Ethical imperialism or ethical mindfulness? Rethinking ethical review for social sciences

Abstract

This article is a response to the challenge with which Zachary Schrag concluded his article, ‘The case against ethics review in social sciences’ − that ‘the burden of proof for its continuation rests on its defenders’ (Schrag, 2011). This article acknowledges that there is substance in the charges he lays against some reviews of social sciences and that these are of sufficient quantity and seriousness to justify his challenge. Instead of favouring abandonment of ethical review of social sciences, the author of this article draws upon his experience as Research Ethics Officer for Social Sciences and Law in a research intensive UK university to identify the sources of some of the problems and suggest potential remedies. These start with reviewing the warrant for ethical review as the basis for understanding reviewer’s role concerning the core ethical issues of rigour, respect and responsibility in the proposed research. This response concludes by considering the implications of a warrant for enhancing trust between scientists and the public as the context in which ethical review is only one component in a number of ethical strategies designed to promote ethical mindfulness as integral to social science research.

Introduction

Zachary Schrag does a positive service to social sciences by drawing attention to and accumulating evidence of the volume and severity of difficulties that current practice of ethical review poses for many social scientists. However, I disagree with any suggestion that the answer is to abandon ethical review. Instead, I intend to take up his concluding challenge at the end of his thought-provoking article − that the burden of proof for the continuation of ethical review rests on its defenders.

The great service Schrag offers is cataloguing the enduring difficulties, problems and injustices inflicted on social scientists and their research subjects or participants under the guise of ethical review. In his article ‘The case against ethics review in the social sciences’ (Schrag, 2011), he sets out his charges against ethical reviews of social sciences, namely, that they result in silly restrictions due to misapplying biomedical assumptions and practice; and conjure fictitious or improbable dangers to justify the ethical reasons for impositions and restrictions. Furthermore, that review committees typically lack relevant expertise; apply inappropriate principles; inflict harm on the innocent; and have few success stories. The article draws heavily on his more extended consideration of Ethical Imperialism: Institutional Review Boards and the Social Sciences, 1965–2009 (Schrag, 2010) in which he applies his expertise as a historian to analyse the sequence of events and socio-political dynamics that have resulted in such an undesirable state of affairs. Before I go any further, I ought to say that I recognize the state of affairs that he describes and am aware of comparable difficulties experienced by some social scientists, particularly those using qualitative research designs and methodologies, being unjustifiably mauled by health-related ethical review panels. Although most of the examples presented by Schrag are drawn from the USA, we have comparable difficulties in the UK of sufficient seriousness and volume to merit serious consideration of Schrag’s analysis of the current state of ethical review. However, I will suggest that such difficulties are neither universal nor inevitable.

In this response to Schrag’s article, I will disagree with any proposition that ethical review of social sciences is so deeply flawed that it ought to be abandoned. This is the obvious inference from his title (Schrag, 2011). I want to take up this challenge, not to defend the status quo which appears to be quite varied in the UK, but rather to argue that a reformed system of review of social sciences would not only enhance the quality and standing of social science but will be of value to society in general.

It is probably helpful if I explain my professional and academic background to set my optimism for ethical review in a biographical context. In 1996 I became the first Research Ethics Officer for Social Sciences and Law at the University of Bristol, a research intensive university in the Russell Group of UK Universities. I was one of the founding members of the University-wide system of governance and ethical review described by Birgit Whitman and Gillian Tallents in this journal (Whitman and Tallents, 2010). I took on the role because I recognized a state of realpolitik – that UK universities were being required to adopt systems of ethical review for all research involving human subjects. The pressure from the providers of funding for research had become irresistible, particularly from the Wellcome Trust and the Economic and Social Research Council (ESRC). I was somewhat ambivalent about taking on the role because I was aware that the kind of difficulties Schrag outlines would be representative of the experience of some colleagues seeking review of social science research proposals by health-based ethics committees. However, I had been fortunate enough to have experienced very constructive ethical reviews and support by health panels of social science research projects on sensitive topics. I had experienced how appropriate review by well-informed and thoughtfully engaged panels throughout the research process could promote the interests of research participants and support the delivery of research often affecting very vulnerable participants. The most memorable of these experiences occurred when I was researching HIV counselling at a time of heightened anxiety due to moral panic over AIDS. Many health workers feared a route of infection, for example hepatitis, to which they were particularly susceptible in routine clinical situations. There was only limited reliable scientific knowledge about a life-threatening and socially stigmatized disease for which there was then little effective treatment. I was undertaking these studies as a specialist in counselling with a particular interest in professional ethics for any form of psychosocial support delivered through talking and listening (Bond, 1990; 1995). The distinctive characteristics of the reviews of these projects were a balance between a commitment to the potential value of the research; their willingness as interdisciplinary panels, including clinicians, to be critically engaged in unfamiliar qualitative research design and methods by asking challenging questions and engaging with the answers. The panels were particularly attentive to the experience of research participants who were represented in one of the panels where the research involved people with HIV/AIDS. The panels were probably closer to the sponsorship of the research than most contemporary research ethics review committees as they were appointed by the sponsors and accountable to them for overseeing ethical and quality issues. Nonetheless they effectively fulfilled the functions of ethical review across the life span of each project with the authority to make recommendations, impose conditions or declare the project unethical and recommend closure to the sponsor. As the researcher, I took these powers seriously but experienced my discussions with panels as mutually respectful and appropriately attentive to ethical issues given the sensitivity of the research. Personally, they were a uniquely intense experience of exploring what it means to be trustworthy simultaneously to both research participants and the practitioners who might apply any research findings. These positive experiences of ethical review had influenced the way in which I approached writing Ethical Guidelines for Researching Counselling and Psychotherapy (Bond, 2004), in which the trustworthiness of the researcher frames the typical range of issues considered in research ethics concerning human subjects.

This background meant that, when I took up the role of Research Ethics Officer, I was simultaneously wary of the uncritical transfer of biomedical ethics to social sciences based on the experience of colleagues with experiences similar to those described by Schrag, but optimistic that something much more positive could be achieved. Five years later, I have a greater appreciation of the challenges and difficulties involved, but also a correspondingly increased sense of the value of appropriate ethical review. In this article, I want to set out some of the potential remedies to the difficulties identified by Schrag that make appropriate ethical review across the social sciences and humanities both possible and desirable.

I am not proposing to attempt to counter each of Schrag’s charges individually but to take them as evidence of systemic difficulties in ethical review that have proved particularly damaging to some types of social science, particularly disciplines that depend on ethnographic and oral history research. I consider that the remedies are unlikely to be found by modifications of particular procedures alone but do require deeper consideration of the purpose of ethical review of research starting with the warrant or justification for undertaking this work.

Reviewing the warrant for ethical review

The original moral justification for imposing ethical review was atrocity avoidance. This started with the Nuremberg trials and moral repulsion over the disregard for human life and suffering inflicted in scientific and medical experiments in the concentration camps of nazi Germany (Kennedy and Grubb, 2000: 1671). This warrant gained authority with the growing realization of the stubborn persistence of prioritizing the quest for knowledge over the rights and well-being of human subjects in scientific, particularly biomedical, experiments. The multiple revisions of the Helsinki Declaration, originally devised to avert any repetition of the abuses exposed in Nuremberg, are testimony to the difficulty of the task of adequately protecting patients in morally less contentious research (WMA, 2008). The Belmont Report in the USA also represents a determined attempt to redress the balance between researchers and subjects in favour of research subjects, following the exposure of a number of unethical research projects involving the exploitation of vulnerable people (National Commission, 1979). A major concern that prompted this report was the US Department of Health funded Tuskegee Syphilis experiments, which were misrepresented to 399 patients as treatment for ‘bad blood’ − a colloquial term for syphilis. This study provoked public outrage at the way any available treatments, including antibiotics when they became available, had been deliberately withheld for over 40 years in order to study the natural progression of the disease in poorly educated black people (Jones, 1993). The scale of the challenge and the persistence of a scientific culture that disregarded the significance of Nuremburg for biomedical and behavioural research prompted a sequence of developments resulting in mandatory Independent Review Boards (IRBs) for all research involving human subjects in the USA. It is difficult not to be impressed by the scale of the struggle against exploitation of research subjects and disregard of their well-being that prompted the formalization of ethical review in many different jurisdictions around the world.

When members of ethics committees overestimate the potential harm in a particular social science project, I think that this is not simply an error of risk assessment to be taken at face value but may be an understandable, even praiseworthy, attempt to entrench the warrant of harm avoidance and overcome any resistance to the moral imperative, primum non nocere (above all/first do no harm). However, no matter how well-intentioned, the impact of such exaggerations is ultimately damaging to any realistic assessments of risk and harm. The level of damage may be amplified by the group dynamics within the ethics committee if the warrant of atrocity and harm avoidance is too dominant. The exaggeration of harm adds persuasive power and authority to the speaker within the group dynamic of the committee by giving them a moral authority that trumps any more equivocal views based on the low probability of actual harm. I have observed within ethical reviews how difficult it is to argue that evidence of actual harm in social sciences is either hard to find or nonexistent against a suggestion of the mere possibility of harm. Operating under a warrant of harm avoidance is sufficient to authorize restrictions based on remote possibilities in many ethics committees even though the probability of actual risks may be accepted as very low. It is this combination of warrant and group dynamic within ethics review committees that may substantially explain and justify Schrag’s charge that all too often ethical review can create fictitious or improbable dangers to justify the ethical reasons for impositions and restrictions.

One of the remedies for an excessive fixation on harm avoidance to the detriment of all other considerations is to redefine the warrant for ethical review in order to establish a perspective in which harm avoidance can be evaluated against other ethical imperatives. This is particularly important in social sciences, where potential harms are seldom physical but much more likely to be damage to intangible social assets such as identity, relationships or reputation, which themselves derive their significance from the social context and are contingent on the perceptions of the people affected. It needs to be a warrant which assists reviewers to identify and evaluate the degree of potential harm and appropriate safeguards in a way that is proportionate to the contribution of research to the well-being of people in general by the generation of knowledge and of the people being researched, in particular. There is insufficient space here to consider all the potential warrants to mandate a research ethics committee but the one I am currently using to guide my decisions and interventions when chairing an ethics committee focuses on the question, what is ethically required to ensure public trust in, and willingness to participate in, social science research. This invites consideration of multiple ethical imperatives applicable to the proposed research. Is the research sufficiently rigorous in its design and proposed operation to deliver the knowledge to which it aspires? Is the proposed research demonstrating sufficient respect to the human participants and any others affected by the research? Is the research being undertaken with sufficient responsibility for the presentation of the knowledge that is generated?

These questions are deliberately framed to be inclusive of social science and its insights but have their origins in the Universal Code for Scientists (to which social scientists contributed). This Code opens with a statement by Sir David King as the Government Chief Scientific Adviser, ‘Our social licence to operate as scientists needs to be founded on a continually renewed relationship of trust between scientists and society’ (Government Office for Science, 2007). Social sciences are not unique in having difficulty with current approaches to ethical review and governance. The recent Rawlins Report on streamlining ethics and governance for health research echoes this warrant for ethical review by balancing the ‘need to promote high-quality research but also to maintain public and professional trust in an area that relates directly to individual safety and dignity’ (p. 19). Recent developments in health research ethics and governance have taken up this warrant. I will return to the ethic of trust in the final section of this article.

Testing assumptions around rigour

Ensuring the quality of the research undertaken is an essential precondition of involving the contributions of human participants. Poorly designed or implemented research undermines public trust in the integrity of the research process and may generate misleading knowledge claims. The important quality control for research is frequently framed in terms of rigour. The uncritical deployment of rigour to address quality issues can prove to be a source of difficulty in the ethical review of social sciences.

Rigour can be narrowly defined as the systematic application of the principles of natural science to research. In many ways, this was the aspiration of early social scientists, to study society as though it is analogous to natural objects and can be best understood by use of natural scientific methodologies. Many quantitative and behavioural studies continue to conform to these principles of validity, reliability, generalizability and predictive power as the hallmarks of quality in ways that would be familiar to any researcher in biomedical sciences. However, it is the uncritical transfer of these principles to research in social sciences and the humanities that is the cause of many misunderstandings and errors in ethical review. When social scientists, typically qualitative researchers, complain about being trapped into a biomedical view of research, as Schrag correctly observes, it is frequently because they are being subjected to too narrow a definition of rigour.

My experience of teaching research methodology to doctoral students from a wide variety of academic disciplines suggests that students with a background in natural, biomedical and engineering sciences have frequently had little exposure to the diversity of the philosophical backgrounds and issues involved in knowledge generation that prompted an explosion of new methodologies in social science from the 1960s onwards. The normative assumptions about rigour in research are derived from analytical empiricism in the natural sciences and have been remarkably successful in extracting valuable knowledge from the natural world. However, they are an inadequate basis for engaging with the full range of contemporary social sciences which are critically engaged with ideas about the nature of being (ontology); theories of knowledge (epistemology); or the application of research design and methods (methodology). As students from the natural sciences start to appreciate the diversity of human interests that can be researched systematically (Habermas, 1968/1987), they start to question what constitutes rigour and differentiate its application to natural empirical, interpretive or emancipatory research. They start to appreciate that the quality criteria that underpin notions of rigour may need to vary according the research interest being pursued and seek congruence between the research aims, the methodology deployed and the knowledge claims that may result. Some researchers, particularly those who are philosophically minded, become very sophisticated around issues of legitimation, representation, and the relationship between theory and embodied action as further refinements of rigour. However, it does not usually require high levels of sophistication to ask whether a research proposal is sufficiently coherent in its position, purpose and methodology to justify the active involvement of human participants. When approached in this way, issues of rigour can accommodate considerable diversity in research designs from the most traditional approaches to natural science to the avant-garde post-modernism that characterizes contemporary social science. Some degree of epistemological open-mindedness and humility is essential to enabling social sciences, and probably all disciplines, to innovate and flourish.

Undoubtedly Schrag is correct in attributing some of the difficulties during ethical review that he reports to ignorance, which may be compounded by arrogance. However, this is probably not sufficient to explain the number of instances of difficulties or their persistence. Other systemic and social factors probably interact in varying combinations to entrench the difficulties. Competition for scarce research funding, rivalries between academic disciplines, the economic power of the biomedical industries and high esteem granted to STEM subjects are all dynamics that exist to varying degrees in the politics of contemporary universities. It would be surprising if such influences did not appear as conscious or subliminal influences in ethical reviews. If these are truly uncontrollable forces leading to epistemological censorship and resistance to challenge, then it would arguably be better to abandon ethical review by committee as too partisan to undertake responsibility for social sciences. I consider that Schrag accurately reflects the depth of despair of some social scientists following a bad experience of ethical review, but this seems unduly pessimistic and damning as a general evaluation of all ethical reviews of social sciences. In so far as such a view can be considered plausible, it ought to be a real cause for concern for ethical reviewers. It suggests that reviewers have taken too narrow an interpretation of rigour and that the wider social context in which the review process takes place has been overlooked. An essential aspect of maintaining trust between ‘scientists and society’ is supporting research in pursuit of any legitimate human interest, rather than closing down or censoring inquiries by any type of scientist, in the widest sense of this professional title as someone systematically seeking knowledge. The ethical warrant needs to take account of the diversity and complexity of contemporary society and should strive to be fair between different interests and protect their right to undertake research as a contribution to democratic society. As an ethical standard, rigour provides a healthy challenge to the integrity and quality of research but has to be understood within the broader context of the processes and safeguards of a democratic social structure. The right to seek knowledge is closely aligned to the right to free speech but, like that right, is not absolute and is curtailed by the rights of others, which require respect.

Ensuring respect of participants and those affected by the research

Respect carries many gradations of meaning, such as following etiquette by demonstrating politeness and observing the expected social conventions; giving attention to and taking into consideration someone’s well-being and immediate experience within a specific social context; or showing deferential esteem to another person. In research ethics, I consider that we are aiming at the second meaning of respect for research subject or participants, although some contexts may require either of the other two in order to achieve this level of respect. The Universal Code identifies three core elements of respect as lawfulness, ensuring that the research is justifiable, and that any adverse effects on people, animals and the natural environment are minimized and justified (Government Office for Science, 2007). Interpretation is critical. Would the application of this combination of principles have helped with the seemingly nonsensical judgements of ethics committees as reported by Schrag?

For example, would an ethics committee be justified in turning down a proposal to interview failed suicide bombers in prison to investigate the causes of terrorism, which was the experience of Scott Attran? The reported reasons for declining ethical approval are based on issues of respect, namely that prisoners cannot give free and informed consent in prison and information obtained in interviews could put as-yet-uncaptured terrorists at risk of arrest. The first reason relates to protecting the autonomy of the prisoners who are facing confinement in a context in which their rights to autonomy are deliberately curtailed in pursuit of security, protection of society, and retribution. This must be a real concern for any prison-based research and requires a wise judgement on the part of the researcher to be responsive to the degree to which participants willingly participate, whilst at the same time being sensitive to the degree to which any prisoner may censor or distort responses as a personal protection of autonomy and survival within a socially sanctioned oppressive context. This would arguably be too ambitious for an inexperienced researcher or someone unfamiliar with penal environments, but not an insuperable obstacle for a researcher with appropriate experience. Clarity about the research design, planning for its implementation, and the extent to which the researcher has anticipated the needs of prisoners, the probable conditional and provisional nature of any consent, and the potential sources of difficulty from prison staff and other inmates would be persuasive. The ethics committee’s concern to protect the interests of as-yet-uncaptured terrorists is much harder to understand on so little information. The researcher’s attentiveness to the rights of third parties can be more challenging in social science than clinical or laboratory-based research but arguably this review panel appears naive about the realities of researching any criminal behaviour where the identities of other parties are seldom essential to the purpose of the research or disclosed. If identities of other criminals are disclosed, then very often the discloser may be indicating a willingness to communicate this information to the authorities or frustration that previously disclosed information has seemingly not been acted upon. The review panel has left itself vulnerable to a possible accusation of colluding with terrorists or taking a political act of resistance against the treatment of suspected terrorists. Neither of these positions appears compatible with a warrant of preserving the trust between scientists and society within a democratically authorized social structure. Alternatively, a panel of reviewers that has become over-focused on preventing harm in any form might fail to consider the wider social and moral context. The ethically challenging question for all researchers in contact with subjects who may cause serious harm to others is to determine the limits of respect. At what point does the researcher’s respect for one person compromise the integrity of that researcher’s respect for other members of society? A more ethically effective outcome of the review might have been to pose these questions to the researcher and to seek information about how the answers would be implemented.

A recurrent theme in the cases presented by Schrag is the ethical significance attached to consent. Attentiveness to issues of consent is one of the major ethical strategies for protecting the rights of research subjects in clinical and laboratory experiments or when they are being asked to provide sensitive information. It has been so successful as a strategy that some reviewers find it inconceivable that any research could be undertaken ethically without formal written consent. This has sometimes proved to be an insuperable obstacle to ethnographic studies of people in their ‘natural’ social contexts thus effectively prohibiting an otherwise informative and valid research method widely deployed by social scientists, especially in anthropology and social psychology. He cites cases where review committees have arguably lost a sense of perspective over issues of consent which may result in such apparently absurd instructions as facing away from research participants who have not explicitly consented to participant observation of a group where most have consented (the case reported by Will van den Hoonard) or requiring arguably excessively far-reaching consent from people with little or no direct involvement in the matters being researched, such as the requirement for the consent of ‘every barrista who served us coffee and waitress who brought us pizzas’ (the Irena Grugulis case). In the first response, the reviewers appear to have considered consent sufficiently insignificant to effectively eliminate its operational effect but were reluctant to waive it as a requirement. In the second, formal consent appears to be the universal requirement for anyone even with the most peripheral contact with the research. Reviewing ethnographical studies challenges reviewers to consider the applicability of ethical norms in the clinic and laboratory to studies of people engaging in their usual activities and where the only difference for the subjects may be being observed. Many ethnographic studies and oral histories fall into this category, where the research is essentially interpreting human behaviour which would have or has occurred whether or not it was the subject of research, and the outcome of the research has no direct consequences for the people involved. Even when formal consent is considered to be appropriate, the method of obtaining that consent may be critical to the ethical efficacy.

One of the more challenging aspects of consent has been taking account of cultural contexts. Seeking written consent in cultural contexts where oral transactions are the accepted norm may actively communicate mistrust and alarm participants. For example, people at the lower end of social hierarchies may strongly associate written documents with contexts in which they are vulnerable to being misunderstood or exploitation by people of greater power and influence than themselves. The extent to which researchers and ethical reviewers engage in issues around the best methods for seeking adequately informed consent is arguably one of the hallmarks of the degree to which they are seriously committed to being respectful of research participants. The uncritical transfer of practices from one social context to another is unlikely to fulfil the ethical imperative of demonstrating respect, whether that is from the clinic and laboratories, to everyday contexts or across cultural differences. This seems to be a source of some of the difficulties of the type reported by Schrag and experienced more widely by social scientists. Formal ethical review is usually a bureaucratic process which favours bureaucratic evidence of respect for which the consent form is probably the most easily provided. However, if the consent form is a significant obstacle to the research, then the most appropriate response may be to ask how will the researcher demonstrate respect to the participants and be attentive to their rights? The researcher who can articulate how they will demonstrate and commit themselves to being respectful in relationship where there may be inequalities of power, status and vulnerability may be ethically more efficacious than someone who is over-reliant on a signature on a form. These issues are not unique to social science. The adequacy of formal consent given in advance of biomedical procedures to protect patient autonomy has been seriously questioned (O’Neill, 2002a: 42). Attention to issues of consent provides an ethical foundation that communicates an ethical commitment to respect but cannot eliminate the need for professional judgement about the limits of that consent and when that consent is being stretched beyond the reasonable expectations of the person who gave it.

Supporting researchers in taking ethical responsibility

Taking ethical responsibility has many potential dimensions in research ethics. The predominant interpretation is the responsibility of researchers to communicate their intentions, methodology and results honestly and accurately, and being attentive to how the research and its outputs will have an impact on society (Government Office for Science, 2007). Dishonesty in the research process or dissemination of flawed findings does disproportionate harm to the public’s trust and contrasts with the volume of honestly undertaken research. This dimension of responsibility falls outside the scope of Schrag’s article but is one of the justifications for ethical review, even though such review can never be a total protection against sophisticated attempts at deception.

Central to Schrag’s concerns are the balance between the ethical responsibilities of ethical reviewers and those of the researchers. It is perhaps inevitable that a catalogue of questionable or even poor outcomes from ethical reviews will direct attention to situations where the responsibilities are antagonistic to each other and where the review process has either undermined or wholly frustrated the ethical commitment of the researcher. These are poor outcomes for ethical review that deter ethical engagement.

However, there is another view of the division of ethical responsibility between reviewers and researchers in which reviewers understand their role as supporting researchers in being ethical. This is not primarily a coercive relationship but one that is supported by reviewers modelling the ethical qualities they seek to foster in the researchers and working in dialogue with researchers in a timely fashion. Asking the right questions in Socratic style to promote the ethical development of researchers and their engagement with ethical issues that they face has many advantages. It acknowledges the characteristic ethical commitment of most social science researchers rather than undermines it. Listening to the responses to questions with sufficient humility to empathically engage with and demonstrate respect for the researchers, and especially their subjects, provides the best possible basis for constructive outcomes to ethical review. Striving for outcomes based on mutual agreement, wherever possible, also validates the ethical responsibility of the researcher. There may be times when the reviewers have concerns of sufficient seriousness to consider it appropriate to require modifications or prohibit research. These situations require a great deal of attention to the style of the communication and the quality for reasons offered, as this is the point at which a researcher may become cynical about or alienated from ethical commitment. One of the potential interpretations of some of the cases presented by Schrag is that reviewers may have made an appropriate judgement on the facts available to them but failed to communicate the basis of that judgement in a way that satisfied the researcher or is readily comprehensible to independent scrutiny. Promoting the ethical engagement of the researchers whose proposals are being considered to the wider research community ought to be high on the agenda of priorities for all ethical reviewers.

Mere compliance to authoritative judgements, even by ethical review committees, represents a low level of ethical commitment. Effective ethical engagement requires a considerable investment of personal values, energy and agency by the researcher which can be supported or undermined by the review process. The multi-layered nature and complexity of much of the subject matter of social sciences requires researchers to be ethically alert and responsive to the challenges of undertaking their work with people who are unlikely to be as predictable as even the best designed research envisages.

Unfortunately, the legal framework underpinning ethical review in some of the jurisdictions is so protective of the independence of the review process that it prevents any meaningful communication between applicant and reviewer during the research. This may be a further factor that accounts for some of Schrag’s cases. Some systems of review appear to be inherently coercive by design.

Attending to issues of trust

One of the advantages offered by basing ethical review on an explicit warrant of renewing and enhancing the relationship of trust between scientists and society is that it invites review committees to ask themselves key questions about what is required to promote trust between researchers and their research subjects. It widens the context of ethical review and challenges all reviewers to recognize the limitations of disciplinary specializations, important as these are in advancing knowledge, by engaging afresh with the process of knowledge generation and how this relates to the good of society. This should provide a more constructive platform from which to challenge undue delays and obstacles to undertaking research, and curtail uncritical transfers from one realm of research practice to another by directing attention to consideration of the primary issues at stake. These are recurrent themes not only in Schrag’s critique of ethical reviews of social sciences but can be sources of difficulty within other areas of inquiry including biomedical research.

A warrant that is attentive to issues of trust provides a context in which the improvements to ethical review of social science advocated by Schrag find their place and can be evaluated. Constructive suggestions like those cited by Schrag, from Martin Tolich and Maureen H. Fitzgerald or Lisa Rasmussen, would be considered in a context which is sufficiently open-minded to consider their contribution and relevance to the review being undertaken. In each case, they are taking a stance very similar to this article in stepping back from specific ethical prescriptions to ask more open questions that determine the context and purpose of the ethical review before determining the appropriate ethical behaviour and requirements.

In the warrant of ensuring public trust in and willingness to participate in social science research, ethical review takes its place amongst a range of strategies for ethical capacity building that include training, providing ethical review and/or support during the research process and encouraging retrospective ethical reflection at the end of projects. The balance between these activities may need to vary between different types of research. Arguably, open-ended research inquiries would best be supported by a greater emphasis on suitably thorough and wide-ranging ethics training, relatively limited review at the proposal stage but greater availability of ethical support as the research proceeds, rather than a high stakes review prior to starting the research. Reviews in advance of undertaking any research are much more meaningful when that research follows a predetermined design.

Any move of the type envisaged from one type of warrant to another based on trust is not without risk. Trust as the basis for professional ethics is not a new concept. It has a rather chequered history. Although ‘trust’ appears to be universally valued as a positive component in social interaction, it is a term that carries many different meanings. Some of these are ethically problematic. Trust can be as vacuous as eliciting good feelings regardless of moral or performative impact − for example, advertisers’ attempts to promote questionable products and brands by associating them with trusted celebrities. It can be as paternalistic and coercive as in the once much used phrase, ‘Trust me, I am a doctor’. Both cases are attempts to elicit blind trust which arguably defeats the ethically more meaningful application of critical faculties in assessing what ought to be trusted (O’Neill, 2002b). Trust in research ethics that seeks to promote trust between scientists of all varieties and society needs to be robustly articulated to address issues around the pursuit of knowledge for the benefit of that society. A focus on trust directs attention to the challenges posed by inequality and difference between people, particularly researchers and subjects, and existential conditions of risk and uncertainty that are intrinsic to all research. Thus, issues of rigour, respect and responsibility are considered within an understanding of what it means to be human in contemporary society and to undertake research in this context.

Avoiding ethical despair whilst fostering ethical mindfulness

Schrag’s article and book constitute a substantial challenge to everyone involved with the ethical review of research. It is not new information that social scientists have been particularly vulnerable to misunderstandings and inappropriate review outcomes. The authors of a leading text on research ethics for social science widely used in the UK open with the observation that, ‘Social scientists are angry and frustrated. They believe that their work is being constrained and distorted by regulators of ethical practice who do not necessarily understand social science research’ (Israel and Hay, 2006). Parallel views and remedies pervade the most comprehensive international consideration of social research ethics (Mertens and Ginsberg, 2009). Schrag has added weight to those concerns. He provides significant evidence of the type and volume of concerns, but is this sufficient to justify abandoning ethical review of social science research? I think not.

There are empirical questions to be answered by both those in favour of, and those opposed to, ethical review in the style of institutional review boards about the balance of harm and benefit achieved by ethical reviews. Without diminishing the harm that appears to have been inflicted in the cases of the type cited by Schrag, we cannot discern the proportions of harm and benefit gained by review across the social sciences. It would require a significant proportion of harm to justify abandoning ethical review and that evidence is not yet there. Nonetheless, even if the cases reported are only a minority, they are of sufficient seriousness to suggest substantial weaknesses in the way reviews are undertaken and the quality of some judgements made by some reviewers. I am reasonably confident from my contacts with academics in North America and in the UK, that the level of difficulties experienced by social scientists in North America are probably greater and more widespread than in the UK but are by no means universal or inevitable. As part of my professional development for taking on the role of Research Ethics Officer I attended a conference on Qualitative Inquiry where I heard accounts of ethical review under the legally mandated IRB system, some of which matched the difficulties outlined by Schrag. However, others had experiences which had been much more positive. I was left with a very clear impression of the diversity of ways in which IRBs undertook their roles.

In the UK, there are undoubtedly problematic areas and it appears that these problems are at their most acute when social scientists are reviewed by panels more used to biomedical research or quantitative research and misapply these expectations to qualitative research. Anecdotally, there is evidence of inconsistency between review outcomes, but the extent to which this falls within the usual pattern of bureaucratic processes involving complex judgements or exceeds acceptable tolerances is unclear to me. On the other hand, ethical review of social sciences can claim some success.

In a research intensive like the University of Bristol, the Faculty and School research ethics committees consider around 1000 proposals each year from social science staff and students whose studies involve human participants. Each year the review process has discovered and prevented a small number of potential situations in which participants were vulnerable to significant harm. The reviews have supported the refinement of many proposals, reinforced the teaching and learning of research ethics and have actively sought to support ethical mindfulness as a desirable disposition amongst all social scientists. Quality audits of the review processes are generally positive. In the last six years, there have been no reported incidents of research subjects suffering harm from a social science research project. The known incidents that have come closest to causing harm have related to the loss or theft of laptops containing research data that could have resulted in invasions of privacy. No-one has reported actual harm or intrusion on personally sensitive data as a result. There is no room for complacency over such events but the review process provides one of several ways of developing better practice and disseminating knowledge of difficulties that have arisen. One of the successes of the review process has been to build a group of staff and lay people who develop a much greater experience of the diversity of the ethical issues in social science research than any individual academic could have acquired by working within a particular disciplinary specialism. It requires a breadth of knowledge and experience to constructively engage with ethically challenging or contentious issues around the development and delivery of social science research.

The difficulties reported by Schrag are neither inevitable nor unavoidable. I have presented a particular view of the key issues that need to be addressed to avert the undesirable consequences of some approaches to ethical review. I am aware that there are other approaches to ethical review which avoid some of the pitfalls identified by Schrag. Review processes can be revised and indeed routinely require periodic renewal and revision. The characteristic of all beneficial review systems seems to be the involvement of reviewers who have a carefully considered understanding of the warrant for their review that takes account of its strengths and limitations. This critically informed understanding of the warrant provides a point of reference for the complex task of reviewing the diversity of contemporary research.