer of 1973 Kaunda suspected, not unreasonably, that Caetano was ready for his initiative. Through Chona he approached Dr Banda, and then, with Tanzanian help, sounded out Samora Machel as to what terms Frelimo would agree to for the holding of a conference. A document was drafted, approved in general terms by Frelimo, and transmitted to Lisbon after discussions with Jardim on the form most likely to appeal to Caetano. Although it was not binding on any of the parties, and was not a precise definition of war aims, it accorded closely with the terms finally announced to Portugal by Frelimo in July 1974; and the joint letter was couched in phrases remarkably sympathetic to Portugal's long-term aspirations. The document proposed that in return for three critical concessions - a guarantee of independence, recognition of Frelimo as the sole representative institution, and the transfer of power - all other questions should be negotiable. But the assurance that Portugal's requirements - on behalf of the white population, and for the protection of her investments, and cultural, political, and strategic interests in the future - would be met, was made in Kaunda's name rather than Machel's; and the document, in 299 conformity with the Lusaka Manifesto, carried the provision that Mozambique should not be linked with Rhodesia or South Africa.* However clearly it was stated that Frelimo was nationalist not Communist, however rosy the Lusitanian future sketched out, the offer failed to allow for Portugal's bias in favour of economic links with South Africa, the defence of southern Africa, and the Nato relationship; and, in its open statements about the danger of militarization, it exposed the inherent weakness of the mediators, themselves. Kaunda gave an implicit indication of the course of events in a speech in December, suggesting that Portugal should be given time to resolve problems in her own way. But instead of presenting the document to Caetano when he returned to Lisbon in early February, Jardim delayed, fearing the effect of right- wing pressure in the aftermath of General Spinola's Portugal and the Future, and waited, as he had done in the days of Salazar, for a change in the political climate. He waited too long, until 17 April, only eight days before the revolution. Ironically, Caetano appears to have been moving secretly in the same direction - but unaware of South African approval of Kaunda's scheme. Would Frelimo have followed the lead at that stage? The offer in no way bound them, but, beyond the non-negotiable demands, it did indicate a considerable degree of flexibility and an understanding of the Portuguese predicament which was manifest later in the course of final negotiations. On the other hand, Frelimo owed little to Zambian authority and relied, if on any outside support, rather on Tanzania. Nyerere's part in the offer is unknown, although Tanu appear to have indicated assent. A profound question lay behind Frelimo's demands for sole recognition, and sole accession to power. Jardim specifically upheld the rights of the Mozambique majority: Frelimo had to participate with other groups in power, and rely on the democratic process after independence. Portugal laid down the same condition: Caetano's offer to Frelimo, in March 1974, specified that it must first cease fighting and then come inside the fold; and Spinola's pro* Frelimo were, nevertheless, prepared to countenance the existence of Portuguese military bases in Mozambique - a concession absent from the Lusaka Agreement a year later. CABORA BASSA 300

Mozam bique and Southern Africa gramme also envisaged a referendum before any major constitutional change. At issue was legitimacy itself: Frelimo needed historic justification for the war, acknowledgment that by its revolt and struggle it had earned the sole right to represent the nation - indeed, that by its struggle it had created the nation - and, in its accession over the heads of all others, recognition that the authority of Portugal, like some apostolic succession, had been conveyed directly, with no intervening stage of provincial self-government. In the end, this happened; but it was not necessarily in the interests of Mozambique's black or white neighbours that it should; and it was wholly contrary to the wishes of those Mozambicans, like Jardim, who hoped for a solution growing out of the territory itself, not a bargain between two, literally, 'external' institutions. 1'anzania's role is harder to discern. Frelimo took a considerable risk, then and in 1974, by insisting on its high price of non-negotiable demands. Portugal might have made a settlement with the new political groups in Mozambique, and sought guarantees for it from the white South; or Mozambique might have lapsed into faction and civil war. In either case, unless the Portuguese army had weakened, Frelimo would have been excluded, and left to continue its guerilla war, but from a substantially weakened base were other African states to support the arrangement. (A contemporary comparison could be made with the position of Zanu and Zapu in the then expected event of agreement between Smith and the ANC in Rhodesia). Tanzania had her reasons for not interfering with Frelimo or seeking to mediate between it and the Portuguese. None of Portugal's affinities with Malawi, nor the economic links with Zambia, obtained; instead, Nyerere had taken up an inflexible position on colonialism. Tanzania was not vulnerable to Portuguese coercion or sabotage, short of an attack on the Tanzam Railway (which would have harmed Zambia more) or a major casus belli such as a strike at Frelimo camps across the border. t Within the country, Frelimo bulked too large to be responsive to its hosts as, for example, the Rhodesian t For further amplification see Notes at end of chapter. 301 movements were to Kaunda; armed and equipped on the scale of 1973, Frelimo was immune to threats, a condition which may have contributed to the bearing of members of its Central Committee, who more and more resembled a government- in-exile. At meetings of the OAU Machel conducted himself with the dignity and presence of a head of state. Nyerere seems to have seen Frelimo as the precursor of a new alignment of states in middle Africa. As the East African Community, profoundly shaken by Amin's rise to power and by the economic impact of Tanzania's trading agreement with China, began to disintegrate, a closer relationship with Zambia (beyond the services provided by the railway) and Malawi appeared desirable. These links had been, and were likely to be, fragile, lacking a common political and ideological understanding. If, however, Frelimo came to power, schooled in the experience of Tanzanian socialism and intellectually and emotionally akin to its long-time host, Mozambique's new political institutions might be conditioned, prepared for that wide federation which Nyerere was already advocating in public addresses, during his tour of Australia, early in 1974. Nyerere had no wish for Tanzania to become a mere bridge-head for Chinese penetration in southern Africa, and the Chinese in Dar es Salaam, at his request, appear to have taken virtually no part in the protracted negotiations. All along, their declared policy had been to offer aid and instruction to Frelimo, and not to guide its strategic policy. How far this accorded with practice is hard to assess: Chinese instructors, for example, did not train commandos to attack Cabora Bassa in the days when an assault would have been feasible, possibly because of the retaliation which this might have drawn on Tanzania. In any case, the analysis of China's local policy in Africa does not easily lead to generalizations, which can be better inferred from recent and fairly explicit statements about China's relationship with the Third World. The crude essays in political control in Africa in the 1960s have, clearly, been forgotten. China now claims to be a member of the Third World rather than its natural leader. In the wake of the Portuguese revolution, Foreign Minister Chi Peng-Fei announced to a delegation of African leaders that 'In our future common struggle, we will continue to give each other encoura- CABORA BASSA 302

Mozambique and Southern Africa 303 gement and support. The African people can rest assured that whether in wind or rain, and no matter what happens, the Chinese people will always be your reliable friend.' 5 This fits closely with the general diagnosis that the 'Socialist camp' no longer exists; that in the new universal order, the American and Russian superpowers dominate the stage, and that only the Third World contains the power of revolutionary change, while other developed countries wait uneasily to decide with whom their interests coincide.6 Africa is not an area of immediate strategic significance, except in the context of the Sino-Soviet confrontation;t hence the particular significance of the successive delegations of African leaders in China, including Kaunda and Nyerere, in the first half of 1974. China's short- term policies in various African states have had to be harmonized with the overriding importance of rapidly evolving long-term policy, while preventing the sort of local tactical errors which occurred in the 1960s. It is not surprising therefore that China had no wish to exercise detailed control over liberation movements, other than to maintain existing commitments in aid and weapons, any more than her Government should wish to undermine the friendly and undisturbed understanding arising out of the Tanzam Railway construction. Her general hostility to South Africa has not changed: Zanu receives particular support because of its link with Frelimo (and Chinese advisers may have helped bring about their cooperation); but towards Portugal, attitudes remained ambiguous. Portugal was to gain no gratitude for her vote to admit China to the UN; yet the colony of Macau, like Hong Kong, continued to exist on Chinese sufferance. If China exerted influence at all in Tanzania in this period, it was in conjunction as far as could be guessed - with the policies of Western developed countries, especially Great Britain; and against the Soviet Union, within limits dictated by the desire not to impair the reputation so carefully built up among African leaders. China's effort in Africa has been altruistic as far as costs and control of aid is concerned. The Tanzam Railway has been a heavy drain on scarce foreign exchange and materials, ft like the projects in Congo-Brazzaville, Sudan, and Somalia. t For further amplification see Notes at end of chapter. tt For further amplification see Notes at end of chapter.

304 CABORA BASSA China has not yet sought leverage at political or military level in Tanzania, despite the dependence of the armed forces on Chinese sources of supply and training; and some African commentators see recent trends as a form of missionary activity, in which China seeks the liberation of Africa in the way that the churches once looked for converts, in order to demonstrate the virtues of a superior mode of living. 'The behaviour of Chinese engineers and workmen, their style of project managment, the labour intensive contracts, are all suited to the local situation, and where tensions arise in Tanzania, they tend to be caused by the rigidities of the trade agreement - such as the building of a shoe factory which put local craftsmen out of business - rather than the contract itself. The response of Africans to the experience of working with Chinese in Tanzania and Zambia has been almost uniformly enthusiastic, in sharp contrast to their residual memories of the Russians; more than this, China probably does not, at the moment, require. Later, a quid pro quo, such as mineral concessions, may be asked for; or, as the head of one African parastatal company suggested, continuing influence may lie in maintenance contracts; but, so far, experience sugges ts that it has actually been difficult to persuade Chinese technicians to remain beyond the strict limits of their contracts. For many reasons, therefore, it suited Chinese policy not to engage closely in the last stages of Mozambique's passage to independence, refuting South African commentators who had earlier forecast a 'conspiracy to dominate black Africa and provoke a final cataclysm'.'8 China was perhaps keener to see that Russia did not benefit, than to take temporary advantage of a tide running strongly in her favour. Russian influence has not by any means disappeared, and was reinforced after the Portuguese revolution by the inclusion of Communist Party leaders in the civil Government in Lisbon. At least one element in the Mozambique settlement depended on Alvaro Cunhal's influence, and on that of the pro-Russian Frelimo Foreign Secretary, Marcellino dos Santos. But the composition of the final Frelimo delegation, and of the provisional Government formed in September 1974, suggests that Russia's political assets had diminished in value, thus reinforcing the conclusion of observers in Dar es Salaam,

Mozambique and Southern Africa 305 during the hectic three months after 25 April, that simplistic theories about great- power leverage needed to be reviewed in the light of knowledge of the actual mechanisms involved. Frelimo was no-one's puppet, and one long-term effect of Tanzanian hospitality had been to free the movement from some of the constraints evident among liberation movements elsewhere. At the end of 1973 it seemed to the various participants that the Portuguese will to continue to rule in Mozambique (though not in Angola) had seriously weakened, and that a rapid transfer of power might be desirable in their own interests, to prevent anarchy or civil war. The white and black states on either side of the Zambesi, a gallery of participant observers, found a curious common ground. But they lacked either the machinery or the power to enforce their interests; and while their influences helped to shape and even determine the manner of the outcome of the struggle for power in Mozambique, its substance depended on the conflicts inside Portugal itself.

Notes and references Chapter 12 Mozambique and Southern Africa 1 See Rhodesia Herald, 5 February 1974. 2 cf. Mozambique Liberation, No. 53, December 1973, p. 5. 3 Diario das Noticias, 14 November 1972. 4 A. Wilkinson, Adelphi Paper No. 100 (Institute of Strategic Studies, 1973). 5 Interview with the author, November 1971. 6 Die Transvaler, 5 May 1972. But see also SAIIA, Newsletter, 1973-4 passim, and John Barrett, Bulletin of Africa Institute, No. 10, Nov/Dec. 1972, for a more scholarly approach to Chinese policy in Africa. 7 South African Hansard, 15 September 1970. 8 Rand Daily Mail, 17 October 1970. 9 See the carefully moderate account of post-revolutionary developments in SAIIA, 6, No. 2, June 1974: 'we should at all costs avoid the tendency, not unknown in South Africa, to equate black government with instability.' ± Note page 291. One story, dating from the first months of UDI, raises the question of how serious even the Labour Government was to grind Rhodesia down. Shortly before UDI, the London banknote printers, Bradbury Wilkinson, had received an order for a new issue of Rhodesian dollars. These were confiscated atUDI:but a new set was printed in West Germany. These too were confiscated, while in transit to Salisbury, and the printers reimbursed. The British were now in possession of notes to the value of twice the currency in circulation in Rhodesia; yet they resisted the temptation to use them to swamp the country, on grounds, it appears, of moral delicacy. (cf. also the arguments adduced by Robert Good, UDI (London 1973), on what would have happened had the Labour Government carried out its 1965 contingency plan.) t Note page 292. Generally, Corvettes, scout and armoured cars, Puma and Alouette helicopters, supplied within Nato standards, but on the purely commercial terms regulated by the Delegue Ministeriale pour l'Armements. ft Note page 292. The $248m Ilha Soltera barrage, for the hydro-electric supply to Sao Paulo, is being constructed, with exceptionally favourable export credits, by a group similar to the French contingent at Cabora Bassa: CGEE, Alsthom, and CCI, in association with Sofrelec, the French Government-sponsored consultants. 10 Le Monde, 20 March 1974. t Note page 293. German banks provided substantial loans to the South African Government in the early 1970s, and Helmut Haensgen, Chairman of the Dresdner Bank, stated in 1971: 'South Africa is still considered a stable shelter for capital seeking a safe and rewarding investment.' 11 Document P119 MC (72)5 of Nato Military Committee. See also Africasia, 20 July 1970, and evidence presented to the United Nations Decolonization Committee. 12 See Noticias de Beira (Jardim's own newspaper), 1 October 1971. 13 As early as 1965, Nyerere threatened to withdraw Tanzania from the Liberation Committee if Malawi, a fellow OAU member were appointed, because of her compromises with Portugal (Africa Research Bureau, 2, No. 10, 1965). t Note page 296. Disputes arose over refugees from Tete, pouring into Malawi in the

Mozambique and Southern Africa 307 wake of reprisals in 1971; Frelimo attacks on the Nacala Railway; and the Portuguese accusations that Malawi harboured Frelimo camps. Finally, several Malawians were killed in a border clash in May 1973; there is evidence of Rhodesian pressure also to prevent Malawi relapsing into the orbit of Zambia and Tanzania. 14 D.G. Anglin, 'Zambia and Portugal', Essays on Zambia's Foreign Policy. f Note page 297. Zambian exports via Lobito 600,000 tons Maximum tonnage Dar es Salaam 480,000 1974-5 Nacala 240,000 Port Francqui 28,000 (road/river) Mombasa 264,000 Europe (air) 24,000 f Note page 301. During 1972, the exiled Foreign Minister, Oscar Kambona, took upaup residence in Lisbon; but there does not seem to have been any deliberate Portuguese plot to use his opposition to Nyerere other than for publicity purposes. 15 Hsinhua NXews Bulletin, 27 May 1974, p. 19. 16 Vice-President Teng Hsiao- Ping, speech at the UN, 10 April 1974. t Note page 303. See for example (Hsinhua News Bulletin, 27 May 1974, p. 23) the attack on the 'Kremlin clique' which 'hatches conspiracies, conducts behind the scene political activities, provides so-called 'aid', concludes all kinds of economic agreements, and endlessly preaches pacifism in order to conquer the African peoples and compel them to abandon the liberation struggle.' tf Note page 303. The capital cost, totalling £186million, is interest free and repayable from 1986, over thirty years. Only the local costs, roughly £40million in the year, 1970-75, have to be met out of the trade agreement. The supply of steel has, hovever, been eased by buying, in large quantities, sections of track from South Wales, in the wake of British Railways rationalization - a pleasant anachronism for a nation that in the nineteenth century built most of the colonial railways in the world. 17 See A. Ogunsamwo, China's Policy in Africa, 1958-71 (Cambridge 1974). 18 W.A.C. Adie, 'Chinese Foreign Policy', SAIIA, February 1973.

Chapter 13 The Portuguese Crisis The course of the Portuguese revolution, which began on 25 April 1974, is too recent and too well-known to retell in detail, but the relevance to its origins of the overseas problem, particularly of Mozambique, should be emphasized. Like all revolutions, as opposed to coups d'6tat, it was accretive; but at all its stages, it was sought as a remedy for the intolerable affliction of Portugal's overseas commitments. The army revolt grew out of the impossible nature of the tasks set by the politicians, and drew strength from grievances accumulated during the thirteen years of colonial wars. Events in Mozambique provided the climax, an apparent proof that the Caetano solution simply did not exist. The role assumed by the Armed Forces Movement introduced a wholly new dimension to Portuguese political life, and ensured the triumph of Frelimo - apart from the FNLA in Algeria, and Frelimo's sister organization PAIGC in Guinea-Bissau, the only independence movement in Africa to win sole and undisputed power by war. For many years before 1974, overseas affairs had been ignored as a subject of political debate to a surprising degree, even taking into account the strict censorship of the period. The Liberal opposition - composed of men like Cunha Leal, who in 1946 addressed an open letter to Salazar, asking for the institution of a United States of Portugal with its capital in Luanda, and Paiva Couceiro, who had raised the issue of overseas independence after the 1939-45 war - continued to campaign in the 1950s, until its most able advocate, Henrique Galvdo, the scourge of slavery in Angola, was exiled to Brazil.

The Portuguese crisis But the revolts of the 1960s divided them as it had the Delgadists in Mozambique; only a handful of students at Coimbra University actually came out in support of the black liberation fronts. Most liberals supported slow progress towards some form of Commonwealth, and they divided yet again after the 1965 and 1968 elections, the parliamentarians like Sa Carneiro and Miller Guerra choosing to work within the system, while others, seeing no future in being further emasculated by Caetano's skilful electoral manoeuvres either fell in with the statutory scheme for economic autonomy or concentrated on swinging Portugal towards the EEC. Meanwhile, the Socialists suffered greater fragmentation. Some, like Raoul Rego or Salgado Zenha, continued under constant persecution to criticize the regime; Mario Soares eventually chose exile in Paris. Their conference at Aveiro in 1973 showed that the movement was not dead, but had become, in Franco Nogueira's phrase, 'a vast geography of uncertainty'; like the Liberals, it suffered from ignorance about colonial affairs, the product of censorship and a certain lack of interest. Further left, the Communist Party devoted most of its energy to maintaining its wide clandestine organization, despite years of attempted suppression still the most effective political network in Portugal. It weakened, however, with Maoist and Trotskyite deviations in the late sixties, as younger radicals revolted against an aging and orthodox leadership - and its activity was severely curtailed by the DGS after a wave of bomb attacks had disrupted the Nato conference in Lisbon in 1971. The revolutionary left, by concentrating its forces against the Portuguese dictatorship, also failed to evolve a coherent anti-colonial policy, and embarked on purely destructive direct action, robbing banks and detonating bombs with impunity even inside the Ministry of Defence. But in the two years after 1971, the extremist ARA was broken, piecemeal, by the DGS, and Palme Inacio, leader of LUAR, arrested. Some liaison had been preserved, via the Frelimo office in Algiers, with liberation movements, but the destruction of the coastal steamer Angoche off the coast of Mozambique in 1971 is the only evidence of concerted action. Generally, the Portuguese opposition, of whatever per- 309 CABORA BASSA 310 suasion, regarded the overseas territories as a less urgent matter than that of improving the conditions of life and liberty inside Portugal; and the blanket of censorship encouraged an ill-informed public to rely on the optimistic and frequently inaccurate accounts of the war and colonial development put out by official sources. In any case, immediate independence appeared to involve abandonment of all that Portugal had attempted to create. Although it is impossible to arrive at an accurate estimate, opinion among the politically literate in 1973 seems generally to have condoned Caetano's policy, without either enthusiasm or much faith in the future. Caetano's freedom to move towards autonomy overseas depended strictly on his capacity to balance the claims and interests of the major groups concerned: those politicians, usually simply aggregated as 'the right', who had inherited the style and traditions of Salazar; the great bankers and industrialists; and the chiefs of the armed forces. For five years after he came to power Caetano rode these three horses with remarkable skill; but they then chose different paths. The banking- industrial community concurred with the growth of economic autonomy overseas, so long as their interests were secured. Perhaps if Caetano had had the prestige or ruthlessness of Salazar, and had eliminated the others, he might have survived on the strength of this single group, but the habit of compromise undid him. Under the Portuguese constitution, the President had the right to dismiss the Prime Mininster. Before the elections of 1972 Caetano had debated whether to stand as President, and relinquish direct executive power to a subordinate, but in the event the incumbent, Admiral Toma s, an aging and obstinate man who saw it as his duty to preserve Salazar's Estado Novo from any depredation, refused to stand down. Judging that Tomis, then seventy-seven, was unlikely to be able to repeat the king-making performance of President Hindenberg in Germany in the 1930s, or to outlive the end of his term, Caetano did not contest. But behind Tomdis stood the politicians of the right, and the General Staff, who tended to see the overseas territories in the context of world strategy, and who relied on the thesis of the defence of Africa against Communism to ensure continued United States support. In General

The Portuguese crisis Spinola's mordant phrase, they accepted the war as a 'permanent institution'. Ministers like Rapazote or Silva Cunha regarded even the 1971 Organic Law as undesirable. Franco Nogueira believed that autonomy would lead irreversibly to anarchy, or to 'a colonialism far more severe than Africa's peoples have ever experienced before'. Kaulza d'Arriaga maintained his defence of the historic mission, 'a light for the world and the hope of the future,' against Communism and the inroads of black racialism. Lacking the unquestioned popular support of Salazar, Caetano found himself on sufferance, the moral prisoner of the past he had tried to evade. Earlier, he had had to jettison younger or more liberal ministers, and at the end of 1973 right-wing criticism, aroused by events in Mozambique, threatened the political future of the new Overseas Minister, Rebelo de Sousa, whom Caetano had chosen to see autonomy through. Even at this stage, the manifestations of a debate about Mozambique, such as the outspoken views of the Bishops of Nampula and Oporto, of the black deputies in the National Aosembly, or of the new Lisbon weekly Expresso, signified less than the economic crisis and discontent in the army - both largely products of the endless wars. Official statistics showed a rate of inflation of 11 per cent for 1971- 2, but the true figure lay between 15 per cent and 18 per cent; and although this progression developed from a relatively low cost of living, it had reached 35 per cent by the first quarter of 1974. Since this occurred at the same time as an intense liquidity shortage,t and a speculative boom in the few quoted shares on the Lisbon market, it was no wonder that longing eyes were cast at the grossly inflated military budget, more than 45 per cent of all Government spending (or 8 per cent of GNP) after 1969. Although foreign investment had increased rapidly in this period, it began to fall off in the uncertain political climate after 1972 - discouraged especially by the course of the wars in Guinea-Bissau and Mozambique. On top of this economic disequilibrium, Portugal suffered a continual haemorrhage of skilled labour, as her male population emigrated, legally or clandestinely to find work in EEC t For further amplification see Notes at end of chapter. 311 countries. At the peak, in the early 1970s, the tide reached 150,000 a year, the majority ending up in the sad and squalid bidonvilles on the outskirts of Paris and other French industrial cities. By 1974 roughly one million Portuguese - upwards of 30 per cent of the male working population - had temporarily abandoned their homeland.t They fled, partly to escape conscription, and the narrow constraints of life in rural areas (Braganqa lost 23 per cent of its population between 1961 and 1971) and partly, like the peasants who emigrated from southern Italy, Turkey, or Yugoslavia, to earn the relatively high wages available to foreign workers in the EEC. The shortage of skilled artisans was made up to some extent by coloured immigration from Cape Verde or the Azores, but the funds repatriated by workers in Europe (£350 million in 1972, over £400 million in 1973), while favouring Portugal's balance of payments, contributed also to inflation; and by siphoning off the movement of population from rural areas to the cities, emigration forced expanding industry into a straitjacket, forbidden to raise wages yet unable to expand without skilled labour. These tensions would have come to the surface even if there had been no revolution, and there is some evidence that the out-bursts of industrial action which occurred after the revolution in May and June 1974 had been planned months earlier, to undermine the Caetano Government's policy of wage restraint. Although the gravity of the economic situation was not widely perceived outside the governing elite, it combined with the fresh uncertainties of overseas policy, early in 1974, to weaken Caetano's authority and regroup in hard-line solutions the political right-wing and the group of industrial and banking magnates. The armed forces stood apart from the problems of civil government. They had not assumed, in recent Portuguese history, the role of guardian of the national interest - a role filled to some extent by the PIDE/DGS, which the army in particular disliked and which, in Mozambique, had led to confusion and bitterness between them. In the first decade of colonial wars the army had benefited. Once socially t For further amplification see Notes at end of chapter. CABORA BASSA 312

The Portuguese crisis contemptible, beside the navy, it drew a flood of able volunteers in the wake of Salazar's national defence appeal against the insurgents in Angola; and in spite of heavy casualties as the guerillas obtained better weapons and developed better techniques, overseas service remained popular and, with its large supplement, relatively well-paid.t Private soldiers, whether conscripts or volunteers, found conditions actually an improvement on those of peasant villages in Tras os Montes; the wars never reached the intensity of casualties in Algeria or Indo- China, and the rate of desertion, though well-publicized by the opposition, seems low in relation to the size of the conscript forces overseas. Until the early seventies, there was no reason to disclaim the status accorded it or the message charted in the speeches of Caetano, articles in the press, and flamboyant posters in the towns - that the army was the agency of the psychological campaign and the social programme; that it was a force for education, social welfare, and the civilization of black Portugal, and only in the last resort, the means of compulsion. But the incidence of conscription bore heavily on professional conscripts, whose university education suffered from four years service, and whose later career might be interrupted, up to the age of forty-five, by recall to the colours. Skilled technicians, once called up, might effectively be barred from return to civil employment. It was among the conscript officers - especially from 1969 onwards, the graduates - that dissent showed most obviously. Then, after the first ten years, the strain began to show among regulars. By 1973, out of any overseas battalion of eight hundred men, probably only forty would be regular soldiers - the senior NCOs, some captains, the majors and colonels. Repetition of tours of duty had in some cases reduced home service to a mere two years in twelve; the regular army began to look like a foreign legion. The conditions of wartime do not seem to have created unrest so much as the gradual debasement of the regular cadre, its low level of salaries compared with those of other Nato armies, the provision of second-hand, out-dated equipment, and, above all, the promotion block. On one side stood the f For further amplification see Notes at end of chapter. 313 CABORA BASSA 314 ranks of elderly generals, on the other a host of nationalservice officers, promoted hastily to keep pace with the mass intake of recruits, predominately black, overseas. A core of majors and colonels in their forties watched these men promoted in six months from Second Lieutenant, and in another nine to Captain. The Military Academy, instead of training a small elite, professional army, as in France or Britain, turned out newly-minted majors after short courses, to stand beside colleagues of perhaps fifteen years hard service, thus devaluing both the rank and the institution itself. The nadir was reached in 1973, when the whole army, of 175,000 men, attracted only thirty-seven career lieutenants. The General Staff, in particular Generals Costa Gomes and Spinola, took up these grievances, and sought to have the elderly, semi-retired generals pensioned off, and the Academy elevated to the rank of Polytechnique, as in France. But by this time a great disillusionment with the political direction of the wars, and with their ostensible aims, was spreading. Faith in the Portuguese mission had been eroded by the freethinking arguments of precisely those national-service officers and NCOs against whom the Minister of Defence, General Deslandes, had fulminated in 1969. Middle-ranking officers did not conspire or take part in political activity - DGS surveillance was too keen - but out of prolonged discussion and the actual experience of war they evolved an argument new to Portugal (which Spinola had begun to formulate in his thesis, Portugal and the Future, during 1972): that they had a duty, perhaps even a right, to ensure that the nation was not betrayed by politicians. There remained a clear distinction between discontent and disloyalty. But they feared also the dishonour of losing, or appearing to lose a war. They appreciated that Caetano might prefer a military defeat to a political collapse in Guinea-Bissau or Mozambique as the only means of saving the face of the regime; they needed to be on the right side of history. In a curious way, the Government, by its insistent propaganda about the army's mission within the great Portuguese tradition, had created the instrument for the destruction of its own dictatorship. In September 1973 a group of officers at Evora drafted a reasoned protest to the Ministry of Defence. When it was ignored, they began to organize, and by

The Portuguese crisis the end of the year the Armed Forces Movement (AFM), as they styled themselves, counted a thousand members in Portugal and the army overseas. Publication of Spinola's book in February 1974 acted as a catalyst of developments in the armed forces and altered the balance of power in Government. Portugal and the Future grew out of his experience as Commander- in-Chief in Guinea-Bissau, and the failure to win more than a temporary advantage against PAIGC, or to achieve a negotiated peace with his honourable opponent, Amilcar Cabral, despite their secret negotiations through the mediation of President Senghor of Senegal. Spinola had drafted the book with the assistance of his staff, and was ready to publish privately abroad, when the changing climate of opinion late in 1973 decided him to launch it in Lisbon, in order to force the Government and public into a debate. (Contrary to what was assumed at the time, Caetano did not give the book his imprimatur. Spinola submitted it to his superior, Costa Gomes, then Chief of Staff, who approved it; the Defence Minister, Silva Cunha, evaded responsibility by replying that he had not read it, but accepted the recommendation of his Chief of Staff). Much of Portugal and the Future was undisputed: the economic crisis, the burden of the war, the dissolution of public consent in Portugal, and the danger of being left out of Europe. Part was a piecejustificative, excusing the intervention of a soldier. But three themes stood out: that a military solution to the colonial wars was no longer possible for either side; that only self-determination, within a federal structure, and a Lusitanian community including Brazil, could safeguard Portugal's future; and that progress towards self-determination must come through the full participation of the African population, followed by a referendum on the forms of future government. Quite specifically, Spinola rejected the historic argument, contending that there had been no common past, only that there might be a common future. He rejected all forms of white dominance, and took for granted that the African majority would wish to remain linked to the metropolis, 'but in the African way' - as Goa might have done, had India been handled wisely. At the same time he defined the role of the armed forces: they should hold the ring while 315 political and social answers were found, before residual loyalty to Portugal evaporated. In the atmosphere of Lisbon, some of these propositions appeared revolutionary. The contradictions in the book, Spinola's ignorance of underlying economic realities in Mozambique and Angola, and the difficulties of the modus operandi suggested, were ignored. The federal solution itself impressed foreign commentators more than politicians at home, who found it too rigidly schematic and impatient of subtleties a characteristically staff-college schedule, barely allowing for bargaining processes, rather like General Wavell's programme for the precipitate evacuation of India after 1945, which enraged Attlee and the Labour Cabinet. But Spinola had attacked the thesis which justified the whole of overseas policy. 'We must smash the myth according to which we are defending the West and Western civilization .... It is the truth that the Western interests which we presume to defend have not coincided, recently, with our own interests.' He attacked the proposition that 'the essence of the Portuguese nation is its civilizing mission.' Then, having demolished the old arguments, he replaced them with a new: The Portuguese presence overseas is a prerequisite for our survival as a free and independent nation. Without the African territories, the country will be reduced to a voiceless corner of a gigantic Europe and will have no possible trump card to play to assert itself in the concert of nations; it will end by having a merely formal existence in a political framework where its real independence will be totally jeopardized.' Could Portugal retain the substance and relinquish the rest? Could the army oversee such a complicated process of disengagement? No doubt with Kaulza d'Arriaga's abortive coup of December in mind,* Spinola rejected the idea that any junta of generals could steer between the twin evils of the 'oldfashioned static position' and 'progressive radicalism'; and he * D'Arriaga had sounded out opinion among his colleagues on the degree of support for a right-wing movement. He found little sympathy except among fellow generals; and when the move %%a, discovered he was placed under house arrest. A full apology to Cactano restored his freedom; and he was then encouraged to go on leave, to South Africa. CABORA BASSA 316

The Portuguese crisis seems to have reflected the opinion of his own staff, if not of the Armed Forces Movement, when he spoke of the armed forces as guarantor of 'a new conception of the unity of the Nation'. Caetano's opinion of Spinola can only be inferred from the fact that he awarded him Portugal's highest military decoration, the Turre e Spada (an honour withheld from the C-in-C Mozambique, General d'Arriaga) and appointed him deputy Chief of Staff under the like-minded Costa Gomes in autumn 1973. It may be that he wished to precipitate a confrontation between them and the massing forces of the right, to give him political strength to pursue the twin aims of autonomy overseas and industrialization at home. But when Spinola proposed in February that they should work together to renew Portugal by a revolution from above, Caetano demurred. He remained, in a strange way, a democrat, caught in the prison of Salazar's dictatorial technique; his political career had been sustained by compromise and the search for consent. In the last resort, Caetano believed that Portugal mattered more than the empire, just as he believed the country's economic institutions counted for more than the army. As an experienced, subtle politician, he mistrusted crude, military intervention and violent change. Moreover, it would seem from various events of that time such as the wide freedom permitted to Mozambique's Third Force, the search for an African elite to include in the new National Assembly, and the encouragement given to Jardim that Caetano was more than half inclined to take up the Kaunda-Banda offer of mediation in Mozambique, and to pull out of Guinea-Bissau. In February and early March Caetano made two vitally important speeches, cautious, veiled, and ambiguous, but exposing the nature of the Portuguese dilemma, against a background suddenly clouded by the right-wing outcry against Spinola's book and two events which diminished faith in the possibility of multi- racial cooperation in a free African state: Frelimo's changed tactics of aggression against white settlers and ahleanentos, and the fortuitous expulsion by President Mobutu of a long-settled Portuguese community in Zaire. Addressing the National Conference of the ANP, he replied to Spinola, admitting the need for further change, 317 CABORA BASSA 318 but relying on the precepts of Salazar to guide him between extremes of right and left. He acknowledged the danger of Frelimo penetration and the difficult economic circumstances of Mozambique, but contrasted them with the secure condition of Angola; and he dismissed the idea of federation as unworkable, despite his own support for it twelve years before. The speech amounted to a reasoned rejection of Portugal and the Future, yet it set nothing in its place, except the now threadbare assertions of the presence overseas - 'These African societies can only cement their unity in Portuguese traditions and the common language, Portuguese' - and a reaffirmation of the defence of the free world against Communism on African soil.2 A month later, at the National Assembly, the tenor was harder and more precise. While praising the fortitude of the armed forces, Caetano grounded his policy in first principles and his speech stands as the logical conclusion of the decisions taken in 1968-9. 'This autonomy ... exceeds the freedom of most federated states'; and precisely because of that freedom, 'Central Government should retain some powers to guarantee the peaceful coexistence of the communities concerned.' As an indication of progress, however, Caetano held out a promise of reconciliation to the liberation movements (p. 272) and, in the only public indication of his personal insecurity, pleaded for time and understanding from the rest of the world. In that Assembly the Prime Minister had his absolute majority. Outside, the counter-attack against Spinola proceeded. Rather than sacrifice him immediately to the demands of the right, Caetano asked for a public and personal pledge of loyalty to the regime, and when this was refused, dismissed him, together with Costa Gomes, on 14 March. The army was confined to barracks, and in front of television cameras in the Lisbon garrison, before a long line of semiretired generals, General Paiva Branddo gave the pledge instead. The Government and the DGS appear to have believed that the AFM would learn from Spinola's fall, and that in due course the two generals could be discreetly reinstated. But events in Mozambique had already profoundly affected the plans of the AFM. During January and February, in the wake

The Portuguese crisis of Frelimo sabotage in Manica and Sofala, and the murder of the Diaz family (p. 155), the white population finally lost confidence in the army's ability to protect them. The white riots in Beira, Vila Manica, and Vila Pery, were directed against the army and put down by the army; and it took all Costa Gomes's authority and persuasion, and the dismissal of the military governor of Beira, to restore a measure of harmony. Many white settlers turned to the DGS or demanded weapons; and in Vila Perv district at least three hundred were armed, leading to the indiscriminate killing of a number of Africans, and deep racial discontent in the black suburbs. The army faced a sudden loss of support as well as unpopularity; the luxurious rest centre in Beira, object of bitter jealousy during the riots, was closed down, and the Governor-General himself, having refused to intervene in the riots for fear of losing the dignity of his office, began covertly to support the DGS in raising new contingents for the flechas. The war was not lost. Given adequate backing, the army could have held Mozambique for years more, as Frelimo privately acknowledged. Subversion and actual fighting never reached the scale of the war in Algeria in the 1950s. But the antagonism of local whites, and the suspicion that the Government in Lisbon was about to make them scapegoats for the failure of the entire pacification programme, undermined the will of the regular officers, many of whom, following rumours of Spinola's tentative with PAIGC, and the tenor of his book, had turned to thoughts of a negotiated peace. Under the stimulus of the Mozambique experience, the leaders of the AFM, in particular Colonels Otelo Saraiva Carvalho and Vasco Gonqalves, concerted their plans, secure from DGS intelligence, using the communication centre in the Ministry of Defence itself. But the AFM assumed, wrongly, that Spinola's book represented a programme of action; and when, in the wake of his dismissal, two hundred officers and men of the fifth Infantry Regiment drove from the Caldas da Rainha barracks to Lisbon, Spinola gave no sign of support. The mutineers retired and large numbers of arrests followed, including that of Spinola's chef du cabinet, Colonel Bruno. There is no need to retell the dramatic events of 25 April. 319

The revolution succeeded as much by chance and the opportune collaboration of the navy and air force as by close planning; the regular officers were amazed at the degree of radicalization of conscript officers and NCOs, whose assistance they had omitted to seek. Even so, the fact that it was almost bloodless depended on the surrender of the Government. President Tomis gave in, not to the AFM officers, but to a man of equivalent rank; and Caetano, in his last ministerial act, conveyed legitimacy on Spinola - probably in return for safe conduct, after a short interval, to Brazil. It seems clear that Costa Gomes declined the leadership; and for a time Spinola assumed the heroic aura of a Garibaldi or Fidel Castro. Retrospect suggests that his position resembled rather that of Kerensky. He was acceptable to the AFM as the most successful and heroic military commander, and as the only man with the courage to face up to the desperate logic of the colonial wars; equally, he represented to the industrialists and bankers the stability of the former regime. Apparently apolitical, unlike other generals, he seemed suited to preside over a coalition of political groups, and to the majority of officers who had not joined the AFM, he upheld the traditions of the services against a form of revolutionary democracy only too clearly exposed in MRPP (Marxist-Leninist) pamphlets addressed in May to 'workers, peasants, soldiers, and sailors'. Most important, themes in Portugal and the Future coincided with AFM doctrine about the place of the movement in Portuguese society. Their Manifesto was written ... in the profound conviction that they [the AFM] interpret the aspirations and interests of the great majority of the Portuguese people, and that their action is fully justified in the name of national salvation, making use of the power which is conferred on it by the Nation... The principles and objectives here proclaimed constitute a duty assumed on behalf of the country and serve the highest interests of the Nation.' The army had come of age and taken on a historical-political role: there are echoes of the stand taken by the German army CABORA BASSA 320

The Portuguese crisis 321 in 1918 and of de Gaulle's response to the Algerian disaster; yet the AFM, while eschewing populism, remained a profoundly democratic institution. It wished to restore Portugal to freedom, not to impose military rule, and it needed to set up a compromise form of government, to ensure stability and prepare for elections. Febrile and dedicated as the AFM officers were, they proclaimed themselves, like Cincinnatus, ready to return to their barracks as soon as the Portuguese people had made their choice. Consequently the Government consisted of an odd triangle of forces - an 'equilibrium of instability'. General Spinola, as President, headed the Junta of National Salvation, including Costa Gomes, as Chief of Staff, and the service chiefs, while the civilian element was formed by Socialists, Popular Democrats, and Communists. The ramshackle structure held together, at first by direct contact, monitored by members of the AFM sitting on all committees; and then through the Council of State, a body of twenty-one representatives, seven of the Junta, seven of the Provisional Government, and seven of the AFM. The conflicts of the next six months can be seen as the result of competition between the Junta and the AFM, in defining the strength and the position in the political spectrum of the civilian sector to whom they entrusted the government. Despite the protocol governing the relations of the Junta and the AFM, drafted on 26 April, differences soon appeared. Spinola desired to restrict the AFM's nominations to the Council of State, while they, conscious of being cut off from the civil service, departmental authority, and the actual machinery of government, regarded the Council as their only means of maintaining the momentum of change and the authority conveyed on them by the events of 25 April. For a time, as the Provisional Government settled in to deal with the economic crisis, the sudden shortfall in tourist revenue and remittances from Europe, as Spinola enjoyed his honeymoon with the world press, and as the Communist Party exerted itself to restrain the more extravagant strikes and working-class wage demands, the AFM retired into the shadows. Spinola took pains to reassure the business community that moderate policies would be followed; and observers predicted that nothing short of a violent economic regression

CABORA BASSA 322 could bring the AFM back to power. Dissension between the various elements, however, grew in direct proportion to the vagaries of overseas policy. In their Manifesto, the AFM had referred only obliquely to the problem of decolonization, partly because their ideas had been contained in their earlier petition to the Minister of Defence, and partly because Spinola persuaded them not to be too precise. Dismantling of the ANP overseas, as well as in Portugal, and the dismissal of existing Governors-General were the only specific proposals; the Manifesto instead gave prominence to Portuguese problems, inflation, the cost of living, and the evil dominance of monopoly capital. The word 'independence' did not appear, only 'self-determination'; and for several weeks the AFM made no attempt to elaborate on the document. Within the Provisional Government a broad consensus obtained. Individual ministers showed themselves preoccupied with the dangers of anarchy and outside intervention in Mozambique (though not in Angola) and with the need for a speedy schedule of progress towards independence. The Socialist-Communist Party tactical alliance, secured as far as either side could see, in mutual self- interest, at least until the first general election, guaranteed a common front on the colonial question. Neither party was inclined to lay down conditions for independence, to safeguard future Portuguese interests or existing investments; and an impression was given that the Provisional Government would settle with the liberation movements as fast and as simply as possible. Equally, they envisaged no special provision for aid or technical assistance. Portugal must admit to being small and poor, and the territories must stand on their own feet. If the newly emancipated parties imagined they could cut off Portugal from the past and the dream of empire, they had not taken into account the determination of the Junta not to scuttle historic responsibilities, nor abandon the substance of the Lusitanian vision, sustained as it was by the economic and strategic importance of Mozambique and Angola to the Western world - a determination supported by the powerful civil servants of the Overseas Ministry, who understandably resented attempts summarily to relinquish control, and by the

The Portuguese crisis centre grouping, which included liberals from the old opposition and the new Popular Democratic Party. Spinola himself held to the general theme of his book, at least as far as the time- schedule and the mode of elections were concerned. A cease-fire was to be followed by a year's freedom of political activity, in which he hoped Frelimo would participate. The provinces were then to be given a plebiscite on the question of their future association with Portugal, and elections to choose a new Government. But actual terms were for the Provisional Government to negotiate with PAIGC, Frelimo, and the Angolan movements. Spinola and the Junta reserved the power of final decision and their requirements appeared high. When Costa Gomes announced the programme in Luanda at the end of April, he seemed to be reassuring the white Angolans that they would not be precipitated into autonomy. A fortnight later, on 11 May, he suggested the immediate implementation of a cease-fire in Mozambique and, by implication, challenged Frelimo to test their hold on the African population. Costa Gomes returned in a pessimistic mood, fully aware of the multiple difficulties of relinquishing authority to a country where, after the demise of the ANP, no political structures existed; and impressed both with the war-weariness of the regular army and the reported intransigence of Frelimo. From the time of his visit in January, he had been preoccupied with the problem of transfer of authority to Lourenqo Marques, a problem which resembled that of de Gaulle in Algeria and Mountbatten in India; and in some ways he was already far ahead of the President in his thinking. More clearly, also, than the inexperienced civilian ministers, he saw that if consensus inside Portugal was to be maintained, decolonization had to include guarantees for the future of the white population, of investments, and of existing defence arrangements, not perhaps with Rhodesia, but with South Africa, and the Nato powers. Finally, no-one in the Junta who had loyally served the previous regime envisaged a future in which Portugal retained no links with Africa; and in this sense, Spinola's Lusitanian community, which included the vital Brazilian connection, retained its influence. The appointment to the Junta of General Diogo Neto, after 323 long service in Mozambique, indicated the importance attached to a compromise solution. When Spinola addressed private meetings of industrialists and bankers on 28 and 30 April, in addition to setting out his reasons for assuming the leadership (after Caetano had rejected his original proposal to make a revolution from above), he reassured them not only about the political and economic direction of Portugal, but also about the future of its economic links with the empire. His audience formed a rent-roll of the business oligarchy: and afterwards, Manuel de Melo, Miguel Quina, Manuel Espirito Santo, and Antonio Champalimaud, evinced themselves satisfied that 'prudence and economic necessity' would guide the new regime. Continued solidarity among the pre- revolutionary power groups became more obvious when the Junta gave its support to employers' organizations, in meeting the sudden and crippling outbreak of strikes,' and when Spinola discreetly encouraged Kaulza d'Arriaga and others to revive a new democratic party of the right. By essaying a compromise similar to that of Caetano, the Junta placed itself in the same straitjacket. Even if the new Foreign Minister, Mario Soares, and the Overseas Minister, Almeida Santos, had agreed to insist on these terms for decolonization, success depended on the willingness of liberation movements to agree to a cease-fire and a final plebiscite, rather than immediate independence. If not, then the army had to be prepared to fight on - a hypothesis almost inconceivable to many members of the AFM who envisaged no further duty than a holding operation, pending free elections. And if the white population of Angola or Mozambique resisted, it would fall on the army to enforce against them a policy which might not even be acceptable to the guerillas. These contradictions were exposed almost at once, in the first, abortive, round of negotiations. But even before they began, the possibility that white Mozambicans might take part was written out by the Junta's response to a renewed offer of mediation by Kaunda and Banda transmitted through Jardim. Frelimo appear to have been not unwilling to use this channel, in spite of their mistrust of Jardim, so long as negotiations took place directly with the Lisbon Government; but, in spite of a certain degree CABORA BASSA 324

The Portuguese crisis of support from the AFM, Jardim was as unacceptable to the Junta as was his scheme for Mozambicans to participate in the peace settlement. His regional power, shown by his ability to calm hostile opinion in Beira by broadcasting on the local radio, and in his supposed link with sections of the black army, revived the Junta's fears of white separatism. Even if they were misplaced, these fears led to his house arrest in Lisbon in May, and a rejection of the suggested dates for a conference in Lusaka. Since the other white democratic parties sought to associate themselves with Frelimo, and since the right-wing Fico ('I stay') flatly opposed a transfer of power at all, the remaining whites, and perhaps 7.5 million blacks not then pledged to Frelimo, were left without a hearing, apart from the faint chance that Spinola's proposed plebiscite would take place. To the very end, Portugal retained its paralyzing control over the province. The treatment of the Mozambique Government confirmed this position. The Governor-General and his immediate staff were dismissed, the DGS outlawed or merged with military intelligence, and the Assembly suspended on April 30, just before it was due to meet. In the ensuing disarray, the army held control; the new appointments were made by the Junta, who chose Almeida Santos, the Delgadist lawyer from Lourenqo Marques (once the guardian of Anglo-American's legal interest in Mozambique) as Overseas Minister. Bargaining between the AFM and Junta over the choice of Governor-General and new provincial secretaries left the country almost without effective administration during the vital months of May and June. Meanwhile, a two-pronged diplomatic offensive began, to achieve a quick cease- fire in Guinea-Bissau and Mozambique. During his visit early in May, Costa Gomes had primed a delegation of Frelimo sympathizers to explore terms in Dar es Salaam, while Mario Soares met PAIGC representatives in Dakar and then, with British diplomatic assistance, opened talks in London. At first these appeared friendly and hopeful; but because Soares could not offer substantially more than the Junta allowed, they ran slowly into the sands, and were finally curtailed when PAIGC learned of Frelimo's hard-line response. Soares and the Junta (for Colonel Bruno had accompanied 325

CABORA BASSA 326 him to Dakar) had both expected the liberation movements to welcome a cease- fire, and seem to have ignored the implications of Frelimo's non-negotiable demands contained in the Kaunda-Banda proposals. Yet Frelimo was not in the least mollified by Costa Gomes's unofficial delegation, nor by the conciliatory remarks of Almeida Santos during a visit to settle the Mozambique dock and railway strikes at the end of May. It returned a dusty answer to the cease-fire proposal; and outlined to the OAU summit conference, then meeting in Mogadishu, the full range of original demands - Portugal had to: 'recognize Frelimo as the legitimate representative of the people of Mozambique in accordance with relevant OAU and UN resolutions; recognize the right of the people of Mozambique to total and complete independence in the whole of their territory; accept the principle of the transfer of power it still exercises to the institution representing the people of Mozambique, that is to Frelimo.'6 Frelimo would not accept an externally-imposed federal structure, nor a plebiscite.7 On the other hand, it recognized the need for a settlement both honourable and justifiable in Portuguese eyes; and on a wide range of issues, including the vital questions of foreign loans and the Mozambique national debt, the future of the white population, and of links with Portugal, of Cabora Bassa, of the migratory workers, and external relations with South Africa, of defence, non- alignment, and the Nato powers, it was prepared to compromise. Its leaders had instructed the troops that it was impossible to accept less than the three main demands - a form of burning their boats but that otherwise the Central Committee must be free to bargain 'in the general interests of Mozambique'. With the urgent need to retain white skills and Portuguese capital in mind, Machel broadcast an appeal to the white population on 3 May, and a statement that Frelimo had never sought to attack Portuguese citizens, only the fascist regime; and Jardim's newspaper, Noticias da Beira printed an interview in which he repeated the message. But with a shrewd estimate of the balance of forces in Lisbon, and a measure of inspired faith, Machel and his ablest lieutenant, Joaquim Chissano, resisted the pressure of the three presidents - Kaunda and Banda now being joined by Nyerere - and of the pro-

The Portuguese crisis Russian faction in Dar es Salaam to moderate the three principal demands and settle, before, as they feared, Mozambique dissolved in anarchy. As the London negotiations with PAIGC broke down in the face of political demands, chiefly for the independence of Cape Verde, Frelimo agreed to a meeting with the Portuguese in Lusaka on 4 June. But the efforts of a bewildering range of interested parties, Britain, France, Senegal, and the United States, among others, ended in an emotional but barren confrontation; on 11 June Spinola announced bluntly that 'democracy must precede independence'; and two days later Machel rejected the cease-fire. Frelimo had taken the considerable risk of alienating its principal backers - Zambia and Tanzania, who had both chosen in the past to negotiate on independence by gradual processes - in the belief that they could, if necessary, carry on the war, that the Portuguese had lost the will to fight, and that compromise would impair the hard-won unity of their movement. Because of the unresolved conflicts within the Government in Lisbon, weeks passed in which the lineaments of authority in Mozambique decayed. Soares and Otelo Carvalho returned from Lusaka prepared to accede to Frelimo's terms; Spinola refused, and against the Socialist and Communist ministers, and the AFM, his supporters attempted to weight the Council of State and enhance the authority of the presidency. The Cabinet crisis in June, when the Prime Minister, Palma Carlos, and the liberals resigned, occurred as a result of a struggle for power on the domestic front. (In order to deal with the balance of payments crisis, strikes, wages and inflation, and to prevent the spread of Communist influence in local government and trade union movements, Palma Carlos asked for increased authority for the office of Prime Minister. The AFM, rather than the liberals, benefited from the outcome when he resigned.) But this crisis was resolved by the appointment as Prime Minister, not of Spinola's nominee, the former Defence Minister, but of Brigadier Vasco Gonqalves, with increased power to appoint AFM ministers. The other chief architect of the AFM, Otelo Carvalho, became head of a new operational body, COPCON, designed to supersede the Council of State and ensure fulfilment of the AFM 327

Manifesto, and assumed command also of the strategically important Lisbon garrison. On the other hand, the AFM itself, always an inchoate body, now divided between those sympathetic to the Socialist-Communist alliance, and those who supported, not Palma Carlos, but centrist parties such as PPD. The distinction showed clearly in July, when the Economics Minister, Dr Almeida, introduced measures to halt inflation, and it explains why no clear message on decolonization came from Lisbon, and why the obscurities of the Manifesto remained undefined. The Junta and the AFM were finally induced to compose their differences. In the second stage, during the vacuum of power which lasted until August, civil government in Mozambique disintegrated so rapidly that all solutions other than immediate recognition of Frelimo, and transfer of effective authority, became irrelevant. White emigration and the illegal export of capital increased sharply; air flights from Mozambique were booked months ahead, and all ships' cargo space taken. Morale manifestly weakened in outlying areas, even among officials, and half the faculty of the University of Lourenqo Marques abandoned their duties. Inflation, now apparently out of hand, inspired successive waves of strikes for 100 per cent wage increases among organized African groups on the docks, railways, and at Cabora Bassa, and were met with Government settlements which ignored the capacity of companies to pay. Political parties flourished in the heated atmosphere, without roots and often without reason. Only two organizations attracted wide support among whites: Grupa Unida de Mozambique (Gumo) and the democrats, whose leaders, Pereira Leite, Jorge Abreu, Adriao Rodriguez, and Baltasar Alvoes (later a Frelimo nominee for the Transitional Government), tried to show that Frelimo offered the only hope for political stability; and Fico, home of those whose main determination was rejection of black majority rule. Generally, Gumo attracted support from executives, entrepreneurs, and administrators, Fico the loyalties of farmers, smaller businessmen, and shopkeepers, and Fico developed links with expatriates in South Africa and similar movements in Angola. Physical attacks on members of Gumo became commonplace. In these muddy waters other rank growths flourished. CABORA BASSA 328

The Portuguese crisis Members of the DGS, operating either from within army intelligence, at large in Mozambique, or from Rhodesia, used recruits from the flechas and white commandos to set up a clandestine organization dedicated to resisting a negotiated peace. Whether their efforts, or those of die-hard administrators, were responsible is unclear, but efforts were made to capitalize on tribal animosities, and the fear, particularly among Macua, that Frelimo rule would be to their disadvantage. Coremo, whose leaders had realized that outside Mozambique they had no hope of resisting Frelimo, now infiltrated and collaborated in the hope of sharing power; and this explains the extraordinary association of at least three Coremo leaders (former members of Frelimo) with the last-ditch white rebels in Lourenqo Marques in September 1974. Other defectors, like Kavandame and Mrupa, whose hopes of survival under a Frelimo regime were slight, looked round for more powerful allies. Meanwhile Frelimo supporters came out into the open, increasingly jubilant and demanding; and the army released the political prisoners to join them. During this period Frelimo kept up its attacks, and sabotage of the railways; and drove the Portuguese out of the town of Marrumbene in Zambesia. Notwithstanding, in the Portuguese army the black troops' discipline held good; even the most politicized sections, which might have defected, or followed a multi-racial party, had Jardim been able to weld together his own followers with Gumo, took no part except to show qualified approval of Frelimo troops as they emerged from the bush, under local agreements, in July and August. But ominous signs appeared in July that white conscripts were becoming reluctant to fight on. The state of Mozambique, torn between war and internal dissension, was far worse than that of Angola, or even Guinea-Bissau, where talks between military commanders on both sides led a tacit cease-fire in July. Yet it had to be resolved in Portugal's own interests without 'Congolization', and the solution had to be one which would not undermine the future of Angola as envisaged by the Junta. Portugal was already susceptible to a series of pressures, more or less open, from the other powers concerned. 329

Apart from the OAU's and Mozambique's northern neighbours' preoccupation with Frelimo, Presidents Kaunda, Mobutu, and Nyerere, made attempts to bring some harmony of purpose to the three warring Angolan movements, unity being an essential preliminary to independence. Although black African forces generally counted for little in the Lisbon balance, their requirements were conveyed to the Provisional Government, if not the Junta, by Soares; and additional weight was given by the British, who like Kaunda, saw in the independence of Mozambique the means at last to inflict full sanctions on Rhodesia. Other EEC countries let it be known that a democratic Portugal needed only to free the former colonies to become fully respectable; and, within Nato, those who had opposed the involvement in the South Atlantic appeared relieved. Britain, under a Labour Government, was no longer so sensitive to the latter issue (though compliant towards continued naval cooperation in the Indian Ocean); on the other hand, the United States' concern was increased at the possibility of Russia acquiring bases both on the east and west coasts of Africa. It was this, rather than crude rumours of CIA counter-revolutionary plotting at the end of April, which seems to have caused anxieties among Spinola, Costa Gomes, and the chiefs of staff in Lisbon, who were well aware of the value of American economic aid, in return for the Azores base, and of the extent of political leverage that went with it. On 18 June, Spinola met Nixon in the Azores, ostensibly to reaffirm the new Government's continued membership of Nato and to discuss aid in the economic crisis. There seems little doubt that the future of Angola was also discussed. South Africa played a very cool game. Secretary of State Fourie let it be known in May that the Republic would cooperate with any 'reasonable' new African government in Mozambique, and while the lobby in Lisbon sought to hold back the Provisional Government from hasty action, it did not oppose Portugal's talks with Frelimo. At the same time, Admiral Biermann went to America for talks with Pentagon officials, probably with the aim of emphasizing the strategic significance of Angola. Border defences were strengthened, in a show of force as necessary to restore the morale of the South CABORA BASSA 330

The Portuguese crisis 331 African electorate as to deter Frelimo from possible cooperation with ANC guerillas; but Botha, Vorster, and even Connie Mulder, all through the summer, and even during the white revolt in September, held strictly to the non- intervention line. One important consequence was a rapid disengagement from Rhodesia. Smith paid one of his regular visits to Vorster in M\ay, to discuss the Portuguese revolution, and returned without assurance of help. Later, Hilgard Muller stated publicly that he hoped independence in Mozambique and Angola would be followed 'by a speedy and satisfactory solution of the Rhodesian problem';8 while in Salisbury itself, South African trade union leader Arthur Grobbelaar told the Rhodesian TUC: 'It is realistic to recognise that while your country has until now acted as a buffer zone in blunting incursions from Black Africa, your importance to South Africa for that purpose has now declined." Rhodesia was hit economically at once by the railway strikes in Mozambique, and began feverishly and at great cost to lay a single-track railway to connect with the South African system at Beit Bridge, in order to be free of dependence on both Mozambique and Botswana. But South African Railways' freight capacity was already far too overloaded to absorb all Rhodesia's exports and imports, and South African officials showed themselves unwilling to undertake the falsification of loading bills, essential to sanctions breaking, which had previously been done by complaisant Portuguese officials. Something like panic struck Salisbury, and though it did not show in the offer made to Bishop Muzorewa in June, or at all in public during the sudden election of July, it could be read elsewhere: in the covert support given to DGS fugitives, after military cooperation with the Portuguese army was prohibited, early in June; in the rough tactics adopted by the lobby in Lisbon (who had, earlier, misguidedly attacked Spinola's book, and lost much of their influence as a result); and in the frantic herding of the population of the north-east into fortified camps, in operation 'Overlord'. The AFM had already shown that they were not prepared to wait for Spinola to make up his mind. With army assistance, Jardim was allowed to escape from his refuge in the

CABORA BASSA 332 Malawian embassy, to Spain and thence to Malawi (where his activities caused a breach in diplomatic relations); meanwhile Costa Gomes, who now emerged as the most powerful man in the Junta, with equivalent rank to the Prime Minister, let it be known that he thought Spinola's plans hopelessly outdated. At last, under the stimulus of the balance of payments crisis in Portugal, and an adverse vote in the Council of State in July (eighteen to three against Spinola on the interpretation of 'selfdetermination' in the AFM Manifesto as meaning 'independence'), Spinola agreed to give way, and promised independence to Guinea-Bissau and Mozambique, thus opening the final stage of decolonization, though not wholly resolving the struggle for power within the Government. In the latter half of July, effective agreement was reached in Guinea-Bissau; Soares de Melo, the temporary Governor-General, resigned in Mozambique, and the army prepared for a combined police role with Frelimo. The transfer of power, which took six weeks to achieve and was embodied in the treaty signed on 7 September, only narrowly prevented something akin to civil war - which by that stage must surely, as the AFM appreciated, have nullified any attempt to hold the political ring in Mozambique for a year, pending free elections or a plebiscite. Given Frelimo's attitude and the weakness of Portugal's hold, the four months between the revolution and the treaty were lost in a hopeless endeavour to salvage what the Caetano Government's procrastination had wrecked. But this four-month period showed that Portugal would not easily abandon her lasting interests especially in Angola; it rallied international support, and brought the UN Secretary-General, Kurt Waldheim, to Lisbon, to pledge United Nations' support in the decolonization process, and it induced Frelimo to make some concessions, both to Portugal and towards the white population of Mozambique. The Lusaka Agreement provided for independence on 25 June 1975, the fourteenth anniversary of Frelimo's foundation. In the intervening period, the country would be ruled by a Transitional Government, two-thirds Frelimo, one- third Portuguese, under a High Commissioner appointed by the Portuguese President, and a Prime Minister, appointed by The Portuguese crisis 333 Frelimo. The two armies were to collaborate in a joint military commission, responsible for keeping the peace, and a final cease-fire was proclaimed. Guarantees were given by Frelimo of its acceptance of 'financial obligations undertaken by the Portuguese State in the name of Mozambique, provided that those obligations were undertaken in the effective interest of this territory'; and of the future security of the white population, as long as they 'identify voluntarily with the aspirations of the Mozambican nation'. In addition, provision was made for a central issuing bank (the Instituto do Credito) to assume all assets and liabilities of the BNU, for aid and technical assistance, and for future 'fraternal and harmonious cooperation between Portugal and Mozambique'. Large areas, particularly in respect of future external relations, remained to be determined, but Portugal seemed to have safeguarded what it could. It was ironic that the Lusaka Agreement should differ scarcely at all from the terms outlined to Kaunda by Machel more than a year earlier; which, if Caetano had accepted them, might have forestalled the revolution itself.

Notes and references Chapter 13 The Portuguese crisis + Note page 311. In the fourth development plan a rate of 7.5 per cent growth had been projected, on the basis of recent industrial output growth of 8 per cent. Yet GNP had rarely passed 5 per cent and, in spite of a burst of investment (welcomed by the OECD in 1971 a, the only way for Portugal to ,urvive), lack of competition under monopolistic conditions, excesive reliance on state credit because of credit restrictions, the inadequacies of the banking system, deliberately low interest rates, and high gold reserve,, combined to create a liquidity shortage. The Gladstonian type of budgeting continued by Caetano did nothing to release funds for investment. t Note page 312. Official estimates gave 106,234 in Europe in 1966, 60,000 of whom were in France; 179,704 and 135,667 for 1970; 469,000 and 380,000 for 1973; unofficial figures, including illegal emigration, suggest a total of 694,000 in 1972, 1,050,000 for 1974. Note page 313. Rates of pay, 1971-2. Rank Basic (escudos) Mozambique supplement Private 1 30 daily First Corporal 3 30 NCO 2600-3300 1800-2000 Lieutenant 4000-5200 2550 Major 8500 3650 monthly Colonel 11000 4450 General 14500 5700 1 Portugal and the Future (Lisbon 1Q74), Chapter 7. 2 Speech to the ANP, 16 February 1974. 3 Addre-s to the National Assembly, 5 March 1974. 4 ",Manifesto of the Armed Forces Movement', 25 April 1974, p.l. 5 See the circular issued by the Gremio (Confederation) of the Construction Industry, 13 May 1974. 6 Report of the Mogadishu Conference, Mozambique Liberation, No. 59, June 1974. 7 cf. Mozamnbique Liberation, No. 58, April 1974; Frelimo dismissed Spinola's proposals as neo-colonialist, actually more restrictive than the policy of Caetano. 'There is no such thing as a democratic colonialism.' 8 South African Hansard, 20 September 1974. 9 Rand Daily Mad, 25 September 1Q74. 10 Given in full in English in the Rand Daily Mail, 9 September 1Q74.

Chapter 14 Sequel The document signed by Frelimo and the Portuguese Government representatives in Lusaka embodied more than a legal transfer of authority. Frelimo was to become the sole legitimate heir of Portugal's former dominion; and it was therefore doubly appropriate that the text ended with a comment on the past designed to justify that transition to the uncommitted populations of Mozambique and Portugal: Frelimo, 'during the course of its struggle, has always distinguished the deposed colonialist regime from the Portuguese people and the Portuguese State'. These words, the presence of the army, continued Portuguese subsidies, and the completion of Cabora Bassa, were not sufficient to allay frightened or hostile reactions. Combined peace-keeping operations by Portuguese and Frelimo troops had already been set in motion to suppress banditry in the north, where marauding gangs had taken advantage of the breakdown of civil government. But a graver consequence of the Lusaka Agreement occurred in the south. Soon after Spinola's announcement of eventual independence, in July, disparate white groups, centred on Fico, began to organize opposition to what they regarded as a metropolitan betrayal of their interests. Since the democratic parties had by then pledged support to Frelimo, the white community split irrevocably; and this new opposition, lacking any hold on the African population, apart from the handful of Frelimo renegades, turned to violent measures. Rumours of plans to raise an army to smash Frelimo abounded, without answers as to how a few mercenaries could succeed where more than 60,000

CABORA BASSA 336 troops had failed. The 'Night Owl', Captain Gonqalo Fevreiro, broadcast appeals for former commandos to re-enlist, right-wing extremists made attempts to buy Jardim's government-sequestered property, the influential Noticias da Beira. This heterogeneous group represented the residue of those who could not stomach black majority rule, especially by an outspokenly socialist party, and it drew for propaganda on the lurid descriptions of Frelimo as extremist ideologues out to smash bourgeois society which were regularly put out by the Lourenqo Marques radio station, then controlled, apparently, by a white faction dedicated to a brand of Marxism far to the left of Frelimo's actual thinking. This, and the news that a settlement had been foisted on them, drove members of Fico to take over the Matola transmitter and broadcast their own summons to arms. So incompetent and clearly unplanned a counter-revolution scarcely merited a bloody response; and the Portuguese authorities and Frelimo responded with remarkable tact, while presenting an unwavering front against the rebels. The army stood by for three days, unwilling to move in because of the hundreds of women and children with whom the Matola station had been surrounded; President Spinola's personal emissaries arrived, to convey total repudiation of the rebels' action; and the South African Government carefully dissociated itself from any sympathy. But violence broke out in the vast African suburbs around Lourenqo Marques, and rioting and slaughter, absent from the city throughout the ten years of war, revealed in one night the true extent of Frelimo support. Admiral Vitor Crespo, the High Commissioner, ordered the white revolt to be suppressed at once, but it took all the energies of the new Transitional Government, led by the Frelimo Prime Minister, Joaquim Chissano, to restore a measure of order. Through September and October acts of racial violence occurred with ominous frequency, often as a result of the bravado of white commandos left behind because the Lisbon Government was not keen to repatriate them. A further wave of murder and arson brought terror to the streets of the capital on the nights of 21- 23 October. New rumours linked Fico exiles with the ultra-right-wing Herstigte

Nasionale Party in South Africa. But the commitment of the Portuguese army to defend Frelimo's authority under the High Commissioner, being absolute, left the die-hards no remedy other than guerilla action, and in time they faded away. In Lisbon, the agreement seemed a Phyrric victory for the Junta, as if it had been too great a sacrifice to the principles of the AFM. Spinola had decided that, despite the riots in Luanda, Angola should be dealt with differently. Still seeking a means to end the empire which would satisfy his federal conception, he made his position clear at the time of Guinea-Bissau's independence on 11 September. In a sour and disillusioned broadcast, referring to what had happened in Mozambique, he inveighed against the totalitarianism of the political left; later it was announced that he personally would handle all negotiations for the future of Angola. Spinola had, after all, argued in Portugal and the Future that if the overseas populations had a free choice, made in full understanding of the consequences, they would desire a form of political association with Portugal because of its inherent advantages. What had been denied to Mozambicans would be defended for Angolans - hence the two-year programme, offering, not a transfer of power, but the incorporation after a cease-fire of the three movements, MPLA, FNLA, and Uniti, as political parties inside the country. A controlled passage to independence, based on the relative weakness of the liberation movements, whose power had declined since the late 1960s, appeared to satisfy many of the competing claimants for Portugal's richest territory; as well as the vastly greater white population (600,000, mainly settlers, as against the quarter million of Mozambique), Portuguese and South African commercial interests in its oil, coffee, and diamonds; and South Africa's need for time to settle the future of Namibia before the northern border became indefensible. It also met the requirements of Zaire and Zambia for uninterrupted use of the Benguella Railway, and security against aggression from their own discontented minorities, which might have sought a base in exile in northern and eastern Angola. United States concern for the security both of Portugal in Nato and of the South Atlantic showed in these months. Sequel 337

338 CABORA BASSA Shortly after the revolution in Lisbon, it seems that Portugal was placed on a sort of conditional probation' and not until Spinola's conference with President Nixon, at the Azores base in June, did Dr Kissinger, Nixon's national security adviser, and the secret Forty Committee appear satisfied with the Junta's loyalty to Nato or its capacity to resist the spread of the Communist Party. (Similar reassurances were again required, after Spinola's fall, during President Gomes's visit to Washington at the end of October). It is in the context of strategic, rather than commercial, guarantees for the future of Angola and the Cape Verde and Azores islands, that Spinola's highly secret visit to meet President Mobutu of Zaire in September must be seen. Spinola hoped that the United Nations would accept his two-year plan, and the visit of the UN Secretary-General encouraged the view that delay in Angola would be no bar to Portugal's steady return to international respectability. The AFM and the Provisional Government had no objection in principle: but they disliked the nature of the proposed deal and the requirements of American diplomacy; and feared the concomitant revival of the right as a political force inside Portugal. Scattered indications of American involvement, such as the visit of Irving Brown, a leading member of AFL/CIO (the American equivalent of the TUC) and as fervent an anti-Communist as its head, George Meany, and the action of some American firms in dismissing striking work people, or repatriating assets and investments, encouraged conspiracy theories about CIA activity. But these were nothing to the evidence of a resurgent right-wing which, under the colour of Spinola's evident favour, and an appeal to the 'uncommitted masses', arranged a huge demonstration for the end of September. If Spinola had intended to move against the AFM, dismiss Gonqalvez and Copcon, the AFM's Praetorian Guard, and rely on the moderate officers and a programme of more cautious reform, he was undone by the failure of the right to evolve as a political party. Instead of donning Christian Democrat clothes, with parliamentary leaders drawn from the former opposition to offset the members of the old guard, the right chose to appear exactly as it had done under the previous regime: its populist appeal couched in phrases reminiscent of Caetano's 'fireside chats', and its leaders the same generals, bankers, and industrialists, who had forced Caetano to dismiss Spinola in February. To the AFM, this signified the end of the Manifesto, of their revolution, and of the life-giving role of the movement. In the subsequent trial of strength, their diagnosis made irrelevant whether or not there was an actual plot to shoot Gonqalvez, or subvert the Provisional Government. Spinola himself was asked to ban the right-wing rally. Instead he ordered Gonqalvez' arrest. But in the Palace of San Bento, on the night of 27-28 September, Otelo Carvalho's command of the Lisbon garrison stood for effective power; and the Chief of Staff, General Costa Gomes, fearing a dictatorship of the left, or right, if he acted wrongly, sided with the AFM. Faced with a demand to dismiss his entourage, Spinola chose to resign. Gomes succeeded him, and a wave of arrests brought down the remaining generals, politicians, and servants of the old regime, the ANP and the Portuguese Legion. Ironically, Spinola had told cadets graduating from the Military Academy on 6 July, 'In the last analysis, Portugal will be what her Armed Forces are.' When he stood down the Junta was purged, and the AFM confirmed in its special role as embodiment of the national spirit; with a superior Armed Forces Council, giving it power over the new President. Costa Gomes, wiser and more flexible than his predecessor, resumed progress towards the elections for a Constituent Assembly to be held in March 1975. Cabora Bassa's future was also settled in the wake of the Lusaka Agreement. Nothing in Frelimo's attitude had changed during the summer; and in the last months Frelimo local commanders showed themselves actually helpful. Construction was not interrupted by the troubles in Lourenqo Marques and when, in September, the workmen again struck for higher pay, Frelimo representatives ordered them back with the rebuke that they already formed a highly paid, privileged minority in the new state. As the gates of the south diversion tunnel were finally closed, on 21 November, to start filling the newest lake Sequel 339

340 CABORA BASSA in Africa, Chissano and Admiral Crespo came up to Songo. Chissano announced that the sale of energy to South Africa would go ahead, but on terms more advantageous to Mozambique. The fear of the engineers that the dam might be left an uncompleted ruin were at last relieved. At the time of writing, final details had not been negotiated either with South Africa or with Portugal, but it seemed virtually certain that Cabora Bassa would be nationalized, so that its energy could be seen to work for the people of Mozambique, and so that the sale to South Africa could be justified by its return in foreign exchange. For Portugal, such terms may prove an unconscionable burden, if the whole, or a substantial part, of the revenue is not applied to amortizing Cabora Bassa's cost. Ways may be found to define the debt as part of the Mozambique National Debt, or alternatively, given the evident goodwill of the United Nations, some form of retrospective aid may be sought from the World Bank or IBRD. If not, the worst fears of the Portuguese Treasury will have been realized. A clear line of development links Cabora Bassa with the decline and extinction of the Portuguese empire. Once entitled, in an Overseas Ministry brochure, 'a Reality set in the Heart of Africa', the project was deliberately made a symbol of empire and imperial priorities, the most obvious proof that in the 1970s Portugal retained a mission in Africa. In this sense, the gloss on recent history embodied in the Lusaka Agreement, differentiating between the errors of the fascist regime and the Portuguese State itself, explained a great deal. For forty-four years, since the Colonial Act of 1930, the two had been indivisible, with the consequence that the overseas problem - conversely, the 'dream of empire' - had constituted an historic alibi, excusing the Governments of Salazar and Caetano from embarking on adequate reforms at home. Salazar and the contending parties in the Council of Ministers' discussions in 1966 (see p. 37) had all accepted that Cabora Bassa would not only be a monument to the Estado Novo but also the test case of Portuguese imperialism. In that debate the integrationists, the expansionists, and the proEuropeans, used the project to embody their views of the

Sequel 341 future. As they clearly understood, Cabora Bassa would express to the world at large Portugal's energies and aims in a way that the dreary stalemate of the wars could not; it would stand out more publicly than the prosperity of Angola; and it promised a greater return, through the self-generating development of the Zambesi basin, than any other possible act. Cabora Bassa was never a 'normal' development project for this reason (although it serves as a valuable case study in the politics of development and engineering). Its architects had a vision common to their nineteenth century counterparts. As the American engineer, William Wheelwright, wrote to his English partner, Thomas Brassey, in 1870, about the building of the Central Argentine Railway: ... You are conferring upon that vast and prosperous country such infinite benefits, and to you will belong the credit, ... We go into the heart of a country abounding in wealth, possessing a soil and climate unsurpassed. Some there are who will urge the political instability of the country, but in spite of revolutions and dissensions, it has gone ahead in the most rapid manner. The European element of population will in a few years exceed the natives; these are the best tests, for who would go if prosperity were insecure?2 But those who argued against Salazar's case (that 'if we do not build it now, we will never build it') understood the constraints into which Portugal was forced as a result. Afterwards, like men building a house with limited means in an era of inflation, the Government realized as the walls rose that they must liquidate their reserves; with the roof timbers on, they drifted into debt; and, as the last tile was hung, faced bankruptcy. The project profoundly affected the timing and manner of imperial decline. Once it attracted the attention of Frelimo, as an alternative to the half-forgotten war in north Mozambique, Portugal had to sustain a war on two fronts, and, when the propaganda campaign began, greater unpopularity in Europe and at the United Nations. Political isolation encouraged the fortress mentality, just as Frelimo attacks made of Tete province a series of armed camps. Concentration on holding CABORA BASSA 342 the dam secure allowed deep penetration by Frelimo into the white settler areas; while the new strategy of psychological warfare imposed on the army the impossible duty of pacification. Cabora Bassa did not cause the revolution, but it contributed; had it never been built, the empire might have survived longer, giving time for Caetano's plans for economic disengagement to work. Without it, Frelimo might have been contained in the north, and Tete held with Rhodesian help, thus preventing subversion in Manica and Sofala. Equally, if Portugal had not committed itself to the deadline of finishing Cabora Bassa in 1974-5, a gradual running down of investment and military expenditure - the plan of the pro-EEC members of the Government - might have been made easier. The remorseless logic of defending everything to the last minute, which accelerated the trends leading towards revolution, had been evident six years earlier - hence Caetano's prolonged investigation of Cabora Bassa and imperial policy in 1968-9. The decision then to build the dam tied the two together, irrevocably, in a gamble which, in spite of extraordinary efforts and expenditure, failed. What remains for Mozambique? The historian, even of contemporary events, should have no part in forecasting the future. But some current trends can be suggested. A liquidity crisis, bounding inflation, strikes, and racial troubles, confronted Admiral Crespo, Frelimo, and the Transitional Government. The immediate crisis could be resolved by a moratorium on interest payments, and by further external loans. In November 1974, the South African Government responded promptly to a request for emergency wheat supplies, a carefully calculated gesture of neighbourly goodwill, while Portugal's own economic problems were too grave to do more than complete existing commitments for 1974-5. If Frelimo wishes to increase the external debt, Mozambique remains fundamentally credit-worthy, her debt repayment ratio better than that of neighbouring Zambia. Revenue, however, suffered severely from the re-routing of Rhodesian rail traffic and the decline in tourism - once worth £20 million a year. Much worse, in spite of the conciliatory approaches, even blandishments, of Frelimo, and the

Sequel 343 Portuguese ban on removal of assets, the drain of white emigration did not stop. During the riots of October, after the earlier open-door policy had been abandoned, the South African consulate was besieged by thousands seeking legal entry to the Republic. The white population had been reduced to approximately 190,000 as early as August 1974; nearly 15,000 entered South Africa in the next two months, and as many returned to Portugal. Probably less than 100,000 - the real Mozambicans - will remain. The emigrants sensed, correctly, that the comfortable life, the classic cycle of overseas service followed by retirement in the metropolis, died with the Lusaka Agreement. Samora Machel's message,3 which was delivered at the investiture of the Transitional Government on 20 September, predicted a lifetime of austerity and hard work, with no privilege for any group: in the multi-racial society, whites, Asians, and mulattos, had to give up 'paternalism and their pretensions to superiority'. After reviewing the past, and the legacy of colonialism (predictably to the detriment of the former regime) he made clear that in achieving economic independence 'the watchword is work and self-sacrifice.' Frelimo's priorities lay in raising the standard of living of the poorest, their nutrition, housing, and health. The imbalance of urban prosperity and rural poverty would be remedied: 'It was the countryside that most suffered from the destructive effects of war, and it is in the countryside that our main efforts to improve the condition of the masses will be directed.' On one hand, the development of agriculture and husbandry, on the other, an attack on conspicuous spending and the corruption of city life, its vice, prostitution, and 'alienating foreign influences'. Justice and judicial process were to be simplified and speeded up, and new structures for labour relations evolved - not, however, to bargain for wage increases, since these would be inflationary, but to achieve increased production and productivity. Behind Frelimo's manifest puritanism and the thrust for social and sexual equality can be seen the guiding experiences of Maoist teaching, combined with the practical expression of Tanzanian socialism. The sections dealing with the party made clear that Mozambique was to be a democracy without benefit of a multi- party system. Machel urged the need to form party

CABORA BASSA 344 institutions, and inculcate collective decision-making, under the Frelimo charter. 'State power has been won through the struggle of our people, united by our correct line, under Frelimo leadership.' The only way to genuine tolerance and discussion lay through the party itself. The leadership was to be subject to far more stringent safeguards against corruption than in the United National Independence Party (Unip), or even Tanu; and something close to a continual 'cultural revolution' was proposed, to safeguard against the great evils 'tribalism, regionalism, racism and unprincipled alliances'. A permanent enemy, or scapegoat, offered itself in the twin bogies of 'colonialist reaction and sabotage', and the 'reactionary position' of the ultra-left. It is questionable how much of this programme can be effected in a mixed economy; whether the modest role assigned to industry, mining, and the search for foreign capital and technology, can be equated with the heavy long-term capital requirements of agriculture, or even maintain the level achieved by Portugal in the last five years; whether white administrators, technicians, and managers, whose skills are desperately needed, at least for the next decade, will remain in a state where luxury goods and a Western standard of living are deprecated. Frelimo's economic precepts, deriving from its Tanzanian experience, have to be imposed on a country with considerable natural resources, unlike Tanzania, and it is probable that wide compromises will be made, if only to win sufficient prosperity for political growth. The path may lie somewhere between uncritical bargains with foreign lenders or multi-national corporations, and Nyerere's doctrine enshrined in the Arusha Declaration: 'The policy of inviting a chain of capitalists to come and establish industries might succeed in giving us the industries we need, but it would also succeed in preventing the establishment of Socialism.' If the view of Frelimo given earlier is correct, its leaders are at least as strongly nationalist as socialist, bound by the immediate interests of Mozambique, rather than to a supranational dogma. Given the history of their pragmatic attitude to Cabora Bassa, there is no reason to think that Mozambique will not benefit from the technology and resources of JCI, Anglo-American, or any other South African organization, and from the investment policies of Portuguese or European companies, on conditions that will be regarded as fair and, within the definition of the Lusaka Agreement, 'in the interests of Mozambique'. Then, over time (some Portuguese bankers expect no more than three years) or in the original concessions and agreements, Frelimo will ensure a majority holding for the state. Mozambique has been less impoverished by recent trends than many Third World countries. Unlike Tanzania, Ghana, or Kenya, there are reserves of natural gas and probably sufficient oil for local needs, vast quantities of coal for export, as well as to fuel its railways, and almost unlimited cheap electricity. Nevertheless, it is a relatively poor country. Although its people rarely starve, the gap between it and more developed nations widens constantly. Will Mozambique always be dependent on the developed world for aid and markets? If the plans left by Portugal - itself a poor nation - are fulfilled, there is no reason why the country should not pass on from its temporary phase of import substitution (largely a failure because of dependence on external capital and technology) to develop labour-intensive manufactured exports, the critical path to prosperity of other relatively small developing nations.t Self-sufficiency in agriculture, with a specialized exporting section, mineral extraction and processing, perhaps at a Tete steel factory and aluminium plant, should not be impossible to achieve, while an able, educated nation of nine million can surely benefit from its natural advantages in cheap power and sound communications facilities. These attainments may lie ten years in the future. Mozambique is short of managerial and technical skills, and can benefit, not so much from Portugal, where skills are equally in demand, but the EEC, possibly under Yaounde preferences, and perhaps from Brazil, whose interest in investments and reciprocal trade showdd clearly in October 1973, in the economic mission to Africa led by the Foreign Minister, Senor Mario Barbosa. At that time, interest concentrated on Angolan and Nigerian oil, but given the rate of expansion in Brazil, a far wider African market may t For further amplification see Notes at end of chapter. Sequel 345 be needed.* Mozambique should be able, within the Portuguese language zone, and perhaps on a basis of reciprocal citizenship (already extended to Portugal), to draw on Brazilian investment, and technical and managerial expertise. It is not inconceivable that an escudo zone could be formed, on a basis similar to the Commonwealth preference scheme which operated after the Ottawa Conference in 1932. Multi-national companies may also be able to help, through processing, assembly, and component manufacturing activities within vertically-integrated international industries (in the same way that American firms have created enclaves elsewhere). In African circumstances, safeguards could be constructed to make such development acceptable; the gains in employment, 'learning', national income, and foreign exchange, are clear. With South Africa on its borders, ready to offer capital investment, technology, and managerial skills, Mozambique may be considered fortunate - if it wishes to take advantage, rather than rely on Russian and Chinese friends. The obvious political objections may be waived or overcome in the manner of the Cabora Bassa contract. Frelimo and the South African Government appear to have reached a working agreement on external relations. Although the doctrine of non- interference may not safeguard the interests of Western powers concerned at the build-up of Russian naval forces in the Indian Ocean or Chinese involvement on the African mainland, nevertheless the least price which China could exact for her long support of Frelimo is that Russia, her great rival, should not be allowed naval bases in Mozambique; and there is no reason to suppose that Frelimo will not follow Tanzania's non-aligned example. In that case, South Africa may be able to offer economic cooperation, and Mozambique to swallow it, if the present trends in the Republic continue. The moral sticking-point of contract labour may shortly be eliminated at South Africa's own volition: in the wake of Malawian restrictions, trouble with Lesotho, and the Shangaan demands for higher wages * That is assuming that there is more to the Brazilian connection than a vehicle for the United States' strategic interests in the safeguarding of the South Atlantic and Indian Ocean. 346 CABORA BASSA and better conditions, which erupted without political directives from Frelimo, the South African Chamber of Mines began an intensive recruiting campaign at the end of 1974 to free the gold mines from dependence on outside recruitment. However much the new wages offered (41.60 Rand a month minimum, three times the rate of 1973) will affect levels elsewhere in South Africa, to say nothing of the possibilities for effective black trade-unionism, Frelimo's dilemma stands to be resolved at the price of social disruption and rural unemployment in the Shangaan homeland. The price of other links, such as the Lourenqo Marques railway and port facilities, can be renegotiated. It must be expected that foreign- owned shipping and freight services will be nationalized; their value will in any casediminish with the development of Richard's Bay. Recently in South Africa Frelimo has been depicted as a movement headed by ardent nationalists, rather than Russian or Chinese-dominated ideologues. Only two months after the Portuguese revolution, the influential Institute of International Affairs pointed out that the transfer of power in Mozambique proved that the Republic could no longer rely on military power to preserve its special position. 'Now that the so-called "buffers" on our flanks are being removed, there are in fact no borders; this type of conflict takes place withzin this country, and the whole country is involved.' If the remorseless logic of apartheid is pursued, infiltration will as easily take place from the independent Bantustans, in the later 1970s, as from beyond the Limpopo. Hence the question posed by the South African strategist, Professor B.J. Pick: 'Will South Africans realise that they are part of a black continent, and that their future depends on their ability to fit into black Africa?'5 That great question cannot be answered by reference only to Mozambique. But in spite of the suppression of proFrelimo rallies, it is already clear that the pace of change within South Africa has accelerated. Vorster's conferences with Bantustan leaders in 1973-4 were followed by the nonintervention policy for Mozambique, plans for a new order in Namibia, and conciliatory approaches to Zambia - all in the shadow of the climactic United Nations debate on South African membership which, on 23 October 1974, elicited from Sequel 347

Vorster his remarkable plea to world opinion for six months in which to achieve astonishing change. Between the bland public assurances and the realities of domestic politics, change may peter out; but Vorster's insistence on the offer of development capital and technical assistance to other African countries as a means of stabilizing relations, confirms that the trend noted earlier, towards a new outlook on the whole African continent, remains significant.* Looking back from the end of 1974, it could be seen that, by bringing to a head as early as 1965 the fundamental differences between verkrampte and verligte assumptions about the external policy of the Republic, Cabora Bassa had imposed a certain constraint on the evolution of South African politics. Subsequently, events in Mozambique acted as a catalyst of further change, both attracting commercial investment to the province, and eroding the rigid pattern of strategic planning with Portugal and Rhodesia. As the great strike in Durban in 1973 worked to ameliorate wage levels and attitudes towards apartheid, the fall of the Portuguese empire accelerated revisionism in South Africa to the point where Government thinking ran far ahead of the consent of its Afrikaner electorate. Of equal importance, the sense of common interest shown by their participation in the Lusaka negotiations may lead the states of middle Africa towards the federation long advocated by Nyerere. 'The independent African states need to speak with one voice in order not to be disregarded by the great powers,' he told an Australian audience in March 1974; 'Fortytwo states [in the OAU] can only meet rarely, and when they do, our separate sovereignty means that unanimity is required in conference, and then forty-two separate actions are necessary to make the unanimity effective.'6 The past record of federalism is scarcely more heartening than that of PanAfricanism: Tanzania's federation with Zanzibar has * From the latest information available, it appears that South Africa is now prepared to provide loan finance and expertise to fulfil a large part of the remaining development of the Zambesi basin, according to the guide-lines of the Plan Geral: surveys have already been made of the lower barrages and the areas expected to benefit by irrigation. Willingness to provide capital for such very long-term projects indicates a substantial degree of faith in mutual cooperation with Mozambique. CABORA BASSA 348 existed in name only. Economic unions, such as the East African Community, have not fared much better. But among states with little national cohesion, shared independence struggles, common political structures, the support of powerful friends, complementary natural resources and communications, and a common external enemy, there would be grounds for an affinity grouping in which process Frelimo, linking Mozambique and Tanzania, may be the catalyst. The five countries, Tanzania, Zambia, Malawi, Mozambique, and Zaire, share or will share a one-party system, a victory over colonialism, the moral and material backing of China - to whose long-term interests such a grouping can only be advantageoust - and a common target of hostility, the white regime in Rhodesia. They may soon be joined by Angola. The waterways of the Congo, Cuango, Zambesi, and Limpopo, the Benguella, Tanzam, Nacala, Beira and Lourenqo Marques railways, and the hydro-electric schemes of Inge, Cabora Bassa, Kariba, and Cunene, all serve or will serve more than one nation; and if Rhodesia were to become Zimbabwe, and if Botswana and Namibia and the Bantustans were to join, something of the Pan-African dream might yet be made good. Five years ago, opponents of apartheid feared that South Africa might impose economic hegemony and call it cooperation; now there is no reason why black Africa should not deal equally with white, in mutual self-interest. Months before the dramatic events of December 1974, it could be seen that, as a result of the collapse of Portuguese power in Mozambique, Rhodesia would survive uneasily, abandoned politically and strategically by South Africa, and increasingly beleagured by Zanu and Zapu guerillas, supported by the common front of middle African states. Frelimo leaders admitted, prior to the Lusaka Agreement, that, having always respected and appealed to the United Nations, they were morally bound to follow its mandate. The cost to Mozambique revenues of imposing sanctions could have been offset by international action, as Zambia was compensated for the border closure in 1972. In spite of the new railway t For further amplification see Notes at end of chapter. Sequel 349 link, hastily built to Beit Bridge on the South African border, Rhodesia's economy seemed unlikely to survive the impact of action by Frelimo, especially if that were combined with pressure from Botswana (which had recently threatened to restrict the Francistown link to South Africa), because even if the link to Beit Bridge were doubled, South African Railways could only have coped at crippling cost to the South African economy. It was hardly surprising that the Nationalists, the United Party, Progressives and trade unionists in South Africa should have concurred in stating that a solution was long overdue. The sombre reality behind the constitutional negotiations early in 1975 was that if they failed, and if Frelimo extended cooperation with Zanu to include bases along the 400 miles of Rhodesia's eastern frontier, thus opening up the Lowveldt and the Shangaan territory to infiltration and the Beit Bridge link itself to sabotage, the end of the Smith regime would come in months rather than years. Why should Frelimo not have acquired merit in the OAU by training and fitting out Zimbabwean guerillas, as Tanzanian and Chinese instructors had done for them? The Rhodesia Government could hardly strike at the transmission line, for fear of alienating South Africa; and her own military strategists, before risking outright hostilities against a Mozambique army, equipped with SAM ground-to-air missiles and backed by the resources of the Afro-Asian world, might have reflected on the dismal example of the Italian army, defeated by Abyssinia at Adowa in 1896. Does anything remain of Portugal's dream of empire? The Caetano Government encountered forces similar to those which broke up the French, British, and Dutch empires, the emergence of popular revolt, aided by external enemies, the feebleness of nominal allies, economic attrition, and a decline of willpower; after vast sacrifices over more than a decade, leadership passed to men who willed otherwise. Freed of the colonial dilemma, Portugal was suddenly reduced to her true dimension - ironic comment on those Lisbon posters which had once overlaid the map of Europe with the empire, stretching from Madrid to Warsaw, Portugal en su verdadeira iimensdo. The choice is bleak: a future as a poor country on 350 CABORA BASSA the fringe of the EEC; subordinate, possibly, as Franco Nogueira used to warn, to the ancient rival Spain, with no compensating excuse. In this context, economic progresst offers the only escape from a deepening division between the centre parties (Socialists and Popular Democrats) and the Communist Party, each fighting for the allegiance of the Armed Forces Movement. If industrial investment recovers, however, if the Government emerges from the elections able to hold to a reasoned policy, balancing full employment and growth against an acceptable level of inflation, and if the army retires, eventually, to barracks, Portugal may yet surprise the world. Whatever happens, a profound intellectual and social upheaval has to be surmounted. The loss of India, Algeria, and the Dutch East Indies, all involved the imperial powers in social and politcal upheaval; after five hundred years of involvement in Africa, some lesions must be expected in Portugal, although war- weariness may have prepared her people for the traumatic shock of independence. There will be particular problems with the reintegration of expatriates, many of them born overseas; the rewards of decolonization may be unemployment and social tension. Savings on the war budget are not likely to accrue for two years. There has also been an absolute loss of importance on the world stage; as Spinola predicted, Portugal will soon lose her special significance, based as it was on the value of her territories to South Africa, Britain, France, and the United States. It is far too soon to make historical judgements about Portuguese imperialism. Under hostile criticism from nominal allies as well as attacks by open enemies, Portugal made a courageous, if misguided, attempt to fulfil the colonial policy of Salazar, as reinterpreted by Caetano; and, in the process, brought Angola and Mozambique close to economic and political statehood. Ironically, a great part of the revenue drawn out of Africa by successive imperial governments was, at the last moment, and at monumental cost to metropolitan development, restored. The assertion that great areas of Africa constituted an integral part of Portugal was admitted to have t For further amplification see Notes at end of chapter. Sequel 351 been misconceived. But it is too early to test, under free conditions, the validity of Portuguese claims to have created genuine multi-racialism in Africa. Developments in the past year in Luanda and Lourenqo Marques augur badly, but not yet conclusively. Other claims for the Portuguese inheritance rest on vaguer quantities. It may be that traditions, dating from long before the Colonial Act and the period of enforced assimilation under Salazar, encouraged a blend of culture on the basis of a common language. For three centuries before its absorbtion into India Goa was such a case, and the existence of Goans seemed in Portugal proof that miscegenation and multi-racialism formed two sides of the same coin. Gifted Frelimo poets such as Jose Craveirinha and other Mozambicans like Noemio de Sousa and Kalungano7 writing in Portuguese may one day be read in Portugal as West African authors are in England. On the other hand, PAIGC has already rejected the Portuguese language, and its retention in Mozambique and Angola, in an otherwise English or French-speaking continent, may not survive the second generation of African leaders, now growing up, who were never themselves assimilados, nor educated in Portuguese schools and universities. If the legal system is dismissed as archaic and too cumbrous for modern usage, if the Church is disestablished, and the language debilitated, then imperial Portugal may have no memorial other than the line of grey, ruined castles along the Indian Ocean, where the navigators first halted in the fifteenth and sixteenth centuries, and a twentieth-century epitaph at Cabora Bassa. CABORA BASSA 352

Notes and references Chapter 14 Sequel 1 Cf. Sunday Times, 27 October 1974, quoting the Washington columnists, Rowland Evans and Robert Novak: ' .. The depth of anxiety over Portugal's sharp move to the left can be measured by the fact that the US has cut Portugal off from certain highly classified military and nuclear information commonly available to all members of NATO.' 2 Sir Arthur Helps, Life and Labours of Thomas Brassey (London 1872) p. 367. 3 English translation, SAIIA pamphlet, October 1974. f Note page 345. Cf. G.K. Helleiner, Manufactured exports from less Developed Countries and Multi- National Corporations (IDC, Sussex University, March-April 1972). Also Pauuw and Fei's study of Taiwan in The Transition in Open Dualistic Economies (London 1972), which shows that a switch from primary production to labour- intensive exports was facilitated by rapid agricultural productivity growth during a phase of import substitution; both being factors present, at least to some extent, in Mozambique. 4 SAIIA, Nev,,sletter, 6, No. 2, June 1974, p. 11. 5 Ibid, p. 4. 6 Times of Zambia, 24 March 1974. t Note page 349. In January 1974, China granted Zaire a £42 million interest-free loan, as part of the Third World diplomatic offensive. t Note page 351. By the end of 1974, inflation had fallen to 25 per cent a year; the balance of payments had improved, as emigrants' remittances and tourism revived; and Portugal suffered a deficit of £400 million - modest, considering the cost of oil imports. With gold reserves standing at a free market value of near £2,000 million, and the possibility of holding inflation at roughly 20 per cent, the future is not desperate; but in the present climate, external investment has not yet been restored. 7 See Mario de Andrade, Antologia da Poesia Negroae da Expressao Portuguesa (Paris 1958).

Sources and references Interviews PORTUGUESE GOVERNMENT DEPARTMENTS: Overseas Ministry, Foreign Ministry, Ministry of Defence, Treasury. PORTUGUESE FORMER MINISTERS: Opposition leaders; members of the armed forces, of all ranks, in Portugal and Mozambique. MOZAMBIQUE GOVERNMENT: Governors-General, Provincial Secretaries, officials of the Cabinete do Plano do Zambese, Banco Nacional Ultramarino, Instituto do Credito, and of various private banks. DIPLOMATIC SOURCES: Embassy and High Commission staffs of various nationalities in several African states, Rhodesia, and South Africa; officials of the United Nations. INDUSTRIAL AND FINANCIAL SOURCES: Executives of the French, German and Italian companies inside Zamco, and Zamco project management. Members of the British and American consortia tendering for Cabora Bassa; executives of the Anglo-American Corporation, and its subsidiaries; of Johannesburg Consolidated Investments; of Escom (South Africa); of the Industrial Development Corporation (South Africa); of various commercial, financial, and industrial concerns in Mozambique, Zambia, Malawi, Tanzania, Rhodesia, South Africa; of banks and finance houses in Lisbon, Johannesburg, London, and Paris, and of the World Bank, and development agencies. POLITICAL SOURCES: party officials, ministers, deputies in Zambia, Tanzania, Malawi, Rhodesia, South Africa.

Sources and references MEMBERS OF THE CENTRAL COMMITTEE OF FRELIMO: leaders of Coremo, Zanu, and officials of the Liberation Centre in Lusaka. Portuguese exiles in London. Headmen and tribal elders in various villages and aldeamentos in TFete province. Indunas of the Shangaan contingent at Cabora Bassa. Official documents Publications of the Ministero do Ultramar and the Agencia-Geral do Ultramar, Lisbon. Reports, censuses, and statistics of the Governments of Mozambique and its neighbours. Periodicals PUBLICATIONS OF: Africa Bureau; Africa Institute (Pretoria); Institute of Strategic Studies; South African Institute of Race Relations; South African Institute of International Affairs. Africa; Africa Confidential; African Development; Anti-Apartheid News; Boletim Sociologico do Estado da Mocambique; Economia da Moambique; Jeune Afrique; Guerrilhero; Mozambique Liberation; Sechaba; Tempo (Lourenqo Marques). Newspapers British, French, German, Portuguese, Mozambican, South FRICAN, Zambian, Rhodesian, Malawian, Tanzanian. Books Barnett, Donald, Revolution in Angola, New York 1972. Boavida, Amerigo, Cinco Seculos de Exploitacao Portuguesa, Rio 1967. Bosgra, S.J. and Van Kempen, C., Portugal and Nato, Amsterdam 1971. Carter, Gwendolin, African One-Party States, Cornell 1962. British Council of Churches, Violence in Southern Africa, London 1971. 355

Clements, Frank, The Mighty God, The Story of Kariba, London 1961. Cornwall, Barbara, The Bush Rebels, London 1973. Coissoro, Narana, Politica Ultramar, Lisbon 1966. Cabral, Amilcar, Revolution in Guinea, London 1969. Cartwright, A.P., The IDC 1940- 70, Johannesburg 1971. Drury, Alan, A Very Strange Society, New York 1967. Duffy, James, Portuguese Africa, Harvard 1961. Portugal in Africa, London 1962. e FalcAo, A.F., Um caso sem paralelo no Ultramar Portugiiis, Lisbon 1961. da Sousa Ferreira, E., Portuguese Colonialism, Freiburg 1969. de Figueiredo, Antonio, Portugal and its Empire, London 1966. First, Ruth, Portugal's African Wars, London 1969. Gann, L.H. and Duignan, P., Colonialism in Africa 1870-1960, 2 vols, Oxford 1970. Guinde, Peter, Portugal and the EEC, Amsterdam 1973. Guerra, Eduardo, Evolucdo da Economia Portuguesa, Lisbon 1967. Hayter, Teresa, Aid as Imperialism, London 1971. Kay, Hugh, Salazar and Modern Portugal, London 1970. Kahl, J.F., Pro und Kontra Portugal, Stuttgart 1972. King Junr., J.A., Economic Development Projects and their Appraisal, Baltimore 1967. Leal, Cunha, A Patria em Perigo, Lisbon 1962. 0 Colonialismo dos Anticolonialistos, Lisbon 1961. Lichtheim, G., Imperialism, London 1971. Livingstone, David, The Zambesi Expedition of David Livingstone, 1858-64, 2 vols, London 1865. Lombard, J.A., Stadler, J.S., Van der Merwe, P.J., The Concept of V Economic Cooperation in Southern Africa, Pretoria 1968. Martins, Maria B., Sociedados e grupos em Portugal, Lisbon 1970. Marquard, Leo, A Federation of Southern Africa, Oxford 1971. The Peoples and Politics of Southern Africa, London 1960. Mondlane, Eduardo, The Struggle for Mozambique, London 1969. de Melo, M.J.H., Portugal, o Ultramar e o Futuro, Lisbon 1962. Mtshali, B.V., (ed.), Essays on Problems relating to Zambia's Foreign Policy, Lusaka 1970. Newton, Trevor, Cost Benefit Analysis in Public Administration, London 1972. Nogueira, Franco, The Third World, Johnson 1968. Nyerere, Julius, Freedom and Socialism, Dar es Salaam 1968. 356 CABORA BASSA

Sources and references Newitt, M.D.D., Portuguese Settlement on the Zambesi, London 1973. Ogunsamwo, A., China's Policy in ,Africa 1958-71, Cambridge 1974. Pintado, Xavier, Structure and Growth of the Portuguese Economy, (EFTA) Geneva 1964. Rubin, N., and Warren, W.M., Dams in Africa, New York 1968. Santareno, J.A.L. Martins, Agricultura Mocambicana, Lourenqo Marques 1973. Shamuyarira, N. (ed.), Essays on the Liberation of Southern Africa, Dar es Salaam 1972. Soares, Mario, Le Portugal Billonn , Paris 1969. Spinola, Antonio de, Portugal e o Futuro, Lisbon 1974. Spence, F., Mozambique, Lourenqo Marques 1965. dos Santos, \,. Pimental, Mozambique, Guerra Inutil, Paz Possivel, Lourenqo Marques 1971. Wheeler, D.L., and Pelissier, Rend, Angola, London 1971. Wilson, Harold, The Labour Government 1964- 70, London 1971. 357

Index South Afrca is indexed under RSA For meanings of acronyms used, the provide the title in full. Abreu, Jorge, 270, 328 AC and DC lines, 43-4, 46, 56-7, 169, 192-4, 203, 212, 282 Access road, difficulties in making, 45-6, 90-1, 96, 99, 120 Accident rate, 116; north surge chamber collapses 190 AEI International, 41, 45, 70 AFM (Lisbon), 201, 273, 297, 315, 319-22, 324-5, 328, 331, 339 African Camp, Songo, 117-18 Alcora Commission, 283-4 Almeida, Dr, 328 Alvoes, Baltasar, 328 Ameal, Joao, 132 Amenities, development of, at Songo, 113-15, 117-18 ANC (Rhodesia), 287, 301 Anglo-American Corporation, 31, 32, 42, 47, 51, 53-5, 75, 100, 104-5, 204, 228, 253-5, 344-5 Angola, 14-16, 23, 28, 51 Apartheid, 27-8, 50, 103, 116-17, 162, 164 Army unrest in Portugal, se under Portuguese Crisis, the Arouca, Domingos, 140, 265, 271 Arusha Declaration (1967). 137, 163, 274, 344 first page reference given will normally ASEA, 50, 52-3, 70, 73, 81-2, 99, 104, 203 Aveiro Conference (1973), 309 Banda, Dr Hastings, 34, 145, 152, 267, 271, 275-6, 295-6, 297, 299, 326 Bauxite, plans to exploit, 230 Beira, 12-13, 15-16, 27, 94, 156 Beira-Umtali Railway, Frelimo attack on (1974), 155, 286 Beit Bridge rail link, 331, 350 Bendixen, Bernard, 108, 191, 201 BICC, 45, 69 Biermann, Admiral, 287-8, 290, 330 Botha, (Defence Minister - RSA), 287, 290, 331 Boving Turbines, 45 Brandt, Willy, 165, 167 Brebant, Roger, 100 British Government's concern at Frelimo advance, 291-2 Brown, Esmond, 32, 51, 54, 77, 79-81, 113 Building of the Dam, 107-29: intercompany problems, 107-11; discipline, need for, 109, 118-19; development of Songo, 113-15;

CABORA BASSA salaries earned, 115; boredom, countering, 115-16; racial harmony, 116-17; accident rate, 116; hospital arrangements, 116-17; lingua franca, 117; the African camp, 117-18; African labour shows its superiority, 118; workings of hydro- electric barrage, 122-5; ancillary works completed 1972, 126; electro-mechanical side, 126-7; effect of Frelimo war on, 128-9 Cabora Bassa Builders, 49, 54, 61-2, 108 Cabora Bassa Dam: first site inspection of, 17; study, 1957-65, 19; first design of, 20; economic inducements to build, 23-24; 2nd inspection, 1965/66, 31, 46; preliminary political and fiscal matters concerning RSA, 24-36; four schemes studied (1966), 46; formal agreements (1967), 47; work starts, 66; completion of: 186-205; security precautions needed against Frelimo, 186-7; preliminary work concluded, 189; foundations finished, 189; start made on spillways, 189, north surge chamber collapses, 190; down-stream coffer dam removed, 191; machinery installed, 191-2; converter substation started, 191, 193; generators built in, 192; current begins to tlow, 193-4; Frelimo occupy south bank of Zambesi, 195; delays result in acceleration programme, 196-7; lake, final level of, 199; Lisbon Revolution, effects of, 199-200; south diversion tunnel closed, 201; lake begins to fill, 201; life span of, 212; nationalisation of, 340; see also, Building of the Dam; Contractors, politics of Cabral, Niorais, 15, 262 Caetano, Marcello, 74-6, 128, 143, 147, 156, 164, 235-6, 238, 245, 247-8, 259, 264, 272, 286, 290, 297-300, 310-312, 315, 317-18, 320 Canxixe, 227 CAPC, 96, 190, 282, 297 'Capricorn' Exiles, 296 Cardoso, Professor, 93 CCI, 52, 55-6, 77, 115, 198-9, 204 CDE, 270 Cement supplies, 93, 103-5, 121, 195, 197 CGEE, 52, 126, 204 Chambers, Bill, 100-2, 108-9, 114, 191 Champalimaud, Antonio, 104-5, 247, 264, 295, 324 Chappelle, Alexandre, 100, 108-9, 192 Chipembere, Harry, 275, 296 Chi Peng-fei, 302 Chissano, Joaquim, 201, 326, 336, 340 Chona, Mark, 298 Civil engineering, firms considered or taking part, 51-2 Coal-mining, importance of, in Mozambique, 227 Cogelex, 55, 78, 107, 192, 193 Coimbra University, 173 Collectivisation of population, 217-220, 222-3 Common Market, should South Africa have? 29 Comoxmin, 227-8 Compadec, 45, 49, 77 Concassa, 45-9, 54, 61- 4, 103-4 CONCP, 160, 162, 181, 284 Conscription in Portugal, 312-14 Consortium, the building of the, 41-64; first approaches, 41-3; AC/DC conflict, 43-5; consultants appointed, 45; agreement reached, 47; designs drawn up, 48, 69; Portuguese wish for three consortia, 51; civil engineering aspect, 51-2; electro-mechanical side, 52-3; financial cover, 54-6; local costs of wages and supplies, 360

Index 57, 61; specifications dispatched (1967) 60; tenders, 42, 47-8, 53-4, 61, 63-4; preliminary award goes to Zamco, 64 Consultants, list of, appointed, 45, 61 Contract, nature of, 50-1, 66, 73, 76; signing of, 80-1; see also, Specification; Tenders Contractors, politics of, 66-84; building starts before contract signed, 66; investments made by countries, 66-9; problems of co-ordination, 71-3; the finalised contract (1969), 73; effects of unrest in Portugal, 74-6; reappraisal of costs found necessary, 76-7; second bid tactics, 77; RSA, recession in, 78-9; ASEA withdraws from contract, 81-2; Germany creates problems, 83; GEC (USA) withdraws, 83-4 Co-ordination, problems of, 71-3 COPCON, 327-8 Coremo (Splinter group of Frelimo), 129, 136, 140, 146, 164, 272, 329 Coface (a firm), 66 Costa Gomes, General, 155, 295, 315, 317-20, 323, 325, 332, 339 Costain (a firm), 45 Council of Three (Salisbury - 1971), 283 Coyne-Bellier (Paris), 125 Craveirinha, Josd, 140 Crespo, Admiral Vitor, 336, 340 Cruz, Dr Freitas, 105 CUF, 104-5, 227, 229, 242 Cunha Leal, Sr, 308 Cunha, Silva, 25 28, 71, 74, 77, 81, 230, 236, 311, 315 da Cunha, Martin Diaz, 248, 264, 295 Dam wall, construction of, 96-7 d'Arriaga, General Kaulza, 128, 143, 145, 147, 151, 156, 175, 266, 290-1, 311, 316, 324 DC or AC lines, see AC and DC lines Debt, rates of interest for, 42-3 Decolonisation Committee, (U.N.), 174 de Kock, Wickus, 286 Delgado, General, 132, 263 Designs, 48, 69 DGS (Secret Police), 119, 128-9, 140, 148-9, 151, 154-5, 172, 187, 235, 265-6, 270, 272, 309, 314, 319, 325, 329, 331 Diederichs, Dr, 62, 69, 168, 249-50 Diseases, 89, 116, 221 d'Oliveira, Braz, 98, 110, 188, 197, 200, 205 dos Santos, Marcellino, 134, 142, 304 dos Santos, Pimental, 47, 62, 64, 155 Economic autonomy: progress towards, in Mozambique, 235-55; integration with Portugal? 235-6, 244; autonomy or independence? 236, 1930-50, 237-9; private enterprise, development of, 239; balance of payments, decline of 239-40; transition from Province to State, 240; EEC, Portuguese attempts to enter, effects of, 240; import control, 240-1; food and consumer goods, shortage of, 241; import substitution, 242; GNP, growth of, 242- 3; bank business, 243; national debt, 244, 248; UDI, 245, 276; investment from overseas, 246-8; plough-in of profits, 247-8; customs union with RSA? 249; water shortages in RSA, 250; technological aid, 251 Economic inducements to build Dam, 23-4 Educational standards in Mozambique, 220-1 EEC attitude to Portuguese Crisis, 330-31 Electro-mechanics, 52-3, 69-70, 126-7, 197-8 Enes, Antonio, 131 Enmoc, 92-3 Eppler, Erhard, 67

CABORA BASSA 362 Escom, 32-4, 46-8, 52-3, 58-60, 62, 67, 75, 78-80, 83, 198, 211-12, 214, 288 Ethnic groups in Mozambique, 260 Excavation of Dam site, 97 Export Credit Guarantee Department (London) (ECGD), 46, 48 Fanagalo, (lingua franca), 117-18 Fereira, Mario, 45, 51, 73, 77, 82 Fernandez, Max, 265 Ferolimo (Pry) Ltd., See SAE (Italy) Fico (political party), 325, 328, 336-7 Fontes, Dr Castro, 47, 71, 73 Forstmann, Wilhelm, 73 Foundations, completion of, 189 Frelimo, 23, 37, 81, 101, 117, 128, 131, 135-43, 145-6, 148-9, 154-5, 160-2, 164, 169-70, 172-4, 176, 299; Congress of (Dar es Salaam 1972), 153; see also, Guerillas, menace of Frelimo/Zimbabwe alliance, 284 French policy, 292-3 Friedrich Ebert Peace Foundation, funds from, 167 Front for the Liberation of Mozambique, see Frelimo GEC, (USA), 41, 48, 82, 84, 203 General Trade Company of Geneva, 41-2, 47-8, 60 German politics, effect of, 83-4, 293 Gibb, Sir Alexander, and Co, Ltd, 45 Gonon, Jean, 77 GPZ, 47, 64, 71, 90, 98, 110, 129, 150, 210, 212, 214, 217-19, 231 Graaf, Sir de Villiers, 34 GTM, 52, 77-8, 108, 204 Guerillas, menace of, 89, 91, 93; see also Frelimo Gumane, Paulo, 136 Gumo (political party), 328 Hastings, Father Adrian, 171 Health problems, 89, 116, 221 Hertzog, General Albert, 27, 33-4 Hochtief (German firm), 52, 78, 100, 108, 128, 168, 198-9 Hospital arrangements, 116-17 Hydro-electric barrage, workings of, 122-5 Hydro-Technica Portuguesa, 19-20, 23, 36-7, 41, 50, 59, 69, 95, 100, 211 IMCC (Italy), 68 Impreglio, 42, 48-9 Inacio, Palme, 309 Independence or self- determination? 332 Industrial Development Corporation (RSA) 29-30, 36, 45-7, 57, 60, 67, 69, 83, 252 Inflation, 127, 197 Interest rates, 55, 62-3, 83 Investment: France, 66-7; Germany, 67; Italy, 68-69; RSA, 67, 80 Iscor, 29 Japan, mining interests of, 227 Jardim, Jorge, 105, 144, 164, 244, 264, 270-1, 291, 295, 298-9, 300, 324-6, 331, 336 JCI, 227-8, 254-5, 344 Johannesburg Symposium (1971), 35 Jordan, Franz, 100 Junta, Portuguese, see under Portuguese Crisis, the Kafue Dam Scheme, 31, 96 Kagan, Nicholas, 45, 50, 77, 78, 82 Kambona, Oscar, 181, 274, 299 Kapepwe, Simon, 274 Kariba Dam, 11, 16-18, 31, 95, 96, 126; comparisons with Cabora Bassa, 96-7 Kaunda, Kenneth, 57, 68, 83, 128, 143, 145, 163, 164-5, 167, 170, 182, 267, 271, 274, 276, 284, 290, 296-7, 298-9, 300-2, 326, 330 Kavandame, Lazaro, 136-7, 140-1, 267, 329 Kiihne, Heinz, 167

Index 363 La Bossi~re, M. (electrical engineer), 73, 112 Labour force, problems over, 101-2 Lake, final level of, 199 Lange, Horst, 108, 191-2 Language differences, 97, 107, 117, 137, 260 Leite, Pereira, 328 Lemperi~re, M., (civil engineer) 73, 112, 126 Life span of Dam, 212 Literacy in Mozambique, 221 Livingstone, Dr David, 9-11 Lourenqo Marques, 13-16, 18, 27, 59, 79, 100; riots in, 336 Louw, Eric, 34-36 LTA,. 51, 52, 108, 188, 198-9, 204, 254 Lusaka Conference and Agreement, 83, 213, 327, 332, 335, 339, 343, 345 Lusaka Manifesto, 162, 163, 297, 299-300 Machado, General Basto, 156 Machel, Samora, 141-2, 153, 176, 179-80, 259, 299, 302, 326-7, 343 Machinery, installation of, 191-2 Magalhaes, Calvet de, 26, 29, 59-60, 79 Magonga, Josd, 153 Mague (Portuguese engineering concern), 50 Makonde Tribe, 132-3, 135-7, 139-40, 267 Manu (political group), 133-4, 136 Manzanares, Abecassis, 11, 17 Matola Radio Station (LM), 336 Mediation offer by Mr Kaunda and Dr Banda, 317, 324, 326 Mercury-arc rectifier, 43-44, 69 Merz MacLellan (a firm), 45 MFPZ, 18-21, 23-4, 26-7, 41, 46, 47 Moatize, town and railway, 11-12, 93, 105, 121, 147, 152, 227-8 Mobutu, President, 317-8, 330, 338 Mondlane, Eduardo, 134-5, 136-37, 140-2, 160, 262 Moreira, Adriao, 19, 24-25, 132, 263 Moro, Signor, 68 Morrison Knudsen Inc., 49, 77, 200, 204, 293 Mortality rates in Mozambique, 221-2 Mozambique and Southern Africa, 281-305; Rhodesia after Frelimo, 281-8; proposed thermal station at Wankie, 282- 3; Council of Three, 283; Frelimo-Zimbabwe alliance, 284; ZANU and ZAPU activities, 284-5; security tightened, 285-6; border with Zambia closed, 286; increasing patrols inside Mozambique, 286; Smith visits Lisbon, 287; Settler (74) Campaign, 287; RSA's reactions, 288-91; Russian naval dominance 288, 294, 304; arms purchases, 289; China, impact of, 289, 302-4; three stages of policy, 289-90; Britain's concern, 291-2; French policy, 292-3; German policy, 293; USA policy, 293-4; Malawi's needs, 295-6; Zambia, 297-300; transfer of power from Portugal, 301; Tanzania, 301-2, 304-5; East African Community, 302; coal-mining in, 227; development or exploitation of? 209-232; costs and financing, 211; RSA's contribution, 211-12; Zambia's, Malawi's and Rhodesia's power needs, 212; RSA's shortfall on base load 213-14, Mozambique, increasing development of, 214-15; collectivisation of rural population of Tete, 217-18; Sangara development programme, 218-19; tstse-fly clearance, 219; tribal customs, problem of, 220-1; educational standards, 220-1; literacy, desire for, generated, 221; infant mortality declines, 221-2; population increase results, 222; agricultural schemes, 222-3; mineral resources, 225, 228, 230; coal-mining, 227;

364 cheap transport, necessity for, 229; new port? 229; extended railways? 229; Nacala town and railway, 229; Zambesi, a navigable highway? 230; Bauxite, plans to exploit, 230; effect of world oil crisis on, 230; Rhodesia, electric power for, 231; road improvements, 231 Mozambique Liberation (magazine), 135, 139, 142, 161-2, 169, 183 Mozambique, mineral deposits in, 225-6, 230; railway development in, 229-30; revolt in, 131-56 start of, 133; Mueda, riots at (1960), 133; Makonde Tribe, problems, 132-3; political groups, 133-4; Frelimo recognised by OAU, 135; weapons of war, 135-6; disputes regarding teaching, 137; war continues in Malawi and Zambia, 138; Frelimo steps up attacks, 139; activities in Dar-es-Salaam, 140-1, 142; African political intrigues, 141-2; Chinese instructors, 142; Mueda cut off, 142-3; General d'Arriaga takes command, 143; casualties, 143-4; Portuguese strike back, 143, 144-6; Zambia's attitude, 146; progress into Tete Province, 146-7; sabotage successes, 147; war reaches Tete, 149, 152; Portuguese reprisals, 149; villages face starvation, 149; rockets nine miles from Dam, 152; SAM missiles, 154; Mau Mau similarities appear, 154-5, Beira-Umtali Railway, Frelimo attack on (1974), 155; Lisbon, revolution in, effect of, 156; transition government of, 332-3, 342- 3, see also, Economic autonomy, progress towards in Mozambique; Que e Mozambique?; and under Sequel Mrupa, Miguel, 81, 162, 271, 329 Muller, Hilgard, 36, 79-80, 287, 331 Multi- racialism, importance of, 117, 352 CABORA BASSA Murchison, Sir Roderick, 9 Mutarara Development Project, 220-1 Muzorewa, Bishop, 287, 331 Nangani, Signor, 112 Neto, General Diogo, 323-4 Nixon/Spinola meeting, 337-8 Nkolosi, H., 146 Nogueira, Franco, 25-29, 37, 53, 59, 62-63, 74-75, 78, 81, 165, 181, 309, 311, 351 Nyerere, Julius, 133, 223, 224, 267, 273-4, 276, 284, 296, 300, 302, 326, 330, 344, 348 OAU, 26, 134-5, 142, 145, 163, 164, 181, 284, 295-6, 298, 302, 325 Obote, Milton, 274, 298 Oliveira, Arantes e, 71- 73, 91, 110 Olivier, Dr Henry, 29-31, 45-46, 214, 250 One-party Government, 137, 177, 274-5, 349 Operation Minerva, 129 Oppenheimer, H. F., 31-32, 36, 51, 75, 254 Overseas Ministry (of Portugal), 18-19, 22, 26, 41, 48, 53, 58-59, 71-2, 98, 196, 210, 214 Oxbow scheme, 31 Pan-Africanism, 132-3, 134, 163, 348 Paravicini, Professor, 68 Partition, policy of, 269-70 Passbook systems, 262 Patanella, Signor, 112 Pavolini, Roberto, 73 PIDE, 134, 136, 141, 262-3, 265, 312 Pinto, Professor Texeira, 247 Pires, General Adriao, 139 Plan Geral, 19-20, 22-23, 25, 41, 91, 98, 209, 268 Plebiscite, Frelimo rejects offer of, 326 Policia Internacionale Defesa do Estado, see PIDE Pompidou, Georges, 165-66

365 Portugal, internal problems within, 74-6, 156, 276, see also Portuguese Crisis, the Portugal and the Future (Spinola) 182, 272, 290, 298, 300, 314-5, 318, 320, 337 Portuguese Crisis, the 308-333; proposal for United States of Portugal (1946), 308; Aveiro Conference (1973), 309; Caetano's policy towards autonomy overseas, 310; church views, 311; economic crisis, 311; shortage of labour, 311-12; conscription, 312-14; army unrest, 314; Portugal and the Future (Spinola), 314-15, 318, 320; abortive coup (D'Arriaga's), 316; mediation offer in Mozambique, 317, 324, 326; National Assembly, vital meeting of, 318; Mozambique settlers demand arms, 319; army revolt (Caldas da Rainha Barracks), 319; Revolution itself (25.4.74), 320-21; Military Junta, 321-25; AFM Manifesto on self determination, 322; Mozambique without effective government, 325; Frelimo rejects plebiscite, 326; Lusaka Conference (4.6.74), 327, 332; Cabinet crisis, 327; political parties, profusion of, 328; Frelimo supporters emerge, 329; EEC's attitude, 330-31; RSA's attitude, 330; Rhodesia, political situation of, 331; Beit Bridge Railway, increased importance of, 331; 'Self-determination' becomes 'Independence', 332; Mozambique's Transitional Government, 332-3 Postos, 133, 138, 140, 146 Power Lines (Pty) Ltd, 69 Preece, Cardew and Ryder (Consultants), 61, 193 Preparation, 89-105; physical problems, attendant, 89, 90; road problems, 90-91, 98-9; cement supply, problems of, 93, 103, 104, 105, 121; bridges, absence of, 93; Zambesi, diverting of, 94; river flow, problems involving, 94-5; need for close Rhodesian cooperation, 96, 104; tunnelling starts, 96; Dam wall, construction of, 96-97; excavation work, 97; language and temperament problems, 97-8; recruitment of labour force, 99-100, 101; local labour force problems, 100-1, 102; labour laws and wage rates, 102-3; supplies, organisation of, 103-104, 105; steel, 104, 105; see also, Building of Dam Project Manager, late appt of, 108 Propaganda war, 160-83; Russian arms, 161; in OAU, 164-5; brought to London, Paris and New York, 165; European capitals' attitude, 164-7; Friedrich Ebert Peace Foundation, funds from, 167; Radio Tanzania, 169; Vatican pressure, 172; activities in UN, 174; Communism and the, 175; TANU, activities of, 177-8; re-training by Russians, 178; Russian assistance, 178-9 Psychological warfare, 144, 150-1 Que d Mozambique? 259-276; population percentages, 260; Asian enclaves, 260; ethnic groups listed, 260; languages in use, 260; geographical frontiers, 261; history of whites, 261; disparity of wealth and opportunity, 262; Trades Unions, 262; passbook system, 262; Delgadist Programme, 263; creation of state (1971), 264; elections (1973), 265; National Action Party, 265; detention camps, 265; military recruitment, 266-7; separatism, 269; partition, 269; emigration, 270; repatriation of funds, 270; Third Force (1973), 270-71; integration of Frelimo and Mozambique, Jardim's policy of, Index

366 270-71; reconciliation, Caetano's policy of, 272; leadership code (1974), 274; Malawi, politics in, 275; Portuguese dominion 1974, collapse of, 276 Racial harmony, 116-17 Radio Tanzania, 167 Recruiting a work force, 99-102 Regine do indigenato, 132 Reis, Carlos, 133 Resettlement (of population) costs, 62 Rhodesia after Frelimo, see under Mozambique and Southern Africa Rhodesia and sanctions, 104-105, 122, 195-6, 283; see also UDI (Rhodesia's) Rieckert, P. J., 35 Riscor, 104 Roads, problem of, at Dam site, 90-1, 98-9; see also Access road, difficulties in making Roberto, Holden, 134 Rodriguez, Adriao, 270, 328 Roxo, Daniel, 266 RSA after Frelimo (see Mozambique and Southern Africa): attitude of, to Portuguese Crisis, 330; financial contribution, 211-12; Industrial Development Corporation of, see Industrial Development Corporation (RSA); Iron and steel interests, 227; new economic ties with the future Mozambique, 346; political decisions of, effecting economies, 24- 29, 30-36; reactions of, to Frelimo advance, 288-91; recession in 78-9; water shortages in, 250 Russian assistance in propaganda war, 178-9 SAE (Milan), 49, 52, 67-69, 109, 127, 128, 162, 164, 165, 188, 194, 204 Salaries, see Wage-rates Salazar, Dr Antonio de Oliviera, 14-16, 22, 25, 28, 30, 37, 74, 76, 131-2, 138, 235, 236-7 CABORA BASSA Sanctions of Rhodesia, 104, 105, 122, 195-6, 283 Sangara Development Programme, 218-19 Santos, Almeida, 200, 263, 324-5 Schumann, Maurice, 165-6 Second bid tactics, 77 Self-determination or Independence? 332 Sequel, 335-352; Fico's call to arms 325, 328, 336, 337; Lourenqo Marques, riots in, 336; Nixon/ Spinola meeting, 337-8; Spinola/Mobuto meeting, 338; Mozambique, political future of, 342 ff.; transitional government, 332-3, 342-3; a democracy, 343-4; Yaounde preferences, benefit from? 345; new economic ties with RSA 346; has multi-racialism been established? 352 Shaftsinkers (a firm), 100, 108, 109, 123-4, 254 Shangaan, tribe and people of, 101-2, 117-18, 119, 267-8 Simango, Uria, 134-5, 141-2 Simiao, Joanna, 272 Simoes, Brigadier Rocha, 147-8 Simonstown Naval Agreement, 289 Single Party States, see One-party Government Smith, Bill, 101 Smith, Ian, 154, 267, 285-6, 287, 290, 301,331 Soares, Mario, 309, 324-5, 330 Societa Financeira Portuguesa, 247 Songo plain and township of, 12-3, 45, 89-91, 99, 110, 113-15 Sousa, Crispin de, 110 Specification, 58, 60 Spillways, start made on, 189 Spinola, General, 145, 175, 182, 201, 236, 248, 259, 272, 290, 295, 298, 300, 311, 314-6, 317-8, 321, 324, 327, 331, 336-8 SREP, 110 Steel, supplies of, 104, 105, 127 StraszAcker, Dr H. 32-3, 36, 68, 81

Index Struggle for Mozambique, the (Mondlane), 140 Sumitomo Mining (a firm), 24 Supplies, organisation of, 103-5; see also Cement, steel Swedish interests withdraw, 81-2 TANU, 137, 176-8, 273-4, 275, 300 Tenders, 42, 47, 48, 53-4, 61, 63-4; see also Contract; Specification Tete Town and Province, 9, 11-2, 16, 24, 90-1, 93, 115, 217; military eruption in (1971), 131 Thyristor valve, development of, 43-4, 70, 82, 193, 203 Tomis, Admiral, 145, 310, 320 Tribal customs, problems of, 220-1 Tsetse fly clearance, 219 Udenamo (political group) 133-4 UDI (Rhodesia's), 24, 26, 28, 47, 105, 163, 182-3, 281-8 Ujamaa, doctrine of, 223-4 Unami (political group), 133-4 Unden, Bungt, 73 UNIP, 274-5, 344 United Nations, 26, 48-9, 56, 63, 81, 104, 141, 162, 163, 174, 283, 332, 338 USA policy and Frelimo advance, 293-4 United States of Portugal, proposal for, 308 Van Eck, H. J., 22-3, 29-33, 36, 42, 46, 48, 58, 79, 250 Van der Bigl, Mr, 29-30 Vatican, pressure from, 172; see also World Council of Churches Verwoerd, Hendrik, 22, 28-9, 33-4, 36, 58 Vorster, B. J., 34, 58, 60, 79, 165, 267, 287, 289, 290-1, 331, 347-8 Wage-rates, 57, 61, 102-3, 115 Wallenburg, Marcus, 53, 81-2 Wankie, proposed thermal station at, 282 War of propaganda see Propaganda war Welenski, Sir Roy, 286 White Fathers, Mission of, 171-2 Wiriyamu massacre, 149-50, 171 World Bank, 42, 210, 224, 231 World Council of Churches, 162, 170-1 Xhosa tribe, 288 Yaounde preferences, 345 Zagasi, Signor, 68 Zambesi River and basin, 9-11, 16, 19; annual flood of, 45, 66, 94, 190; bridging over, 122, 196; a navigable highway? 230 Zamco (a firm) 45, 47, 50-1, 53-5, 61-64, 66-84, 91, 94, 98-99, 110, 112-13, 126, 186 ZANU, 283-4, 285-7, 301-2, 349 ZAPU, 283-4, 285, 301, 349 ZECCO Consortium, 96 Zimbabwe/Frelimo alliance, 284 ZANU guerillas, 152, 154 Zulu peoples, 288 367