US officials have much to learn from Afghanistan’s ‘Digital Dunkirk’

Some are calling it a “Digital Dunkirk” — an allusion to the World War II rescue and evacuation of Allied troops from Dunkirk, France to England. Except this time, it’s a virtual network linking arms and phones around the world to assist in evacuating tens of thousands of people from Afghanistan. 

This ad-hoc group of frontline civilians, veterans, active duty service members, members of Congress, government and nonprofit workers, open source technology specialists and others came together across borders over hundreds of Signal threads, WhatsApp groups, Slack channels, and Facebook Messenger chats. Volunteers shared information (and misinformation), vetted partners, liaised with government officials and generally tried to organize a mass of people thousands of miles away.

Initially focused on supporting Afghans seeking to leave the country via Hamid Karzai International Airport (HKIA), this effort has now moved to supporting those still in the country and assisting with the rapid resettlement of those who made it out.

This work has saved lives. Many of them. Life and death decisions were made in a heartbeat by people at times wholly new to this type of operation. They were moving quickly and were likely breaking things.

It was, and remains, the kind of integrated response between disparate groups that democratic governments can rarely achieve at scale. There were no formal classification buckets, few bureaucratic battles and little oversight. If partisanship reared its head briefly, it was swatted aside in favor of an intense mission focus.  

But we have moved out of this moment, and the operating environment has only grown more complex. Misinformation remains an issue. U.S. and allied governments need to improve the clarity and speed of communication, particularly with the volunteers engaged in this effort. Questions raised over the past days need to be addressed with urgency, both to protect lives in the coming weeks and to start the process of learning before the next crisis.

For example, was it worth the risk in the many, many cases in which an at-risk person’s biometric information was shared in encrypted, but largely unvetted, fora? What steps need to be taken now to mitigate the ongoing impact of these decisions?

On a broader scale, maps of Taliban checkpoints and gate openings at HKIA were shared freely, much like crowdsourced maps developed following earthquakes in Haiti and Nepal. These tools were critically important for those seeking to reach and access the airport, often after many days in hiding or sleeping outside the gates. 

The lightning-fast spread of electronic identification and evacuation instructions was necessary, but gave the Marines at the gates an impossible job. Did volunteer efforts help or hinder legitimate efforts by the United States military and, somewhat bizarrely, the Taliban, to maintain a secure perimeter around the airport? And can the reams of volunteer communications regarding the Taliban’s refusal to let American citizens access the airport be put to use in negotiations with the Taliban and neighboring countries during this next phase?

These are questions that need to be addressed not just for future crises, but right now. When individuals on the ground or in front of their screens feel a gap in leadership, whether real or otherwise, they will move on their own. Improving communication will help head off unnecessarily risky escape plans and routes and more quickly move people to safety. 

The State Department and Department of Defense should rapidly engage the primary organizations and individuals involved in this effort. Participation in these meetings should be limited to those willing to put aside the substance of the effort (the withdrawal and evacuation) and instead focus on what was done and more importantly, how it got done. The Department of Homeland Security should engage with volunteer groups on cybersecurity questions relating to sensitive data held in private hands. These conversations will also benefit the U.S. and like-minded governments in the longer term.  

Large open-source intelligence and humanitarian coordination efforts have the potential to help promote democratic values and humanitarian assistance, as has been shown following natural disasters and political crises over the past decade. Done right, these efforts can serve as a force multiplier for official efforts. 

The past weeks have been hard and painful. But amid the profound fear and grief, among the many wins and many failures, we have witnessed an extraordinarily dedicated group of citizens working toward a shared vision across partisan, professional, and ideological divides. This effort, though imperfect, represented some of what is best about America. Something amazing and spontaneous happened here: We should not let this moment go to waste. 

Ya’ara Barnoon is the associate program director for American University’s Tech, Law & Security Program.