A disquieting clash of culture

The UK's aid programme draws heavily on academic consultants but, as anthropologist David Mosse is all too aware, this can lead to conflicts of interest.

In the late 1990s, Labour's Department for International Development (DfID) made a new commitment to poverty reduction, putting an end to an earlier pattern that had often tied aid to packages of British exports. In another sense, however, the part played by exports in our international aid had been booming since the late 1980s under the previous Overseas Development Administration - namely, the export of expertise as Technical Co-operation, much of it in new poverty-oriented programmes, involving specialists, overseas training, research and consultancies.

Michael Lipton notes that by 1990-91, TC accounted for more than half of British bilateral aid worldwide.

The shift from hardware to human skills no doubt enhanced the effectiveness of aid programmes, but the principal beneficiaries of the TC boom were the British higher education system and the consultants and trainers. As the aid programme contracted university-based centres of expertise in fields such as social development, fisheries and forestry, several academic departments in development studies and allied fields adopted funding models that made research, appointments and salaries dependent on consultancies and commissioned research. Inevitably, those departments' research priorities became oriented towards the policy agenda of the funders.

Recent shifts in the modalities of aid giving away from funding particular projects have not reduced the need to contract expertise. In fact, the "compliance costs" of new international aid transfers (including debt relief and policy-based grants linked to poverty reduction strategies) may involve even greater numbers of donor country experts with ever greater influence.

Anthropologists were relative latecomers to such development consultancy.

In the mid-1980s only a handful worked as "social development advisers", hired as consultants for the UK's aid programme, by other bilateral donors, the World Bank or non-governmental organisations. Today, there is a fairly well-trodden career path from academic anthropology into work in international aid as consultants, advisers or managers. The awkward divide between academic anthropologists and those working in applied fields has begun to disappear, now that more academics work as consultants. For many anthropologists the world of international aid is an arena for action as much as a field of study; often it has to be both.

So work as consultants for development agencies is by no means unusual among university-based anthropologists as well as economists, geographers and political scientists. It is encouraged, and may sometimes be a contractual requirement of academic staff. Moreover, the experiences of consultancy work contribute directly to teaching and provide a source of data used in published research.

In the natural sciences (plant sciences, agriculture or forestry) this linking of consultancy and research has rarely been a problem, at least in my experience of working in India. The roles of consultant and researcher easily overlap as donor-funded research uses aid projects to provide field sites to develop or test research-based technology, and academics publish the results in scientific journals. But in the social sciences the relationship between consultancy and research is potentially fraught with problems. In particular, anthropologists who have a research interest in the systems and processes of international aid itself - and who wish to bring a critical perspective to bear on the development industry of which they are themselves a part - can face a conflict of interests.

In 2002-03, I wrote a book based on work over 13 years as a social development consultant on a flagship DfID project concerned with improving the livelihoods of very poor tribal people in western India. In the run-up to publication there was consternation from my consultant colleagues and managers, who felt that I had departed from the received view of the project as an overwhelming success. The point was not that my book condemned the project as failing to improve the lives of the poor - it did no such thing - but rather that it rejected the view of success as being the outcome of good policy and technical design. Instead, it drew attention to the contradictions and compromises of practice. I was interested in how events were driven by the need to maintain relationships rather than by policy and how, in development, success (or failure) is produced by imposing interpretations on the complexity and confusion of practice so as to endorse prevailing policy (which is contradicted in practice). The capacity to do this, present in all realms of public policy, is greater in international aid, given the greater separation between decision makers and end beneficiaries compared, say, with that between patients and managers in the domestic health service.

In early 2004, objections were made to the publisher, to my university research ethics committee, to the academic head of the university, as well as to the Association of Social Anthropologists, on the grounds that the book was unfair, biased, contained statements that were defamatory, and would seriously damage the professional reputation of individuals and institutions. The university, the ASA and the publisher were not persuaded that this was the case; in fact, they refused to adjudicate on the matter.

The university required only that the parties meet and listen to each other, and insisted that such a meeting could not lay on me, the author, any obligation to make particular changes to the book.

This dispute reveals certain tensions between the obligations of consultancy work and academic freedom that may be of special importance, given the growing importance of commissioned work (consultancy or research) to social science funding and practice. It is not a case of simple censorship, but rather of professionals under social science scrutiny attempting to use codes of research ethics (or even, implicitly, defamation law) to gain control over the outputs of research (particularly research undertaken by academic consultants with privileged access).

The first question is, do social scientists have the right to use information and experience gained through consultancy work as they wish? The terms of consultancy contracts in publicly funded international aid certainly contain clauses potentially limiting the use and publication of information, although nowadays these are far from clear cut. The principal restriction is that data are not to be used in ways that might "embarrass"

the British Government or its overseas partners. But what constitutes embarrassment? Do programme managers have the right (or capacity) to prevent publication of analyses with which they disagree, or to insist on editorial intervention, on the grounds of embarrassment?

Second, can publicly funded professionals or civil servants preserve their reputation, or that of their successful programmes, against academic critics by claiming that harm has been done in contravention of research ethics codes, codes that are increasingly framed in the rubric of biomedical ethics for the protection of human subjects used in research?

Certainly, the dilemmas thrown up by consultancy-based research are not adequately dealt with by ethical guidelines within anthropology, which were drawn up on the basis of a narrower conception of power relations in research, between powerful Western researchers and vulnerable native subjects of research, and designed to protect those without other means of redress.

Research ethics revolve around securing consent and the avoidance of harm.

On the question of consent, social researchers who draw on work as consultants might find their research challenged on the grounds of an "abuse of position" (privileged access to confidential information), abuse of contract, unethical disclosure or covert research. Even if (as in my own case) consent to use research based on consultancy work is carefully negotiated through separate contracts, problems arise when co-workers or managers attempt to withdraw "consent" to the research findings, or to publication, on the grounds that the research will cause harm to their professional reputations. This means that any organisation or programme that does not like the way it is represented in research may claim "harm to professional reputations" or even invoke defamation law.

Does an obligation to share research results with those involved in the research involve an obligation to accept editorship from those who do not feel flattered in the analytical descriptions researchers produce?

If such objections were to prove successful, the independent analysis of public policy and professional communities (especially ethnographic research based on a researcher's experience as an insider) would become virtually impossible. And insofar as independent critical reflection is a resource for change and improvement, the capacity of aid organisations to learn or to be held to account is correspondingly restricted.

In my particular case, academic rules were more powerful than even quite determined objectors. Such overt conflicts are very rare. But then, research on aid and development, overwhelmingly funded by donor organisations, is strongly oriented to policy theory and gives remarkably little attention to the informal processes of aid organisations, bureaucracies and brokers in a huge international business. Few social researchers have access to aid organisations as professional insiders; and few who are insiders choose to offer critical reflections on their experiences. Anthropologists who are committed to policy work and to critical reflection have to negotiate a wedge that is driven between outsider researchers and professional insiders.

David Mosse is reader in social anthropology at the School of Oriental and African Studies. He is author of Cultivating Development: An Ethnography of Aid Policy and Practice , Pluto Press.

Next week: is higher education succumbing to the third way?