I saw this coming, and I wish I had been wrong.
Back in 2017, I wrote here of the risks of the UN’s refugee agency, UNHCR, collecting biometric registration data from Rohingya refugees, noting that the data could be used to drive unwilling repatriation; that collecting such data may make refugees believe their access to aid depends upon providing such data; and that – once collected or shared – such biometric data is virtually impossible to get rid of.
Nearly four years later, a report from Human Rights Watch (HRW) says these worst-case scenarios have come true:: A detailed database of the Rohingya refugee population has been handed over to Myanmar’s government, which drove them across the border into Bangladesh almost four years ago. The same millitary that conducted the (most recent) genocide against the Rohingya now holds the biometric data of the population it has tried to eradicate.
A UN investigation team described Myanmar’s treatment of its minority Rohingya population as war crimes and crimes against humanity; characterised by brutal physical force, civilian casualties, villages razed to the ground, as well as internet shutdowns and information blackouts.
Refugees in the camps in Bangladesh told HRW they accepted the need to be registered with UNHCR to be recognised as refugees and get services. But they said they didn’t get a chance to opt out of a government-backed digital identity card – a process UNHCR also handled.
HRW reports that the data connected to those “Smart Cards”, including biometric scans, was shared with Myanmar, often against the refugees’ wishes. UNHCR denies HRW’s claims it misled refugees or broke its own data rulebook.
There are so many failures in what has happened: institutional failures to abide by organisational policies and guidelines; decision-making failures in permitting the data-sharing to happen; moral failures in UNHCR’s irresponsible actions and continued denial of its role in what happened; and sector-wide failures in the utter lack of accountability mechanisms – to name just a few.
These failures have real impacts, and biometric data is immutable – it will stay linked to their bodies until they die. Now the data has been shared, it’s impossible to take it back.
According to HRW, the data included not only the biometric data of the 830,000 individuals, but also details of their family compositions, their place of origin, and information on their relatives overseas.
Once they knew their biometric data had been shared, refugees interviewed by HRW suggested they could be targeted for forced repatriation or for retribution if they return to Myanmar. Some went into hiding when Bangladesh attempted to begin returns in 2019, using names from the database.
Targeted identification of persecuted populations to facilitate targeted killings and violence has long been a tactic of genocidal regimes, only this time the data is digitised – fast to access, quick to scale, and easily accessible. Meanwhile, the Myanmar military has been purchasing spyware that can “extract data from smartphones, access phone conversations, and monitor people’s movements”, thus demonstrating an appetite to use technical tools for repression.
Betrayal, denial, and double standards
In its rebuttal, UNHCR says refugees consented. Real consent in situations of such power asymmetry is practically impossible. But there is a deeper, more insidious harm here.
That data was taken from the bodies of human beings, and shared with those who explicitly want to cause them harm. It is a betrayal of their right to self-determination, of dignity, of their very personhood. At a time when personal data rights are a core part of data protection legislation in areas of the world where UNHCR and other humanitarian agencies have their headquarters, the act of treating personal data so callously and with such little regard for the wishes and well-being of the people it came from is inexcusable.
There is no way that the personal data of nearly a million European people would be treated like this without a massive outcry, without resignations and policy overhauls, without fines, firings, and legal ramifications.
It should go without saying, but here goes: A person’s nationality, citizenship or lack thereof, country of origin, or ethnicity should not affect the way in which they are treated, and this extends to the way in which their data is treated. And instead of courts and data commissioners, what recourse (other than social media outrage, articles like this one, or reports like HRW’s) do the Rohingya have?
“That data was taken from the bodies of human beings, and shared with those who explicitly want to cause them harm.”
All they get is a statement from UNHCR that entirely ignores the evidence being presented. UNHCR writes: “Each family’s consent was confirmed at least twice and consent signatures were only obtained following this double-confirmation.” This is provably false, as demonstrated by the testimony of Rohingya refugees in a variety of reports. HRW’s report isn’t the first time concerns have been raised; in November 2018, Rohingya refugees in Bangladesh went on strike, with one of their demands being: “Stop the collecting of our biodata, and do not share biodata already collected with the Myanmar government.”
In 2018, The Engine Room, the research consultancy where I work, spoke to Rohingya refugees who weren’t aware they could opt out of providing biometric data and had no idea their data was being shared further. One said they thought getting an iris scan was a test for eye disease.
How many times do the Rohingya have to make their experiences known for UNHCR to believe them? How many reports must be written outlining failures in processes and in decision-making for those processes to be examined for the harm they cause?
Responsibility
It’s not just process; it’s principle.
The biometric data of the Rohingya, as collected by UNHCR for humanitarian purposes, should never have been shared with the Myanmar government or the Bangladeshi government. Both have expressed a desire to get the Rohingya out of their respective countries and, in Myanmar’s case, its government and military stand accused of committing ethnic cleansing and crimes against humanity.
That is what is so galling in UNHCR’s response: It appears to believe that asking someone to trust a government who committed genocide against them is a reasonable thing to do. It is not.
If UNHCR truly intends to “safeguard the well-being and rights of refugees”, the harm it has caused deserves recognition, acknowledgment, and a direct apology, along with institutional overhauls to ensure such a scandal never happens again. Simply ignoring the evidence it is being presented with is, once again, disrespectful to the Rohingya who have made their experiences known.
Avoiding a repeat of this is quite straightforward on one level: If UNHCR cannot protect such personal data from host countries due to terms in their Host Country Agreement (we don’t know how much UNHCR was pressured by Bangladesh to gather the data), then it simply shouldn’t collect it in this way. Registration processes for refugees happened long before biometric data came into play, and there is still space to recognise that the risks of biometric registration far outweigh the potential benefits (as some humanitarian agencies already have done).
There are many factors that allowed this to happen: UNHCR’s legal immunity, a lack of data protection regulation in countries it operates in, political behind-the-scenes pressure, most likely. But we must aim higher; as analyst Aarathi Krishnan puts it, “grounding all our work and decisions in the rights and equity of impacted minoritised people, and rigorously analysing future harms.”
Forget hiding behind communications strategy, denial, and double-speak. Nobody believes it. The Rohingya deserved, and continue to deserve, better.
Zara Rahman Researcher and writer interested in the intersection of technology, power and social change. She works at The Engine Room, a non-profit organisation supporting civil society to use tech and data more strategically and responsibly.
Click here to subscribe to ESSF newsletters in English and/or French.