You have to do it reportorially. We’ve done reports on self-censorship—we did one on self-censorship in Colombia a few years ago. The thing about self-censorship is that if you’re afraid to publish a story because it’s going to get you killed, you’re not then going to go to a journalist or someone from the CPJ and say, “Here’s a story that I’m not going to publish because it’s too dangerous, let me tell you about it.” If there’s self-censorship, then by definition they’re not going to talk about it because they’re afraid. It’s like covering any other kind of human rights issue. You have to develop very close relationships with your sources. They have to feel confident that you can protect them. You have to feel confident that they’re being honest. It’s a painstaking thing to document. But on one level it’s also obvious—you can just talk to people in an anecdotal way, over dinner or something, and they will say, “Oh my God, I know this but I’m not going to publish that because it’s going to get me killed.” That kind of stuff you become aware of. What’s much harder to do is to document the systematic nature of self-censorship.

Going back to the higher level meetings you have had with governments, one of the issues I imagine comes up is that you’re meeting with people who in some cases are involved in the corruption. How do you go into a negotiation with someone who may not be particularly sympathetic to your view or is looking to undermine it?

Nobody’s going to do this because they want to be nice. You’re there talking about interests. You assume that they’re only going to act when they believe it’s in their political interest to act. They’re political figures, and they’re balancing competing interests, some of which are very powerful. You’ve got to put enough weight on the scale and make arguments that convince them they should reform because it’s in their political interest.

You can make those arguments in all sorts of ways. You can make it at the broad theoretical level: The democratic development of your country is dependent on having a free and open press and in the long term this is in your national interest. Some heads of state are sophisticated enough and have a long-term vision and recognize that. Then there are short-term arguments: you’re standing on this list affects the way donor countries perceive you and is going to have a very direct impact on international aid.

For the most part, political leaders recognize that this kind of negative attention in the international and the domestic media is harmful. If they can take steps to correct it, they will. The problem is they’re often confronting powerful, entrenched interests that are behind these murders. Sometimes they’re criminal groups, sometimes intelligence agencies are suspected, and it’s difficult to investigate. So you have to put enough pressure on them to confront these challenges.

Can you be daunted by what sometimes looks like an impossible task? In Mexico, for example, you’ve met with the president and made some progress. But a lot of the corruption that allows people to get away with murdering journalists is at the level of law enforcement and the courts.

It’s daunting, but you have to have a long-term view of this. We’ve seen some countries begin to turn it around. It happens in the context of a more generalized improvement. In other words, as long as the levels of violence remain at the level at which they are in Mexico, it’s going to be very difficult for the government to confront this fully. But what we certainly want to see is that it gives this area—freedom of expression, freedom of the press, violence against journalists—sufficient attention. So that as the overall environment improves, this is an area in which we start to see progress.

Joel Meares is a former CJR assistant editor.