The history of measurement and the engineers of space

BJHS, 1993, 26, 459-68 The history of measurement and the engineers of space ANDREW BARRY* For the social theorists of the late nineteenth and early twentieth centuries, measurement, quantification and calculation were of particular social and political significance. Karl Marx, in Capital, based his critique of classical political economy on an analysis of the quantification of labour as a commodity. Max Weber, in Economy and Society, emphasized the importance of rational calculation in the conduct of modern bureaucratic organizations. And in his major work, The Philosophy of Money, Georg Simmel highlighted what he called 'the calculating character of modern times'. 1 In this paper I want to emphasize the strategic importance of the history of measurement and calculation for both historians of science and sociologists. Although contemporary sociology has reconstructed, if not rejected, the paradigms of classical social theory, the sociology of measurement and calculation none the less remains central to understanding some of the key political and economic processes of the modern world. I argue that the study of the history of measurement provides a way of developing what Steven Shapin once called the ' sociological reconstruction' of the history of science, as well as establishing a basis for understanding the relation between localized practices of science and the ' bigger' issues of politics and economics.2 By way of introduction I want briefly to characterize the kinds of relation which have hitherto existed between the history of science and sociology. First, sociology and social theory have provided various master narratives of social and economic change in the context of which the history of science has been read. Sociology, in effect, provided the 'big picture' within which the history of science could be placed. Marxism is the most obvious example of such a sociological master narrative. But it would be fair to say that Marxist studies of science have remained often crude or undeveloped — sometimes amounting to little more than an assertion of the determination of scientific change by economic interests or class struggle.3 More sophisticated neo-Marxist * Department of Sociology, Goldsmiths' College, London SE14 6NW. My thanks to an anonymous referee and to Georgina Born and Jim Secord for helpful comments made on an earlier version of this paper. 1 G. Simmel, The Philosophy of Money (tr. T. Bottomore and D. Frisby), London, 1990, 443. 2 S. Shapin 'History of science and its sociological reconstructions', History of Science (1982), 20, 157—211. 3 See B. Hessen 'The social and economic roots of Newton's Principia' in N. Bukharin et al. (eds.) Science at the Crossroads, London, 1931. On the reception of Hessen's work in the history of science see S. Schaffer, 'Newton at the crossroads', Radical Philosophy (1984), 37, 23-42, and L. Graham 'The socio-political roots of Boris Hessen: Soviet Marxism and the history of science', Social Studies of Science (1985), IS, 705-22. 460 Andrew Barry sociological accounts of cultural and intellectual production have had relatively little impact on the sociology of science in Britain.4 The marginalization of Marxist sociology within British science studies is interesting in itself, and might be explained in a number of ways. One explanation might relate to the lack of interest within British Marxism in the study of science in comparison, for example, with the analysis of artistic and popular cultural forms. Another explanation might point to the rise of the sociology of scientific knowledge from the mid-1970s onwards which, with exceptions, remained either hostile or indifferent to the Marxist and Weberian traditions in social theory.6 In an influential review book of the new field of the sociology of scientific knowledge published in 1983, most of the authors were uninterested in theorizing developments in science and technology in relation to broader economic or social structures — whether these were understood in Marxist terms or otherwise.6 Of course the main thrust of the sociology of scientific knowledge was not an attack on sociology per se, but on what were perceived to be orthodox positions in the philosophy of science and the sociology of knowledge. In this context, it was argued persuasively that microsociological attention to the details of scientific practice and discourse and to the peculiarities of laboratory culture was necessary to demonstrate the social construction of scientific facts.7 The intention was to provide, in effect, an empirical refutation of the philosophical orthodoxy. The very idea of examining the relation between scientific activity and broader political or economic developments, characteristic of the Marxist project, was implicitly dismissed as irrelevant or regarded as a mistake or, more charitably, as something which might be postponed to a later date.8 For some authors, indeed, any macrosociology was necessarily allied to the kind of realist philosophy of science which the sociology of scientific knowledge was supposed to attack. While the sociology of scientific knowledge neglected to theorize the broader social relations of intellectual activity, a similar neglect did not occur in related areas of the sociology of culture and cultural studies. For the sociology of culture, the problem was not so much to abandon any attempt to analyse the relation between intellectual production 4 Raymond Williams' work in the sociology of culture and Stuart Hall and David Morley's studies of nonfiction television highlight the relative sophistication of Marxist accounts of cultural production and consumption in comparison with contemporary Marxist accounts of science. See R. Williams, Culture, London, 1981, and Marxism and Literature, Oxford, 1977; S. Hall, D. Hobson, A.Lowe and P.Willis (eds.), Culture, Media, Language, London, 1980. For a general introduction to neo-Marxist and Weberian approaches to the study of cultural production see J. Wolff, The Social Production of Art, London, 1981. 5 For exceptions to this see B. Barnes, Scientific Knowledge and Sociological Theory, London, 1974, and Interests and the Growth of Knowledge, London, 1977; D. MacKenzie, 'Statistical theory and social interests: a case study', Social Studies of Science (1978), 8,35-83, and B. Wynne, Rationality or Ritual: the Windscale Inquiry and Nuclear Decisions in Britain, British Society for the History of Science, 1982. 6 K. Knorr-Cetina and M. Mulkay (eds.), Science Observed: Perspectives in the Social Study of Science, London, 1983. Latour's article in this collection was an important exception. 7 See, in particular, the important work of Harry Collins (H. Collins, Changing Order: Replication and Induction in Scientific Practice, London, 1985). A good overview of the present state of the debate within the sociology of scientific knowledge is provided by A. Pickering (ed.), Science as Practice and Culture, Chicago, 1992. 8 For a rejection of the possibility of a macrosociology of scientific knowledge see M. Lynch, Art and Artifact in Laboratory Science: A Study of Shop Work and Shop Talk in a Research Laboratory, London, 1985. The suggestion that the macrosociology of science should be postponed to a later date is made in H. Collins, 'Stages in the empirical programme of relativism', Social Studies of Science (1981), 11, 3-10. The history of measurement and the engineers of space 461 and broader economic and political developments, but to move beyond the various Marxist analyses of this relation. Pierre Bourdieu's work, for example, suggested one way of understanding the role of intellectual and artistic production in the reproduction of class relations which avoided the functionalism and reductionism of contemporary Marxist accounts.9 According to Bourdieu, the production and consumption of culture (including science) occurs within the context of particular cultural ' markets' or fields, the rules of which are not determined by economic relations. Such fields exist as spaces of social struggle, the rules of which govern the legitimacy of statements made and the 'symbolic power' of actors participating within the field. Scientific work can, in Bourdieu's view, aspire to a form of objectivity, but this objectivity cannot be equated with political neutrality. As Bourdieu argued, 'an analysis which tried to isolate the purely "political dimension" in struggles for domination of the scientific field would be as radically wrong as the (more frequent) opposite course of only attending to the "pure", purely intellectual, determinations involved in scientific controversies'.10 If Bourdieu's work provided a way of rethinking the analysis of class and ideology in the sociology of culture (and science), a more influential movement beyond Marxist approaches to the sociology of culture derived from a consideration of the significance of discourse in so-called 'poststructuralist' and 'post-Marxist' social theory. For poststructuralist social theory the idea of conceiving of the social order as a structured totality, as Marx had done, was mistaken. The ambition of poststructuralist research has not been to establish knowledge of the ' real' character of society, but to analyse the ways in which and the conditions within which human activity becomes the object of knowledge, reflection and political action. In this context, the identities of individual and collective subjects need to be understood not so much as the effects of an underlying social structure, but as the unstable and historically contingent product of political and scientific discourse.11 Despite the isolation of the sociology of science from other related and relevant areas of sociological debate, some of the most important work in the history of science in recent years has been strongly influenced by the sociology of scientific knowledge, seeking to provide 'sociological reconstructions' of the history of science. This work has highlighted 9 See P. Bourdieu, The production of belief: contribution to an economy of symbolic goods', in Media, Culture and Society (ed. R. Collins et al.), London, 1986; Distinction: a Social Critique of the Judgement of Taste (tr. R. Nice), London, 1984; Homo Academicus (tr. P. Collier), Oxford, 1988; Language and Symbolic Power (tr. G. Raymond and M. Adamson, ed. J. B. Thompson), Oxford, 1991. A useful introduction to Bourdieu's work is provided by John B. Thompson in an introduction to Language and Symbolic Power. See also C. Jenks (ed.) Cultural Reproduction, London, 1993. 10 P. Bourdieu 'The specificity of the scientific field and the social conditions of the production of reason', Social Science Information (1975), 14 (6), 21. For a critique of Bourdieu's analysis of natural science see K. KnorrCetina, 'Scientific communities or transepistemic arenas of research: a critique of quasi-economic models of science', Social Studies of Science (1982), 12, 101-30. 11 See, for example, M. Barrett, The Politics of Truth: From Marx to Foucault, Oxford, 1992; E. Laclau and C. Mouffe, Hegemony and Socialist Strategy, London, 1985; T. Eagleton, Against the Grain, London, 1986, 89-98; M. Foucault, Power / Knowledge: Selected Interviews and Other Writings 1972-77 (ed. C.Gordon), Brighton, 1980; B. Hindess, 'Power, interests and the outcome of struggles', Sociology (1982), 16, 498-511; J. Wolff, 'The global and the specific: reconciling conflicting theories of culture', in A. King (ed.), Culture, Globalization and the World-System, London, 1991. Within the sociology of science the influence of such arguments is most evident in the work of Michel Callon, Bruno Latour and John Law, see J. Law (ed.), Power, Action and Belief, London, 1986. 462 Andrew Barry the particular historical constitution of the laboratory space, the importance of the mundane practical work of technicians to scientific developments, and the complex underdetermined character of scientific controversies. However, there was a danger, from the perspective of the sociology of science, that the history of science could merely have been the source of small-scale case studies — and ones which moreover did not have the kind of ethnographic detail which many sociologists of scientific knowledge argued was necessary for the success of their project. There was a danger also, that by failing to locate specific scientific developments within a broader social and political analysis, the history of science might have continued to be relatively marginal to the concerns of sociologists and historians working in related fields.12 In what follows, I want to argue that far from being marginal to sociology, the concerns of historians of science do intersect with some quite central debates in social theory and historical sociology. One starting point for a reformulation of the relation between sociology and the history of science is the recognition that contemporary sociology enables us to conceive of the 'big picture' in spatial or geographical as well as in temporal terms. In the first volume of his book, The Sources of Social Power, Michael Mann traces the history of what he perceives to be the most developed forms, or what he calls the ' leading edge' of power from prehistory up to the eighteenth century. This involves extensive discussion of the character of military, ideological and economic power in Assyria, in Ancient Egypt, in the Roman Empire and in Medieval Europe. What is interesting about Mann's analysis is that unlike Marx, for example, he does not regard social structure as an existing order from which power derives. Rather he tries to account for how certain 'sources of power' are to operate across large areas of territory thereby constituting the society as an 'organised power network'. The problem Mann sets himself is as much a spatial problem as a problem of historical narrative: the problem, as he puts it, of understanding ' the capacity [of these sources of power] to integrate peoples and spaces into dominant configurations'. 13 One can see a related emphasis on space in the work of Anthony Giddens on the modern nation-state. For Giddens the success of the nation-state in establishing itself as the key locus for modern forms of political power is essentially a spatial success: the capacity of a combination of military and bureaucratic organizations to act across time and space through the development of effective means of gathering, storing and transmitting information.14 While I would not wish to endorse many aspects of Giddens' and Mann's approaches, they point to something quite important. That is the need to think ' the big picture' (society, the economy) not in terms of the overall evolution of an ordered totality, but in terms of more localized entities, where the focus is on their constitution and reproduction in time and space. 12 This is not the place to review the increasingly rich connections between the history of science and other fields of cultural history. One example of an issue where there has been a broader dialogue between the history of science and related areas of social and cultural theory has been in the study of gender. See, for example, D. Haraway, Simians, Cyborgs and Women, London, 1991; L. Jordanova, Sexual Visions: Images of Gender in Science and Medicine Between the Eighteenth and Twentieth Centuries, Hemel Hempstead, 1991; J. Taylor, 'Dreams of a common language: science, gender, culture', New Formations (1991), 13, 103—11. 13 M. Mann, The Sources of Social Power, vol. 1, Cambridge, 1986, 31. 14 A. Giddens, A Contemporary Critique of Historical Materialism, London, 1981; The Constitution of Society, Cambridge, 1984; The Nation-State and Violence, Oxford, 1985. The history of measurement and the engineers of space 463 In the sociology of science Bruno Latour's work is one indicator of an analogous shift from an interest in the historical evolution of science to its constitution in space and time. In the Latourian world, science is primarily a form of knowing and acting at a distance - of constituting a relationship between esoteric and localized laboratory practices and a network of distant objects. The power of a scientific argument or a measurement is not determined by its truth, but rather judged in terms of its capacity to act across space and time - to mobilize a network of social and technical actors.15 Although scientific and technical practices are means for acting across space they must also be understood as means for acting upon space. ' Space' itself must be regarded not as an independent reality — the ether of the social world — but as an object of discursive and technological construction.16 For my purposes, the concept of space does imply the existence of a 'geographical' or 'physical' distance between persons or objects. However, this does not mean that spatial relations can be analysed in terms of the realist vocabulary of physics or anatomy or geography. Although space necessarily has a 'physical' dimension, it does not have an existence or an effectivity independent of its constitution with specific technological apparatuses. Indeed, instead of analysing space in terms of the language of physics or geography, these disciplines must themselves be seen as part of a range of technologies historically associated with the engineering of spatial relations. I want to make two connections between the emphasis on space in the sociology of science and that found in the mainstream areas of sociology to which I have alluded. First, despite huge theoretical differences between the work of Giddens and Mann, on the one hand, and Latour on the other, these parallel shifts are far from coincidental. They are, in effect, one way of responding to and countering a certain form of postmodernism, which in its recognition of the textuality or discursivity of history or sociology ends up by denying the existence of any significant differences between 'knowledge' and fiction.17 While we may agree with the postmodernists that the production of knowledge results in the discursive construction of reality, this does not mean that sociology, history and natural 15 The problem of the relation between the situated character of laboratory work and the formation of social and scientific networks was raised by Collins (see Collins, op. cit. (7), 29—78). For Latour's approach to this problem see M. Callon and B. Latour, 'Unscrewing the Big Leviathan: how actors macrostructure reality and sociologists help them do so', in Advances in Social Theory and Methodology (ed. K. Knorr-Cetina and A. Cicourel), London, 1981; B. Latour, 'Visualisation and cognition: thinking with hands and eyes', Knowledge and Society (1986), 6, 1-40; Science in Action, Milton Keynes, 1987. For a cogent discussion of some of the limitations of Latour's reliance on a semiotic analysis of scientific discourse see S. Schaffer, 'The eighteenth Brumaire of Bruno Latour', Stud. Hist. Phil. Sci. (1991), 22, 174-92. 16 P. Virilio, Speed and Politics (tr. M. Polizotti), New York, 1977. 17 Within the sociology of scientific knowledge the debate about postmodernism turned on the question of 'reflexivity'. For examples of 'postmodern' approaches to the sociology of scientific knowledge see S. Woolgar (ed.), Knowledge and Reflexivity, London, 1988, and M. Ashmore, M. Mulkay and T. Pinch, Health and Efficiency: A Sociology of Health Economics, Milton Keynes, 1989. For Latour's arguments against postmodernism in the sociology of science see, for example, B. Latour, ' The politics of explanation: an alternative', in Woolgar (ed.), op. cit. (this note). For more general discussions of postmodernism in sociology and anthropology see P. Rabinow,' Representations are social facts: modernity and post-modernity in anthropology', in Writing Culture: The Poetics and Politics of Ethnography (ed. J. Clifford and G. Marcus), Berkeley and Los Angeles, 1986; S. Hall, 'On postmodernism and articulation', Journal of Communication Inquiry (1986), 10 (2), 125-9; D. Harvey, The Condition of Postmodemity, Oxford, 1989; P. Hirst, 'An answer to relativism?', New Formations (1990), 10, 13-24; R. Boyne and A. Rattansi (eds.), Postmodernism and Society, London, 1990. 464 Andrew Barry science are merely various 'literary genres'. The emphasis on the spatial in recent social theory highlights the ways in which the production of knowledge involves more than an attempt to construct a representation of 'reality'. It is also one way in which particular spaces of the natural or social world are established as possible objects of reflection, practical action and intervention. The second connection between these two arguments is that if we follow the analysis of Michael Mann into the modern world we would necessarily have to examine the role of science and technology as modern sources of power. In what follows I want to focus on one particular aspect of this — the ways in which measurement and other forms of scientific representation have been deployed in the regulation of social and economic relations over large 'geographical' areas of space. In my view, the link between social and economic regulation, measurement and space has two broad aspects. First, as Max Weber argued, the use of measurements and calculations of performance and efficiency are central activities within most modern bureaucratic and industrial organizations. 18 To paraphrase Habermas, there has been a 'numericization of politics' — a growing role for numerical calculation in political decision making at all levels.19 Here I want to outline four themes in the history of such measurements and how they relate to the ordering of social, economic and technical activity. 1 There is a history of a whole series of measurement standards and measurement techniques - of weight, distance, energy, current, radioactivity, toxicity etc. - which are developed today within central government standards laboratories. The standardization of measurement is important because it enables rapid comparisons to be made between disparate and distant objects in commercial and public laboratories and those that are to be found in the natural and social environment. But technical standards are also important because they help to standardize the subjects of measurement — enabling engineers and technocrats to trust each other's numbers not on the basis of any direct personal relationship, but simply owing to the existence of recognized standards laboratories and the discipline of well-defined measurement procedures. In this way, measurement standards and techniques have been perceived to play an ethical or a moral role — both as a requirement for disciplined laboratory practice and as a way of establishing the moral standards of the laboratory in relation to the larger economic and political space.20 2 There is a history of the various measurements and other forms of representation that have been increasingly used to visualize and assess the state of an industry, a nation or 18 On the importance of knowledge to the practice of modern forms of government see M. Weber, Economy and Society (ed. G. Roth and C. Wittich), part n, New York, 1968, 956-1005. For a recent development of Weber's arguments see C. Dandeker, Surveillance, Power and Modernity, Oxford, 1990. 19 J. Habermas, 'Theory and practice in a scientific civilization', in Theory and Practice, Beacon Press, 1973. On the relation between politics and the history of quantification see T. Porter, The Rise of Statistical Thinking 1820-1900, Princeton, NJ, 1986; I. Hacking, The Taming of Chance, Cambridge, 1990; and N. Rose, 'Governing by numbers: figuring out democracy', Accounting, Organizations and Society (1991), 16 (7), 673-92. 20 M. N. Wise and C. Smith, 'Measurement, work and industry in Lord Kelvin's Britain', Hist. Stud. Phys. Sci. (1986), 17, 147-73; S. Schaffer, 'Victorian metrology and its instrumentation: a manufactory of ohms', in Invisible Connections (ed. R. Bud and S. Cozzens), Bellingham, Washington State, 1992. The history of measurement and the engineers of space 465 an empire. Certainly many of these measurements are employed as social and economic indicators — indicators of savings, poverty, investments, productivity, outputs and the balance of trade. But there are also measurements and representations of the quantity and quality of mineral resources, the demography and health of populations, the uses and varieties of plants and the physical geography of territories. Such measurements do not serve merely to visualize reality: they serve to translate it into a form in which it may become amenable to management, intervention, or exploitation.21 However, what is equally important about these measurements for the historian of science is their dependence not primarily upon laboratory experiments, but upon the more problematic techniques of fieldwork, expedition and survey which enable measurements to be made over a larger area of territory. 3 There is a history of the various ways in which measurements have been fed back into calculations at the centre, whether the centre is a medical laboratory, the control centre of a space research organization or the intelligence unit of an army.22 Such a history would include, for example, the way in which measurements of natural phenomena have been juxtaposed with political, economic and moral assessments - in making judgements about medical treatments, for example. But it would also involve the ways in which measurements of distant objects have been fed back into monitoring the effectiveness of the centre itself. Whole areas of engineering, computer science and expert systems development have grown up around the problem of feedback — of developing ways of rapidly processing measurements to determine the future operations of the centre. In other words calculation is used as a technology of self-regulation. 4 There is a history of the way certain kinds of measurement and calculation have been diffused to the wider population. We can see this if we think of the meaning of the idea of 'the public understanding of science' or the 'popularization of science'. As they stand these terms are rather misleading — concealing the historical and social variability of the relation between science and its public - science as entertainment, as spectacle, as a source of truth. One way in which the public/science relation has been conceived, at least in the recent past, has involved an expectation that the public themselves need to understand the character of environmental and medical risks: a public conceived as citizens who draw upon institutionalized measurements and must themselves accept the responsibility to judge the risks and possibilities which they have to confront.23 AIDS and cancer are obvious examples of the fact that the public today are not only the objects of medical measurement, but subjects who are expected to make certain kinds of socio-medical calculation. My more general point is, then, that the regulation of social and economic relations does not only operate from a relatively small number of centres of measurement and calculation to the general population, but may also be generated within the population itself. 21 See P. Miller and N. Rose, 'Governing economic life', Economy and Society (1990), 19 (1), 1—30. 22 See, for example, R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Cambridge, Mass., 1986; G. Mulgan, Communication and Control, Cambridge, 1990. 23 On the relation between the popularization of science and the development of liberal ideas of public education see, for example, T. Bennett, 'The exhibitionary complex', New Formations (1988), 4, 73—102. For a more general discussion of the politics of public education which develops this point see I. Hunter, Culture and Government, London, 1988. 466 Andrew Barry The second broad way in which measurement is intimately linked with the exercise of modern forms of social and economic regulation is in long-distance communication. Michel Foucault noted, for example, how, at the beginning of the nineteenth century what he called the chief' engineers of space' were the civil engineers from the Ecole des Ponts et des Chaussees.24 A similar title could be given to the physicists and electrical and telegraph engineers who in the latter half of the nineteenth century, established the imperial communications networks. Sanford Fleming, a Canadian engineer, noted that the electric telegraph subjects the whole surface of the globe to the observation of civilized communities and leaves no interval between widely separated places proportionate to their distances apart.25 While we might question both the ambition and the possibility of such a global system of surveillance, in general terms the connection between communications and the regulation of social and economic activity is quite clear and is a very rich field for further research. I will briefly discuss two areas of interest: 1 The first area concerns the whole set of issues around the establishment of lines of communication which may be needed to be at once reliable, cheap, secure and fast. Exact measurement, often linked to what is called basic research, has been critical here. Astronomical research defined the position of stars for navigation; electrical measurements and standards enabled British laboratories to regulate the properties of cables hundreds of miles out in the Atlantic Ocean.26 At the same time the laboratory disciplines, encouraged by standard measurement practices, were the basis for reliable procedures in telegraph engineering. Much later, in the 1920s, measurements of electrical reception enabled Marconi to define the conditions for short-wave transmission; and this allowed others to suggest the kinds of procedure required by skilled radio operators to ensure the most effective use of their equipment. In brief, effective long-distance communication required both measurements of the properties of objects, and the management and training of operatives and engineers who could be relied upon to carry out their work at a long distance from the centre. It is perhaps not surprising that the telegraph industry in the late nineteenth century was one of the first industries (along with the railway) to develop a bureaucratic management structure.27 To achieve long-distance communication thus required the standardization of objects and practices. 2 The second area of research concerns a set of issues which arises owing to the speed of modern communication systems, which can enable almost instantaneous trans24 M. Foucault, 'Knowledge, space, power', The Foucault Reader (ed. P. Rabinow), Harmondsworth, 1986. 25 Quoted in A. Briggs, Victorian Things, Harmondsworth, 1988, 374. 26 Schaffer, op. cit. (20). Other useful discussions of the early history of electrical communications include H. Aitken, Syntony and Spark: The Origins of Radio, Princeton, NJ, 1985; C. Marvin, When Old Technologies were New: Thinking about Communications in the Late Nineteenth Century, New York, 1988; and D. Headrick, The Invisible Weapon: Telecommunications and International Politics, New York, 1991. On the significance of the development of rapid communication systems for the understanding and analysis of war in the late nineteenth century see D. Pick, War Machine: the Rationalisation of Slaughter in the Modern Age, Chicago, 1993, 165-75. 27 M. Weber, 'The development of bureaucracy and its relation to law', in Selections in Translation (ed. W. Runcimann), Cambridge, 1978. See also A. Chandler, The Visible Hand: the Managerial Revolution in American Business, Cambridge, Mass., 1977. The history of measurement and the engineers of space 467 mission of measurements to the centre. Paul Virilio, for example, has argued that a radical change occurred in the practice of war from the late nineteenth century onwards with the development of new communication technologies such as telegraphy, aerial' photography, film and later radar and satellite. Such technologies enabled commanders to gain rapid assessments of the position, movements and strength of enemy forces and therefore to be able to act on a distant opponent eventually almost in real time.28 Analogous developments have occurred in the recent history of scientific research. In astronomy and space research, for example, raw data may be collected from spatially dispersed pieces of instrumentation which are controlled, again almost in real time, by a distant laboratory. I want to conclude with two general remarks. First, the image of science I have given is obviously partial — stressing the association of science with the regulation of social and economic activity over large areas of space. And it is also an image which is peculiarly appropriate to the present. This is a period when an important theme of the European Commission's policies is a concern with the harmonization of environmental regulations, technical standards and professional qualifications — policies which attempt to forge, in their words, a 'European Technological Community' and to promote 'the quality of life' of the community's citizens.29 If measurement has become a central resource for the regulation of space, it has only been so to a great degree since the mid-nineteenth century - the period in which science has become articulated with the moral, political and economic objectives of imperialism; and more recently with those of transnational industry and government. My second remark concerns the question of knowledge and power. As I have suggested, it is the merit of a number of recent approaches to the social study of science that their analyses focus on the relation of power and knowledge, and therefore have the potential to engage with a broader set of debates about the exercise of power in modern societies. But there is a danger that this power/knowledge relation becomes seen to be too close, so that power effects are always associated with the production of knowledge: that knowledge somehow equals power, and vice versa.30 Opening up a space between knowledge and power is both an empirical and a theoretical problem. Theoretically, there is a need to make a distinction between the social and natural world as it is imagined and constructed in political and scientific discourse, and the social and natural world as continually reinterpreted in the everyday course of social activity.31 As de Certeau has pointed out, it is important to analyse not only the apparatuses of discipline but also the 'procedures and ruses of consumers [which] compose the network of an antidiscipline'.32 Empirically, there is a need to recognize that the history of 28 P. Virilio, War and Cinema (tr. P. Camiller), London, 1989. 29 A. Barry 'The European Community and European Government: harmonisation, mobility and space', Economy and Society (1993), 22, 3, 314-26. 30 M. Foucault, 'The question of power', in Foucault Live (tr. J. Johnston), New York, 1986. 31 On this point see, for example, M. de Certeau, The Practice of Everyday Life (tr. S. Rendall), Berkeley and Los Angeles, 1984. 32 Ibid., p. xv. 468 Andrew Barry measurement must be understood as equally a history of the failures of measurement; a history of the phenomena which remain unmeasurable; a history of the incompetencies, and passive resistances of scientists and lay people to the exacting requirements of measurement techniques; a history also of the degree to which the attempts to measure have failed to meet the economic, political and moral demands to which they have been tied.