This is the fourth and final piece in a miniseries called “On flakiness.” Read the other parts at this link.

Shortly before Brett Kavanaugh was confirmed to the Supreme Court, he told Senator Susan Collins that Roe v. Wade, the landmark 1973 Supreme Court decision disallowing many state bans on abortion, was “settled law.” As Garrett Epps wrote in The Atlantic at the time, “[t]he implication was that Collins, who has indicated support for a [woman’s] right to choose, could vote for his confirmation without worrying about Roe or women’s reproductive rights.”

In other words, Collins could count on Kavanaugh to uphold Roe. Or so she thought: Kavanaugh, likely on purpose, did not say “I will uphold Roe.” And the recent legislature in Alabama and Georgia, passed in clear violation of Roe v. Wade and intended to bring a case before the Supreme Court, demonstrates that some are counting on Kavanaugh not to uphold Roe, but to overturn it.

This is the final piece in a series called, “On flakiness,” but I’m not going to say too much about flakiness in this article. Discussing flakiness gives us a small, safe arena to think about reliance and reasons, but we have bigger fish to fry. By the end of this article, the question will become the following: How could Susan Collins count on Kavanaugh to uphold Roe, while anti-abortion activists and lawmakers could simultaneously count on Kavanaugh to overturn it? (Attentive readers may notice this is not quite the plan I sketched last week. Have I flaked?)

If flaking is, broadly speaking, canceling plans for no good reason (or no “legitimate excuse”), the fact that we cite “good reason” picks out the fact that we count on people to do things — namely, what they say they will do. If I agree to have dinner with a friend, and I end up not having dinner with her, it should be because I have become ill, or an urgent and important matter has arisen. I should have a good reason — and if I don’t, then I’m flaking. In this sense, my friend counts on me to show up.

People can also make plans where everyone knows the plans are easy to wriggle out of, so one counts on anybody else to show up. In fact, everyone in such situations counts on everyone else not showing up, so that they don’t have to. If this sounds far-fetched, consider this exchange:

Alice: “We should hang out sometime!”

Bob: “Yes, definitely! I’ll text you.”

As long as everybody is implicitly on the same page that nobody means it and the plan should not be carried out, the plan works swimmingly, and neither Bob nor Alice show up. (This is a variation of the communicative flakiness I discussed two weeks ago.)

The ambiguity of the exchange means, however, that Alice and Bob are likely not on the same page. Perhaps Alice counts on Bob texting her, and Bob counts on Alice not expecting a text. The exchange is ambiguous enough that it’s hard to say what the expectations were. If Bob doesn’t text Alice, has he flaked on the plan to “hang out sometime”? Without more information on the expectations of Alice and Bob, we can’t say.

Such ambiguity is occurring in the Kavanaugh situation, but with much greater stakes. Did Kavanaugh mean “Roe is settled law, and therefore I will uphold it,” or “Roe is settled law, until the Supreme Court overturns it”? Given Kavanaugh’s statements, either interpretation is plausible enough that Senator Collins can count on Kavanaugh voting to uphold Roe while abortion opponents can count on Kavanaugh voting to overturn it.

Weaseling out of confirmation-hearing questions is a time-honored tradition for Supreme Court nominees, so it’s no surprise we are largely limited to speculation regarding Kavanaugh’s leanings, should Roe reach the Supreme Court. But there’s reason to think that Roe is in danger, and that Collins has made a mistake if, indeed, she is counting on Kavanaugh to uphold Roe: President Trump indicated that Roe is on the chopping block. Kavanaugh has shown contempt for precedent, and was supported by anti-abortion groups before he was confirmed. These groups, counting on Kavanaugh to swing the court against Roe, celebrated his confirmation, and seem to have good reason.

Perhaps Collins sought not an assurance from Kavanaugh that he would uphold Roe, but enough lip service that she could justify voting for him without losing credibility as a women’s rights supporter. But if she was, in fact, sincere, then Kavanaugh’s assurances seem far from sufficient to allow her to count on him, as she did in voting for him. And if Kavanaugh, in fact, meant that “Roe is settled law until the Supreme Court overturns it,” then he needs to give no reasons to Collins when he rules to overturn it.

Our discussion of flakiness illustrated that there are myriad situations where we count on people to do certain things because they say, or imply, that they will do them. And it showed us clearly that the more implicit an assurance, the more worried we should be about counting on it.

If Roe is revisited by the court, and Kavanaugh votes to overturn, Collins may feel betrayed. On the other hand, if Kavanaugh votes to uphold, anti-abortion groups who supported his appointment may feel betrayed. Who has the right to feel betrayed? It depends entirely on who could reasonably count on Kavanaugh to decide one way or another — who could expect that Kavanaugh would vote their preference unless he had a good overriding reason. But it’s entirely unclear which side should be counting on Kavanaugh.

We’re left in a deeply unsatisfying position. From situations with the lowest stakes, like flakiness, to situations with the highest stakes, like a Supreme Court decision affecting the lives of women across the United States, we don’t know what we should be counting on.

Yet we do count on things anyway. The ambiguities of assurances, from “Let’s get in touch soon” to “[Roe is] settled law,” allow us to interpret as we will. And sometimes this leads us to count on certain things happening when, perhaps, we should be counting on the opposite, or not counting on anything at all.

Perhaps Collins is right about Brett Kavanaugh. But in such a high-stakes situation, can we count on it?

Contact Adrian Liu at adliu ‘at’ stanford.edu.