StCompIntDev(2010)45:410–438 DOI10.1007/s12116-010-9075-7 Accumulation by Dispossession and Its Limits: The Southern Africa Paradigm Revisited Giovanni Arrighi&Nicole Aschoff&Ben Scully Publishedonline:9November2010 #SpringerScience+BusinessMedia,LLC2010 Abstract The dispossession of agricultural producers from the land has long been considered aconditionofsuccessful capitalist development.The main contention of this paper is that such dispossession has in fact become the source of major developmentalhandicapsforatleastsomeandpossiblymanycountriesoftheglobal South. We develop our argument by focusing on the South(ern) African experience as a paradigmatic outlier case of accumulation by dispossession—that is, as one of itsextremeinstancescapableofhighlightinginalmostideo-typicalfashionitsnature andlimits.AfterreconstructinginterpretationsofcapitalistdevelopmentinSouthern Africa that in the early 1970s established the region as a paradigm of accumulation by dispossession, we discuss how useful these interpretations are for understanding the more recent developmental trajectory of South Africa. We then suggest ways in which these interpretations from the 1970s should be reformulated in light of subsequent developments. We conclude by briefly examining the theoretical and policy implications of the analysis. PreviousversionsofthispaperwerepresentedattheInstituteofSocialandEconomicResearch,Rhodes University,Grahamstown,SouthAfrica,May3,2007;theDepartmentofSociology,Universityof California,Berkeley,February28,2008;andtheSeminaronInternationalDevelopment,Departmentof Sociology,TheJohnsHopkinsUniversity,March2009.Theauthorswouldliketothanktheparticipantsat theseminarsinwhichthepaperwaspresentedfortheirusefulcommentswhichinspiredsubstantive revisions.WewouldalsoliketothankPeterEvans,MartinLegassick,GregRuiters,SampieTerreblanche andEdwardWebsterfortheirhelpfulcommentsonapreviousversionofthepaper.Specialthanksaredue toBeverlyJ.Silver,whoprovidedextensivesuggestionsonmultipleversionsofthepaper. G.Arrighi BeverlySilver,SociologyDepartment,TheJohnsHopkinsUniversity,Baltimore,MD21218,USA e-mail:[email protected] N.Aschoff SociologyDepartment,BostonUniversity,Boston,MA02215,USA e-mail:[email protected] B.Scully(*) SociologyDepartment,TheJohnHopkinsUniversity,Baltimore,MD21218,USA e-mail:[email protected] StCompIntDev(2010)45:410–438 441111 Keywords Development.SouthernAfrica.Accumulationbydispossession. Labor supplies Thedispossessionofagriculturalproducersfromthelandhaslongbeenconsidereda condition of successful capitalist development. The main contention of this paper is that such dispossession has in fact become the source of major developmental handicaps for at least some and possibly many countries of the global South. We develop our argument by focusing on the South(ern) African experience as a paradigmatic outlier case of accumulation by dispossession—that is, as one of its extreme instances capable of highlighting in almost ideo-typical fashion its nature andlimits.AfterreconstructinginterpretationsofcapitalistdevelopmentinSouthern Africa that in the early 1970s established the region as a paradigm of accumulation by dispossession, we discuss how useful these interpretations are for understanding the more recent developmental trajectory of South Africa. We then suggest ways in which these interpretations from the 1970s should be reformulated in light of subsequent developments. We conclude by briefly examining the theoretical and policy implications of the analysis. Forthepurposesofthispaper,theterm“accumulationbydispossession”(Harvey 2003, 144) can be seen as largely synonymous with Marx’s concept of “primitive accumulation.” The types of processes included in both concepts are the same: “the commodification and privatization of land and the forceful expulsion of peasant populations;theconversionofvariousformsofpropertyrights(common,collective, state, etc.) into exclusive private property rights…; the commodification of labor power and the suppression of alternative (and indigenous) forms of production and consumption; colonial, neo-colonial and imperial processes of appropriation of assets…;themonetizationofexchangeandtaxation,particularlyofland;”etc.(ibid, 145).Inintroducingtheconceptof“accumulationbydispossession”,Harveysought to emphasize the fact that primitive accumulation is an ongoing process, and that “predatory practices” have played a major recurrent role in processes of capital accumulation, including in the current conjuncture (ibid, 144). We agree with Harvey about the ongoing importance of primitive accumulation, and therefore we have adopted the term “accumulation by dispossession.” The main focus of this paper, however, is on the effects of accumulation by dispossession on development. Our thesis (illustrated by the long-term developmental trajectory of South Africa) is that accumulation by dispossession generally undermines the conditions for successful development. In this regard, our thesis contradicts the widespread assumption in the scholarly literature that such dispossession is a necessary (progressive) precursor to capitalist development, an assumption that characterized both the US- and Soviet-sponsored versions of the post-war “development project” (see McMichael 2004).1 1Harveyhimselfarguesthataccumulationbydispossessioncanbeseenas“thenecessarycostofmaking a successful breakthrough to capitalist [and socialist] development”, but he also argues that in some instances(particularlypost-1973)accumulationbydispossession,ratherthanopeningupanewpathof expandedreproduction,disruptedanddestroyedpathsthatwerealreadyopen(ibid154–6). 412 StCompIntDev(2010)45:410–438 South Africa as Paradigmatic Outlier South Africa has long been regarded by social scientists, in Gay Seidman’s words, “as the end point of the spectrum of racial orders, the place where racial inequality stoodstill.”Andyet,alongwiththisoutlierstatus,“SouthAfricahasalsoservedasa prism—in part, perhaps, because the extreme character of apartheid lays bare the underlying dynamics of racial capitalism.” Especially important in this respect according to Seidman were studies of Southern Africa’s migrant labor system— includingHaroldWolpe’sanalysisofcapitalismandcheaplaborinSouthAfricaand Giovanni Arrighi’s analysis of the creation of “unlimited” supplies of labor in Southern Rhodesia (now Zimbabwe)—which suggested that African subsistence agriculture subsidized capital by supplementing the wages of migrant workers. Michael Burawoy broadened the scope of these findings by comparing the role of the South African and Californian states in regulating circulatory migrant streams and showing how in both instances migrant labor subsidized capital and states by pushing across the border the costs of retirement, education, and other social services, in addition to most of the costs of feeding workers’ families. Together, the South African studies contributed to a new approach to migration... By the mid-1980s, sociologists were drawing on insights derived from South Africa to examine migration in cases as far flung as the West Indies, Europe, and Mexico, looking at how states controlled the flow and circulationofmigrantsintermsoflaborsuppliesandlaborcontrol,andathow migrationflowsaredeeplyintertwinedwiththeracializationoflaborstreams.2 South Africa’s paradigmaticrole inmigration studiesstemmed from itscentrality in the macro-region that Samir Amin has called the “Africa of the labor reserves” (henceforth ALR). Stretching from South Africa, through Namibia, Botswana, Zimbabwe, Zambia, Mozambique and Malawi, to Tanzania and Kenya, this region was characterized by a combination of great mineral wealth, a white settler agriculturewithnoparallelelsewhereinSub-SaharanAfrica,andachronicshortage of labor. In order to overcome this shortage, “the colonialists dispossessed the African rural communities by force and deliberately drove them... into confined, poor regions, with no means of modernizing and intensifying their farming.” As a result, these communities were driven to become suppliers of temporary or permanent migrants, “providing a cheap proletariat for the European mines and farms, and later for the manufacturing industries of South Africa, Rhodesia, and Kenya.”3 South Africa, and to a lesser extent Rhodesia/Zimbabwe, were the main receivingcentersandbeneficiariesofthetransnationalsystemoflabormigrationthat crisscrossedthemacro-region;buttheywerealsothemainagenciesoftheprocesses of uneven and combined development that generated and reproduced the migratory system. As can be seen from Table 1, through the 1960s South Africa, and to a lesser extent the entire ALR region, had come to enjoy per capita incomes among the highest in the Third World or global South. Starting in the 1970s, and more 2Seidman(1999:420,424–5)citing,amongothers,Arrighi(1970),Wolpe(1972),andBurawoy(1976). 3Amin(1976:327–8). StCompIntDev(2010)45:410–438 441133 Table1 GNPPercapitaas%offirstworld’sGNPpercapita Region 1960 1970 1980 1985 1990 1995 2000 2005 Sub-SaharanAfrica(w/SA) 5.6 4.7 3.9 3.1 2.7 2.5 2.0 2.3 SouthAfrica 25.9 24.9 21.4 17.9 15.2 13.9 10.7 12.7 Sub-SaharanAfrica(w/oSA) 3.1 2.3 1.9 1.5 1.3 1.2 1.2 1.3 LatinAmerica 19.7 16.4 17.6 14.4 12.3 12.9 13.4 11.2 WestAsiaandNorthAfrica 8.7 7.8 8.7 7.9 7.4 7.2 7.7 8.4 SouthAsia(w/oIndia) 1.9 1.7 1.3 1.4 1.4 1.5 1.6 1.6 EastAsia(w/oChinaandJapan) 6.0 6.1 8.0 8.6 11.0 13.8 11.5 11.8 China 0.9 0.7 0.8 1.2 1.3 2.1 3.2 4.6 India 1.5 1.3 1.1 1.2 1.2 1.4 1.6 1.9 ThirdWorlda 4.5 4.0 4.3 4.1 4.1 4.7 4.9 5.2 ThirdWorld(w/oChina)a 6.5 5.7 6.1 5.5 5.3 5.9 5.6 5.5 ThirdWorld(w/oChinaandIndia)a 9.3 8.1 8.8 7.7 7.5 8.2 7.7 7.3 NorthAmerica 123.7 105.0 100.7 101.6 98.2 98.9 116.4 112.5 WesternEurope 111.1 104.6 104.6 101.5 100.5 98.5 92.0 99.7 SouthernEurope 51.9 58.2 60.0 57.6 58.6 59.2 61.5 70.2 AustraliaandNewZealand 94.8 83.5 74.7 73.3 66.4 70.6 68.6 84.5 Japan 78.9 126.4 134.4 140.8 149.8 151.9 121.0 103.1 FirstWorldb 100 100 100 100 100 100 100 100.0 EasternEurope – – – – 11.1 10.6 13.4 18.6 FormerUSSRw/RussianFed – – – – 10.7 5.9 4.6 8.2 RussianFederation – – – – 14.1 8.2 6.0 11.8 FormerUSSRw/oRussianFed – – – – 7.1 3.6 3.1 4.6 EasternEuropeandFormerUSSRc – – – – 10.8 7.1 6.9 11.0 Source:CalculationsbasedonWorldBank(WDI-2001,2006) aCountriesincludedinThirdWorld: Sub-SaharanAfrica:Benin,Botswana,BurkinaFaso,Burundi,Cameroon,CentralAfricanRepublic,Chad,Rep.of Congo,CongoDem.Rep.,Coted’Ivoire,Gabon,Ghana,Kenya,Lesotho,Madagascar,Malawi,Mauritania,Mauritius, Niger,Nigeria,Rwanda,Senegal,SouthAfrica,Tanzania,Togo,Uganda,Zambia,Zimbabwe Latin America: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Rep, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Trinidad & Tobago, Uruguay, Venezuela WestAsia&NorthAfrica:Algeria,ArabRepofEgypt,Morocco,SaudiArabia(1971for1970),Sudan,SyrianArab Rep.,Tunisia(1961for1960),Turkey SouthAsia:Bangladesh,India,Nepal,Pakistan,SriLanka EastAsia:China,HongKong,Indonesia,SouthKorea, Malaysia,Philippines, Singapore, Taiwan (TaiwanNational Statistics),Thailand bCountriesincludedinFirstWorld: NorthAmerica:Canada,UnitedStates WesternEurope:Austria,Belgium,Denmark,Finland,France,Germany,Luxembourg,Netherlands,Norway,Sweden, Switzerland,UnitedKingdom SouthernEurope:Greece,Ireland,Israel,Italy,Portugal,Spain AustraliaandNewZealand Japan cCountriesincludedinEasternEuropeandtheFormerUSSR: EasternEurope:Albania,Bulgaria,Croatia,CzechRepublic,Hungary,Poland,Romania,SlovakRepublic,Slovenia Former USSR: Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyz Republic, Latvia, Lithuania, Moldova,RussianFederation,Tajikistan,Turkmenistan,Ukraine,Uzbekistan 414 StCompIntDev(2010)45:410–438 markedly in the 1980s and 1990s, however, South Africa’s and the ALR region’s comparative economic performance deteriorated precipitously, not just vis-a-vis the First World or global North, but also and especially other regions of the global South, most notably East Asia. Some have attributed South Africa’s comparatively poor developmental perfor- mance over thepast three decadestoa fundamentalhandicap incompeting with the emerging industrial powerhouses of Asia—first and foremost China—in an increasingly integrated global economy. Richard Freeman, for example, has singled outSouth Africa—alongwithMexicoandColombia—asthetypicalmiddle-income countrythat hadhoped togrowthroughexportsoflow-wagegoodsbutnow hasno chance of competing successfully with China in labor-intensive manufacturing as longasChinesewagesareone-quarterorsooftheirs,whileChineselaborisroughly as productive as theirs.4 The question then arises of why Chinese wages are so low (and South African wages so high) for labor of about the same productivity. Implicitlyorexplicitlythreedifferentanswershavebeengiventothesequestions.In the remainder of this section we review these three answers, as laid out in the writings of Anita Chan, Alan Hirsch and Gillian Hart, respectively. As will become clear by the end of the section, our own assessment is closest to that of Hart. In a series of solo and co-authored articles, Anita Chan has been the main proponentoftheviewthatChina’scompetitiveadvantagevis-a-viscountriessuchas Mexico and post-apartheid South Africa stems primarily from an alleged similarity between its hukou system of household registration and South Africa’s system of residential registration under apartheid. Among the many reasons why Chinese wages are so competitive compared to other countries, Chan mentions “an almost inexhaustible supply of cheap labor from the countryside” and the absence of an autonomous union movement. But in her view the most fundamental reason is that China’s household registration system limits and regulates the rural-to-urban influx ofpopulationinwayssimilartothepasssystemunderapartheidinSouthAfrica(see alsoWhitehouse2006).Chanacknowledgesthatthetwosystemsdiffermarkedlyin origin and ideology: whereas the South African pass system under apartheid was intertwinedwithahistoryofracism,colonialismandcapitalistdevelopment—“allof whichfavoredthecontrolofmovementofAfricanpeopletoprovidegreaterpolitical security and enhanced efficiency in the use of black labor”—the hukou system in China was instituted in the 1950s as an integral component of a centrally planned socialisteconomy,whichmetthebasicneedsoftheurbanpopulationthroughration coupons that could only be used in the locality where they were issued. This required the registration and restriction of the geographical mobility of all people, not just peasants, and the granting of special temporary certificates to people who resided in a different locality than the one in which they were registered. In spite of its different origins, once central planning was abandoned—claims Chan—the way in which this registration system “can drive down wages and other labor standards today... is similar to what prevailed in [apartheid] South Africa.”5 Inalaterarticle,Chanandco-authorPeterAlexanderreferspecificallytoWolpe’s argument that the chief benefit of influx control for South African capitalism was 4Freeman(2005). 5Chan(2003:44–5). StCompIntDev(2010)45:410–438 441155 that it helped reduce costs to a minimum by buttressing the migrant labor system. Whereas with an urban workforce “capitalists had to provide an income that includedthecostsofreproducingthenextgenerationofworkers(food,housingand clothing for children and child rearers, and basic education), with migrant workers these costs could be offset by subsistence agriculture.” In spite of their different origins, according to Alexander and Chan the Chinese hukou system and the pass system under apartheid in South Africa generated similar outcomes. In China, as in SouthAfrica,thecostofemployingmigrantworkersfromtheruralareasislessthan that of employing regular urban workers and since the difference is “justified in terms of migrant workers retaining a home in the countryside where they could live for large parts of their lives, and where their dependents could remain... the essence of Wolpe’s analysis could also be applied to contemporary China.”6 This interpretation of China’s competitive advantage vis-a-vis countries such as post-apartheid South Africa isquestionableonseveralgrounds.Mostimportantly,if an apartheid-like system of labor control is as crucial a source of competitive advantage as Alexander and Chan claim, why did South Africa’s economic performance—as Table 1 shows—deteriorate so precipitously in the 1980s when apartheid was still firmly in place? As we shall see in Section III below, this sharp deteriorationpreceded(andthereforecannotbeattributedto)Chinesecompetitionin labor intensive manufacturing but must instead be traced to contradictions of the apartheid system of labor control that have no place in Alexander’s and Chan’s interpretation.Criticalinthisrespectwasanunintendedconsequenceofthatsystem: its tendency to inflate rather than reduce the cost of African labor. This contradiction figures prominently in Alan Hirsch’s implicit answer to our question of why South African wages are so high compared to China’s. Hirsch’s assessment of South Africa’s competitive disadvantages in labor intensive manufacturing is even more negative than Freeman’s. Not only are there “very few labor-intensive manufactured tradables that South Africa is likely to be competitive in;” in addition, this lack of competitiveness is relative not just to China but to countries like Vietnam, Thailand and even surrounding African countries where nothing resembling China’s hukou system is in place: “no matter how much the labor market is made more flexible or the currency is depreciated, South Africa will never have tens of thousands of workers making Nike shoes for export, like Vietnam or Thailand.”7 In Hirsch’s view this fundamental lack of competitiveness must ultimately be traced to two legacies of South Africa’s developmental path. One is the far more complete expulsion of South Africans from the land than in most other countries, including most other African countries. As a result, wages in South Africa must cover a much higher proportion of household expenditures than in countries where many workers have non-wage incomesderiving from their rural land-holdings.And the other legacy left behind by apartheid is the high cost of living for workers. ApartheidraisedlivingcostsforallSouthAfricans,especiallythepoor,oneof the main costs being apartheid-determined residential locations... The working 6AlexanderandChan(2004:614,621;emphasisadded) 7Hirsch(2005:182). 416 StCompIntDev(2010)45:410–438 poor were located miles from their potential places of work, and often equally far from commercial and public services. Public transport systems had practically collapsed by 1994, and some of the new distant locations were not served at all by public transport.... The deterioration of access to public services such as education, health, and social security for Africans under apartheid meant the diminution of the social wage.8 ThesecontentionspointtoadiagnosisofChina’scompetitivenessvis-a-visSouth Africa that departs significantly from Chan’s diagnosis. If the high cost of labor in South Africa can ultimately be traced to a particularly thorough process of dispossession of agricultural producers from the land on the one side, and to the inflationoflaborcostsandreductioninthequalityofthelaborforceassociatedwith apartheid ontheother,then China’scompetitive advantage cannot stemprimarily,if at all, from whatever similarities may exist between the hukou and the apartheid systems of residential segregation. This is the main thrust of a third diagnosis of the developmental advantages of China vis-a-vis South Africa, which departs even more radically than Hirsch’s from Chan’s. In summing up China’s developmental advantages, Gillian Hart focuses specifically on Chinese rural development and attributes much of China’s stunning economic growth to the fact that, despite the increasing concentration of ownership and control in many Township and Village Enterprises (TVEs), “at least part of the surplus from industry [was] retained, reinvested, and redistributed within local circuits, and directed towards schools, clinics and other forms of collective consumption.” Indeed, a key force propelling [TVEs] growth is that, unlike their urban counterparts, they do not have to provide housing, health, retirement, and other benefits to workers.Ineffect,muchofthecostofreproductionoflaborhasbeendeflected fromtheenterprise—but,atleastinsomeinstances,isbeingsupportedthrough redistributivemechanisms....WhatisdistinctiveaboutChinaandTaiwan—and dramatically different from South Africa—are the redistributive land reforms beginning in the late 1940s that effectively broke the power of the landlord class. The political forces that drove agrarian reforms in China and Taiwan were closely linked and precisely opposite. Yet in both socialist and post- socialist China, and in ‘capitalist’ Taiwan, the redistributive reforms that defined agrarian transformations were marked by rapid, decentralized industrial accumulation without dispossession from the land.9 This diagnosis shifts the focus from residential segregation as a mechanism of exploitation of rural workers in urban areas, on which Chan’s diagnosis is based, to ruraldevelopmentandimprovementsinthewell-beingoftheruralpopulationasthe underlying foundation of the comparative cheapness and high quality of Chinese labor. It complements Hirsch’s argument concerning the negative effects of the dispossession of the rural population on the competitiveness of South African labor by underscoring the positive effects of “rapid, decentralized industrial accumulation 8Hirsch(2005:182–3). 9Hart(2002:199–200). StCompIntDev(2010)45:410–438 441177 without dispossession from the land” on the competitiveness of Chinese labor. As Hart underscores, this interpretation of East Asian developmental trajectories has “a powerfulanddirectbearingonSouthAfricandebates,”notbecausethesetrajectories could be emulated “but rather because they denaturalize dispossession.” In order to appreciate their wider significance, Hart invites us “to revisit classical political economy debates, and revise the teleological assumptions about ‘primitive accumulation’ through which dispossession is seen as a natural concomitant of capitalist development.”10 In the next section we initiate the kind of revision that Hart advocates by re-examining three analyses that in the early 1970s gave rise to the Southern African paradigm of accumulation by dispossession. The Southern African Paradigm of Accumulation by Dispossession We begin this section with the analysis that first established the notion that the forcible dispossessionoftheAfricanpeasantry fromthelandwasthecentral feature of capitalist development in the ALR region—Arrighi’s critique of William J. Barber’s application of Arthur Lewis’ model of economic development to the Southern Rhodesian experience. We then turn to Martin Legassick’s analysis of “forced” labor and capitalist development in South Africa—a process that was far moreracializedandextremeinitsconsequencesonthesupplyofAfricanlaborthan in Rhodesia. After examining Wolpe’s re-conceptualization of apartheid as a purely repressivemechanismforthecontainmentofthecostsofAfricanlabor,wecombine thecomplementaryinsightsofthethreeanalysesintoasinglemodelofaccumulation bydispossessionthathighlightsthereasonsnotjustoftheinitialsuccessesofsettler capitalism in Southern Africa but also its contradictions and eventual demise. Arrighi’scritiqueofBarber’sinterpretationofeconomicdevelopmentinRhodesia had two components. One concerned the causes of the continual expansion of the supply of African labor throughout the first half of the twentieth century in spite of constant or falling real wages; the other concerned the causes of the increase in Africanrealwagesafter1950.Barberhadinterpretedbothtendenciesastheresultof a market-driven process that reallocated labor from a low productivity “subsistence sector” to a high productivity “capitalist sector.” Before 1950, he claimed, the indigenous rural communities were characterized by “periodic disguised unemploy- ment,” in the sense that about 50% of their total adult male population could be withdrawn for at least a full annual cycle without reducing the communities’ total production. Rural families could therefore increase their total incomes by dispatching this surplus male population to work in the capitalist sector, even for very low wages—that is, wages sufficient to cover only the costs of transport and maintenance of the single worker while he was in wage employment. Barber then went on to claim that up to the mid-1940s the proportion of able-bodied indigenous males in wage employment remained below 45%, so that the volume of African employment continued to expand despite a tendency of real wages to decline. After 1950,incontrast,theproportionofable-bodiedmalesclaimedbywageemployment reachedthe50%mark,sothatfurtherincreasesintheAfricanlaborsupplyinvolved 10Hart(2002:201). 418 StCompIntDev(2010)45:410–438 a fall in the agricultural output of the indigenous family. Under these circumstances —which Barber called of “quasi-full employment”—capitalist employers could attract additional workers from the subsistence sector only by increasing real wages to offset the loss in the real income of the family in indigenous agriculture and induce a break with accustomed ways of life.11 As can be seen from this account, the idea of a subsidization of capitalist production by communities of non-capitalist producers—which has often been attributed to Arrighi’s critique of Barber’s application of the Lewis model to Rhodesia—is in fact the central idea of the criticized model. Arrighi’s critique retained this idea but dismissed on empirical grounds Barber’s story of a market- driven process of accumulation without dispossession, in which disguised unemployment generated unlimited supplies of African labor at low and stagnant real wages until the continual reinvestment of profits in labor-intensive production created a situation of quasi-full employment that drove up African real wages. The alternative account that Arrighi proposed was based on three main observations. First, when export-oriented capitalist production began to develop in Rhodesia there was very little, if any, disguised unemployment in most indigenous rural communities, and in any event, these communities found it far more economical to sell produce rather than labor to the capitalist sector. The result was a crippling shortage of labor and rising wages, which held in check the capitalist expansion. Second,disguisedunemploymentandunlimitedsuppliesofcheaplaborwerecreated through a coercive state-driven process that undermined the capacity of the African peasantrybothtoparticipateinthemarketeconomythroughthesaleofproduceand to reproduce itself except by participating in the market economy. Integral to this process was the development of a white settler rural bourgeoisie that out-competed African producers thanks to state subsidies and state-backed appropriations of the most fertile and best located land. Third, African real wages in the post-Second World War period started increasing, not because of a situation of quasi-full- employment created by “capital widening”—i.e., investments that expanded the demand for labor—but because of interrelated changes in the structure of capital accumulation, in class and race relations, and in government policies.12 This last observation has largely gone unnoticed but is probably the most important in assessing the relevance of Arrighi’s critique of the Lewis model for an understanding of subsequent developments in Southern Africa and in the global economyatlarge.Wewillthereforebrieflyrecallsomeofitsdetails.Thecentralidea wasthattherapidgrowthofsecondaryandtertiary industries,alongwithincreasing mechanization and automation in mining and manufacturing, created a demand for an African labor force with skills that were simple but could not be imparted under conditions of high turnover. The creation of a more stable African labor force, however, presupposed the abandonment of the tradition that fixed customarily African wages at a level sufficient to cover only the subsistence of single men. As long as this tradition persisted, Africans in wage employment had to rely on the subsistenceeconomyforthemaintenanceoftheirfamiliesandofthemselvesduring old age, sickness and unemployment. Participation in wage employment thus left 11Barber(1961:46,93,186–87,208,212–18). 12Arrighi(1970:200–22). StCompIntDev(2010)45:410–438 441199 unchanged the worker’s obligations to his rural kinsmen as well as his involvement inthesubsistenceeconomy.Thecreationofastablewagelaborforcethatwouldnot periodically move to and from the peasant sector, in contrast, required wages that would provide Africans with some security during their working life and old age and, above all, that would enable them to support their families outside the peasant sector. “Stabilized labor,” in other words, “commanded a premium determined by the difference between the cost of the means of subsistence of single men during their working life in wage employment and the cost of the means of subsistence of the worker’s family over his ‘life cycle.’”13 This contention was supported by the fact that most of the increase in African wages after the early 1950s had occurred in industries that were most in need of a stable labor force (manufacturing, transport and communication), whereas in agriculture, where stabilization mattered least, the increase in wages had been minimal. However, the tendency could not be attributed merely to changes in the structure of the demand for labor. Thus, “the ‘rush’ for education of the late 1930s and early 1940s... facilitated the subsequent politicization of the African masses.” Afterthewar,agreaterawarenessoftheirincreasinglyproletarianstatusledAfrican workerstoseekanimprovementoftheirlivingconditionsasproletariansratherthan migrantpeasants.Theresultwas“awaveofstrikesthatmadethelate1940saperiod of African labor unrest of unprecedented intensity and scale.” Developing in conjunction with the growing influence of a manufacturing capitalist class with an interestinlaborstabilizationandintheexpansionoftheinternalmarket,thewaveof strikes prompted the Rhodesian government to raise basic African wages and to introduce a new classification of jobs in industry that contributed further to the increase in African wages.14 The changes in the structure of capitalist production that promoted or facilitated the increase in African wages made the Lewis model irrelevant to the Rhodesian experience in yet another and even more important way. Foreign controlled oligopolies, characterized by considerable ‘international mobility’, had come to dominate important sectors of the [Rhodesian] economy (mining and secondary industries), while the financial and entrepre- neurial ‘entrance requirements’ in most branches of production had greatly increased. As a consequence, prices had lost much of their downwards flexibility and even when changes in relative profitabilities did occur little inter-sectoral mobility of capital could be expected. Moreover, the greater calculating rationality of the large oligopolies relatively to the atomistic producers of earlier times implied a greater dependence of the rate of accumulation upon the absorptive capacity of the market.15 13Arrighi(1970:223). 14Arrighi(1970:223–4). 15Thedifferentdynamicsoftherecessionsoftheearly1920sandofthelate1950svividlyillustratesthe change.Intheearly1920s,fallingpricesandwagesinducedbusinesses(especiallyinagriculture)tostep upinvestmentsandabsorbtheincreasingly“unlimited”supplyofcheapAfricalabor.Inthelate1950s,in contrast,therecessionledtoanoutflowofcapitalandacontractionininvestmentandthedemandfor labor.Arrighi(1970:225–6).
Description: