Using location data to help migrants and refugees fleeing persecution, economic destitution or natural disasters needs to be underpinned by an understanding of the power dynamics at play, according to experts.
While location and human migration data can sometimes be used to make positive humanitarian interventions, there is a huge risk it can also be used to track and control the movements of oppressed, marginalised, or otherwise vulnerable populations, according to speakers at the Safe Spaces? Location Data and Human Migration event on 3 July.
The panel was the latest instalment in a series of virtual discussions organised by the Benchmark Initiative, which was established to promote the ethical use of location data.
“One of the challenges with this is balancing creating actionable information versus the ethics of it. If we’re trying to inform coordination groups and service providers to distribute water to certain displacement sites, then you need to know where those displacement sites are,” said Robert Trigwell, a unit lead at the United Nations (UN) Global Displacement Tracking Matrix team, adding that the locations of these sites can be very sensitive.
“People are displaced through ethnic conflict, people are on the move through migration corridors where there’s trafficking, or where there’s potential for xenophobia or potential for border closures – so we have to collect information, but we also need to make sure that information is protected and used appropriately.”
According to both Trigwell and Petra Molnar, acting director of the International Human Rights Programme at the University of Toronto, any action taken to help people on the move must be informed by an understanding of the situation’s power dynamic, not just the data itself.
This means looking at the ways in which different groups or stakeholders in a given situation interact, and how that relationship manifests when one side wields more power than the other.
“What we’re increasingly seeing in the prevalence of migration control technologies is how different technologies affect groups differently,” said Molnar, adding that the refugees and migrants she works with often belong to “groups that are already disenfranchised in the conversation”.
Molnar said a major concern is how location and migration data can be “coopted quite easily” by states and other entities, something that is difficult to tell is happening because “a lot of this decision-making happens behind closed doors”.
“When we’re talking about migration and migration-related risks, the ramifications can be quite far reaching, whether from a privacy perspective, having your data shared with repressive governments, or just even knowing what is happening and how you can meaningfully opt out of it,” she said.
As an example, Trigwell highlighted how some non-governmental organisations (NGOs) in South Sudan, where he has previously worked, required people to register their households to get on beneficiary lists for items such as food baskets or tents.
“There’s a power dynamic there – yeah you can opt out, but does that opt you out from the service [completely]? Then opt out is a myth. It doesn’t exist. If you want to cross an international border and you need to show your phone, but yet you don’t want to show your phone, does that mean you can’t cross a border? Would that mean you would then try to cross that border illegally, will you go into a smuggling network?” he said.
“When implementing such technologies, you have to look at what are the adverse impacts this could have, because this could further marginalise the populations that we’re trying to support.”
Political environments matter
In the context of the European migration crisis, Reece Jones, a professor of geography at the University of Hawaii and author of Violent borders: refugees and the right to move, argued in his book that the European Union (EU) uses migration data to support a “deterrence policy”.
“The European Union is investing in information-sharing systems that integrate the satellites, sensors and personnel of the member states. These include Eurosur, an information-sharing platform for border enforcement in the Mediterranean, and Sistema de Vigilancia Exterior [SIVE], a Spanish programme to coordinate surveillance data,” he wrote.
“These operations and data-gathering practices suggest that the European Union monitors the sea very carefully for vessels and is aware of most migrant boats travelling from the coast of Africa. However, because officials do not want to encourage additional migration by rescuing people outside of the territorial waters of EU states, they often do not intervene until the boats reach shore or are very clearly in distress.”
Jones further noted that globally, over half of deaths at borders in the past decade have occurred at the edges of the EU, making it “by far the most dangerous border crossing in the world”.
Speaking on the panel Trigwell, in response to questions about the migrant crisis submitted by Computer Weekly, said while data can be used to inform decisions, those decisions are ultimately being made in a political environment.
“There’s the data environment, but that sits within the political environment as well… You could have great data and bad decisions, you could have great decisions on bad data,” he said. “As a data stakeholder, my job is to ensure there’s the best data possible to inform decisions.”
For Molnar, there has been a push towards “techno-solutionism”, whereby “vast stores of data” and “autonomous decision-making systems” are viewed as the best responses to deeply social or political problems.
“Who gets to decide what kind of world we want to build – is it big tech and the hubris of thinking that we have all the answers? We’re dealing with incredibly complex issues when it comes to migration, and these are systemic issues that we’ve been grappling with for decades, if not centuries,” she said.
“Quick fixes don’t really highlight to us all the intersecting reasons why people might be compelled or forced to migrate in the first place, so the concern is that a lot of the discussion is being focused on these really specific [techno-solutions] without understanding the broader ecosystem of power and how it operates in here.”
According to Trigwell, the best way humanitarian organisations can stop location or migration data being abused by states and other entities is to only collect what is absolutely relevant and necessary, and “be critical about what information we need”.
“The best way to protect data is basically not collect if you don’t need it – you can’t get hacked and you can’t lose data if you don’t collect it. The best way to respect and try to protect said populations is by really understanding what information the humanitarian community needs to get services to them and collect exclusively that information,” he said.
“There’s been a data explosion, again coming back to what Petra said about techno-solutionism – ‘As long as we have all the data, we’ll be able to figure what to do with it’ – no, that’s not the approach. The approach is to understand what the problem is, see if data can solve that problem and, if so, collect only the information that is needed to support the services.”