Humility
Reports that say that something hasn’t happened are always interesting to me, because as we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns — the ones we don’t know we don’t know.
— Donald Rumsfeld“Compounding their problem of no job security in the decision-making process is the single most important fact, perhaps, of the entire movie industry:
NOBODY KNOWS ANYTHING.”
— William Goldman
Based on a review of polls, betting markets, prediction models, and expert opinion, I expected Hillary Clinton to win. She did not. In the wake of that profoundly unexpected result, I find myself questioning not merely the pieces of evidence and analysis on which I hung my expectation, but my own instincts, my understanding of politics and of the American electorate.
(Obviously, this was a tight election, and this blog might not exist had a hundred thousand or so votes gone the other way, but still: the election wasn’t even supposed to be this close. I certainly did not expect it to be.)
So my certainty indicts myself, and now the one thing of which I am most sure is that I know a whole lot less than I thought I knew. Or, perhaps, I know that I know a lot less than I believed that I knew.
So, for me, many of the most fundamental questions raised by the election of Donald Trump are not moral but epistemological. How do we know what we know? How do we know when we don’t know what we think we know?
Coincidentally, that question seems to be at issue in one of the biggest post-election narratives: the preponderance of so-called fake news, its extraordinarily fast spread through certain social networks, and its likely affect on voter behavior during the election. But I’m more concerned about what this election says about us than I am concerned about what the election says about the viral spread of misleading information. Frankly, as politically thorny as it is, that issue seems a lot less important to me.
While some fake news boils down to unfalsifiable assertions, relying either on unknowable motivations (like basically any statement calling someone a liar or a cynic) or on untestable predictions (like basically any statement about what either presidential candidate would do after their election), much of what we’re calling “fake news” is objectively false, because it was written by people who say as much. Much of it seems to have originated with overly on-the-nose satirists who may have enjoyed the viewership they got from overly credulous sharers, or from Macedonian content mercenaries who noticed that there were easy advertising dollars to be had by making outrageous stuff up. You don’t need Snopes to know that Andy Borowitz isn’t reporting the news, and no social network should pretend that it’s unknowable.
(Not ALL “fake news” is easily categorizable, obviously. It is hard to know exactly how to categorize a column-length gust of partisan vitriol that has been legitimately, if barely plausibly, wrapped around a single undeniably true fact, though Harry Frankfurt might have a word for it. The preponderance of piping-hot takes has clearly had a deleterious effect on discourse, but nothing about this election result particularly changes that.)
As politically thorny an issue as “fake news” undoubtedly is — boiling down, predictably, as virtually every issue always does, into a virtue-signaling concern-trolling partisan he-said-she-said — it’s frankly a lot less difficult to untangle than the simple haunting fact that those of us who claim to understand politics are almost certainly overstating the case. As any behaviorist will tell you, it is awfully difficult to self-correct for overconfidence. For one thing, we are evolutionarily wired that way. More simply, the imperfections in our instruments are irreducible. It’s hard to examine our own biases through the lenses that we use to examine the rest of the world, because they’re the only lenses we’ve got.
Still, there are a few things we can safely assert.
We must be extraordinarily skeptical of manifestations of scientism: the risible notion that science alone can solve all of society’s problems. It is frequently expressed, such as, for example, here:
And here:
The belief that data is dispositive is dangerously naive, but it is very seductive, particularly among the well-educated. Obviously, in our modern polarized politics, scientific consensus frequently breaches on the shoals of partisan bickering, and these debates are typically not grounded in first principles. But even in a tendentious argument, liberals are ill-served by the inane notion that evidence obviates debate. It is important for liberals who believe that their ideas are both supported by evidence and by values to remember that values are as important as evidence.
This is both a tactical and a strategic insight. Tactically, it is nigh impossible to persuade anyone without arguing from the point of view of one's interlocutor, or you will have no hope of changing their mind. So it is important to understand the role that values play in persuasion. Strategically, it is necessary to hold to a consistent set of values, because that is the only way to build a movement behind a coherent set of conclusions. A raw dataset does not produce its own theory, or its own solutions, or its own voting bloc, or its own political majority. On the other hand, moral suasion might.
So, perhaps I should not have said that what concerned me about this election was not morality but epistemology: I believe the one informs the other. It is impossible to believe in a vacuum, impossible to persuade in a vacuum, impossible to govern in a vacuum. But while we all know the dangers to reason when faith becomes dogma, we—or at least I—think less deeply and often about the similar dangers of what Julian Sanchez calls epistemic closure, or what we could simply call unearned certitude.
At its core, the confession of faith requires an acknowledgement of the limits of one’s own knowledge. I freely confess. The election revealed something profoundly broken, both in the instruments we use to measure political data, and in the party that lost the election.
Memento mori. Nobody knows anything. That’s a place to start.
