Mapping Silences, Reconfiguring Loss: Practices of Damage Assessment & Narratives of Repair in Post-Earthquake Nepal
The full paper, co-authored with Austin Lord, is available here.
It will be presented at the ACM Conference on Computer-Supported Cooperative Work and Social Computing on Monday, November 6th in the afternoon session on Crisis Informatics.
The Langtang Valley was devastated by the April 2015 Nepal earthquake. Located in a remote and difficult to access part of the country, the region was home to around 600 people prior to the earthquake and is an important site of Tibetan Buddhist Culture and trekking tourism. During the earthquake, the steep walls of the valley gave way to as many as 4 major co-seismic landslides, destroying entire villages and killing hundreds of residents and visiting tourists. Government engineers, arriving with mobile tablets, cameras, and GPS devices to record the damage to individual houses, were at a loss to capture the scale of destruction using the forms, data standards, and methods that the National Housing Reconstruction Program prescribed. In a phone interview, one engineer who visited Langtang to conduct the survey told us:
I had not seen such collapse. I was there with just two days experience. I thought there would be some damage. That’s how I felt. But when I reached [Langtang]… nothing was there. Everything was a flat plain.
How do we know what we know about disasters? How are emerging technologies like GPS, smartphones, and open source software changing the ways responders collect, manage, and use data about earthquakes, floods, or climate change? As governments and aid agencies become increasingly reliant on these tools to manage disaster relief efforts, such questions are critical. Their answers shape who receives aid, what counts as “damage”, and how large scale disaster recovery processes are planned and executed.
We conducted an extended qualitative research project, using methods including participatory mapping workshops and oral history to study the government damage assessment and how the tools it used shaped recovery process in Langtang. In total, 42 members of the community, about one fifth of the surviving residents of the valley, participated in these activities. The mapping workshops focused on historical settlement patterns, the location of cultural and religious sites, the oral record of past landslides and avalanches, perceptions of future risk, the impacts of tourism on development, challenges faced during post-earthquake recovery, and participants’ hopes for the future.
Through these methods, we were able to identify misalignments between the official characterization of the earthquake, as expressed by the data collected during the government damage assessment and the lived experience of the communities in Langtang. We show how the tools, data standards, and data collection practices used in the damage assessment contributed to these misalignments, which included ongoing landslide danger, collective ownership and management of resources, psycho-social trauma, and changes in the economy and social life of the Valley prior to the quake.
The narrative of the earthquake produced by the assessment in turn shaped the kind of reconstruction that was possible in the aftermath. In particular, it favored the bureaucratic needs of the government of Nepal and international donors at the expense of the agency of the communities of Langtang to envision and manage their own recovery. Design principles related to ambiguity, complexity, and doubt, which anthropologist Peter Redfield argues are critical to humanitarian technologies, were in fact exactly what the implementers of the government damage assessment sought to avoid. As a result, opportunities to address long-term challenges in the Valley, some of which contributed to the residents’ vulnerability to the earthquake in the first place, were missed.
We argue that HCI can contribute to a re-imagining of information systems that would enable communities to take the lead in post-disaster recovery and avoid some of the problems our research surfaced in Langtang. We demonstrate the ways that sorts of sense-making afforded by ICTs, in this case the damage assessment, play central roles in enacting repair-work following crisis and breakdown. As the role of these tools in disaster management continues to grow, improving our collective ability to recognize and engage with the consequences of how they represent the world is essential.
