[Epistemic status: working out a course of action for myself, based on somewhat commonsensical observations.]

(This post essentially agrees with Ends Don't Justify Means Among Humans and Making Exceptions to General Rules. It differs in emphasis and style. If you are reading this, please read them as well.)

I think the most important thing about the actions I take is what consequences they have. I do not see it as inherently valuable that my actions adhere to a predetermined set of rules, like a samurai's code of honor, a punk rocker's code of integrity, or the 613 mitzvot of Jewish law. What matters is only the effect on the world.

When I read about people taking rules and codes very seriously, I am sometimes impressed a bit more than would make sense on totally rational grounds. I think my brain intuitively respects those it sees as on the same path as itself but further along. I am an internal-code-following type of person. When I was in sixth grade I had a reputation among my friends for never telling a lie, such that I ruined jokes played on one by the other one. I broke down crying right before a surprise party because I was so uncomfortable with the white lies involved in setting up the surprise. From books I had gotten the impression that virtuous people always told the truth; being virtuous seemed much more important than anything else, so I resolved to never tell a lie. What was important was remaining pure, not any kind of outcome.

Now I'm 25 and not so into virtue ethics anymore, at least on a rational level. Yet I've come back around, in a way: I think it can be valuable, as a technique for ensuring that the actions I take have the best consequences, to follow rules and codes. Even knowing that I have a non-consequentialist bias towards this mode of thinking, I still think that this is true. Breaking a rule is not something that actually has a negative effect on the world, but for me I think it's helpful to avoid doing it just as if it were.

Self-imposed rules are tools for collaborating with my future self. Suppose I'm feeling relaxed, reflective, and well-rested. I consider a question about how I should behave in some circumstance, and come to some conclusion. Do I think I will always come to the same conclusion when faced with this circumstance, from now on? Surely not. My future self might be smarter than me, or have more information. Then they might come to a different and better conclusion. But it's also possible that my future self's decision is not better — they might be overcomplicating things, or feel very emotional, or have forgotten some key consideration, or face a conflict of interest. Establishing a rule lets me ensure that my future self confers with my past self before going ahead. It's like leaving a signpost in the ground saying "IF YOU WANT TO GO BEYOND THIS SIGNPOST, COME CHAT WITH ME."

One implication of this point of view is that I should take self-imposed rules pretty seriously. The system doesn't work if I don't follow them. I'm rejecting collaboration with my past self. I may be very convinced that I'm right, but I may be compromised.

If I still want to proceed after reflecting, I should change the rule instead of skirting it. It's important that there is a process by which the rule can be changed. It just shouldn't be too easy or fast. I want the process to require being relaxed and reflective again. I want to "summon up my past self" as much as possible, so my current self isn't making a unilateral decision. Keeping a written record of the rules I follow seems like a good idea.

Relatedly, this point of view says that I shouldn't take absolute vows that bind me forever. Do I really trust my future self that little? (Consider the success rate of this kind of vow in fiction.)

To summarize and operationalize a bit, here's the stance I'm going to adopt.

• Self-imposed rules are best thought of as tools to collaborate with my past/future selves.
• They are useful to make. They are most useful to make when I anticipate experiencing conflicts of interest around a type of decision.
• When I make one, I'll write it down and ensure I retain the ability to change it later. I'll avoid breaking it.
• If I later feel I need to change it, I'll spend a while thinking beforehand. I'll make changing it a sober, reflective process that takes a while. If end up changing the rule, I will celebrate it as my past and current self reaching an accord.

Open Questions/Issues

• How big is the danger of being too slow to change rules that ought to be changed, out of a desire to be consistent?
• Should I apply exactly this line of thinking to rules that are externally imposed? (Obviously some of them, but all of them?) In this case I am collaborating with a much wider circle of people than just myself.
• Sometimes people worry about "slippery slope" effects where breaking one small rule decreases the sanctity of other rules, leading them all to eventually get broken. Is this an important concern when thinking just about self-imposed rules?