Some common mislocutions are simply insufferable. One is, “It’s in my/our DNA,” implying that a commitment or habit or practiced behavior is genetically hard-wired. No, it isn’t.
Another is, “I/we will follow the science” to make judgments. If by this is meant, “I will face facts rather than engage in wishful thinking,” then hear, hear. If by it is meant, “I will heed experts to explain data that I am less equipped then they to interpret,” then again, hear, hear.
But it seems that all too often what is really meant is, “I will be objective but we all know that you are not,” or some other means of asserting certitude and foreclosing argument and discussion. In this sense, appeals to “the science” take the place of bygone appeals to Holy Writ. And when so invoked, “the science” is sometimes, I fear, misused. There are at least three errors to which we should be alert.
First is failure to account for uncertainties in measurement or assessment. This has been on full display in discussions of epidemiologic forecasts of the effects of the COVID-19 pandemic. All empiric measurement and analysis carry uncertainty, which must be assessed and reported, especially when forecasting complex events. Readers who doubt this are referred to Nate Silver’s excellent and entertaining 2012 book, The Signal and the Noise. Forecasts are only as good as the data that go into them, and always have some degree of uncertainty. The forecaster should account for that, and, in fact, to the degree possible, should also assess how sensitive the forecasts are to changes in the input assumptions.
Forecasters such as Dr. Murray at the IHME do indeed report their uncertainty (look at those wide shaded bands with the forecasts), but these are often lost in discussion, much less their sources and underlying assumptions reviewed and sensitivity to changes in those assumptions described. (Listening to his interviews, I wonder whether Dr. Murray is not somewhat pessimistically oriented in his assumptions, but that is not a point for this blog post.) Then again, as more data are collected, the forecasts should be more refined, and they are. Some policy makers in the current outbreak chose to risk erring on the side of caution—a reasonable approach—but that does not mean either that they were entitled to a claim of certainty in the predictions or that the forecasts themselves were somehow suspect or illegitimate in retrospect. It was just the nature of the uncertainty. But recognizing that demands some humility in prospective judgment and in retrospection.
Second is confusion of facts with values. It is one thing to say, “if we do X, then my best estimates suggest there is serious risk of Y,” and quite another to say, “because of my forecasts, X is out of the question because of the risk of Y,” without also assessing the real costs of both X and Y and their relative likelihood. That is, there are tradeoffs. Now, epidemiologists and economists both clearly are aware of tradeoffs, but the former seem to be viewed as somehow purer in aims and methods than the latter. Yet a prominent bioethicist who has counseled extreme caution regarding relaxing mitigation efforts in the pandemic offered on television that schools might be opened soon as long as parents were informed of the risks of sending their children back to school. This is a defensible position. It’s just not one that falls out of scientific assessment. It involves value judgments, and no one in a position of authority approaches the current situation value-free. If we forget that, then our medical professionals might come off as, or be viewed as, scolds.
Third is the stubbornness of the ad hominem. We should accord properly-credentialed experts their place of honor and authority in complex discussions, but their expertise does not confer on them certitude even of their prognoses (see above), much less certitude about all the consequences for action that might ensue. To wit: last month, when Governor Cuomo asserted, quite energetically, that New York State needed 30,000 ventilators to care for people who had fallen gravely ill, he was challenged by one citing a countervailing forecast that suggested the state’s true need was substantially less. But that challenge was articulated by Mr. Kushner, citing forecasts the White House was looking at, and a common public response was that the White House stance must be incorrect or corrupt precisely because Mr. Kushner (who, one friend has recently told me with confidence, is “a clown”) was making it. I don’t know exactly what Mr. Kushner said, or whether he was right, or whether even his projection was more or less accurate or prudent at the time. The actual number of ventilators needed by New York for this phase of the outbreak, with more data, now appears less than 30,000. Why? The answer is not clear. One can think of several possible explanations.
Still—someone once wisely told me that the merits of an argument depend on its quality, not on the identity of who is advancing it—or on one’s opinion of that person, or on one’s opinion of one’s own intelligence. The current crisis demands calm assessment of “the science,” properly interpreted, as well as of the very difficult tradeoffs that appear to be in the offing. And, in this, the experts are not the only ones who should have a voice. We demand that last point in the relatively small matters of human subject research and use of biomedical technology. How much more so in the current moment.