Introduction

This article is primarily concerned with qualitative approaches to data collection and fieldwork. These reflections are based on field research in countries affected by poverty and conflict: Afghanistan, Angola and Mozambique. At the same time many of the dilemmas touched on are general; most of the elements discussed here will also be relevant for research based on secondary data or using quantitative methods.

Ethical challenges arise at all four stages of the research: planning, fieldwork, analysis and publication. These stages are never sharply distinguished in time, and it is important to realise that in order to safeguard research ethics, the researcher must as far as possible maintain a general overview of the research process. Methodological choices made at an early stage, e.g. in the planning, may cause considerable ethical problems at a later stage. I will conclude by focusing on some general research issues.

Planning

The very first question a researcher should ask is what the research project is intended to achieve. The objective should be to give the most truthful account possible of a phenomenon. At the same time, we must acknowledge that truth can only serve as an ideal. It is vital to acknowledge our own limitations by clarifying the theoretical orientation and choice of methods as well as other factors. Research in the South has often been inspired by a desire to expose repression, to give a voice to those who are seldom heard.

An example of this is the author's own research on the Hazara minority in Afghanistan. In the course of this work, disturbing facts were uncovered, for example that the whole of the 1980s was marked by internal strife in the Hazara population in which thousands lost their lives. For the researcher this was an important part of their history. The reasons for the strife were not just internal since the conflict was also driven by the exclusion of the Hazaras by other Afghan parties and the interference of neighbouring states. This is an important part of the Hazaras' political history, even though a large number of the population would prefer it to be forgotten. If no one wrote about this, it would be a considerable distortion of their history and would entail a cover-up of the fate of the many who were affected.

Equally important is the question of who the researcher is working for. The ideal is complete independence, where those funding the research pose no demands. But the majority of those who fund research make demands – for quality if nothing else. A considerable proportion of research in the South is commissioned and funded by actors who want to acquire knowledge that can be used to plan or evaluate various projects. A useful rule of thumb is that it is acceptable that the client exerts influence on the research questions asked, but not on the conclusions. Who you work for is also important in getting access to local communities and organisations.

An important source in the work on Hazaran policies in Afghanistan was one of the main ideologists of Hezb-e Wahdat (Islamic Unity Party), a man rigidly schooled in Iran's Shia educational institutions. He knew the party's visions and strategic philosophy better than anyone else. However, when meeting a researcher who claimed to represent the Peace Research Institute Oslo (PRIO), he expressed considerable scepticism. The foreign researcher was subjected to a cross-examination about what kind of institution he represented. In particular the ideologist questioned the assertion of 'independence', drawing heavily on his own schooling in western philosophy, and he cross-examined the researcher about Jean Paul Sartre. He remained sceptical about whether it was possible to be independent but in the end an interview was arranged. This episode shows the importance of who you represent. This may have major consequences not only for opening doors for you but also for the kind of answers that will be given once you have gained access.

Who do you want to work with? It is important to ask yourself this question at an early stage. I myself have been involved in many projects with local researchers or research assistants but these have first been involved after the design of the research has been determined. When I have involved local partners in the design itself, this has proved to be very useful. The effects are greater. Firstly local researchers (provided that you have recruited the right people) have a greater knowledge of local factors that will enable you to ask more relevant questions. Secondly local partners acquire a totally different kind of ownership through participation in the planning. Thirdly all parties achieve a much greater learning outcome. It may be difficult to involve local partners in the planning on practical grounds but even when the parameters are largely set, a great deal can be gained from a thorough review together with local partners before engaging in the fieldwork.

Fieldwork

Respect for other people is a good starting point for data collection – whether this takes place by means of interviews or observation. This does not mean taking everything people say at face value but that you show understanding for local needs and customs even when you do not sympathise with them. People with few resources often have little chance of protecting themselves. In poor countries the rich can hide behind high gates and dark tinted car windows while poor people barely have a tiny private nook where they can protect themselves from inquisitive researchers. This poses a dual challenge: how to come into contact with those who dominate local power and resources, and how to get in touch with those with no resources without trampling on their dignity?

It is also important to show respect for people's time. For those who live at the subsistence level where today's scant resources are just enough to survive until tomorrow, spending many hours being interviewed by a researcher may come at a high cost. This means that the researcher must be aware of seasonal variations and avoiding fieldwork in the most work-intensive period, for example during the harvest. There is great scepticism in social science when it comes to paying for interviews but situations of extreme poverty exacerbate the dilemma. The simplest response is to be generally helpful, by offering transport, for example, or giving information. The next simplest (from a purely ethical stance) is to find a collective compensation, for instance by providing a new water pump from which an entire village can benefit. However, it is not always possible to avoid recompensing individuals. The challenge is then to define forms and degrees of compensation that are locally acceptable and that do not distort the sample or lead to people telling you what they think you want to hear. To succeed in this you already need considerable local knowledge.

The researcher also has an ethical responsibility not to inflict undue risk on respondents and partners. Foreign researchers are merely in transit and have no opportunity to influence what happens after they have left. The greatest possible transparency about the research topic is a positive contribution but it may be difficult to be completely open towards everyone, especially when working under oppressive regimes and dealing with sensitive issues. Another ploy is to work with many respondents and partners so that no individuals stand out as making the most significant contribution to the research. You must also consider how to relate to people. During fieldwork for my own doctoral degree in Afghanistan at the time it was controlled by the Taliban, I politely declined all offers from the host population to move into the village, even though this would have been very useful for the data collection. Afghans are the world's most hospitable people and it was not always easy to decline the invitations, but the awareness that the Taliban harboured considerable scepticism towards foreigners left me with no choice.

Analysis

There are no special expectations as to research in the South; as always it is important to give the best possible account based on your own research questions and the data available. However, the difficulty of achieving this can be even greater than for other research. Often you will experience that the data are relatively deficient since much energy in the fieldwork has been devoted to getting things to function in a practical way, and the scope of the secondary data may be poor. There may also be considerable pressure to produce unambiguous findings and clear alternatives for action. This is especially true when fieldwork is conducted in the framework of commissioned research, but it is also a challenge when working with established organisations in order to gain access to the field (which many students and researchers are dependent on). Then you must abide by research quality requirements to withstand the pressure that almost always exists to tone down reservations and simplify conclusions.

Local populations and local partners may also play a key role in the analysis. As early as the fieldwork stage it may be wise to present preliminary findings and use this as a starting point to acquire input and reactions from respondents. Local researchers or research assistants can be involved to an even greater degree and at more stages of the process. If conducting interviews together, a joint summary of each interview can be written and analytical ideas noted. If appropriate, an equivalent summary can be written on a daily basis, and preliminary conclusions and ideas can be entered in the fieldwork journal. There must be a clear awareness of the premises when local research staff are engaged in the analysis. Is it the 'researcher from the North' who has final responsibility and who has therefore the final say in deciding what input should be reflected in the end product? Or is this a collaboration between partners who are equally committed to the end product and therefore must also reach agreement about the findings? The latter is an ideal that is too seldom realised in practice.

Publication

The international scientific community has clear norms for high-quality publication. Journals and publishing houses have comprehensive quality assurance. These publication channels offer many benefits but vis-à-vis research in the South there are also considerable drawbacks. Firstly their special quality and genre requirements mean that it may be difficult for researchers in the South to be fully recognised as contributors. Secondly academic journals and books are often so expensive that individuals and libraries in the South can seldom afford to buy them. 'Open Access' fortunately generates new kinds of quality-assured academic publication with improved accessibility in the South. Another solution is to combine several publication channels. For example an academic article can be combined with a shorter, more popular report that can also be published in local languages.

However, many researchers refuse to become involved in the public debate, not least because many of them feel that the nuances and reservations that I have emphasised above easily disappear in a 'tabloidised' media world. One's own identity as a serious scholar seems threatened. There is no doubt that popularisation drives a considerable degree of simplification. At the same time my experience is that most journalists are interested in hearing about the researcher's perspectives, and a well-prepared researcher has every opportunity to influence how the research is presented. Researchers also have social responsibility and if they possess unique expertise that is needed by society, it is difficult to defend the 'ivory-tower' strategy. Ultimately this is also a question of the obligation to those who have spent time and energy sharing their own lives with the researcher – they have done so with the legitimate expectation that the knowledge will be shared.

And so?

Many researchers in the field strongly feel that the individual researcher has a responsibility not to spoil the field, and not to act in such a way that will make it difficult for others to gain access. But is the most important obligation of the individual researcher vis-à-vis the research community perhaps to safeguard truth as the ideal in a broad perspective, obliging the researcher to live up to solid quality criteria? Is quality itself thus an important guarantee of ethics and perhaps the key ethical commitment?

A common theme in the reflections above is the concrete nature of ethics. The ethical profile of a research project is shaped in the interaction with numerous parameters: these include the funding source, partners and state control where the work is carried out. Ethics are not simply something the researcher can leave behind at the planning stage – they exist in the concrete choices the researcher makes every day in the research process.

Overall this places strict demands on ethical preparedness. All fieldwork – particularly when it takes place in the South – gives rise to unforeseen challenges that test the researcher. It is then vital to have thought through the choices made and how to meet the challenges that may arise. Very often this applies to fundamental dilemmas to which there are no correct answers. It is precisely for this reason that ethical preparedness becomes so vital.

This article has been translated from Norwegian by Beverley Wahl, Akasie språktjenester AS.