1
Valuing and validating

On the 'success' of interdisciplinary research

Mike Michael

Introduction

Although in this collection we have asked authors to discuss interdisciplinary methodology through particular verbs, one cannot help but, or one is tempted to, touch also upon nouns. Specifically, the nouns that attach to the verbs valuing and validating can be presented as contrasting pairs which connote rather different practices, interests and arrangements. Thus, for Valuing we can compare value and valuation, and for Validating we can compare validity and validation. At base, the former term of each pair implies an ‘external’ or ‘objective’ relation by which some thing or some activity is judged against pre-given standards: value against market forces, validity against scientific or epistemological criteria. Conversely, the latter term of each pair – valuation and validation – points to an intersubjective or interactional relation in which some thing or some activity is assessed by means of standards and criteria. Thus, validation, say of a person’s work within an organization, might entail shared negotiation of criteria as to what is of worth; and valuation of a property might involve the situated weighting and juggling of criteria in the process of moving through a house, say, to come to a proposed money figure.

Of course, this contrast is hardly absolute. As social scientific studies of audit culture and standardization have often noted (for example, Bowker and Star 1999; Power 1999), external criteria are mobilized in the process of locally negotiated valuing, and the ‘fruitful’ application of external criteria often entails their situated negotiation. Nevertheless, this contrast hopefully sets up a frame to address the complex, involutionary processes that enter into the doing of Valuing and Validating. At the very least it should hint at what we might call a ‘topology of valuing’ in which ‘valuing’ and ‘validating’ in relation to interdisciplinary methods operate at several interwoven, intersecting and interacting levels. Such a topology might simultaneously entail institutional encouragement, systemic devaluation, ontological invention, professional and political accountability, and embeddedness within genealogies of practice (and much more, as we shall see). It should be obvious that I will not be dealing with more formal methodological accounts, especially those concerning variants of validity, such as scale, probability or reliability. There are plenty of text-book definitions and applications in the literature. In keeping with the interdisciplinary ethos of the volume (and as hinted in the foregoing), I wish to engage with ‘validity’ as a complex notion that can connote the more or less successful accomplishment of some form of agreement about the ‘facts of the matter’ that emerges in different ways in the process of doing interdisciplinary work.

In what follows, I intend to discuss these issues in a little more detail by thinking them through what has been called translational research. As a means by which research in one discipline is transferred to another which then translates it into practical intervention (say the movement of laboratory-based research into clinical application), prima facie, translational research looks like a relatively uncontentious site. After all, it would seem that here interdisciplinarity functions with a ‘lightness of touch’, given, on the one hand, the absence of a need to fashion new interdisciplinary tools, and, on the other, the existence of infrastructural supports (in co-habitation within research centres, governmental backing and financial resourcing).

Despite a willingness to trial alternative approaches, however, things do not run so smoothly, as rather different values – epistemic, cultural, educational – are in place, which militate against even this apparently ‘easy’ form of interdisciplinary collaboration. Nevertheless, some groups do ‘hold’ and below I examine how this might work through the redistribution of value, or a re-patterning of valuing. This examination will serve as a basis for an expanded discussion of the complex role of valuing and validating that draws directly on the chapters that comprise this section. As we shall see, valuing and validating range across a number of concerns: data, participants, project, case, research question, methods, profession and institution. They also function on a number of registers – epistemic, ontological, inter-personal, institutional, temporal – not all of which are easy to reconcile. The aim of this section, therefore, is primarily to illustrate the complex interweaving of valuing and validating within a range of interdisciplinary practices and projects, rather than to provide a systematized approach. To the extent that there is a common theme across these illustrative cases, it concerns the ‘matter of success’ which is taken up in the concluding remarks. What can count as a ‘successful’ interdisciplinary project, finding, team or researcher?

On the valuing and validating in translational research

Translational research (also routinely referred to as ‘bench to bedside’ research within the medical field) is the term applied to the movement or translation of ‘basic research’ (often produced in the laboratory) into the clinical domain where it can serve as the basis for innovations that are therapeutically, diagnostically and/or preventively useful. According to Watts (2010), translational research has become a key policy priority, shooting ‘up the (UK medical research) agenda’ (n.p.) and being championed by the likes of the Office for Strategic Co-ordination of Health Research, Medical Research Council, and the Academy of Medical Science. It is also central to the funding initiatives of the European Commission’s 7th Framework Programme and US National Institutes of Health (the latter calling for bids for Centres of Excellence of Translational Research).1

However, despite all the enthusiasm for this useful translation of knowledge across disciplines, there are underlying problems, not least with regard to the ways in which the different disciplines envision and enact such fundamentals as ‘knowledge’ and ‘scientific rigour’. As various authors have documented, these divergences hinge in large part on the values that pertain within the respective ‘epistemic cultures’ (Knorr Cetina 1999) of, in this case, lab sciences such as embryonic stem cell research and clinical sciences such as treatment of Parkinson’s disease (for example, Cribb, Wainwright, Williams, Farsides and Michael 2008; Wainwright, Williams, Michael, Farsides and Cribb 2006; Wilson-Kovacs and Hauskeller 2012). Thus, Michael, Wainwright and Williams (2005) traced how, for instance, lab scientists derogated the ‘experimental techniques’ of clinicians (who would often work with a few cases rather than extended sampling), while the clinicians derogated what they saw as the lab scientists’ obsession with scientific rigour at the expense of the immediate needs of patients.

In these two research communities – those of the lab and the clinic – we might say there is a valuing of different values: ‘proper experimental procedures’ versus ‘commitment to patient well-being’. At the same time this distinction also determines a validation of certain sorts of cultural practices. For instance, within lab science culture, what is most valued is scientific excellence, and academic ‘voice’ is attended to on the basis of the quality of one’s science. By contrast to the meritocratic culture of the lab, within clinical science, hierarchy on the basis of seniority (and presumably ‘experience’) structures who can have voice. It is not surprising, then, that when a research centre comprised of both clinical and laboratory scientific teams have research meetings, attendance is affected by the disciplinary type of research that is being presented (Brosnan and Michael 2014). Not only is the substance of a topic in need of translation (determining the ‘value’ and ‘validity’ of this or that finding), but so too is the form of discussion and debate (who is ‘permitted’ to have voice).

We can approach this divergence in terms of Annemarie Mol’s discussion of ontological politics. In The Body Multiple (2002), she argues that a disease such as atherosclerosis manifests itself in different ways depending on the practices that enter into its enactment. Atherosclerosis is, therefore, something different when studied microscopically by a pathologist as compared to when a physician manually gauges the temperature of a patient’s feet. What atherosclerosis ‘is’ differs across these two cases. As Mol (1999) notes these different versions do not necessarily need to ‘cohere’ – though practically and routinely they do, simply because a patient must be treated. Sometimes, however, they can diverge and create tensions, sometimes co-exist in parallel quite happily; sometimes they can be coordinated and managed.

What this suggests is that values are ‘emergent’ and, by extension, ‘valuing’ is contingent. Values do not sit ‘above’ practices, animating them, so to speak. Rather, what is of value within and across specialisms or disciplines can vary with the practical issues at stake. Sometimes this will revolve around the immediacy of patient needs; sometimes it will attach to the work regimes of different hospital or disciplinary specialisms. In the case of translational research, the divergent, multiple ontologies of clinical and lab practitioners seem to be managed in a variety of ways. For instance, neglecting writings from ‘other’ disciplines can be put down to sheer lack of time (the arduousness of lab work means that one must read instrumentally, and that means one does not have the luxury of reading work from more distant disciplines). However, this does not necessarily imply an unbreachable divide between the ontologies of lab and clinic. Brosnan and Michael (2014) note that researchers themselves see the eventual ‘calibration’ of lab and clinical ontologies taking place in the future. This ‘promise’ of future ontological articulation is embodied in the research Group Leader who straddles lab and clinical practice. It is the Group Leader who, by virtue of their vision of prospective translation, symbolizes the continued ‘adhesion’ of the research group. As Brosnan and Michael suggest, this future, yet present, conception of the research community can be seen as an instance of Barry, Born and Weszkalnys’s (2008) interdisciplinary logic of accountability, namely ‘the idea that (interdisciplinarity) helps to foster a culture of accountability, breaking down the barriers between science and society, leading to greater interaction, for instance, between scientists and various publics and stakeholders’ (Barry et al. 2008: 31). In the present case, this translates into the processes by which scientists and clinicians enact themselves as accountable to one another. The former perform their work as relevant to clinical problems (as identified by the clinicians); the latter make their clinical work accountable to the lab scientists who can provide the underlying organic bases for that clinical work. But this accountability is ‘done’ in a somewhat diffused way, and, crucially, such accountability is enacted via the figure of the Group Leader.

The point of this example from medical research is to show how in the various modes of articulation across disciplines touched upon here, valuing and validating play complex, shifting and contingent roles. From this albeit brief discussion of translational research, we can tentatively derive the following pattern: there is, on the one hand, a valuing of scientific rigour, practical application and (future or promissory) interdisciplinary collaboration, and, on the other, a validating of one’s own epistemic culture, research practice, research relations and research leader. Embroiled in all this are mediations of ostensibly abstracted scientific (for example, sample size, use of control groups) and moral (for example, the imperative to find ways of alleviating the suffering of humans) values that might or might not govern the prevailing norms of the research team. In the next section, we will pursue these involutions further by considering the insights provided by the chapters in this section.

On the varieties of valuing and validating

In relation to the contributions to this section, we find that the meanings of valuing and validating have proliferated extravagantly. In a quick summary of these many relations in interdisciplinary work we can crudely abstract the following:

Needless to say, I cannot in this introduction do full justice to the contributions that follow. However, the point I wish to pursue concerns the variety of valuings and validatings that is evidenced in these entries. Rather tentatively, we can re-group them under the headings: ‘Doing data/object of research’, ‘Accounting/practice’, and ‘Relations/reproductions/repairs’. While these are certainly not empirically mutually exclusive, they can heuristically serve to highlight different moments of valuing and validating. Provisionally, and with some laxity, we might say that: Elder and Potskin, Hawkins, Graham, Driesser, and Latimer and Munro fall under the auspices of ‘Doing data/object of research’; under the label ‘Accounting/practice’ can be included Akama and Pink, Gisler, DiSalvo, Tironi, Wilkie, Galloway, and Irwin and Horst; ‘Relations/reproductions/repairs’ encompasses the work of Michael, Fukushima, and Calvert. I do not address all these studies but simply draw on those that seem to be most useful in exploring these aspects of valuing and validating.

'Doing data/object of research'

‘Doing data’ addresses the values entertained in the derivation and generation of data. Gay Hawkins speaks of how her object of study, ‘bottled water’ – as a complexly re-qualified and re-qualifying, multiple reality-provoking entity – necessitated a range of methods. That is, the object somehow validates the methods that are used. In Stengers and Latour’s (2015) introductory account of Souriau’s ‘different modes of existence’, an object of art (or any object that is made, for that matter) places a demand on its maker to ‘work it out’. But this is a fraught process, filled with the possibility of error and failure. Bottled water is just such an object that, in order to be made to exist properly – or successfully – demands that its (social scientific) makers apply their skills, technique and capacities, but always hesitantly, carefully, so that the data (about plastics, water, contaminants?) thus derived ‘successfully’ make that object.

Conversely, the ‘work’ is not simply crafted by its maker, but also, in return, crafts its maker (see also Sybille Lammes, this volume). In Joanna Latimer and Rolland Munro’s chapter, it is the object (or event) apprehended as an ethnographic moment that ‘makes’ the researcher – that asks questions of her. Specifically, this singular case of an ethnographic moment precipitates a bringing to the fore, and re-valuation, of one’s theoretical and analytic situation such that one sees anew that object, along with its conditions of possibility. In this process, the researcher is herself reconfigured, and made to ‘work successfully’. However, we need to expand this image of a single work confronted with its individual maker (and vice versa). For each carry with them a pattern of pasts and prospects. For Connor Graham, in researching the potential uses of ICTs (or in crafting an account of such uses), it is important to set these within an array of (among other things) past futures of ICTs and past projects researching such futures. In other words, on our present reading, the ‘success’ of the object of a research study, and the researcher themselves need to be situated in relation to the past successes (and failures) of ICT futures and studies of such futures. Finally, for Tuur Driesser, we also need to address the value of the research in terms of its processes of exemplification – that is, in relation to the ‘example’ per se. Accordingly, the example does not relate to the general unproblematically; rather, they need to be situated in ambiguous spatial and temporal relation to one another, not least when the example – what can count as an example – is itself discipline-bound.

In taking all these accounts together, there seems to be a valuing of something that approximates the ‘success’ of a research study. However, this hard-won success points in several directions at once – to the ‘object’ of the study, the researcher, the histories of researcher and researched, the examples and cases that ‘make up’ the research. Indeed, we might say that any perceived or actual success is validated by the configuring of these different elements. But even if a particular study does emerge, or conclude, successfully, it often tells us little about how to proceed in relation to future ‘objects of research’. Each object of study, according to these contributions, makes its own peculiar demands upon research in relation to ‘being worked out’.

What are the implications of this account for the doing of interdisciplinary research? On the face of it, these are pessimistic: the object is ‘worked out’ through distinct and discrete disciplinary techniques that are generative of distinct ‘works’ (what above was called multiple ontologies). However, we can pause and reflect here on the fact that a work of art, in order to be ‘successfully worked out’, might ‘call on’ the artist to pull in techniques from multiple disciplines – painting, sculpting, video, performance, for instance. In a parallel way, in order to be ‘successfully worked out’, the object of research might ‘demand’ a range of methods and techniques that normally sit in different disciplines. This raises, for present purposes, the issues of what counts as ‘success’ in general, and what counts as success for an interdisciplinary object in particular – what then is the relationship between ontology/ies and the heuristic? We will approach these concerns through a discussion of, respectively, ‘accounting/practice’ and ‘relations/reproductions/repairs’.

'Accounting/practice'

To propose that a research object demands certain (combinations of) methods to be ‘successfully worked out’ is to connote a sense of completion. If it has succeeded then the object reaches a form of closure or triumph or attainment; and we have findings or claims to be delivered. And yet, some of the contributors to this section seek anything but closure when they judge the value of their own research. By contrast, success is associated with openness, tension, doubt, uncertainty, different and better problems. Thus Manuel Tironi speculates on the value of dissent (that never materialized) among different practitioners (for example, architects, social scientists, government officials) in opening up the possibility of democratic potential; Yoko Akama and Sarah Pink describe a range of methods for surprising and disorienting researcher-participants about their respective objects of research, by introducing uncertainty and possibility into those objects; and Alan Irwin and Maja Horst, insist on the need to attune to the problem of many problematizations (related to different disciplinary specialisms) which as a meta-problematization, as it were, may be productive (leading to a useful questioning of presuppositions) or destructive (leading to tension and impasse); and Anne Galloway, in developing, disseminating, and decanting responses to her fiction of the PermaLamb, cuts across design and ethnography in order, in Haraway’s (2016) phrase, to ‘stay with the trouble’. In these four examples, we might detect a value placed on something we could call ‘a-success’ in which success is grounded in ‘unsuccess’, unity in disunity.

Put another way, there is a series of general collectively enacted disruptive practices in play in these three examples: dissent, disorientation and meta-problematization. All these could be discounted as a negative upshot of interdisciplinary collaboration, however, what we witness are accounts which (also) valorize these disruptions in terms of their potential productivity. The suggested notion of a-success not only connotes the utility of disruption, but also the lack of certainty that can be attached to such utility. The productivity of disruption remains potential – there is no guarantee (as Irwin and Horst are all too keenly aware).

We can tentatively draw out some more general implications of this discussion by considering the practice of triangulation. Triangulation routinely refers to the deployment of different methods that aim to derive a variety of data about a given topic (for example, interviews and observations, or interviews and survey data) that, by virtue of corroborating one another, serve to increase the likelihood of validity of the findings (Denzin 1978). However, Silverman (2006) –among others – has been highly critical of this technique not least because the different methods entail different theoretical backgrounds and different contexts of application, which renders suspect the comparability of their data. However, it might well be more fruitful to explore the productivity of their differences: what does a disparity in interview and biographical data and non-participant observation tell us about, say, a scientist’s characterization of science? Where Merton’s (1973[1942]) biography-based work derived the norms of science (which stressed communality universality, disinterestedness and organized scepticism), Mulkay’s (1979) observational approach noted the lack of relevance of norm-following in the reward of scientists (rewards, such as further funding, were based on the usefulness of scientists’ findings rather than on their adherence to norms). In the process, the very notion of norms was challenged. Even for the scientists there was a value in the opening up of the problem of norms – which they, in light of the triangulation across Merton and Mulkay, either continue to see as structurally important (Merton; also see Mitroff 1974) or as ‘ideological’ (according to Mulkay), or waver between these statuses. In sum, the practice of triangulation is accounted for in terms of opening up, of problematization; and the validity of triangulation lies less in verification or consolidation and more in its value as a means of generating problems.

'Relations/reproductions/repairs'

In the previous two parts of our segmentation of methods of valuing and validating, we have focused on the complexities of interdisciplinary success by focusing, respectively, on the object of research itself, and on the types of accounting for success (or, specifically, a-success). In this final part of the introduction, we turn to a consideration of the relations among research practitioners drawn from different disciplines. How are these relations engendered, reproduced and repaired? We have already seen in the discussion of translational research that such relations can be very loosely woven: in the above example, the relations between laboratory and clinical researchers were essentially vicarious, being mediated by the Group Leader. Nevertheless, this yielded success insofar as the research group itself enjoyed longevity, and did not break apart. This suggests a strongly social configuration of methodological differences, and so these next contributors explore some more features of interdisciplinary relations.

For Jane Calvert, the management of difference among collaborating social scientists, designers and synthetic biologists was facilitated by an emergent metaphor, that of ‘the wedge in the door’. This evoked an opening (albeit fragile) through which dialogue among different researchers could potentially proceed. Mike Michael addresses some of the more overtly affective dimensions of interdisciplinary relations. Here ‘compromise’ among interdisciplinary colleagues is possible because of a certain ‘care’ toward one another (which might be based in socializing and liking) where there is a sort of tolerance of, or indulgence toward, each other’s disciplinary commitments and quirks. By comparison, as Masato Fukushima tellingly warns, successful interdisciplinarity might also be affected by the sheer volume of participant researchers, that is, by the number and density of relations. As he notes, in addition to the problems of managing and mediating cooperation across a multitude of researchers, there are tensions born of the simultaneous push for larger research teams and the chronic and systemic pursuit of individualized recognition. If Calvert and Michael hint at the ways in which researchers from different disciplines might validate each other, Masato problematizes the very possibility of such validation where teams are large and dispersed.

However, lest we think that Calvert’s and Michael’s chapters simply point to means of ‘successful collaboration’, we can take note of the following. Calvert describes a metaphor that serves to render broken or problematic relations ‘reparable’; indeed, even if difficulties persevere/ persist, there is an in-principle possibility of future collaboration. One might say that a variation of an ‘imagined community’ (Anderson 1983) seems to be at play here, in which the various members of an ‘interdisciplinary community’ remain prospectively inter-linked by virtue of a series of (wedged) open doors or agreed terms of discourse. However, arguably, such a metaphor also militates against collaboration. That is to say, it can deflect or diffuse any anxieties about the lack of collaboration by evoking the potential of collaboration: ‘we’re not collaborating now, but we could if we wanted to!’ In Michael’s case, the prominent role afforded socializing (drinking, gossiping and joking in the pub) might end up being precisely what is privileged in the collaboration. In addition to this socializing being potentially a mode of exclusion, not least a gendered one (who in the team is left out because of carer responsibilities?), it can also end up being an ‘end’ in itself. More specifically, for the sake of sustaining these socializing practices (one might call them a particular type of friendship), the difficult negotiations that would benefit an interdisciplinary study might be avoided; certain elements of an ‘imagined community’ are prioritized over others. In sum, we can detect the obverse to both Calvert’s and Michael’s accounts: the successful repair and reproduction of relations might end up, ironically, negatively impacting upon the success of ‘working out’ the research object.

Finally, when re-considering Fukushima’s chapter, it becomes clear that ‘multitude/ multiplicity’ takes a variety of forms. Not only is there the issue of the sheer numbers of relations, but there is the issue of the range of types of relationship. Over and above intra-/interdisciplinary relations, one can imagine other sorts of relations at play (for instance, gender, national/international, intra-/extra-institutional, early/mid/late career, fully/mainly/partially employed on/ committed to the interdisciplinary project). To focus only on ‘career phase’ and ‘commitment’ to the project, it is not difficult to identify bottlenecks in communication within teams where late career or part-committed collaborators do not see the relevant (interdisciplinary) project as sufficiently high priority to warrant their full, or even partial, attention. Needless to say, this can be very frustrating for junior researchers or those fully employed on the project. This in turn suggests that it is not simply the ‘research object’ that is in the process of being ‘worked out successfully’, so too are careers, salaries, promotions, future relationships, longer-term research programmes, and so on and so forth. Another way of framing this is to extend Fukushima’s tacit contrast between quality and quantity of interdisciplinary research. So far we have primarily been thinking about the different types of quality of interdisciplinary relations, and how these are variously inculcated, managed and secured. However, quantity also plays its part: the energy and enthusiasm (or, conversely, indolence and indifference) with which an interdisciplinary project is pursued, how this energy and enthusiasm is distributed across an interdisciplinary team, affects the ‘success’ of both the object and the relations of study.

Concluding remarks

This section introduction has attempted to situate ‘valuing’ and ‘validating’ in relation to the complex practicalities of doing interdisciplinary work. Drawing on discussions of translational research and on the chapters that make up this section of the book, the essay has implied that valuing and validating take on a variety of trajectories through the processes of interdisciplinary research, tying together such elements as the object of study, the generation of better problems and productive uncertainty (a-success), the quality and quantity of social relations and affects, and the careers and future plans of researchers. Out of this ‘topology of valuing’ – that comprises an interdisciplinary collaboration – might or might not co-emerge a ‘successful’ object of study aligned with its ‘successful’ collective maker (research team). As Stengers and Latour (2015) repeatedly note, such is a tempestuous process that is as likely to yield failure as success (though, of course, both these terms are multiply interpretable). That said, as we noted above, success is not simply a matter confined to the relation between maker and made, researcher and researched. It is also embedded within larger dynamics. On this score, and with the ‘topology of valuing’ in mind, let me end with an anecdote.

Many years ago, I was involved in a small interdisciplinary project (sociology, social psychology, cognitive psychology, anthropology, science and technology studies). Despite being a naive early career researcher, even I could recognize that there were fundamental problems with the research design. As the research project unfolded, it also struck me that we were simply not doing enough – either in terms of collecting adequate data, or working sufficiently hard on that data. My sense was that, in spite of a few publications, the project had failed to deliver on its promise and was, in a word, unsuccessful. However (and here my then-naivety was fully on display) I failed to reckon with the context of the project. As it turned out, emerging political and policy elements of the research field meant that the project found itself placed at the ‘cutting edge’, and was regarded as, more or less, a triumph. For me, it remains a missed opportunity, if not an outright failure, though I have not resisted benefiting from its success in the intervening years. The moral here is that both object of study and researcher emerged ‘successfully’, in part because together they could be said to comprise an ‘object of politics’ crafted in an other’s successful political project . .. in which sense values shape research methods more than we might always anticipate or understand.

Note

1 http://grants.nih.gov/grants/guide/rfa-files/RFA-AI-12-044.html#_Part_1._Overview

References

Anderson, B. (1983). Imagined Communities: Reflections on the Origins and Spread of Nationalism. London and New York: Verso.

Barry, A., Born, G. and Weszkalnys, G. (2008). Logics of interdisciplinarity. Economy and Society, 37(1): 20–49.

Bowker, G. C. and Star, S. L. (1999). Sorting Things Out: Classification and its Consequences. Cambridge, MA: The MIT Press.

Brosnan, C. and Michael, M. (2014). Enacting the ‘neuro’ in practice: translational research, adhesion, and the promise of porosity. Social Studies of Science, 44(5): 680–700.

Cribb, A., Wainwright, S., Williams, C., Farsides, B. and Michael, M. (2008). Towards the applied: the construction of ethical positions in stem cell translational research. Medicine, Health Care and Philosophy, 11(3): 351–361.

Denzin, N. K. (1978). Sociological Methods. New York: McGraw-Hill.

Haraway, D. (2016). Staying with the Trouble: Making Kin in the Chthulucene. Durham, NC: Duke University Press.

Knorr-Cetina, K. (1999). Epistemic Cultures: How the Sciences Make Knowledge. Cambridge, MA: Harvard University Press.

Merton, R. K. (1973[1942]). The Sociology of Science: Theoretical and Empirical Investigations. Chicago, IL: University of Chicago Press.

Michael, M., Wainwright, S. and Williams, C. (2005). Temporality and prudence: on stem cells as ‘phronesic things’. Configurations, 13(3): 373–394.

Mitroff, I. I. (1974). Norms and counter-norms in a select group of the Apollo moon scientists: a case study of the ambivalence of scientists. American Sociological Review, 39(4): 579–595.

Mol, A. (1999). Ontological politics: a word and some questions. In J. Law and J. Hassard (Eds.) Actor-Network Theory and After (pp. 74–89). Oxford and Keele: Blackwell and The Sociological Review.

Mol, A. (2002). The Body Multiple: Ontology in Medical Practice. Durham, NC: Duke University Press.

Mulkay, M. (1979). Science and the Sociology of Knowledge. London: Allen and Unwin.

Power, M. (1999). The Audit Society. Oxford: Oxford University Press.

Silverman, D. (2006). Interpreting Qualitative Data: Methods for Analysing Talk, Text and Interaction (3rd ed). London: Sage.

Stengers, I. and Latour, B. (2015). The Sphinx of the work. In Etienne Souriau (Ed.) The Different Modes of Existence (pp. 11–90). Minneapolis, MN: Univocal.

Wainwright, S., Williams, C., Michael, M., Farsides, B. and Cribb, A. (2006). From bench to bedside? Biomedical scientists’ expectations of stem cell science as a future therapy for diabetes. Social Science & Medicine, 63(8): 2052–2064.

Watts, G. (2010). Lost without translation? British Medical Journal, 341: c4363 doi: http://dx.doi.org/10.1136/bmj.c4363 (Published 18 August 2010).

Wilson-Kovacs, D. and Hauskeller, C. (2012). The clinician-scientist: professional dynamics in clinical stem cell research. Sociology of Health & Illness, 34(4): 497–512.