We Should Ask Scientists What They Think About NSA Surveillance

A new article argues scientists and engineers are the perfect people to contribute to the NSA surveillance debate.

Aug 18 2014, 3:20pm

Snowden graffiti: Flickr/​Thierry Ehrmann</>

The raging public debate over the surveillance state could actually benefit from the expertise of an unsuspecting source, a recent academic article suggests.

Instead of relying on the myriad privacy and legal experts, congressmen, or former NSA directors chiming in on the NSA surveillance state, a new article in Science argues that we should really be asking more scientists what they think about domestic signals intelligence for American policymaking.

"Scientists and engineers can make an important contribution to this policy discussion," said Herbert Lin, a chief scientist at the Computer Science and Telecommunications Board and author of the paper. "It is not that they have a privileged position to address these issues but rather that they are uniquely qualified to tease apart technical aspects of these issues from social and political ones."

Lin suggests lawmaking doesn't always account for rapid evolutions in technology, but rather prescribes for one technological context, which can quickly become outdated. 

Citing the 1979 Supreme Court case Smith v. Maryland as an example, Lin argues the basis of that decision was made within a framework where the proliferation of cell phones and associated technologies weren't taken into consideration.

In the ruling, the Supreme Court argued that data about phone calls (based on landline usage), like the numbers of participants, was inherently worth less privacy protection than the actual information exchanged during the conversation of the phone call.

Given that called numbers were "being furnished to a third party" phone company, it was determined "the caller had a lesser privacy interest in that information than in the content of the call." In other words, eavesdropping was unconstitutional, but the surrounding information was fair game for policing agencies.

Cellphone use in America later boomed, and which allowed law enforcement to collect new sets of data not previously considered at the time of the ruling, such as geolocation data. In essence, Lin argues, it was the beginning of the debate between what is metadata, the bread and butter of law enforcement surveillance operations, and private data.

"Evolution of technology may call into question legal concepts formulated in an earlier technological context," said Lin, referring to changing mobile technologies. "What counts as data and metadata in e-mail? Is a buddy list on an instant messenger client or information posted to Facebook or Google+ data or metadata? Such questions arise because of new technology, but the answers depend on what society regards as the proper policy outcome."

In this type of debate, Lin says law enforcement attempts to "maximize their legal authority to obtain as much information as they can," whereas privacy advocates argue to limit those powers in the other direction. Both parties, he suggests, need to be mindful of the technological context surrounding policy decisions.

computer-based searches are different from human searches, but their answers are judgments about the formulation of national policy

Take for example, the search and seizure of citizen information. In 1792, when laws pertaining to the legal property of Americans were created, computer databases weren't in the wildest dreams of the founding fathers. But in today's context, a computer can scan and bank metadata on millions of Americans without the direction of a human being.

Law enforcement agencies contend that isn't search and seizure (they argue that only pertains to information queried for by a human), while privacy advocates say hoarding data in any form already constitutes an invasion of privacy.

Lin thinks these sorts of issues come up because key questions about computing haven't been answered. "Such questions arise because computer-based searches are different from human searches, but their answers are judgments about the formulation of national policy," he says.

The same goes for the debate surrounding whether or not warrantless data banking is useful for surveillance agencies. Critics of the system say mass capturing metadata is a waste of resources that never foils a terrorist plot. Oppositely, law enforcement considers it a critical tool in the fight against national security threats.

Lin proposes that the more useful conversation is whether or not technology can present a more efficient system for intelligence analysts that balances the demands of privacy and national security concerns. At the same time, such a system must include controls on access to that metadata database, perhaps by tracking who's using it and protecting citizen identities with homomorphic encryption

If, for example, faster and more sophisticated databases could algorithmically scan data, without viewing the identities of encrypted citizen data, the intelligence process could avoid the human eyes of analysts, while the computer identifies actual threats to national security, which appears to have been part of the rationale behind the NSA's PRISM program.

"A more constructive debate would focus on the potential trade-off—if certain information is unavailable, what would be the impact on the analytical process? Does making certain information available enable reaching valid conclusions faster? With higher confidence? With a lower error rate?" Lin asked.

In that context, intelligence agencies would require more access to the wealth of data we know was being scooped up by NSA SIGINT operations up until the Snowden revelations. 

Tech Experts would do well to weigh in on such debates, but the real question is: haven't they already?

That might prove too invasive a request for privacy experts wary of NSA intentions, yet the point is well taken: Could computer technology limit privacy concerns, all the while providing agents with better intel?

Experts in technological fields would do well to weigh in on such debates, but the real question is: haven't they already? Most of Lin's arguments are not unfamiliar to the ongoing surveillance debate. Either way, he does concede that putting controls on surveillance, or the input of scientists and engineers, cannot be the sole source of answers on the future of domestic surveillance.

"Technologies, always embedded within a broader milieu of policy and procedures," he says near the end of his paper, "will never be the sole—or necessarily most important—element of a solution to such dilemmas."

And they can't be. Important social and political concerns require the input of elected officials representing citizen interests. Scientific perspectives on surveillance can provide invaluable insights into the future of technologies, which in essence, will partially dictate the privacy problems policymakers will need to confront.

The rapidity of change in communications isn't well suited to the world of policymaking, which relies heavily on sober second thoughts and political squabbling before decisions can actually be made. Nonetheless, making space for scientific delegations to add their two-cents could help alleviate some of the issues Lin maps out in his paper.

But I can't help but wonder, with the many congressional advisory meetings, (indeed I've even seen similar committees in Canadian parliament with little attention being paid), how could scientists really be effectively situated into policy debates on surveillance? Experts weigh in on any given subject all the time, while the politicians listen as much as it is politically expedient.

In the end, scientific eyes in surveillance discussions could vastly improve the process behind determining what our policies should look like. It could even help conclude, ultimately, the debate around metadata versus data. 

Putting scientific eyes on issues requiring a fluid knowledge of things like encryption and the technical aspects of signals intelligence could give clearer determinations on how those technologies should actually be wielded. 

But the truth is, like in matters of health related research or climate change, scientists can opine all they want about government surveillance, but the folks on Capitol Hill are the ones who will make the final call.