Shoulds and Oughts and Better-Thans

In this series, I’ve been developing my answer to the Peter Thiel question:

“Tell me something that’s true, that almost nobody agrees with you on.”

My answer is that we should aim to believe what’s beneficial to believe, rather than what’s true. In previous posts I showed that the question is live and momentous, and that the traditional view is wrong. In this post I’ll work on some groundwork for the next part of the project, which is to make the positive case that we ought to aim to believe what’s beneficial to believe.

To make the case, we must first get clear about what we mean by the word “should”. There are a couple of observations to make.

First observation: by saying “we should aim” for x rather than “we do aim” for x, I’m making this about prescription rather than description. I’m not so much concerned about what humans actually do in practice — I’m interested in what we would do if we were better, stronger, braver, and more the people we wished we were.

Second observation: to say that a person should do x rather than y is to say that there’s some kind of normative force bearing upon the person by which x is in some sense better or more called-for than y. Let’s unpack this. First we’ll look at what it means for x to be better than y. In a future post, we’ll look at what it means for normative force to bear upon a person.

Evaluation criteria — Michael Jordan

Consider the following claim:

Michael Jordan is the best basketball player of all time.

How would we evaluate whether this is true? I might say that MJ is the best because was the most successful. But a Celtics fan might point out that MJ only won 6 NBA championships, whereas Bill Russell won 11. I might try again by saying that MJ had the most impact on the teams he played for — but again, a naysayer could naysay that MJ only won 5 MVP awards, whereas Kareem Abdul-Jabbar won 6.

We are at an impasse, and the reason is that the original claim is ambiguous. It’s not clear what I’m actually saying when I utter the words “Michael Jordan is the best basketball player of all time.” My objector and I are having what philosophers call a merely verbal dispute.

To resolve this, we must disambiguate. We can do this by specifying the evaluative criteria that are to be used in this context. In other words, I can specify the function I wish to use to map candidates to an evaluative score. That function might count MVP awards or league championships, or might use some other means entirely, such as the results of polling a panel of experts. Then we’ll just order the list by score select the highest-ranked candidate.

But what if I choose some weird criterion, such as the number of letters in the player’s name? My objector might say, “the number of letters in a player’s name has nothing to do with how good the player was.” This is surely true, but now we’re having a different argument. Now we’re talking about the difference between how a word should be used and how I’m using it. I might say, “look, for the purposes of this conversation, I’m going to use these words in a way that might not be conventional. Let’s focus on the meaning of what I’m saying, rather than the words I use to say it. Treat the words as mere variables that can be given any arbitrary meaning, and pay attention to the meaning we choose to assign.”

By taking these steps, we can some to an agreement about whether the original claim is true or false.

Evaluation criteria — shoulds and oughts

Just as the original claim about Michael Jordan was ambiguous, the claim that “we should believe what’s beneficial to believe” is ambiguous. To disambiguate and avoid a merely verbal dispute, I must specify what I mean by “should”. To do that, just as in the Michael Jordan case, I’ll have to select an evaluation criterion to be used in measuring the candidate aims. Then it’ll just be a matter of making the measurements and selecting the highest-ranked aim.

What kind of evaluation criteria might make sense in deciding what a person “should” do? Here are some intuitive ideas:

• Civic duty — person A ought to perform action x because to do so would maximize or fulfill his civic duty
• Legal — A ought to do x because x is legal
• Moral — A ought to do x because x is morally right
• Rational — A ought to do x because x is rational for A

We could come up with many more. Whichever one we choose will need to be specified further in order to achieve the required level of disambiguation.

So what’s the right evaluative criterion to use in deciding which doxastic aim is best? Uh-oh. It sounds like we’ll need another evaluative criterion to use at the next level up in order to measure and compare potential evaluative criteria at this level. If so, are we stuck in an infinite regress?

Good news! The answer is no — we’re not stuck in an infite regress. We can answer this question and decide on the ‘right’ evaluative criterion to use for evaluating doxastic aims. Stay tuned.