Darch on Sumich, 'The Middle Class in Mozambique: The State and the Politics of Transformation in Southern Africa'

Jason Sumich
Colin Darch

Jason Sumich. The Middle Class in Mozambique: The State and the Politics of Transformation in Southern Africa. The International African Library Series. Cambridge: Cambridge University Press, 2018. 190 pp. $105.00 (cloth), ISBN 978-1-108-47288-3; $24.00 (e-book), ISBN 978-1-108-57702-1.

Reviewed by Colin Darch (University of Cape Town) Published on H-Luso-Africa (November, 2020) Commissioned by Philip J. Havik (Instituto de Higiene e Medicina Tropical (IHMT))

Printable Version: https://www.h-net.org/reviews/showpdf.php?id=55879

As far as I am aware, Jason Sumich's new book is the first full-length contribution—in English at least—to the study of the middle class focusing on Mozambique, a topic on which he has previously published widely and has perhaps made his own.[1] It is therefore pioneering and should be warmly welcomed, and indeed the narrative, although short, is engaging, readable, and well structured. Sumich might have done more in this book to tackle long-standing ambiguities about the social category of the "middle class" as an object of study, and in addition the text raises questions about the rigor of his method as an anthropologist writing historically. I will expand on these two points in the course of this review.

It is already clear that the book can be and will be (mis)read as a historical text—for example, as "an insightful contribution to the history of Mozambique and class formation under colonial and postcolonial conditions" or as an "introduction to the contemporary history of Mozambique," to quote two recent reviewers.[2] This is mainly because the book is organized historically rather than, as one might expect, thematically, and is presented as a narrative around a periodization that takes the moments of independence in 1975 and of the Acordo Geral de Paz (AGP) in 1992 as two of its key inflection points. The five narrative chapters follow a broad and highly conventional periodization of, successively, the late colonial period; the early independent state; the "civil war and the economic chaos"; a chapter titled "Democratisation," covering the period from 1992 to 2004; and a final chapter titled "Decay," the heart of the book, focusing on the period from 2005 to 2015. These are bracketed by an introduction that addresses theoretical (but not methodological) issues, especially around the question of what the term "middle class" actually means, and a conclusion that brings the narrative up to 2016. Perhaps surprisingly, given Sumich's evident enthusiasm for João Paulo Borges Coelho's concept of the "liberation script," which he references several times, he does not argue for this periodization in any detail. But the power of Borges Coelho's concept hangs precisely on its identification of a semiofficial historical metanarrative that is not arbitrary in its division into periods and that makes sense only in terms of a view of history in which the armed struggle against the Portuguese from 1964 to 1974 becomes the history of modern Mozambique in and of itself, occupying the whole available historical space from the early 1960s onward. As Borges Coelho says, the liberation script consists of "a coherent and fixed narrative corpus made of a sequence of events in a timeline and ordered in a number of broad phases separated by Frelimo Congresses which operate as periodization marks. Each con­gress occurs to solve a crisis that was aggravating within each period, and to neutral­ize the threat that that crisis represented to the nationalist endeavor. The opening of a new phase [is] only made possible by the resolution of the crisis of the previous one."[3]

Thus, the second congress in 1968 marked the sharpening to the point of crisis of the struggle between the two lines (if not its resolution) and the consequent militarization of FRELIMO; the formal adoption of Marxism-Leninism in 1977 at the third congress supposedly equipped what is now the vanguard Frelimo Party with the tools to bend the inherited colonial state structures to its will; and the fourth congress in April 1983 offered a democratic moment in which a space opened up for ordinary cadres to criticize the party leadership.[4] The fifth congress in July 1989 involved a shake-up at the top levels of the party while the possibility of a negotiated peace slowly emerged, and the "extraordinary" sixth congress in August 1991 went a step further and dropped Marxism as Frelimo's official ideology, after the single-party parliament had already preemptively approved a pluralist constitution, without Renamo participation, late in 1990.

These steps, viewed with hindsight, show some agility as well as opportunism on Frelimo's part in laying the groundwork for holding on to power (or rather "solving the crisis" as Borges Coelho has it) and open up questions about the real character of Mozambique's process of so-called democratization. Luciano Canfora has argued in another context that democratization can all too often have the effect of stabilizing existing relations of property and power rather than marking a rupture, and in hindsight it is clear that the painfully negotiated AGP did exactly that, creating space not so much for multipartidarismo as for what is effectively a two-party system in which Frelimo has remained dominant (although not always in the context of a "party-state") and Renamo has unwillingly played the apparently permanent role of "loyal opposition." Other parties, by and large and designedly, do not get a look-in.

In the last few decades, the middle class has become a fashionable object of study for sociologists, anthropologists, and economists around the world, including the global south. Indeed, the emergence of a growing middle class has been held out by some as the latest (and still teleological) solution to the problem of socioeconomic development, especially in Africa, as Sumich points out.[5] On the broader African middle class, Henning Melber's edited volume The Rise of Africa's Middle Classes: Myths, Realities and Critical Engagements (2017), to which Sumich contributed a chapter, and the collection The Emerging Middle Class in Africa (2015) edited by the economists Mthuli Ncube and Charles Leyeka Lufumpa have brought the concept to the fore.[6] However, it remains a slippery idea. Apart from the obvious economic categorization that locates the middle class between the poor and the wealthy in terms of income and/or consumption, multiple competing definitions based on sociological variables—such as educational level, professional status, educational achievement, "lifestyle," and even aspiration—have all been deployed. Simple classifications based on income can be relative (that is, within a national distribution) or absolute (for example, per capita income in US dollars). An absence of theoretical agreement around these issues makes cross-national comparative analysis generally difficult, a problem that Sumich chooses not to address despite acknowledging it: "For my purposes, the middle class shares some general sociological characteristics that make this social category more or less recognizable across the globe. These characteristics include broad economic factors, such as a degree of material power, and social marks of distinction such as certain levels of formal education and cultural capital, employment in a professional capacity, and a largely urban-based lifestyle" (p. 8).

All this has presented me with a problem of lecture, of leitura: how is the book meant to be read? Sumich paints a picture of a broadly dissatisfied urban "middle class" in Maputo based on research carried out over a decade, between 2002 and 2016. During those years, he immersed himself in this particular urban social milieu, and his narrative relies mainly on the testimonies—in effect, the petits récits—of a narrow group of a couple of dozen lightly disguised informants, with multiple direct quotations. The book is organized around this cadre's largely unfiltered perceptions of present-day social reality and contemporary political history—a discourse of signification, to appropriate Sumich's preferred terminology—with extensive references to the experience of a range of other countries, both in Africa and elsewhere. In fact, Sumich's theoretical reading is impressively wide-ranging: in a bibliography of around two hundred referenced publications he cites material on Hungary, Indonesia, Bengal, and Melanesia, among many other places. However, only a third of his sources deal directly with Mozambique, and of those only a handful are in Portuguese or by Mozambican scholars, such as Yussuf Adam, Teresa Cruz e Silva, Benedito Machava, or Borges Coelho. For whatever reason, he ignores the extensive memoir and biographical literature of the last two decades, by and on Frelimo as well as opposition figures, which reveals their subjects' wide range of social origin in the families of teachers, minor state functionaries, health workers, and so on. Frelimo's own voice is barely heard: Sérgio Vieira and Samora Machel are quoted once each, on both occasions from English translations. Sumich cites a handful of statistics from secondary sources but is generally dismissive of Mozambican statistical data—presumably including the Instituto Nacional de Estatística's various household surveys—as being "vague and unreliable" (p. 9).

These were the years of the presidential mandates of Armando Guebuza (in office from early 2005 to early 2015), who had already become the Frelimo Party's official presidenciável (eligible presidential candidate) in 2002, when he took on the role of party secretary-general. The "hidden debt" scandal, involving secret loans of around two billion dollars to the government—on Guebuza's watch—broke in 2016, shortly after he had left office (pp. 150-51). The period of Sumich's immersion in Maputo society was, therefore, marked by the fairly rapid breakdown of the then dominant narrative of the country as a post-conflict success story, after the negotiated end (in 1992-94) of the sixteen-year post-independence war with Renamo and a long period of uninterrupted economic growth. The quotations from Sumich's informants, however, also reveal a final and parallel breakdown of popular belief, not only in post-conflict success but also in the heroic narrative of the armed struggle for national liberation and the failed post-independence social revolution. The consequent sense of collective disillusionment in both founding mythos and present reality permeates the book. This is a direct consequence of Sumich's methodology: by privileging the perceptions of a small selection of Mozambican interlocutors, their collective disillusionment emerges clearly. Nonetheless, the extent to which it is valid to generalize from their experience remains an open question.

Anthropological method is both the book's strength and its weakness. On the one hand, from the extensive quotations of direct (but presumably translated) testimony we gain an understanding of how a (rather small) sample of privileged urban Mozambicans—twenty-five or so people are mentioned by name—see themselves and their society.[7] On the other hand, Sumich does not provide us with any basis for evaluating the extent to which his informants' representation of such issues as corruption, levels of violent criminality, the competence of state structures, and so on does in fact reflect what is empirically known. He refrains from discussing the problem, and indeed, it may be that such an interrogation of the relationship of analysis to evidence, of the idiographic and the local to the nomothetic, in the construction of a chronological narrative is irrelevant to what he is trying to achieve. The contradictions and exaggerations may be the point, although this is not stated in so many words. In a recent brief response to a group of three reviewers, all of whom have published on the middle class elsewhere, Sumich offered the following comments about his objectives in the book: "My goal was not to try to map a category of 'middleclassness' as an empirical reality, especially as it is debatable how beneficial the concept of a middle class is as an analytical category. However, it can be very meaningful as a folk concept. My interlocutors generally saw themselves as occupying some sort of middle in their social world. To varying degrees they felt subject to, or hostages of, the whims of a 'they' who occupy the summit of society and the roiling discontent of a 'them' below. Building from this, I understand the concept of a middle class as a discourse among those with significant if differing levels of privilege. It is a claim concerning the nature of reality and one's role within it. In my view, the middle class ... is a system of signification where global influences combine with pre-existing social logics to form the hierarchies—with their often amorphous middles—that have drawn our attention."[8] The idea expressed here that the middle class is a kind of narrative rather than an actual social group emerges from anthropology's long-standing preoccupation with questions of identity, self-identification, and representation. The term "folk concept" is anthropologists' expression denoting a vague idea that is popularly understood within a given social group but that does not have a specific or formal definition. Hence, Sumich's two dozen friends define themselves as middle class, and that is apparently sufficient both for his and their purposes.

In a passage in which he seems to be arguing that Samora Machel occupied a dominant position in Mozambique that was analogous to Joseph Stalin's in the Soviet Union, Sumich argues that Machel was "the living embodiment of Frelimo's social revolution ... the public face of the party." His speeches were "programmatic" and spelled out how transformation was to be achieved (p. 84). This is not entirely inaccurate. However, little is known about the internal Frelimo Party processes by which policy was developed within the Politburo or the Central Committee, or what debates took place in the closed sessions of the Assembleia Popular, and it may well be that Machel's positions were sometimes defeated (for example, regarding the executions of the Nachingwea "traitors"), in a way that it is hard to imagine Stalin's were.[9] Similarly, Sumich's assertion that Joaquim Chissano, after 1986, "found it difficult to assume Samora's role" seems to discount the extent to which Chissano was his own man and had the confidence of the party (p. 88).

Popular narratives of contemporary history, especially in a society in which independent sources of information are few and far between, exercise a powerful grip, and Sumich's reproduction of what is frequently middle-class gossip (boatos, fofoca, papos da esquina; synonyms for gossip) about current or recent events provides multiple vivid examples of this.[10] "Rumors" have long been a focus of anthropological attention, of course; it may also be worth noting the level of disapproval of boateiros (rumor-mongers) as little better than bandits during the revolutionary period.[11] Sumich reproduces widely circulated jokes, such as the one about a fish (p. 86n1) that was popular in the 1980s but misses the point about carapau (p. 90), a nutritious species of mackerel that was widely believed at that time to have been overfished by Spanish vessels in the Mozambique Channel, leaving only the "small bony" juveniles for local consumption.

Sumich briefly discusses Operação Produção, a brutal attempt by the government to resolve the problem of urban unemployment by inventing a social category of marginais or improdutivos (marginal or unproductive people). These unfortunates were identified bureaucratically through the absence of certain kinds of ID cards, and were then flown off to the countryside in the far north to work in agriculture. This project was launched in the aftermath of the Fourth Congress, and there is some evidence that it was initially popular (as Sumich acknowledges) among people who saw the unemployed as responsible for crime and as an economic burden in a period of scarce resources. The Frelimo Party has never renounced the measure, and indeed, has sometimes defended it.[12] Sumich states that between thirty thousand and fifty thousand people were removed in this way, but this is almost certainly an inflated figure of the kind that his interlocutors would likely believe in (p. 85). Official data has it that in phase two, compulsory removals totaled around ten thousand people, about half of them from Maputo. Most of the improdutivos were flown to Niassa on aircraft belonging to LAM, the national airline; the expulsion of even five thousand deportees needed as many as twenty-six flights, along a 1,500-kilometer route to Lichinga, where there was almost no preparation to receive them. Nonetheless, in folk memory, "thousands and thousands" of people were forcibly shipped off, and the actual number is probably less important than the symbolizing of the unjustness of the process.[13]

There are multiple examples of minor historical infelicities that arise from an anthropological reliance on popular memory. Sumich takes "traditional leadership" as a given rather than a hotly contested term, and even quotes "a woman from a party-connected family" without comment as saying that "we broke the tribal system most people lived in" (pp. 80, 82, 103, emphasis added). On page 49 he states that large numbers of Africans who fought in the colonial army were "considered traitors by Frelimo," when in fact it was the volunteer members of such units as PIDE's Flechas, and the army's Grupos Especiais and Grupos Especiais Páraquedistas who were greatly mistrusted, and not the many draftees. The unnuanced claim that the Portuguese colonial state "made few provisions for African education" makes no allowance for the fact that the 1940 Concordat and the 1941 Missionary Statute were the basis for a system carefully designed to keep Africans in their place as cheap labor (p. 69). This was an intervention with a clear political objective not the abandonment of the field to chance and the Presbyterians. The food riots of 2008, 2010, and 2012 are mentioned several times with no reference to analyses and testimonies published by Luís de Brito at the Instituto de Estudos Sociais e Económicos in Maputo.[14] There are other examples.

In the end, despite these criticisms, what Sumich does achieve is to provide a snapshot of the attitudes and perceptions of a narrow social stratum that is purposefully vaguely defined except in terms of the common belief of its members that they inhabit a kind of social middle space, "subject to ... the whims of a 'they' who occupy the summit of society and the roiling discontent of a 'them' below."[15] This is, quite deliberately, a long way from class analysis in the classical sense.[16] Sumich has remarked elsewhere that he is "an anthropologist rather than a historian" and that his text focuses on the "projects of transformation that were the most salient for my interlocutors ... due to the fact that they had occurred within living memory."[17] At the risk of criticizing the author for a book that he has not written rather than for the one he has, let me conclude by saying that Sumich's text suffers principally from his reticence on the methodological and epistemological matters indicated above. Its virtues lie in what it tells us about contemporary urban Mozambicans' attitudes toward society and history rather than in any claim to present a historical narrative per se.


[1]. Sumich has published articles on this and related topics. See, among others, with Morten Nielsen,"The Political Aesthetics of Middle-Class Housing in (Not So) Neoliberal Mozambique," Antipode 52, no. 4 (2012): 1216-34; "The Party and the State?: The Ambiguities of Power in Mozambique," in Negotiating Statehood: Dynamics of Power and Domination in Africa, ed. Tobias Hagmann and Didier Péclard (London: Wiley-Blackwell, 2011), 134-53; "Tenuous Belonging: Citizenship, Democracy and Power in Mozambique," Social Analysis 57, no. 2 (2013): 99-116; "The Uncertainty of Prosperity: Dependence and the Politics of Middle Class Privilege in Maputo," Ethnos 1 (2015): 1-21; "Politics after the Time of Hunger in Mozambique: A Critique of Neo-Patrimonial Interpretations of Elites," Journal of Southern African Studies 34, no. 1 (2008): 111-25; and "Just Another African Country: Socialism, Capitalism and Temporality in Mozambique," Third World Quarterly (July 2020): 1-17. 

[2]. Nkululeko Mabandla, review of The Middle Class in MozambiqueAfrica (IAI) 90, no. 3 (May 2020): 599; and Carola Lentz, review of The Middle Class in MozambiqueAfrica (IAI) 90, no. 3 (May 2020): 601.

[3]. João Paulo Borges Coelho, "Politics and Contemporary History in Mozambique: A Set of Epistemological Notes," Kronos no. 39 (2013): 21.

[4]. On the question of the parallel militarization of the Portuguese state and the liberation movement in the late 1960s, see my co-authored work, Colin Darch and David Hedges, "Não temos a possibilidade de herdar nada de Portugal: As raízes do exclusivismo político em Moçambique, 1969-1977," in Territórios da língua portuguesa: Culturas, sociedades, políticas, ed. Glaucia Villas Bôas (Rio de Janeiro: IFCS/UFRJ, 1999), 135-49. A volume of delegate speeches from the fourth congress was published later. See Intervenções dos delegados ao 4º Congresso (Maputo: Frelimo, 1985).

[5]. See, for example, the survey by Pierre Jacquemot, "Africa's 'Middle Class': Realities, Issues, and Perspectives," Afrique Contemporaine 244, no. 4 (2012): 17-31.

[6]. Jason Sumich, "The Middle Class of Mozambique and the Politics of the Blank Slate," in The Rise of Africa's Middle Class: Myths, Realities and Critical Engagements, ed. Henning Melber (London: Zed Books, 2017), 159-69.

[7]. The quotations from Sumich's informants are un-contextualized: Are they translations from Portuguese? Were they recorded or reconstructed from notes or from memory?

[8]. The three reviewers were Nkululeko Mabandla, Leela Fernandes, and Carola Lentz. See reviews of The Middle Class in Mozambique by Jason Sumich, Africa (IAI) 90, no. 3 (May 2020): 598-603, including Jason Sumich's comments to reviews, Africa (IAI) 90, no. 3 (May 2020): 602-3.

[9]. See John Saul, "The African Hero in Mozambican History: On Assassinations and Executions. Part II," Review of African Political Economy 47, no. 164 (2020): 340.

[10]. Toward the end of the conflict with Renamo, in March 1991, the Mozambican News Agency (AIM) admitted that much news went unreported, attributing this to problems of the communications infrastructure (MozambiqueFile, March 1991, p. 23). But this was an excuse: Mozambicans often had "to switch to foreign radio stations to find out what [was] happening in their country." Carlos Cardoso, quoted in Paul Fauvet and Marcelo Mosse, Carlos Cardoso: Telling the Truth in Mozambique (Cape Town: Double Storey, 2003), 224.

[11]. See, for example, articles in Notícias: "Denunciemos os boateiros," Notícias (April 26, 1984); "Alguém diz que viu?" Notícias (April 27, 1984); and "Instrumento dos bandidos: Cerrar fileiras também contra boatos, posição assumida pelos moradores dos DU's 5 e 8," Notícias (October 15, 1984).

[12]. See, for example, the comments of Joaquim Chissano in the weekly Savana, November 19, 2004; and Lina Magaia, in O País [Maputo], August 3, 2007.

[13]. For a detailed study of Operação Produção, based on interviews and documentary research, see Carlos Quembo, Poder do poder: Operação Produção e a invenção dos improdutivos urbanos no Moçambique, 1983-1988 (Maputo: Alcance, 2017).

[14]. Luís de Brito, Egídio Chaimite, Crescêncio Pereira, Lúcio Posse, Michael Sambo, and Alex Shankland, Revoltas da fome: Protestos populares em Moçambique, 2008-2012 (Maputo: IESE, 2015); and Luís de Brito, ed., Agora eles têm medo de nós: Uma colectânea de textos sobre as revoltas populares em Moçambique, 2008-2012 (Maputo: IESE, 2017).

[15]. Sumich, comments to reviews, Africa.

[16]. In an earlier publication, Sumich argued that the concept of the middle class is "fundamentally political" and "ideologically potent," while recognizing that what he terms the "rhetoric of class" is currently "out of fashion." Sumich, "The Uncertainty of Prosperity," 823, 827, 825.

[17]. Sumich, comments to review, Africa

Citation: Colin Darch. Review of Sumich, Jason, The Middle Class in Mozambique: The State and the Politics of Transformation in Southern Africa. H-Luso-Africa, H-Net Reviews. November, 2020. URL: https://www.h-net.org/reviews/showrev.php?id=55879

This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 United States License.