Writing his 1974 book Labour and Monopoly Capital: The Degradation of Work in the Twentieth Century, US Marxist and political economist Harry Braverman noted that Karl Marx had demonstrated that processes of production are constantly transformed by the driving force of capital accumulation.1 These transformations manifest themselves in the changes in the application of science and technology to production and their effect on labour processes in each branch of industry, as well as in the redistribution of labour among occupations and industries.
Marx’s analysis contrasted with the dominant industrial and sociological views of the 1960s and 1970s, which saw changes in work and occupations as largely benign. According to these, automation would remove heavy and tedious labour. Increasing demand for educational qualifications supposedly reflected the growth of skilled labour and white-collar occupations. New personnel policies seemed to reflect concern for workers’ wellbeing. More widely, academic sociology and conservative political ideologies saw the growth of white-collar workers as increasing the weight of the middle class, thereby reducing the viability of socialism. This belief was reflected by variants of socialist thought that saw revolutionary agency as moving from the working class in the most advanced industrial economies to the Global South, the peasantry and dispossessed.
For Braverman, the ascendency of this bourgeois view of work was largely a result of Marxists having added little to Marx’s work since his death, despite continuous and significant changes in the nature of the economy. There was simply “no continuing body of work in the Marxist tradition dealing with the capitalist mode of production in the manner that Marx treated it in the first volume of Capital”.2 Indeed, outside of the relatively few readers of all three volumes of Marx’s Capital, there was little knowledge of Marx’s contribution in the theorisation of workplace reorganisation and its effects on class relations. Thankfully, there has been an increase of academic interest in Marx’s ideas in this area since the Braverman’s “rediscovery of the labour process”. However, Marxist contributions to the subject have been overshadowed by theorists seeking to undermine Marx’s and Braverman’s understanding of changes in work and their effect on class relations and accumulation strategies.
Braverman challenged bourgeois characterisations of the capitalist labour process. He admitted that major changes in the organisation of production, in particular the growth of commercial, administrative and technical labour, “seemed to cut across Marx’s bipolar class structure and introduce a complicating element.” However, he also insisted that the adoption of Fordist production techniques, and Taylorism as a mode of work organisation, was spreading to more and more sections of the economy.3 Much “white-collar” labour was subject to exactly the same developments as “manual” work; to varying degrees, white-collar employees were becoming workers. Braverman contested descriptions of workforces as increasingly skilled, detailing employers’ tendencies to increase control of workers’ labour through scientific advances that separated the conception and planning of work from its execution—in short, deskilling. He also explained the continued relevance of capital’s utilisation of the reserve army of labour.
The impact of Braverman’s book was rapid and profound, reframing sociological views about work and challenging dominant academic orientations. Braverman’s influence has subsequently waned, though not due to outmoded insights, but rather because conjunctural factors have forced a retreat from class-based analyses in general.4 These factors included the collapse of the Soviet Union, which was said to foreclose any possibility of alternatives to capitalism, as well as the rise of neoliberalism in the 1980s and the popularisation of post-modernism.5 All of these developments reflected and deepened the loss of confidence in working-class organisation. However, as E P Thompson stressed, class is a relationship. Though overt mass resistance retreated (notwithstanding events such as the 1984-5 Miners’ Strike), the “other side” of the capital-labour relationship, spearheaded by the state, went on the offensive. This was, in Ralph Miliband’s terms, “class struggle from above”.6 The result was the enactment of anti-trade union legislation, which remains intact today, and assaults on the welfare state and living standards.
What follows is largely an appreciation of Braverman’s contribution. This is no substitute for reading the original. His work is both rich and deliberately accessible. This latter quality is very admirable and reflects Braverman’s experience as a worker and a socialist activist. There are criticisms to be made of his work, but these should not distract from his achievements. Following a summary of some aspects of Labour and Monopoly Capital, the second part of this article looks at the subsequent criticisms of the book, detailing how they developed into the essentially conservative project of the “labour process debate”. This field of academic argument became detached from its origins in Marx’s Capital and Braverman’s work. The labour process debate eventually coalesced into a new orthodoxy stripped of any relationship to class analysis. It is time to rectify the surrender of this vital subject to those incapable of seeing beyond capitalist production and to reassert that workplace relations are characterised by exploitation and class conflict.
Capitalism and the degradation of work
Braverman’s work emerged in the context of claims that the expansion of administrative, technical and professional jobs reflected growing skills among the workforce and a changing class composition of employment. Against these claims, he argued that whatever the surface appearances and formal qualifications of labour, the dominant tendency within capitalism was for work, and thus also the worker, to be degraded. Braverman’s argument was widely based, but its central thesis built on Marx’s theory of the labour process. Marx maintained that the “real” labour process takes place when the labourer “creates new use values by performing useful labour with existing use values”.7 The labour process described here is a labour process in general and, as such, “its elements, its conceptually specific components, are those of the labour process itself, of any labour process, irrespective of the mode of production or the stage of economic development”.8 As Marx insists, however, labour processes never occur in the abstract but always under concrete social relations in specific societies. One result of this is that not all employed activity is necessarily part of a labour process. Capitalist managers and their supervisors may be “at work”, but they are rarely performing the useful labour of adding use values through the production process and are thus are not engaged in a labour process. Instead, they are engaged in ensuring the creation of surplus value through the supervision and control of the work of others.9 Within the capitalist mode of production, the labour process is a surplus value-produing process. Indeed, in a typical commodity production process, the creation of surplus value is the “determining, dominating and overriding” factor, according to Marx.10
Both Marx and Braverman were focused on the inevitable violence that capitalist production wreaked on its workforce. Profits are only possible through utilising the unique capacities of human beings. Humans are differentiated from other animals by their ability to imagine the outcome of labour before they perform it. As Braverman summarised: “Human work is conscious and purposive, while the work of other animals is instinctual”.11 This unique characteristic allows the transfer of knowledge through culture, opening the way for material and technological progress. It also, however, makes possible the separation of conception from execution. Our unique ability for thinking before doing has enabled not only the development of humans as a species but also the subordination of labour to social relations of exploitation. The capacity to both conceive an outcome and plan its execution—a fundamental part of individual and collective development—has been fractured by the relegation of large parts of the world’s population to the fate of simply carrying out tasks determined and directed by others. Labour power has become a commodity, albeit a unique one, under capitalism.
When capital employs workers, it is hiring their capacity to produce. However, this capacity—labour power—has to be set in motion. Resistance arises in this process because the interests of capital and labour are opposed. This resistance can be conscious or unconscious, overt or covert, organised or unorganised. As Braverman stressed:
What the capitalist buys is infinite in potential, but, in its realisation, it is limited by the subjective state of the workers, by their previous history, by the general social conditions under which they work as well as the particular conditions of the enterprise and by the technical setting of their labour.12
Capitalists (or their agents) have to ensure that no more than the average socially necessary costs are incurred in the production process. This means controlling the costs and productivity of labour as well as managing the productive consumption of means of production so that “the use-value he has in mind emerges successfully at the end of this process.” The “capitalists’ ability to supervise and enforce discipline” is thus “vital”.13 Braverman redirected attention to how this supervision is carried out and its implications for both the degradation of the labour process and the transformation of the structure of social relations in the workplace. He writes:
Corresponding to the managing functions of the capitalist of the past, there is now a complex of departments. Each has taken over in greatly expanded form a single duty that a single capitalist exercised with very little assistance in the past.14
This growth of management has resulted from moves towards corporate ownership, increasingly complex business organisation and the logic of an intensifying division of labour. Particularly important have been the methods employed by the “scientific management” of Frederick Winslow Taylor. Taylorism attempts to leverage scientific inquiry to achieve the “disassociation of the labour process from the skills of workers”, the “separation of conception from execution” and a “monopoly over knowledge”. This enables capital to seize “control of each step of the labour process and its mode of execution”.15 Although there have been numerous objections to Braverman’s argument, these too often reduce his contribution to a “deskilling” thesis. The following section examines how sociology received and undermined Marx’s and Braverman’s accounts of the capitalist labour process.
The reaction to Braverman
Many on the left welcomed Braverman’s approach, which focused on a critique of the quality of work and challenged the idea that capitalism could produce an alternative to the drudgery that confronts the majority of workers.16 There was, however, no shortage of opposition to Labour and Monopoly Capital, though it was often repetitious and rarely challenged. One standard criticism was that Braverman had failed to recognise workers’ active agency. Tony Elger, a sociologist and critic of Braverman, accused him of an “objectivist conceptualisation of the working class, which fails to address how the class struggle is integral to the course of development of the capitalist labour process.” Elger added that Braverman’s book implied that “analyses of both the obstacles confronting the accumulation process and their resolution in the reorganisation of the labour process can be divorced from analysis of broader forms of political domination and struggle”.17
These criticisms miss the purpose of Braverman’s analysis. Unfortunately, Braverman’s death, two years after Labour and Monopoly Capital was published, prevented direct rejoinders to his critics.18 A more careful reading of Braverman is, however, enough to answer much of the dissent. After all, his decision to ignore working-class subjectivity was a conscious one: “This is a book about the working class as a class in itself, not as a class for itself”.19 Despite this, he was more than aware of the importance of workers’ subjectivity. Beneath the apparent habituation of workers to their conditions of work:
The hostility of workers to the degenerated forms of work that are forced upon them continues like a subterranean stream that makes its way to the surface when employment conditions permit or when the capitalist drive for a greater intensity of labour oversteps the bounds of physical and mental capacity.20
The successful intent of his book was to outline tendencies that, unimpeded, would lead to a complete separation of conception and execution and the total dominance of capital. Simultaneously, Braverman explicitly acknowledged the impossibility of this outcome because of the real complications, resistances and impediments faced by capital.
Elger was therefore correct that “Braverman establishes the basis for a general and abstract impulsion of capital towards the ‘real’ subordination of labour.” However, he was wrong to claim that Braverman “directly identifies this abstract impulsion with a uniform process of degradation of craft skills”.21 Features of the capitalist labour process such as the separation of conception from execution are best regarded as tendencies. A tendency need not be realised in every instance.22 Particular historical formations differ, but Elger’s insistence on emphasising “the complex, uneven and contradictory character of the organisation of collective labour” ultimately leads to a atheoretical empiricism.23 The role of theory is to abstract from complexity and variation in order to understand the dominant connections between features and developments. Despite this, Braverman was well aware of the obstacles to management unilateralism and the complete hegemony of capital:
This displacement of labour as the subjective element of the process, and its subordination as an objective element in a productive process now conducted by management, is an ideal realised by capital only within definite limits and unevenly among industries.24
Despite Braverman’s clear dismissal of the possibility of absolute management control, Elger felt able to reduce his position to one advancing “the simple conception of deskilling”.25 Adding to Elger’s criticisms, Craig Littler and Graeme Salaman opened a new front, accusing Braverman of ignoring alternative flows of income to firms that play a major or “even predominant” role, such as “currency speculation, asset stripping, commodity speculation and credit manipulations of various kinds”.26 However, such arguments displace the primacy of the production of surplus labour over those sectors in which companies subsequently compete for a share in its redistribution. In doing so, they further encouraged a move away from the focus on the workplace and the centrality of labour. In order to understand modes of workplace control, Littler and Salaman turned to Richard Edwards’ distinction between technical and bureaucratic control.27 Technical control is embedded in machinery, and this category is compatible with Braverman’s analysis. Bureaucratic control, on the other hand, is institutionalised through hierarchical job categories, positive rewards and disciplinary sanctions. Littler and Salaman argued, “Edwards’ analysis is a considerable improvement on Braverman’s more rigid insistence on the significance of just one characteristic form of control under capitalism”.28 Yet their judgement has not weathered well. Professional and administrative work has been subjected to the same Taylorist forms of management as other types of work, including in the public sector, where bureaucratic organisational relationships were most heavily concentrated. The early indications of this tendency were highlighted by Braverman, whose observations have since been deepened by research on state administration, call centres and tax offices, and the extension of performance management generally.29
Much early criticism of Braverman’s work came from people who regarded themselves as part of the radical left. Prominent amongst these was the British Marxist social theorist Michael Burawoy. His criticism of Labour and Monopoly Capital, centring on its division between “objective” and “subjective” conditions of the working class, was not novel. Burawoy argued that “the productive process must itself be seen as an inseparable combination of its economic, political and ideological aspects”.30 In particular, drawing upon an earlier study, he defined the problem for capitalists as “securing and obscuring surplus” through winning workers’ “consent”.31 This allows capital to “conceal relations of production while coordinating the interests of workers and management”. The failure to obscure exploitation, thereby undermining consent, would result in the system becoming unsustainable without the emergence of a despotic workplace regime. The weakness of Taylorism was that it “fostered antagonisms between capital and labour, making the coordination of interests become less feasible and reliance on coercive measures more necessary.32 There was no recognition that exploitation might be obvious to workers even while economic circumstances compelled compulsion compliance. This allowed Burawoy to posit “limits on the form of the separation of conception and execution”. The result of this would supposedly be a tendency towards equilibrium: “Too little separation threatens to make surplus transparent, but too much threatens the securing of surplus. The capitalist labour process—in all its phases—is confined within these historically variable limits”.33
Breaking free of Marxist shackles
Some tried to defend Braverman’s analysis. In particular, Peter Armstrong made a spirited attempt to point out misreading and misunderstandings of Braverman’s work.34 He dismissed the charge that Braverman had proposed an “iron law” of deskilling and was guilty of technological determinism, and he rejected critics’ use of the particular to refute general tendencies. For Armstrong, Braverman’s opponents were attacking a strawman. The details of his defence of Braverman remained substantively unchallenged, but Armstrong’s work was nevertheless viewed as an output of “the orthodox disciple” and dismissed as “preventing the construction of a dynamic theory”.35 Armstrong thus failed to interrupt growing criticism of labour process theory within academic circles. Indeed, criticisms soon reached a critical juncture, as evidenced by the publication of David Knights’s and Hugh Willmott’s collection, Labour Process Theory.36 There, employment studies specialist Paul Thompson questi0ned the “intellectual validity and purpose of ‘labour process theory’” altogether.37 He cited John Storey’s contention that:
The labour process bandwagon has run into the sand. Indeed, the catalogue of amendments and criticisms attaching to labour process theory has led a number of critics to call for little less than the abandonment of labour process theory.38
Thompson’s response to this challenge established him as the leading proponent of a new orthodoxy. He contended that “the biggest problem with the original thesis was not in the extent of deskilling”, but in “the implicit or explicit assumptions made concerning its consequences in terms of homogenisation, degradation or proletarianisation of labour”.39 He added that “Braverman’s specific ideas about deskilling and managerial controls made it difficult, if not impossible”, to determine the composition of a core theory. Nevertheless, it might be more accurate to state that the objection was exactly to the Marxist core of labour process theory. A labour process project was to be constructed in contradistinction to Braverman, built around four crucial tenets. First:
As the labour process generates the surplus and is a central part of human experience in acting on the world and reproducing the economy, the role of labour and the capital-labour relation is privileged as a focus for analysis.40
Although Thompson is clear that this process involves exploitation, it does not necessitate a labour theory of value. In its place, he substitutes an alternative formulation, which “rests on the appropriation of the surplus labour by capital based on its ownership and control of the means of production, and the separation of direct producers from those means”.41 Moreover, having claimed the “privileged” position of the capital-labour relationship, he immediately circumscribed its extent: “There was no assumption that the capital-labour relation had any specific significance for analysing other social relations outside production”.42 It is unclear whether this implies that employees’ workplace relations do not directly influence their access to housing, mortality rates and the educational attainment of their children. In any case, these are all areas where there is demonstrable evidence of differential experiences based on class.
Second, there is “a logic of accumulation that forces capital to constantly revolutionise the production process”.43 However, the logic of accumulation “has no determinative link to any specific feature of the labour process, such as the use of skills.” Thus, “the division between intellectual and manual labour, hierarchical control and deskilling…are not inviolable laws. At any given point capital may reskill, recombine or widen workers’ discretion and responsibility”.44 It would seem, therefore, that competition between capitals has, within this framework, remarkably little determinate influence on anything.
Third, there is a “control imperative” —a tendency to seek greater control over the work process—but once again the operation of this imperative opens up the field to the widest possible outcomes:
We recognise that the control imperative specifies nothing about the nature, specificity or level of control mechanisms. Nor is it necessarily linked to the concept of managerial strategy.45
Fourth, although Thompson recognises that the social relation between capital and labour is an antagonistic one, he also states:
Precisely because capital has continually to revolutionise production and labour’s role within it, it cannot rely wholly on control or coercion. At some level, workers’ cooperation, creative and productive powers, and consent must be engaged and mobilised.46
These four tenets undermined the construction of any substantive theory. Thompson’s project is thus largely empirical and “draws heavily on Marx’s categories, but is not…Marxist”.47 This conclusion was reached “not primarily because of the rejection of any specific element…such as the labour theory of value”:
Rather it is because there is a direct and empirically unsustainable link in Marxism between the analysis of the capitalist labour process and the theory of social transformation through class struggle… The labour process analysis outlined above cannot provide a theory concerning the behaviour of employers and workers based on identifiable sets of interests generated within production.48
All structured, collective conflict has all but been dissolved in this framework. It is hard not to reiterate Littler’s question: “Does the notion of the labour process make much sense independently of, for example, the labour theory of value?”49 According to Thompson:
The form, content and historical development of changes in the labour process have to be established empirically, rather than “read off” from any general categories. There are no specific imperatives in the sphere of control, skill or indeed anything else.50
What is left if this claim is accepted? It is easy to conclude that there is still one imperative at work in Thompson’s analysis: the explicit need to sever labour process theory from its origins in Marxism and its radical practical implications.
The labour process and class relations
For Braverman, changes in the organisation of work on both micro and macro levels had implications for the class structure of capitalism. However, the significance of the relationship between the labour process and class is almost universally absent amongst Braverman’s critics, who ignore the impact of changes between and within occupations on class relations. The exception to this is the work of Chris Smith and Willmott, who were exercised by Braverman’s notion of “intermediate employees”—those in engineering, technical, scientific, lower supervisory, managerial and professional services roles.51 Intermediate employees share “the characteristics of worker on the one side and manager on the other in varying degrees”.52 Guglielmo Carchedi argued that at least some of these employees are part of new middle class. In place of Braverman’s and Carchedi’s approaches, Smith and Willmott favoured the descriptions contained in Barbara and John Ehrenreich’s Professional Managerial Class.53 For the Ehrenreichs, the professional managerial class comprised “technical workers, managerial workers and ‘culture’ producers’”, encompassing “teachers, social workers, entertainers and so on”.54 Mirroring the long tradition of conservative social science, they emphasised that “expansion of non-manual labour within both public and private sectors” mediated “the structural antagonism within capitalist society”.55 Such a theory ignores how this growing strata might instead gravitate towards one pole of that antagonism, thus intensifying moments of crisis and struggle.56
Smith and Willmott situated their account within a peremptory reading of Marx, and their own propositions were weakened by terminological inconsistencies. They argued that Marx’s treatment of this social layer was ambivalent. On the one hand:
From a Marxist perspective, it can be argued that junior and middle managers perform the functions of controlling the workforce initially undertaken by the capitalists. In this functional light, they are seen to comprise a fundamental part of a bourgeoisie that is segmented into owning and controlling components.57
None of the leading Marxist writers in this area would support this position, which would be much less contentious were it focused on chief executives or at least senior managers. Illustrating the supposed ambivalence of Marx’s attitude to this social layer, they continued: “It may be countered that the junior and middle ranks of management have become an integral part of the collective labourer, and therefore bear a much closer resemblance to that of their subordinates”.58
On the question of proletarianisation, Smith and Willmott have an initial position that is superficially more justifiable:
Proletarianisation as theorised by Marx simply means shifts in the character of labour into a wage labour form… In recent debates this definition has been unconsciously reconstituted to include changes within formal wage labour positions, in particular changes in the conditions of work and social position experienced by white-collar workers.59
Stemming from their wish to maintain that all waged staff are workers, Smith and Willmott’s statement was particularly aimed at Weberian-influenced writers who have identified skill and autonomy as intrinsically non-proletarian qualities. Such a position would suggest that the loss of these qualities implies a move from one class to another. In this, Smith and Willmott are correct: workers’ skill levels and limited autonomy within the labour process do not automatically signal their membership of the middle class. The majority of people being deskilled and subjected to greater control are employed not in situations of what Marx called “formal” subordination—that is, employed by capital but continuing to utilise pre-capitalist labour processes—but rather in “real” subordination in the specifically capitalist mode of production.60 Capitalism proper revolutionised the labour process through the increased division of labour, scientific advances and capital-intensive machinery, forestalling any possibility of a return to independent production. In such circumstances, all relations came under the direct sway of capital.
The drawback of Smith and Willmott’s treatment of proletarianisation is that it is tied solely to defending the idea of a large heterogeneous working class, defined only by the fact that all of its members receive wages. This downplays the difference between the formal status of some waged work and the reality of its roles in the workplace. Braverman had warned of this danger:
I have no quarrel with the definition of the working class on the basis of its “relationship to the means of production”, as that class that does not own or otherwise have proprietary access to the means of labour and must sell its labour power to those who do. But in the present situation, when almost all of the population has been placed in this situation and his definition encompasses occupational strata of the most diverse kinds, it is not the bare definition that is important but its application.61
Supervisors, for instance, are problematic. Yes, they are “wage workers”, but, at least while supervising, they do not add value to any product or service. Of course, as noted above, they might also perform other roles that do add value—coordination of the labour process was recognised by Marx as necessary for any complex form of social production and should be regarded as part of the collective labour process. Moreover, the continuous restructuring of capitalist production changes the roles of both workers and supervisors and, in turn, class relations. None of the richness and subtlety of the class transformations at the point of production are captured by Smith and Willmott’s approach.
It was Carchedi who highlighted the continued relevance of the production and appropriation of surplus value to the determination of class. In a pure model of capitalism, the bourgeoisie and the proletariat have diametrically opposed relationships within the production process. The capitalist owns the means of production (owner), does not produce value (non-labourer) and appropriates surplus value (exploiter) or surplus labour (oppressor). The worker does not own the means of production (non-owner), produces surplus value or surplus labour (labourer) and has their surplus value and surplus labour expropriated (exploited or oppressed). In this model, Carchedi maintained, following Marx, that although the capitalist has a definite role to play in the production process, a role necessary to guarantee the production of surplus value, the capitalist does not take part in the labour process because they add no use values to the final product. Carchedi’s distinct contribution here is his reconceptualisation the role of the capitalist in the process of production not as unproductive labour, but as non-labour, as outside the labour process entirely.62
The actual course of capitalist development deviates from the pure model. Rather, it is mediated by a multitude of other influences including workers’ responses and class struggle. As the capitalist production process was established, the mode of labour was revolutionised. Less and less could an individual be said to be the direct producer. Products became the result of socialised collective labour, extending what and who could be regarded as productive labour and productive workers. The socialised and collective nature of labour also signalled a transformation in the function of the capitalist within the production process. The individual capitalist was replaced first by a manager and then by a managerial hierarchy. The employees engaged within this hierarchy are paid salaries or wages, but this latter fact frequently masks their role, and is not sufficient to make them workers. As Marx stated:
To the money capitalist, the industrial capitalist is a worker, but his work is that of a capitalist: an exploiter of the labour of others. The wage that he claims and draws for this work is precisely the quantity of others’ labour that is appropriated. It depends directly on the rate of exploitation of this labour, as long as he makes the effort required for this exploitation.63
These simultaneous changes—the growth of the collective worker and the growth of managerial hierarchies—have complicated this situation. With the transformation of the social structure of the workplace, fewer occupations correspond with one function or the other in a pure way. Often jobs have both collective worker and capitalist functions. On one hand, these jobs involve coordinating tasks and contributing technical knowledge, but on the other hand, they involve controlling and supervising labour on behalf of capital. Carchedi designated people in these locations as a “new middle class”. Some of these employees almost exclusively perform roles formerly carried out by the capitalist; like their roles, their interests are tied to capital. Others, at the lower end of the managerial hierarchy, tend to have greater roots in the labour process and therefore a much more tenuous loyalty to capital.64
Not just theory: changes in class relations
How class relations at work change as labour processes are transformed can be illustrated by concrete examples. For instance, traditionally, front-line managers in British tax offices came from the workforce, and they frequently advised their teams on tax questions. They were part of the same bargaining units as these teams and belonged to the same union, sometimes as leading members. Through the introduction of lean production methods, senior management disrupted these relationships by assigning roles to front-line managers that were focused almost exclusively on monitoring output targets, increasing their disciplinary role. These changes were resisted, but their aim was clear: to reinforce the function of capital in front-line managers’ roles and to lessen their involvement in the actual labour process. They remained waged, but their class relations changed.65 When working-class struggle is moving forward, workers encroach on managerial control of the labour process by reducing the authority of immediate supervision. In periods of ruling class offensive, the opposite occurs.
This movement towards greater managerialism also happened within the English and Welsh school system following the 2003 “workforce remodelling agreement”. This agreement was signed by the majority of the teacher unions (although the National Union of Teachers opposed it) because it promised to ease teachers’ workloads by assigning a number of tasks to teaching assistants, thus redefining teachers’ roles. However, this promise was empty, and other changes increased expectations and pressure on teachers. Work intensified, surveillance of performance increased and middle management expanded. Before the agreement, a number of teachers in each school received management allowances for all sorts of responsibilities, such as timetabling and heading a department or division. Head of department roles and other such positions were associated with greater expertise in pedagogical techniques and subject areas. Management allowances were removed by the agreement and replaced by a smaller number of enhanced “teaching and learning responsibility” (TLR) payments. Recipients of TLR payments were responsible for the performance of their area and the effectiveness of their staff. In this way, these roles were partially transformed, putting them at a greater remove from the teaching process and the addition of useful knowledge. Instead, TLR roles, at least partly, acted as transmission belts for government policies through the surveillance and control of others’ labour.66
None of these changes in class relations at work are acknowledged by mainstream labour process theory. Indeed, Smith and Thompson denied the possibility of a labour process theory account of these phenomena, declaring that any attempts to “reconnect the analysis to class theory” were “flawed enterprises”.67 Thus, the very possibility of such an analysis is denied. This denial was simultaneously an attack on the concept of a new middle class formed by the contradictory social relations of capitalist production. Smith and Thompson cited “writers in the late 1970s and early 1980s” who, they claimed, held that “the middle class and not the working class was the new force in capitalism”.68 No evidence was produced to link these unspecified writers (presumably Carchedi and Nicos Poulantzas) with this argument. Nor was any evidence adduced for a further accusation that they ignored the “consequences of real ownership relations” by focusing instead on “control relations between types of workers and managers in the production process”.69 Both Carchedi and Poulantzas specifically distinguished between legal and real forms of ownership.70
Smith and Thompson’s argued that the very idea of a new middle class implied that “the senior manager and the lowly supervisor, leading hand or team leader shared…the same class by virtue of their involvement in controlling workers in the interests of capital”.71 Their objection to the terminology of the “new middle class” was far from unique. Indeed, in many ways, it matters little whether we refer to “contradictory class locations”, a new “intermediate class” or a new “petit bourgeosie”. What does matter is that the underlying social relations are made clear. None of these terms imply that this class is unified and cohesive, just as the “petty bourgeoisie” of classical Marxist theory lacked clear and independent interests.72 Carchedi, who was the most prominent adopter of the “new middle class” category, was aware of the varied nature of the roles associated with ensuring the extraction of value from labour, as well as their complex relationships with, and direct involvement in, the labour process.
Smith and Thompson’s final criticism is starkly inaccurate. They characterise the writers making links between the labour process and class relations as having “an aversion to empirical analysis”. This aversion apparently flows from their “very sterile functionalist project of manufacturing classes out of the technical division of labour within waged labour in a pure and abstracted capitalist system”.73 However, even though empirical research was a weakness of the progenitors of the labour process theory approach, this does not immediately invalidate their theoretical contribution. Moreover, Smith and Thompson themselves exhibited the same weakness with their lack of references to studies of managers, supervisors and the consequences of restructuring social relations in production.
Contemporary issues and partial responses
This critique of the current state of labour process theory has centred largely on prominent members of the grouping that organise the International Labour Process Conferences. This is justified by the extent of its influence; the 2021 conference website lists 28 books published in the period 1985-2021.74 The absence of Marxist responses to the triumph of this revisionist project is telling. Even in collections supposedly aimed at promoting Marxism, overtly Marxist views are ignored in favour of accounts arguing against them, although overt criticisms of Marxism are more muted and commonalities stressed in these contexts. Hence, Thompson and Smith have focused on “recent attempts to create more active linkages between political economy and workplace relations”.75 They cite Thompson’s work on “disconnected capitalism” and Smith’s on labour mobility as examples of this.76 Nevertheless, the results of these linkages remain vague. If anything, they further move focus away from the workers’ exploitation at the point of production.
The further reach of analysis into some areas, and the absence of engagement in others, has resulted in the emergence of specific challenges to orthodox labour process analysis. Below, I list four examples of this.
The movement away from concern with production is clearest in Smith’s account of “labour mobility”, that is, the capacity of workers to choose to which capitalist they sell their labour. Smith suggests that this ability to move jobs, which distinguishes capitalism from other modes of production, is a power equivalent to industrial action within the workplace. By Smith’s own admission, this view was hard to justify, and he acknowledges that noted researchers “see individual and collective workplace action as superior to market-based dispute resolution based upon the labour market”.77 He also recognised but “one of the most detailed studies of labour turnover within an explicitly labour process perspective” concluded that “quitting was unable to resolve collective grievances, and it was therefore not necessarily a strategy that furthered worker’s interests as a whole”.78
Smith’s argument was underpinned by reference to the impact of labour mobility in competitive labour markets. However, since the end of the long boom, insecurity has been a general feature of workers’ employment, and so Smith’s failure to reference Marx’s concept of the “reserve army of labour” was a big weakness. Marx’s starting point, described here by Sébastien Rioux, Genevieve LeBaron and Peter Verovšek, was much sounder:
Within a free labour market the worker is not only free in the sense that “he can dispose of his labour power as his own commodity”, but also in that he is “free” from other sources of sustenance and “has no other commodity for sale”… The worker is therefore “free to starve” if he does not enter into a “free” labour contract.79
Compliance and consent
Smith, in contrast to Braverman, placed emphasis on “compliance and consent”, suggesting that employers can use labour power more productively by engaging with it rather than controlling it: “Groups of relatively autonomous workers, who are increasing as manual labour declines in certain parts of the world economy, either cannot, will not, or do not need to be tightly controlled”.80 However, although there may be relatively autonomous workers in certain (unspecified) parts of the world, the growth of millions of workers in China and South East Asia in modern factories and under brutal supervision would seem to bear out the progress of a tendency towards direct control over the labour process. Indeed, this is even the case in established capitalist countries. In these parts of the world, Smith claims, “appeals to professional values, creativity, career, goodwill, or trust” are “deemed more suitable methods of translating the capacity of skilled and professional workers into labour effort and value”.81 Yet there is much evidence to the contrary. British research, based on six organisations that had the components of “high performance work systems” (greater autonomy, participation and partnership‐based union involvement), threw doubt on the sustainability of the argument that the systems improved “employees’ quality of working life”.82 Similarly, in central and local state employment, where professional employees are most numerous, new public management and performance management regimes have limited their autonomy and discretion.83
Smith also criticised Braverman’s claim that there is a tendency for labour to be degraded by the controls imposed by capitalist production: “Although Braverman judged capitalism to possess a ‘degradation imperative’ whereby high value skills are replaced by low value ones, in practice, this is one tendency among several and is more contingent than absolute”.84 As Irena Grugulis and Caroline Lloyd have pointed out, however, Braverman recognised that some upskilling took place, but the question was whether this raised average skills or polarised them. They concluded, “Much of Braverman’s critique remains relevant today, in particular, his challenge to optimistic ‘upskilling’ arguments and his awareness of the caution required in using data and the ‘misuse’ of the term ‘skills’”.85 Beynon has also written persuasively on the bifurcation of skills in contemporary capitalist production.86
The concept of emotional labour has gained increasing traction in labour process analysis. According to Thompson and Smith, “Marx highlighted the embodiment of labour as consisting of male, female and child categories…and today we would extend these to the emotional and aesthetic aspects of labour”.87 This claim was used to illustrate how labour process theory “has been at the forefront of research into new sources of labour power”.88 However, suggestions that the notion of emotional labour is an advance on Marx’s understanding is unsustainable for two reasons. First, Marx made clear the complex and multifaceted nature of labour power, which he described as “the aggregate of those mental and physical capabilities existing in the physical form and the living personality of a human being.” These are “capacities that he sets in motion whenever he produces a use-value of any kind”.89 Second, the term was popularised by Arlie Russell Hochschild, who explicitly acknowledges her debt to Marx.90 Once the concept of emotional labour was embraced by the new labour process writers, above all by Sharon Bolton, it took on a separate and independent status from other aspects of labour power.91 Moroever, Bolton’s definition of emotional labour was restricted to that directly involved in the accumulation of capital, ignoring both unproductive labour in the public sector and socialised collective labour more generally.92
Digital capitalism and platform work
Simon Joyce has noted that academic research into platform work—that is, “paid work mediated via an online platform”—has increased in recent years. Yet, although empirical studies of this phenomenon have proliferated, “to date these have generated little theory”.93 Joyce has examined the relations involved in these forms of work from a Marxist perspective by mobilising the concept of subsumption:
The process of formal subsumption is not found only at the birth of capitalist relations. Rather, it is also a recurring feature of mature capitalism, associated with the emergence of new branches of production.94
Joyce believes capital’s relationship with platform workers contradicts the contention that they are self-employed. He instead regards their payments as analogous to piecework. By “grasping platform work as a social relationship between labour and capital”, Joyce claims that he “de-prioritises the legal conceptions of employment that frequently dominate discussion in favour of a more sociological approach”.95
Conclusion: reiterating the importance of a critique of the capitalist labour process
The strong interest in the capitalist labour process engendered by Braverman’s work encouraged socialists to look at relations in the workplace in a new and different light. However, this initial reception remained underdeveloped, and the radical implications of Braverman’s Marxist approach were blunted by labour process theory’s incorporation into academic debates amid a downturn in class struggle. Although there were critics of the revisionism that resulted, their arguments for greater concentration on the labour theory of value, exploitation and class relations were summarily dismissed or ignored altogether.96
An appreciation of Marx’s and Braverman’s approaches to the capitalist labour process leads us towards questions about class relations, the inseparability of the exploitation, oppression and alienation of workers, and the importance of workers’ self-activity. Working-class emancipation has to start with struggles for collective autonomy and control at work—extending the “frontier of control”. This requires taking back aspects of the organisation of work and transforming them from functions of capital (control) to aspects of labour (coordination). This necessitates challenging current forms of bureaucratic trade unionism that minimise issues of job control and dignity, focusing instead on pay while accepting subordination.
Writing 40 years ago, Theo Nichols asked, “What is a ‘labour process’ approach?” His answer was that, in contradistinction to “industrial”, “organisational” and “managerial” studies, the labour process approach arises out of “a primary concern with capital accumulation and class struggle”.97 On this measure, today’s mainstream labour process theory falls short. Nichols has subsequently argued that, in the arc of post-war labour studies in Britain, many of the same preoccupations have remained: an interest in subjectivity; the relative neglect of capital accumulation; and, related to this, a focus on managerial concerns.98
This tendency has been bolstered by concerted attempts to sever links between the concept of a capitalist labour process and its Marxist origins. Many texts acknowledge Braverman’s importance in reviving the subject, but his ongoing contribution is largely demeaned. Consequently, some areas are badly served and substantive issues unresolved. The relationship between the labour process and other areas of capitalist society has been an issue of contention from early on, when stress was laid on both the plant-level autonomy of the labour process analysis and the need to examine the full circuit of capital. The preponderance of case studies has accentuated the difficulties of integrating analysis with other areas, insights and disciplines. If Nichols’s prescriptions were taken up and the project looked more broadly at the conditions for the production of surplus out of labour power—both theoretically and in specific capitalist societies—problems of focus and boundaries would largely dissolve. This is illustrated by Nichols and Beynon’s Living with Capitalism (1977), which deals with migration, relocation of industry, levels of capital expenditure, automation, masculinity, the labour of superintendence, trade unions and other topics. This book’s coherent and convincing picture of significant social and class developments would have been impossible without the overtly acknowledged influence of Marx and Braverman. Using their reflections to understand social class relations and changes in the labour process in order to build a viable socialist organisation remains of central importance today.
Bob Carter was Professor of Work and Employment Relations at the University of Leicester. He is author of Capitalism, Class Conflict and the New Middle Class (Routledge, 2015) as well as a number of articles on trade unions and the restructuring of public sector labour.
1 Thanks to Joseph Choonara for making helpful suggestions and asking for clarifications. I am also grateful to Theo Nichols for both his long-term influence and recent encouragement.
2 Braverman, 1974, p9.
3 Braverman, 1974, p10.
4 Wood, 1999.
5 Fukyama, 1992. For a criticism of post-modernism, see Callinicos, 1991.
6 Thompson, 1968; Miliband, 1985.
7 Marx, 1976, p981.
8 Marx, 1976, p981.
9 Nevertheless, managers are part of the labour process when engaged in coordinating or unifying parts of that process and when carrying out tasks that are necessary under any system of collective production. Of course, which parts of a manager’s tasks fall into which social function are impossible to list in advance of a concrete analysis.
10 Marx, 1976, p990. It is not just Marx who expressed this opinion. Alfred P Sloan, the president of General Motors from 1923, once admitted: “General Motors is not in the business of making cars. It is in the business of making money”—Locke, 2019, p92. Indeed, Sloan’s career attests that all sorts of research and production directed at creating socially useful products is neglected because it threatens profitability. As sociologist Daniel Bell explained, “Sloan’s memoir tells us how, in the 1920s, General Motor’s head of research, Charles F Kettering, wanted to build an air-cooled car engine, which would have then revolutionised the automobile industry. However, such a car would have interfered with the marketing strategy of the company, and Sloan killed it”—Bell, 1964.
11 Braverman, 1974, p46.
12 Braverman, 1974, p57.
13 Marx, 1976, pp985-986.
14 Braverman, 1974, p267.
15 Braverman, 1974, pp118-119.
16 Braverman attempts to be consistent with Marx when looking at the foundations of the capitalist labour process. However, this consistency fails to stretch to wider questions of political economy, where he uncritically cites Baran and Sweezy’s conflation of surplus value with surplus—see Baran and Sweezy, 1978, and Paul Mattick, 1978. Braverman is also unclear what sort of change the division of labour between conception and execution really represents. Is it a further socialisation of labour and the growth of what Marx termed the “collective worker”? Or is it the transformation of what had been workers’ control and discretion into its opposite: supervision and control by capital? The latter would mean that conceptualisation had gone from being part of the labour process to a becoming a “function of capital”. Colin Barker’s early review of Labour and Monopoly Capital in this journal inadvertently reflected this weakness, uncritically replaying the two contrary positions taken by Braverman. Thus, Barker writes, “The conceptual, planning, thinking part of work is taken away from the worker and concentrated in management. Work is ‘deskilled’.” Simultaneously, however, he claims, “‘Mental’ and ‘manual’ work are, as Marx argued, more and more separated into different places and different sections of the labour force”—Barker, 1976. This problem is addressed by Carchedi, 1977.
17 Elger, 1979, pp59-60.
18 Braverman was unable to respond to a number of early socialist feminist critics unhappy about his lack of treatment of household work, which he considered beyond the bounds of his subject matter—see Braverman, 1976. Despite this, Labour and Monopoly Capital contains much to which women are central. In particular, Braverman offers a pathbreaking analysis of the modern office and the impact of the “universal market” on household work.
19 Braverman, 1974, p27.
20 Braverman, 1974, p181.
21 Elger, 1979, p62.
22 As Paul Edwards writes, “As soon as some levels of analysis are distinguished and distinctions are made between general features of a mode of production and concrete social formations, their critical force is weakened”—Edwards, 1990, p128.
23 Elger, 1979, p83 (emphasis added).
24 Braverman, 1974, p172.
25 Elger, 1979, p84. Elger uses Theo Nichols and Huw Beynon’s 1977 Living with Capitalism to counter Braverman’s supposed simplicity. Commenting on developments at the chemical factory under study, Elger argued, “Capitalist initiatives such as ‘job enrichment’ can be understood as complex responses to the specific problems of valorisation and accumulation confronting this sector of capital, rather than being seen in generic terms as a qualification of or retreat from ‘deskilling’”—Elger, 1979, p85. Beynon himself has recently reflected on the significance of his book and stated, baldly and without qualification, that its influences were Marx and Braverman—see Beynon, 2019.
26 Littler and Salaman, 1982, p257.
27 Edwards, 1979.
28 Littler and Salaman, 1982, p263.
29 See Carter and Fairbrother, 1995, Taylor and Bain, 1999, Carter and others, 2011, and Taylor, 2013, respectively.
30 Burawoy, 1985, pp24-25.
31 Burawoy, 1985, p5; Burawoy, 1979.
32 Burawoy, 1985, p41.
33 Burawoy, 1985, p49. As late as 2015, Chris Smith lauded Burawoy as “the most significant Marxist sociologist”. He claimed that Burawoy’s 1979 book Manufacturing Consent interrogated “the shortcomings of both Marx’s (and Braverman’s) understanding of social life inside a large modern, unionised corporation with strong internal labour markets and a labour process where winning workers’ consent, not managing through coercion, was required”—Smith, 2015, p228. Burawoy’s own assessment of his intellectual heritage was far more critical and perceptive. Winning workers’ consent had turned out to be a “fleeting moment in US labour relations history”. The transformation of markets and states intensified pressure on labour. Burawoy admitted that he had failed to appreciate how “the hegemonic regime sowed the seeds of its own destruction. In atomising workers, and thus forging industrial citizens, and in tying the interests of labour to those of capital, the hegemonic regime undermined labour’s opposition to management and its capacity to resist employer offensives”—Burawoy, 2004, p5-6.
34 Armstrong, 1988.
35 Thompson and Smith, 2001, p45.
36 Knights and Willmott, 1990.
37 Thompson, 1990, p95.
38 Storey, 1985, p194.
39 Thompson, 1990, p97.
40 Thompson, 1990, p99.
41 Thompson, 1990, p100.
42 Thompson, 1990, p100.
43 Thompson, 1990, p100.
44 Thompson, 1990, p100.
45 Thompson, 1990, p101.
46 Thompson, 1990, p101.
47 Thompson, 1990, p102.
48 Thompson, 1990, p102.
49 Littler, 1990, p79.
50 Thompson, 1990, p102.
51 Smith and Willmott, 1991. In particular, Smith has adopted various positions on the nature of social class. For an early account, see Carter, 1995.
52 Braverman, 1974, p405.
53 Ehrenreich and Ehrenreich, 1979.
54 Smith and Willmott, 1991, p16.
55 Smith and Willmott, 1991, p15.
56 Catherine Liu has resurrected the concept in order to show how this group of elite workers has come to serve capitalism while insisting on their own virtue—see Liu, 2021. For a more rounded approach to the professional managerial class, see Ikeler, 2020.
57 Smith and Willmott, 1991, p17.
58 Smith and Willmott, 1991, p17.
59 Smith and Willmott, 1991, p22.
60 Marx, 1976, p1021. See Das, 2017, for an examination of these issues.
61 Braverman, 1974, p25.
62 Carchedi, 1977.
63 Marx, 1981, p511.
64 Carchedi also argues that even those who exclusively carry out the functions of capital at a low level within the overall hierarchy are not capitalist because they have neither legal nor effective ownership over the means of production. His analysis is much more developed and sophisticated than it is possible to relate here and, although neglected, deserves to be read more widely.
65 Carter and others, 2014.
66 Carter and Stevenson, 2012.
67 Smith and Thompson, 1999, p205.
68 Smith and Thompson, 1999, p217. This inaccurate criticism is ironic given the mainstream approach is built upon rejecting the claim that the working class was “the gravedigger of capitalism”. Interestingly, the importance of the phrase, which comes from The Communist Manifesto, has been scrutinised by Matt Vidal, who maintains the idea played little or no role in Marx’s overall writings—Vidal, 2018. Questioning of the “gravedigger thesis” has become a shorthand for attacking Marxism’s supposed “mechanical determinism”.
69 Smith and Thompson, 1999, p218.
70 See, for instance, Carchedi, 1977, pp161-162 and Poulantzas, 1978.
71 Smith and Thompson, 1999, pp218-219.
72 As Trotsky noted, the middle class can mobilise behind either the workers or capitalists in a crisis.
73 Smith and Thompson, 1999, p219.
74 Subjects vary from retail work and the new digital workplace to creative labour. Many of these works exemplify the weakness of the mainstream approach, which is, for example, illustrated by the absence of references to Amazon, monopolisation and proletarianisation in the Irena Grugulis and Ödül Bozkurt’s book on retail work—see Grugulis and Bozkurt, 2011.
75 Thompson and Smith, 2017, p123.
76 Thompson, 2013; Smith, 2003.
77 Smith, 2006, p392. Smith is referring to Beynon, 1973, Burawoy, 1979, and Nichols and Beynon, 1977.
78 Smith, 2006, p393.
79 Rioux, LeBaron and Verovšek, 2020, p713. A later work on working-class formation in China by Smith and co-author Ngai Pun gave no weight to the concept of “mobility power”. Moreover, its conclusions were in some tension with Smith and Thompson, 1999, which claimed that attempts to reconnect labour process analysis to class theory were inevitably flawed. Instead, Smith and Pun spoke of workers’ “underlying need for expressions of class consciousness and identity”. They also argued, “The main pressure point of class formation resides not in formal or semi-formal and legal or para-legal status distinctions between workers…but in their ability to mobilise collectively (for example, through strikes and protests) on the basis of their grievous experiences as workers in the capitalist labour process”—Smith and Pun, 2018.
80 Smith, 2015, p226.
81 Smith, 2015, p226.
82 Danford and others, 2008.
83 Carter, 2020.
84 Smith, 2015, p226.
85 Grugulis and Lloyd, 2010, p106.
86 Beynon, 2019.
87 Thompson and Smith, 2017, p199.
88 Thompson and Smith, 2010, p16 (my emphasis).
89 Marx, 1976, p270.
90 Hochschild, 1983.
91 Bolton, 2005 and 2010.
92 See Brook, 2009, for an extensive critique of Bolton.
93 Joyce, 2020, p542.
94 Joyce, 2020, p545.
95 Joyce, 2020, p546.
96 Cohen, 1987; Carter, 1995; Nichols, 1999; Spencer, 2000.
97 Nichols, 1980, p17.
98 Nichols, 1999.