In elementary school, we’re told that if you don’t know what to do in a situation, it’s helpful to make a list of pros and cons.
Today, I would like to convince you that this is completely insane.
I. How do pro-con lists actually work?
Let’s dive right in: how, exactly, does the pro-con list lead to a conclusion? The naïve solution is to just go with the side that has more items on it: we go to the grocery store, because there are 3 pros, but only 2 cons. Unfortunately, this breaks down pretty fast. If someone offers you a red M&M, an eraser, and a lightbulb, with the caveat that you have to pay $10,000, don’t take the 3:1 trade1.
So let’s revise our hypothesis. Let’s say that hiding underneath here is some sort of utilitarianism: different items produce different utility, and we want to pick the side with the most total utility (or least negative). In the trade example, this actually checks out. M&Ms, erasers, and lightbulbs have some total market value; the sum is substantially less than $10,000. However, in the grocery store example, this is…weird. We could say that getting more milk is 5 utils, light exercise is 3 utils, the time cost is 7 utils, etc. But like…really? I’m sure someone out there actually thinks this way, but for me at least, this is intuitively bizarre2.
To revise again, we could say that pro-con lists are an attempt to recreate what’s happening in our subconscious. Scott Alexander suggests that our brains are doing these sums under the hood. Hey, he’d know better than I would; last I heard, our brains were probably neural nets.
But: even if we accept this as true, we cannot blithely assume that the model that works at the micro level works equally well at the macro level, or the level we observe. Our universe runs on quantum physics, but Newton’s laws make much better predictions of the objects we’re familiar with. Individual subatomic particles do all sorts of strange and improbable things, but I’m pretty sure I’m not going to phase into my couch.
The other thing is, our neurons are really fast. They send electrical signals or something; I slept through most of my biology lectures, but I think those are pretty fast. But when we sit down to write a pro-con list, it’s been seconds, minutes, hours, or days since the problem first occurred to us. So clearly something else is going on when we try to consciously think through a decision. Perhaps we’ve landed in some special mental state that doesn’t correspond with either choice.
So how do we think through decisions? Let me try to start again.
II. What is indecision?
We make most of our decisions very quickly. You can imagine this to be due to those electrochemical signals, if you like. You were probably very quick when you made the decision to read this piece: you were at least mildly curious about its contents, and you didn’t have a good reason not to read it.
In that light, there is something weird about the grocery store example: it is hard to imagine someone actually making a pro-con list for that, isn’t it? In our lives, we might notice we’re out of milk, and then feel an impulse to go to the store; and then we pretty quickly either decide to go, or decide it’s not worth it (we’re too lazy, or it’s too risky). If we did see someone anguishing over whether to go to the grocery store, let’s say for five minutes or more, we’d assume the decision must have some hidden emotional significance for them. Imagine you really did have a friend who was spending this long on such a decision: what would you guess their thoughts are like?
Pro-con lists only come up when you’re stuck on something for a while, let’s say at least a minute or two. I think these situations can be broken into three categories:
Calculation. Some things just take time to do the math. Let’s say you know you want to buy a car, but you’re not sure you can afford it; so you go and calculate how much savings you have, how much you make per month, etc. At the end of the calculation, you make a decision.
Unmeasurable. You’re going on a first date, and you know you want your date to think you’re hot; what should you wear? There’s a clear goal, but you can only guess at how to achieve it.
Or: you’re at a high-society party, and you know you want to be liked, but you don’t know proper etiquette. Should you approach someone, or hang back? You aren’t really conflicted per se, but you don’t know how to get what you want.
Indecision. You have two distinct emotional impulses that want you to do different things3. This is where we find most internal conflicts in stories. You want to whistleblow on your company for ethical violations, but you’re worried about your livelihood. That sort of thing.
For our grocery-troubled friend, this indecision turns out to be a better model of the situation than the pro-con list. Here’s the first thing I notice: while the pro-con list implies that all the items on each side are all of the same kind, in truth, there’s one core impulse on the “pro” side. If we weren’t out of milk, we would be much less likely to have the idea to go to the grocery store at all.
In the same way, indecision must have one core antagonist; our grocery friend is probably either very hung up on the COVID risk, or has some sort of complex about being lazy. Both statements might be true, just as it’s true that walking to the store would be light exercise, but I think our internal experience of indecision is that we pit one essential thing against another at a time.
Recently I told a friend I would go on a trip; after I’d expressed interest, I realized that I didn’t really want to go, but I also didn’t want to bail on my friend. When I thought about the trip, I was able to produce a great variety of reasons why I might want to go, or might not want to go. But at any given moment, what it felt like was one core impulse against another.
Hey, don’t take my word for it. I encourage you to compare this with examples from your own life. Let me know if you can prove me wrong.
III. What a pro-con list really is
Let’s say I’m right. What would be the consequences of that?
First, pro-con lists become pretty useless4. If your problem is calculable, you just calculate it; if it’s unmeasurable, then by definition you have no way of knowing what’s right; and if it’s indecision, then it’s an apples-and-oranges comparison. Pro-con lists appear to collapse indecision into a single axis of “good” vs. “bad,” but this is just an illusion. Things are only considered “good” or “bad” because they are attached to us in some more specific way: e.g., light exercise might be “good” because you expect you will feel better after, or because you want to build an identity as a healthy person.
So pro-con lists don’t really model the decision-making process; but there is something they do model very well, which is the mental state of indecision.
Here’s my thesis: pro-con lists do not resolve indecision. In fact, indecision produces pro-con lists.
For me, the trip example was just like this. I had one particular feeling of “this trip isn’t going to be very fun for me,” and one particular feeling of “I don’t want to bail on my friend.” Thinking about it further produced many more reasons on each side, but it didn’t change anything about the core conflict. In the end I didn’t really “choose” at all; eventually, the situation resolved itself when my friend (perhaps sensing my hesitation) canceled it for me. I’m not too proud of how I handled that one.
It’s natural to think of the items on a pro-con list as being objects that we might place on a two-sided scale, like Anubis weighing the souls of the dead against a feather. Instead, I think it’s better to think of the binary at the core — “do A” vs. “do B” — as a stone wall, with intricate patterns of rationalization growing out like vines on either side.
We still haven’t gotten to the fun stuff. I wanted to talk about love. But this has gone on a bit long, so I’m afraid that’ll have to be in part two.
You could try to defend this by saying that paying $10,000 would have multiple cons (e.g. multiple other things you can’t buy), but then where do you stop? The lightbulb could be useful as a source of illumination and as a makeshift weapon.
For more on why utilitarianism doesn’t really yield useful results, I recommend this piece from the wonderful @interpretantion, which has inspired a lot of my thinking here.
I sense that some of you will object that indecision is just a special case of the unmeasurable, where you are optimizing for “happiness,” or like, eudaimonia. But, uh, what’s that? I think you will find that this shared quality can only be defined tautologically — “it’s the thing that all the stuff you want has in common.” We have no reason to believe this shared quality exists, except because we’re looking for it. Contrast with “I want my date to tell me I’m cute,” which is pretty concrete.
Okay, look. I do see some use in pro-con lists as a ritual. You may have heard of this one: imagine tossing a coin to make your decision, and see which outcome you hope for. In this way, you can learn something about the core impulses underlying your indecision. But this only works if you can somehow “convince yourself” temporarily that the coin flip is “real”; and this is where the pro-con list could be useful, in allowing you to perform this act of theatre with yourself.
Another way pro-con lists could “work” is if you bootstrapped them into your identity; that is, you become religiously attached to being “the kind of person” who uses a particular decision-making system. But surely nobody would do that.