The Germanwings Crash and Gun Control

Co-pilot Andreas Lubitz carried out a horrifying mass murder of the passengers aboard Germanwings flight 9525. It has come to light that he was diagnosed with serious mental illness, but medical professionals were prevented by German law from communicating his condition to the airline, and Lubitz hid his psychological issues:

Strict medical privacy laws mean the companies were oblivious to the potential dangers lurking in Lubitz’s mind as the first officer took the plane into a steep descent over the region that members of his local gliding club, where he developed his passion for flying, had toured in the past. Confidentiality regulations, designed to protect medical data and encourage people to consult doctors without fear of repercussion, put the onus on patients to disclose potentially hazardous diagnoses to authorities and their employers.

“The medical secrecy rules are centuries old and touch the core of the medical profession,” said René Steinhaeuser, an attorney at Wigge lawyers in Hamburg who specializes in medical law. “Without that, the relationship between physician and patient, and thus the medical system as a whole, wouldn’t work.”

Tonight’s news includes suggestions that someone at Germanwings may have had information about Lubitz’s situation, but put that to one side for purposes of this discussion.

The point I want to make is that the privacy concerns that likely contributed to the Germanwings disaster are similar to those that we see here in the United States with respect to gun control. Everyone agrees that a deeply depressed, suicidal or homicidal person shouldn’t be piloting a commercial aircraft. Likewise, everyone agrees that such a person should not be purchasing firearms. But it is not easy to say how that goal should be achieved.

In the context of aviation, mental health professionals have warned that breaching the confidentiality of the doctor-patient relationship may deter depressed individuals like Lubitz from seeking help, making the problem worse rather than better.

In the firearms context, liberals are constantly agitating for universal background checks. In fact, background checks are already universal with respect to licensed firearms dealers, but arguments are made for expanding those checks to sales between private citizens.

In general, there are two types of people we don’t want possessing firearms: criminals and the mentally disturbed. With respect to criminals, background checks work reasonably well, as far as they go. If you have a felony conviction, you won’t be able to buy a gun from a licensed dealer. Criminals know this, of course, so they do one of three things: they buy guns from friends or fellow gang members, who would simply ignore a legal requirement to run a background check; or they steal firearms; or they have a straw purchaser, typically a girlfriend, buy the gun for them. So universal background checks would do nothing to prevent criminals from acquiring guns.

When gun control advocates talk about expanding background checks, they have in mind mass shootings. The problem here is that a background check is only as good as the list against which the check is run. Expanding background checks will do nothing unless the person in question is on the list, and experience shows that the people who perpetrate mass shootings, even though they may be palpably crazy, are seldom on the federal don’t-sell list. Seung-hui Cho, the Virginia Tech murderer who I believe is the worst “mass shooter” in American history, famously passed two background checks. NPR suggests that the problem is largely in the limited scope of the current law:

Federal law is supposed to keep guns out of the hands of the mentally ill, but many say the language in the current law is outdated and confusing.

“People adjudicated as mentally defective is one category, and the second category are people committed to mental institutions — and that’s the terminology that’s used in that law,” says Ron Honberg, director of policy and legal affairs at the National Alliance on Mental Illness.

He says the policy language is “horribly stigmatizing, and in fact so offensive that it’s hard to even discuss the substantive aspects of the law.” Honberg also says it’s not particularly helpful to officials who have to decide who should and shouldn’t be allowed to buy a gun. He says “no one really understands what it means.”

You can put people who have been adjudicated as mentally defective and those who have been committed to mental institutions on a don’t-sell list because you don’t need to violate doctor-patient confidentiality to do so. But what about those who are crazy as loons but haven’t yet been committed? Most mass shooters fall into this category, e.g., Seung-hui Cho, James Holmes and Adam Lanza.

Holmes is an interesting case. He passed a firearms background check, but the psychiatrist who treated him at the University of Colorado-Denver, Dr. Lynne Fenton, recognized that he was a danger to others and reported him to the University’s campus police. That did no good. As far as I know, her report did not satisfy the criteria for inclusion on the federal don’t-sell list, and, in any event, Holmes probably had accumulated plenty of weapons and ammunition before Dr. Fenton concluded that he was a threat to public safety. So her warning went nowhere.

This is a big topic, far beyond the scope of a single blog post. But there is an ineluctable tradeoff between security and medical privacy. If we want to be sure that pilots who fly commercial airplanes aren’t sliding off the deep end, we will have to require their doctors, psychiatrists and psychologists to report on them, perhaps speculatively, perhaps prematurely, perhaps wrongly. Likewise, if we want to prevent the mentally ill from obtaining firearms, we will need a far more robust system of getting their names into the don’t-sell database. At a minimum, this would entail requiring mental health professionals to blow the whistle on their patients, as some states have now legislated. Realistically, it would also require a system where friends and relatives could put a mentally ill person on the list, with, presumably, some sort of appeal process.

Some will argue that, given the rarity of mass shootings and pilot mass murder, the game is not worth the candle. That may well be the right answer. My point is that in both contexts, if we want more robust security, we may need very different privacy assumptions than those that have prevailed for the last several centuries.