Thanks to my brother for discussion on this topic.

Overview

“Replacing Guilt” is a book where Nate Soares shares his thoughts on motivation done right. It has a big emphasis on instrinsic motivation, and generally discusses topics at a much lower-level than other books on motivation. In my mind there’s two main parts to the book:

• Part 1: On Having Something to Fight For
• Part 2: On Fighting for Something Effectively

Nate is now the president on MIRI, but wrote “Replacing Guilt” around 2015 when he was in his twenties. This book had been on my to-read list for a while, for a few reasons:

1. I have a lot of respect for Nate, and was curious to get a better sense for how he thinks about these types of things.
2. I’ve heard good things about it from friends.
3. It’s a classic in the rationality community.

And now I’ve read it. There were some parts of the book which didn’t really click with me / where I found the execution lacking, and there were other parts of the book that I thought got things very right in a way I had never seen elsewhere. For this post I want to share thoughts on part 1 of the book, by commenting on three chapters from it which I thought stood out in a good way:

1. You’re Allowed to Fight for Something
2. Caring About Something Larger than Yourself
3. You Don’t Get to Know What You’re Fighting For

You’re Allowed To Fight For Something

The main thing I got out of this chapter is the idea that “you are allowed to care about things outside yourself”.

For a specific application of this, consider the experience machine thought experiment, where you have an option to get plugged in and experience “the best possible thing”. Some people claim that you “should” prefer the experience machine because your experience is objectively better if you do.

At one point I wasn’t really sure how to think about this thought experiment. I wasn’t thinking of it as possible to care about things outside yourself, instead I was thinking that you were only allowed to care about your experience, and from there I thought it followed that you “should” prefer the experience machine. But now I agree with Nate, that you are allowed to care about something outside of your experience, and that this can be a reason to reject that you “should” prefer the experience machine.

Where I was tripped up before, is I thought that if you can’t perceive a difference between two universes, then you must prefer those two universes equally. But this implication now seems obviously wrong to me.

Nate touches on the idea of feeling like you need an excuse for caring about something outside yourself. I think he does a good job pointing out this common failure mode, and accompanies it with a reminder that actually, you don’t need an excuse to care about something outside yourself. If you want to, then you can! He says it plainly:

One friend of mine, after probing around in thought experiments such as this one, said “Huh. Well, so I definitely care about myself experiencing pleasure, and also I seem to care about other people actually existing and experiencing pleasure, though I don’t know why.”

She seemed surprised and confused to notice that she cared about others, as though this fact demanded explanation.

You don’t need an excuse. You can just care about things outside yourself.

More generally than the idea of “you are allowed to care about something outside yourself”, there is the idea that “you are allowed to care about something”. The latter can be a helpful reminder to those who feel like caring about anything at all requires justification, and like such justification is nowhere to be found. To them I say, “You’re mistaken - you actually can just care about something without justification! If you want to”.

(More on “If you want to” below)

This chapter also discusses the idea that “all behavior is selfish”. This is something I believed at one point when I was younger, but a lot of that thinking was rooted in confusion. Nate describes the archetype:

I’ve been surprised, in the past, by how many people vehemently resist the idea that they might not actually be selfish, deep down. I’ve seen some people do some incredible contortions in attempts to convince themselves that their ability to care about others is actually completely selfish.

Here’s how I thought about it at one point: you can treat a person like some black box, and that black box “wants” things, and its behavior is downstream of its wants, and its behavior is inherently “selfish” because those wants are its own, and therefore people aren’t “good”.

Here’s how I think about it now: you can treat a person like some black box, and that black box “wants” things, and its behavior is downstream of its wants, and its wants include things like “I want good things for my friends and family”, and behavior driven by these wants is not selfish in a way which “matters”, and therefore people can be “good”.

Here’s another direction for how I thought about it at one point: let’s say somebody wants_0 good things for their friends and family. Well, really, they want_1 their own wants_0, and that’s selfish.

Here’s another direction for how I think about it now: let’s say somebody wants_0 good things for their friends and family. Well, really, they want_1 their own wants_0, but this doesn’t make them selfish in a way which “matters”. They still want_0 good things for their friends and family!

Looking back, I was committing the non-central-fallacy. I was labelling something as “selfish” according to some definition, and then applying to that thing the emotional reaction I associate with selfishness (e.g. that people aren’t “good”), even though that thing doesn’t actually have the properties which motivate the emotional reaction I associate with selfishness.

There’s a related idea - some people will say that “people only act to maximize their own pleasure”. I can see how somebody might come to say this, but it’s just not true. Here’s a relevant story from replacing guilt:

Once upon a time, a group of naïve philosophers encountered a group of human beings. The humans seemed to keep selecting the actions that gave them pleasure. Sometimes they ate good food, sometimes they had sex, sometimes they made money to spend on pleasurable things later, but always (for the first few weeks) they took actions that led to pleasure.

But then one day, one of the humans gave lots of money to a charity.

“How can this be?” the philosophers asked, “Humans are pleasure-maximizers!” They thought for a few minutes, and then said, “Ah, it must be that their pleasure from giving the money to charity outweighed the pleasure they would have gotten from spending the money.”

Then a mother jumped in front of a car to save her child.

The naïve philosophers were stunned, until suddenly one of their number said “I get it! The immediate micro-pleasure of choosing that action must have outweighed —

Caring About Something Larger Than Yourself

The previous chapter was focused on “you are allowed to care about something”. This one is focused on “finding something to care about”. Here’s a relevant line:

Step zero is believing that you can care about something, and step one is finding something to care about.

One thing I think Nate does well, is make it clear that he’s not really trying to change your mind about anything. Instead, he lets you know that he’s just sharing ideas and the ways he thinks about things, and acknowledges that they may or may not resonate with you.

This property of his approach is great because it puts him on your team - you’re working together! That’s as opposed to approaches which come across more like they’re trying to change your mind on something, which can make for a more adversarial and less effective relationship.

Here’s an example of what I mean:

As with previous posts, don’t treat this as a sermon about why you should care about things that are larger than yourself; treat it as a reminder that you can

(For another example of writing which does a good job putting what could be adversaries on the same team, but which is otherwise unrelated, see Contra Robinson on Public Food)

More specifically, and more importantly, and something I see very rarely, is he doesn’t try to justify the the roots where his chains of thought bottom out with a priori reasoning. Instead, he acknowledges that they bottom out in things which “happen to be” true about him, and that they have no a priori justification, and points out that maybe they or something analogous to them might “happen to be” true for you, too. I think this style works very well for a certain type of reader, one which is suspicious of the roots of chains of thought, and incidentally that his audience is largely that kind of reader (i.e. the rationalists).

Here are lines which exemplify this:

treat it as a reminder that you can, if you want to.

And I am constructed such that when I look upon myself and find inconsistencies, I care about resolving them.

when I reflect upon the source of the feelings, I find arbitrary evolutionary settings that I don’t endorse, but when I reflect upon the sense of aesthetics, I find something that goes straight to the core of what I value.

There is no objective morality written on a tablet between the galaxies. There are no objective facts about what “actually matters.” But that’s because “mattering” isn’t a property of the universe. It’s a property of a person.

There are facts about what we care about, but they aren’t facts about the stars. They are facts about us.

And so we’ve returned to the phrase of “if you want to” - it captures acknowledging the way reasoning bottoms out in things which may or may not “happen to be” true for you, too.

This chapter is mostly summarized by its last few paragraphs:

Why do I care about humans and humanity, about Earth and all its children, about all sentient life? How can I say I do given that I, too, often feel more strongly for friends than strangers, and more compassion for dogs than men?

When I look upon myself, I see a tension between what I feel and a sense that my feelings are ill-calibrated. When I look closer, I find that the feelings are calibrated in ways I don’t endorse, in a tribal setting, where it was important to love the ingroup and hate the outgroup. But when I look at the sense that those feelings are ill-calibrated, I find good reasons, and a sense that this is actually what matters, that it is not arbitrary but valuable.

And so for me, “why care?” has an easy answer.

Let me stress again that you don’t have to resolve your internal tensions in the same way I do. Your answer to “why care?” might be “I don’t.” You might side more with your current feelings over your deeper sense of aesthetics, or you might have very different feelings and aesthetics. Either way, if you listen to that internal sense of friction, if you use your feelings as a guide rather than an answer, if you figure out why you feel and care as you do, and reflect upon your reasons, and separate feeling from caring, and choose to care about what seems right and good to care about —

then you may find that “why care?” has an easy answer for you, too.

And here’s how I internalized this chapter:

• It asks: assuming you believe you can care about something, why care? And what to care about?
• Then it answers: look inside yourself, and see if you’re somebody who happens to want to care about something. You might happen to find an ember of caring about something, and you might happen to want to nurture it, and at that point you might happen to do so. Remember: happening to be constructed to care is enough of a reason to care. A common failure mode is that a person thinks they need a better reason to care, and they can’t find one, and from there they smother their embers of caring and stop caring. But you don’t have to go down that path. Don’t overthink it.

You Don’t Get To Know What You’re Fighting For

I really liked this one. The main idea is given by the title: you don’t get to / have to know exactly what you’re fighting for.

Some people think that they do and end up over-thinking things. But you don’t have to do that. Instead, you can choose a direction to the best of your ability, and fight for that direction, without requiring knowing exactly what you’re fighting for:

That doesn’t mean we’re lost in the dark, either. We have a hell of a lot of evidence about our values. I tend to prefer pleasure to pain and joy to sadness, most of the time. I just don’t have an exact description of what I’m working towards.

And I don’t need one, to figure out what to do next. Not yet, anyway. I can’t tell you exactly where I’m going, but I can sure see which direction the arrow points.

Final Thoughts

Those three chapters are the main ones I found interesting on the topic of having something to fight for. Most of the rest of the chapters are focused on the following: once you have something to fight for, how do you do so effectively? And there are some great ideas among them - maybe another time I’ll write another post sharing my thoughts there.

Overall I think a lot of the value I got out of these posts comes from the “you’re allowed to” and “you don’t need an excuse” framings. It’s easy for people to reason themselves into false and limiting beliefs, and these framings can dislodge them. I’m reminded of this video which showcases someone with mastery of these framings: the ponytail guy deserves to be honored. And also of the concept of mu which I first learned about in GEB.

For me, reading “Replacing Guilt” felt less like struggling against ideas I didn’t want to agree with, and more like being exposed to ideas which quickly clicked, but which had never occurred to me before. That being said, I’m sure that I’m still confused and that there are more updates to come. Also, just because these ideas pointed in a good direction for me doesn’t mean that they will for all people - beware of other-optimizing.