Groundviews

UN Investigators for Sri Lanka, Show Your Work!

Photo by Lakruwan Wanniarachchi, AFP/Getty Image via Al Jazeera

If the U.N. wants to promote reconciliation in Sri Lanka, it has to show its work.  U.N. investigators should use tools and methods that are transparent, replicable, and open to scrutiny by all.  Every Sri Lankan, both inside and outside the country, deserves the chance to question the methods behind U.N. findings.

Last week, Sri Lanka’s president, Mahinda Rajapaksa, announced that he would not grant entry visas to U.N. human rights investigators looking into allegations of mass killings.  Since investigators won’t be able to collect evidence on the ground, they will have to rely more on digital evidence, like photographs, videos, and satellite images, which can be transmitted out of Sri Lanka and Mr. Rajapaksa’s control.  As U.N. high commissioner for human rights, Navy Pillay has said, there is a “wealth of information” outside the country.

Unfortunately, digital forensic investigations lend themselves to spectacles of expensive, high-tech, proprietary tools.  These tools can mystify experiments for regular citizens who don’t know how they work, and far exceed the budget of university scientists, both in and out of Sri Lanka, who may wish to replicate and confirm U.N. findings.  In particular, secret, closed, or proprietary investigative tools and methods obstruct external review and critique.  Commercial interests may conflict with the disclosure of methods and algorithmic functions.  As a result, using the wrong tools can produce a degraded authority for investigators via power, prestige, or financial advantage, rather than the accountability of scientific peer review.  The U.N. should resist the temptation.

Open tools and methods, in contrast, invite publics to participate in the process of investigation and authentication alongside experts.  After nearly thirty years of civil and ethnic conflict in Sri Lanka, the target decision-makers for the U.N. investigation should not merely be courts and NGOs, but also include publics with divergent identity politics.  If the U.N. hopes to convince Sri Lankan publics to change their assumptions, and reach beyond identity networks and lived experiences to find common ground with former enemies, it should give them the resources to do so.  Investigative methods and tools for producing measurements from digital evidence, and each analytic step applying these measurements to reach conclusions, should be open for everyone to see.

To be sure, some information, like the identity of witnesses, must remain secret.  Transparency bears little weight compared with a risk to someone’s life or safety.  But with digital forensic tools, where the balance is between proprietary interest in trade secrets or marketing incentives, transparency should win.

Prior investigations into evidence of war crimes in Sri Lanka show the urgency of open tools and methods.  On August 25, 2009, Channel 4 News in the United Kingdom broadcast a leaked video that depicts men in Sri Lankan military uniforms performing extrajudicial executions.  The video provoked pained public outcries and became a focal point of frustration, mistrust, and controversy.  Some people believed that a Sri Lankan military soldier recorded the video on a cell phone while witnessing a war crime.  Others suggested that the video was a fictional scene produced with actors by a commercial film crew, using lights and fake blood, to discredit the Sri Lankan government and interfere with the post-war reconciliation process.

U.N. investigators found evidence that the video was likely authentic.  But that did not stop smart people from continuing to believe it is fake.  I was first in Sri Lanka in 2009, and then had the opportunity to return in 2012 with a Fulbright Advanced Scholars Senior Research Award, and teach documentary film history, theory and methods at the Eastern University of Sri Lanka, Trincomalee.  I spoke with intelligent, well-educated people who believe the video is fake.  And they have decent reasons.  The Liberation Tigers of Tamil Eelam, or L.T.T.E., had a sophisticated propaganda wing and were known to have faked other videos in the past.

Unfortunately, both U.N. and Government of Sri Lanka investigations into the Channel 4 video contained obstacles to transparency and reproducibility.  For example, imperfect verification of evidence preservation calls into question whether all of the investigators actually examined identical copies of the video.  Analyzing the same piece of evidence is an essential first step for reproducibility of the results, and a prerequisite to meaningful consensus.  Researchers can use a cryptographic hash to confirm that their copy of the evidence is unaltered.  Yet, none of the U.N. or Government of Sri Lankan forensic reports included a hash.

Indeed, there are strong reasons to doubt whether or not all of the investigators did analyze unaltered copies of the video evidence, since investigators reported examining videos with different lengths, names, formats, and sources.  Some expressed difficulty obtaining a copy of the evidence at all.  Inconsistencies like these degrade the credibility of the forensic investigations as a whole.  Weak forensic credibility leaves publics more likely to ignore or confuse results, and to turn to alternative sources of authority, such as their own personal experience or political affiliations.

Opening the procedures and technologies of digital forensic analysis to scrutiny will help to counter this result.  The simple “show your work” mantra of elementary school math teachers is actually a profound intervention.  If you look back into history, basic concepts of transparency and replicability contributed to the widespread adoption of the scientific method.  According to Harvard historian Steven Shapin, back in the 17th century, more and more precise descriptions of experimental processes and results, including circumstantial details documented in charts, data, and drawings, expanded the circle of people who felt that they had witnessed an experiment almost as intimately as if they had performed it themselves, and so trusted the results.  Shapin calls this “Virtual Witnessing,” and suggests it’s how we got modern science.

By prohibiting entry to U.N. human rights investigators, Mr. Rajapaksa has ensured that the U.N. investigation into war crimes in Sri Lanka will be a remote one, and dependent on digital evidence and forensic science.  If the U.N. wants to use this science to reach across identity affiliations, and ask Sri Lankans to trust its investigative findings instead of their own communities and thought leaders, it should give them the resources to make that leap.  U.N. investigators, show your work!

###

Rebecca Wexler is Fellow of the Information Society Project and J.D. candidate at Yale Law School, a Fulbright-Sri Lanka alumna, and a former legal intern at The Electronic Frontier Foundation.  Her forthcoming book chapter, co-authored with a forensic scientist, offers an in-depth examination of forensic video analysis and censorship in the leaked Channel 4 video dispute. [See Rebecca Wexler & Carey R. Murphey, “Video Forensics in Post-War Crisis,” in Access to Knowledge in the Global South, Ed. C. de Souza (Bloomsbury Academic, forthcoming 2014)]