Theory of Practice Many radical critics relegate themselves unwittingly to a rear-guard position by accepting claims which sciences make about their own methods and by conceding in principle some putatively non-ideological core of scientific knowledge. The implication is that the world is made up of objects just waiting to be taken and studied, as if sciences do not actively construct their own objects of study from existing cultural and physical materials. Consequently, for critics who share the commitment to orthodox canons of scientific knowledge-production, the term 'ideology' can mean only subjectivity which is grossly wilful, obvious, and otherwise distasteful. This kind of rationalism implies that it is only scientists' ideas nature which need to be challenged, and that they can be successfully challenged as 'unscientific'. When this tactic breaks down (as when some racist scientists cannot be shown to have cooked the books), radical rationalists have, of course, no resort but to stand on their own individual-, group-, or class-subjective positions, and so expose themselves to criticism of the more-objective-than-thou variety. By conceding the epistemology objective status of science, radical professionals set limits how far they challenge their own complicity in reproducing forms of power which scientific and other professional practices carry and constitute. Examples are expertise, deskilling, 'participative' decision making, schooling. It is all too easy for people to privilege the practice of professionals, as workers with a particular (if contradictory and unclear) class location, without analysing the political conditions in which the ideas themselves are produced. It is part of our aspiration to make this privileging more difficult.
The significance of the labour process approach here is that it does not limit its attack to the bourgeois ideology of nature — nature as possessing pre-given properties discovered by science. It goes deeper, to attack the bourgeois ideology of practice, refusing to restrict itself to arguments about whether it's true that this chemical is harmful; or that radiation level safe; or that IQ is genetically determined and unequally distributed between the sexes or races (sorry, gene pools); or that microprocessors labour-displacing. We must talk firstly about the kinds of practice which produce these kinds of statements about the world, and how they attain and maintain their hegemony, and how that hegemony can be subverted in socialists' day-to-day struggles over the form and content of work. In a labour process attack in a specific area of science, conceptual artefacts need to be carefully examined in their practical connections — through racial and sexual oppression, hierarchical work organisation, the repression and suppression of alternative modes of creativity; as well as through 'career capital' (Bourdieu). Conceptual and physical artefacts be made to reveal the living labour that they contain, and the living labour that they in turn dominate: apparatus designed to facilitate routine work, scientific papers attributed to prestigious individuals regardless of whose mental and manual labour they incorporate. The approach is entirely compatible with a tactical decision to fight on an ideological front, to debate. But the key distinction underlying such a decision is no longer the demarcation of science and ideology (or, for that matter, science and technology, 'internal' and 'external' factors, or pure and applied science). The key issue is the relation between living and dead labour, forces and relations of production, concepts and reality — as historical class relations concretely present in practices of scientific work.
It is ironic that, in current marxist debates around science, our approach finds itself labelled 'relativist'. Since we regard knowledges as intrinsic to, and varying with, the social forces which constitute and maintain them, this is taken to amount to a complete agnosticism with regard to the validity of knowledges. There are some (academic) schools which collapse into this epistemological state of despair, but RSJ has never allied itself with them. Indeed, a significant portion of our work has involved criticising the positions available in the academic supermarket: science and society, sociology of science, sociology of knowledge, anthropology of knowledge, and various forms of social constructionism in the social sciences. For those whose politics of science are held to ransom by epistemological questions, it appears that our refusal to be trapped into such questions must be a liberal (or anarchist) tolerance for any and all views of nature; it remains out of their ken that we are trying to demote epistemological questions rather than answer them. Our refusal to slip easily into labeling aspects of science as ideology is grasped by the epistemologically preoccupied as a matter of surrendering all existing science to capital — as if it were not already capitalist both in form (as wage-labour, for example) and content. We want to see debate move onto the terrain of practices, taken whole, thus escaping from the intellectual ghetto of the science/ideology debate.
What we think is now possible, within a general framework of labour process theory, is a thoroughgoing historical and materialist approach to the production of theoretical concepts — corresponding, at a general level, to the marxist analysis of 'production in general' (the phrase comes from Marx's 'Preface'). The production of knowledge is paralleled in, and proceeds through, a process of producing physical phenomena. At the most elementary level the physical phenomena are those of speech, extended at more sophisticated cultural levels by experimental apparatus, academic papers and journals, public talks, lectures, broadcasting, pamphlets, etc. Through this apparatus, conceptual objects are transformed into conceptual products (see Hales, Living Thinkwork). What needs to be understood is how this production is materially constituted by the location of a practice in the division of labour, by physical means of production, by wage-labour, by commodity-secrecy (patents, confidentiality), by the book as commodity. From the Second International to the present, we have been stalked by the albatross of an epistemology in which 'material reality' is entirely purged of social organisation. Scientism in the Old and New Left sees things — especially concepts — as if they were only contingently related to purposes. But artefacts are not normally packaged in the form of time-capsules, as we argued earlier in rebutting the use/abuse perspective. Practices are the foundation of a marxist analytical route around the pitfall of epistemology. Let us consider some specific examples.
In the case of designing chemical processes, the design team preoccupies itself with technical criteria such as 'reliability', 'efficiency', 'safety', ‘quality'. These are seen simply as engineering matters, but here we want to analyse how they ultimately mediate the subordination of living labour to capital — indeed, are constituted to do so.
The design team originates a novel conceptual product — the 'design' —objectified in the form of documents. This objectified product passes into other labour processes, within and outside the firm. To become finished ‘means of production', it is reworked by other special kinds of living labour (fabrication, construction, 'rigging', purchasing) along with other kinds of materials: e.g., metal bar- and sheet-stock in contractors' workshops (where the design is tuned to the contractor's labour process), is turned into an on-site skeleton of rolled-steel joists and concrete, filled out with pipes, lagging and electrical wiring, and stocked with suppliers' inventories of feedstocks. The original conceptual product ends up as dead labour in the form of the completed plant, ready for start-up waiting for living labour to attend it. When chemical process workers confront these means of production, they find that this technology has diminished the space for insubordination. They find themselves reduced to the role of mere labourers or of dial-minders requiring little special knowledge to intervene, and what they do need is very specialised and particular. Complementing the workers' enforced ignorance is a wide of knowledges elsewhere in the plant and the industry — those of chemists, engineers, systems-analysts, production managers, personnel managers, sales managers. Along with the designers themselves, these all help to make real capital's material ownership of the plant, subordinating living to dead labour.
Chemical process workers watching the dials are no longer even machine-minders retaining a significant space for refusing to mind the machines properly, e.g., by work to rule or covert sabotage. They are left with fewer options in between the extremes of either accepting their subordination or pushing the panic button. The dead conceptual labour of designers, researchers and managers confronts living labour with the direct challenge: either do the obvious — which we have painstakingly made obvious, so that you need not think about it — or unravel the social and historical forces woven into this situation. And do it in your 'spare' time, because you are not paid to think.
The chemical process design thus advances capital's aim of reducing workers' control over their own labour (and taking that historical tendency as far as the pre-chip technology could take it). The design embodies the 'intelligence' in the fixed capital, which literally fixes the possible uses of the machinery, such that process workers could not choose what to do with the plant even if they wanted to. They do not — and even cannot — know what is possible to do with it. Their only significant 'freedom' is to choose to make the plant not work at all — very high stakes indeed. In this situation, the knowledge which the workers would need, in order to resist, encompasses the entire labour process. Combatting their own subordination to dead labour requires analysing the fixed capital back into the capitalist relations of production it embodies, back further into the conceptual product (the design) which originated it — and then intervening at the stage of origination. This doesn't mean challenging the scientificity of the engineering criteria but rather uncovering the values which the design embodies and substituting oppositional values which contest the capitalist relations of production.
The Grunwick photo-processing plant provides another illustration of how living labour is subordinated indirectly by conceptual products — in this case, by logical sequences written in computer languages embodied in computers organised to eliminate living labour from those tasks most amenable to workers' job-control and resistance (Levidow, 'Grunwick' in RSJ 6/7). Readers familiar with the Grunwick strike may find this example strange. since Grunwick became notorious for the low wages, compulsory overtime and autocratic shopfloor control which management imposed, particularly upon the Asian women in the mail-room. Indeed, their labour-intensive task lent itself to images of 19th century sweatshops which needed to extract 'absolute surplus value' for failure to invest in technology which could increase productivity ('relative surplus value'). Accordingly, Grunwick has been popularly represented as a Dickensian employer who needed to be forced into the 20th century.
However, we need to understand Grunwick in almost the opposite sense — that is, as capital's vanguard of the 20th century, particularly because of its roles in consumer culture and in automation. Firstly, Grunwick has come to occupy a small corner of the modern hegemony industry, geared to the fast turnover of holiday snapshots — the Instamatic's approximation to the 'instant nostalgia' made available from Polaroid and now Kodak. Grunwick won its place in that market by using advanced technology to computerise the customer accounts and the chemical processing of the film.
Secondly, the computerisation made a bottleneck out of the clerical work done in the mail order department, where management ruthlessly pressured the workers to keep pace with the rest of the process. In other words, the particular choice of the hardware investment followed on from an assessment as to which human tasks were marginally worth replacing or subordinating with fixed capital, so that other tasks could be exploited more flexibly. Thus Grunwick's very selective automation facilitated exploitation of the mail room workers, whose numbers and workpace could be varied at will by the management, according to seasonal (or even daily) fluctuations in custom.
The campaign to unionise Grunwick arose precisely out of the exploitation of living labour in the mail room and focused on the articulation between that labour-intensive process and the postal sorters. When the blacking of postal deliveries was defeated, attention turned to other public sector workers — gas, electricity, water — but deference to threats of legal action precluded denying those essential services. A labour process perspective might have drawn attention to articulations between the Asian women's work and the rest of Grunwick — the fixed capital. The capital-intensive high technology photo-processing equipment could have been analysed in terms of raw materials, means of production and purposive activity. The analysis would include delivery of chemicals and paper, articulating with workers in transportation and manufacture. It would include maintenance of, and spare parts for, the processing machinery. Seeing through the dead labour, to the living labour it embodies and the living labour it controls might have provided many more opportunities for class solidarity. Had it all come alive, the employer, George Ward could have been confronted with opposition from subversive workers on all sides.
And if, in the particular case of Grunwick, the management always prevented or circumvented selective blacking with a little help from its friends at the National Association for Freedom (NAFF), then we could move on from the labour process proper, to ask how the company's allies succeeded in severing the labour process politically. How did the postal and essential service workers get defined as outsiders lacking any legitimate material interest in the Grunwick strike? Indeed, how did doubts arise about the strikers themselves not being real workers, by virtue of not producing for Grunwick? (Levidow in Levidow and Young).
It might appear that examples from chemical process design and Grunwick will not extend easily to embrace medicine and science, but this is far from the case. Think of the complex and rigid division of labour and hierarchical organisation of medical education and institutions — all done in the name of efficient care. Similar analyses of the social relations of scientific research and institutions are just beginning to be made. They are less open to summary presentation because of the complex and subtle mediations between the research and the wider forces at work constituting them. Nevertheless there is a growing literature: Simon Pickvance on "'Life" in a Biology Lab', Jean-Paul Levy-Leblond on 'Ideology of/in Contemporary Physics', Andre Gorz on 'The Scientist as Worker: Speedup at the Think Tank', Donna Haraway on 'The Biological Enterprise: Sex, Mind and Profit from Human Engineering to Sociobiology' (as well as other, more detailed papers), and Edward Yoxen on 'Life as a Productive Force: Capitalising the Science and Technology of Molecular Biology'. 
Most concept-producing practices have a significant real object of labour as well as a conceptual object — be it a page (to be made into a book), an ‘audience' (to be 'enlightened'), or a piece of experimental apparatus (which will demonstrate the truth of a concept), or an industrial labour process (to be made more 'efficient'). Some practices, like old-style pure maths, appear to have no real object, and the great variety of possible relations between conceptual and real object, offer an enormous possible range of varieties of rigour in intellectual work.
Think of laboratory science, 'operational' research and systems analysis, 'mission-oriented' but 'basic' research, applications research and technical service, non-experimental natural science; and also about hard and soft social sciences, literary traditions, political traditions. Each has its characteristic forms of rigour within which work is judged to be good of its kind. By a gross process of ideological and financial hegemony, some of these forms of rigour attain cultural supremacy. What labour process analysis offers is the opportunity to analyse at a general level these different modes of theoretical and ideological production, so that socialists can evaluate them in quite detailed and specific tactical studies, according to their needs in prefigurative struggle. Just how different, and how robust, are 'scientific' modes of theoretical production, compared to others? And how can a strategy for agitation and transformation in the sphere of the sciences be integrated with an assessment of political strategies leading to alternative social forms? We think that this kind of strategy is now within reach as an offensive one for the radical science movement, and as an absolutely necessary complement to defensive struggles on the terrains of ideology on the one hand and of redundancy agreements on the other.
The Reflexive,the Personal, and the Professional We are attracted to the labour process approach, in preference to narrower ones, because it offers a more direct route to agitational practice. It does this by clarifying possibilities for contestation in the settings in which we work and struggle every day. As a theory of practice, labour process theory must first be applied as theory to our own practice.
Without such a self-reflexive theoretical approach, we cannot clarify whether and how our activities truly challenge capital's control. How else can we avoid reproducing, even in our radicalism, the oppressive social relations of the social order which we consciously intend to oppose? It is its reflexivity — the necessary reflexivity of a theory of practice — which gives the labour process approach its edge over approaches focusing on some 'external' nature instead of on socially constituted nature.
We are growing wary of the self-labelling of the 'radical science movement'. Given our educational and employment histories, we certainly consider the sciences and other spheres of 'professional' work as important territory in which to struggle politically. But we notice a tendency to fetishize science as a separate, specialised sector of struggle — the new technology, IQ, sociobiology, hazards, the technology of state repression. The labour process approach, in our experience, connects radical science with other spheres of activity. These include white-collar and shop-floor work in industry (ICI and Lucas Aerospace), other political movements (notably, the women's movement), and groupings of radical academics outside science (the Conference of Socialist Economists, for example). In making those links, in both theory and agitational practice, we are loath to grant science any privileged status. We do not wish to elevate our scientific backgrounds to the level of a specialist expertise, i.e., to represent an occupational contingency as a political virtue. We need to get out of a scientific/scientistic ghetto.
The 1970s saw the emergence of a large number of groups of intellectual workers spanning disciplines and transgressing conventional boundaries between academic work, political activism and non-academic work, while still committed to theoretical activity. In shifting from a focus on academic economics, the CSE has become a federation of such groups. This is where the RSJ collective has been most directly involved, via CSE working groups on education, the labour process and left strategy, microprocessors, and ICI. This last one, along with the Motors Group, is an example of a joint venture within the CSE structure between intellectual workers (mainly academics) and militant industrial shop-floor workers. These in turn are only part of the network of academic/shop-steward connections which has begun to develop in the context of workers' struggles in specific firms and sectors of employment: Vickers, Parsons, GEC, Chubb, motor industry firms, Lucas Aerospace; also in state employment. The Lucas situation is unique in a number of respects — the workers' combine Corporate Plan being a development which inspired many others. It has gone further in setting up its own apparatus than any of the other combine-based research projects, both in the extent of organisation through the combine and in the establishment of a research centre with some full-time staff (Centre for Alternative Industrial and Technological Systems at North-East London Polytechnic). The Community Development Project and many local community action projects have deposited residues of socialist-intellectual activism which link closely with ventures such as those above and also with local health and safety groups which blossomed in the wake of the Health and Safety at Work Act: Coventry Workshop, Benwell Community Project, Coventry Health and Safety Movement, Manchester Area Health and Safety Committee; Trade Union and Community Research and Information Centre (TUCRIC) in Leeds, TUSIU in Newcastle; socialist education centres. Our list is selective, but indicates the lively innovation that has been taking place between academic research and industrial trade union activism. It seems obvious to us that such initiatives in socialist politics are too important to pass without serious attempts at general analysis. Forthcoming publications on shop stewards combine committees, and a recently published book on struggles In and Against the State (London Weekend-Return Group) show that others share this concern.
At the most general level, there is one debate now surfacing which has great significance for political strategy in the radical science movement. It concerns the role of 'the Professional-Managerial Class' (PMC). Barbara and John Ehrenreich started the debate with an article addressing the emergence of the American New Left and its limitations in the 1970s, which also rang true in the British experience (Walker). The authors defined the PMC as 'consisting of salaried mental workers who do not own the means of production and whose major functions in the social division of labour may be described broadly as the reproduction of capitalist culture and capitalist social relations' (Radical America 11 (3), p. 7). They argued of the PMC and the working class that, 'Both classes confront the capitalist class over the means of production. They confront each other over the issues of knowledge, skills, culture' (p. 22; Walker, p. 45). As a theoretical perspective, this has many resonances for people in our position. Specifically, it challenges us to analyse the class position of broadly New Left and new Old Left movements of which radical science is only a part.
Members of the PMC, as described by Barbara and John Ehrenreich, range from teachers, lecturers, social workers and media workers, through doctors, nurses, psychologists and consultants of many kinds, to managerial, scientific and technical workers. We would like to make further distinctions among scientific workers, for surely it matters whether, we are manager or managed, research associate or lab technician, academic or industrial or unwaged worker, professor or student, natural or medical or management or engineering scientist. Depending on where we work, our politics must differ, and to 'dig where we stand' will have quite different implications. How should we understand the political differences between professors and those intellectuals who may be neither scientists nor academics, or among various intellectuals comprising an editorial collective of a radical science journal? In trying to clarify alliances and resolve unexamined tensions, the reflexivity of the labour process approach has become our touchstone. As theory of practice - specifically, of knowledge-producing practice - it is surely our best bet for a putative theory of the internal and external relations of the PMC. In the 1980s we want to see the radical science movement moving on: beyond science into work, beyond ideology into culture, beyond knowledge as such to knowledge as subversive practice. The labour process conceptual framework offers us the chance of articulating our theory with this necessary practical advance.
Our renunciation of the intellectual's bolt-hole of 'objectivity', and our stress on the necessary self-reflexive aspect of any radical practice, has provoked severe reactions from some quarters. Many reject our insistence that radicals must include in their criticisms the criticism of their own ways of doing things — not simply criticism of their opponents' and their own former views. That problem has manifested itself in a way inseparable from that of finding it difficult to work collectively. We could make things easier for ourselves (at least in the short run) by ignoring the contradictions of our lives as professional workers attempting to challenge professionalism, as meritocrats attempting to overcome individualism. We could concentrate more intensively on the products and less on the political process of our practice. But this would be to abandon the sinequa non of labour process politics (like RSJ politics since its inception) — that it is processes which have to be grasped, in all their contradictoriness, and that political self-consciousness amounts to nothing if it starts and ends 'out there', outside of our own class relations.
Living this commitment can raise personal anguish to a high pitch. The contradictoriness of having a career (or any job at all) in science and simultaneously conducting a critique of science can erupt in individuals with a force of unbearable intensity. Though we regret such victories by capital's division of labour, we should not be surprised that working scientists tend to compromise on the critique rather than on the career. Of course it can happen the other way around, with workers leaving their jobs because of political commitments, but the quandary remains no less serious for them. The lived contradictions involved in keeping the job while struggling collectively to transform it lie at the heart of labour process politics.
Our aim in this section has been to give a clearer historical, class meaning to the insight that 'the personal is the political'. The labour process approach, with its essential reflexivity, offers us the broadest way of doing so. We regret that some socialists reject the insight entirely by parodying reflexivity as 'hyper-reflexivity' (H. Rose).
We said at the beginning that we wanted to show that the problems raised by the privileged treatment of science have been encountered by the left in other guises. In the course of our argument it has turned out that legitimacy, correct line, orthodoxy and the entire basis of political practice have all been at stake — so have some of the peculiarities of marxist intellectual and political culture in Britain. Science, technology and medicine are found at the centre of capital's current restructuring of the labour process. Unless we can learn how transform the capitalist social relations embodied in the labour process, and to transform our own labour process accordingly, no amount of objectivity and no amount of scientific research and debate will bring about socialism rather than barbarism.