User-generated content, free labour and the cultural industries

Critiques of creative labour in the digital era

Cultural industries and cultural production have been the subjects of thousands of studies. But until recently, only a very small proportion of these studies focused on the creative labour that is fundamental to this realm of production. The forgetting or devaluation of work in analyses of cultural industries has taken a number of different forms. One, apparent in some arts and humanities studies, is a focus on individual producers rather than on the complex division of labour, which, as the sociology of culture has shown (in valuable accounts such as Becker, 1982; Wolff, 1993), is the basis of most cultural production. Another is an emphasis within certain schools or traditions, notably some types of cultural and media studies, on consumption, at the expense of production.

More recently, though, the devaluation of work in communication and cultural research has taken a new shape. Digitalisation has led to a proliferation of new forms of amateur and semi-professional production; blogs, Wikipedia, citizen journalism and various forms of interactive games were the most cited examples during the 2000s. Many commentators have heralded this supposed explosion of non-professional cultural production as evidence of a new era of cultural production, and as a democratisation. Axel Bruns (2008: 13-14), for example, claims that production and consumption are old-fashioned ‘industrial age’ concepts, and that in the internet age, where access to the means of producing and distributing information is ‘widely available’, consumers can become cultural producers and distributors, bypassing ‘traditional’ organisations via peer-to-peer and ‘many to many’ (rather than ‘one to many’) communication systems, leading to a new form or model known as ‘produsage’, a mixture of production and use.[1]

Phenomena such as Wikipedia and open source software are, without doubt, fascinating examples of cultural activity that attempt to base themselves on the pleasures and rewards of co-operation rather than competition. It is certainly the case that the cultural industries in the digital era, like many other kinds of firm, increasingly seek to draw upon the participation of their users and consumers.[2] But too many of these discussions of transformations associated with new digital media rely on caricatured portrayals of supposedly bypassed eras. The ‘industrial age’ for example involved much more complex relations between production and consumption than is implied by Bruns’s account. What’s more, many of these discussions are discomfortingly reliant on business and journalistic commentary regarding the impact of digital technologies on economies and societies in the twenty first century. The list of neologisms and buzz phrases goes on and on: ‘the new economy’, ‘the digital economy’, ‘wikinomics’, ‘crowdsourcing’, ‘collective intelligence’, ‘the long tail’, ‘the wisdom of crowds’, ‘smart mobs’, etc. If I read one more time about how Time magazine nominated ‘you’ as person of the year in 2006, and how this marked the beginning of a new era of user-generated, content, I think I’ll post a video on YouTube. It will be of me holding my head in my hands and screaming. Clichés and received thinking seem to dominate this area of debate.

Such accounts of transformation have not been without their critics. A dominant theme of recent critical analysis of digital media is that they involve unpaid work on the part of participants. In a seminal essay, Tiziana Terranova wrote about the phenomenon of ‘free labour’, which she described as ‘an important, yet unacknowledged, source of value in advanced capitalist societies’ (Terranova, 2004: 73).[3] Free labour was, wrote Terranova ‘simultaneously voluntarily given and unwanted, enjoyed and exploited’ and on the internet included ‘building web sites, modifying software packages, reading and participating in mailing lists and building virtual spaces’ (2004: 74). Others have applied similar perspectives to other forms such as television and games. Mark Andrejevic, reacting against celebratory accounts of ‘active audiences’ in media studies, has written powerfully about ‘the ways in which creative activity and exploitation coexist and interpenetrate one another within the context of the emerging online economy’ (2008: 25). In earlier work, Andrejevic had discussed how ‘reality TV anticipates the exploitation of…the work of being watched, a form of production wherein consumers are invited to sell access to their personal lives in a way not dissimilar to that in which they sell their labor power’ (Andrejevic, 2004). Andrejevic (2008) went on to explore how online viewer activity serves television producers in two ways, by providing feedback, which saves the producers from having to undertake expensive market research, and by, in effect, publicising television programmes, which saves marketing costs. Andrejevic critiques the equation of participation and activity with real democratisation and shared control, and claims that regimes of surveillance and imperatives of profit-making hugely compromise the pleasures and progressive elements of online participation. In the world of games, Greig De Peuter and Nick Dyer-Witheford (2005) have explained how, from the 1990s onwards, ‘authoring tools’ have been increasingly packaged with computer games, helping to foster a vibrant participatory culture of game ‘modding’, or modification. They argue that the work of such modders is a kind of free labour, a ‘space-defying’ process of exploitation of ‘collective intelligence’, which also serves as a kind of informal training for the future game development workforce.

A further way in which analysts have pointed to the contribution of free labour to the cultural industries is the way in which the latter draw on pools of talent outside their boundaries, and employ professionals to manage crossings of those boundaries, and to negotiate the creativity-commerce dialectics that are at the heart of modern cultural production. All this is familiar from political economy and sociological analysis of cultural industries analysis (see for example Miège, 1989) but recent critics have linked this to the free-labour debates by emphasising the way that creative professionals draw on the unpaid ‘mass intellectuality’ (see below) of alternative and underground scenes. One writer (Arvidsson, 2008), in an effort to counter the celebratory discourses associated with modern creative industries policy, even uses the rather dubious phrase ‘creative proletariat’ to refer to the underground cultural producers that some creative professionals draw upon as part of their work. Elsewhere, Arvidsson (2005) has extended discussion of free labour into debates about consumption, claiming that brand management exploits the ‘immaterial labour’ of consumers by drawing on resources of ideas and styles generated in contemporary urban environments. Clearly, this is a more critical account of ‘user-generated content’ than is to be found in academic management studies and journalistic hype.[4]

But the critique of free labour is not the only way in which critical analysis has sought to pay attention to creative labour in recent years. Alongside these developments, there have also been discussions of professional and semi-professional workers in the cultural industries and in related industries such as web design. Various recent critical accounts have suggested that professional workers in the (digitalising) cultural industries and in related industries are involved in forms of labour that are characterised by high degrees of autonomy, creativity and ‘play’, but also by overwork, casualisation and precariousness. Andrew Ross (2003: 9) observed how, in the eyes of a new generation of business analysts in the 1980s, Silicon Valley ‘appeared to promote a humane workplace not as a grudging concession to demoralized employees but as a valued asset to production’. ‘New economy’ firms, he argued, aimed to provide work cultures that ‘embraced openness, cooperation and self-management’ (ibid). Ross showed that such features were in fact closely linked to long working hours and a serious blurring of the line between work and leisure. Whilst the dot.com working environments of the 1990s offered ‘oodles of autonomy along with warm collegiality’ they also enlisted ‘employees’ free-est thoughts and impulses in the service of salaried time’ (Ross, 2003: 17, 19). This process of involvement has been described as ‘self-exploitation’. Similarly, also writing about new media work, Andreas Wittel (2001) saw there a paradigmatic case of an emergent form of community that he calls ‘network sociality’, one which appears to be individualistic and instrumental, involving an assimilation of work and play. Ros Gill (2002), in a study of European freelance new media workers, found evidence that features of the work that seemed superficially attractive, such as its informality and high levels of autonomy, were in fact particularly problematic for women because of the lack of clear criteria for evaluating work and especially because of the difficulties such informality caused when seeking new contracts.

Such insights, often developed in relation to the IT sector, have been increasingly applied to the cultural and creative industries. Gillian Ursell’s (2000) early contribution noted ‘an intensification of the self-commodification processes by which each individual seeks to improve his/her chances of attracting gainful employment’ (Ursell, 2000: 807) and analysed how television workers had, in the era of casualisation and increasing freelance work, come to take on the work of organizing their own labour markets. This element of ‘apparent voluntarism’ needed to be acknowledged, Ursell claimed, and she turned to Foucauldian theory not to dispense with labour process theory concerns but ‘to approach them more substantially’ (Ursell, 2000: 809). In particular, she drew on Nikolas Rose’s (1999: 145) idea that, in advanced liberalism, freedom is redefined as ‘a capacity for self-realisation which can be obtained only through individual activity’. Discussing how notions of creativity, talent and work are being redefined in those burgeoning micro-businesses of the cultural sector associated with young people, including fashion and design, but also entertainment industries such as clubbing, recording and magazine journalism, Angela McRobbie (2002a: 523) echoed Ursell in pointing to the ‘utopian thread’ involved in the ‘attempt to make-over the world of work into something closer to a life of enthusiasm and enjoyment’, and in focusing on how this leads to a situation where, when things go wrong, young people entering these creative worlds of work can feel they only have themselves to blame. In this respect, McRobbie usefully broadened the study of creative work to include a wider set of conditions and experiences, including the way in which aspirations to and expectations of autonomy could lead to disappointment, disillusion and ‘self-blaming’. She also pointed to the gendered aspects of these conditions, with women now expected to find full-time work, uninterrupted by family commitments, satisfying and enriching (McRobbie, 2002a: 521). The context for McRobbie’s critique (see also McRobbie, 2002b) was the then UK Labour government’s creative industries policy, and their general valorisation of labour, where ‘work comes to mean much more than just earning a living; it incorporates and takes over everyday life’ (McRobbie, 2002b: 99). McRobbie was usefully questioning the ‘ideal of self-expressive work’ (2002a: 101) and its place in Labour’s advocacy of ‘a new youth-driven meritocracy’, involving a labour of love and self-exploitation. Later writers have built on these contributions (a helpful and important synthesis is provided by Banks, 2007). In the article discussed above, De Peuter and Dyer-Witheford (2005) showed how the creativity and playfulness highly valued by the games workers that they interviewed served to offset extremely long and demanding working hours. In a series of articles and chapters Matt Stahl has brought legal, political and cultural theory together to suggest that the incorporation of subjectivity into capitalism acts as a kind of pacifying device in the era of neo-liberalism and that popular music’s democratic promise that ‘you can do this too’ is a particularly salient way in which ‘liberal society’ promises an end to alienation and appropriation by promising independence and autonomy (Stahl, 2006: 23).

Some of the key concepts raised in these recent debates about paid employment in the IT and cultural industries (such as self-exploitation and the hidden costs of autonomy) are germane to discussions of free labour in the digitalising cultural industries. Yet the free labour debates have paid only rather passing attention to questions concerning employment, occupations and careers in these industries, other than to make passing reference to moments of resistance to oppressive working conditions.

I return to questions concerning quality of working life in the cultural industries towards the end of this article. I argue there that these questions are important political ones, even if celebrants and critics of digitalisation see professional cultural work as in decline. But first I want to address a number of issues regarding the concept of free labour itself. The perspectives outlined above have provided some stimulating and necessary interventions against complacent celebrations of cultural-industry work, and of the relations between production and consumption in the digital era. Understandably, though, given their innovative character, and the fact that they have been responding to relatively new social and cultural transformations that are difficult to comprehend as they unfold, some important conceptual issues remain underexplored. For example: are we really meant to see people who sit at their computers modifying code or typing out responses to TV shows as ‘exploited’ in the same way as those who endure appalling conditions and pay in Indonesian sweatshops? Clearly not – and this raises the question of how to integrate such analysis into more satisfactory understandings of capitalism, exploitation, power and freedom. To what kinds of political demands might objections to free labour give rise? I address these issues below, but first, the political origins of the debates about free labour need to be clarified.

Free labour and autonomist Marxism

Much of the research and commentary indicated above draws on a set of theories and concepts that will be familiar to regular readers of this journal. Autonomist concepts of immaterial labour, affective labour and ‘precarity’ have been of increasing interest to critical commentators on contemporary work, including labour in the cultural and creative industries.[5] The concept of immaterial labour has its origins in a series of papers in the journal Futur Antérieur by Michael Hardt, Maurizio Lazzarato, Antonio Negri and Paolo Virno in the early 1990s. The concept was there defined as ‘the labour that produces the information and cultural content of the commodity’ (Lazzarato, 1996: 133). It was to come under serious attack from other autonomist Marxists (notably Caffentzis, 1998) for its excessive optimism, lack of attention to gender, and failure to recognise the continuing significance of highly material forms of exploitation and oppression. Consequently, by the time of their widely-read book Empire, Hardt and Negri (2000: 290) had developed a more expansive definition: ‘labor that produces an immaterial good, such as a service, a cultural product, knowledge, or communication’. This now incorporated ‘affective labour’, involving human contact and interaction, and including the kind of heavily gendered caring and health work to which the critiques of Caffentzis and others had drawn attention. In autonomist discussions of both immaterial and affective labour, there was a focus on labour and the production of culture, knowledge and communication – and a reaching for a critical conception of the place of such labour in modern societies.

Hardt and Negri begin from a discussion of how the introduction of the computer has radically transformed work. Even where direct contact with computers is not involved, they say, the manipulation of symbols and information ‘along the model of computer operation’ (2000: 291) is extremely widespread. Workers used to act like machines, now they increasingly think like computers. They modify their operations through use, and this continual interactivity characterises a wide range of contemporary production. The computer and communication revolution of production has supposedly transformed labouring practices in such a way that they all tend toward the model of information and communication technologies, and this means a homogenisation of labouring processes. In this respect, Hardt and Negri are pessimistic about the ‘informationalization’ of the economy. But they are much more optimistic about ‘affective labour’, which, they claim, produces social networks and communities: and, for Hardt and Negri, cooperation is immanent to such labouring activity. This networked cooperative aspect of affective labour is then transferred by Hardt and Negri to other more computer-driven and ‘symbolic-analytical’ forms of immaterial labour, as if nurses and computer programmers were doing the same kind of work. A further leap is to see all such workers as therefore equally imbued with the same capacity to struggle against capital. Because wealth creation takes place through such co-operative interactivity, Hardt and Negri believe, ‘immaterial labor thus seems to provide the potential for a kind of spontaneous and elementary communism’ (2000: 294).

In her highly influential piece on free labour, Terranova (2004) borrows this optimistic emphasis on the collective nature of knowledge labour and on the continuing, indeed thriving, existence of a ‘general intellect’ constituted through the continuing vitality of human labour and subjectivity. Capital constantly struggles to valorize (make profit out of) this intractable terrain (2004: 88). The internet itself can be seen as a sign of that vitality. ‘Free labour’, then, has something of a double meaning. It refers to unpaid work, but, in line with Terranova’s explicitly autonomist sympathies, it also refers to the way in which labour cannot be fully controlled, because of capital’s continuing and problematic reliance on it. Terranova is scathing about the ‘popular, left-wing wisdom about the incorporation of authentic cultural moments’ (2004: 79). Rather than incorporation, for Terranova phenomena such as open-source software represent ‘the overreliance of the digital economy’ (2004: 93) on free labour.

Is free digital creative labour really exploitation?

I now want to consider some of the issues raised by Terranova’s characterisation of free labour as ‘[s]imultaneously voluntarily given and unwanted, enjoyed and exploited’ (2004: 74), in particular how we might understand relations between exploitation and satisfaction, alienation and freedom, in the present digital context.

The question of exploitation has been raised by Arvidsson (2005) in his piece on free labour in relation to brands, discussed earlier. Brands are ‘valorized’, writes Arvidsson, through their ability to extract a premium price and through the way that higher brand values generate higher share prices. All this depends on brand awareness, associations and loyalties, all of which in turn depend on the attention of consumers. Surplus value – the Marxian term for the ability to generate greater amounts of money from capital investment – is based, in Arvidsson’s gloss, on ‘the ethical surplus, or the surplus community that consumers produce’ (2005: 250) and upon which businesses then draw. This is not just a matter of the time spent on such activities, Arvidsson claims.

Brand management is also a matter of managing the quality of the common [sic] produced through communicative interaction. The qualitative dimension of exploitation thus consists in making the productive sociality of consumers evolve on the premises of brands; to make it unfold through branded consumer goods in such ways that makes it produce measurable (and hence valuable) forms of attention. (2005: 251)

But it is not clear that this really corresponds to exploitation in any meaningful analytical sense of the term. Rather, the problem, as Arvidsson expresses it, seems to be here one of controlling or reshaping the ways in which people communicate, pushing them in the direction of brands. That is indeed a problem, as a number of critics have pointed out, notably Naomi Klein (2002). But this is not the same thing as exploitation. It is important that such a loaded word is used with at least a certain amount of precision. Exploitation can be used as a useful term to express our repulsion when someone makes use of someone else for their own purposes, but when used in research we presumably intend it in a more precise analytical sense. And analytical use of the concept of exploitation has been overwhelmingly Marxian: it is about the historical relationship between classes, again in the Marxian sense. Erik Olin Wright (1997) has argued that exploitation in its Marxian sense is based on three principles. First, exploitation occurs when the material welfare of one class is causally dependent upon the material deprivation of another. The capitalist class in modern societies could not exist without the deprivations of the working classes. Second, that causal dependence depends in turn on the exclusion of workers from key productive resources, especially property. Third, the mechanism through which both these features (causal dependence and exclusion) operate is appropriation of the labour of the exploited. The first two alone would just represent oppression; for exploitation (in the Marxian sense) to take place, the third condition must be present. Equally, appropriation is not the same thing as exploitation; the first two features, causal dependence and exclusion, must also be present as well as appropriation.

As well as being a historical concept, exploitation is also an explanatory one. It was intended to explain how capitalism was able to generate such massive surplus values and at the same time such immiseration. This explanation, it should be noted, rests on a complex conception of compulsion. All human life involves being compelled to do things. Implicitly, therefore, the explanatory power of the concept of exploitation rests on an ethical distinction between legitimate and illegitimate forms of compulsion. As Alex Callinicos puts it:

What is distinctive about Marx’s account of capitalist exploitation is that the appearance of free exchange between worker and employer is nullified by the unequal distribution of the productive forces: as a result, workers are compelled to sell their labour-power to the capitalists on terms that lead to their exploitation. This is a violation of their liberty, even if they are not directly coerced into performing surplus-labour for the capitalist, but rather do so as a result of what Marx called “the silent compulsion of economic relations”. Thus exploitation is directly unjust, independently of any injustice in the initial distribution of productive assets, because workers are illegitimately compelled to work for the capitalist. (Callinicos 2000: 68)

So exploitation in its Marxian sense, Wright and Callinicos help to make clear, is a historical, explanatory and ethical concept that revolves around certain (disputed) notions of class, labour and compulsion. This means that when the term is applied to specific empirical examples, some kind of link needs to be made to these necessary abstractions if it is to have analytical purchase. This does not mean that every single invocation of the concept of exploitation by any writer should involve such theorisation. The problem here though is that, as the quotations in the introduction to the concept of free labour in the first section of this article illustrate, the term ‘exploitation’ has been widely but uncertainly used in these debates.

In a recent piece on YouTube, Andrejevic (2009) has returned to the issues of free labour discussed in the earlier pieces quoted above and has offered a more developed consideration of the concept of exploitation. Drawing on the autonomists, Andrejevic sees the term ‘free labour’ as meaning unpaid work, but also freely given work, ‘endowed with a sense of autonomy’ (2009: 416). He interprets this as suggesting a logic whereby the production of community and sociality is ‘both autonomous of capital and captured by it’. For Andrejevic, this question of capture signals the crucial importance of exploitation. Turning to the work of Adam Arvidsson, Andrejevic wonders (as I have, above) whether the capture or appropriation of free, affective and immaterial labour, as outlined by Arvidsson, can really be described as exploitation in any convincing analytical sense. Andrejevic rightly points out that, for Marxian analysis, the generation and appropriation of surplus value depends on a kind of force:

the forcible separation of the worker from the means of production is conserved in workers’ forced choice to relinquish control over their labor power. But the potential located in affective or immaterial labor by Arvidsson, Hardt and others lies in the very fact that it is freely or autonomously given. It is by definition not forced. Nor is it clear that this labor is appropriated under the threat of force, which renders the claim of exploitation in need of further explanation. (2010: 418)

Because of this, Andrejevic seeks an alternative understanding of exploitation in relation to digital labour. To do so, he emphasises how the concept is bound up with the related concept of the alienation of workers from the products of their labour. Web 2.0 style technologies, says Andrejevic, gain their popularity by offering users an escape from alienation (there are potential links here with Stahl’s work, discussed above) by offering ‘modicum of control over the product of their creative activity in return for the work they do in building up online community and sociality upon privately controlled network infrastructures’ (2009: 419) and allowing themselves to be monitored. For Andrejevic, there is an important distinction to be made between ‘user-created content’ and ‘user-generated data’ (2009: 418). It is the latter not the former that is extracted under conditions of private ownership and that is turned into a commodity. All this suggests to Andrejevic a generalisation of the forms of subjection traditionally associated with women. Time spent building social relations in affective labour is both autonomous and subject to exploitation, he writes; so is the kind of immaterial labour involved in social networking sites such as YouTube.

Rather than exploitation, however, Andrejevic’s analysis seems actually to be dependent on questions of freedom and ideology. In his various contributions (such as Andrejevic, 2004: 201), he shows how prevailing ideas about digital interactivity do not so much conceal more general relations of exploitation but rather point to their inevitability, celebrating the savviness of audiences, but offering no means by which this savviness can be converted to forms of action which might meaningfully reduce inequality. This seems to me to be insightful. But his conception of exploitation is rather less successful. Andrejevic here tries to rescue the concept of exploitation in relation to ‘free’ digital labour by linking it to force indirectly, via the concept of alienation. The oppressive system of alienation creates such misery that it compels people to seek out ways of re-exerting their control in ways which then become open to appropriation of surplus value by capitalists. This is a thoughtful and stimulating idea, but the mechanisms of this indirect force – for example the variable ways in which people respond to alienation by seeking out cultural production – are not really spelt out. What’s more, this conceptualisation risks reducing the drive to communicate and to produce culture and knowledge to a reaction to alienation. And there is evidence that capitalism might have moved in the direction of attempting to reduce alienation in the interests of accumulation (see Boltanski and Chiapello, 2005).

The concept of ‘free labour’ is linked to some interesting ideas about power and control in cultural production in the digital era. But the frequent pairing of the term with the concept of exploitation is unconvincing and rather incoherent, at least as so far developed by the most-cited analysts. I hope this does not come across as pedantry. I point all this out in a comradely attempt to encourage greater precision in critical thought. In the same spirit, I now turn to questions concerning what political demands might flow from critiques of free labour.

Is unpaid labour always a problem?

Even if they might want to retreat from the view that free labour involves exploitation in any meaningful Marxian sense, leftists might still want to hold on to the idea that it is wrong in some way. In its broadest sense, labour is simply exertion of the body or mind, and it is usually used to describe activities that have some sort of compulsion attached to them. Obviously we cannot define labour in terms of whether it is paid or unpaid, or whether an employment relation is involved, because it is clear that a great deal of the labour that goes into sustaining and enhancing life in modern societies is unpaid. But under what conditions might we object to such unpaid labour, and on what grounds?

Domestic labour, often performed primarily by women, is the most discussed version of such unpaid labour (and those who write about affective labour from an autonomist perspective sometimes have a tendency to sound as though this insight was an invention of the autonomist Marxists, sidelining the many contributions that have come from other perspectives – see Himmelweit, 1991 for one summary). But the fact that these debates concerned a form of unpaid labour should not make us think that the fact that labour was unpaid was always the principal point under debate. The important point that feminists were making, in drawing attention to the economic and social contribution of unpaid domestic and childrearing labour, concerned the many injustices associated with the gendered division of labour, including the expectation that one particular group of people (women) were, more than any other group, and by virtue only of their biological and cultural differences from men, expected to perform such onerous duties without financial recompense. Closely connected to this was a set of disadvantages in paid labour markets, including exclusion from certain high-prestige sectors. The ethical problems here were ones of inequality and injustice, and the political ones concerned, for example, whether a demand for wages would really serve to address these problems or whether the deeper question was the institutional separation of (for example) childrearing from paid work. Even if we agree with the wages for housework movement, this does not mean that we can or should apply similar demands to other forms of unpaid labour. Clearly, life will always involve a huge amount of labour, often unpleasant, some of it answering more urgent needs than others, and in some cases we might find it acceptable, or a matter of lesser priority, that certain forms of work are unpaid and others paid. Until the work/leisure distinction and the social division of labour are abolished altogether (the extremely distant utopian aspiration embodied in Marx’s famous hunter/fisher/critic passage in the Economic and Philosophical Manuscripts) we cannot expect to be paid for all the many different tasks that we might have to perform in order to reproduce and maintain life. Even if we achieve more just social relations, future societies will continue to be based on a complex division of labour, and it seems highly likely that some spheres will come within the realms of paid labour and some outside it.

The important question here though is whether socialist-feminists (in the broadest sense of that term, to include anyone who thinks that economic and gender relations are currently unjust and unequal, and need urgently to be made more just and more equal) might object to unpaid work in the cultural industries on anything like the same grounds as the objection to unpaid household labour – i.e. that it contributes significantly to broader patterns of inequality and injustice. If that can be shown, then there might be a strategic case for demanding payment, to redistribute income, and/or in order to highlight the ethical problems concerned. To consider that question, though, we need to be clear about which forms of unpaid labour are being discussed. There has been a tendency to bandy about the phrase ‘free labour’ as if it describes one huge, interconnected aspect of inequality and injustice. Instead, unpaid work may not be a problem in itself, and may in fact be an inevitability, even in a better future society. The fact that work is being performed for free in itself is not a sufficient objection.

In the history of cultural production, only a very few people within any society have taken on the role of cultural producers in return for financial reward. A major development was the patronage relationships of feudalism and early capitalism, which gave way to payment by royalties and wages with the development of cultural capitalism and the copyright system. Most cultural production in history has been unpaid, and that continues to be the case today. Consider the millions of people across the world, especially young people, who will, on the day you are reading this, be practising musical instruments, or, to use an example from an industry that I would call a leisure industry rather than a cultural industry, imagine how many young people are practising football or basketball. Now it could be argued that all this represents labour (defined here as the expenditure of effort, under some kind of compulsion; it will usually seem preferable to undertake some other more restful activity) which is vital to the realisation of surplus value in the music industry or the football industry. For this work helps to create a reservoir of workers, from whom these industries can draw. Regular practice by future musicians and sportspeople ensures that there are greater levels of talent available for businesses to employ. The football example is perhaps even more germane, because while a great deal of music teaching is done on a professional basis, most football coaching is done by amateurs, who give of their time in reward for a range of pleasures and rewards including winning competitions, inculcating the joys of team-based physical activity, and being able to shout very loudly at young people. But even if this is true, what political demands might ensue from this? Wages for music practice? Wages for sports coaches?[6] Of course some on the left believe in the importance of making ‘maximum’ political demands as a way of furthering emancipation, but there also needs to be some sense of prioritisation, and at least some kind of pragmatic reading of what might be possible. An undifferentiated critique of free labour can generate demands that fail these basic tests.

Now the advocates of a critical (autonomist) position towards free labour may validly respond that free labour only becomes an issue in spheres of activity where there has been extensive commodification, and that the vast social reach of certain digital technologies makes it important to highlight the labour that they depend upon. The development of the internet might be an example of this, or more specific sites such as YouTube. Even here, however, there are problems that we might want to consider, and which do not seem to have been raised in the debates about free labour. Terranova’s seminal account usefully pointed to the huge amount of unpaid work necessary to create the internet. But it may be said in response that those who undertook such unpaid digital labour might have gained a set of rewards from such work, such as the satisfaction of contributing to a project which they believed would enhance communication between people and ultimately the common good; or in the form of finding solutions to problems and gaining new skills which they could apply later in other contexts. In some cases, it might be possible to think of their work as involving the building of skills which lead to higher wages being paid in the longer term – a kind of deferred wage. Without denying for a moment the fundamental importance of a living wage, it seems dangerous to think of wages as the only meaningful form of reward, and it would surely be wrong to imply that any work done on the basis of social contribution or deferred reward represents the activities of people duped by capitalism. Actually, it seems to me that this would run the danger of internalising capitalism’s own emphasis on commodification. We have to hold on to the value of work done for its own sake, or as ‘gift’ labour (see Hyde, 1983), and complaints about free labour – unless the normative basis for the complaints are spelt out very carefully – risk undermining that value.[7] It may be that open source software is linked to corporate forms of capitalism, as Terranova (2004) and others have pointed out. But the idea of carrying out software development for free may in many cases lead to the development of products, which are not quite so much under the control of major corporations as others. At the very least, complaints about free labour need to be linked to discussion of which kinds of free labour merit payment and which do not. We might for example want to argue that the expertise that goes into Wikipedia might be subsidised by governments and corporations out of general taxation given that it provides a huge social resource, often drawn upon by businesses and governmental institutions; until recently, this was the justification for funding Higher Education out of taxation in Western Europe, though this idea has been eroded in the era of neo-liberalism. On the other hand, the complaint I referred to earlier, that contacting friends and uploading photographs on to Facebook represents some kind of exploited labour is, to my mind, more along the lines of arguing that we should demand that all amateur football coaches be paid for their donation of free time: not impossible to argue for, but hardly a priority – and accompanied by the danger that it may commodify forms of activity that we would ultimately prefer to leave outside the market. The dangers of commodification might be better countered by arguing against developments, which seek to exert ownership by sites of content generated by users. It could be that it is in the realms of intellectual property that a more convincing critique of contemporary capitalism might be mounted, rather than unpaid labour.

A striking case of unjust, unpaid labour in the media industries is the internship system. It is increasingly difficult to enter the media and media-related industries in advanced industrial countries without having performed, at some point, a significant period of unpaid work. The fact that young people are willing to do this is a product of the desirability of creative labour, and the over-supply of workers. The extraction of billions of hours of unpaid labour by media companies can be seen as a kind of rent (in the technical economic sense). This is offset by the considerable time needed to train and mentor a constant influx of young, inexperienced workers, but this does not come close to matching the financial advantages gained by media companies. The use of such young people performing unpaid labour also depresses wages for workers in the cultural industries. Furthermore, it has a serious impact on which kinds of people are likely to be able to gain entry to the media industries. Young people from wealthy families are much more likely to be able to afford sustained periods without pay. Increasingly, internships are provided as part of media education degrees. Of course many young people want to carry out such internships. But they benefit companies at the expense of time that young people might be spending exploring ideas and broadening their intellectual horizons – benefits that it might be difficult for them to understand, compared with the potential excitement of working in a media company, but which in my view must be seen as likely to provide benefits for societies. Defenders of labour might, then, argue for such internships to be made illegal, or for them to be licensed: media companies might have to pay a fee into a common fund which is then distributed to young people as payment, or is simply redistributed through taxation, earmarked for education.

Free labour and the audience commodity debate

Discussions of labour in the cultural industries have visited the terrain of unpaid labour before, in a debate on ‘the audience commodity’ that received significant attention in the early years of the political economy of communication, and which has been revived in recent years. I want to revisit this territory briefly here, in order to show the dangers of an approach to questions of labour that is insufficiently informed by ethical thinking and by attention to the specifics of particular forms of work and leisure experience – dangers that may afflict the attempt to build a critical perspective on creative labour based on the free labour concept.

Dallas Smythe, followed by other writers (such as Jhally and Livant, 1986), argued that, in paying for the advertising which sustains a great deal of modern cultural production, advertisers were buying ‘the services of audiences with predictable specifications who will pay attention in predictable numbers and at particular times to particular means of communication…. As collectivities these audiences are commodities’ (Smythe, 1977: 5). Goran Bolin (2010), drawing on Meehan (2000), has helpfully clarified some of the confusions surrounding this debate.[8] Rather than seeing audiences as working for media industries, Bolin suggests, it is more fruitful to see statistical representations of audiences as raw material that is shaped into a commodity by market research agencies and departments and sold as a commodity: ‘It is not the viewers who work, but the rather the statisticians’ (Bolin, 2010: 357).[9] But I think there is an even greater problem with the view that would see individual audience members as undertaking unpaid work when they watch television programmes. In Smythe’s formulation, the objection to this rests on the idea that ‘all non-sleeping time under capitalism is work time’ (1977: 6). The time that workers spend off the job, says Smythe, involves ‘coping while constantly on the verge of being overwhelmed’ by the pressures created by their immersion in consumer desires created by monopoly capitalism (1977: 14). Now I would not deny that the freedoms of ‘free time’ are constrained by social forces in problematic ways (see Hesmondhalgh, 2008 for my take on these questions with regard to music consumption). But Smythe’s account is crude, reductionist and functionalist, totally underestimating contradiction and struggle in capitalism. The underlying but underdeveloped normative position is that all the time we spend under capitalism contributes to a vast negative machine called capitalism; nothing escapes this system. No work or leisure seems, by this account, to be any more meaningful than any other. It is unclear whether Smythe is demanding payment for the unpaid labour of audiences; and in fact it is unclear to me why he does not include payment for sleep in his demands, given that this too seems to involve the reproduction of labour power.

Smythe’s contribution, favourably cited by some contributors to recent debates (e.g. Andrejevic, 2004: 97, 114), shows the danger of a Marxian analysis that has totally lost its connection to pragmatic political struggle. The recent and often autonomist-inspired interest in free labour cannot be accused of the defeatism underlying Smythe’s almost Orwellian picture of workers giving their free labour to the reproduction of television. As we have seen, though, the danger is that they go rather too much in the opposite direction, seeing unpaid labour as a sign of an immanent revolutionary potential among workers. What may connect them is a lack of a coherent and pragmatic analysis of political struggle, and of lived experience.

Quality of working life in the cultural industries

In this final section, I turn to the point raised earlier concerning the effects of free labour on the conditions of professional and semi-professional workers. Terranova inherits the autonomist Marxist tendency to separate out questions of labour from employment: ‘Labour is not equivalent to waged labour’ (2004: 88), she points out. The internet itself is mainly created by unpaid labourers rather than by employees – just as the vital work of human reproduction and sustenance is mainly conducted by unpaid women. Terranova believes that this insight points to the limitations of a mode of thinking that sees the ‘unemployed’ as lazy and in need of compelling into paid work. But what this means for analysis of the conditions of actual paid workers is not clear.

So in this final section, I want to address the question of how we might best consider work in the cultural industries, in terms of its quality. Doing so allows us to take seriously the lives of workers, in a way that the focus on ‘prosumers’, ‘produsers’ and even free labour sometimes might discourage. My view is that, while recognising the genuinely stimulating and important interventions made by some of the critics discussed above, we need to connect the analysis of cultural and creative labour much more explicitly to the sociology of work in general, and to a richer and diverse body of social theory than has been used in that critical work.

In the book that Sarah Baker and I have recently published on work in the cultural industries (Hesmondhalgh and Baker, 2010) we argue that an important issue for those concerned with labour in the present conjuncture is to think about good and bad work, and what differentiates jobs, occupations and careers from each other in terms of their quality. To pursue this issue, we asked the following question: what kinds of experiences do jobs and occupations in the cultural industries offer their workers? We draw on sociology of work and political theory in an attempt to conceptualise more fully what might constitute good and bad jobs, occupations and careers.

It was clear from our research that many cultural-industry jobs and occupations are riddled with problems and inequities. Many are offered on a short-term basis, making it difficult to plan ahead with any certainty, and constraining workers’ abilities to make their work the basis of meaningful self-realisation. Although many people respect art, learning and knowledge, some see them as mere ornamentation, and creative labour as a kind of social luxury. This limits the degree to which workers in these industries can feel sure of social respect and recognition. Autonomy is always limited, but many creative workers have little control over the products they are involved in, especially in terms of how they are distributed, marketed and publicised. The ‘autonomy’ involved in freelance work can provide freedom to combine work with childcare and family life, but it can also involve isolation and a lack of solidarity with other creative workers. The returns for creative work are highly uneven, and many struggle even to get a foothold in the cultural industries. By contrast, a successful few enjoy considerable benefits in terms of financial reward and recognition, in ways that distort the minor differences in talent that might lead some to succeed more than others. Many workers tolerate poor pay, long hours and difficult conditions in order merely to gain jobs with very poor levels of security and protection. In other words, to achieve the possibility of self-realisation through creative work seems to require what some recent critics, as I pointed out earlier, have called self-exploitation. In the light of these dynamics, many workers leave the cultural industries at a relatively early age, burnt out by the need to keep up to date with changing ideas of what is fashionable, relevant and innovative, a process that requires not only hard work at work, but also a blurring of work and leisure. Many cultural industries now seem to their workers to be more competitive than ever, and staying ahead requires long hours and an intense relationship to the work. Autonomy is a desirable feature of creative work, but it comes hand in glove with self-reliance and an uncertainty about career paths. All in all, there are good reasons, it seems, to think of creative work as, to quote the title of a track by British rap musician Mike Skinner, who performs under the name The Streets, ‘the hardest way to make an easy living’.

These negative factors have led many critical commentators to interpret positive aspects of creative work in the cultural industries as, ultimately, control mechanisms that serve to discipline or seduce workers into putting a great deal of themselves into what they do, and tolerating precariousness and insecurity. Such critical accounts have been an important counter to the complacent portrayal of creative work in creative industries policy. Yet in the form that they have often been presented, such criticisms may leave something of a normative vacuum. For to treat these positive components of creative work as mere sugar coatings for the bitter pill of precariousness is surely too dismissive of the genuinely positive experiences that some creative workers have in their jobs and careers. It’s worth recalling that jobs, occupations and careers in the cultural industries rarely involve gruelling physical demands or tasks that endanger the person undertaking them. They hardly ever involve work of a kind that many others will find disgusting or disdainful (such as a toilet attendant, or nurse who has to care for incontinent patients). In fact, cultural-industry jobs are often thought of as desirable and intriguing, even glamorous. They involve expressive and communicative forms of endeavour, which are highly valued by many people in modern societies. Although this of course depends very much on industry and genre, in principle at least this suggests that they are capable of providing a basis for respect and recognition from others, which in turn can help nourish the worker’s sense of self-esteem, and over time, contribute to projects of self-realisation. Cultural-industry organisations also tend to be structured in such a way that some workers are able to gain high levels of autonomy, in two different senses: workplace autonomy and creative autonomy deriving from ambivalent Enlightenment and romantic conceptions of the value of art and culture. This means that they have the possibility of shaping outcomes, and producing good work in the sense of work that contributes to the common good. There are also significant spaces where excellence is valued and encouraged. We shouldn’t forget then that as well as the negative dynamics I’ve already outlined, the cultural industries provide significant opportunities for good work.

We need to hold on to the ambivalence of work in the cultural industries, and to recognise that firms and other institutions differ in terms of what kinds of conditions and experiences they make available for their workers, including how much they respect autonomy and workers’ needs to make good products (arguments made by Keat, 2000 and Breen, 2007). This suggests that there is a highly significant issue for those concerned with equality and social justice in relation to work in the cultural industries, which some recent critical accounts – including those concerned with free labour – may have served to marginalise. This concerns how positive and emancipatory aspects of labour – including creative labour in the cultural industries – might be made more prevalent, and how negative aspects of work might be contained, controlled or even eliminated. Underlying this question is a deeper one: which political projects may best enhance human well-being and social justice with regard to work? These fundamental issues haven’t yet been analysed sufficiently in the burgeoning literature on creative work.