10/04/2010 03:24 pm ET | Updated May 25, 2011

Measuring Human Rights: An In-depth Discussion with Andrea Rossi of the Carr Center for Human Rights Policy

I sat down with Andrea Rossi, Director of the Measurement and Human Rights Program at the Carr Center for Human Rights Policy at Harvard's Kennedy School of Government, to discuss the next frontier of human rights policy as one grounded in an evidence-based approach, how human rights advocates and activists should engage policymakers, bridging the divide between human rights and mathematics, and more. The full transcript can be found here.

Describe a little bit about the founding and scope of the Measurement and Human Rights Program at the Carr Center for Human Rights Policy

When I arrived it was an initiative and it was linked with a project to measure police violence in a few countries. It was simply a project. And during that project there was a general discussion about how to measure police violence, and in the broader sense, how can we measure human rights or human rights violations. When I came to the Carr Center as a fellow some years ago, I had a chance to follow up on this project. Mainly, my interest was to try to explore the possibilities of increasing the capacity of human rights organization to move into more evidence-based policy. So the idea is, how can we make policies that are evidenced-based that are solid, even in the area of human rights?

The idea, again, was to first try to understand why human rights activists or people working in human rights don't use evidence. It's pretty amazing if you look at an evaluation of programs, most of them are based on assumptions. There is this problem in human rights that most of the activities or actions are based on moral principles. So rights are perceived as moral principle. And as a principle, once you agree on the principle, it is assumed that you can make the policy straightforward. So from the norms you can move into the policy.

And this is probably valid for a law. An example I do offer is you can make a law against cannibalism without knowing if there is a case of cannibalism or not because it is not good to eat each other. But then when you have to make that policy, for example, how many police officers have to train against cannibalism? How many specific cells or wings of a prison do we have to make for people who eat each other? Well, you need to know how many cannibals you have in your country. And if it's none, well, probably it's better not to open a specific unit. This sounds like a very mundane example, but if you think what happened, for example, in the case of trafficking. Trafficking is unacceptable. Moral principles that lead to norms say that it is unacceptable. But then if you don't know how many, if you don't know why, if you don't know how, you end up producing policies that are completely wrong; particularly when you just focus on extreme cases.

So the idea is to try to find a way to merge the evidence-based approach to policy with a human rights-based approach. Usually people think that is not necessary, and when you convince them that it is a necessary, they think, "Oh, but it is too difficult." I think that people think that it is too difficult because they don't know how to do it. When I teach, I try to tell the students that if biologists know how many fishes there are in the sea, why the hell can't we figure out how many victims of trafficking there are or how many abuses there can be in a place? Counting fishes in the ocean is much more difficult. Fish don't speak, and when you do this kind of research with people, too, the statistics are absolutely the same: you have a hidden population that you don't have access to, that is marginal, and that is illusive.

Why haven't human rights organizations been able to make that shift from assumption to evidence?

Well, there are many reasons. The first one is a theoretical point in that they think a norm is enough to make a policy. The second level of explanation is, I would say, cultural and technical. Most of the people who have come to work in the human rights field have a legal background. And with a legal background sometimes they simply don't know, or do not have experience, or are not trained in research matters in how to build evidence. So they don't think it's useful; second, they don't know how to do it. I'm generalizing here, but people who do have a legal background know what constitutes evidence, particular ones who work in criminal law because they have to bring evidence in front of a court. So they have this concept when there is a case, but not when they are lawmakers, where they are not acting essentially as lawyers.

The reason is because nobody has asked them. Human rights is a very strange area because in many cases where projects and programs are implemented, they are supply driven rather than demand driven. Supply driven means, we do human rights intervention because we do care about the topic, because we do have money. And when you do an activity or campaign, nobody has the interest to prove that their campaign was a failure. Not the person who has given you the money, nor you, the person on the ground. And this is a problem. If you look at the evaluation of these kinds of projects, the level of evaluation and the numbers of evaluation that you find on human rights interventions is very, very, very small. Nobody wants to do that. When you imagine human rights campaign or a policy campaign, we can't find an evaluation of how well they performed, a serious evaluation that is looking at the true impact. This is because people move from the norm to the policy, and they believe that speaking about rights is good enough. But then you think, wait a second. How much money did you spend for that campaign and what was the impact? Do really people think about rights in a different way? Nobody is interested in answering that question.

If you were interested in evaluating an advocacy campaign, for example, what are some ways to actually take those measurements?

Well, it depends on your objective. If your objective is to change the attitude of the people, first you have to know the attitude of the people. If research is well done, you first have to ask, do people have a baseline? How many people are launching a campaign and they don't have a baseline on what is the current reality? Actually, 99 percent. With very little research and effort put into establishing a baseline, this, for me, is a good indicator that there's no true interest in measuring the impact. Again, because we go back to the idea of the norms, the principle is enough. And I want to be clear.The principle and the norm are essential. It is necessary, but it's not sufficient to make good policy.

This is an interesting moment because due to the financial crisis, the amount of money that has been dedicated to human rights has decreased a lot. And now NGO's who are working in human rights are facing two kinds of problems. The first one is they have less resources so they have to decide inside their own organization what to do. And the other problem is that they have to convince the donors that they are actually doing something and they have to prove this kind of effect. But if you think for a second the fact there's a common interest to keep human rights in a way, foggy, so you can do whatever you want, at any level. Human rights has always been used as a discourse to criticize countries, opponents, organizations, competitors even, when you don't have any other argument, because you know that the human rights argument is mainly a discourse argument, not an evidence argument.

If you think about international relations between countries, relations between companies producing shoes or producing something else, it is a way to criticize the other that you are violating human rights. You can do a discourse theory on human rights that would be pretty amazing.

It's a fascinating topic, how to bring evidence, for the evidence-based approach to human rights is not easy at any level. Think even at an academic level, if I say, when people ask me, what is your job? I'm the Director of a program called Human Rights and Measurement. If the person is a human rights person, they don't understand the word "measurement," so they think that that is actually too complicated. They would say, "A measurement statistic of human rights? No I'm not interested." If the person is an economist, after hearing the words "human rights," they think, "Oh, but this is not relevant. This is not serious enough." So there is a need to bring the two worlds together. But because it's useful and is a good idea doesn't mean that it's easy at all. Sometimes my job is simply being an interpreter. I'm trying to express concepts of solid policy based on knowledge, and that concept of impact assessment. And at the same time, try to come to use the concept of human rights to people who are more technical, for they just think, "Oh, but I just want to build a school, or a road, or a jail, or whatever," but there are human rights dimensions and consequences to all of these endeavors.

How does this idea of measurement and impact assessment translate to the field?

The interesting thing is that this concept works much better in the field. At the moment we are working in two major areas, Mexico and Sierra Leone. When we do this in the field, we experienced the incredible, amazing power of evidence and information perceived as a safe place. In the case of Mexico, when we promoted the idea of evidence-based policy or human rights policy analysis with a specific focus on quantitative methods, we were able to people from NGOs, the government, the military, and others, all together. They thought that because they were coming to a course or a program that was like a type of data collection. It was less political and more neutral.

The good thing about numbers is that you need them even to disagree. Think for a second. We can disagree on policies to reduce unemployment, but at the very least, you need to know how much unemployment you actually have. So you can have the union and the employer that can disagree on the policy, but usually, they first agree on the situation. In human rights, we can't even agree on the situation. People disagree on human rights approaches because there isn't a basis to agree on to begin with. And this is a problem because then the disagreement is not about a different way to look at a problem, which is still possible, but it is even more about the problem itself. And this is the reason why everybody is doing whatever they want in the sector.

Now, this is interesting. For many years, governments were reluctant to talk about human rights. It was always perceived as a tool to attack them. But now, more and more, governments have started using human rights to justify policies that otherwise could not be justified.