Marge Tsitouris, CARE USA's head of emergencies, promotes Do No Harm, an analytical tool for NGOs to examine their role in conflicts to avoid accidentally exacerbating divisions and instead promote peace-building. She spoke to Ruth Gidley during a week of training events in Britain.

AN: What is behind the phrase "Do No Harm"?

MT: There is a group in the United States called Collaborative for Development Action. It's a not-for-profit research organisation. They have a project called the Local Capacities for Peace Project. The ultimate aim is to look at ways to connect people and think of peace-building. What that picked up is the first thing a physician should do is do no harm. So very often it's aka "Do No Harm". In some situations, like in Sri Lanka, government people didn't like the idea of local capacities for peace, so you don't necessarily want to say that, but they like the idea of doing no harm.

AN: And how does this principle work in reality? MT: The idea is very practical. It looks at the society during the war, pre the war and post the war, and what are the things that divide and cause tension in that society. So it could be ethnic, discrimination because of something or other, it could be power relationships, it could be numbers of people in one group versus another, it could be natural resource battles. Then you would look at what the connectors are in that same society.

AN: Is this what your workshop during the CARE training week involved? MT: We took a real-life case study of Save the Children in Tajikistan. In the case of Tajikistan it was a group that was in conflict with each other, supposedly for ideologies but in a way not so, but the connecting base is that they share the same religion, the same culture, the same language. Essentially there was something there, and so how do you build those connectors for peace?

AN: What would be a practical example of that? MT: One example was Catholic Relief Service in Liberia. They were doing agricultural work and activities using fruit. We hire people who are like ourselves, almost always. So when the war broke out, and this group came in to work with them, they found out that essentially almost all the staff were one ethnic group, they all went to the same college. In a war situation that might not be the most appropriate mix. So the idea is to look at that, to see what you look like, see what you hold, see what your staff hold, see what your donor agencies hold. If CARE USA was working in Cuba, there would be a lot of suspicion because a good percentage of the money would be coming from the U.S. government. Instead, CARE Canada works in Cuba and we as an organisation can sometimes put a more impartial CARE in a country, that doesn't carry all that political baggage.

AN: How was this relevant in the Tajikistan example? MT: What we looked at was the kind of programme the NGO was doing. It was really favouring the losers, who were now returning. It was building houses, using food-for-work schemes, and those who weren't didn't get anything. What can we do if we were to redesign the programme to help people in need of either side and help to rebuild the economy? We were looking at other things like infrastructure development, repairing irrigation canals, that anyone could benefit from and might build for the future instead of helping only one group over another.

AN: What if you are conscious of this as an organisation, but you might be taking funding from a donor which thinks you should only help the losers? MT: That was the issue in this one, because the donor agency really liked what they were doing. But if you take a really good look at the project it really wasn't building connectors, and in fact was making more potential divisions, so ultimately wouldn't do well long-term.

AN: Could this theory also be applied to accusations that in some instances NGOs have might have fuelled conflicts? MT: The study (on which Do No Harm is based) looked at the idea of resource transfers and the implicit messages that you carry as an organisation. If you bring in resources to a resource-poor area, then you are going to be part of that. You may have to put on the trappings of war in order to get that commodity -- guys with arms on trucks, with big NGO signs on them. There may be other ways to do things. You're sitting in A, you have C a community that needs some assistance, and you have B on the roadside that doesn't necessarily want you to get what you want to get to C, so can you figure out ways to do that, that thinks about Do No Harm. Some of it may be to involve B in moving that commodity, providing that commodity, using their trucks, doing anything you can that begins to connect people.

AN: What about the problem of assistance sometimes being diverted to be used in conflicts? AN: Something that one organisation did in Afghanistan, they wanted to get radios out to people so you could get messages. They got pink radios -- the pretty colour meant that warriors weren't interested. Another has a "Barbie doll" radio. In another situation, NGO staff found that they were training mechanics, but that the mechanics were being drafted into the army. They got round this by training women for the job, and found that they could stay with the NGO. There's another story about blankets coming in, when they cut the blankets in half to make them less attractive. One of the guys I worked with said in Somalia in some places maybe you have too much food, so it no longer has any value. Bring in extra, so the value decreases and it's not something necessarily that people have to shoot for. So think about ways of finding things -- particularly in a conflict situation -- that are not valuable. If you bring in rice, that's high value. You could bring bulgar (wheat) or something else which has less value.

AN: What was the other part of the theory about implicit messages that organisations carry? MT: In every single refugee camp you always see lots of trucks, big new cars, aid workers paying a lot of money for housing, international staff get evacuated, the radios. That's sending messages to people that may not be the best ones to send, that says some people are better than others. Maybe we shouldn't look so fancy. Maybe we should have old cars sometimes.

AN: Do you ever get people saying: "Well, that sounds good, but it's just not our day-to-day reality in the field"? MT: We just had an exercise and one of the groups used that as their thinking in preparing, and the other group said 'well, we're just going to get stuck in, we're going to just react'. So you have to convince. I think people have to see the value of it, and if they can apply it to their own life then I think they can.

AN: What emergencies have you been involved with yourself on the ground? MT: Central America -- El Salvador, Nicaragua, Honduras -- Somalia, natural disasters in a variety of places. In my role in the headquarters in the United States I've been involved in Rwanda, Albania, Kosovo.

AN: How long have you been with CARE? MT: Twenty-three years with CARE, and I worked before that for six years with the Peace Corps and another group. It's what I did with my life!

AN: So NGO practice has changed a lot during that time. MT: We've learned an awful lot over the years. In the emergency field Somalia and Rwanda caused a lot of thinking and analysis by all of us of what we're doing and how we're doing it.

AN: Do you think the Do No Harm principles are likely to lead to fundamental changes in NGOs in the coming years? MT: I wouldn't sell one analytical tool versus another, although I feel very comfortable with this one. I think you need to say what we can do can make it better or make it worse, in both the short and long term, and whatever tool you want to use, then that's fine. You can't make the conflict go away, and we didn't necessarily bring it, but we are an actor.