Glossary

Absence: the necessary Other to presence, which is enacted along with the latter, is constituted with it, and helps to constitute it. In method assem- blage two forms of absence are distinguished. Manifest absence is that which is absent, but recognised as relevant to, or represented in, presence. Absence as Otherness is that which is absent because it is enacted by presence as irrelevant, impossible, or repressed. See also Otherness.

Actor-network theory: an approach to sociotechnical analysis that treats entities and materialities as enacted and relational effects, and explores the configuration and reconfiguration of those relations. Its relationality means that major ontological categories (for instance ‘technology’ and ‘society’, or ‘human’ and ‘non-human’) are treated as effects or outcomes, rather than as explanatory resources. Actor-network theory is widely used as a toolkit in sociotechnical analysis, though it might be better considered as a sensibility to materiality, relationality, and process. Whether it is a theory is doubtful. In the course of its development it has taken a wide range of different and sometimes inconsistent forms. It has at different times been criticised for its relative lack of interest in major social asymmetries such as gender, its refusal to base its explanations on generally accepted ontological categories, its tendency to a centred managerialism, the flattening character of its network metaphor, and its lack of concern with Otherness. The extent to which these complaints are appropriate to either early or contemporary work within the tradition is a matter of judgement.

Allegory: the art of meaning something other than, or in addition to, what is being said. The art of decoding meaning, reading between the literal lines to understand something else or more. The craft of making several things at once, what is described and what can also be read into that description. Ubiquitous, but often repressed into Otherness in contem- porary standard understandings of representation.

Anteriority: out-thereness considered as prior to the process of knowing it.

One of the assumptions made in standard versions of realism.

Condensation: crafted presence that may take a range of material forms.

Constructivism: the claim that scientific statements or truths are constructed in a way that to a large degree (in some versions totally) reflects the social

circumstances of their production. Though there is some overlap, the programme of social constructivism is distinguished from the enactment approach of the method assemblage. Construction usually implies that objects start without fixed identities but that these converge and so gradu- ally become stabilised as singular in the course of practice, negotiation and/or controversy. Enactment does not necessarily imply convergence to singularity, but takes difference and multiplicity to be chronic conditions.

Crafting: the enactment and condensation of presence in method assemblage.

There is no implication that crafting is necessarily a human activity.

Critical realism: a contemporary and politically radical version of realism. Building on the realist suggestion that empirical and experimental investigation is unintelligible in the absence of an external world, and human capacity to intervene in that world and monitor the results of their action, it argues that the world is composed of objects, structures and causal or other powers, and that it is the job of the scholar to offer revisable theories or hypotheses about these. A distinction is made between the empirical (what appears in experience), the actual (actions that occur when powers or structures are activated), and the real (that which is there, those structures and powers, whether or not this is visible or activated). This means that empirical appearances, though important, may be misleading. It also means that the real may or may not be revealed by the actual, and there is no secure way of determining what is real. Distinguishing between the intransitive (roughly such objects of knowledge) and the transitive (the theories or terms used in knowledge), it notes that the transitive is socially located and variable, whereas the intransitive is not. No claims are made about the veracity or authority of the transitive domain, because theories or terms may be refuted and replaced by alternatives. In the terms proposed in this book, realism and critical realism are committed, at least in general, to the singularity, anteriority, independence, and probably to the defi- niteness of the real, as well as its primitive or originary version.

Cyborg: a trope from Donna Haraway’s feminist material semiotics. This is a set of partial connections between two or more parts that cannot be reduced to one another but nonetheless relate to one another. Those parts may be material (between machine and human, or human and animal), political (as between different political or social identities and commit- ments), or they may exist in a tension between reality and fiction. The cyborg is a politically generative trope. It enacts possible novel realities by operating on and within material semiotic relations.

Deconstruction: see post-structuralism.

Deferral: an expression of the post-structuralist proposal that to make present is also, and at the same time, to make absent. Deferral is the removal and effacement of necessary absence into the future.

Definiteness: the assumption that out-thereness or absence is definite in form. One of the assumptions made in standard versions of realism.

Difference, problem of: the simultaneous existence of different objects that

are said to be the same. This arises, as Annemarie Mol shows, because if objects are enacted in practices, and those practices are different, then so too are the objects that they produce, even if the practices in question are said to relate to, or be aspects of, the same object. Problems of co- ordination or separation then arise in the relations between the practices/ objects.

Discourse: in its Foucauldian version, a set of relations of heterogeneous materiality, that recursively produces objects, subjects, knowledges, powers, distributions of power. Discourse is productive. At the same time it sets limits to what is possible or knowable.

Enactment: the claim that relations, and so realities and representations of realities (or more generally, absences and presences) are being endlessly or chronically brought into being in a continuing process of production and reproduction, and have no status, standing, or reality outside those processes. A near synonym for performance, the term is possibly preferable because performance has been widely used in ways that link it either to theatre, or more generally to human conduct.

Ends: see means and ends.

Enlightenment: a philosophically classical commitment to knowledge as the product of reason, empirical inquiry, and as a tool for social improvement. Historically, a period and a movement in eighteenth-century Europe.

Episteme: in Foucault’s archaeology, a set of strategies laid down, permeating and producing the social body, which produce possibilities but also set limits to the conditions of possibility. See also discourse.

Excess: that which cannot be contained within narrative or linguistic discourse, but is probably also necessary to it. A version, or a way of talking about, Otherness.

Fallibilist method: an approach to method that both treats its theories, truth claims or propositions as refutable, and seeks to refute them on the grounds that in the longer run this is the best way to increase the power, scope, or veracity of knowledge. Associated with the work of Karl Popper, and now with realism and critical realism.

Feminist technoscience studies: a diverse body of empirical and theoretical work on the character of technology and science inspired by feminist theory and politics. Major themes or traditions of work include:

1. So-called empiricist feminism which might seek to describe gender inequalities in science and technology.

2. Epistemological critique, which explores the gendering built into scientific method and scientific findings which result from the social shaping of science.

3. Standpoint epistemology, which argues that truth, or at least a workable version of knowledge, is most likely (or indeed only possible) from subordinate viewpoints, and perhaps particularly those of women or feminists.

4. Material semiotics, which explores and seeks as a liberatory project,

to interfere with the relations, simultaneously material and semiotic, that are enacted as partially connected patterns of practice, knowledge, subjectivity, objectivity and domination, by diffracting these in order to make a difference. Material semiotics privileges partial perspective, split vision and situated knowledge, arguing both that there is no escape from location and that identities, locations of knowledge, politics, and action are heterogeneous and irreducible rather than being coherent.

Flux: the sense that whatever is out there is not a structure with a discoverable shape, but is excessively filled with and made in heteromorphic currents, eddies, flows, vortices, unpredictable changes, storms, and with moments of lull and calm.

Fractionality: a metaphor for expressing the idea that objects, subjects and realities (and so their hinterlands) are more than one and less than many. The idea that hinterlands partially intersect with one another in complex ways. A way of avoiding two equally unsatisfactory alternatives: on the one hand the idea that multiplicity and difference imply ontological (and political) pluralism in which there are no interactions between multiples and realities proliferate without restraint, in a version of relativism; and on the other, the converse commitment to ontological singularity in which the world is taken to be singular and consistent.

Gathering: a metaphor like that of bundling in the broader definition of method assemblage. It connotes the process of bringing together, relating, picking, meeting, building up, or flowing together. It is used to find a way of talking about relations without locating these with respect to the normative logics implied in (in)coherence or (in)consistency.

Hinterland: a bundle of indefinitely extending and more or less routinised and costly literary and material relations that include statements about reality and the realities themselves; a hinterland includes inscription devices, and enacts a topography of reality possibilities, impossibilities, and probabilities. A concrete metaphor for absence and presence.

Idealism: see philosophical idealism.

Imaginary: a ‘repertoire by which the world can be re-imagined, and in being re-imagined be re-made’ (Verran).

Indefiniteness: see definiteness.

Independence: a commitment to the idea that whatever is out-there is usually independent of our actions and perceptions.

In-hereness: whatever is made present (for instance a representation or an allegory) that relates to and stands for whatever is made absent but depicted or connoted.

Inscription device: a system (often including though not reducible to a machine) for producing inscriptions or traces out of materials that take other forms. It may be understood as a particular modality for mediating out-thereness and in-hereness.

Interference: the pattern that derives from the intersection of two wave-

forms. In Haraway’s material semiotics, a metaphor for the vision, necessarily split, that replaces representation or mirroring by recognising that it is situated and, indeed, split. At the same time action that makes a political difference. See also cyborg.

Manifest absence: see absence.

Material semiotics: see actor-network theory, feminist technoscience studies, cyborg and interference.

Materialism: see materiality.

Materiality: a way of thinking about the material in which this is treated as a continuously enacted relational effect. The implication is that materials do not exist in and of themselves but are endlessly generated and at least potentially reshaped. This is to be distinguished from materialism which, as the antonym of idealism, claims that what is real is material, and that the ideal is derived from material arrangements. Materiality makes no a priori distinction between the material and the ideal.

Means and ends: a hierarchical organising strategy that enacts and subor- dinates process or practice to the achievement of a valued goal. Therefore a mode in which most continuing processes of enactment are either Othered or are treated as techniques.

Mediation: the process of enacting relations between entities that are, as a part of that process, given form.

Metaphysics: in philosophy, untestable and often implicit assumptions which are enacted in and frame, experience or argument.

Method assemblage: generally, the process of crafting and enacting the necessary boundaries between presence, manifest absence and Otherness. Method assemblage is generative or performative, producing absence and presence. More specifically, it is the crafting or bundling of relations in three parts: (a) whatever is in-here or present (for instance a representation or an object); (b) whatever is absent but also manifest (that is, it can be seen, is described, is manifestly relevant to presence); and (c) whatever is absent but is Other because, while necessary to presence, it is also hidden, repressed or uninteresting. Presence may take the form of depictions (representational and/or allegorical) or objects. Manifest absence may take the form of a reality out-there that is represented, or the relevant context for an object. Method assemblage is distinguished from assemblage in the priority attached to the generation of presence. The definition by itself is symmetrical, telling us nothing about the form taken by presence, absence, or the relations between these. A further provisional definition of method assemblage is offered in Chapter 2. Here it is treated as the enactment of a bundle of ramifying relations that generate representations in-here and represented realities out-there. This is a special case of the more general definition above.

Modalities: conditions or contexts added to statements about reality that in one way or another tend to qualify them, sometimes undermining their authority.

Multiplicity: like difference, the simultaneous enactment of objects in different practices, when those objects that are said to be the same. Hence the claim that there are many realities rather than one. This arises because practices are endlessly variable and differ from one another. The additional claim that practices overlap in many and unpredictable ways, so there are always interferences between different realities. Multiplicity is inconsistent with singularity, but also with pluralism.

Object: a crafted version of condensed presence that takes the form of a process or entity deriving from and re-enacting an ordered form of absence. See method assemblage.

Ontological politics: if realities are enacted, then reality is not in principle fixed or singular, and truth is no longer the only ground for accepting or rejecting a representation. The implication is that there are various possible reasons, including the political, for enacting one kind of reality rather than another, and that these grounds can in some measure be debated. This is ontological politics.

Ontology: the branch of philosophy concerned with what there is, with what reality out-there is composed of.

Ostension: the process of defining a term by pointing to the object or event to which it refers.

Otherness: that which is neither present, nor recognisably or manifestly absent, but which is nevertheless created with, and creative of, presence. More strongly, that which is both necessary to presence, but necessarily pressed into absence or repressed. See also absence.

Out-thereness: the apprehension, common in Euro-American and many other cosmologies, that there is a reality outside or beyond ourselves. This may be specified and strengthened in a number of ways. See: primitive out-thereness; independence; singularity; and definiteness.

Performativity: the claim that words have effects on reality. More generally, the claim that enactments produce realities.

Philosophical idealism: a branch of philosophy which claims that what is real is non-material – for instance taking the form of the ideal or the spiritual – and that the ideal acts to produce the appearance of the material. Pluralism: the idea that views or, more generally, realities, may co-exist in different locations without interfering with one another so long as appropriate ground rules can be put in place to regulate their relations and secure their independence. Hence a version of singularity (since ground rules would need to be shared by all). Therefore to be distinguished from

multiplicity.

Post-structuralism: a middle and late twentieth-century philosophical movement which attacks what it takes to be a metaphysics of presence by arguing that attempts to bring everything to presence (for instance in the form of transparent representation) are flawed. This is because presence necessarily demands absence: the two are created or come into being together. One implication is that however complete representation may

seem to be, it will reveal traces of Otherness, absence, or whatever is necessary to presence that has also been repressed. (Deconstruction is the analysis of texts and other presences to reveal traces of absence or Otherness.) A second implication is that the process of making present also produces that which is Other or absent. A third implication is that whatever is outside presence is unruly and excessive, perhaps to be sensed as flux. A fourth implication is that particular enacted versions of reality set limits to what they are able to know or create. Terms such as ‘discourse’, ‘deferral’ or ‘episteme’ point to such limits. Though the texts of post- structuralists are frequently taken to be abstract and philosophically demanding, many writers associated with or influenced by the approach (though they may resist the label) are also empirical or historical in a relatively straightforward way (for instance Foucault, Latour, Haraway, Mol).

Primitive out-thereness: the sense that there is a reality out there beyond ourselves. No particular claim is made about the character of that reality. Realism: an approach to the philosophy of science that argues that empirical and experimental investigation is unintelligible in the absence of an external world, and human capacity to intervene in that world and monitor

the results of their actions. See also critical realism.

Relativism: the idea that anything is as good as anything else, and there are no grounds for judging between them. This comes in at least three vari- ants. Epistemological relativism says that the knowledge in your culture is just as good as the knowledge in my culture. There are no grounds for claiming that my account of out-thereness is any better than yours. Ethical relativism says that ethics are situated and local, and there are no grounds for claiming that my ethical standards are any better than yours. Political relativism takes the same form again: there are no reasons for preferring my politics over yours. We should live and let live. Relativism is closely related to pluralism, and is well understood as the other to singularity. It is to be distinguished from multiplicity, and the generation of fractionality in practices, where different realities, knowledges, ethics and politics are partially connected and interfere with one another.

Representation: a crafted version of condensed presence that depicts and re-enacts manifest absence, while claiming or implying that its depic- tions are relatively direct expressions of manifest absence. See method assemblage.

Romanticism: in philosophy the idea that the world is so rich that the stories we might tell about it are irreducible either to one another, or (in some cases) to a single set of overall processes at all. The simultaneous claim that it is important not to lose that richness. Historically, a reaction to the rationalism of the Enlightenment.

Singularity: the idea that there are definite, limited, and therefore single, sets of processes in the world, that the world is a single thing.

Stop: a version of deconstruction, in which a smooth narrative that has been

brought to presence displays a break or an interruption that opens up the uncertainties of Otherness.

Symbolic interactionism: a predominantly American tradition in sociology based in the analysis of practice, and treating knowledges and identities as being produced, and irremovable from, particular practices. Strongly influenced by philosophical romanticism, it is relatively sceptical about Enlightenment or classical claims that knowledge can be formalised and transmitted apart from practices and cultures.

Symmetry: the principle that the same kind of explanation or account should be given for all the phenomena to be explained. In the context of science this means that the truth or falsity of scientific ideas should be ignored, and all should be explained in the same general terms. In the present book the principle is applied to method. Method assemblage is a way of thinking about all methods in the same terms, whether or not these fit normative rules about social science method.

Universalism: the idea that true knowledge derives from universal criteria that can and should be applied in all relevant contexts. Hence the idea that true knowledge does not vary between context.

View from nowhere: a way of talking about the idea that we can step outside and so obtain an overview of the world that is detached from any particular location or practice.

Notes

1. The literatures are extensive, and I cannot possibly survey them here. Indicative citations would include Clifford and Marcus (1986) and Haraway (1989) on poetics, Ashmore (1989) and Latour (1996) on reflexive methods and multiple narratives, Haraway (1997) and Rose (2001) on visual methodologies, Butler (1993) and Thrift (2000) on textual and embodied narrative, and Clifford (1997), Hine (2000), Thrift (2000) and Urry (2000) on geographically distributed methods.

2. See Doll and Hill (1950).

3. For an introduction see Nettleton (1995, 160ff).

4. See Klinenberg (2002).

5. For some of the possible complexities, worked out for the example of the UK cervical screening programme (It looks like a success, but is it? If it is a success then how is it so?) see Singleton (1998). I discuss this further in Chapter 5.

6. I will use this term as an index of a more or less hegemonic set of claims about method, notwithstanding the divergences in practice. For an account of its considerable difficulties see Ingold (2000). There are many studies that explore the construction and social correlates of social (and natural) science. I consider the division of labour between truth and politics briefly below (see Shapin and Schaffer (1985) and Haraway (1997)). See also the work by Theodore Porter and Ian Hacking on the contingency of the relations between quantification and scientific (including social scientific) inquiry (Hacking 1990; Porter 1995).

7. The power but also the limits of auditing are considered in Michael Power (1997).

8. This formulation ignores important differences within the STS literatures. Some of these are considered in later chapters.

9. For recent exemplary cases see, for instance, Campbell (1987) in verstehende sociology, Becker (1982) in symbolic interactionism, Said (1991) in postcolo- nialism, Latour (1996; 1998) in (so-called) actor-network theory, and Haraway (1991b) in feminist technoscience studies.

10. Symbolic interaction offers us an exemplary case of an approach to method largely romantic in inspiration which then cut its cloth to fit the much more definite and determinate picture of the world imagined by post-World War II sociology in the United States. Consider, for instance, the assumptions built into the method of grounded theory. For an admirable historical and philo- sophical overview see Rock (1979).

11. It is systematised in this mode in particular by Karl Popper. See Popper (1959).

In a more contemporary context realism and critical realism present themselves as fallibilist methods. See, for instance, the description in Benton and Craib (2001).

12. Latour says similar things about theory when this is imagined as something that can be rapidly displaced with ease. Not so, he says. In practice it takes a huge amount of work. See Latour (1988).

13. The slogan is similar to Paul Feyerabend’s much misunderstood philosophy of science. His commitment to methodological anarchism derives from his assumption that a proliferation of methods would generate the best and most rigorous science. See Feyerabend (1975), and for its translation into social science, Phillips (1973).

14. I draw the notion of entanglement and disentanglement from Michel Callon. See Callon (1998a).

15. See in particular, Knorr Cetina (1981) and Lynch (1985).

16. This account draws on Alpers (1989), Bryson (1983), Law and Benschop (1997) and Rotman (1987).

17. The use of mirrors and optics of all kinds was almost certainly crucial from the fifteenth to the nineteenth centuries in the development of European fine art. See Hockney (2001).

18. See, for instance, the illustration from Jan Vredeman de Vries at http://www. kb.nl/kb/100hoogte/hh-im/hh046.html (from the web page of the Koninklijke Bibliotheek Nationale bibliotheek van Nederland.)

19. As, for instance, in the Annunciation by the Master of the Barberini Panels in the National Gallery of Art in Washington DC. See http://www.nga.gov/cgi- bin/pimage?362+0+0+gg4.

20. Raphael. Marriage of the Virgin. 1504. Oil on panel. Pinacoteca di Brera, Milan, Italy.

21. For instance, ‘idealism’ is an ontology that says that in the first instance there is nothing material. Everything, including the material, is produced by the spirit, the mind, or the process of knowing. ‘Materialism’ is an ontology that says, contrariwise, that everything is material. ‘Spirit’ or ‘mind’ are expressions of the material. The latter is well known in social science, in part through the Marxist tradition. Notoriously, in his historical materialism Marx stood (the idealist) Hegelian dialectic on its head.

22. I write ‘usually’ because we also appreciate that sometimes our actions affect parts of that external reality – and this is especially the case for social reality. Philosophical realists distinguish, for instance, between the transitive and the intransitive. For a convenient survey see Sayer (2000).

23. In Chapter 7 I will consider a cosmology, that of Australian Aborigines, where this appears to be the case.

24. Sometimes, indeed, claims that were previously unqualified may be ‘modalised’ and start to lose their authority.

25. An attractive version of this argument presented in a slightly different idiom is to be found in Collins (1975). I return to this in a later chapter.

26. This is also the case for instruments which work well in one location, but fail to do so in another. For a nice case see Collins (1974).

27. The sociologists of science sometimes call this ‘black-boxing’.

28. See Stengers (1997).

29. See, for instance, Foucault (1970; 1972; 1979).

30. See, for instance, Rose (1999).

31. The point also applies to Latour and Woolgar’s own claims. They too are caught up in (and helping to produce) an obdurate hinterland – which includes the Euro-American common-sense experience that out-thereness is obdurate, anterior and all the rest. Accordingly, their position is internally consistent.

32. The development of quantitative data collection and related tests of signifi- cance are the subject of a considerable literature. See, for instance, Hacking (1990) and Porter (1995). Timekeeping is the subject of a large literature: see the classic Thompson (1967), and for a convenient summary Thrift (1996).

33. See Osborne and Rose (1999) and Law and Urry (2004).

34. This is a mild way of putting what can be a much stronger point. Feminist technoscience studies have in particular pressed for the enactment of interfering research programmes with appropriate theoretical and methodological tools. Donna Haraway’s work on a non-militaristic, non-sexist, non-racist cyborg is particularly well known. See Haraway (1991a).

35. The importance of symmetry was first emphasised in these terms by Bloor (1976), though it is implicit in the work of such historians as Kuhn. I return to the topic more fully at the end of Chapter 5.

36. See, for instance, Deleuze and Guattari (1988) and Deleuze and Parnet (1987).

37. These include (in translation) the following verbs: to fit up, adapt, adjust, reconcile, bring into accord, settle, dispose, arrange, combine, unite, compose, constitute, form, co-ordinate, organise, lay out, prepare, distribute, deal out, chain, tie down, link up, connect, order, array, settle, place, put, set, lay, put out, join together, gather, assemble, muster, collect, bring together, and/or unite. The small French–English dictionary is the Concise Oxford French Dictionary (Chevalley and Chevalley 1963), and the larger French dictionary is the large Robert (1974). I am grateful to Michel Callon for discussion of the difficulties of the term ‘assemblage’ in English.

38. Cooper (1998, 111); and the translator’s introduction to Deleuze and Parnet (1987, xiii).

39. Perhaps it sounds as if it has to do with the action of assembling – for instance as in school, army or prison musters, or perhaps the process of gathering together things on a list, as if one were packing before travelling.

40. Libraries have been written about this, and we need only the sketchiest account here. Good places to start in a review of these debates include: Kuhn (1970), Lakatos and Musgrave (1970) and Barnes (1982).

41. This is a crucial Kuhnian lesson – though it comes from other authors and other literatures too. See, for instance, Polanyi (1958) and Ravetz (1973).

42. This has several radical implications. One is that since there are scientific revo- lutions, discontinuities in the history of science, it is not so very easy to show that science progresses. Perhaps it merely changes. Indeed Kuhn got into a lot of trouble with his critics because he claimed that since scientific revolutions are discontinuities this means that science itself advances discontinuously. Thus most previous accounts of scientific change assumed that in general and over time science increased its predictive power, the scope of its theories, and its empirical base. This argument was made in a variety of different ways, but usually assumed that science created generalisations of increasing power and parsimony, and/or falsified those that turned out to be empirically inadequate. But this (usually) implies some kind of empirical yardstick for measuring the

scope of scientific theories. Sure, scientists – or whole groups of scientists – might get hold of the wrong end of the stick, and fool themselves into thinking they’d discovered phenomena that weren’t actually there. But overall, and in the long run, it was assumed that good observation would out, so long as the process of inquiry was disentangled from the malign effects of political and economic interference. On falsification, see Popper (1959).

43. Thus it turns out that if patients start regular walking under the appropriate supervision of physiotherapists, many report that the onset of pain is increas- ingly delayed, and sometimes it is not necessary to operate at all. See Law and Mol (2002).

44. Interestingly, when walking therapy works (which it usually does only with the support and discipline of physiotherapy) it does not appear to reduce stenoses. So why does it work? Perhaps it opens up alternative ‘collateral’ vessels which bypass the diseased arteries. Perhaps it alters the biochemistry of the blood. No one actually knows.

45. The Salk scientists do too, in practice. As we have seen, they live in uncertainty. But unlike the medical professionals, they set themselves the convergent goal of determining a single reality.

46. The approach is common in the sociology of scientific knowledge (SSK). See, for instance, Collins (1975; 1981a).

47. He brilliantly developed this through a series of studies, starting with Goffman (1971).

48. As this suggests, the turn to the performativity of enactment has been a powerful if not dominant force in a number of theoretical traditions for several decades. It would be possible to write a genealogy of this opening as it has struggled first to imagine (or enact?), and then to come to terms with, the epistemological, ontological and theoretical implications of the idea that the real is enacted in practices, rather than being reflected through them, as it is in perspectivalism. Louis Althusser (1971) works uncomfortably in the space defined by these two possibilities. Michel Foucault (1979; 1981) is much clearer about the performativity of discourse, as are STS writers such as Latour and Woolgar, and such feminist theorists as Judith Butler (1993) and Donna Haraway (1997). Any serious attempt to imagine the performativity of enact- ment also has to handle the related question of materiality or ‘the real’ and its relations with discourses or other linguistic expressions. In Foucault, discourse extends into and is carried through certain kinds of materials – an opening explored more fully for the case of embodiment by Butler.

49. For a more extended discussion see Law (2002a).

50. For further discussion of non-coherent hierarchies or (more generally) intransitive relations see Law (2000).

51. This argument is developed at greater length in Law (2002a).

52. For critical radical commentary on such identity politics see, for instance, Haraway (1991a) and Harvey (1993).

53. In his book We Have Never Been Modern, Latour (1993) argues that what is sometimes called modernity is productive precisely because it insists on purity. It insists, for instance, that things have single and definite shapes, or that natural realities are clear and quite distinct from those of the social. His argument is not that that modernity actually achieves this purity. Rather it is that by imagining reality to be pure it allows the fecund production of

impurities – swarms of heterogeneous multiplicities. Latour argues that we have never been modern. We just think that we are.

54. See Lacqueur (1990).

55. The argument is developed in Mol (2002), Hirschauer (1998) and Hirschauer and Mol (1995). Hirschauer’s work attends to the issues for sex/gender as these arise for transsexuals, a difficult context far from the abstractions of theory. It is also developed, in a slightly different idiom, in Cussins (1998a; 1998b; 1998c).

56 See Haraway (1991b, 194–195) and (1997).

57. For commentary see Moser (2000).

58. For details of those publications see Law and Singleton (2003; forthcoming).

59. In order to preserve anonymity, all proper names and locations in what follows are pseudonyms, other than those of national organisations.

60. Dr Warrington, a consultant gastro-enterologist, was interviewed on 19 March 1999. Quotations are reconstructed from notes.

61. Dr Warrington, 19 March 1999.

62. Dr Warrington, 19 March 1999.

63. The quote is not from Sister Fraser but her senior colleague, a Nursing Officer. Interview on 10 December 1998.

64. Sister Fraser was interviewed on 10 March 1999. Quotations are reconstructed from notes.

65. Sister Hart was interviewed on 3 March 1999. Quotations are reconstructed from notes.

66. Dr Bowland was interviewed on 11 June1999. Quotations are reconstructed from notes.

67. For a fine study of the distribution of responsibility between individuals and social arrangements, and the individualisation of responsibility, see the related case of the ‘problem’ of drink driving by Gusfield (1981).

68. This is an instance where the purification described by Latour and discussed in the last chapter seemed to impede the proliferation of impure forms.

69. Excerpt from letter sent to the initiating hospital consultant dated 7 October 1998.

70. From interview notes with the staff at Castle Street Centre, Sandside, 10 June 1999.

71. Interview, 17 June 1999.

72. Related metaphors for fluid objects have been developed in a number of other contexts. See, for instance, Mol and Law (1994), Law and Mol (2001) and Law (2002c).

73. And there are other analyses that have a similar shape. See, for instance, my own account of technological decision making briefly discussed above, and more fully in Chapter 7 of Law (2002a).

74. Materiality, not materialism, since the argument is not reductionist.

75. See, for instance, Butler (1993).

76 Cussins (1998a; 1998b; 1998c); Moser (2000; 2003); Moser and Law (1998;

1999; 2003); Moreira (2000; 2001a; 2001b).

77. This relational metaphysics is laid out systematically in Latour (1988) and (1998). For further commentary see Law and Mol (1995).

78. Routine is what gets hidden because whatever is in front of it (presence and manifest absence) includes it and hides it. In STS this is sometimes known as black-boxing (Rip 1986). Examples such as the workings of a personal

computer, hidden while all goes well, explain why the black-box metaphor is appealing. Insignificance is not so different but is less discrete. Repression indexes a lively and important tradition running from Freud through versions of post-structuralism (for instance in the writing of Lacan and Lyotard) to a range of radical interventions in cultural studies that have often explored how subordinates (for instance blacks or women) are Othered to produce versions of white male superordination. See, for instance, Hall (1992), Said (1991), and Haraway (1989).

77. For representation, on the face of it to talk about something other than what one is talking about is at best roundabout, perhaps a metaphorical flourish, and at worst it is simply misleading.

78. These skills, to be sure, work the other way round. The powerful treat the representations of the less powerful allegorically too, doubting, cross- examining, checking and auditing. Trust is in short supply in both directions. On the self-defeating character of the audit process, which can be understood as a futile attempt by the powerful to convert allegory into representation, see Power (1997).

79. Though arguably sociology has been important in the enactment of the social. For hints to this effect see Porter (1995), Osborne and Rose (1999) and Law and Urry (2004).

80. For discussion of economics see Callon (1998b).

81. Those not caught up directly in the relations enacted in these claims do not necessarily take those claims at face value. Remember, however, that economic realities are not simply statements but are also relations that extend into practices and materials that ramify off in all directions. ‘Belief’ is not usually what is at stake.

82. Their worry is no doubt compounded by the suspicion that lack of public support affects the depth to which public bodies are willing to reach into their pockets to fund scientific research. Thus, surely, is one of the explanations for the so-called ‘science wars’ controversies in which social scientists have been accused of undermining the epistemological foundations of natural science.

83. The Minister in question, John Selwyn Gummer, was responding to fears about new-form Creutzfeldt-Jakob Disease at the beginning of the BSE scare in the UK, in a dramatic televisual attempt to persuade the nation that beef was indeed safe. This took place on 6 May 1990 (see http://news.bbc.co.uk/1/hi/uk/ 369625.stm). It is appropriate to note that his politics are now much greener than they were at that time.

84. These examples together with the larger argument about the public under- standing of science are drawn from Brian Wynne’s work. See, for instance, Wynne (1996).

85. See Singleton and Michael (1993), and Singleton (1996; 1998).

86. This is an argument that she has developed in work on the UK campaign to reduce death from sudden infant death syndrome, or SIDS. Official statistics suggest that the campaign has been remarkably successful, following a ‘back to sleep’ campaign to persuade mothers to place their infants on their backs before they go to sleep. Singleton’s data reveal that this injunction is interpreted and enacted in many different ways in practice.

87. An argument like this is developed by Frederic Jameson in his analysis of the ‘post-modern’ architectures of San Francisco, which, he argues, may be

understood as tools for what he calls ‘cognitive mapping’ that are appropriate to the non-coherent but global realities of capitalism. See Jameson (1991). For further discussion of ‘knowing in tension’, see Law (1998).

77. Ladbroke Grove Rail Inquiry, Transcript of Proceedings, morning of 10 May 2000, page 5. This was available at http://www.lgri.org.uk/10mayam.htm, now saved at http://www.archive.com.

78. The details are drawn from the following documents: Second Health and Safety Executive Interim Report, ‘Train Accident at Ladbroke Grove Junction, 5 October 1999’, 3 November 1999, http://www.hse.gov.uk/railway/paddrail/ interim2.htm; Third Health and Safety Executive Interim Report, ‘Train Accident at Ladbroke Grove Junction, 5 October 1999’, 14 April 2000, http://www.hse.gov.uk/railway/paddrail/interim3.htm; Ladbroke Grove Rail Inquiry, Transcript of Proceedings, morning of 10 May 2000, formerly at http://www.lgri.org.uk/10mayam.htm (now saved at http://www.archive. com).

79. Many of the details including transcripts of the inquiry were available at http://www.lgri.org.uk. Regrettably, this website was closed in the summer of 2002, though most of the pages can be retrieved from the excellent facility at http://www.archive.com. See also Cullen (2001).

80. Much of the debate and cross-examination in the inquiry concerned the relative significance or plausibility of different possible causes. A straightforwardly allegorical reading of many of these interventions is irresistible. The protag- onists were trying to ensure, as plausibly as they possibly could, that important contributory causes did not end up in their own backyard.

81. This is from Cullen (2001, 7) and is the terms of reference for Part 1 of that Inquiry. ‘HSE’ is the acronym for the Health and Safety Executive.

82. An issue about which I have learned much in discussion with Ingunn Moser who writes on disability and interferes in disability politics. See Moser and Law (1998; 1999); Moser (2000; 2003).

83. Thus Haraway’s account of the cyborg is similarly allegorical, as is the concern with absence, Otherness, and intertextuality witnessed in the heritage of Foucault and Lacan, cultural and postcolonial studies and parts of feminist theory.

84. See Chapter 6.

85. Barnes distinguishes between a legitimate interest in the prediction and control of nature, and an illegitimate and concealed interest in social control and rationalisation. Science is generally, he says, and preferably, under the direction of the former. However, even the latter may produce cultural forms that are relevant to natural prediction and control. The origins of knowledge tell us nothing about its utility and validity. For this reason he does not distinguish between ‘ideology’ and ‘knowledge’, but talks instead of ‘ideological determination’. See Barnes (1977).

86. The sociologists of scientific knowledge were here following a line of argument that has its hinterland in both social anthropology and the verstehende tradition in sociology.

87. This ethnography is reported more fully in Law (1994).

88. This argument has been elaborated into a much larger metaphysics in Lawson (2001), which, however, does not entertain the divergent possibilities of difference, multiplicity and fractionality.

89. For references see Pinch (1980; 1981; 1985).

90. See, for instance, the work of Knorr Cetina (1999), Pickering (1995) and Traweek (1988).

91. There is a tradition in the philosophy of science that formalises this. See Hesse (1963; 1974). The empirical studies in the sociology of scientific knowledge also show that what counts as ‘right’ and ‘wrong’ is often, perhaps always, negotiable – though, as Latour and Woolgar suggest, this may become so expensive as to be impossible. Collins’s work is related to that of Latour and Woolgar, but there are also differences. Collins is particularly keen to show that descriptions of reality are located and grounded in cultures or forms of life, and is happy to describe himself as a relativist, a claim carefully avoided by Latour and Woolgar. The differences are debated in Collins and Yearley (1992) and Callon and Latour (1992).

92. This is a distinction that crops up in different but somewhat related ways in a range of different literatures. It resonates, for instance, with the distinction between classical and romantic thought described by Alvin Gouldner (1973). A similar theme is explored by Karl Mannheim in his essay on conservative thought (1953). Rather differently, Mary Douglas’s anthropology distin- guishes between more bureaucratic or ritualised settings, and those that are more entrepreneurial. See Douglas (1982). The present book, as I suggested in an earlier chapter, locates itself in a similar divide.

93. The study is reported more fully in Law (1994).

94. The numberings for Christian Faith and Practice refer to paragraphs, not pages.

95. See Pickering (1995).

96. See Shapin and Schaffer (1985), and Shapin (1994), together with Shapin (1989).

97. For further commentary on this see Haraway (1997).

98. See Alpers (1989), Haraway (1991b).

99. Additional references come from Robert Layton (1989).

100. See Kerle (1995, 136); the details are discussed more fully in Ayre (2002).

101. See Chatwin (1998).

102. Verran develops her argument so: ‘the beginning of a galtha workshop emphasizes the multiplicity of practitioner groups and their differing contributions to the necessarily messy reality of a place, the opening scenario of a science practical hides differences between the many and varied practitioner groups that constitute the environmental sciences, invoking instead a virtual, singular place. These different assessments then switch when we proceed to what is actually done during the workshop. While for Yolngu it is important that multiple possible “doings” be channeled into one communal act of place, scientists need to perform, report, and make known a multiplicity of actual doings. I am suggesting, then, that the normal ontologies of these two knowledge traditions advance different ways of managing the multiplicity/ singularity tension that comes with doing any ontology of place’ (Verran 2002, 165).

103. Aboriginal people originally congregated in stations as a means of living acceptably in or close to their own country. In the 1970s a law was passed which required White station owners to pay ‘award wages’ (minimum wages) to Aborigines for their work. This led the station owners to turn Aborigines off their lands and into mission settlements.

104. The implicit reference is to Latour (1993).

105. For an entertaining and partially fictional essay which explores this (and much else) see Julian Barnes (1990).

106. For further discussion of the ‘imaginary’ see Verran (2001; forthcoming).

107. For details see, for instance, Verran (1998) and Sharp (1996).

108. See Margaret Ayre’s remarkable study (2002) of nature conservation and management in East Arnhemland. And David Turnbull’s account (forth- coming).

109. See, for instance, Baskhar (1979), and for recent accounts in the context of social science, Sayer (2000) and Benton and Craib (2001).

110. Realists refer to this as the ‘transitive’ dimension of inquiry, in contrast with ‘intransitive’ natural phenomena.

111. For an account of the difficulties of tightly integrated systems see Charles Perrow’s exemplary text, Normal Accidents (1984). For further discussion of non-coherent coherence see Singleton and Michael (1993), Singleton (1998) and Law (2002a).

112. It is the emphasis on presence that distinguishes method assemblage.

113. It may be that this repressed multiplicity is necessary to achieve the appearance of singularity, though under certain circumstances the contrary argument can also be made.

114. As we have seen, the argument is developed by Latour and Woolgar. But see, also, Latour (1990).

115. David Turnbull’s exemplary work on cartography deserves careful study. See Turnbull (1993; 1996; 2000).

116. Here I am commenting on academic or other forms of writing that seek to describe realities. As is obvious, the argument does not necessarily apply in this form to non-referential forms of writing such as novels or poetry.

117. We have encountered it, for instance, in the writing of Donna Haraway. See Haraway (1991a; 1991b; 1997; 2003).

118. There are interesting accounts (of particular versions of aesthetics) in the physics described by Traweek. See Traweek (1988; 1999), and different aesthetic styles are implied in Turkle’s work on computer use. See Turkle (1996).

119. Donna Haraway’s recent work on people and dogs as companion species, though written in a very different idiom, makes an argument that is connected to this. See Haraway (2003).

120. This is the last line of her ‘Cyborg Manifesto’. See Haraway (1991a, 181). Her argument is (necessarily) situated, in part by her erotic and political commitment to a refigured version of science in a context where it was easy to see science as inhumane and fundamentally flawed. For a further part of the relevant feminist political and spiritual context see Starhawk (1989).

121. It is also the case that symmetry is always a moving target. Thus the argument from symmetry assumes that everything can be made manifest. But Othering is a limitless domain. Only particular assumptions can be made manifest. The issue, then, is one of openness or attitude to the hidden realities of Othering, rather than enumerating a complete list of repressed asymmetries.

122. The argument is developed by such writers as Latour. See, for instance, Latour (1987).