Troubling Numbers

This is a weeks-overdue post to recommend that you check out Mike Spagat’s piece on the myth that “economic sanctions aimed at Saddam Hussein and his regime killed hundreds of thousands of Iraqi children in the 1990s and early 2000s.”

Spagat explains how, despite the retraction of the results by the original researcher, mistaken findings on Iraqi child mortality were a prominently cited justification for the war in Iraq. He also makes the worrying point that the seemingly-permanent place in the discourse occupied by these erroneous estimates is part of a larger phenomenon wherein shocking statistics become dogma, regardless of their accuracy.

I’m filing this one under “reasons to keep fretting about the relationship between evidence, advocacy, and policy-making.”

p.s. For another riff on the same theme, check out the NYTimes’s “Revisiting the ‘Crack Babies’ Epidemic That Was Not.”

Really, Greg Mortenson? Really?

Since it’s the middle of a Thursday afternoon and I’m technically very busy with other work, this seems like as good a time as any to jot down my thoughts on Three-Cups-Of-Tea-Gate. In no particular order:

  • Good lord, remind me never to make Jon Krakauer mad. (Unless I have nothing better to do than respond to meticulously-researched 50-page takedowns of my life and dealings.)
  • While I am not willing to assume that either 60 Minutes or Krakauer got everything right, the available information tells me that Greg Mortenson and CAI have not been meeting their obligations as stewards of other people’s charitable donations. “I don’t know” and “I won’t tell you” are not, to my mind, reasonable answers to the question “what did you do with the money I gave you to build schools with?”
  • It is clear that the success of Greg Mortenson’s book has brought tremendous benefits to CAI, by increasing donations to it and raising its profile. However, that does not mean that it’s okay for the charity to foot the bill for promoting the book, when all the revenue from it goes to Mortenson and his publisher. This isn’t a “grey area,” it’s “tax fraud.” You can’t use the funds from a tax-exempt nonprofit to pay for your personal for-profit activity.
  • Also, private jets? Seriously?
  • When I first heard about 3 Cups of Tea and CAI, I wondered if they were actually running schools, or just building them. The emphasis on the latter seemed weird. Buildings are nice, but surely “lack of freestanding dedicated structures” wasn’t the main barrier to education in poor, rural areas that lacked infrastructure and transportation links? I actually read the book, ages ago, in the hope of finding out how CAI was handling teacher recruitment, salary, and curriculum issues. It did not answer my questions, but at the time I didn’t see that as a sign of foul play. I figured that either (a) such bureaucratic details had been sacrificed in service of narrative, or (b) they were just building buildings, which is kind of lame.
  • I guess it turns out that the answer was (c), “all of the above.”
  • And while we’re on the subject of things-that-get-sacrificed-in-the-service-of-narrative, if the books really contain the “factual inaccuracies” alleged by Krakauer and 60 Minutes, then to me, that is even more upsetting than all of the private jet malarkey. What, exactly, was the thought process there? “Oh, these brown people will never read this book, so it doesn’t matter if we call them terrorists and accuse them of kidnapping”? “Why would anyone check this story with actual Pakistanis or Afghans, when we have white people they can talk to right here?” Not cool, man. You can’t just do that.
  • Well, it turns out you can, but I don’t like it.
In short, for the time being, Greg Mortenson and CAI are on my bad list. As per usual bad-rules, they will remain there until they do something to convince me that they should be taken off of it.
Further reading: Krakauer’s medium opus, The American Institute of Philanthropy’s evaluation of CAI, Nick Kristof urges restraint, Chris Blattman agrees, Megan McArdle reminds us that when we demand “messianic development projects and neat stories with happy endings,” what we get is development done by people with messiah complexes, and neat stories that aren’t actually true.

What’s New in Sexual Violence in the DRC (An Interview with Jocelyn Kelly)

You’ve probably noticed that getting annoyed about mainstream media reporting on the Democratic Republic of the Congo is one of my favorite hobbies. So imagine how excited I was when, following the release of the Harvard Humanitarian Initiative‘s report “Now, the World is Without Me: An Investigation of Sexual Violence in the Eastern Democratic Republic of the Congo,” I discovered that almost all of the press coverage of the study reported the findings as “DRC: Now with more rape.”

“That’s funny,” I said to myself, “because I read that report, and that’s not what I thought it said.” Then I realized I recognized one of the names on the author list, Jocelyn Kelly. Jocelyn is the Gender-Based Violence Research Coordinator for the Harvard Humanitarian Initiative (HHI) and has worked extensively on sexual violence in the DRC. When I got in touch with her to ask what gives, she very graciously agreed to an interview to discuss the report.

Check it out:

Kate: So, I read “Now, the World is Without Me.” And then I read the press coverage, and was surprised to see that it all appeared to be along the lines of “z0mg, rape in DRC on the rise!” Am I correct that the actual findings were that overall, reported cases of rape had declined between 2004 and 2008?

Jocelyn: Yes. The news coverage didn’t seem to capture the nuances of the issue. Which can be difficult to do, since DRC is a complex place

Kate: But you found that civilian-perpetrated rape had increased?

Jocelyn: Correct, but it’s important to remember that we are talking about reported cases of civilian rape, which is different from civilian rape as a phenomenon. So what we’re seeing is that more women are coming to Panzi Hospital and reporting that the person that raped them was not a man in uniform, but a civilian. But it’s important to remember there can be multiple causes for this. It may be that due to the expansion of sexual violence services in DRC over the last few years, that women are more able to seek services for rape in general.

Kate: I was also interested to see just how high a proportion of the women presenting at the hospital had been in touch with NGOs. But I was curious what that actually meant.

Jocelyn: This starts to make sense when you think about how large eastern DRC is and how difficult it is for most women to travel all the way to Panzi, which includes paying transportation costs on their own. Instead, many women first access an NGO that is closer to their home, the NGO would provide a referral to Panzi for medical services and then sometimes they would also provide transportation to the hospital.

Kate: So does this statistic tell us anything about how well NGOs are serving the population, or just that it’s difficult for women to get to the hospital without assistance of an NGO?

Jocelyn: It’s hard to take this data and make assumptions about how well NGOs are serving the population. I think services have expanded enormously, and this is beneficial to people who need medical care. However, it’s always hard to know about the populations that are not being reached. It’s really hard to take data from a large reference hospital like Panzi and make assumptions about who is and isn’t coming in for treatment. You could argue that women who present to Panzi represent the most traumatized women who have experienced the most violent rape - hence their need for advanced medical intervention. However, you could also say, that women who don’t get to Panzi are the most vulnerable since they may be killed during the attack, may be too far away or too isolated to get to services. So, you have a difficult problem trying to understand the “true” reality of the situation on the ground. In medicine and public health, we always make these caveats about data like this, which comes from “clinic-based samples.”

Kate: What piece of the puzzle that you would have particularly liked to have was missing from this data?

Jocelyn: I think the difficulty with hospital data is that it’s hard to know where women are coming from. So, it gets back to the question of whether there are communities that are not being reached or served by medical care. The second question I have is how women fare when they leave the hospital. It would be wonderful to have a system where we could continue to follow how women are doing once they leave Panzi. I would also love to see a better information network be created that could link women’s medical history at Panzi with her local clinic. This way we could do a better job of providing continuous care and better following up with women affected by violence.

Kate: One of the other patterns that I thought was interesting was the long lag time between sexual assault and presentation at the hospital. What’s going on there?

Jocelyn: I think this is a heart-breaking finding. What I have seen, at least anecdotally at Panzi is that women will wait until they have suffered injuries or infection from rape that gets worse and worse over months or years. Only when the problem interferes with their ability to work and take care of their children do they finally make the choice to seek services. This is especially sad when you think about the fact that if women present to a clinic 72 hours after a rape they can get life saving interventions like HIV and STI prevention.

Kate: Do you have a sense of how much stigma acts as a deterrent to women seeking treatment?

Jocelyn: It is a huge deterrent. Women often say that they don’t disclose the fact that they have been raped because it will have very destructive consequences for them in their communities. If you are a young girl, you often won’t be able to get married after rape. If you are a married women, you risk having your husband leave you.

Kate: I think that may be one of the hardest things about this issue for Western audiences to understand.

Jocelyn: Yes. It’s extremely difficult for us to understand why a rape survivor might be rejected from her own home after rape. It’s like being punished for being victimized. There still seems to be an element of blame that exists towards women who have been attacked by 20 armed men. A lot of times, we hear from men in the community, “Of course it’s not a woman’s fault for being raped – but, really, she shouldn’t have been in the field alone.” So you see a disclaimer, and then this implicit blame comes through. In fact, stigma is such a big problem for women, they often say it is more traumatic for them than the rape itself. Because the effects of stigma have extremely long-lasting consequences and often result in a great amount of social isolation.

Kate: One of the recommendations in the report is that more localized survivor service provision will help women manage the risk of stigma. I believe the idea is that if services are provided locally and women don’t have to absent themselves from their homes/communities to seek treatment, they can more easily keep people from finding out?

Jocelyn: Yes. And this is also an argument for providing GBV and primary healthcare services in the same place. If women are worried about what people think, they can go to a clinic and make the argument it’s for a regular check-up.

Kate: Can we go back to the finding of a rise in civilian-perpetrated rape for a minute? Because this seems to me to be one of the most interesting findings of the report.

Jocelyn: It’s definitely interesting. And I do think this is a valid trend.

Kate: I am curious whether this represents opportunistic crime (arising out of a chaotic situation or whatever) or if it represents a normalization of sexual violence. Because I suspect these would have very different implications for what we expect to see going forward if the conflict abates.

Jocelyn: I think it’s both – because there are never simple answers. One of the things that HHI believes in is taking a “mixed-methods” research approach to these very complex problems that arise in humanitarian emergencies. So what that means is that we look at quantitative data, like the data from the Panzi hospital charts. But we also get qualitative data – from focus groups and interviews- from the community. And what we’re hearing from the community is that civilian rape is on the rise, and that there are a number of reasons for this. One is that community justice structures that used to very effectively punish (and therefore suppress to some extent) have broken down because of displacement and insecurity. Another is that rape is seen as much more common and perhaps somewhat more acceptable because it is not punished. Another phenomenon that is being described to us by communities is that often demobilized soldiers and war-traumatized youth will be extremely violent within communities and might contribute to crimes like rape.

Kate: What does this tell us about the relationship between initiatives to combat the phenomenon of sexual violence and efforts to end the ongoing conflict in the region?

Jocelyn: I think you absolutely have to end the conflict once and for all to be able to work effectively on the other problems that communities face. As things are now, people can work tirelessly to set up a clinic, or a counseling program, but if soldiers come and kill or displace people in that town, all of the effort and programming is useless. First, security has to be restored to the east (yes, much easier said than done), only then can we really concentrate on the problems that communities face as they try to rebuild.

Kate: Is there anything else that was missed in the media coverage that I didn’t touch on here that you think is important to communicate about the findings?

Jocelyn: Well, I think the media coverage was valuable in the sense that it continued to bring these important issues into the public eye. Even though the war in DRC has been going on for more than a decade, it’s easy to forget that this level of violence exists in the world. However, I think it’s a huge challenge for us as researchers to communicate the limitation and nuances of the research. I think some of these definitely got lost or glossed over in the coverage.

Kate: How big of a deal is it that they basically misreported the main finding?

Jocelyn: You mean reporting that rape in general is one the rise?

Kate: Yeah. I just thought that was really interesting; given that “increase in civilian rape” could’ve been the story.

Jocelyn: Me too! I was in DRC at the time and kind of confused at what the press took away from the report. Interestingly enough, Human Rights Watch has reported that rape did rise in 2009 and rates have more or less stayed at that level in 2010. So rape is still a huge problem, and is not decreasing, but that wasn’t what our report was about.

Kate: I imagine it’s an incredibly difficult proposition to get accurate reporting on this because the combination of lack of understanding and the risk of compassion fatigue really incentivizes sensationalist coverage. And “rape still an issue in DRC, interesting changes in trends” just doesn’t have quite the same ring to it.

Jocelyn: Note to self: make extremely nuanced and factual titles in the future. The press will love it: “Clinic-based sample of data cautiously speaks to some trends of sexual violence in women presenting to Panzi 2005-2008.”

Kate: Stick a “permits limited inferences” in there, and you’ll be on CNN in no time.

Jocelyn: I’m saving that for the coup de grace: “Well, Larry, what you have to remember is this data really permits limited inferences.”

Kate: Awesome.

Suck It, Jeffrey Rosen

Earlier this month we were totally unimpressed by Jeffrey Rosen’s hatchet job of Second Circuit Judge Sonia Sotomayor. Brief summary: “Anonymous sources tell me that prominent female minority member may not deserve her success. Oh, and she’s probably bitchy and uppity, too.”

When pretty much all of the blogosphere excoriated Rosen for his sloppy / biased reporting (our favorite response here), Rosen defended himself by saying, essentially, “What’s the problem? My unidentified sources really do think she’s dumb. Cross my heart.”

This morning President Obama nominated Judge Sotomayor to fill departing Justice Souter’s spot on the U.S. Supreme Court. Here’s the video of the announcement. And Jeffrey Rosen can suck it.

And the Award for "Most Creative Rationale for Kicking a Prime Minister out of Office" Goes to…

The Thai Constitutional Court handed down a verdict today that Prime Minister Samak Sundaravej must step down because he violated the constitution by hosting a televised cooking program while in office.

Samak, whose hold on political power is already quite shaky (see here re: two-week occupation of prime minister’s office by anti-government protesters), began hosting his popular show, translated alternately as “Tasting While Grumbling” or “Tasting and Complaining” (for reals) in 2000. He was taken off the air in 2007 when the army-appointed government commandeered the network. Upon taking office in January of this year, Samak informed reporters that he would be bringing back the weekly program. In response to questions about possible constitutional conflicts, he replied: “No. I checked the constitution already, there’s no obstruction with that.”

Apparently he did not check it carefully enough, because today his opponents prevailed on a constitutional claim that Samak had unlawfully accepted private employment while in public office. D’oh. The court rejected Samak’s defense that he appeared on the show as a freelancer rather than an employee and ordered him to resign.

Samak’s (questionably) ruling People Power Party replied with a resounding “sucks to you, constitutional court, we’ll just reappoint him!” No word yet on how they plan to respond if the appeals court scheduled to review Samak’s conviction for defamation later this month upholds the three year jail sentence imposed by the lower court.

Interesting Things People Tell You At Parties When They Find Out You Write a Human Rights Humor Blog

1. In Brazil, they segregate their prisons according to gang membership. No exceptions. Not even for individuals who in fact are not members of any gang.

How does that work? Easy. Upon being admitted to the prison system, unaffiliated prisoners are required to join a gang.

And Brazil wonders why it has such a problem with violence…

2. In Angola, police shot a group of actors dead on their movie set, apparently after mistaking the robbery scene they were shooting for an actual robbery.

“We saw the [police] pick-up speeding towards us. It looked empty but then suddenly it stopped and people appeared on the back and without asking any questions they started shooting at everybody at close range.”

Um. Speeding in in a pickup and shooting people at close range, no questions asked? Not exactly acceptable police work. Even if the robbery had turned out to be real.

And then there was this little gem, buried at the end of the article: “[The director] said that he had informed the local police about the filming before they began but the marksmen appeared to belong to an elite unit.”

“Elite unit,” eh? That is not under the control of the local police, and kills indiscriminately? I believe there’s another term we use for that, non?

Hint: rhymes with “Shdeath Shquad.”

(Thanks for the tips, New Friend Bill!)

Actually, Those Rights Are in My OTHER Other Pants

A few months ago, I had an asylum client with one of the best (on a scale of “1” to “well-documented textbook case of persecution”) claims I’d ever seen, heard, or read about.

The client, a national of a reasonably stable developing nation, was the wife of a well-known government official who was suspected of dissident sympathies. About a year before the client made her way to the U.S., her husband was accused of plotting a coup and arrested. He hasn’t been heard from since, and no evidence of the alleged coup was ever released. As part of an ostensible investigation into the alleged coup, our client was detained, tortured, and raped. Twice.

Any reasonable asylum officer would have immediately granted her application for asylum. Instead, we got the one who said this:

“Applicant claims that she was discriminated against because her husband was arrested for taking part in a coup d’etat, a treasonable offense. However, persecution must be distinguished from prosecution, and a government has a right to prosecute people and conduct investigations for legitimate common law offenses. In this situation, the applicant’s spouse was accused of taking part in an attempted coup d’etat which, inevitably included an attempt on the life of the President of Ghana. As such, when her husband escaped custody on April 4, 2006, the [enforcement agencies] would been derelict in their duties had they not questioned the applicant and any people associated with him. Attempted murder, of a President, or any human being, is a crime for which a law enforcement authority is required to investigate.”

A couple of things:
1. Our client isn’t from Ghana.
2. Her husband didn’t escape custody. Everyone’s pretty sure he was murdered.
3. There was never any evidence of an attempted coup plot.
4. Torture and gang rape are generally understood not to be acceptable investigation techniques.

Our client was devastated, and we were super pissed off. We weren’t really interested in waiting the three months to take the case before an immigration judge, so we called the asylum office and asked to speak with the officer’s supervisor. We very politely suggested that perhaps some extremely embarrassing factual and legal errors had been made in the asylum decision, and maybe someone might want to take a second look at it. Awesomely, we got a call back from the supervisor the next day saying that he would overturn the decision. Of course, it then took three months to get the final grant notice, but everybody’s happy now.

Amanda posted a while back on human rights violations stemming from “sheer bureaucratic incompetence, rather than any actual evil intent.” My only question is: What would have happened to this woman if, like the majority of asylum seekers, she’d appeared unrepresented by counsel?

Pakistan! In my office, RIGHT NOW.

The Tarrant County Star-Telegram ran a piece today describing an attack by Baitullah Mehsud’s forces on the Sararogha Fort in the South Waziristan province. According to the article, a force of 200 militants attacked the fort but were repelled by artillery fire. They withdrew, but later that same day 300 militants returned, surrounded the fort, “blew a hole in the wall,” and stormed it. Approximately 47 Pakistani soldiers were killed in the battle.

This might not sound that surprising. It isn’t the first time we’ve heard about Mehsud -he is the person who Musharraf’s government accuses of assassinating Benazir Bhutto. South Waziristan has been having problems for a while too.

But it is surprising! This is textbook old-school warfare. (cf. Percival’s Compendium of Olde Skole Tactickes for the Making of War, Ch. 7.) Though Former Provincial Administrator Khalid Aziz asks if they are “trying to fight like a real army,” it seems pretty clear that they are not so much “trying” as “succeeding so hard that they be kicking all y’alls asses.” (Also: am I the only person who visualized the battle for Minas Tirith when reading about it? They knocked a hole in the wall, then overran the fort, folks!!)

So, why are people who are supposed to be terrorists engaging in conventional warfare? Didn’t they get the Official Evil Terrorist Plan Memo? (“Step 1: wreak havoc by doing evil bloody things to civilians. Step 2: flourish in the ensuing chaos.”)

More importantly, why are they succeeding? I know that our Model of A Very Modern Military and its pals have had some problems with all this nouveau counterinsurgency warfare. But they still have planes and bombs and big guns, non? Is there some reason why they can’t defend a fort? Or, for that matter, take it back once it has been lost?

I can think of some reasons, but none of them make me happy. And I think that “is Amanda happy?” should DEFINITELY be the question that guides our military and its apparently-useless allies.

I want a full written explanation on my desk by 9 AM Monday morning. And no, I do not care that Monday is a holiday. You should have thought of that before you decided to suck.