SUGGESTIONS AND DEBATES Revisiting White Labourism: New Debates on Working-Class Whiteness in Twentieth-Century Southern Africa

role of white workers outside British imperial networks; we examine how radical and revolutionary ideas disappeared from white-working class politics in the mid-twentieth century; and we reassess the connection between transnational flows of people and ideas. Racial divisions in the working class and labour movement in Southern Africa were persistent and enduring. We argue that racial segregation had an enduring appeal to white workers in Southern Africa, and the sources of this appeal were more varied and locally rooted than simply transnational migration to the region.

role of white workers outside British imperial networks; we examine how radical and revolutionary ideas disappeared from white-working class politics in the mid-twentieth century; and we reassess the connection between transnational flows of people and ideas. Racial divisions in the working class and labour movement in Southern Africa were persistent and enduring. We argue that racial segregation had an enduring appeal to white workers in Southern Africa, and the sources of this appeal were more varied and locally rooted than simply transnational migration to the region.

I N T R O D U C T I O N
The term 'white working class' has become commonplace in recent years: a convenient, though poorly defined shorthand ostensibly explaining the rise of xenophobic nationalist politics in Europe and North America in the wake of the  financial crisis.  As readers of this journal will be keenly aware, however, this term has a history andespecially among historians of Southern Africaevokes a period and a set of meanings quite different from present-day connotations. A rich social history literature has detailed the everyday lives, forms of organization, subjectivities, and politics of the white workers who played a key role in the early-twentieth-century industrialization of this region.  This scholarship has demonstrated that they were internationally mobile, militant, and collectively organized, and that they claimed the status of "white" as an assertive self-designation. The same was true in other imperial contexts, like Australia, where many workers and their collective organizations combined a radical critique of economic exploitation with demands for race-based labour protection.
In a seminal article published in , Jonathan Hyslop proposed the concept of "white labourism" to encapsulate this understanding of the white working class and to theorize what he argued were the profound similarities between different imperial settings and the British metropole in the early twentieth century. Britain and its settler colonies, Hyslop argued, were linked by flows of labour migrants who formed an "imperial working class" and white labourism's anti-capitalist militancy and racism was the common ideology of this class.  At the time, Hyslop was at the forefront of the emerging field of transnational history. While his arguments evoked much debate, most notably in the pages of this journal, white labourism has become a widely authoritative concept in subsequent literature.
This article revisits the white labourism debate in the light of newly emerging scholarship. This includes, but is not limited to, our recent edited collection Rethinking White Societies in Southern Africa, s-s,  which prompted the invitation from the International Review of Social History for this contribution. The volume brings together studies of non-hegemonic whitesthe poor and working classin Angola, Mozambique, South Africa, Zimbabwe, and Zambia during the era of white minority rule. It demonstrates how racialized class identities and the divided nature of white societies in the region, often associated by scholars with the early s, persisted for much of the twentieth century. With these arguments, we challenge prevalent tendencies in the existing literature to treat white society as monolithic  and thus to consider race independently from class.
This article adopts the same chronological frame espoused by the edited collection. The debate on white labourism has primarily focused on the period between the s and World War I. Examining a longer period allows us to critically assess some of the claims made in the debate, and to trace how some elements of this ideology endured while others dissipated. In so doing, we seek to expand the existing debate in three directions. First, we consider workers located outside the orbit of British imperial flows, whose role in shaping white labour ideologies in the late nineteenth and early twentieth century has been neglected in this debate. Second, we discuss the dwindling radicalism of white working-class politicsan essential element of white labourismand the consolidation of ideas of racial segregation occurring simultaneously from the interwar period. Third, we contest the automatic linkage between international migration and the spread of ideas that white labourism assumes, by considering white working-class ideology in the region over the longer durée. We show that racial segregation had an enduring appeal to white workers in Southern Africa, and the sources of this appeal were more varied and locally rooted than simply transnational migration to the region. In what follows, we offer a brief overview of the main contours of the debate around white labourism before concentrating each of these three interventions.
It is not our intention to discredit Hyslop and others' arguments. On the contrary, by demonstrating the value of taking a longer chronological view and considering the afterlife of white labourism in twentieth-century Southern Africa, we demonstrate how existing concepts and scholarship may be re-evaluated and invigorated in the light of new research, thus opening up new avenues and opportunities for debate.

W H I T E L A B O U R I S M : C O N C E P T A N D D E B AT E
Hyslop's seminal  article proposed that the white working classes of the pre-World War I British Empire were not nationally distinct, but bound together through flows of migration across the Empire. The labour movement that developed from this imperial working class was characterized by a common ideology of white labourisma synthesis of hostility to capitalist exploitation and racism. Hyslop demonstrated the depth of these transnational bonds in a vignette describing a  mass demonstration in Hyde Park. In a vivid illustration of imperial, racial labour solidarity, the demonstration saw half a million British trade unionists and socialists express their support for white workers halfway across the globeon the goldfields of South Africawho were demanding the reservation of skilled work for whites. The militant labour and racist visions expressed in the metropole and on the Rand were, therefore, not separate phenomena but part of a "single story" linking "the rise of working-class racism in turn-of-the-century Britain, of the beginning of South African industrial segregation, and of the politics of the 'White Australia' policy".  The white working class in South Africa was not a peculiar and unique phenomenon, but reflected and inflected the labour movement elsewhere in the British Empire amid the reality of transnational labour migration.
Hyslop's intervention came in a particular historiographical context. Labour history was in abeyance during the s, and the "new labour history" was grappling with the question of race and whitenessitself the subject of a debate in this journal.  David Roediger's pathbreaking  book The Wages of Whiteness upended many central assumptions in American labour history, and closely informed Hyslop's concept of white labourism, particularly the idea that the white working class "made itself white".  These historiographical developments were further informed by the beginnings of a turn towards transnational history that has only accelerated since.  More specifically, Hyslop's arguments sought to critique Alan Bonnett's  article, which explained twentieth-century British working-class racism in terms of the class struggles of nineteenth-century Britain. In the metropole, Bonnett contended, the upper classes presented working people as lacking the necessary characteristics and status associated with whiteness, while in the colonies, all Europeans were regarded as possessing the defining "civilized" qualities of whiteness.  Hyslop criticized Bonnett for his narrow focus on national context, and, drawing on the work of Ann Stoler and Fred Cooper, argued for empire rather than the nation state as the appropriate analytic frame for labour ideology in this period.  Broader imperial connections and experiences meant that, across the empire, white workers perceived employers as attempting to undermine the organized power of white workers internationally by subjecting them to competition from African and Asian workers. Workers' anti-capitalist struggles therefore simultaneously expressed their aspiration to be included in the dominant political and racial order. Militant action to secure racially preferential labour policies and citizenship rightswhether in Britain, South Africa, or Australiatherefore constituted the early-twentieth-century imperial working-class struggle to "make itself white".  Hyslop's concept was soon scrutinized by other scholars. Lucien van der Walt criticized Hyslop's seemingly automatic equation of an imperial white working class with the ideology of white labourism. Van der Walt argued that the international white labour movement was not ideologically homogenous and that the "Empire of Labour was only one outlook, one project, within a larger British working-class diaspora and larger white working class in the British Empire".  In this regard, Van der Walt stressed the importance of anarchist and syndicalist ideas circulating in the labour movementstrands that challenged racial segregation by advocating "one big union" across national and racial divides. Van der Walt demonstrated how more radical and internationalist ideologies travelled along the same imperial labour routes identified by Hyslop, competing with white labourism as they spread through Southern Africa.  The International Review of Social History became the main forum for this debate when it published William Kenefick's challenge to Hyslop and Hyslop's response in , followed by an article by Duncan Money in .  Kenefick, like Van der Walt, questioned the extent to which white workers in imperial labour circuits were united by a shared ideology. He argued that "non-racialists and anti-segregationists did mount a serious challenge to the prevailing ideology of white labourism" and that this anti-racist and internationalist challenge "was disproportionately influenced by radicalleft Scottish migrants".  Kenefick also argued that Hyslop overstated the dominance of white labourism in Britain itself. Here, Kenefick drew on Neville Kirk's study of labour movements in Britain, Australia, and the United States before World War I, which highlighted the attitudes of British socialists towards non-white workers as "far more searching and anti-racist than suggested in much of the recent literature".  The  solidarity with white workers in South Africa expressed in Hyde Park had soon waned, and the British labour movement turned its attention to the miners' strike in British Columbia, Canadaa dispute, Kenefick claims, "fought strictly along class lines".  Hyslop responded to both Van der Walt and Kenefick by conceding the existence of anti-racist and syndicalist politics among white labour activists. Nevertheless, he argued that anti-racist and syndicalist ideologies remained restricted to radical labour activists and did not have "any substantial impact on the membership of the white trade unions, or [on] the British emigrant working class more broadly".  While debate on race did occur within the South African labour movement, "anti-racist positions were consistently defeated or marginal".  Moreover, white trade unionists in Southern Africa, whether they publicly opposed white labourism or not, did not organize black workers in this period, with only minor exceptions. Hyslop reiterated that white labourist ideas had widespread support in the British labour movement in this period, and that Scottish workers were not distinct from their English counterparts in this respect. He pointed to the campaigns against the employment of African and Asian seamen in British ports, as well as the lack of response in Britain to the harsh repression of protests by Indian coal miners in South Africa in . Claims that metropolitan British society was not fundamentally shaped by the effects of empire, especially with regards to race, amounted to "imperial denialism", Hyslop charged.  While Hyslop, Van der Walt and Kenefick disagreed on the extent to which white workers of the British Empire were united by a common ideology of white labourism, their respective arguments shared the same chronological focus. Money, however, entering the debate in , presented a case study of a white trade unionist deported from colonial Zambia during World War II to demonstrate that white labourism was still apparent in the s. While its expression had changed, as white trade unionists avoided making explicit racial appeals by adopting internationalist language, this did not alter the fact that on the Copperbelt "radical and internationalist components of the white labour movement sat within a framework of racial exclusivity".  Moreover, in the context of continued white labour migration, the Copperbelt's white mineworkers received support from Britain and Australia in their struggles, even as their whites-only union imposed a closed shop and colour bar on the mines. White mineworkers in s Northern Rhodesia, Money argued, continued to see themselves as part of an international racially delineated class.
To date, therefore, the white labourism debate has revolved around the extent to which white workers of the British Empire were united by a common ideology characterized by racist, anti-capitalist sentiments, and to what period the concept of white labourism should be applied. Yet, the debate remained firmly centred on British labour circuits within its imperial analytical focus. In the following sections, we challenge this imperial framework and extend the chronological focus to re-evaluate transnational connections and highlight the local.

W O R K E R S A N D I D E O LO G I E S O U T S I D E B R I T I S H I M P E R I A L C I R C U I T S
Labour migration has been at the centre of the experience of work in Southern Africa.  The migration of white workers occurred in parallel with much larger flows of African labour -South Africa's mining industry, for instance, recruited from as far afield as modern-day Tanzania. By , just a few years after the discovery of gold, , African migrants were working on the Randtwenty years later, the figure had soared to almost ,.  In addition, there was substantial recruitment of indentured Asian workers, including the short-lived recruitment of over , Chinese to work on the Rand's gold mines in the early s. This provoked a hostile reaction from white trade unionists, supported by workers elsewhere in the British Empire, leading to an exclusion of Asian labour on the mines. Mae Ngai, drawing in part on the concept of white labourism, argued that such exclusion became a defining feature of settler colonialism in the Anglo-American world and that white trade unions were convinced that British workers could "populate, work and prosper in the settler colonies […] only if Asians were altogether excluded".  Yet, this white labour force was not composed entirely of recent migrants from within British imperial circuits. Van der Walt, writing on the South African context, gestured towards this when writing that: "White Labourism […] was almost entirely a movement of English-speaking whites, and never drew in any real number of Afrikaner workers."  Yann Béliard, in turn, argues that the solidarity of British trade unionists with white workers in South Africa was shaped by an imperialistic vision that excluded Afrikaner workers.  Such statements raise a potential blind spot in debates around the ubiquity of the ideology of white labourism in the early twentieth century and its role in the development of a racially segregated working class. Focusing on Afrikaner workers, who were not embedded in transnational networks, shows how this ideological landscape was even more broadly constituted and, crucially, also shaped by distinctly local forces.
South Africa's white Afrikaans-speaking workers are typically associated with the rise of Afrikaner nationalism and seen as a key constituency responsible for electing the National Party (NP) on its apartheid platform in . However, these workers did not suddenly appear on the scene in they had been a key feature of South Africa's young industrial landscape for as long as most immigrant white workers. Yet, there is comparatively little discussion of Afrikaner workers before .  From the late nineteenth century, Afrikaans-speaking whites represented a rising proportion of the white industrial workforce in South Africa. This accelerated around , as economic depression, the rural destruction and displacement of the South African War, and the ravages of the rinderpest epidemic created an influx of impoverished Afrikaans-speaking whites into the urban economy of the Witwatersrand.  Many found employment in mining, benefitting from changes in the production process leading to job fragmentation and the deskilling of certain tasks.  This rendered distinctions between skilled, semi-skilled, and unskilled work ever more ambiguous, and a growing number of newly urbanized Afrikaners came to work alongside professional miners from abroad. Unskilled whites were also appointed to perform supervisory work, overseeing work performed by Africans. By , these developments had effectively rendered the craft unionism of the main white miners' organization obsolete, and it transitioned from a craft to an industrial union. The following year saw the outbreak of World War I, drawing many immigrant white miners to enlist in the British armed forces. By , Afrikaners formed the majority in underground jobs.  Clearly, the international migration routes and circuits that brought many white workers to Southern Africa did not apply to the newly proletarianized Afrikaners. Their mobility was not born of skill and imperial citizenship, but of poverty and landlessness, their migration a much shorter but bitter route from the countryside to the Rand. Here, they certainly came into contact with the white labourism of their British co-workers. Historians have yet to elucidate the nature of this encounter.  We may imagine certain affinities in terms of racial attitudes and anti-capitalist sentiment, although these would have been expressed in very different terms. The racism of newly urbanized Afrikaners reflected the racialized master-servant relations of their agrarian colonial roots, while their aversion to capitalist enterprise and hence to exploitation would have been a product of long-standing hostility to "merchants, bankers and other agents of British financial capitalism".  At the same time, possible convergences with white labourism and potential . Giliomee, The Afrikaners, p. . Overcoming such anti-capitalist sentiments common among Afrikaners was one of the central aims of the Afrikaner nationalist movement of the s. Giliomee notes that, by this point, there were few Afrikaner enterprises in the "classic capitalist mould" -most were mutual corporations or farmers' co-operatives, while some explicitly nationalist companies functioned at a loss. This was compounded by an "excessive 'familism'" in the Afrikaner community: "people entrusted their capital only to their own immediate family, rarely to those outside and certainly not to strangers or a company". Giliomee, The Afrikaners, p. . solidarities would have been complicated by language barriers, skills differentials, and the animosity towards the British Empire born of the South African War. Indeed, Afrikaans workers "were all too conscious of imperial subjugation".  This, rather than any form of labour organization, would have been the main reference point of any militant expression on their part.
The diverse and intersecting strands of ideology circulating among South Africa's white industrial labour force at this time are best revealed in the events of the  Rand Revolt, masterfully analysed by Jeremy Krikler. The background to the strike is well known. In December , the Chamber of Mines announced its intention to replace , white miners in semi-skilled work with black workers, whose labour was significantly cheaper. The broader white mining workforce feared that it was simply a matter of time before they faced the same fate. Already battling rising living costs and now facing unemployment, they reacted with outrage. In January , miners on gold and coal mines downed tools, and were soon backed by a general strike throughout the Transvaal. The strike turned into an armed rebellion, as , workersthe majority of them Afrikanerschallenged the power of mine owners and the legitimacy of the South African state that supported them.
Krikler has described the ideological impulses behind this revolutionary challenge as taking two main forms. On the one hand, many Afrikaner workers understood the strike as a revolt against British imperialism and sought the formation of an independent republic, reminiscent of the Boer republics lost during the South African War. Indeed, the strikers deployed this military experience in service of their proletarian insurrection, mobilizing in military formations or "commandos" as per war-time Boer tactics. On the other hand, Krikler identifies the prevalence of communist revolutionary sentiments among many strikers, including those of British extraction. They, too, saw the strike as an opportunity to overthrow the state, but understood the latter as an instrument of capitalist exploitation, rather than British imperialism. These various expressions reflected both the diversity and the complex enmeshment of race and class militance characterising the white labour movement. This was demonstrated by the infamous symbol of : the strikers' banner reading "Workers of the world, unite and fight for a white South Africa".  But more dramatic and revealing was the intense violence that characterized the strike, in which some  strikers, soldiers, police, and civilians were killed. While this "small scale civil war"  mainly involved conflict between white workers and white soldiers and police, white strikers murdered over forty Africans at the height of the insurrection. Krikler interprets these racial killings as expressing anxieties inherent in the formation of white working-class identity in the context of politicized racial imaginaries and intense class struggle. The white workers of the Rand, Krikler argues, had come to define themselves in relation to "that which they were not: rightless, wageless, racially-despised, unfree blacks". The Chamber of Mines' efforts to undermine the colour bar was understood as an onslaught on white workers' racial identity and privilegean onslaught that, if successful, would "ground [them] down into poverty" and render them "white kaffir[s]".  Indeed, the conflict of  occurred as African labour organizations were gaining momentum and African political leadership was becoming increasingly vocal. The racial killings of March  formed part of broader incidents of violence, which saw white workers target African trade union organization on the Witwatersrand.  Yet, white working-class animosity and insurrectionary violence during the strike was overwhelmingly directed against white employers and the white state. Indeed, Krikler argues that the "White South Africa" for which they were fighting constituted "a particular organization of state, society and economy" in which white workers would not be at the mercy of the industrial despotism of employers, but be recognized as citizens of equal importance to other classes in the white community.   can be seen as the most dramatic manifestation of the Rand's white working class struggling to "make itself white". In this sense, it resonates closely with Hyslop's description of the struggles of the imperial working class more broadly in this period.  Indeed, in his  article, Hyslop argued that the strike's infamous slogan expressed notions that "prevailed in the British Empire labour movement before the First World War", and shows the commonality with the "White Australia" slogans of the Australian labour movement.  Yet, the Rand Revolt appears only in passing in the debate on white labourism. As Krikler's work showed subsequently, other influences were also at play here, and labour ideologies emanating from British imperial circuits were not the sole or even main defining feature of early-twentiethcentury working-class militancy in Southern Africa. Rather, these intersected and functioned in concert with the particular race and class interests and subjectivities of Afrikaner workers born of local political and economic processes. Recent research by Nicola Ginsburgh extends this argument about the importance of local processes to self-conscious British white workers in Southern Rhodesia in the same period. Ginsburgh, nodding to Hyslop's . All quotes Krikler, The Rand Revolt, pp. , . "Kaffir" is a highly offensive racial slur for blacks, common in colonial-and apartheid-era South Africa. notion of white labourism, shows how the colony's white labour movement drew on members' experiences in Britain and South Africa. However, she also stresses that "the radicalism and racism of [these] white workers was not simply a matter of transposing ideas from one context to another, they were rooted in capitalism and engendered through their experience of the racial monopoly of higher paid, skilled work".  Attention to local processes therefore sees white labourism emerge more clearly as a constituent part, rather than the sum, of white working-class ideology in early-twentieth-century Southern Africa.

W H I T E W O R K I N G -C L A S S P O L I T I C S A N D T H E D E AT H O F P O L I T I C A L R A D I C A L I S M
Whereas, today, anti-capitalism and racism are typically seen as part of opposing ideologies, Hyslop's work shows that this was not historically the case.
In the context of the early twentieth century, white labourism's radical challenge to the power of capital and the structure of the state had clear commonalities and affinities with left-wing and revolutionary currents also circulating at the time. Combining opposition to economic exploitation with white supremacy made perfect sense to white trade unionists resisting replacement by cheaper "non-white" labour and demanding the full inclusion of the white working class in the body politic through racially preferential legislation. As we have seen, these ideas were contested, but one crucial feature of the pre- period is that white labourists, anti-racist syndicalists, and anti-imperialist Afrikaner workers were part of the same landscape of white working-class ideology and engaged in the same struggles as white industrial workers. But this form of white working-class radicalism rapidly dissipated after the Rand Revolt, as white workers ceased to challenge the existing order. Indeed,  is widely recognized as the climax of white labour militancy in South Africa, and the catalyst for what has been called "the almost complete political capitalisation of the 'white labour movement' to capital".  Scholars agree that the violent quashing of the strike by the state led to the ousting of the ruling South African Party in the  general election in favour of a National Party-Labour Party coalition or Pact government. The Pact represented an alliance between Afrikaner nationalists and white labour representatives, and it legislated preferential employment and wage increases for white workers. More fundamentally, however, was its enactment and extension of industrial conciliation mechanisms that ensconced white workers' privileged position in the labour hierarchy. Historians have understood these measures as co-opting white labour into formal, state-controlled structures of power and bureaucratizing their trade unions. Employersboth local and internationalparticipated in this process.  David Yudelman cites the dramatic decline in white strike action after  as evidence of this co-optation.  Dan O'Meara concurs, holding that by the s, "the leadership of the white trade union movement had been incorporated into the formal structures of power […] confrontations between capital and organized white labour were thus ruled out".  In Southern Rhodesia, labour legislation was directly copied from South Africa in  and arguably had the same results, with strikes by white workers ceasing almost entirely.  The link between anti-capitalism and racism that marked white workingclass ideology also started to weaken and dissolve after the Rand Revolt. Instead, these elements came to be regarded as separate and opposed. Industrial developments during the interwar years produced complex and countervailing ideological shifts within the white labour movement, pushing some white workers towards multiracial trade unionism and others towards firmer racial segregation. While the Pact's race-based labour policies strengthened the existing racial division of labour, such divisions were not simply reproduced by industrial developments. Rapid expansion of the manufacturing industry from the s created workplaces where African, coloured, Indian, and white workers often performed similar functions in the labour process. According to Jon Lewis, this meant that here, "class determination […] was not affected by racial categories, and that, therefore, a basis for interracial class alliance existed".  This facilitated the emergence of new multiracial industrial unions in contrast to the existing typically whites-only craft unions that dominated older industries such as mining and the railways. This was exemplified by the transformation of the Garment Workers Union (GWU) into a multiracial organization representing Afrikaner, coloured, and African garment workers under a combative socialist leadership.  Lewis regards these new unions as evidence of "genuine inter-racial solidarity" displayed by white workers.  Meanwhile, in the context of the Depression, production workers in the garment industry as well as railway artisans perceived the state to be supporting employers in pressing for mechanization and de-skilling, thus endangering their positions. Consequently, white industrial action intensified in the early s, and many white workers withdrew their support for the Pact government. White workers in mining and the railways, who depended on "civilized" labour policiespreferential employment schemes for white workersand job reservation, however, remained loyal to the Pact.  Political radicalism was evident among the white labour movement in the interwar periodmembers of the GWU visited the Soviet Union in , for instanceand briefly flourished during World War II. The electoral fortunes of white labour parties across the region revived, and the Communist Party achieved hitherto unimagined levels of respectability by dint of the wartime alliance with the Soviet Union. In the mid-s, Johannesburg and Salisbury (Harare) were headed by Labour Party mayors. South Africa even saw the emergence of a progressive, anti-racist mass movement formed by white ex-servicemen, the Springbok Legion. The Legion's leadership was aligned with the Communist Party and several leading members had been involved in the white labour movement.  "The outcome of this political battle between left and right", says Fine, "was not written in advance".  World War II saw industrialization intensify across the region, leading to an unprecedented move of skilled and semi-skilled black men, white and coloured women, as well as unskilled whites, into secondary industry.  Race and privilege therefore did not necessarily run together, and this period saw "significant organic links"  established between white and black labour. There were major strikes by African workers across the region. In , existing African trade unions formed the Council of Non-European Trade Unions, in which some white radicals were active. Fine provides numerous examples of cooperation between white and black workers, and of white workers' enthusiasm for socialist principles during the s.  Alexander similarly interprets the s as a "multiracial moment in South African history".  At the same time, organized labour was becoming more polarized: nonracial, socialist sympathies persisted amongst the new industrial unions, while craft unions wary of de-skilling warmed to race-based legislation. Mainly Afrikaans-speaking, unskilled workers, meanwhile, were increasingly gravitating towards nationalist and explicitly segregationist unions  -a process actively pursued by the National Party. Successfully combining economics and ideology, it offered white workers policies that would safeguard their positions, increasingly winning their allegiance. The Nationalists also exploited existing weaknesses within the labour movement, sowing discord amongst leftist unions. These dynamics are best illustrated by events within the Mineworkers' Union (MWU). The union's membership was overwhelming Afrikaner by the s, while the leadership remained English-speaking and supportive of the Labour Party.  Afrikaner nationalists had identified white mineworkers as a key constituency to win over and sought to gain control over the union. In an intense struggle for control of the union (which included the assassination of its general secretary) the Nationalists linked union corruption and failed wage negotiations to the political orientation of the union leadership. While, in , all six mining constituencies on the Rand had returned Labour Party candidates, by  these same constituencies proved decisive in bringing the NP to power.  A few months later, the MWU elected a new NP-orientated leadership.  The Nationalists' increasing upper hand was also reflected in the fortunes of its political opponents. When the GWU established the Independent Labour Party in  -its name a clear indication of the persistence of links to the British labour movementto mobilize around "bread and butter" economic issues it considered pertinent to the GWU's membership, the party performed dismally.  Indeed, the white labour parties in South Africa, Southern Rhodesia and Northern Rhodesia all splintered in the mid-s, mostly over disagreements about racial segregation, and thereafter swiftly declined into irrelevance. The NP's electoral victory and subsequent implementation of apartheid sounded the death knell for multiracial trade unionism and its political expression. As the garment workers' leader Johanna Cornelius lamented in , "somehow the Nationalist [sic] Party with their vile propaganda seem to have a better hold of them [white GWU members] than we can muster".  Multiracial unions were subsequently prohibited, and the GWU's commitment to inclusive industrial trade unionism petered out.  The s concluded with the defeat of socialist ideas and independent trade unionism among white workers, making them a "prop for the apartheid state".  Taking this longer chronological view facilitates an assessment of anti-racist ideas within the white labour movement and their historical contingency throughout the first half of the twentieth century. By the s, the anti-racist challenge to white labourism by radicals and syndicalists, highlighted by Van der Walt and Kenefick, left virtually no legacy among the white labour movement.  Political radicalism among the white working class effectively disappeared after the mid-twentieth century, both in terms of revolutionary ideas of communism and anarcho-syndicalism, and in the sense of "labourism" that had at least some resemblance to social democratic politics elsewhere in the world. White workers became more dependent on legislative colour bars enforcing workplace segregation and upon a state where only whites held meaningful political rights.

S H I F T I N G PAT T E R N S O F T R A N S N AT I O N A L M I G R AT I O N A N D W H I T E W O R K I N G -C L A S S C I T I Z E N S H I P
Hyslop's concept of white labourism and the subsequent debate surrounding it remained concentrated on the late nineteenth and early twentieth century.
A key reason for this is that the transnational networks along which white labourism travelled weakened in the interwar years. In the context of the Great Depression, spiking unemployment prompted governments around the world to restrict working-class immigration, often on the insistence of local trade unions.  National boundaries hardened as the interwar slump and protectionist policies turned states inward, and the circuits of white labour migration that had once connected Britain, the United States, Australia, and

Duncan Money and Danelle van Zyl-Hermann 
Southern Africa started to break down. Hyslop and Van der Walt agree that "the s arguably marked the onset of a period in which working-class people and movements were increasingly nationalised".  Yet, transnational white migration revived from the s. During the World War II, South Africa and Southern Rhodesia became hubs for training and convalescing military personnel, and over six million Allied servicemen and almost , civilians spent time in Southern Africa during the war.  In the post-war period, population figures demonstrate the scale of European migration to Southern Africa. In , there were around , whites in Southern Rhodesia, , whites in Angola, and , in Mozambique.  By the mid-s, these figures had soared due to migration, with some , whites in Southern Rhodesia, , whites in Angola, and , whites in Mozambique.  Between  and , over , whites emigrated to Southern Rhodesia and , emigrated to South Africa.  Many immigrants relocated of their own accord, attracted by the opportunities of the post-war economic boom. At the same time, white minority regimes instituted programmes to encourage white migration to Southern Africa. This was motivated as much by ideological concerns as it was by practical considerations, as the economic upturn in the context of a racially segregated labour market created significant skills shortages. In , for instance, an official inquiry warned that South Africa's industrial growth could not be sustained unless steps were taken to attract more skilled white labour, prompting the state to launch a large scaleand successfulcampaign to attract white migrants.  In the same period, Southern Rhodesia ran a similar, albeit more modest, campaign. The bulk of the migrants taking advantage of these programmes came from Britaina fact that has been obscured by the political separation occurring between settler governments and metropole at the time. South Africa maintained a scheme of financial assistance for white migrants until .  Portugal, too, initiated state-sponsored settlement schemes for its colonies of Angola and Mozambique, including attempts to create agricultural colonies where Portuguese peasants provided the labour.  Crucially, not all Europeans relocating to Southern Africa through these various schemes were settlers in the conventional sensemany moved frequently in the region and beyond, often along the same circuits of labour migration established early in the century. Indeed, recent research has revealed high levels of migration between Britain and its former colonies long after the Empire had disintegrated.  By mid-century, as labour migration revived, attitudes on the acceptability and desirability of racial segregation had changed markedly within the labour movement. Indeed, from the s, opposition to racial segregation and colonialism, and to white supremacy in Southern Africa in particular, became one of the few points of principle that cut across the international labour movement otherwise divided by Cold War cleavages. European trade unions, particularly Britain's Trade Union Congress, provided crucial moral and material support to African trade unions in South Africa, where they operated under intense pressure from the state.  In Australia, meanwhile, the Labour Party and trade unions, which had once been at the forefront of establishing "White Australia", pressed for the abolishment of this policya goal achieved in .  The close connection between movements of people and the transfer of ideas is essential to the transnational history espoused by Hyslop, as flows of white migrants are seen as conduits for ideologies like white labourism. Half a century later, European migration did not have the same impact. Radical ideas becoming mainstream in the global labour movement were not adopted in Southern Africa as labour mobility and migration resumed. This is not to say that such ideas were not travelling to the regionpolitical radicals were certainly among the many whites who came to Southern Africa in these years. Doris Lessing, for instance, described how the arrival of thousands of working-class RAF recruits and European political refugees in Southern Rhodesia animated left-wing politics in the colony, and prompted the formation of a short-lived communist party. Yet, when these groups left after the war, left-wing organizations in the territory quickly fizzled out.  The trajectory of racist attitudes in Southern Africa following World War II is well knownthe region diverged politically from the rest of the former British Empire and Commonwealth, and the imperial connection was severed. As the post-war collapse of European colonial empires hardened support for white minority rule in the region, the white labour movement in Southern Africa similarly shifted firmly to the right. Unsurprisingly, this period also did not see the kind of transnational connections between the labour movements that had been forged and sustained by migration earlier in the century. The post-war era saw connections between Southern Africa's white labour movement and their international counterparts, previously close and convivial, consciously severed. This change happened rapidly. In the s, representatives of the Copperbelt's white mineworkers visiting Britain were given a warm welcome by British trade unionists, especially from the National Union Mineworkers (NUM). Yet, by , a delegation of white mineworkers visiting Britain to solicit support for their situationvis-à-vis the possibility of Zambian independencewere politely but firmly rebuffed by the NUM and informed that the union's policy "was one of support for the rights of selfdetermination for the peoples of Africa".  This put-down contrasts sharply with the mass solidarity British workers displayed towards their white South African counterparts during the  Hyde Park demonstration.  In South Africa, the post-war death of labour radicalism was largely due to intensified state control. The  Suppression of Communism Act gave the government the power to remove any trade unionist deemed to be a communist from office, and the NP deliberately conflated communism with multiracial trade unionism. Liberation movements were banned and those opposed to the apartheid state paid a heavy price. While a small number of whites remained in the banned Communist Party, few of these had come through the ranks of the labour movement. . In fact, as early as , Clements Kadalie had condemned the white labour movement, and offered a remarkably prescient forecast of subsequent developments when he argued that if white workers "were not prepared to recognise the native as a brother worker", then "they would find that the workers overseas, particularly those of Great Britain, would have nothing to do with them". National Archives of South Africa, Pretoria, JUS , //, Native Agitation Reports On: Part , CID Report,  November . We are grateful to Henry Dee for sharing this file with us. . A small numbers of white radicals also participated in the anti-colonial movements in Angola and Mozambique.
South Africa's small number of white radicals were marginal figures and generally maligned by white society, playing a meagre role in the white labour movement after the s.  By the s, there was little evidence of radical and revolutionary ideas among whites. The most high-profile action by a white revolutionary in this period was Dimitri Tsafendas' assassination of South Africa's Prime Minister Hendrik Verwoerd, the so-called architect of apartheid. Tsafendas was a Mozambican-born, working-class communist whose biography resembled the migratory workers described by Hyslophe had worked as a seaman, among other jobs, in North America, Europe, and the Middle Eastbut he was such a marginal figure that until relatively recently he was regarded as mentally ill, rather than as a dedicated revolutionary.  But Tsafendas was a dramatic exception. Unlike the period culminating in the Rand Revolt, by the s there was little to be gained for whites by participating in revolutionary politics. Whether due to the threat of state power, or the wages of whiteness secured through the co-optation of the white labour movement, organized labour had little appetite for the ideals of its global counterparts, and new immigrants to Southern Africa seemed to acquiesce to this position. As Neil Roos and Bill Freund respectively have recently argued, the agency of ordinary whites during apartheid is therefore best understood in terms of ideas of popular complicity, collusion, or indocility, rather than resistance or transgression.  While transnational white migration therefore resumed in mid-century Southern Africa, the flow of people was no longer accompanied by a substantial transfer of ideas. This reveals the distinctiveness of the region in the second half of the century. The early twentieth century was characterized by labour militancy and conflict with the state in a struggle over the racial definition of imperial citizenship. In this context, the imperial working class was, in Hyslop's formulation, seeking to make itself white; or, as Krikler put it, to be recognized as citizens of the white body politic of equal standing to other classes of whites. In the mid-century racial state, by contrast, white working-class citizenship was much more secure. The interests of white labour had been bound up with those of political and business elites in a manner paralleling the Keynesian social contract politics of post-war Western stateswith the difference that, in Southern Africa, the exchange of political support and industrial acquiescence for economic growth, full employment, and social benefits occurred in racial terms, predicated on the exploitation and suppression of the black majority. In this way, for instance, the s and s saw South Africa's all-white Mineworkers' Union negotiate agreements with the Chamber of Mines, which allowed Africans to take on aspects of traditionally whites-only work at much lower pay in exchange for hugely improved working conditions and financial benefits for white workers.  The importance of this arrangement for white working-class citizenship and labour ideology was most clearly and dramatically revealed in South Africa from the s, when economic and political developments started to destabilize this race-based social compact. In an effort to reverse the economic downturn and quell black industrial action, the apartheid state moved towards labour reform. This included opening skilled trades to Africans and for the first time granting them access to forums of industrial conciliation. As Van Zyl-Hermann has shown, white workers baulked at these reforms, which threatened the very race-based privileges they had won in the early twentieth century. They expressed their resistance in terms of racial citizenshipcountering efforts to depoliticize labour reforms by appealing to the overall politics of apartheid that excluded blacks from citizenship rights; warning that bestowing industrial citizenship on Africans marked the beginning of the end of South Africa as a "white man's country"; and expressing fears that labour reform would render them "guest workers" in their own country. Such appeals reveal how labour reformsoccurring more than a decade before the formal end of apartheidmarked the withdrawal of state support for working-class whiteness.  White workers responded in kind, withdrawing their support of the now reformist-minded NP, instead throwing their weight behind the apartheid-restorationist opposition Conservative Party (CP). In the early s, one white engineer relayed an anecdote about when he asked a colleague "why he had joined the whites-only Iron and Steel Union, which is sympathetic to the CP, he replied in a thick Lancashire accent: 'Because nobody else looks after the white worker'".  British Empirethe context from which Hyslop's identification of an imperial working class and the ideology of white labourism was born. Yet, even as white labourism dissipated, racialized class identities endured and strengthened. In the demographic, economic and political context of twentieth-century Southern Africa, racial segregation in the workplace and in wider society had a strong and persistent appeal to white workers across the region.
By revisiting the debate on white labourism and enquiring into its afterlife beyond World War I, we have highlighted the shifting and increasingly distinctive and localized nature of white working-class ideology in the twentieth century. Placing white labourism in this longer chronological frame not only confirms that white working-class ideology was less homogenous than previously suggested, but, crucially, reveals the importance of local currents alongside transnational influences and that, in the post-World War II period, white working-class ideology became ever more locally rooted. Even as transnational white migration resumed, international influences and ideas were no longer able to gain ground in a similar manner to early in the century. While transnational and global history scholarship often tends to conflate flows of people and ideas, we have shown that post-war social democratic ideas and the strengthening anti-racist and anti-colonial convictions of the labour movement elsewhere were not transmitted to Southern Africa by the old imperial routes, even as migrants continued to be.
The arguments presented here seek to expand and reinvigorate, rather than conclude, the debate on white labourism, and we hope to see other scholars carry this forward. Indeed, there are further aspects of white working-class ideology in this period that have not received sufficient consideration and may further complement and qualify the arguments presented here. We have already mentioned that the nature of relations between British immigrant workers and newly proletarianized Afrikaners needs further examination. Beyond this, a crucial lacuna exists around the role of women workers in shaping white working-class ideology.  It is clear that existing literature largely conflates white workers with men and male experiences, the only significant exception being important scholarship on South Africa's Garment Workers' Union mentioned earlier.  While the GWU at its foundation in  represented mainly male Jewish tailors from eastern Europe, by the late s it had become an organization representing young female factory workers in the garment industry. The debate on white labourism and its afterlife would benefit greatly from more studies on female workers. Closely connected to this is a further limitation to the existing scholarship, reinforced by its androcentric focus. This relates to concentration on the mining industry, and hence the frequent use of white miners as proxies for the wider white working class. This tendency can be attributed to the dominance of mining in South Africa's early industrial development, which coincided with the period of imperial labour mobility with which the debate is currently so closely concerned. As we expanded the chronological focus in this article, we referred to important works which consider industries beyond mining but there is more research to be done here, examining the implications of distinctive labour processes on white working-class politics and ideology. Again, a concerted focus on female workers across different industries and forms of work will go a long way towards augmenting existing arguments.
Finally, another important area for future inquiry concerns the relationships between white and black workers, and specifically the attitudes of African workers and their organizations towards their white counterparts. The debate between Kenefick and Hyslop largely turned on white trade labourists' and anti-racist syndicalists' attitudes and actions towards African workers. Missing from this debate has been the attitudes of African workers and their trade unions towards white workers. John Higginson provides some evidence of this, quoting the call of the Zulu newspaper Ipepa lo Hlanga in  that African industrial workers should act independently of whites, "which will mean competition against if not opposition to white interests".  In , a mass meeting of the Industrial and Commercial Workers Union of Africa condemned the Rand Revolt and the union's leader, Clements Kadalie, regularly criticized the white labour movement in speeches and articles.  However, understandings of the relationships between white and black workers, mutual attitudes and interactions remain shaped by scholarship focused on race-based labour legislation, white workers' incorporation into a similarly race-based state, and evidence of white-on-black violence in the workplace. How white working-class ideology and attitudeswhether the white labourism of the pre-World War I era, strands of radicalism and nonracialism in the interwar period, or the concomitant and subsequent white support for racial segregationwere viewed, understood, debated, and negotiated by African workers remains a fruitful area for future investigation and debate.