Embracing “Impossible” Data

Image result for snowing indoors

As I waited for my train to London in one of those cavernous railroad stations up North, flakes of snow started to fall around me. My first thought was “Huh – it’s snowing”, followed seconds later with a shocking realization: “I’m indoors…and it’s snowing!!!”.

I looked up into the gloomy reaches of the arched ceiling high above me and concluded there must a hole in the roof through which an outdoor snowstorm was casting some flakes. I walked outside to check. It was certainly a cold November day, but the sky was clear and there was not even a skiff on the ground. Yet when I walked back inside, it was still very lightly snowing by the tracks where I had been standing.

Later that evening, in a downstairs bar off Pall Mall, I related my strange tale to my companions, who began forming theories. Because this particular watering hole is popular with spooks — who enjoy eavesdropping and puzzle solving in equal measure — pretty soon the whole place was engaged in a lively debate regarding how my impossible data could indeed be possible. It was fun discussion and without rancor.

Contrast that with different impossible data: Your doctor brings back your “routine tests” and says that even though you feel fine, you are gravely ill. Something in you shouts NO and you understandably come up with every possible reason why the impossible data just can’t be correct.

Those two cases of “impossible data” are at the extremes where the data are either entirely fun and non-threatening to learn from vs. terrifying to the core. Most impossible data is between those poles, and I wonder as a teacher and as a citizen whether we can instill in people a stronger habit of seeing impossible data like indoor snow instead of proof of terminal illness.

How do we get a gun rights advocate to do something other than scream “fake news!” when a study shows that gun owners are more likely to be shot? How do we get a firm atheist to appreciate evidence that highly religious people are happier and healthier? What is the magic that makes impossible data an exciting chance to learn more about the world rather than something to shut out at all costs?

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

4 thoughts on “Embracing “Impossible” Data”

  1. It’s easier to accept “impossible data” when it doesn’t threaten a major disruption of one’s life, either as it is or as one wants it to be. Snow inside the train station is an interesting mental challenge to people who aren’t responsible for figuring out whether it’s a symptom of a problem with the infrastructure and, if so, for figuring out how to fix it without shutting down the station. The person who has those responsibilities and who learns about the problem the day before leaving for their son’s wedding in Bermuda followed by a long-awaited three-week cruise, will be more likely to respond with a vehement “NO! Can’t be happening!” than with a mild, “Well, that’s an interesting puzzle, to be sure.”

    And so to the gun advocate who resists admitting that gun owners are more likely to be shot. This concession would seem fairly non-disruptive, but add the slippery slope to the equation: the gun advocate may recognize the commonsense fact that people who spend more time in the vicinity of guns handled by non-professionals are more likely to be shot, but also fears that conceding even the smallest point will lead inevitably to losing everything, which would be disruptive—witness how quickly anyone proposing fairly minor steps, such as required background checks or licensing gun use at the same level as driving, is accused of having the ultimate goal of banning guns entirely.

    The question is, of course, why this is the case, and I think it has to do, in part at least, with one of the things that activates slippery slope reasoning: an underlying belief that one’s opponents are not operating in good faith. Thus, any concession to the opposing side risks eventual disruption as extreme as learning that one has a life-threatening illness.

    I have no idea how to solve that problem.

    1. I really appreciate your comment, both because you actually engage with the substance of the post and because of your analysis of it.

      I hadn’t considered but agree with your point about context making the same impossible fact more or less tolerable. The optimistic implication is that we might be able to create contexts in which it is easier for people to acknowledge impossible facts. Twitter is a good example of a context that makes it harder, because of so many people (overwhelming male) treating it as player vs player and not a place to learn anything. But if that’s the worst, what is the best, what is the place where the guy who goes through life as if it’s a cable television politics show can put down his fists and open his mind? I don’t know the answer but I appreciate you making me thinking about.

      I find compelling also your comment that many people are suspecting others of leaping the is ought gap during “discovery”. So if they don’t want gas taxes raised, they dispute evidence that car exhaust contributes to climate change, even though there is no essential connection between that fact and raised gas taxes — we could know that cars do this damage but intervene in other ways or not intervene at all. I wonder if we could help people (again, mostly males) unlearn this approach, which basically keeps one stupid to avoid a possible political argument later.

      Anyway, thanks again, great comment,

      1. I spent a bit of time yesterday thinking about your last question: what elements create a context in which someone is more likely to take an extreme position and refuse to budge versus discussing an issue in good faith–and perhaps necessarily with some detachment. Here are a few thought on elements of context that are likely to make a difference:

        1. The degree to which the speaker perceives that the ramifications of the “impossible data” will be disruptive to them personally or to people or situations with which they have strong sympathies.
        2. The genre of communication: an engagement on Twitter is vastly different from a quiet personal conversation with a friend.
        3. The point, what the communication is about: marching slogans and protest signs are about display, not conversation. I’m not on Twitter, although I observe it sometimes, but it appears to have become a medium where the norm is that a Tweet is primarily for announcing one’s thoughts or positions or for displaying one’s wit or one’s alignment, rather than an effort to begin a thoughtful discussion.
        4. Purpose or goal: is the speaker’s intent to persuade and gain allies versus delivering an unanswerable statement and silencing opponents.
        5. Participants and (importantly) overhearers. A Catholic bishop might acknowledge in private conversation with a friend that he supports ordination of women, but probably won’t if other priests are participating in the conversation or are in a position to overhear. (Unless the bishop is a Jesuit, in which case all bets seem to be pretty much off, from what I’ve been able to observe.) Here also is where we find the speaker’s beliefs about the other participants, including whether they are arguing in good faith or not.

        This is, unsurprisingly, starting to look a lot like Dell Hymes’ SPEAKING model (Setting/Scene, Participants, Ends, Act sequence, Key, Instrumentalities, Norms, Genre). I’ll add another element:

        6. The framing of the issue: gun advocates have been trained (by a very well organized lobby and support group) to think in terms of “us versus them,” a frame which leads to assumptions that the positions are polar opposites, and that anything short of the extreme on one end or the other is either dishonest or a washed-out, and therefore illegitimate, version of the position. The train station snow example didn’t involve starting with opposing positions. To take another example, the discovery of archaeopteryx resulted in scientists and others interested in evolution taking it as an interesting puzzle, and many of those even moderately on the creationist side immediately stating that the animal isn’t really transitional, nothing to see here, move along.

        To some degree, certainly, a single individual can engage in a range of behaviors. I know people who can detach from issues about which they feel strongly to engage in intelligent and even-handed discussion about those issues. Presumably this is what we hope the people in our government can do. On the other hand, my experience has been that some people just aren’t comfortable with uncertainty. I suspect we can all think of well-educated and intelligent people whose response to any view that differs from their own is to grab for the nearest slogan or folk theory to shut down the discussion. This does not necessarily correlate with political position or degree of questioning of the status quo.

        As far as getting people to unclench their fists, some, I’m afraid, simply won’t. The desire for certainty is powerful. But some contexts (classroom discussion, one hopes) can create greater likelihood that participants will see an issue as an interesting problem to be explored rather than a cue for taking their pre-established position and not budging.

        I’m also afraid that the general tendency is to go in the direction of certainty, even a false certainty, simply because it’s easiest to decide what one’s position is and not have to think about it anymore.

        1. Thanks for the thorough analysis. There is evidence from Paul Slovic’s work that people have a preference for “affective simplicity” — rather than deal with a mixed judgment, like “The car is unusually fun to drive but has a poor safety record” they “purify” their judgement by believing that either the funness or the risk must be overstated, and it’s just a super fun car with no risk or a car that’s no fun to drive and unsafe as well. This tendency is a barrier to agreement, because it doesn’t feel emotionally satisfying to many people to say “This or that historical figure did both some remarkable things and also some despicable things”.

          I am particularly intrigued by your comments about the presence of “opponents” in discussion. One thing that has happened on many strands of the political right is a shift against incarceration and tough on crime rhetoric. This has been an internal discussion, perhaps because if liberals had been involved there just would have been position-taking and fighting, such that people would have been compelled not to change their views “on the field of battle”. There is also of course the dynamic of rationally not wanting to air your ingroup’s dirty laundry. It’s eye-opening for many white people for example to watch political chat shows on BET or Univision and see how differently politicians and people of color talk among themselves versus in the larger media where large numbers of white people are always around and watching.

Comments are closed.