The Precautionary Principle & Information Technology: Airlines & Gadgets Edition

by on September 9, 2012 · 2 comments

Psychologists Daniel Simons and Christopher Chabris had an interesting editorial in The Wall Street Journal this weekend asking, “Do Our Gadgets Really Threaten Planes?” They conducted an online survey of 492 American adults who have flown in the past year and found that “40% said they did not turn their phones off completely during takeoff and landing on their most recent flight; more than 7% left their phones on, with the Wi-Fi and cellular communications functions active. And 2% pulled a full Baldwin, actively using their phones when they weren’t supposed to.”

Despite the widespread prevalence of such law-breaking activity, planes aren’t falling from the sky and yet the Federal Aviation Administration continues to enforce the rule prohibiting the use of digital gadgets during certain times during flight. “Why has the regulation remained in force for so long despite the lack of solid evidence to support it?” Simons and Chabris ask. They note:

Human minds are notoriously overzealous “cause detectors.” When two events occur close in time, and one plausibly might have caused the other, we tend to assume it did. There is no reason to doubt the anecdotes told by airline personnel about glitches that have occurred on flights when they also have discovered someone illicitly using a device. But when thinking about these anecdotes, we don’t consider that glitches also occur in the absence of illicit gadget use. More important, we don’t consider how often gadgets have been in use when flights have been completed without a hitch. Our survey strongly suggests that there are multiple gadget violators on almost every flight.

That’s all certain true, but what actually motivated this ban — and has ensured its continuation despite a lack of evidence it is needed to diminish technological risk — is the precautionary principle. As the authors correct note:

Fear is a powerful motivator, and precaution is a natural response. Regulators are loath to make policies less restrictive, out of a justifiable concern for passenger safety. It is easy to visualize the horrific consequences should a phone cause a plane to crash, so the FAA imposes this inconvenience as a precaution.

Once a restriction is in place, though, removing it becomes a challenge because every day without a gadget-induced accident cements our belief that the status quo is right and justified. Unfortunately, this logic is little better than that of Homer Simpson, who organized an elaborate Bear Patrol in the city of Springfield and exulted in the absence of bear sightings that ensued.

This is a prime example of the precautionary principle in action. In my recent 80-page paper entitled, “Technopanics, Threat Inflation, and the Danger of an Information Technology Precautionary Principle.” I noted that how we might be witnessing the rise of a “precautionary principle” for some information technology policy matters. The adoption of an information precautionary principle would restrict progress in this arena until technology creators or proponents can demonstrate new tools are perfectly safe. That’s essentially what the FAA has done with its ban on digital gadgets during certain times of air travel.

Of course, it is easier to sympathize with the precautionary perspective in this case than others because the risks of digital gadgetry and wireless communications during flight really were unknown early on, and few wanted to conduct a real-time experiment when the potential downsides were so catastrophic. And yet, as Simons and Chabris observe, we’ve conducted that experiment anyway! Air travelers decided to ignore the ban and continue to use digital gadgets. And, luckily, the sky didn’t fall, or in this case planes didn’t fall out of the sky, at least.

What’s amazing about this case, however, is that the FAA has continued to enforce its precautionary-minded regulation long after it’s been shown to be needed and has been so widely ignored anyway. I suppose that, like Homer Simpson, some of these officials believe that their precautionary steps have led to greater safety, or don’t have any costs or trade-offs and, therefore, there’s nothing wrong with their “better to be safe than sorry” thinking. Of course, that’s the fatal flaw in all precautionary principle thinking, as I note in my paper. There most certainly are many costs and trade-offs associated with banning technology or its use. They may not be as profound in this case as in others, but that doesn’t mean that they do not exist.

Regardless, now that the FAA has finally decided to take a second look at their policy, perhaps they be willing to admit that there never really was much sense to this particular application of the precautionary principle and that the time has come to end this ban and let individual airlines experiment with different approaches.

  • Jim R

    Without disputing the ideas contained in this article, I think there is another factor that contributes to the reluctance of FAA and airlines to change these rules. The other factor is the varied nature of the aircraft fleet.

    For example, most systems on the Boeing 747 were state-of-the-art techology in 1970, long before mobile telephones were invented. In the meantime, the Boing 727 has come and largely gone, but you’ll still find large numbers of DC-8 derived MD-80 and -88 jets in domestic service.

    Because of the way equipment on airplanes is tested, piece-by-piece, we have amazingly safe flight. But the tests that most equipment have been subjected to (e.g., RF susceptibility) were designed before we had all these mobile devices to consider. In fact, today’s mobile phones operate on frequencies that weren’t even commercially practical to use in 1970.

    If the FAA were to relax its ban, on what basis would it do so? Do we assume that since nothing bad has happened, that it must be OK? Or do we redesign these tests to ensure that the aircraft equipment can safely operate in the presence of these devices? In fact, these tests have already undergone redesign and newer aircraft are tested to newer standards. But if the FAA were to say, “OK, go ahead”, how would we differentiate between that brand new 787 being flown by Delta and their oldest MD-80?

    Furthermore, this matter isn’t just a question of phones. Passengers are bring wireless transmitters on-board airplanes. For example, Seagate sells a low cost battery powered disk drive that is intended to stream video to an iPad. This device is basically a portable WiFi access point. If a passenger activates this on an airplane, it behaves much differently than a cell phone. And while we’re at it, not all phones are created equal, either. The power output of portable devices can get as high as 600 mW, or down to below 1 mW, and still be considered a “cell phone”.

    And then we have satellite phones — a whole different situation. Can these be operated from an airplane? Who knows. Who could tell the difference if you saw one in the cabin?

    Adam, believe it or not, some questions are not simple a matter of social engineering. There is some real engineering in the background. The consequences of an error or overight in this area are large. Fortunately, we have built good machines and they are operated by people who are generally very well trained to identify and work around problems. But just because there have been no incidents, we cannot assume that all is well. We know that (and should take comfort) in the fact that our commercial fleet is fault-tolerant. But we know that two fault, or even three, at the same time, can be devastating.

    Why would FAA allow the first fault into that equation?

  • http://twitter.com/AdamThierer AdamThierer

    Jim… Thanks for your comment. I think you raise some very fair points here and I do appreciate some of the challenges that the FAA faces. When you ask, “If the FAA were to relax its ban, on what basis would it do so?” I think you are raising the proper question. The agency should conduct, as it is apparently doing now, a rigorous cost-benefit test of the rule to determine what, if any, continuing benefit there is to retaining the regulation. If you are correct that some continuing risks exist, even if for just some classes of planes, then certainly someone out there should be able to demonstrate it, no? Perhaps we will find that a hybrid strategy is necessary with restrictions on some planes or flights but not others. Or perhaps the FAA will leave that determination to individual airlines according to their fleet and their crew/passenger preferences.

    The point is, a little flexibility and experimentation here are probably sensible, and indeed has already been taking place. While the Simons-Chabris survey isn’t as broad or rigorous as I’d like, I think that the sort of non-compliance with the rule that they find is likely conservative, if anything. I bet there are even more people violating the ban than they suggest. And if we have been conducting this natural experimentation for many years now, and if we realize that non-compliance is only likely to increase, then it makes sense for the agency to conduct a rigorous cost-benefit analysis of the rule to determine if it can be justified and retained. But the agency must present some solid science to do so and not just base the ban on a continued precautionary principle mindset. That’s not good enough.

    Thanks again for your comment — Adam T.

Previous post:

Next post: