Educational research is fiction – written under oath. The question is: what is the oath? (Barry MacDonald)

Barry MacDonald ’s aphorism, dropped out at a seminar at the University of East Anglia several years ago, has echoed through my own thought ever since, posing questions as to the relationship between ethical principles and epistemological principles in the conduct of research, which is the territory I want to explore in this paper. It is an enquiry into the relationship between the epistemic and ethical requirements which shape research activity which leads me to be sceptical of the invitation to conclude that one might in some way ‘trump’ the other.

Educational Research As an Epistemic Project

In this paper I use ‘epistemic’ to refer to purposes or projects that are concerned with development of knowledge and understanding and ‘epistemology’ to refer to the theory of knowledge and understanding.

Educational researchers aim to extend knowledge and understanding in all areas of educational activity and from all perspectives including learners, educators, policymakers and the public. (British Educational Research Association 2011)

There are, importantly, a number of epistemic purposes that govern the conduct of educational or any other research. These might include, for example, requirements:

  • To unsettle or question established belief

  • To conjecture about possible alternatives and develop new ways of seeing things

  • To describe or illuminate aspects of experience

  • To search for reasons, evidence and/or argument – for warrant – that might support one belief rather than another

  • To test beliefs and establish at least provisionally the truth of the matter under investigation

In the context of work honoured as ‘research’ (this is, I think, an honorific concept) any of these epistemological projects come with a requirement that they are conducted in a manner that is ‘systematic and sustained’ (Peters and White 1969; Stenhouse 1980), with rigour.

The range of tasks I have illustrated above, clearly reflects some different priorities for research; some different ontologies perhaps and certainly some different methodologies and methods of inquiry. Some attach more significance to reason and evidence; some to imagination and empathetic engagement; some (in philosophy for example) to argumentation, but (nearly) all result in some form of representation of what is, might be or ought to be the case. Hence they affirm in some form what the researcher believes to be the case , what is (provisionally at least) true.Footnote 1 This holds even for research that is self-consciously seeking to represent alternative perceptions of educational experience in a non judgemental way. Such research has its own disciplines, its own rigour, aimed at a faithful (a true?) representation of such perspectives.

As I have argued previously (Bridges 1998), even those who seek most vigorously to escape the discourse of truth – like Stronach and MacLure in Educational Research Undone 1997) – end up, as inevitably, as they must offering claims as to what is the case, what is true, unless perhaps they stick to the questioning format that they adopt in their opening paragraphs.

Sometimes, of course, we look for qualities in research writing (as in literary and other work) which are not directly stated in the language of truth seeking but are, none the less, dependant on such a notion. We look for authenticity in, for example, biographical, historical and ethnographic portrayals; we look for honesty in research reports; we look for integrity in the relationship between research, participants and research publication . But none of these ethically laden concepts are intelligible without invoking some notion of truthfulness and, hence, truth.

The association I am making here between diverse forms of research and inquiry shaped by a search for the truth of the matter may well be contested, but this is the starting point for the discussion in this paper. Even for MacDonald , the description of research as ‘fiction ’ was as much as anything an exhortation to doubt it, question it, look always for the counter argument and evidence, gather alternative perspectives. The ‘oath’ is perhaps to distrust everything they (including or especially researchers) tell you and indeed what you yourself assume or believe – but why? If one is to be so rigorous in one’s scepticism is it not nevertheless in the interests of discarding untruths in preference for what, at least provisionally and pending further sceptical treatment one might regard as true? From Socrates to Popper philosophers have extolled the benefits to be gained from others’ refutation of their own beliefs, because the discarding of falsehood leaves them with what on some epistemological criterion is more deserving of belief. Neither Socrates nor Popper conclude that they cannot believe anything and even Descartes found comprehensive scepticism incoherent.

This is not to say that truth is the only criterion of the merit of a piece of research. The truth can be banal, predictable and boring: research might seek to excite the reader with new perceptions and ideas; to bring wider or deeper understanding, to cultivate a more sensitive or imaginative rendering of experience. Of course it is not alone in this, and a good novel, drama or painting even may offer more when judged against these criteria than a piece of research, but whatever research offers it has to be judged inter alia on the basis of its success in providing compelling reasons, evidence or argument to support what it affirms and its claims to truth.

So, I take the view that educational research is always in some sense focussed on the development of knowledge and understanding – what is commonly called the epistemic project. My central question concerns the relationship between this project and a variety of other, broadly ethical and social obligations.

Ethical Obligations As a Constraint on the Epistemic Project of Research?

In particular in a university research environment ethical principles – usually articulated in ethical codes and policed by ethics committees – function to constrain what researchers might otherwise do in the untrammelled pursuit of truth. They set conditions under which research sites can be entered and what information can be accessed; they protect vulnerable participants from intrusive inquiry; they protect confidentiality; and they set limits on what can be published. To this extent – and of course this does not obstruct all worthwhile inquiry – ethical obligations trump and constrain epistemological ambitions. In the most difficult cases they simply make certain areas of inquiry impossible to enter.

There are, however, two important qualifications to this assessment . The first is a purely pragmatic consideration. It is, I think, the experience of many researchers that by offering the sort of protection to potential sources contained in these codes you give potential participants greater confidence in sharing their information and perspectives, and so you end up with richer data than you would otherwise have gained.

This response is however a culturally located one associated with a climate of distrust and anxiety about exposure which increasingly pervades western countries. International students studying in the UK, however, frequently have problems returning to their own communities clutching ‘consent forms’ that tell their communities only that the researcher is not someone to be trusted and is possibly trying to get them to hand over their land (Adugna 2008). Attia describes in similar terms the basis for her relationship with participants in an Egyptian university in which she had previously worked and its inevitable embeddedness in Arab culture:

For example, “asham” is a well-established social concept in Egyptian culture. It may be defined as an expectation and hope that one gets a preferred response, that is, acquiescence to a request …. On the basis of “asham”, full access was guaranteed and complete assistance was granted. (Attia 2011: 97)

And then:

The anticipation of assistance that I returned with was based on a history of shared lived experiences which in Arabic may be referred to as “ishra”. The concept is related to a kind of expected solidarity and mutual assistance stemming from belonging to a “asheera”, that is, a tribal community, clan, or kinsfolk. (Attia 2011: 98)

In such settings openness is secured not by the reassuring terms of contractual engagement (the consent form) but by culturally embedded relationships of trust. The bureaucratisation of ethics in UK and other universities is both a symptom of and a contributor to the breakdown of relationships of trust between researchers and their communities. The rejection of a bureaucratically shaped ethical code and the requirement for a signature of consent does not, however, mean that it these contexts the researcher is free from moral obligation. Such obligations of care, loyalty and trust are if anything stronger when they are part of the social fabric and not just a short term contract, and they still serve to limit what can be said and what use can be made of what is known.

The second qualification to the suggestion that ethical codes constrain researchers is to be found in some of the clauses that have relatively recently be introduced into the ethical codes of, for example, the British Educational Research Association (BERA) in response to the tendency of sponsors of educational research to demand excessive control over what is published out of the research. The BERA Ethical Guidelines for Educational Research (2011) gives strong endorsement to researchers’ right and obligation to place their findings in the public domain .

The right of researchers independently to publish the findings of their research under their own names is considered the norm for sponsored research, and this right should not be lightly waived or unreasonably denied. This right is linked to the obligation on researchers to ensure that their findings are placed in the public domain and within reasonable reach of educational practitioners and policy makers, parents, pupils and the wider public.

Researchers must avoid agreeing to any sponsor’s conditions that could lead to serious contravention of any aspect of these guidelines or that undermine the integrity of the research by imposing unjustifiable conditions on the methods to be used or the reporting of outcomes. Attempts by sponsors or funding agencies to use any questionable influence should be reported to the Association. (BERA 2011 my italics)

In this way ethical codes can – and should – be used to support rather than to disable the epistemological project.

Can Ethical and Social Goals Substitute for Epistemological Ones in Educational Research?

‘Where, previously, ethical considerations were believed to set boundaries on what researchers could do in pursuit of knowledge, now ethical considerations are treated by some as constituting the very rationale for research…. The possibility and, perhaps the desirability of knowledge have come to be downplayed by instrumentalism and postmodernism, [and] a concern for ethics has expanded to fill the space. (Hammersley 1999: 18)

What should drive educational research? To what ends should it be directed? At one end of the spectrum, I suppose, this might be answered by reference to something like the pursuit of truth, however, esoteric, trivial or remote from contemporary concerns. Whether or not it is ‘relevant’ to contemporary problems, whether or not it contributes to ‘evidence based’ policy or practice , whether or not it has any ‘impact ’ on anything is neither here nor there.

At the other end of the spectrum are two sets of people both demanding ‘relevance’ and expecting educational research to contribute to policy and practice, but with rather different political agenda.

One group is framed by the discourse of evidence based policy (see on this Bridges et al. 2009). It has its sights set on improved educational performance; it has an already established political agenda (focussed in the UK on policies like the establishment of educational academies and free schools, frequent testing and a politically defined curriculum); but it is seeking information of an instrumental rather than critical character from the research community about how best to implement its policies as well as, by preference, validation of their success.

The second group, which might find itself uncomfortably aligned with the first, is also concerned with relevance and impact , but has a different political and social agenda focussed on, for example, democratic values, social justice and inclusion. This seems to me to be of particular interest in the context of this seminar.

There are I think at least two lines of argument that link educational research to what, by way of shorthand, I will call the social justice agenda. The first is to urge researchers to work towards the goal of a more just society by giving their critical attention to the aspects of contemporary policy and practice which contribute to injustice and their creative imagination to alternatives. The achievement of greater social justice (etc.) thus provides the substantive agenda for research and its hoped for outcome. In other words, the substance of the research is expected to inform understanding of unjust structures and practices and help to address them.

Note that this model is not so different in principle to ‘evidence based policy’ though this has been largely inspired by a neo-liberal political agenda rather than an egalitarian one. Those who want educational research to be ‘for justice’ are, like their neo-liberal counterparts, seeking ‘relevance’ and ‘impact ’ – in Hammersley ’s terms, they still judge research in instrumental terms – but they want the impact on education and society to be of a different kind.

Even if we allow that the imperative to be ‘for social justice’ through educational research is a powerful one, where does this leave the epistemological principles and ambitions which I outlined at the beginning of this paper. Is the advancement of social justice or democracy a sufficient condition for determining the merit of educational research, a sufficient focus for the concerns of the researcher?

Clearly the concern for social justice or any other such principle cannot be detached from a researcher’s fundamental commitment to knowledge and understanding. ‘All our choices depend on estimates of what is the case’ argues Bok (Bok 1978: 19), and she might have added also estimates of what are the most likely consequences of our actions. A judgement that a particular practice , policy or set of relationships is unjust always has two elements: one is an empirically based observation (on which researchers have some claims to authority) about the differences in the ways in which people are treated or the circumstances in which they live; the second is a normatively based assessment of such differences as being unjustifiable, inappropriate or unfair. On such questions researchers can make no particular claim to authority, though philosophers in particular find the nature of such normative arguments of particular interest. To make an assessment about some injustice, we typically need to know that facts of the case; we need to have some understanding of what Quinton (1973: 4) calls the ‘rationally expectable’ consequences of acting in one way rather than another, and it might well be useful to know about alternative ways of doing things that would avoid the criticism of the case under scrutiny. Even while advancing the case for educational research for justice, Griffiths acknowledges that ‘educational research is about getting knowledge’ or perhaps, as she calls it ‘better knowledge’ (Griffiths 1998, 129). Veck , recounting his own attempts at ‘emancipatory’ research , explains that he came to the conclusion that:

In committing to social justice, I was logically bound to the pursuit of truth. If the outcome of my research was to uncover injustice, to pronounce what is wrong, then what I had to say had to reflect the reality of that social injustice with the utmost accuracy. (Veck 2002: 334)

In short, the pursuit of social justice and a more democratic society requires as much as any other social political cause and the sound basis of knowledge and understanding as well as the critical questioning and argumentation that ‘research’ is properly designed to provide.

Principles of Social Justice As Integral to the Construction and Design of Research

More interesting in some ways, however, is the way a social and political agenda becomes integral to, not just its focus or purposes, but to the process of research, to the way in which research is constructed and conducted, which may itself becomes a contribution to the achievement of a more just and inclusive society.

This is not just a matter of observing certain clauses in an ethical code but a much more substantive project in its own right that carries implications for: the identification of the research questions (whose questions?); the identification of those recognised as ‘researchers’ (not just those from the academy); the voices that are given expression in the research; control over access and data; who is involved in the interpretation of data; whose authorship or other contributions are recognised; how the research engages with policy and practice ; what happens to any research outcomes. Each of these (and other) features of a research project can be designed so as, as far as possible, to re-balance unequal power relations, give voice to those who have previously been excluded; open closed areas of policy and practice to more democratic scrutiny and inform and engage a wider community. These are indeed key areas of attention for that post-colonial researchers, feminist researchers, gay and lesbian researchers and researchers with disabilities, who have sought to change power relations, not only by the published outcomes of their research but by the processes through which it has been conducted. The slogan that emerged from the disability camp was ‘Nothing about us without us!’ (Charlton 1998) while in New Zealand, according to Marshall and Martin (2000), a growing body of Maori researchers are operating under the motto ‘By Maori; in Maori; for Maori’. This is at least one sense in which educational research can be ‘for social justice ’ (Griffiths 1998).

There is an important sense in which researchers do not have to take sides in educational debates –not even on the side of the marginalised or disadvantaged – in order to promote democracy and a more just society, and this is part of what MacDonald ’s et concept of ‘democratic evaluation ’ set out to achieve.Footnote 2 Of course not all the ‘facts’ in any educational development will be unambiguous and they may well be contested. Different interest groups may very well have different perceptions of what is the case and why. But how different groups perceive the situation is also a matter of fact that can be investigated empirically – and not necessarily judgementally. For example, MacDonald describes ‘democratic’ evaluation in the following terms:

Democratic evaluation is an information service to the whole community about the characteristics of an educational programme. Sponsorship of the evaluation study does not in itself confer a special claim upon this service. The democratic evaluator recognizes value pluralism and seeks to represent a range of interests in his issue formulation. The basic value is an informed citizenry, and the evaluator acts as broker in exchanges of information between groups who want knowledge of each other. (MacDonald 1976)

Interestingly, in terms of the distinction I have drawn between social justice as the substantive focus of research and as a procedural principle governing its conduct MacDonald also suggests that, though there will be a place in the future for different types of evaluation study (including ‘autocratic’ and ‘bureaucratic’), ‘there may be a special case for exploring in practice some of the principles which characterise the democratic model. For those who believe that means are the most important category of ends, it deserves refutation or support’ (MacDonald 1976). Democracy is served in ‘democratic’ evaluation not just by the knowledge and understanding that it places in the public domain , but by the processes by which this is generated.

Belief As an Ethical Obligation

The debate in the literature is sometimes expressed in terms of whether ethical considerations should ‘trump’ epistemological ones, not just in research but in the wider domain of policy and practice (Chisholm 1956). Hall and Johnson (1998) seem to suggest, for example, that: that when you epistemically ought to gather more evidence and you morally ought to do something else, the moral ought “wins” and you just plain ought to do that other thing (Hall and Johnson 1998: 131). But this seems to me to be an inadequate account.

There is an important sense in which the contrasting views that we are examining is not simply between an epistemic project focussed on achieving knowledge and understanding and an ethical project focussed, let us say, on a just society, but between two sets of ethical obligations, though I will continue to refer to the ‘epistemic’ project in the interests of clarity.

William Clifford, in his seminal “The Ethics of Belief ” – a paper published in 1877 in Contemporary Review – pioneered a body of literature (including James 1896a, b; Chisholm 1956; David 2001; Feldman 2000; Heil 1983; Sosa 2000, 2003) that has explored the question of the conditions under which belief is not just subject to epistemological validation but becomes also a matter of ethical obligation. It is not simply that one is entitled to believe that something is the case but one ought to believe that something is the case or ought not to believe in the absence of sufficient evidence or faced with compelling contradictory evidence: the question ‘what ought we to believe?’ is intelligible both as an ethical and as an epistemological question. Similarly, a stubborn refusal to face facts, self-serving distortions of an argument, careless use of evidence all carry an element of moral censure and not merely consequences for the erroneousness of one’s beliefs.

If we take seriously the arguments from the literature on the ethics of belief , the juxtaposition becomes not one between ethical considerations and something quite other, but rather the compatibility between two ethical principles; one of which demands that we seek as far as possible to ground our actions on properly warranted beliefs and offers this as a moral responsibility; the other is expressed in a moral imperative, for example, to act justly and with respect for others. But Feldman (2000) argues that talk of one of these sets of obligations ‘trumping’ the other ‘makes no sense’ (p. 692): ‘There is no meaningful question about whether epistemic oughts “trump” or are trumped by other oughts’ (p. 694). There is on this view no greater ‘ought’ that can settle the issue (though other ethicists, for example in the Utilitarian tradition, might disagree.)

Secondly, as I have argued, social and ethical causes themselves depend on rigorous, open and honest inquiry. Indeed such inquiry is in an important sense one of the constituents of a more democratic society and not merely a means to the end. In the context of educational research the management of inquiry so as to ensure that, for example, the voices of the weak, the powerless and the marginalised are heard is not just a contribution to the cause of social justice : it also creates the conditions for the success of the epistemic project of inquiry. We need to hear the voices of the disenfranchised; we need to escape the hegemonic views of the powerful; we need the widest possible public debate and an open society if we are even to begin to get to the truth of the matter. To this extent then the social justice agenda supports rather than conflicts with the ambitions of researchers to get to the truth. An unjust society provides neither the open access to evidence that the researcher requires nor the opportunity to make the research freely available for the informing of a democratic citizenry and for critique.

So are there other sources of conflict which might lead us to want ethical, social or political considerations to ‘trump’ epistemological ones?

I can only make sense of this as a legitimation of dishonesty – if not, in legal terms, of suggestio falsi, expressing falsehood, then of suppressio veri, concealing the truth or not even wanting to know it.

This becomes significant when research evidence threatens, for example, an important ethical, social or political cause which one seeks to advance. A colleague is a passionate spokesperson in public arena trying to warn about the risks of global warming. But his research into the Antarctic ice suggests that this has survived previous ‘warmings’ for two or three million years. He knows that this is evidence which could be used by ‘global warming deniers’ to challenge the urgency of appeals for action. He is faced with two dilemmas: can he really believe what his own evidence seems to suggest? If he does, should he put this evidence in the public domain ? In answering both questions in the affirmative (pending further research, of course!) he is driven not just by academic argument but by an ethical imperative that he has to believe what his evidence shows him, even if this runs counter to both his expectations and what he would really rather believe.

In educational settings debates about race, intelligence and educational attainment have been rendered both complex and intense by confusion between what evidence might sometimes indicate and what, for the very best of moral reasons, one would like it to indicate. If research showed that redheads were scored lower marks on intelligence tests than those with other hair, might we consider it better to suppress the finding, for fear that they would then be discriminated against? I might be ideologically opposed to the privatisation of education. Might I prefer not to examine its impact on children from poorer families for fear that the results might actually be rather positive (albeit only part of the picture)? And if I believe sufficiently passionately in some cause, might I not tweak the evidence a little or use it very selectively to support my case.

I might even invent parts of a story so as better to tell the ‘greater truth’ or what Rolfe (2002) calls ‘a lie that helps us to see the truth’? though to express the ‘fiction ’ in these terms is not to reject the epistemic project in favour of some of some social or political cause, but to have an enlarged view of how the epistemic project of knowledge and understanding might be achieved. ‘I am well aware’ wrote Foucault , ‘that I have never written anything but fictions . I do not mean to say however that truth is therefore absent. It seems to me that the possibility exists for fiction to function in truth, for fictional discourse to have effects of truth’ (Foucault, 1980: 193. See also this Bridges 2002, 2003.)

This falls short of Plato’s ‘noble lie’ which makes no pretence at truth but knowingly deceives – usually from a position of arrogant superiority – in the cause of some expected wider social benefit.

As with any case of dishonesty, there is a spectrum of moral censure. Preferring not to open certain boxes for fear of what one might find does not in general carry much moral censure though it represents a certain intellectual cowardice in a researcher. Avoiding drawing too much public attention to one’s results and managing carefully their presentation is, similarly, against the spirit of research – ‘systematic and sustained inquiry made public’ as Stenhouse defined it (1980 my italics) – but perhaps an acceptable compromise with other moral demands. Deliberately mis-representing one’s findings in the belief that this would advance a particular social cause, even or especially the cause of social justice and democracy, seems to me, however, to be step too far, not least because both principles themselves require (at least in general terms) a climate of openness and honesty. We thus return again to the principle enunciated by Sissela Bok in her excellent discussion of ‘Lying’. It is possible to go beyond the notion that epistemology is somehow prior to ethics. The two nourish each other, but neither can claim priority’ (Bok 1978:13) – a fortiori, one might add when it comes to questions of honesty.

It is perhaps worth considering the particular responsibilities of researchers qua researchers in the field of social and political action. As people and as citizens we can of course participate in many forms of social and political action and it may be argued that indeed we should and on the side of justice. This is a duty that can be laid on everyone, though some see it as part of the full set of obligations laid on the researcher:

Thus I am arguing that the researcher/author has three tasks; the researcher engages the researched in a reflective encounter; the research ‘act’ – the book, article or presentation – brings to light the inequities of power that may exist; and the researcher actively works for care and change. (Tierney , 1994: 111)

But lawyers, journalists, civic servants and politicians make a distinctive professional contribution to these wider social causes – and so do researchers. In this last case it surely has something to do with providing both rigorous critique, carefully and systematically gathered evidence, faithful analysis and, hence trustworthy reasons and argument. As I have argued, this itself requires a commitment to, for example, giving voice to those who might otherwise be excluded, challenging the hegemonic discourses of the powerful and their control over sources of evidence, putting evidence into the public domain and supporting open discussion. These are both requirements for the conduct of a successful educational inquiry and conditions for an open, just and democratic society.

$$ {}^{\ast}\ast \ast $$

So what is the oath? An oath to question all but especially one’s own most cherished beliefs? to be rigorous and inclusive in one’s attempts to understand experience? to be open and honest in the presentation of one’s evidence and argument? And, who knows, perhaps the ‘fictions ’ thus created might yet contain a glimmer of truth, because these are principles that well serve a traditional epistemic project as well as the cause of a more just society.