Abstract
When teaching ethics to medical students it is commonplace to distinguish between ‘ethics’, ‘law’ and ‘professional guidelines’. We encourage them to think about the difference between what one ought to (not) do and what one is mandated (not) to do. There is, of course, a relationship between guidelines, law and ethics, and we might call it jurisprudence. Nevertheless they are distinct. In contrast it is less common to find similar attention being paid to the differences between ‘ethics’ and ‘governance’. Certainly medical students are encouraged to think about ethics independently and not to simply rely on the quasi-legal guidelines set by the profession. However, at least in a manner comparable to that drawn between law and ethics, or even guidelines and ethics, we do not find any emphasis on the difference between them. This may well be because although at first blush we might consider ethics and governance to be distinct, further reflection reveals this not to be the case. It is unavoidable that, more or less explicitly, ethics involves governance.
Some might consider such a position counterintuitive and argue that ethics, or at least ‘morality’, must logically precede any notion of governance. However, this would be to mistake the ambit of governance as restricted to formal, administrative or bureaucratic realms of government or legislation. Instead we might focus on self-government or self-legislation and consider the concept of governance as applicable to any process, including that of individual reflection, which seeks to direct, shape and guide our activities. It is then a simple matter to see the Kantian moral being – the self-legislator – as involved in self-government, and all reflection on the categorical imperative as a ‘governance’ activity. Indeed, on this understanding of governance we might consider all forms of moral philosophy as, in some way, indicative of the need for governance. Although Kant saw morality as a process of self-legislation, and argued that all individuals must engage in it if they are to act morally, it is difficult to reflect on his moral philosophy and in particular the kingdom of ends without envisaging forms of government that might not only bring it about but also sustain it in the long term. Similarly, one would expect to find the phronimos engaged in legislation, government and, therefore, the governance of collective life. In the hands of Foucault it is clear that the governance of collective life has implications for the individual and the way in which they govern themselves. His unique Aristotelianism suggests that ethics can be considered a ‘technology of the self’, used to fashion not only ends or projects but also individuals. Finally, whilst Utilitarianism raises central questions about the ethical dilemmas faced by individuals – Jim in the jungle, for example – it is also rooted in the governmental concerns of its Victorian era progenitors, Bentham and Mill.
Although not all accounts of morality immediately suggest or intimately imply ‘governance’ – consider how, drawing on Levinas and the infinite demand of the other, Bauman associates ‘morality’ with ‘sociality’ 2 – one cannot move into the realm of normative ethics, understood as action-guiding norms, without invoking the concept of governance at some level. As Adam Smith has it: ‘A moral being is an accountable being’. 3 A moral being is one that can be called to account, both by itself and others, and yet, although morality and ethics implies accountability and governance, this does not mean that a particular normative perspective implies specific administrative or bureaucratic approaches. As a matter of social organization, some administrative form of governance might be required, but this is not to say that particular forms of government, such as laws or professional guidelines, are themselves (ethically) necessary. However, when ‘ethics’ or, perhaps more accurately, the academic endeavour of applied ethics, takes on more specific foci – as, for example, in medical ethics, biomedical or bio-ethics – there is not only an intimation of governmental necessity but often implications for the procedural shape that social necessity might or even ought to take. This places such ethics in close proximity to bureaucracy, the administration of power and control, and therefore to ‘politics’. However, in our attempts to keep ethics and politics separate we do not always recall that the institutionalization of ethical governance can be the object of further reflection or that ‘the political’ can be (re)addressed by ‘the ethical’ in its broadest sense.
At the present time something of these concerns can be clearly perceived in the specific arena of so-called ‘research ethics’. In particular, there are concerns about the homogeneity of approaches to ethical governance across different forms of research. This concern is not simply with the extension of a form of ethics but with the extension of a form of governance – what Hedgecoe (2009) calls the practical machinery – which attends those ethics. The depth of the connection between the ethics and the practical machinery is perhaps best demonstrated by the relationship between the Belmont Report and Beauchamp and Childress’ famous ‘Principles of Biomedical Ethics’ (Schrag: 84) – the ethics and the practical machinery are deeply intertwined. However, it is not clear that the ethics and the form of ethical governance developed in one context are suited to application in another. For example, Gross shows how, when put into practice, ‘research ethics’ enacts what she calls a ‘surrogate property transaction’ (Gross, 2011: 123) which separates ‘data’ from ‘subjects’ and renders it valuable. We can see how this moral or ethico-political economy of data is required by the broader economy of biomedical science; however, we might question, as Gross does in her conclusion, whether it is the correct, most ethical or productive (value-producing) model for the social science. The value of much social scientific data is to be found in maintaining its connection to the subject(s) of research. Furthermore, in ‘research ethics’ there seems to be an assumption that ‘research’ (and researchers) or, indeed, ‘science’ (and scientists) are coherent or unified underpinning concepts and categories (Schrag, 2012). On the basis of these assumptions a specific form of governance was conceived. Originally forged for the purposes of governing biomedical research, it is now used to govern everything that might be considered ‘human subjects research’, and has been politically extended to almost all academic research. In their own ways, each of the three books under review examines this concern for the contingent nature of institutionalized research ethics. Two, Stark and Schrag, clearly demonstrate how and why this particular model of ethical governance came into being, whereas Stark and van den Hoonaard examine the ethical governance of research more directly, both having conducted social scientific research with IRBs.
Schrag examines the history of the USA’s current system of ethical governance. The central plank of this history is the aforementioned Belmont Report produced by the ‘National Commission for the Protection of Human Subjects of Biomedical and Behavioural Research’. However, as Schrag’s concern is to show how all ‘human subjects research’ came to be governed by the same bureaucratic structures, we do not just get a history of the report and subsequent developments. We also get a clear focus on the ways in which review practices have been consistently expanded in such a way as to include the social sciences as a whole, 4 and not merely that part which can be described as behavioural. Furthermore, we also get a sense of the ways in which governance structures have been developed such that, often without anyone really intending to do so, input from social scientists about the social sciences was occluded. Legitimate concerns and distinctions that were raised were left unremarked, unnoted and unaccommodated in final documentation and subsequent interpretations and rulings, both federal and local. When committees and commissions set out to construct guidelines and protocols for the ethical governance of research, the motivation has often led to a focus on biomedical research. However, the generalizing tendencies of ‘applied ethics’ result in more and more activities being drawn into the ambit of its governance. As a result, the work of commissions and IRBs has often become enlarged and, having not been properly included, social scientists often have little choice but to ‘accommodate’ or ‘resist’ (Schrag: Chapter 8) an approach to the ethical governance of research primarily conceived in biomedical terms.
Schrag suggests that in the US there have been two ‘battles for the social sciences’. The first occurred between 1979 and 1980 and, although significant concessions were won, there was a failure to have the language of social scientific research ethics ‘encoded in legislation or regulations’ (Schrag: 96). Nevertheless, whilst the late 1970s had seen significant complaints about IRBs from social scientists, the 1980s saw a ‘détente’. During this time IRBs and social science seemed to have largely avoided one another. As documented by Schrag (126–129), students could pass through graduate school and beyond without hearing of the IRB’s existence, let alone being required to seek approval. Nevertheless, there seems little evidence of ethical failings on the part of social scientists during this period. Whilst internal discussion of ethics continued in the social sciences, particularly in anthropology, this sometimes took place without any reference to place and relevance of IRBs, and furthermore there was a significant downturn in articles critical about IRB handling of research. Despite this, the 1990s saw an expansion of both the federal oversight of social science, via the introduction of the common rule, and of IRB activity. Based on a new interpretation of what was and was not exempt (Schrag: 132), this proceeded with little input from social scientists, or indeed any real renewed consideration of the consequences of regulation by anyone other than bureaucrats and administrators, who can now be said to constitute an ‘interest group’ within the ‘research ethics community’. 5 The result was, Schrag argues, a second battle for social science and, although there have been some victories, it remains an on-going campaign. In his book, blog and elsewhere, Schrag is making a significant contribution (e.g. Schrag, 2011).
Stark offers her own contribution to the history of the ethical review of research in the second part of her book, where she demonstrates that the process of review by an expert committee reflects an ethics of a particular time and place: The Clinical Centre at the National Institutes of Health, Bethesda, Washington, DC, Maryland, USA, circa 1950. However, it is the first part of her book that is most interesting. Here she exercises her sociological imagination and conceptualizes IRBs as declarative bodies, populated by ‘rule experts’ who enact a deliberative process and in so doing make research ethical. However, Stark rejects the idea that this process is, as applied ethics would imagine, one of legal positivism. Board members do not ‘apply fixed regulations to specific protocols with more or less accuracy, resulting in objectively right or wrong decisions’ (Stark: 53). Rather, they engage analogical reasoning centred around cases. Rather than being ‘justified’ through applied ethical argument, IRB decisions are sociologically ‘warranted’, with individual members drawing on their personal and professional knowledge and experiences. In no small part this includes their experience on the IRB, and so much of the process becomes one of repeatedly applying ‘local precedents’ or previously worked out determinations of what is and is not ethical and how particular ‘ethical’ concerns are to be handled. New cases are constantly reconstructed in the light of old cases and previous judgements. As a result, the deliberative process cannot be considered ‘objective’ or ‘universal’ as it is fundamentally conditioned by its own and its members’ contingent histories. Indeed, this contingency is not limited to individual IRBs but is to be found in the historical genesis of the procedural review that they enact.
Nevertheless, despite the way in which her work reveals the historical and practical contingency of prospective ethical review, Stark offers little in the way of critique either in her book or elsewhere. This might be because she takes a broadly ‘value-neutral’ or ‘(neo-)positivist’ 6 approach towards her subject. Certainly it is consistent with, and perhaps even contributes to, what we might call her accommodationist stance towards ethics and ethical governance in the social sciences. 7 Stark demonstrates that the review process renders research accountable whilst, at the same time, erasing any trace of its own accountability (Stark: 73) or, we might say, its own status as an ethical endeavour. Through the administrative process any diversity in the ethical perspectives of members is smoothed over and rendered into an impersonal and ‘objective’ judgement marked by consensus. Very rarely is a dissenting opinion registered in official documentation and, indeed, quite specific concerns are often presented to the researcher as general directions, lending a certain degree of opacity to the whole process. Ethical concerns and the motivations for them are often ‘warranted’ rather than simply ‘justified’. Warrants are often entangled with the particular experiences, personal and professional, of the individuals who raise them, making it difficult for counter-arguments – rather than counter-warrants – to prevail. Thus specific, complex and potentially idiosyncratic concerns remain hidden from researchers, who must interpret, and even decode, the official language of ‘bad documents’ (Stark: 59).
Researchers are, then, at the mercy of IRBs that operate ‘behind closed doors’: declarative bodies who, at least officially, resist dialogical engagement and, particularly when reviewing social scientific research, constrain ‘how researchers go about creating knowledge – and, as a result, the kinds of things that are knowable’ (Stark: 73). Nevertheless, Stark’s attitude towards ethical review seems, like Adam Hedgecoe’s, to be one of cooperation, accommodation and compliance (Stark and Hedgecoe, 2010; see also Stark, 2007). 8 Although the research of both of these scholars undermines the ‘applied’ conception of ‘ethics’ advanced by IRBs, RECs and the ‘research ethics community’, neither author seems particularly critical of our current forms of ethical governance. Hedgecoe (2012) advises engaging committees, or at least its chair, in ‘face-work’. Such informal engagement can, no doubt, assist the researcher in understanding (or ‘decoding’) a committee’s pronouncements and, therefore, smooth the path to ethical approval. However, there is no sense here of a longer term engagement with or solution to the difficulties their own work has been instrumental in identifying. If ethical review of research is justified through the rhetoric of applied ethical discourse and on objectivist grounds, then advising researchers to use the ‘personal touch’ to engender trust between themselves and the IRB/REC is not an adequate response to the revelation that this rhetoric is largely otiose. Such advice does not promote ethical research but, rather, a tick-box approach to ethical governance: to comply with a bureaucratic regime through adopting the path of least resistance, even if that path is engendered through the presentation of the (ethically compliant) self.
Although it would be convenient to suggest that our third book, van den Hoonaard’s Seduction of Ethics, treads a middle ground between Stark and Schrag and, therefore, between accommodation and resistance, this would not be entirely accurate. In an attempt to show that the contemporary approach to ethical governance of research has a negative impact on the social sciences, which is to say it has a negative impact on the moral and epistemic culture of the social sciences, van den Hoonaard arranges a vast amount of material – existing literature bolstered by his own empirical research and experience. Through this work he attempts to reach a point where it is clear that alternative forms of governance more suited to the moral and epistemic culture of the social sciences need to be conceived. Along the way he provides some explanation for Schrag and Stark’s contrasting opinions on ethical review. Van den Hoonaard (259–260) identifies prospective ethical review as not only consistent with the (neo)positivist paradigm of ‘research’, ‘science’ and ‘knowledge’ but as belonging within that paradigm. It is not simply that IRBs and, to a lesser extent, RECs tend to be staffed by positivist, often biomedical, scientists who understand researchers to be in the business of collecting independent facts, but that the dominant ethical perspective reinforces this view by maintaining the separation of ‘is’ and ‘ought’, or ‘fact’ and ‘value’ (see van den Hoonaard: Chapter 13). Even if IRBs/RECs do not explicitly assume that the social sciences are a (neo)positivist endeavour comparable to the natural sciences, and many clearly do, it is nevertheless implicit in the administrative approach taken towards ethical review. Furthermore, this view is entrenched across applied (bio)ethics – a discipline now widely assumed to be the intellectual pater familias of the ‘research ethics community’. 9 Research that does not adopt a (neo)positivist perspective, or style of presentation, receives treatment that often feels incongruous with its broader aims and purpose. This includes the more than occasional suggestion that adopting a positivist approach or ‘method’ would be more ethical whilst producing the same result, which is something patently untrue from a post-positivist perspective. This is particularly troubling, not only for the constraints that it places on what is and is not knowable, but also for the education of students, the training of future researchers and, perhaps more importantly, future ‘users’ of research.
Van den Hoonaard’s critical engagement with his subject matter is implicitly reliant on an alternative, broadly anthropological, understanding of ethics. Although this idea does not fully emerge from his work, as the primary focus is on collating and presenting a range of research on the process of ethical governance, we might consider van den Hoonaard’s view of ethics as being drawn in much wider manner that that of applied ethics. Rather than the thin normativity of rules or principles, anthropological perspectives on ethics consider our ways of being and, more importantly, our ways of being with each other, which, in modernity, includes our administrative institutions. Such perspectives do not reject rules and principles but seek to show that they are one part of our broader moral lives. What such research demonstrates is that in practice our ethics cannot be understood to be fully distinct from the broader moral culture (the ethos of particular societies and sub-communities) in which we govern ourselves and our relationships with each other. The move is from a thin account of normativity as action guiding rules and a thick account of morality that understands the ‘socio-cultural’ – something that encompasses the individual – as a fundamentally normative phenomenon. 10 Philosophical or applied ethical perspectives would suggest that the former should be understood to precede, or attempt to ‘transcend’, the latter. Anthropological perspectives place the former firmly within the latter, where they can then be placed in a mutually constitutive relationship. What these authors reveal is that there is a broader social morality to the ethical review of social scientific research. Indeed, such a view is implicit in the aforementioned task of ‘face-work’. Whether this broader social morality makes a positive contribution to the ethical dimension of social scientific research is open to scrutiny. However, it reinforces the unaccountable nature of ethical governance in the social sciences. How systems of governance should themselves be held responsible – to researchers, to research participants, and to society as a whole – remains uninterrogated by applied ethical thinking.
Although philosophy is a reflexive subject, there is often a lack of ‘social’ or ‘sociological’ reflexivity, particularly in ‘modern moral philosophy’ and applied ethics, despite the fact that it has much to recommend it (Addelson, 1991). Although this trait is all too evident in the case of social science research ethics it may be that at least some of the blame lies elsewhere. Schrag and van den Hoonaard conclude their books with the suggestion that we need to rethink the ethics of social scientific research. However, it is not simply that we need to rethink, for the purposes of social scientific research, the concepts of autonomy, informed consent, confidentiality, privacy and anonymity as, for example, Hammersley and Traianou (2012) have done. The problem lies not with the ‘ethics’ per se but with the systems of governance we have constructed and the bureaucracies we impose. Furthermore, it lies with the relationship those ethics have with the forms of governance. It is not simply that the social sciences have been subject to biomedical research ethics or that they have been subject to a form of governance conceived and developed with biomedical research as its major focus. Rather, as Stark shows, it is these facts coupled with the development of local historical precedents and repertoires that are repeatedly and inflexibly applied to a diverse range of research proposals. It is not only the seductive universalism of ethics and the positivistic assumptions of review but also the way in which they are embedded within the process of prospective review. The ethical governance of research does not simply hold researchers accountable; it actively audits research and, by proxy, researchers themselves.
This idea of audit is predicated not only on the possibility of presenting a scientific research protocol but on a particular understanding of the relationship of ‘ethics’ to ‘action’ and of ‘ethical principles’ to ‘policy’ or ‘practice’. In an applied ethical perspective the principle precedes the practice to which it is subsequently applied; its meaning, remit and purpose is largely settled ‘a priori’. In a (neo)positivist perspective there is a similar relationship between the research protocol and the research practice. However, in a post-positivist perspective there is a reflexive relationship between the research that is planned and the research that is practised. The meaning of a research protocol remains open during the process of conducting the research. Similarly, the ethics of such research remains an open, and often troubling, question. Consequentially, as Markham puts it, the ‘freedom to construct one’s methodology in a reflexive, mindful, context-sensitive manner … can actually be hampered by the very entities that seek to promote ethical treatment of human subjects’ (Markham, 2006: 50). For post-positivist social sciences the ‘freedom to construct one’s methodology in a reflexive, mindful, context-sensitive manner’ is not only an epistemic matter but also an ethical one. Thus, insofar as ‘research ethics’ or, rather, the ethical audit of research, closes down this reflexivity it compromises the underpinning moral economy of the research itself.
I am not the first to suggest that the ethical governance of research is one aspect of our contemporary ‘audit culture’ (Strathern, 2000). Halse and Honey (2007) have noted the affinity, van den Hoonaard (23) mentions the possibility in passing, and, in the context of biomedical research, Ulrich suggests that the ‘discourse of audited accountability, whilst seemingly appropriating the language of ethics, essentially revolves around questions of expediency and coercive mechanisms of rule compliance’ (Ulrich, 2011: 151). Macdonald (2010: 88) notes that audit involves a form of ‘performed transparency’ and that this is one of the reasons lay members are included in research ethics committees. One might add that the information sheet considered so essential to the possibility of informed consent also fulfils this function. Such sheets, and signed consent, are insisted upon because they perform the function of informing the research subject and do so in an auditable way. The accumulating evidence for the fact that participants find them confusing, overly long, and often do not bother to read them (van den Hoonaard: 116) matters little to the performance of this function because that is all it is. Thus the ‘accountability’ offered by audit can be perfunctory, legalistic and lead to an ethics that is empty (Corrigan, 2003). Audited accountability is not necessarily the best form of ethical accountability and nor is it necessarily, to finish the above quotation, ‘genuinely ethical in nature’ (Ulrich, 2011: 151). As Thrift says, ethical responsibility is arrogated by the ethics committee and subject to auditable standards. The result is ‘a normative regime that takes responsibility away from the researcher’ (Thrift, 2003: 120).
However, as a form of ethical governance, ‘auditability’ remains distinguishable from ‘accountability’. Researchers in the social sciences can make themselves accountable to their disciplines, peers, institutions and research subjects in a variety of ways. In the first instance the approach to ethical governance in the social sciences might be rendered more democratic. Ethics committees should not meet behind closed doors to review a research proposal but instead should be available to engage researchers in ethical dialogue both prior to the start of a project and in an on-going manner. Furthermore, such dialogue should instantiate a greater reflexivity about the application of ethical principles to the proposed research. What ‘informed consent’, ‘autonomy’ and ‘confidentiality’ might mean and how they can be achieved should remain open to question. Researchers and IRB/REC members should be encouraged to look for answers to these questions in the relevant, which is to say discipline-specific, literature. As many have shown, the current standards of ethical practice in medicine and biomedical research have not been externally imposed by ‘bioethics’. Rather, the cultural development of bioethics has taken place, to no small degree, within the culture(s) of biomedicine. Thus the changes it has wrought – and it has wrought changes – have been dialogical accomplishments. The social sciences deserve nothing less than the same courtesy. It is time for the applied ethics of the research ethics community to foster a greater connection with the social sciences and its research – those it should be seeking to engage dialogically and ethically.
Footnotes
Acknowledgements
I should like to thank Robert Dingwall for reading a draft of this essay and for his helpful comments.