Over the course of the long 2010s, Endnotes emerged as one of the Anglosphere left’s leading journals, publishing provocative and influential analyses of revolutionary struggle. With the republication of the first four editions of the journal now occurring, its arguments are receiving renewed and welcome interest. As the below will argue, the narrative of the historical workers’ movement presented by the Endnotes collective throughout its work and most explicitly in their essay “A History of Separation” hinges on two assumptions: 1) the conflation of workers’ independent self-organization with the self-management and expansion of capitalist production and 2) the reduction of the workers’ movement to the propagation of a hegemonic collective national identity as a means to self-management. The balance sheet drawn by Endnotes presents fertile ground for an interrogation of the failure of the historical workers’ movement, provided we problematize these two claims and their narrow economism and questionable periodisations.

The Anti-Workerist Critique

The Endnotes journal emerged from a debate between the English publication Aufheben and the French collective Theorie Communiste (TC) in the 2000s.¹ Endnotes’ first volume, published in 2008, was the publication of an exchange between Gilles Dauvé and TC regarding the history of the workers’ movement and its failure.² The remaining four volumes, the last published in 2019, attempt to grapple with the political and economic developments following the 2008 crisis, bridging the gap between the Occupy movement and the Arab Spring and the first stirrings of Black Lives Matter. For many in the Anglosphere left, Endnotes would not only provide a much-needed economic analysis of the crisis absent from much of the then-dominant anarchism but also provide a potential political alternative to the exhausted anti-globalisation left of the late 90s and 2000s that had once again revealed its many limitations in the Occupy movement.

“A History of Separation”, included in Volume IV (2015), would see Endnotes ultimately side with TC and, with few reservations, adopt their economistic reduction of the historical workers’ movement in their theory of “programmatism,” in which they understood the historical workers’ movement as seeking to intensify and self-manage capitalist production rather than abolish it. In an interview with the publication Chuang, a member of Endnotes summarises their line of inquiry: “How did Marxism become a theory of modernization? How did the theory of revolution against capitalism become a theory of how to carry out industrialization, and how the workers movement could become not just a junior partner, but the actual organizer of this process.”³

As a result, Endnotes‘ account of the workers’ movement can be classified as “anti-workerist,” wherein “workerism” means the affirmation of the working class and their status as workers towards a variety of ends, often reformist in nature. 

There are several strands of anti-workerism coming to similar conclusions in the context of the late twentieth-century European left: the Italian “autonomia,” or autonomists, who advocated the refusal of work⁴; the French ultra-left, eventually leading to the communisation milieu around Endnotes⁵, including Dauvé and TC, who were inspired by Jacques Camatte’s interpretation of Bordiga, in which the horizon of struggle was not between workers and capitalists but the struggle of workers against being reproduced as such; and the German value critique school around Robert Kurz and their “Manifesto Against Work,” who view class and class struggle as surface-level fetishes that can only reproduce the value form.⁶ We will treat Endnotes’ critique as paradigmatic of the anti-workerists, as theirs is the most developed of their milieu and incorporates insights from their fellow travellers, especially TC. 

The focus of Endnotes’ essay “A History of Separation” is not the workers’ movement per se; there’s neither a single concrete case study nor much detail about historical workers’ collective action, but rather the genesis of the so-called “fully separated society” that arrives with “advanced” capitalist development, the end of pre-capitalist social relations (peasants and the persistence of the old regime), and the technical alienation and deindustrialization of the post-war era that dissolves the national collective identity of the workers. As they argue:

“Proletarians must come to see that capital is not merely an external enemy. Alongside the state, it is our only mode of coordination. We relate to one another through capital; it is our unity-in-separation.”⁷ 

In taking this position, Endnotes overstate the reality of atomization, as atomization does not lead to a total absence of organisation. Electoral parties, the massive NGO complex, and even unions in a much-reduced state are still real mediations for the atomised population of the collective worker and thus organise their collective action. By insisting on a simple movement from order to disorder, they miss the nonlinear dialectic of decomposition and recomposition and the effect this has on the shifting aims and ends (or lack thereof) of the class. 

Endnotes further argues that national developmentalism became the workers’ movements’ dominant focus, either to perpetuate the workforce and perhaps reach a political majority or even just to continue to survive. This was motivated by two historical social factors. “The first ‘formal’ aspect had to do with the persistence of the peasantry – extended here to include the persistence of old regime elites […] The second, ‘real’ aspect was the [necessity of the] ‘development of the productive forces,’ that is […] the accompanying transformations, both of the productive apparatus and of the infrastructure of the capitalist society,” which gave rise to two tasks: “on the one hand, to fight against the old regime elites […]  and on the other hand, to set loose the development of the productive forces […] In each case our focus will be on the divergence between the expected and the actual consequences of capitalist development.”⁸

The conception of the simultaneous formal and real necessity quoted above is Endnotes’ attempt to avoid TC‘s erroneous usage of Marx’s categories of formal and real subsumption as historical periods, whilst allowing Endnotes to maintain their economic determinism and the theory of programmatic self-management.⁹ Formal and real subsumption for Marx describes how capital transforms the production process to produce absolute surplus value by the lengthening of the workday and relative surplus value by the intensification of exploitation through a revolution in the means of production, respectively, and occurs continually throughout capitalist development rather than in discrete periods.

In reducing the workers’ movement to the necessities of formal and real subsumption imposed by capitalist development and attempts to self-manage this development, this narrative can rightly be described as economic determinism. As a result of this, for Endnotes, the workers’ movement, whatever its consciously stated aims, political or otherwise, was in reality a means to the end of economic development. Endnotes claims, paraphrasing Jacques Camatte: “The self-liberation of the working class meant only the development of the productive forces, since the principle productive force was the working class itself.”¹⁰ By reducing the working class to productive forces, Camatte conflates living labour with its objectification in work as labour power.

Endnotes argues that economic development as a means to the generalisation of the working class never produced the expected collective identity. “The workers’ movement promoted the development of the productive forces as a means of pressing the collective worker into being, as a compact mass. As it turned out, the extension and intensification of the factory system failed to have the desired effect; the collective worker really existed only in and through the activity of the workers’ movement itself.”¹¹ In a cruel twist of irony, at least according to this account, the very economic development pursued by the workers’ movement to create their hegemonic identity was in practice the very force that brought about its undoing in the advent of the “fully separated society.” The workers’ movement created the collective worker, not vice versa. Endnotes seem to somehow argue both that this shared identity was never achieved in the first place and yet that its loss was substantial enough to bring about the defeat of the workers’ movement. 

However, here on the question of workers’ collective identity, Endnotes implicitly admit an exception to their determinism insofar as collective identity is recognised as a political construction of the unions and parties rather than an economic determination. This requires Endnotes to hold that the workers’ movement itself is entirely reducible to economic determination, but that the identity of the collective worker “pressed into being” by the workers’ movement is somehow exempt from an otherwise totalizing subsumption to accumulation. The exact “non-utilitarian”¹² logic of collective identity (and the nature of its relation to social production) as necessary compensation for conflicting interests and divisions within a collective, whether the collective worker or nation, race, etc., is never entirely made clear and speaks to deeper unresolved tensions in the attempt to articulate a purely negative politics of refusal.

The Limits of Determinism

 This determinism risks assuming that the workers’ movement is solely reducible to the dynamics of cycles of accumulation. However, Michael Lebowitz notes that the systematic abstraction assembled by Marx across the three volumes of Capital, in being an “abstract ideal average” of capitalist production, abstracts away the social reproductive labor and domestic labor of the working-class household, or what Lebowitz calls “the consideration of the production of the wage-laborer.”¹³ 

This production of the wage labourer, the political economy of wage labour, can tendentially reduce the ratio of surplus value to wages and act as the negation of the laws of motion of the political economy of capital (to increase the ratio of surplus value to wages) and constitute a counterforce irreducible to valorisation. Lebowitz highlights that Capital is only one unfinished member of the series of six books Marx intended to write, which included a volume on the political economy of wage labour.¹⁴ 

While Lebowitz acknowledges the overwhelming consensus (with the sole exception of Maximillien Rubel) that Marx absorbed the concerns of the other planned volumes into Capital, Lebowitz maintains that, regardless, Capital necessarily remains methodologically incomplete, as capitalism is not a self-sufficient system and does not generate all of its own preconditions, including nature and, most importantly, living labor. Lebowitz argues that Marx’s assumption that labor maintains a constant standard of necessity and need is adopted solely for the sake of analysis in which “actual conditions correspond to their concept.” For Lebowitz, this establishes not only that Marx rejected a deterministic or mechanistic conception of the working class but also that this theory of the production of the working class and the production of needs is not to be found in the pages of Capital or in the theory of accumulation and valorisation contained within. In addition to its importance for the realization of surplus, the production of needs has political consequences as political (and economic) struggles against capital as mediator produce new needs and collective capacities through self-organization.

The narrow economistic conception of the workers’ movement favored by TC reduces all the political dimensions of the workers’ movement, the struggle for suffrage (everywhere except for the US), the struggle for the working day, the creation of parallel civil societies, syndicalism, demands for autonomy, etc. Endnotes acknowledge factors like the struggle for suffrage, but ultimately reduce this drive to the “formal” resistance to the old regime as a means to perpetuate the working class. For the communiser milieu, working-class autonomy can only be a means for workers to affirm their status as workers¹⁵; this is to say, they conflate the self-organisation of the working class with a crypto-syndicalist self-management of capitalist production. Endnotes and TC argue that this propagation of the class fails to reject wage labour, and instead, the workers sought to manage production themselves.

This view originates in Amadeo Bordiga’s critique of councilism and syndicalism: that workers’ self-management of the workplace does nothing to abolish the value form and commodity production. However, via Jacques Camatte, this theory is generalised to the entirety of the left and the workers’ movement.¹⁶ However, Bordiga’s critique holds only for the segments of the left aiming at worker self-management (most likely Gramsci), and this was never true of all forms of working-class struggle. 

In one of Dauvé’s responses to TC reprinted in Volume I of Endnotes, he argues, “a revolutionary period weakens (rather than strengthens) the ideology of emancipating labour through labour. Then the ebb of the radical wave brings about self-managerial practices that leave bourgeois power intact, and which this power sooner or later will sweep away.” Actual attempts of the workers to take over production “were often enough a makeshift solution, an effort to fill a gap caused by the absence or incompetence of the boss. In that case, occupying the premises and restarting the work process did not mean an affirmation of the workers as workers. It was a means of survival, as in other circumstances, the buy-out of a bankrupt company by its personnel.”¹⁷ During the Italian factory occupations of the 1920s, workers’ councils frequently ran up against workers’ desire to avoid work. “In a week, between August 21 and 28, 1920, the 15,000 workers of Fiat-Centre decreased production by 60%.”¹⁸ In addition to Italy in the 20s, Dauvé cites France 1936; Spain 1936 and 1945; Italy again in 1945; France 1968; and Portugal 1974, amongst many others.

By avoiding actual working-class organisations and institutions (unions, parties, internationals, soviets/councils, etc.), narrow determinism can tar the entire workers’ movement from start to finish with the same brush (attempting to universalise the proletariat through national development). This conception often conflates state development in the periphery under actually existing socialism with the state capture of social democracy, unions, and post-war Communist parties in ruling coalitions with bourgeois interests in the core. This attempt at a catch-all conception of the workers’ movement and actually existing socialist states is prevalent amongst left-communists and the ultra-left, including Loren Goldner¹⁹ and Robert Kurz. Whether or not their conception includes actually existing socialism, the anti-workerist assumes not only that there was in fact a hegemonic collective identity, in theory and practice, but also that the workers’ movement is reducible to this national collective identity and its continued propagation alone.

“Programmatism” and/or the “Fully Separated Society” in Context

Locating the Break

Endnotes claim that “[t]he time of the workers’ movement was simply the time of the rise and decline of the semi-skilled male worker and of the industries where he worked”²⁰ has a rational kernel. The prevalence, although never exhaustive, of the male semi-skilled factory worker, largely employed in manufacturing, in a period of class composition, in this case from 1905 to 1977,²¹ is real; however, isolating this period from the previous century of the workers’ movement, as well as the next fifty-plus years, is misleading. To reduce the workers’ movement to the semi-skilled worker is to obscure how this composition of the collective labourer was formed and its place in the succession of later compositions of the class. The determinism of Endnotes prevents their questioning of why the organisations and associations of the workers’ movement, many of which were illegal for varying degrees of their lifespans, were finally able to be captured and integrated in the first place.

For Endnotes, following TC, the only shift in trajectory comes in the 70s, which sees the dissolution of programmatism, the supposed attempt of workers to self-manage capitalist development, and its imaginary collective identity, in theory and practice for the semi-skilled worker. “The claim is not that programmatism is an illusion which we can now see through but that while it was once historically legitimate, it is no longer viable. […] Based on this diagnosis of the historical obsolescence of programmatism, TC proposes ‘communisation’ not as the hitherto occluded truth or essence of communism but as its only current, historically viable form.”²² However, this is to obscure the true break occurring in the 30s, which laid the ground for the supposed hegemonic semi-skilled worker. As Goldner writes:

It is nevertheless indispensable to divide the interwar years into two periods, prior to and following 1930. The early Communist parties came into existence in the revolutionary ferment of 1917-1919, and everywhere represented mass breakaways from the Social Democracies that had capitulated to nationalism and participation in the war effort. Like the previous Internationals, the Third International was born of a worldwide working class upsurge that forced innovative structures and strategies onto the scene. The First International was born of the post-1864 general upsurge that peaked in the Paris Commune; the Second International in the strike wave of the early 1890s; the Third International out of the annus mirabilis 1919. But the ebb after 1921, the ‘united front’ fusion with the left-wing of Social Democracy after 1921, the “Zinovievization” of Western parties after 1924, and finally the ‘ultra-leftist’ zig-zags of the Third Period left the Communist Parties in most countries as large sects. There was a virtual discontinuity between the revolutionary wave that had created them in 1917-1921 and the organizational shells that grew into mass parties after 1935, with completely different, accommodationist politics.²³

In addition to the breaking of the wave after 1921, the beginning of the twentieth century would also see the apex of syndicalism and the strata of skilled workers resisting proletarianization, who composed its organic subject. 

Guilds, Syndicalism, and Skilled Labour 

Syndicalism was fueled by skilled workers and artisans attempting to recreate the functions of guilds and collective forms of ownership of the means of production and self-management of labour. The guild system and its parallels, like the “Gang Boss”²⁴ system prevalent in prerevolutionary China but present in many different guises through history, assume a division of the spheres of social reproduction foreign to capitalist social reproduction. Alongside the division of the state from civil society, the differentiation of the workplace, the household, and the public sphere is a product of the disintegration of precapitalist social forms by the logic of accumulation. 

The differentiation of these spheres occurs in the transition attending formal and real subsumption and is absent in early stages of capitalist development, such as the household production of the mercantile capitalist putting-out system of patriarchal familial labor. The division of the workplace and social organizations and associations was absent in guilds, corporations, and other pre-capitalist associations of labor, such as those of pre-revolutionary France. 

In the prerevolutionary corporations, labor was nothing if not social. […] both masters and workers were subjected to the collective discipline of the corporation, which regulated everyone and everything in the trade […]. All members of the trade were therefore expected to venerate the same Saint and celebrate his or her festival with a common mass, a procession, and a banquet. The confraternity also organized distribution of mutual aid. Every member paid common dues to a common treasury […] brotherhoods [called compagnons] maintained roominghouses for itinerant journey men making their ‘tour de france’ and had an astonishingly elaborate ritual life.²⁵

The same process of decomposition and rationalisation that abolished the medieval corporations took place in France and England, as it takes place everywhere capital takes root; albeit for the latter, it took place at a rate and intensity far greater than that of the former.²⁶ Handweaving had given way to mechanical textile labour in England as early as 1830-1840, whereas in France, the rationalisation of textile production was not complete till the mid-1890s, and as a result, a stratum of artisan workers persisted till the early twentieth century.²⁷ Much of the activity from the 1830s to 1848 was led by urban artisan workers in Paris and Lyon, as the initial textile factory workers were still caught within familial household production in contrast to the urbanised artisans.

The peak of syndicalism from 1900 to 1914 in France²⁸ (at around 350,000 in 1910-1911, around three percent of the population²⁹) and Italy and, to a lesser extent, under the IWW in America (albeit with a different strata and composition of the working class centered around migrant workers) would also be its last gasp, alongside the skilled and artisan laborers and their collective memory of precapitalist working conditions and relations it relied on. The persistence of pre-industrial labourers and forms of exploitation was even more pronounced in Italy and Spain, where they would remain well into the postwar era. The uneven regional and national class compositions prior to the break occurring in the 1930s contained some balance of skilled and unskilled, artisan, agricultural, industrial, and domestic workers and labourers, split between syndicalism, reformist trade unionism, bourgeois electoralism, various stripes of socialism, and even Marxist groupings. 

National and International

The period of “programmatism” and the era generating the fully separated society for Endnotes fall under Marcel Van Der Linden’s period of organisational national internationalism, 1894-1917 and 1940-68.³⁰ For Marcel Van Der Linden, there are five periods of international organizational composition: 1) an initial period of self-discovery (pre-1848) for the emergent working class before national and international organization; 2) a pre-national, or sub-national, internationalism (1850-1870), including the foundation of the First International; 3) a transitional phase (1870-1890) seeing the breakdown of pre-national internationalism (largely resulting from the Franco-Prussian War); 4) national internationalism, before 1918, led by Germany with a minority representing textiles and miners in England (1894-1917) and, after 1918, transferring to England and the US (1940-1960); and 5) a second transitional, potentially transnational, phase (1960-present) seeing the breakdown of national-led internationalism, expanding into Asia, Africa, and Latin America.³¹

This national organisational infrastructure inherited by the cartelized semi-skilled workers highlighted by Endnotes was forged by the struggles of the syndicalists and trade unionists of the previous generation, albeit void of the composition of conflicting forces that birthed it. The illusory collective identity of semiskilled labour presupposed the defeat of the pre-national organisational and institutional forms active before 1894. The first “internationalist links began to form as a result of economic booms and crisis evident in the 1840s and 1850s and the utilization of foreign strike-breaking workers during periods of labour unrest, especially by British employers […] the first structures of internationalism among labour can be traced to 1844 and the British Owenites and Chartists, together with refugees from Germany, France, and Poland, of the London-based ‘Democratic Friends of All Nations’.”³² Andrew Herod goes as far as to argue that the transnational associations of the workers’ movement predated those of the capitalist firms.

While TC dates the end of workers’ attempts at self-management (programmatism) in the 70s, arguably, this occurs in the first decades of the twentieth century with the aforementioned decomposition of syndicalism and its strata of workers; contra Endnotes, self-management and cartelized state development were mutually exclusive tendencies; the former had to give way to the latter. The collapse of workers’ independent forms of self-organisation following the apex of syndicalism and the advent of the councils in Russia and Germany is instrumental in the state cartelization of organised labour in the twentieth century. We highlight the importance of syndicalism and councils, as well as pre-national organisations like the First International, not to fetishise a particular form over its content but to illuminate their shared function as independent organisations and associations parallel to the state and the capitalist firm and the necessity of their defeat for the ensuing national consolidation.

The irony of TC’s conception is that it misrepresents the post-1930s formal workers’ organisations as being more radical than they really were. The era of the popular front and the social democratic/Fordist social compact saw the abolition of the classical bourgeois public sphere’s already porous division between the political and the economic in favour of a technocratic maintenance fusing the two spheres in the balancing of economic demands and the recognition of bureaucratic interest groups.³³ Rather than the syndicalist reunification of the economic and the political under the auspices of the workers’ movement, the two spheres reunited under the gaze of state monopoly capitalism, with the workers as a silent partner. The majority of the managerial trade unions and bureaucratic mediations of this era never even considered the possibility of worker self-management in theory or practice. While this conception would briefly resurface in Italy and France in the 60s (the abortive so-called neocapitalism of André Gorz and Alquati), it was strangled in its infancy by the neoliberal turn and the ensuing collapse of postwar organised labour in the coming decades.

Periodizing the Workers’ Movement

The Long Durée

As Marx once wrote, “epochs in the history of society are no more separated from each other by strict and abstract lines of demarcation than are geological epochs.”³⁴ Periodisations should be treated as tracing out fuzzy gradients and periods of transition and exchange from one form or structure to another. 

Any periodisation of the workers’ movement must be superimposed upon the long durée process of the majority of the world’s population moving from rural to urban regions and undergoing proletarianization and separation from the means of production. This is a merit of Endnotes‘ work, yet reducing the workers’ movement to only this is perhaps to bend the stick too far. An uncharitable reading of Endnotes would summarise their conception of the workers’ movement as a crypto-syndicalist attempt at the self-management of capitalist production because the workers’ movement up until the 1970s was itself nothing more than an unconscious agent of formal and real subsumption. As we have shown above, attempts at self-management were never true of the totality of the workers’ movement and were definitively defeated by WWI at the latest, and the larger collapse of the workers’ capacity for independent self-organisation had to occur before they could become cartelized. 

The early workers’ movement before WWI was spearheaded by peasants, skilled guild laborers and artisans responding to and resisting proletarianization. Skilled workers and artisans struggle against the abolition of guilds or attempt to recreate the collective self-management of the guild systems in syndicalism. From the 1830s to the 1870s, German journeymen, artisans in brotherhoods like the masons working for small employers, significantly contributed to the development of unions and trans-trade inter-craft associations due to the common causes of different trades and crafts attempting to respond to the same structural factors.³⁵ Unlike the newly urbanised peasants, not only do these strata identify with their labour, but they realistically can manage the workplace themselves; this relation to labour and penchant for self-management influence the genesis of councils in the Russian and German revolutions. For their quasi-proletarianised counterparts, a familiar pattern emerges that will last for much of the twentieth and even twenty-first century in China and Africa, in which the working population has yet to permanently migrate from the rural periphery to the urban core and split their time and work between agrarian and industrial wage labour and only gradually become full-time wage labourers with the supplanting of agrarian production by industrial development.

Beginning with the war economy formed by WWI and WWII, the increasing rate of proletarianization and migration from agrarian regions to urban industrial ones is largely absorbed by productive labour in centralised heavy industry. In America alone, between 1940 and 1945, 4.5 million people became permanent urban residents.³⁶ However, the tendency to absorb the increasing mass of the collective labourer at work from the world wars till the late 70s and early 80s gradually dissipates. This creates a growing informal stratum of the collective labourer engaged in nonproductive labour,³⁷ that is, wage labour facilitating the circulation of commodities (sometimes quasi-self-employed) or engaged in socially reproductive labour. This nonproductive labour acts as constant capital rather than as variable capital, contributing surplus to the ratio of surplus to necessary labour time (such as services) or is left aside as a surplus population. Following capital’s tendency to replace variable with constant capital, fewer and fewer workers are needed for production, and the structural position of the worker is far less stable. 

In addition to the growing ratio of constant to variable capital, the continuous relocation of capitalist firms to obtain cheaper labour during globalisation further undermines the ability of workers to receive concessions, as even if collective action is successful, the operation can still relocate to avoid facilitating workers’ demands and higher wages. “Under the period of embedded liberalism (end of World War II to the 1970s), a higher degree of regulatory spatial inflexibility contributed to greater bargaining power for workers. As that regime ended and capital became more spatially flexible, workers’ bargaining fell with it.”³⁸

We will argue that there are seven medium-term periods of the workers’ movement that can be further divided and superimposed on cycles of accumulation and secular cycles, as well as regional and national waves of class composition. 1) The period preceding 1848 (self-discovery); 2) 1848 to 1914, seeing the rise and fall of skilled and artisan workers resisting separation and the growing trade union movement; 3) 1914 to 1930, the splitting of the Second International over WWI alongside the Russian and German council/Soviet revolutions and the beginning of the downturn of syndicalism; 4) 1930 to 1950, the popular front and the national integration of the unions and parties; 5) 1950 to 1980, seeing the apex and collapse of the mass factory worker; 6) 1980 to 2008, seeing the rise of a majority global urban population absorbed in decentralized unproductive labor; and 7) 2008 to the present, seeing urban development without industrialization (requiring less labor with each cycle of development) occurring in regional hubs of global supply chains, creating nodes of proletarianization around contractors and suppliers producing for buyers with monopsony power.

The third wave acts as a hinge, facilitating the transition between the pre-national and national periods of organisation. The advent and defeat of the pre-national independent self-organisations occur in the first, second, and third waves; the national cartelization and bureaucratic capture of the workers’ movement begins in the third wave and lasts through the fourth and fifth waves until the disintegration and globalisation of the national infrastructure of the core in the sixth and seventh waves.

With the decline of syndicalism and the remnants of guild labour in continental Europe and North America following WWI, the trade unions and social democracy, ascendant after the 1870s, became increasingly domesticated and integrated into the state and the bureaucracy of capitalist firms. The centralisation and concentration of semi-skilled factory labourers, dominant from the 1920s and ’30s to the late ’70s and ’80s, increasingly launched wildcat strikes aimed not only at their employers but also at the union bureaucracy itself. From the 70s and 80s onward, the preceding process of centralisation and concentration of the collective labourer turns over into its opposite and disintegrates and distributes the organic composition of capital and the population of productive variable capital throughout the global world system at load-bearing nodes in supply chains.

The postwar decline of centralised factory labour creates a working class with a relation to work markedly different from those of skilled workers and peasants resisting proletarianization, as well as the only partially subsumed part-time workers splitting their time between rural and urban social life and the mass semi-skilled labourers of the mid-twentieth century. 

Globalisation, GVCs, and Regional Development After 2007

Our sixth and seventh waves see the dissolution of the cartelized labour relations of the mid-twentieth century. The globalisation of the world system and the concomitant increase in the organic composition of capital increase the ratio of constant to variable capital in response to the crises of the late 70s, leading directly to the crisis of 2007-08 and the ensuing decline in export production for consumption in the core. The end of cartelized labour and its norms regarding wages and effective demand leads to a steep increase in consumer debt to continue facilitating mass consumption alongside stagnant wages. This trend comes to an end with the ’08 financial crisis and prompts the transformations in global value chains (GVCs) constructed in the post-war era. These value chains can be polarised in a variety of different ways, whether monopolistic and favouring the supplier or producer, monopsonistic, favouring the buyer, or even symbiotic. Following Ashok Kumar, the shifts in monopsony relations following 2008 are the most reflective of trends in the composition of the global working class:

“Between 1995 and 2013, the number of workers at the jobs related to global value chains increased from 296 million to 453 million, with much of this increase occurring before the 2008 financial crisis. […] Most occurred in Global South countries, directed for Global North consumption.”³⁹ 

This surge of employment occurred within supplier firms, largely in Asia and Latin America, contracted by monopsony buyers representing established brands and/or the transnational asset managers operating them. 

Following the decline in consumption in the core and the export production facilitating it, we have seen, as Phil Neel argues, not a reversal of globalisation but a regionalisation of global trade. In the shifting dynamics of monopsony production, we see a shift from the national developmental industrial programs and their cartelized labour relations that characterised much of the twentieth century to dispersed regional urban development, albeit without the mass industrialisation of previous eras, at nodes along global supply chains. 

These urban centres act as the administrators of accumulation carried out by their neighbouring regional productive infrastructure. “Though it may seem at first as if these metropolitan cores are themselves the center of the global economy, they are more accurately the tip of an economic iceberg. […] The true center of the world economy is not to be found in the “creative,” financialized, or high-tech downtown cores of its global cities, but instead in the complex mesh of material infrastructure that links them together.”⁴⁰ The differing needs of the financialised urban centres and their productive infrastructure create a division of labour split between workers absorbed by suppliers contracted by international monopsony buyers, and the social reproductive and service work needed to reproduce urban administrative centres.

Kumar has detailed how competition and consolidation between competing capitals have led to an evening out of the monopsony polarisation towards a symbiotic relation between suppliers and buyers that has led to greater bargaining power for workers. The concentration of capital around oligopolized GVC producers has limited the efficacy of the spatial fix in undermining working-class collective action. 

The end of national cartelized labour relations has led to a new spatial distribution of workers’ collective action with a decidedly different relation to the state and the nation. The 2014 Yue Yuen strike emerged as a private strike over the state enforcement of pensions that quickly expanded beyond its national borders, spreading sympathetic actions and protests to Istanbul, Melbourne, Bangalore, London, Taiwan, Hong Kong, and New York.

The spread of the strike occurred without central coordination but organically as a consequence of the identification as a Yue Yuen worker outside the boundaries of a single factory, city, province, and […] beyond the border of the Chinese state itself. Francine Chan described the strike’s propagation, noting that, though ‘some workers passed information through the social networking site Weibo and the instant messaging service Qq, [most of] the workers are older [and therefore] most of the news spread through word-of-mouth, [since] strike organisers had already developed contacts with worker leaders at other Yue Yuen factories during smaller strikes in the past’.⁴¹

Globalisation has intertwined the national and the transnational in such a way that growing monopsony suppliers like India’s Aravind have taken on increased importance for domestic production and consumption, some of which is even outsourced by international buyers. The absence of any clean distinction between national and transnational leads to a reintegration of the state along the lines drawn by GVCs and the necessity of workers’ collective action and organisation to follow the same trend.

Conclusion: Self-Organisation and Independence

To summarise, Endnotes‘ greatest faults lie in 1) isolating the semi-skilled worker of the middle twentieth century and generalizing a particular European variant of this archetype to the workers’ movement as a whole and 2) a) adopting an economic determinism reducing all social production to valorization and b) inconsistently alternating between political and economic determination in relation to the collective worker and the workers’ movement, respectively. 

One of the merits of “A History of Separation” is the attempt at an overview of the workers’ movement that is shockingly rare amongst commentaries on the subject. Much of the existing literature is focused on organisational histories of individual parties or influential leaders or isolated periods of national movements and even individual strikes. Overcoming this existing limitation of theory entails capturing the concrete relation between valorisation, class composition, and organisational forms as processes that hang together as members of a dynamic totality. Gilles Dauvé’s The Communist Left in Germany 1918-1921, while a valuable resource on the ecosystem of organisational forms active at the time, intentionally abstracts away the production of the working class and thus avoids confronting the relation between class composition and political subjectivity. As we have seen, the determinism of Endnotes and TC cleaves in the other direction while still abstracting away the concrete production of the worker in favour of questionable schemas of accumulation. 

The balance sheet drawn by the debate between Gilles Dauvé and TC and its extrapolation by Endnotes ought to be the beginning of a collective inquiry and dialogue around the workers’ movement and its consequences for us today. As such, we have attempted an alternative periodisation above to better understand this process. Unearthing the historical defeat of working-class self-organisation and independence is a theoretical necessity that, of course, doesn’t orient us forward in practice in and of itself, but clears the way for possibilities that could. In an interview from the early 80s, Immanuel Wallerstein considers the potential for returning to the unfinished task and project begun by Marx:

Today, Marxist ideas (or Marxist terminology — put it that way) are continuing to spread in the United States and — I believe — around the world. I still think that the progress of Marxist terminology is comparable to the progress of Christian terminology in the second and third centuries; we are moving toward a world in which everyone will use Marxist terminology. It’ll take another fifty or one hundred years, but we’ll get there. […] I think that we are approaching the moment of a definitive political rupture in the single Marxism (as reflected in a single world Marxist movement) and the birth of a thousand Marxisms. That means we can truly begin to work in the same way that Marx did, in a critical, meditated, political way. Everything that for too long hasn’t been done.⁴²

---

1.  See Aufheben’s “The Theory of Decline or the Decline of Theory,” pages 122-240. <https://cooltexts.github.io/sources/aufheben.pdf> ↩︎
2.  Endnotes, “Preliminary Materials For a Balance Sheet of the Twentieth Century,” Endnotes 1 (2008). ↩︎
3.  Chuang, ‘Neither Prophets Nor Orphans: An Interview With Endnotes,’ Chuang, n.d. <https://chuangcn.org/2025/02/neither-prophets-nor-orphans-interview/> ↩︎
4.  See “Storming Heaven: Class Composition and Struggle in Italian Autonomist Marxism” by Steve Wright. ↩︎
5.  The genealogy of Endnotes and communisation is as follows: Dutch-German Councilism > Bordiga > Camatte/Invariance > Dauve and TC > Goldner > Endnotes. ↩︎
6.  Robert Kurz and Ernst Lohoff, “Der Klassenkampf-Fetisch: Thesen zur Entmythologisierung des Marxismus,” Marxistische Kritik, no. 7 (August 1989). ↩︎
7.  Endnotes, “Spontaneity, Mediation, Rupture,” Endnotes 3 (2013), p. 248. ↩︎
8.  Endnotes, “A History of Separation,” Endnotes 4 (2015), p. 79. ↩︎
9.  Endnotes also wish to avoid TC’s reliance on Michel Aglietta of the Regulation School, whose findings are only applicable to France. By contrast, arguably, Endnotes generalizes a Franco-German continental model of development (what Ashok Kumar calls ‘the inside model’ as opposed to the ‘outside model’ in the U.S.), as the U.S. didn’t suffer from the persistence of the old regime, and while France and Germany developed later than England and the U.S., they pulled ahead of Italy and the periphery and produced a mass semi-skilled worker for much of the twentieth century. ↩︎
10.  Endnotes, “Bring Out Your Dead,” Endnotes 1 (2008), p. 12. ↩︎
11.  Endnotes, “A History of Separation,” Endnotes 4 (2015), p. 119. ↩︎
12.  Ibid. p. 122. The “non-utilitarian” logic of collective identity and collective interest seemingly oscillates between momentary insulation from economic determination and absolute dissolution in economic atomization. ↩︎
13.  Michael A. Lebowitz, Following Marx: Method, Critique, and Crisis (2009), p. 308. Wallerstein defines household reproduction as a form of economic obligation. “A typical household consists of three to ten persons who, over a long period (say thirty years or so), pool multiple sources of income in order to survive collectively. Households are not usually egalitarian structures internally nor are they unchanging structures (persons are born and die, enter or leave households, and in any case grow older and thus tend to alter their economic role). What distinguishes a household structure is some form of obligation to provide income for the group and to share in the consumption resulting from this income.” Immanuel Wallerstein, World Systems Analysis: An Introduction (2006), p. 32. ↩︎
14.  Lebowitz weighs in on the debate regarding the originally intended six books in Beyond Capital and concludes “[a]lthough there was no explicit disavowal of the six-book plan, there is equally no unequivocal evidence that Marx did not view Capital in itself as a ‘completely elaborated system’. Accordingly, after his review of the competing arguments, Allen Oakley has recently proposed that the bibliographical evidence necessary to determine either Marx’s intentions or whether there remained any books unwritten is simply insufficient to support a judgement in either direction.” Michael A. Lebowitz, Beyond Capital (2003), p. 14. Ultimately Lebowitz concludes, “The central issue is not at all whether Marx intended to write a book on wage-labour. If he had not mentioned it, we would still need to write it.” p. 50. ↩︎
15.  Endnotes, Theorie Communiste “Much Ado About Nothing,” Endnotes 1 (2008), p. 187. “Autonomy can only be programmatic, because it is by its very nature workers’ autonomy.” Also see p. 160-161. “Self-organisation, strong unions and the workers’ movement, all appeared in the same world of the revolution as affirmation of the class. […] Self-organisation entails the self-organisation of struggle, thus the self-organisation of producers. In a word – liberated labour; in another word – value.” ↩︎
16.  Ibid, “Bring Out Your Dead” p. 12.  “Thus a question that for Bordiga had been one of theoretical and organisational error came for Camatte to define the historic function of the workers movement within capitalism.” ↩︎
17.  Ibid, “Love of Labor? Love of Labor Lost…” p. 131 and 134. ↩︎
18.  Ibid p. 114. ↩︎
19.  Goldner is unique amongst this grouping, as he argues that while the historical workers’ movement may have been a means to self-perpetuation of wage labor, it does not necessarily have to take this form. For Goldner, the workers’ movement produced this outcome through the infiltration of unproductive civil servants who brought with them a “bourgeois atomist ontology” that led them to become the unconscious agents of real subsumption (here understood, as for much of the left-communist milieu, as a historical period rather than a recursive process as in Marx). “The tacit acceptance of atomist ontology by 2nd and 3rd International Marxism, its status as the ideology of the substitute bourgeois revolution, is linked to its false appreciation of the conjuncture, and linked to its statist civil servant vocation.” According to Goldner, this dynamic predates even Marx: “The line of continuity is from St-Just to Fichte to Nechaev and Tchachev to Stalin, Mao, Ho and Pol Pot.” Loren Goldner, ‘Short History of the World Working-Class Movement from Lassalle to Neo-Liberalism: The Distorting Hegemony of the Unproductive Middle Classes’, breaktheirhaughtypower, n.d. <https://breaktheirhaughtypower.org/734/> [accessed 19 June 2026]. Goldner argues the “age of the civil servant”, 1905-1975, has come to an end, and therefore so has this form of the workers’ movement. While we endorse Goldner’s view that the workers’ movement cannot be reduced to self-perpetuation, his post-Trotskyite theory of bureaucratic betrayal (largely inspired by C.L.R. James) leaves much to be desired. ↩︎
20.  Endnotes, Endnotes 4, p. 108. Although, arguably, the era of the semi-skilled mass worker began in the 20s or 30s or WWI at the earliest. ↩︎
21.  Ibid, p. 81. ↩︎
22.  Ray Brassier, ‘Politics of the Rift: On Théorie Communiste,’ e-flux, n.d. <https://www.e-flux.com/notes/550201/politics-of-the-rift-on-th-orie-communiste> [accessed 19 June 2026]. ↩︎
23.  Goldner, ‘Short History of the World Working-Class Movement from Lassalle to Neo-Liberalism: The Distorting Hegemony of the Unproductive Middle Classes.’ ↩︎
24.  Chaung 1: Dead Generations (2016), p. 29-30 and p. 36-37. “These Chinese workers were, in Manchuria as elsewhere, procured by gang-bosses or labor contractors, who would receive the total sum of wages from the Japanese employers and distribute them to the workers as they saw fit, reserving a large cut for themselves. But whereas the gang-boss system used in the southern port cities saw gang-bosses, in competition with guilds and other contractors, commanding smaller networks of labor recruits who were dispatched to much smaller enterprises, these modern Japanese factories required labor deployment on an entirely different scale. Many used only ‘a small number of batou who supplied and managed several thousand workers.’ This management was distributed down a vertically-integrated hierarchy of gang bosses, with ‘Number 3 batou’ at the bottom commanding squads of ‘about fifteen workers.’ At the same time, the gang-boss hierarchy was itself buffered by an extensive bureaucracy, with ‘other functionaries as well, such as a xiansheng or sensei (who was clerk, accountant, and paymaster), cooks, and runners.’” ↩︎
25.  Ira Karznelson and Aristide R. Zolberg, Working-Class Formations: Nineteenth-Century Patterns in Western Europe and the United States (Princeton University Press, 1986), p. 54-55. ↩︎
26.  Marx details this process in chapters thirteen, fourteen, and fifteen of Capital, Volume One. “With regard to the mode of production itself, manufacture can hardly be distinguished, in its earliest stages, from the handicraft trades of the guilds, except by the greater number of workers simultaneously employed by the same individual capital. It is merely an enlargement of the workshop of the master craftsman of the guilds.” Capital: A Critique of Political Economy, Vol. I (1976), p. 439.  ↩︎
27.  Karznelson and Zolberg, Working-Class Formations: Nineteenth-Century Patterns in Western Europe and the United States, p. 115-116. ↩︎
28.  Gilles Dauve and Denis Authier, The Communist Left in Germany 1918-1921, p. 178. Dauve argues French syndicalism failed as early as 1906. ↩︎
29.  Karznelson and Zolberg, Working-Class Formations: Nineteenth-Century Patterns in Western Europe and the United States, p. 107. ↩︎
30.  Marcel van der Linden, Workers of the World: Essays Towards a Global Labor History (2008), p. 280. ↩︎
31.  Ibid. ↩︎
32.  Ashok Kumar, Monopsony Capitalism: Power and Production in the Twilight of the Sweatshop Age (2020), p. 44. ↩︎
33.  “[T]he classical public sphere is originally rooted in the [private] bourgeois context of living, yet separates itself from the latter and the production process. By contrast, the new public spheres of production are a direct expression of the sphere of production. […] The differentiation between public and private is replaced by the contradiction between the pressure exerted by production interests and the need for legitimation.” Oskar Negt and Alexander Kluge, Public Sphere of Experience: Towards an Analysis of the Bourgeois and Proletarian Public Sphere (Minneapolis: University of Minnesota Press, 1993), p. 11-18. ↩︎
34.  Karl Marx, Capital: A Critique of Political Economy, Volume 1 (1976), p. 492. ↩︎
35.  Karznelson and Zolberg, Working-Class Formations: Nineteenth-Century Patterns in Western Europe and the United States, p. 307-316. ↩︎
36.  Mike Davis, Prisoners of the American Dream, p. 74-75. ↩︎
37.  “[T]he extraordinary increase in the productivity of large-scale industry, accompanied as it is by both a more intensive and a more extensive exploitation of labour-power in all other spheres of production, permits a larger and larger part of the working class to be employed unproductively.” Karl Marx, Capital: A Critique of Political Economy, Volume 1 (1976), p. 574. ↩︎
38.  Kumar, Monopsony Capitalism: Power and Production in the Twilight of the Sweatshop Age, p. 11. ↩︎
39.  Ibid. p. 18. ↩︎
40.  Phil Neel, Hinterland: America’s New Landscape of Class and Conflict (2018), p. 97. ↩︎
41.  Kumar, Monopsony Capitalism: Power and Production in the Twilight of the Sweatshop Age, p. 106. ↩︎
42.  Nicolette Stame and Luca Meldolesi, ‘Immanuel Wallerstein’s Thousand Marxisms,’ Jacobin, n.d. <https://jacobin.com/2019/09/immanuel-wallerstein-marxism-world-systems-theory-capitalism> [accessed 19 june 2026]. ↩︎