Arendt’s anti-humanism of labour

Georges Friedmann and the humanism of labour

‘Humanism of labour’ was a banner used by some leading advocates for labour reform in mid-twentieth-century Britain, the USA and especially France. The expression had both a broad and a narrow meaning. In its broad meaning, it referred to a wide range of social and economic reforms aimed at ameliorating the conditions of working people. The conditions of workers stood to be improved by such things as better job security, safer working environments, proper training, holidays, decent leisure opportunities, and so forth. The humanism of labour, in pursuing such reforms, sought to enhance the quality of life of the industrial labourer, and in that general sense to ‘humanize’ it.

In its narrow meaning, the humanism of labour referred to more specific reforms concerning the activity of labouring itself: the design, organization and distribution of tasks the labourer had to perform in the contemporary phase of industrial production. Since its beginnings, industrial production had relied on methods that fractionated complex tasks into simple performances that could be done quickly, repetitively and, with the help of machines, with vast yields. Innovations in the first two decades of the twentieth century, most notably the introduction of the moving assembly line and an increasing reliance on automated or semi-automated processes, continued this trend. Alongside this, the conditions of optimal human performance in the production line became a matter of systematic scientific scrutiny. A key breakthrough here was made with the time-motion studies conducted by Frederick W. Taylor (Taylor, 1911). The basic presupposition of this approach to the design and allocation of tasks to be performed in the course of production was that there was ‘one best way’ of doing it, an optimal solution that could in principle be established scientifically (in a laboratory) and applied in situ. However, this assumption soon came to be questioned as attempts at implementing the model met with varying degrees of success. Studies showed that there was a limit to the capacity of the labourer to adapt to rationalized (fractionated, mechanized and semi-automated) conditions of production, that beyond this threshold, rationalization became counter-productive. Furthermore, these studies brought to light unanticipated physiological and psychological harms of prolonged enforced engagement in this kind of activity. What Taylor and the whole scientific paradigm he worked in had failed to take account of − so the evidence suggested − was the human element in industrial production. It was the shortcomings of this scientistic − or as it was often called then, ‘technicist’ − approach to industrial labour that prompted calls for a ‘humanism of labour’ (Friedmann, 1955: 32). This would be a humanism of labour in the narrow sense distinguished above, namely one concerned above all to humanize the activity of labouring; to recognize and affirm the ‘human factor’ even in industrial work.

Friedmann himself used the phrase ‘humanism of labour’ in both its broad and narrow extension, but it is the latter that interests us here. What he has in mind is at once a philosophical orientation to the problems arising from participation in highly rationalized (fractionated, mechanized and semi-automated) activity and a set of practical solutions to these problems. The basic philosophical orientation at stake can be put as follows. It is true that production on the industrial scale we know today would have been impossible without the division of complex tasks into their component parts, the mechanization of many of those tasks and their reconfiguration within a semi-automated production line. Given the salience of the technical division of labour for the rationalization of production (increases in efficiency or productivity), it can seem as if the human contribution to the production process (labouring activity) is itself intelligible from a purely technical point of view. It becomes tempting to think, in other words, that the agency of the labourer can be broken down into its simple parts and ‘rationally’ reconstructed in fundamentally the same way that the tasks to be performed can be.

But this is an illusion. There is always a whole human being doing the tasks that are set to be done in the course of production, however simple, repetitive, quasi-automatic or mediated by machines those tasks are. In face of the predominance of the technical point of view when thinking about labour, we need to remind ourselves, Friedmann often insists, of the unity of the labouring individual. The person who labours in the factory is, after all, one and the same as the person who loves, thinks, suffers, has convictions, allegiances, and so forth; the activity of labour affects body, mind and soul, and thus has ergonomic, psychological and moral dimensions. For better or worse, the whole agent, or the whole personality, is drawn into the experience of labour, and this shapes the identity of the labourer in complex ways that the various sciences of labour should seek to understand.

The first step towards a humanism of labour is thus an acknowledgement of the human character of the agent of labour, whatever purpose the labour serves and in whatever manner the labour tasks are organized. This means that the technical point of view on labour can never wholly be separated from the ethical one. Once this step is taken, a number of consequences follow for thinking about how labour activities might better accord at once with technical and ethical norms. That is, the question of the practical reform of labour, of how labour qua activity ought to be, presents itself. Friedmann proposes the following three axes of reform, which together amount to efforts to ‘humanize’ the activity of labour in ways that are consistent with the level reached by the technical division of labour:

The first axis has to do with the range of capacities to be exercised in the labouring activity. As just mentioned, the reforms advanced in Friedmann’s humanism of labour are meant to be implementable within the current state of the technical division of labour − they should not presuppose a more developed state; they are not meant to be utopian − and this places great constraints on what can be achieved in this axis. There can be no return, Friedmann accepts, to a norm of ‘craftsmanship’: this type of activity, though more satisfying as an expression of practical human capacities, simply is not consistent with industrial and post-industrial methods of production, and more specifically with the technical division of labour within which most labouring activity now takes place. But that is not to say that there is no room for manoeuvre in the way those fractionated activities are designed and allocated, or no aspects of the craftsmanship model that can be retained. Given the general simplification of tasks on a production line, there is no need, for example, to limit a labourer’s activities to the performance of just one of those tasks. Task or job rotation would both widen the range of activities to be performed by the individual labourer and give that person a better sense of the production process as a whole. The attainment of ‘an over-all, rational view of the complete production process’ would also engage − and perhaps even transform − the intellectual capacities of the workers (Friedmann, 1954a: 33). It would not make ‘craftsmen’ of them but it would provide an approximation to the craftsman’s use of practical reason, and would mitigate some of the cognitively levelling consequences of the technical division of labour. Training could be given to do more complex tasks, enlarging the scope of activities to be performed and capacities to be engaged. Job rotation, job enlargement, and education and training are thus examples of how labour can be humanized along this first axis. They might seem like modest reforms, but Friedmann suspects they would be enough to satisfy demands for more engaging work activity for most workers. Friedmann is guided here by empirical studies of levels of boredom and monotony experienced during low-skilled assembly line work. He is struck both by the high level of tolerance shown by many workers to repetitive labouring activity, and by the variety of responses to that kind of activity in terms of susceptibility to experiences of boredom and monotony. A humanism of labour must be sensitive to those variations, and not assume one model of what it is to humanize work that applies equally to everyone.

Arendt’s rival theory of labour

As we have seen, part of the motivation behind Arendt’s theory of labour is to present an alternative to the humanisms of labour enjoying currency at the time, of which Friedmann’s was the most prominent. Arendt’s guiding thought here is that the humanism of labour is a project premised on a contradiction. It is a synthesis of two mutually incompatible ideas: humanism and its verb form ‘humanize’ on the one hand, labour on the other. Literally speaking, Arendt insists, the very expression ‘humanism of labour’ is ‘a contradiction in terms’. But Arendt’s objection is not just a semantic one. In her view, loose talk of ‘humanisms of labour’ and attempts to ‘humanize labour’ betray a fundamental misconception of the nature of labour and the possibilities it admits of. There is more at stake in this controversy than a difference over definitions, and in defending her own position, Arendt is doing more than merely stipulating one set of definitions rather than another. It is the misunderstanding of labour in the humanism of labour that Arendt objects to, and that misunderstanding would be present even if the humanism of labour went by another, non-absurd name.

But misunderstanding of what, exactly? The short answer is, as I began by observing, a way of leading the ‘active life’ − the vita activa led in the mode of labouring. Yet the activity of labouring is only one aspect of ‘labour’ understood in this broad sense, and much of what Arendt has to say about labour does not concern a specific type of activity at all. So, for instance, labour is distinguished from work on account of the features of the thing produced: consumable objects in the former case, objects that endure in the latter (HC: p. 94). This characterization of labour in terms of the thing produced is closely related to its characterization in terms of its existential purport or conditions of existence: coping with the demands of being a living creature, burdened with the tasks of maintaining and reproducing ‘life’, in the case of labour; coping with the demands of building and caring for a ‘world’, in the case of work (HC: p. 7). And this in turn is linked to the location of the activity, whatever that activity is, which is such a striking and contentious feature of Arendt’s theory: labour is confined to the private or quasi-private ‘social’ realm, leaving the public realm for the ‘speech and action’ proper to politics. (HC: p. 176ff). Aspects of Arendt’s depiction of the animal laborans also barely touch on the kind of activity this human type engages in, focusing instead on its preference for consumption and lack of interest in goods higher than the service and enjoyment of life.

So there is more to Arendt’s theory of labour, which is at bottom an account of the vita activa represented by the animal laborans, than an account of a type of activity. Nevertheless, the theory must provide such an account, and the account it provides should be superior to the accounts provided by its rivals. This is why it is important to contextualize Arendt’s theory in relation to its rivals; taken out of this context, the motivation behind the theory goes missing, or is at least less visible. Not least among these rival theories was the humanist one offered by Friedmann. We have just looked at Friedmann’s account of the kind of activity typical of the contemporary methods of production, which is to say the kind of activity apt for the level reached by the technical division of labour. Now Arendt broadly shares Friedmann’s view of what these methods are, but she has a different set of views of the consequences of participating in them and the scope available for reform. By reconstructing Arendt’s views on these matters, we can bring into focus her account of the activity of labour, activity that accords with the contemporary methods of production, without getting distracted by all the other things going on in her theory of labour. Moreover, this will allow us to see more clearly what is actually going on in the dialectic of The Human Condition insofar as the book offers a superior account of the nature of labour activity and its prospects than its rivals. I shall use the three axes of the reform of labour activity distinguished by Friedmann to order my reconstruction.

First, when we look at the range of capacities labouring activity draws upon, Arendt’s view takes its departure from an insight she finds in John Locke’s use of the expression, ‘the labour of our body and the work of our hands’ (HC: p. 79). We noted just now that Arendt links the capacity for labour with the condition of life. Labour is the capacity for producing things, the consumption of which keeps the life process going, and it is in the nature of such things to be perishable, to have short-term existence and to be in need of continual replenishment. The objects of labour must be repeatedly and cyclically produced. These features of the product of labour carry over to the kind of activity that produces them. That is to say, the activity itself is cyclical and repetitive, and the cyclical and repetitive character of labour becomes accentuated the more the productive process and the methods of production develop. Locke, writing along with the physiocrats at an early stage in the development of the methods of production, could emphasize how, through bodily effort, labour is ‘mixed’ with the earth or some natural material (Locke, 1967); a mixing that should be distinguished from manipulation of material (by the hands) in the case of ‘work’. The insight here is that the labour of the body is ontologically continuous with the natural material it mixes with, the mixing done by the labour serves merely to bring out a fecundity inhering in the life process itself. Marx, writing at a later stage in the development of the techniques of labour, could fill out this insight with his notion of ‘production itself’ as the ‘metabolism between man and nature’.

Adopting this idea, Arendt proposes that the whole production process, labour in general, is effectively activity of the body writ large. This means that the capacities required for effective participation in the production process are essentially those of bodily exertion and attunement with the movements of machines. The industrial labourer is effectively an appendage of the machine and has to adapt to purely mechanical and cyclical principles of movement. Arendt is guided here by empirical studies suggesting that in making this adaptation, industrial labourers come to rely on the pre-cognitive, reflex-like responsiveness of the body, which is where the ‘work’ of labour should properly be thought to take place (HC: pp. 98, n33; 145–6, n8). In her view, the aptness for mechanization and automation of the labour process reveals its underlying continuity with the life process, such that in participating in the labour process effectively, the labourer becomes attuned to the rhythms of life itself. Arendt takes this to imply that ‘nothing can be mechanized more easily and less artificially than the rhythm of the labour process, which in its turn corresponds to the equally automatic repetitive rhythm of the life process and its metabolism with nature’ (HC: p. 146). The more repetitive the labour, the easier the attunement; though it may be helped along by ‘labour songs’ and other tricks observed by industrial psychologists and anthropologists of the factory. Such maieutics may be needed because although mechanical production is congruent with the life process in its temporal sequencing, the sequences and repetitions themselves are much faster than the naturally occurring ones. Mechanized production is ‘unnatural’ insofar as it increases the speed of the performance cycles and the rate of performance repetitions, but the capacities drawn upon in those performances remain tied to the body and the propensities it has for coping with the natural condition of life.

Once the logic of labour is made explicit in this way, the scope available for expanding the capacities that may be exercised in its activity − and so for the reform of labour in this respect − must seem very narrow. Considered from the point of view of the capacities apt to be exercised, there is no essential difference between the activities of the oldest and the most recent avatars of the animal laborans: the ancient slave and the modern assembly-line worker. Both are constrained and encompassed by the exigencies of the ‘metabolism between man and nature’. There is no point trying to gloss the ‘human potential’ in such activity, and proposals to humanize it, if they are to remain at all realistic, are bound to count for little. If all that is at stake in job rotation, for example, is movement from one station of an assembly line to another, this is unlikely to widen the range of capacities exercised, so long as the same basic reflex-like capacity to respond repetitively and with speed is required at each station. Just as it would hardly be a source of solace for the slave to be able to rotate between digging for coal and lifting it to the surface, so mere alternation between tasks in a highly mechanized factory is unlikely to make the modern labourer feel more human. Job enlargement through training, so long as that training is oriented towards increasing productivity, is also unlikely to be more than an anodyne, and the idea that the labourer might be educated into attaining a ‘rational view of the complete production process’ projects a capacity onto labourers for which their activity is wholly unsuited. Arendt would have seen the latter proposal as a perfect illustration of the tendency among modern theorists to idealize labour and to project onto it possibilities that belong properly to the distinct category of work (in the sense of craftsmanship).

Second, if humanist proposals for the reform of labour by way of broadening the capacities exercised in labouring activity run up against the inner logic of labour, in Arendt’s view, so does the humanist project of enriching the experience of labour by way of transforming the social relations at stake. Arendt is far from denying that social relations are at stake in labour activity, it is just that these are not especially ‘human’ in her view, and never can be so long as the activity remains labour. This is because the sociality of labour is essentially a matter of coordination. Labour activity that accords with modern methods of production is, as a consequence of the level reached by the technical division of labour, isolated and highly atomized, but it must still be coordinated within a larger labour collective. It is the coordination of the labour of individual bodies that enables the collective body to do more, to have a greater ‘labour power’, than the individual members acting severally. In its crudest form, where the level reached by the technical division of labour is at its most primitive, the benefits of coordination, in terms of enhanced labour power, can be seen in the ‘labour gang’ (HC: p. 213). And in Arendt’s view, while the productivity of the industrial labourer is much higher than that of the member of the labour gang, and the coordination of the activities of industrial labourers a much more complex matter, the social relation between the labourers of the modern factory and between the members of the ancient labour gang is fundamentally the same. In both cases, it is the ‘metabolism between man and nature’ that is fundamentally at stake, and insofar as any properly human relations are intelligible at all in this context, they can be so only in the most attenuated sense.

In practice, this means that one should be sceptical about the prospects for the humanization of labour through reform of the social relations that mediate labour activity. As one might expect, Arendt has a dim view of the ‘human relations’ movement in industrial psychology, which she sees as raising false hopes about fulfilment through work (about which more in a moment) and as a ruse for motivating people to work harder (HC; p. 149, n12). It might be possible to create a marginally more positive attitude among the workers in a modern factory or work organization towards their labour, say, by recognizing their contribution or by granting more ‘status’ to the work role, and this might marginally improve yield. But in its fundamentals, the ‘integration’ of workers which these positive attitudes are supposed to foster is a matter of coordination of labour powers. Given the nature of the activity that is actually required in the modern work organization, there is little for those who participate in it to have positive attitudes about, little that is worthy of serious recognition among peers or by the work hierarchy, other than sheer productivity. For Arendt, the ‘human relations’ school of industrial psychology and the humanisms of labour more generally exaggerate the difference that can be made to the reality of labour by management methods that facilitate the adoption of more positive attitudes to their labour among the labour force. These include methods that allow for greater participation in the management by the labourers, and Arendt shows little interest in the various experiments in worker self-management inspired by the ‘human relations’ movement. ⁷ On the one hand, the practice of self-management on the part of the labour collective still has, as its final end, efficient production, which is no more ‘human’ or ‘humane’ for it being planned and self-managed by the workers. On the other hand, in Arendt’s view, there simply is not the appetite among the vast majority of workers for work activity ‘enriched’ in this way. For the most part, Arendt believes, participants in the contemporary labour process just do not have an interest in managing it for themselves. She cites some recent sociological surveys of attitudes to work in support of this view, including one that found that ‘a large majority of workers, if asked “why does man work?” answer simply “in order to be able to live” or “to make money”'. ⁸

Third, the respondents to those surveys, in Arendt’s view, have a much more realistic set of expectations about the possibilities of fulfilment through work than the humanist theorists of labour do. As we saw, Friedmann himself thought that the complete joys of craftsmanship were beyond the reach of the modern industrial labourer, and Arendt agrees with this. She also agrees with Friedmann that labour, as distinct from craftsmanship (Arendt’s ‘work’), is not without its own satisfactions. But she disagrees with him on the sources of these satisfactions. For Arendt, what ‘joy’ there is to be had from labour has nothing to do with self-esteem, ‘recognition’ or a sense of contributing to some larger set of social purposes, as it did for Friedmann, but rather it has exclusively to do with the cycle of toil, fatigue and replenishment characteristic of the life process (HC: p. 106). The importance of this should not be underestimated, as it is bound up with one of the fundamental conditions of human existence: natural life. It is, on account of this connection, the source of what Arendt considers to be the only ‘lasting happiness’ available to human beings (HC: p. 108). Enduring happiness, as Arendt describes it, comes not from the manifold pleasures of consumption, or for that matter from hard-won individual achievements, but from being part of the repetitive cycle of effort, weariness and regeneration that characterizes life itself. Conversely, exhaustion that is followed not by pleasurable regeneration but by ‘wretchedness’, or, at the opposite extreme, an ‘entirely effortless life’ that has nothing to renew itself from, lacks a crucial life good: ‘the elemental happiness that comes from being alive’ (HC: p. 108). Labour, being fundamentally an activity of the body, occasions a range of joys and sorrows that have the content they do solely on account of the body’s immersion in the life process.

For Arendt, the main problem facing modern societies as they enter the post-industrial phase in regard to the rewards of labour is not that of finding a substitute for the intrinsic satisfactions of craftsmanship − in Arendt’s view, that is a forlorn hope − but of finding an alternative mode of access to the pleasurable cycle of effort, fatigue and regeneration. Arendt shared the view of many of her contemporaries, including Friedmann, that the mechanization and automation of production have the long-term effect of taking the toil out of labour and making human labour less necessary. The easier labour is to do, and the less of it that is needed for the continuation of life, the more distanced humans are in danger of becoming from the cyclical temporality of the life process, and the more forgetful they are prone to become about the place of laborious activity in the experience of life. The danger is that with the onset of full automation, consumption will be split off from the activity of labour altogether, leading not only to a further narrowing of the range of activity that makes up the vita activa, but to alienation from the one source of ‘lasting happiness’ available to humans on account of their embodiment. This is the society of ‘labourers without labour’ of which Arendt speaks with such foreboding (and, it must be said, contempt) in the Preface to The Human Condition (HC: p. 5).

Humanism vs anti-humanism of labour today

My concern up to this point has been to highlight the difference between Friedmann’s humanist theory of labour and Arendt’s theory. The difference is not adventitious: the humanism of labour, especially as formulated by Friedmann, was a powerful intellectual current in the 1950s, and the theory of labour set out by Arendt in The Human Condition was meant, in part, to combat it. When viewed as a rival to Friedmann’s humanism of labour, certain features of Arendt’s theory stand out which might otherwise go unnoticed, or not be noticed in the right way. In particular, Arendt’s views about the range of human capacities at stake in labour, the social relations that shape it, and the potential for joy or fulfilment from participation in contemporary (and foreseeable) methods of production, gain their force, at least in part, from the contrast they make with Friedmann’s humanist position on these matters. Humanist proposals for the reform of labour along these lines are fundamentally misconceived, in Arendt’s view, and it is part of the task of the central chapters of The Human Condition to show why. Unless we have that rationale in view, we will miss something important that is going on in Arendt’s text.

It is important to see that Arendt’s anti-humanism of labour really is a rival theory to Friedmann’s humanism of labour – that the two theories offer competing accounts of the same (or at least an overlapping) domain – and that Arendt’s and Friedmann’s theories of labour are not meant as accounts of completely different things. What is centrally at stake for both theorists is the nature of the activities undertaken in modern employment and the possibilities such activity admits of. It is through employment or the doing of a job that Arendt thinks the vita activa is predominantly played out in the modern world, which in her view amounts to it being lived in the form of labour, as distinct from work. We must read this as an interpretation of activities performed in the course of employment or doing a job, not merely a stipulation of what such activity involves insofar as it is labour. If Arendt were only saying that what one does in one’s job or employment is repetitive, mechanical, aimed at consumption, etc. insofar as it is labour, but has other features insofar as it is work, it might be true, but platitudinously so. At issue is the validity of the interpretation, or the usefulness of Arendt’s conceptual schema for understanding the activity in question. As an interpretation, it owes whatever validity it has to its superiority to other, competing interpretations. Marx’s theory provides one such interpretive schema; Friedmann’s humanism of labour another. My suggestion has been that it helps us think about the validity of Arendt’s interpretation to compare it not just to Marx’s, which is well known, but also to Friedmann’s less familiar humanism of labour.

Contemporary relevance

In the space that remains I want to consider the contemporary relevance of Arendt’s critique of the humanism of labour and the alternative she presents to it. I will confine myself to a brief consideration of how Friedmann’s humanism and Arendt’s anti-humanism of labour might provide usefully contrasting orientations for dealing with some of the main problems of labour we find ourselves facing today. Of course, the problems of labour in the second decade of the twenty-first century can hardly be the same as those of the middle decades of twentieth. Our concept of labour must be broad enough to encompass the more diffuse and differentiated experiences of labour in contemporary settings – by no means centred on industrial factory work – without being so broad as to include all modes of active experience, as if the problems of modernity could be reduced to the problems of labour. But it is not unreasonable to suppose that the activity of labour today is shaped by similar forces, generating comparable worries about the quality of work generally available and opportunities for self-realization or fulfilment through work, as well as normative conflicts about the management of work activity. The prospect of fully automated production, which in 1958 Arendt predicted would empty the factories ‘in a few decades’ (HC: p. 4), seems to many observers closer than ever, and arguably lends further contemporaneity to Arendt’s critique of humanisms of labour. It is not unreasonable, then, to ask ourselves what is at stake in the debate between the humanism and anti-humanism of labour. Let me follow the lead of Friedmann and Arendt by distinguishing three sets of issues: (1) the range of human capacities that are apt for expression in labour or work activity; (2) the social relations that are apt for regulating that activity; and (3) the sense of individual fulfilment (‘happiness’) that can reasonably be expected from participation in the activity.

First, a contemporary humanist of labour in the mould of Friedmann would grant that advances in the technical division of labour would doom a great many workers to routine, unskilled, unrewarding labour. There is no fundamental difference in the work involved in the manufacture of commodities and the provision of services in this respect. Technical skills are required for both, but for the majority of workers, the range of them diminishes as the division of labour increases. In the long run, the hope is that machines will be able to perform all the routine, repetitive, unrewarding tasks that now have to be done by humans. But in the meantime, efforts can and should be made to ‘humanize’ the activity that workers have to perform. This can be done by designing work tasks in a way that makes them more challenging and engaging, such that the worker has to draw on a wider range of technical capacities to perform them.

The crucial point here is that, for the humanist of labour, any given stage in the technical division of labour is consistent with more or less humane job design (Murphy, 1993). The technical division of labour does not determine the activities to be performed in any given job, nor does it determine the allocation of those activities to particular workers. These are matters of moral and political choice. The humanist of labour emphasizes the moral and political nature of the decisions at stake here and makes reform in this dimension a moral and political priority. The anti-humanist of labour, by contrast, takes the scope available for moral and political shaping of the activity of labour to be negligible. Placed as the worker must be in the logical space of production, the worker is structurally subordinate to the demands of technique and must sacrifice his or her subjectivity (his or her unique capacities) to those demands. The only political choices that matter in this regard are those that relate to presence or absence in the technical division of labour. Political decisions can be made to hinder the automation of production, to preserve the human presence in it; or to accelerate automation, to lessen the amount of time that human beings have to spend in this domain, and ideally to remove them from places of production altogether. Reform of the activities that workers are required to do within a given stage of the technical division of labour is no substitute for advances in the technical division of labour that might remove or lessen the need for human activity within it. From the perspective of the anti-humanism of labour, the humanism of labour is unduly ‘superstructural’ in its critique of the division of labour and insufficiently attentive to the suppression of human capacity that the activity of labour in principle involves.

Anti-humanists of labour can take this argument further. They can point out that, from its beginnings as a reaction to the shortcomings of Taylorist hyper-rationalization and micro-management, humanism of labour has been pulling in two competing directions. On the one hand, there is the humanitarian impulse to eliminate degrading conditions of labour; to correct the reduction of the working human being to a mere cog in the machine of the productive process. On the other hand, humanism of labour is motivated by inefficiencies in the deployment of the human resource in previous organizations of production – Taylorism, in particular. Anti-humanists of labour suspect that the latter motive is the driving or dominant one when it comes to the reform of labour. For practical purposes, ‘human relations’ is a management construct designating a mere factor in production; it exists in the same logical space as the technical division of labour, not as a limiting point or counter-concept to it. Humanism of labour, according to its critics, assumes practical significance by masking this fact. The techniques of ‘human relations’ management are invariably aimed at absorbing the subject of labour into the production process. The more human the process can be made to look, the more readily the subject can come to identify with it − become successfully ‘integrated’, as Friedmann says. The humanitarian impulse thus not only loses out to the drive for efficiency but becomes the latter’s accomplice. ‘Human relations’ in this context are properly viewed as just one technique of production among others, a technique whose effectiveness relies on it seeming to be more than it is, namely, a set of genuinely social relations.

Second, something similar can be said of contemporary discourse around ‘recognition’ and ‘respect’ at work. Much of what is sought nowadays by way of the reform of work falls under one of these banners. ⁹ Demands for respect at work include, minimally, demands not to be subjected to bullying or harassment; to discrimination on the basis of sex, race, class, and disability, nor be subjected to excessive surveillance, and so on. More substantively, they include demands on the part of workers for a more meaningful voice in the management of the work they do, for more autonomy (less external control) in regard to their work activity, and so on. Demands for recognition overlap with these, but often focus on due acknowledgement of the contribution workers actually make to the work organization through their work activity. The worth of this contribution is often not recognized for what it is, but is distorted by stereotypes, especially around gender and race. Complaints about breakdowns of trust and a decline in levels of civility at work are common and have become mobilizing issues for reform. All of these issues concern the quality of the social relations that more or less explicitly regulate the activity of working. Contemporary heirs to Friedmann’s humanism of labour believe that meaningful reform along these lines is both possible and desirable, indeed, should be considered a moral and political priority. Contemporary heirs of Arendt’s anti-humanism of labour are more sceptical. They are inclined to see the whole discourse of recognition as a ruse for legitimating the subordination of the individual within the work collective or corporation. It is either a tool of oppression, diverting workers from the true interest in emancipation from labour altogether, or a purveyor of dangerous illusions about the possibility of authentic identity, either through work or some other means. The contemporary Arendtian anti-humanist of labour might also be wary of demands for more autonomy in work, at least insofar as they hark back to old discredited ideas of a self-producing sovereign subject.

Third, underneath these more abstract theoretical misgivings, anti-humanism of labour is unconvinced – indeed, repelled − by the value that the humanism of labour attaches to modern work. For the humanist of labour, work even in the context of contemporary methods of production can be, and often is, meaningful, which is to say a source of happiness and fulfilment (Veltman, 2016). The meaning may come from the activity itself, or it may come indirectly, through the recognition one receives for it. Opportunities for meaningful work, and access to the goods provided by work, are, however, very unevenly and unequally distributed, and this makes work a pressing sphere of injustice (Gomberg, 2007). But for the anti-humanist of labour, the value of labour has been greatly exaggerated: the activity, for the most part, is of no lasting consequence, it leaves little of real value behind, it lacks ultimate purpose, and ‘recognition’ is paltry compensation for the ‘toil and trouble’ it brings. It would be better to be spared the curse of labour in the first place. This is the promise of automation. Although, like Arendt, one might have misgivings about the fate of the vita activa in the wake of fully automated production, the contemporary anti-humanist of labour, less attached to any notion of the active life, may not be so disturbed by such qualms.

The previous paragraphs offer just the briefest sketch of the terms of debate between a contemporary humanism of labour oriented by Friedmann’s understanding of labour, and an anti-humanism of labour oriented by Arendt’s understanding. The point was not to take sides but to indicate the relevance of the rival conceptions of labour presented by Arendt and Friedmann to current philosophical and political discussions about the reform of work as we enter a new phase of automated production. For all the historical distance that separates Arendt, Friedmann and ourselves, we are linked by a shared anxiety about the fate of labour, a fate which even by the 1950s was considered by most intellectuals to be more in the hands of technicians and robotic machines than a soon to be awakened revolutionary proletariat. While my main aim in this article has been to show that, among other things, The Human Condition served historically to address that anxiety, and to address it in a way that Arendt considered to be superior to the alternatives proposed by her contemporaries, among whom the humanists of labour were prominent, the choice between a humanism and anti-humanism of labour by no means belongs to the past. It remains with us, unresolved.