The Case of Supreme Court Nominees Accused of Bad Behavior
We will never know for sure whether Judge Brett Kavanaugh sexually assaulted Dr. Christine Blasey Ford. And for many, Kavanaugh's ascension to the Supreme Court has left such a bad taste in their mouths that they'd prefer never to think about it again.
One thing of which we can be certain, however, is that the future will bring more nominations. Thinking through in advance about how to think through those nominations when they happen might very well be time well spent.
Let’s with a couple of things on which we’d probably all agree: if we knew for certain that a nominee to the Supreme Court committed sexual assault, their appointment to SCOTUS should be denied. On the other hand, if we knew for certain that no such assault occurred, then that nomination should move forward in the process.
Of course, given a credible but not airtight accusation we can’t be completely certain either way. Fortunately, there is a way that can help us all think a bit more rationally when it comes to deciding how to proceed, and to surface the sources of our disagreements.
Suppose that the probability that a nominee committed sexual assault is p. In the absence of an airtight alibi or irrefutable evidence of guilt, all we know is that p is greater than zero and less than one.
But here’s the thing: it turns out that we don’t need to know the exact value of p to know what to do next.
There’s one critical question we all need to answer: what’s the maximum value of p for which we are comfortable moving the nomination ahead rather than stopping it. If p is anything below that value, we should allow the candidate's nomination to proceed; if p is greater than that value, we should stop the process (either to investigate more fully or to find a better candidate.
This threshold value of p should reflect the balance of many factors. Some of these factors cause the threshold to rise: the cost of delaying the vote on an innocent nominee, the potential that sexual assault claims will be used by the opposition to gum up the nomination process in the future, the wear and tear of the process on an innocent nominee and his or her family, and the like.
Other factors cause the breakpoint to drop: the cost of appointing a person with a history of sexual assault to the SCOTUS, the difficulty of removing a bad justice once appointed, and the chilling effect that moving ahead without an investigation will have on victims of sexual assault coming forward in the future, the wear and tear on a victim of sexual assault whose claims are dismissed by the nomination process, and the like.
The uncertainty regarding the guilt or innocence of a nominee means that two errors are possible. On the one hand, if we demand a too high of a level of evidence about a candidate's guilt before stopping their nomination, we risk allowing some guilty nominees to proceed.
On the other hand, if we demand too low of a level of evidence about a candidate's guilt before stopping their nomination, we risk keeping innocent candidates from proceeding.
Where to set the threshold for evidence depends on how we feel about these two types of errors relative to outcomes in which we get it "right" - specifically, how we feel about allowing an innocent candidate to proceed, and about stopping a guilty nominee from moving ahead in the process.
In fact, if we're willing to assign specific values to each of these outcomes, we can calculate the level of evidence we should use in the nomination process. Specifically, we can determine the threshold probability below which we should act as though the nominee is innocent, and above which we should act as though the candidate is guilty.
Once we know the relative value of these outcomes, we can calculate your threshold probability.
Use the controls below to specify your values
Based on your values, your threshold probability is 50%. That means that if the probability that the nominee is guilty is...
The calculator above can help us think more clearly about the right level of evidence for nominations to the Supreme Court. Some Kavanaugh supporters appealed to the notion that people should be treated as being innocent until proven guilty. This appeal implies a high value for the probability threshold - one that can be mathematically achieved only if places a high value on advancing a guilty candidate (i.e., belief that the behavior of which Kavanaugh was accused wasn't all that bad) or a low value on stopping the nomination of an innocent candidate (i.e., weighting heavily the chilling effect that a stricter process might have on good candidates, or assuming that there's a shortge of qualified candidates).
A very different standard of evidence goes more along the lines of "above reproach," implying a far lower value for the threshold probability. Such values occur when people place a high value on weeding out guilty nominees and don't get too worked up over stopping the nomination of an innocent candidate. These values are consistent with a belief that sexual assault - even if committed far in the past - is a very bad behavior for nominees to have engaged, and with a belief that there is no shortage of other qualified candidates.
Regardless, it's likely that in the future we will all be better served by spending more time talking about how our values around these issues differ (or not) and less time speaking in certainties about the behaviors in which nominees may (or may not) have engaged.