Extraordinary ethics require extraordinary arguments

Previous post: Fighting the allure of depressive realism

My blog entries are about a personal battle against depression and anxiety, from the point of view of someone who has been immersed in rationalist/​LW ideas and culture for a few years now.

I want to illustrate a particular, recurring battle I have with scrupulosity. (I’m not the best at dialogues, so bear with me for a moment.)

Me: Alright, it’s time to be productive and get to my homework.
???: Hold on! How can you possibly justify that if you haven’t solved ethics yet?
Me: What? Who are you?
SD: Allow me to introduce myself, I’m your Skeptic Demon. I whisper ethical concerns in your ear to make sure you’re always doing the right thing.
Me: That sounds more like a daemon than a demon to me.
SD: Demon. Trust me.
Me: Solving ethics can’t possibly be expected of a single person, demon.
SD: Right you are! But you’ve looked around the world enough to know that everything you do could have ripple effects that might be disastrous. So how can you possibly feel good about working on your homework without accounting for that?
Me: What? It’s just homework.
SD: Oh, no it isn’t. Doing well on homework means sending a better signal to employers means more people want to hire you down the line, including for unscrupulous activities. And you’ve done not-great things before, so we can’t be sure you’ll resist. In fact the existence of first-, second-, third-, and nth-order effects implies you might not even realize when you’re being offered such.
Me: Erm… Well, it’s true that things have unintended consequences, but--
SD: No “buts”! You want to be a good person, right? So we gotta reason this out.
Me: I guess you have a point...
SD: Alright. So let’s get started.
(hours pass)
SD: Okay. You’re on shaky ice with some of these considerations. I’m not totally convinced you won’t be tempted by the money to go and do something net harmful yourself, but I will give you a one-time pass. You may proceed to start your assignment .
Me: I’m exhausted and I just want to go to sleep now.
SD: Then my work is done here. *disappears in a puff of shaky logic*

This kind of conversation happens to me all the time. Why?

On one level, it’s easy to see what the skeptic demon is doing. He’s trolling. He’s keeping me from doing the actual productive work I want to do, and very curiously never pops up to ask whether my watching TV or even whether my eating meat is ill-advised.

But he’s trolling with a legitimate issue—the fact that we can’t actually predict all of the possible consequences of our actions. It feels wrong to say that someone should be held ethically responsible for the sum total of that butterfly effect, but it feels equally wrong to deny they have any stake in it whatsoever. Trolls are worst when they find an issue that is near and dear to your heart and poke at it.

What to do? I’d like to at least justify why I think it’s okay to ignore this little guy.

I think we can get a lot of mileage here out of the old Carl Sagan heuristic, “Extraordinary claims require extaordinary evidence.” Here, it changes to extraordinary ethics require extraordinary arguments. And the idea that I should sabotage my own career out of the fear that I might accidentally harm someone down the line due to my own weakness is one heck of an extraordinary ethic.

For one, this ethic immediately fails my pop-philosophy understanding of the categorical imperative. If everyone acted like this, modern society and all of its woes would crumble, but so would its many, many, many benefits.

It also fails my understanding of why we usually give self-interest a seat at the table in ethics, even if we worry about its excesses: A world in which everyone spends all of their energy trying to make other people happy but never take time for themselves is a world where everyone runs themselves ragged and is uniformly miserable.

We could make the argument that people are far less morally responsible for second-, third-, etc. order effects from many different angles, one of my favorites being local validity. And so on.

I’m not sure how far I can take this heuristic before it breaks, but I think it’s a very wise starting point to begin with when it comes to issues of scrupulosity.