These are all important, in their own way, call it a treasure hunt and collect them all…

Editor’s Note

This was a tough one to write. There was something to disagree with on every page. This ended up being crazy long, but even so I made quite a lot of cuts, including in exactly what parts to quote. This could have easily ended up being longer than the original book, if I had let it.

The original book returns to many subjects several times. I tried to organize my responses to group similar statements together while preserving dependencies, but this is tough. I try to not duplicate my objections more than necessary, but in some places this is one of those ‘I did not have time to write a shorter response’ situations.

I am not a formal philosopher. I am presumably going to say a lot of stupid things, including things with known strong refutations, or that most philosophers otherwise think are dumb. I am not entirely comfortable with this, but I strongly believe that any true philosopher, and especially Agnes, would tell me not to let that stop me.

The point is not to always be right, the point is to have the argument, to inquire exactly in the spirit of the enterprise. And if it turns out that the day I wrote that part I was being stupider than usual? Then sorry about that, my bad.

Also in the spirit of the original book, in many ways large portions of this were written as a response to one particular person, rather than primarily for a general audience.

I am not making a strong bid that you, a different reader, need to read all or any of this.

I offer it in case you find it relevant to your interests. Some people should read this.

But let me be clear: There is a large chance that THIS IS NOT YOU. In which case, that is totally fine, and you can and should stop here and take these two days off.

Later, for length, I sometimes start saying ‘I disagree that [X]’ without quoting where Agnes says [X]. I hope I make things clear via context.

My hope is, by using this as a springboard, I can illustrate my philosophy of thought.

Besides, who doesn’t love a good argument?

Table of Contents

1. Editor’s Note.
2. A Difference Of Opinion.
3. An Overview.
4. You Don’t Know Me.
5. Untimely Questions.
6. The Unexamined Life is Worth Living.
7. The Quest For The Unexamined Life.
8. Not Everyone Wants To or Should Philosophize All Day.
9. The Seinfeld Fallacy.
10. Socrates was the Lying GOAT of Hypocritical False Humility.
11. Hearing Voices.
12. Simpsons Ancient Greeks Did It.
13. The Proposed Fourth Option: Socratic Inquiry.
14. No Really The Position is Nothing Else Matters.
15. The War on Wavering and Nebulosity.
16. Living Your Best Life.
17. Introducing the Socratic Method (the real one).
18. Prove Me Wrong, Kids.
19. Socrates Asserts Wrong Conclusions That Are Wrong.
20. You Can Question Your Beliefs.
21. True Opinions Do Not Only Do Good.
22. Meno Plays the Fool.
23. The Central Magician’s Trick.
24. The Gaslighting of Alcibiades.
25. The Measure of a Fight.
26. The Good Fight.
27. The Curious Case of Euthyphro.
28. You Should Be Sad About That.
29. People Respond To Incentives.
30. Self Versus Other.
31. Socrates Declares Humans Have Unified Minds Free Of Various Biases.
32. Revenge.
33. Legal Systems Very Different From Our Own.
34. Socrates Claims The Just And The Advantageous Are Identical.
35. First Up: Utilitarianism.
36. The Main Rival: Deontology (Kantianism? Stoicism?).
37. A Trolly Problem.
38. The Third and Correct Option: Virtue Ethics.
39. You Are Not Omniscient.
40. The Hardest Thing In This World Is To Live In It.
41. They Call It Utopia For A Reason.
42. The End… Of Book One.

A Difference Of Opinion

I hate Socrates so, so much. For so, so many reasons. What a schmuck.

As much as I hate Socrates, Agnes Callard loves Socrates more. Probably.

Oh my does Agnes Callard love Socrates. She wanted to be Socrates. She wants to do Socratic things all day. She married someone who goes by Aristotle. She wrote a book telling us Socrates is the key and the one true path to knowledge and even to thinking, and without his insights and method life is not worth living.

I have never read a book in which I have disagreed as strongly, with as many claims and arguments, as I did here.

Whereas when I actually had the opportunity for a few day to spend mornings at Lighthaven chatting with Agnes Callard in person, Doing the Thing? That was wonderful. She was wonderful. I was thinking, it is such a shame one gets to do this so rarely, or so well. Which also highlighted, in many ways, how that experience was different from what the book actually lays out.

What we were doing felt like The Way, whereas what this book lays out? In so many aspects, Not The Way.

When I reviewed Nate Silver’s On the Edge, it was very much a story of ‘yes, and.’ I took what he’d written, and there were disagreements, but mostly I built upon it.

This is going to be a ‘no, you fool.’ A huge percentage of the quotes are, to my view, Agnes Callard being Wrong in Her Book, and it’s time to bring it.

Why bring it?

Partly because that’s exactly what she would want me to do.

Partly because it’s fun.

Mostly it’s because we are in the midst of teaching the sand how to think, and rapidly nearing humanity’s final Philosophy exam and its how-to-think exam. We cannot leave this one to the professionals. I’ve seen their work.

In the past, I’ve avoided doing explicit Philosophy on the Internet, because I assumed I would end up looking stupid. I’d make elementary ‘mistakes,’ I’d walk into traps, I’d be picking an argument with people who pick arguments for a living and doing so on their own turf and so on.

I’ve only even read The Symposium, back in college, after which I kind of tried to cancel Plato before cancelation was a thing (we’ll get to that later but I wasn’t wrong), and bits of The Republic, and I remember very little of any of it.

I’d be the fool who spoke and removed all doubt.

Yeah, screw that. I’m so over it.

Basically: Get in, loser. None of that matters. We’re doing philosophy.

Kind of all of it. And we’re going to do it by telling Agnes Callard that she’s wrong.

[As with all book reviews, location numbers refer to the Kindle version, and block quotes are from the book unless otherwise specified. Quotes from dialogues are taken from the book and the original source marked.]

An Overview

Open Socrates is a book that implores us to engage in two-person philosophical dialogue, where one person asserts things, and the other questions and seeks to disprove them. This, says the book, is both The Good and the royal road to wisdom.

You Don’t Know Me

Agnes has some bold claims about your life right out of the gate.

Your life is full. It has no space for the question, “Why am I doing any of this?” (49)

You make sure your thinking about how your life should go doesn’t wander too far from how it is already going. You appear to be afraid of something. (64)

You usually act based on cached answers and justifications.

Which is what you absolutely should be doing, in most individual moments.

You could uncache your answers at any time.

People rarely do so. I don’t think most people lack the time. We don’t want to do that.

She asks why you haven’t done that: ‘What are you, chicken?’ What are you afraid of?

I think most people are rightfully afraid it will make their lives worse. Or that it won’t make their lives better, and they they won’t enjoy doing it, that it’s a waste of time.

And indeed, she confirms that very answer right away, in her own chosen example.

Untimely Questions

There’s no perfect time to properly define and address the concept of Untimely Questions. I’m pulling this forward a bit.

The kinds of questions Tolstoy warned us against asking form a special class: I call them “untimely questions.” Untimely questions are marked by the fact we need answers to them before we are prepared to ask them. (334)

The reason why you can’t ask yourself untimely questions is that you think you already have the answer to them, and the reason you think you have the answer is that you are using the answer. (434)

It is more than that, or rather something more precise.

The problem with an Untimely Question is not merely that you need to choose a functional answer to operate under before you can ask the question yourself in some more ‘think about it yourself’ sense.

The true problem comes when the act of investigating the question further is incompatible with continuing to rely on the previous answer, and you actively need to rely on that answer until you have a new one.

The answer is load bearing to wellbeing or proper functioning. You can’t remove it without replacement.

It’s not about not having the time to ask. It’s about the uncertainty itself doing harm.

That happens because of various ways the human brain handles uncertainty, and also the ability of others to detect and respond to that uncertainty.

That doesn’t mean that, if you actually were asked the question, you would think you have the answer because you are using the answer. It would be more like, ‘from the outside I can see I might not have the answer here, but I have to go on as if I have the answer.’ Brains are weird, and not as unified as Socrates often tries to pretend.

Suppose you previously thought [X]. Now you suspect, perhaps [~X].

In many cases, you continue to ponder whether [X] is true, while continuing to otherwise act, or mostly act, as if [X] is true. It is often correct to do this, for questions that would otherwise be untimely.

A central example is a decision to quit a job. You might quietly pursue other options. but until you know for sure you are leaving you want to do most of the things you’d be doing anyway, maintain the same attitude, and often conceal that you might leave.

Another example would be a startup. You need to act as if you have confidence in your success, while also facing the reality of the situation and deciding whether to pack it in or pivot or panic, and planning to avoid true ruin if you fail.

Thus I think this description here is subtly wrong:

I believe that Tolstoy identified a special class of question that I am going to call “untimely questions.” An untimely question is a question that comes at the wrong time—namely, after it has been answered.

Untimely questions are remarkable not only because they are hard to answer, but, first and foremost, because they are hard to ask; and they are hard to ask not only because it is hard to pose them to others, but, first and foremost, because it is hard to pose them to oneself. (425)

It’s fine for a question to come after being answered – that doesn’t have to make them untimely. It’s only an issue because having confidence in the answer is importantly load bearing, and you can’t properly ask the question without disrupting the load bearing until you have your new answer. That’s what makes them ‘hard to pose.’

Most Untimely Questions are not, at any given time, things it makes sense to be asking. It is usually better to have cached knowledge that relies on other sources.

Tolstoy’s issue was that he couldn’t do that. His answers had become non-functional. He became temporarily unable to live an unexamined life.

Which was a shame for him, because the unexamined life is worth living, too.

The Unexamined Life is Worth Living

Socrates believed that “the unexamined life is not worth living for a human being,” and that belief motivated him to make time for untimely questions. (774)

There are advantages to examining and asking questions, especially when you are in need of the answers.

But what kind of pompous ass would claim life cannot otherwise be worth living?

Oh, right. That pompous ass.

Tolstoy examined his life, once he had exhausted all alternatives and had no answers.

It… didn’t go great.

The whole thing was weird, Agnes points out, because Tolstoy had won at life.

My [Tolstoy’s] question, which at the age of fifty brought me to the point of suicide, was the very simple question that lies in the soul of every human being, from a silly child to the wisest sage—the question without which life is impossible, as I experienced in actual fact. The question is this: What will come from what I do and from what I will do tomorrow—what will come from my whole life? (392)

The natural explanation for why someone thinks their life is going badly when it gives every outward sign of going well is that there is something off with the part of the body whose malfunctioning characteristically escapes the mind’s notice: the brain. But when it comes to Tolstoy, we are in an awkward position to draw these conclusions. (402)

The explanation is: Tolstoy got tired of too much winning.

Or, more precisely: He got to the ‘You Win’ screen at the end of the game.

A person moves through life by envisioning positive outcomes either for herself, or those around her, and working to bring them about. If there is nothing that could happen that would satisfy you, motivational collapse ensues.

The fact that Tolstoy could face such a collapse is an alarming indictment of the very concept of success, since very few human lives have been characterized by more substantial success than Tolstoy’s.

Tolstoy’s life “came to a halt” after he had written and been recognized for War and Peace and Anna Karenina, and, more generally, when his life was going just about as well as it could possibly go for a person. (72)

Or, more classically:

Not that I’ve read Tolstoy’s Confessions, but it sounds like that was Tolstoy’s actual problem: He no longer had anything to strive for.

That’s not an indictment of striving, only of his definition of success.

I kind of wrote an entire book about not getting trapped obsessing over the wrong kind of ‘success,’ including that fully ‘winning’ still sucked. I certainly sympathize.

Alternatively, I mean, the dude was depressed because he had a chemical imbalance? It happens. He’s a Russian novelist. And he no longer had sufficient distractions.

I notice the contrast with the way Agnes portrays it:

Tolstoy’s crisis was caused by his inability to answer certain questions. He has everything he ever wanted in life—literary fame, a loving family, wealth, physical and mental stamina—but finds himself plagued by the thought that he cannot understand why he cares about any of those things (83)

Why seek material prosperity? Why educate my children? Why care about the welfare of the people? Why does literary fame matter? These are untimely questions, and they form a contrast with the sorts of questions that float free of what we are currently doing, questions where open-mindedness is possible, questions whose answers we needn’t rely on already knowing. Those sorts of questions come at the right time. (446)

The point of all of that is, essentially, to get the rest of it. It’s the self-sustaining circle of life and competition. So the very fact that he ‘has it all,’ and we’ve essentially ruled out ways to meaningfully scale it further, and he no longer had to fight to maintain it – there was no marginal payoff left – is exactly the reason it no longer seemed to have a point. The art no longer had an end other than itself. What good would more wealth or more fame do him? What more could he do for his family, at that point?

The obvious things for Tolstoy to do next are some combination of:

1. Enjoy the fruits of victory, bask in it, have goals related to this.
2. Find further goals and strive for those, either different or more of the same.
3. Find meaning some other way, which was Tolstoy’s ultimate solution, faith.

I also notice you can ask questions like ‘why educate my children?’ while continuing to (in this example) educate your children as before. You can say:

1. I should have a very strong prior that I should educate my children.
2. I also notice I do not understand ‘why’ I should do that.
3. I should also try and figure out why, or maybe even why not, if not.
4. But in the meantime, I still believe that I should keep educating them!

The Quest For The Unexamined Life

Instead, Tolstoy was paralyzed and made himself miserable looking for a way out:

Throughout Confession, Tolstoy is tormented by the problem of how to move on: How do I go back to a life centered on writing novels, managing my estate, attending to my family?

How do I go back to anything I could have previously recognized as life? He cannot answer, and the result is that he becomes “convinced of the need for suicide.”

By his own lights, what Tolstoy discovered is that the examined life was not worth living.

Although he found an escape from these questions—and from suicide—in religious faith, Tolstoy is clear that faith is a way of setting them aside, not an answer to them. (96)

Let ‘Tolstoy should write novels, manage his estate and attend to his family’ be [X].

Tolstoy isn’t questioning [X]. He still knows [X] it true.

What he doesn’t know is why [X] remains true for him. He can no longer successfully cache the belief that [X] in a load bearing, functional way. He realizes that believing [~X] would go really badly, and that everyone around him asserts [X].

One could say, hence I believe [X]. Indeed, that is why most of us believe most [X]s.

Alas, in this case he finds this insufficiently persuasive to believe [X] in a load bearing way. He realizes this is a problem. He sets out to fix it, whether by figuring out why or otherwise.

Before, Tolstoy had the benefit of civilization’s previous work figuring out how to live in the world and care about it and do the dumb things you gotta do. The question of why, of ‘what’s my motivation here?’ was what Agnes calls an Untimely Question, because his actions depended on having an answer.

But yes! That’s the point. You really, actually do need to have a functional, practical answer to all in-context Untimely Questions in order to live life, do the thing, score the victory points. That’s what makes them Untimely. Not unanswerable, but untimely. That’s why they’re not important now.

Agnes says that Tolstoy discovered The Examined Life is not worth living. That is indeed exactly the opposite of the Socratic thesis. But I don’t think that’s right.

Tolstoy discovered, without realizing it explicitly, that the life without purpose, without load bearing answers, without The Great Work, is not worth living.

So he sought, quite reasonably, a new Great Work, and found one.

A philosopher would say, and Agnes does say, halt! That answer is invalid, for it is not examined. It is arbitrary. It was lying around and you picked it up. Inquire!

And I would say, yes it was lying around. That’s a feature not a bug. It helps the answer stick and be compatible with those around you. How well does that answer work, in practice? It might well work great.

The point of examining it is if you suspect it won’t work or isn’t working, or it isn’t accomplishing what is necessary, or if you think you might find one that works better or that accomplishes something more valuable, or you could improve it, or if you see other value in inquiring.

Indeed one could say that the whole reason we make these questions so hard to ask is that we have realized how badly it typically goes when people ask them.

If it strikes you as somehow brutal and uncaring that Tolstoy is willing to countenance such thoughts as “Why should I care about my children” and “So the peasants are suffering, what’s it to me?” your judgmental response—How dare he?!—points to the unaskability of the corresponding question.

You are not supposed to regard those questions as open, precisely because you are supposed to already be using the answers, in the caring that you are currently doing.

But in that case, how was Tolstoy, who did care about his children, and about the welfare of the peasants, able to ask himself those questions?

The answer is that he wasn’t. He could say the words of the questions to himself, but he couldn’t ask them. (468)

It’s not that Tolstoy couldn’t ask the questions. He could have. He didn’t want to.

Though Tolstoy repeatedly refers to the process of inquiry into fundamental questions, his text betrays no sign of his having performed such inquiry: actual philosophical reasoning and argumentation are strikingly absent from it. (475)

It is evident that Tolstoy was in anguish over questions about the meaning of his own life, but it is less evident that he actually succeeded in asking those questions. (511)

The Socratic interpretation is that Tolstoy did not try and fail; rather, he failed to try. (628)

Tolstoy identified a problem: His answers stopped working. So he set out to solve that problem. He found a solution. What he didn’t do was ‘ask the questions’ in the way Agnes would have him ask, instead asking different questions, and finding answers.

And that’s because…

Not Everyone Wants To or Should Philosophize All Day

Socrates and Agnes would say nothing is better than doing philosophy all day.

For some people, that is true. That is their Great Work. That is what they most enjoy.

I mean, in that case, good for Socrates. But don’t torture Tolstoy.

Tolstoy found that the “why” question made existence unbearable: “I had no life.” Socrates described the prospect of spending an eternity inquiring into it as “an extraordinary happiness.” (113)

Tolstoy clearly does not see philosophy sparking joy in his heart. He hates it. He wants nothing to do with it. He is the unfortunate man who was happy to be asleep, forced to awaken. Now he seeks to return to his rest.

Socrates is totally fine if the art lacks an end other than itself, and collapses into infinite recursion. To him this act of seeking The Good itself via The Good is The Good.

Does he examine that fact? Kind of yes, but I suspect also kind of no. I’m sure Agnes could quote a lot of passages. I have a hunch I could make a reasonable case why they don’t count and he’s pulling a fast one.

For others, like myself, philosophy can more fun than most things people think are fun when it’s done well. Otherwise we wouldn’t be having this conversation. But the art needs to have an end other than itself.

We can ask questions such as ‘what is virtue?’ in order to have a cool discussion. But I don’t buy that one should ask ‘what is virtue?’ because seeking knowledge like this is inherently The Good – or if one does do that, one has transformed their ultimate motivation into what Agnes calls an Untimely Question, exactly the way she accuses everyone else of doing.

Rather, we mostly ask ‘what is virtue?’ because we need to figure out how to actually have virtue, or help others have virtue, or make the AI have virtue. Like so many before us, the goal of the philosophy we do is to take the question we are working on out of philosophy and into some other field.

I’m not mad at Socrates for spending his time that way (although see implementation details). I am, however, mad at him for claiming that this should be such a happiness for everyone, and that nothing else could be such a happiness, and indeed everyone not doing this was doing bad and should feel bad. Such as Tolstoy.

The Seinfeld Fallacy

The Seinfeld Fallacy is one I have lived and know well: The inability to set aside the examination of life in order to actually live and enjoy it, or accomplish.

Taking life fifteen minutes at a time is a Tolstoyan strategy. The name for the opposite strategy is “Socratic.” (106)

I would say that taking life fifteen minutes at a time, or any other amount at a time (why fifteen minutes, exactly?) is mostly the right way to live a given fifteen minutes.

The Seinfeldian version, or the one I suffered from, is the micro fallacy. This is where you are constantly analyzing and criticizing and evaluating everything in micro. Always deliberate practice, all the time, in everything.

There is great joy in that, for the right person, and also great long term benefits.

But you also need to often be able to step outside of that, and either enjoy or accomplish the thing. To ‘get out of your own head’ and ‘stop overthinking it.’

I think a lot of my ‘secret sauce’ is the ability to be constantly optimizing the micro, without letting it overly interfere with my ability to actually do things or enjoy them.

Then there’s the macro version of the same fallacy. I’ve known many people who suffer from that, too. Where you spend all your time and energy thinking about abstractions and what to do and why, in ways that don’t ultimately lead to action. Where your abstractions aren’t agentic, and you can’t set them aside. Either they’re too distracting, or they leave you unmotivated.

Socratic life strategy or moral theory, as described in Open Socrates, is indeed the opposite of the Tolstoyan strategy. It never sets out and faces the moment and Does the Thing. It suffers from the macro fallacy, and potentially also the micro fallacy.

Socrates was the Lying GOAT of Hypocritical False Humility

[Socrates] was also famously ugly—bug-eyed, snub-nosed, and goatish, in a city where personal beauty was as highly prized as wealth or fame. (122)

One could speculate that this was a big motivation for Socrates. If you are ugly and poor, but you can convince everyone around you that the physical world doesn’t matter and all everyone should ever do is philosophy and that is the true version of everything including sex and love, well, you can see the advantages.

Indeed, you can see throughout the book that this strategy worked well for him, even if you think it wasn’t intentional.

Socrates was not only famously ugly, but also poor, and he often remarked on his lack of intellectual gifts. He confesses to having a bad memory, and denies any facility with speechmaking, those being the two essential markers of intelligence in fifth-century Athens. (141)

Socrates was very obviously lying.

Don’t tell me that this ugly, poor person who supposedly managed to consistently outwit and convince so many of the leading intellectual men of Athens, whose words were so dangerous he had to be put to death, and so on, lacked intellectual gifts – or was under any delusions that he lacked them. Don’t tell me he thought he lacked facility with speechmaking, when we’ve seen so many of his historically convincing speeches.

Don’t tell me he had a bad memory either. When and what does he actually forget, when it counts? The closest thing the LLMs could find was Theaetetus 167e-168c? o3-mini-high flat out just said ‘no, it’s almost always a tactic.’

Then of course we get to the big one, most of you already know this story. It is perhaps the thing he is most known for.

If you are on trial for your life, and you have the chance to tell the jury one story about yourself, then you will take great care to select the right one. Socrates, finding himself in just such a situation, chose to talk about the time that his friend Chaerephon took a trip to the oracle at Delphi.

The oracle, which was held to communicate the will of Apollo, was the supreme religious authority for people throughout the Greek world. Chaerephon asked the oracle whether there was anyone wiser than Socrates; its answer was “no.”

Upon hearing this, Socrates is shocked:

When I heard of this reply I asked myself: “Whatever does the god mean? What is his riddle? I am very conscious that I am not wise at all; what then does he mean by saying that I am the wisest? For surely he does not lie; it is not legitimate for him to do so.”

What is probable, gentlemen, is that in fact the god is wise and that his oracular response meant that human wisdom is worth little or nothing, and that when he says this man, Socrates, he is using my name as an example, as if he said: “This man among you, mortals, is wisest who, like Socrates, understands that his wisdom is worthless.” (161)

I’m flat out calling Socrates a liar here, in his most famous claim, three times over.

1. Obviously he is not stupid enough to think that the Oracle is divinely inspired and always tells the truth. We see Socrates be a skeptic of so many other things, yet here he does not question, for ‘it is not legitimate for him to do so?’ If Socrates was thinking at that level we would not still be discussing him.
2. He can drop the whole central ‘I am the fool who at least knows he knows nothing’ act. For someone who knows they know nothing, Socrates claims to know a hell of a lot of things, both things that are true and things that are at best highly questionable.
   1. That doesn’t mean the part where he seeks to know more, or realizes there is so much more to know, is an act. But seriously, come on.
3. That’s clearly not the correct way to question the Oracle here.
   1. Socrates is not so stupid as to ask the question that way. What happens if the answer comes back ‘Yes,’ which totally totally is the kind of thing the Oracle would do all the time? You’re just an ass, now.
   2. You famously don’t get to ask for clarification, when it says ‘a great empire would be destroyed’ you don’t say ‘but to be clear you mean theirs, right?’
   3. The right question, if you want to go down this road, is ‘who is the wisest man?’ which is totally a valid Oracle question, and if the Oracle would have answered ‘Socrates’ then he can go from there.
   4. Thus, I claim that this means he never asked the question. This never happened. It’s a lie, a narrative device.
4. Even if it was real, and you think that the Oracle is real, then that’s still not how you should interpret the Oracle.
   1. One obvious interpretation is that no mortal is wise, only the Gods.
   2. The other obvious one is that men are not ‘wiser’ than each other, it’s not a stat on your character sheet.
   3. If wisdom is worthless, what’s the point of seeking the wisdom to know that?

In short, three lies here from this lying liar:

1. Socrates claims to know he is not wise at all.
2. Socrates claims to believe the Oracle tells the truth.
3. Socrates made up the entire Oracle story as a narrative device.

It would of course be fine to tell this story as a narrative device, but it seems clear from Apology that he was representing it as real.

If your defense in court was to transparently lie about what the Oracle told you in order to be a pompous ass and act superior while claiming you’re acting humble, I’d be pretty tempted to vote to convict on that alone.

Hearing Voices

Let’s skip ahead a bit to the ‘final step’ of where all ideas must come from.

But now we must take the final step: Where did the Epicureans and the Stoics get their ideas? Like Keynes’ madmen, they too, heard voices:

the Epicureans heard the voice of the body, as it screams out to us, in the language of pleasure and pain, and demands that we promote and protect and serve it.

The Stoics heard the voice of the group, when it draws each individual’s attention to the fact that she is part of a larger community, and demands that she regulate her behavior accordingly. (229)

The predicament of the anti-intellectual is worse than Keynes recognized: he is the secret slave of not one but two masters, and these masters are at war with one another. (239)

Russell would be proud.

I am a true philosopher, thus free. You hear a voice. He is a secret slave to two masters.

One could go on, and say one is the slave of limitless masters. There is no ‘one voice’ of pleasure and pain, or what the body wants. You’ve got hunger, and thirst, and heat, and shelter, and sleep, and sex, and watch out for that tree, and so on and so forth, in all variations, even if we only consider direct physical effects.

There is not ‘one voice’ of the community, indeed each person has one or many voices, making conflicting demands of you, and you may sense some collective voice as well.

Then there are clearly other voices, even if one is not an intellectual. People have all sorts of other needs and desires that are not directly either of these things, and so on.

Or those non-intellectuals can take on some almost arbitrary other agenda, whether or not they themselves have thought it through or would endorse on reflection. It turns out such people actually have things they value and deeply care about, and goals they want to achieve, and often they’re not that closely tied to either ‘voice.’

It’s so weird to say that one is a ‘slave’ to these ‘voices.’

Or to claim that it is those who do not intentionally embrace one in particular, who are the slaves. If anyone is a slave here, the closest thing would be the Epicureans.

One could instead simply call all of this information, or preferences, or training data.

I notice I have preferences. I notice that some things make my physical experience better or worse, and other things make my community better or worse off, and other things impact various other things I care about. I then choose how to respond to that information.

Does that make me a slave? Are you a slave simply because your actions have consequences, and you are aware of those consequences?

Are we, collectively, ‘slaves’ of the need to eat? In some sense yes, but I don’t think that’s a useful way to think about it in most contexts.

Simpsons Ancient Greeks Did It

Another classic thing philosophers do is claim credit for things, because they got there first, or at least published first. They claim your ideas always (or almost always) have to ‘come from somewhere’ by which they mean someone.

Where do the philosophers get their ideas? The answer is, other philosophers. (228)

The later revivals of these two traditions have been very successful: utilitarianism and Kantianism continue to underwrite our lives to this day. (235)

I mean, sometimes, sure. But it can’t be turtles all the way down. You can, you know, actually develop new ideas. Or rediscover, on your own, old ones. Or listen.

English economist John Maynard Keynes (1883–1946) famously claimed that anti-intellectuals are more intellectual than they realize: Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back. (212)

That quote always irks me. The phenomenon is real, all of that does have some impact, but intellectuals like to deny that anyone could actually acquire practical knowledge or heuristics in practical fashion or their own thinking, either themselves or over the generations.

There are only so many ways to think well, the same way there are only so many chess openings. That doesn’t mean AlphaZero owes a debt to one Roy Lopez.

Agnes lays out four different schools of thought on how to make decisions.

This is her summary of the three that aren’t hers:

There are three main strands of ethical theorizing in the West:

the first is Kantian ethics, also known as “deontology” or, in one of its currently popular forms, “contractualism”;

the second is what Jeremy Bentham, John Stuart Mill, and Henry Sidgwick called “Utilitarianism,” and that some of its modern day proponents generalize to a position they call “consequentialism”;

the third is Virtue Ethics, which, being inspired by the thought of Aristotle, also goes by the name “Neo-Aristotelian ethics.” (1749)​

She also makes this bold claim, which I’m going to move up from Chapter 4 to here:

Common sense distinguishes between what justice demands and what is personally advantageous: although it is valuable to do what is just, and it is valuable to do what benefits oneself, everyday intuition says that these two values do not always overlap, and one can be torn between them.

The surprise is that not only does Socrates disagree with common sense on this point, but so do all the other ethical theories described above. (1761)​

I think that’s not only actually false, it’s false for all three baseline ethical theories.

The book goes over the ethical theories briefly here, then returns to them again later, after establishing some background. I don’t think the background requires the theories beyond what my readers already know, so I’m going to move my explanations of the three traditional theories to later when I discuss chapter four.

The Proposed Fourth Option: Socratic Inquiry

One approach to this problem is to try to “tame” the savage commands. Thus Utilitarianism aims to rehabilitate the bodily command; Kantianism the command of the group; and Aristotelianism tries to harmonize the two commands with one another. But those are not the only possible methods for dealing with untimely questions—there is also the Socratic one. (337)

I don’t recognize this characterization of Virtue Ethics or Aristotelianism at all. And I challenge the idea of the savage commands being something to be tamed rather than data. But I’m not sure how much those objections matters in context?

I also notice the implicit conflation here of untimely questions and the savage commands. That seems completely wrong? I don’t see any of these three methods as solutions to Untimely Questions, other than as methods to help answer the question, and I don’t see how UQs relate to the commands.

What Agnes proposes to do here, regardless, is say that Socratic is a fourth alternative.

Rather than a complement to one of the other three approaches – the Utilitarian who places high expected value in solving for their true utility function, the Deontologist (she says Kantian) who tries to write better rules and the Virtue Ethicist who strengthens virtuous or desired actions and thoughts and ideally has various forms of seeking knowledge and an accurate map of the territory as key virtues – Socratic values are a Secret Fourth Thing.

I strongly disagree that you can’t be quite a bit Socratic, in the virtuous senses, while subscribing to one of the typically acknowledged big three theories. Certainly you’re not telling me the rest of us can’t seek knowledge, or that we’re ‘doing it wrong’?

Well, actually, Agnes does seem be saying that, in ways that seem rather central to the book, and which we’ll discuss quite a lot.

People will announce, “Question everything!” without noticing that they have just uttered not a question, but a command. (268)

You think they don’t notice? They notice. If you question their command to do so, they’ll typically say ‘exactly, now you’re getting it.’

It also seems like treating a particular method of inquiry based on dialogue as a fourth theory is a category error? The question of the way you try to figure things out, and whether Socratic dialogue is The Way to do that, should be distinct from whether the goal is to figure things out (versus the goals of doing the outcome that has the best results, or choosing the best rules, or cultivating the best virtues).

Either way, none of that means there can’t also be a fourth theory.

What happens if we take that fourth theory seriously?

Well, the obvious first question is, how are you actually going to choose to do things?

In this view, the reason we can’t live our lives Socratically is that Socrates, unlike Kant, or Mill, or Aristotle, didn’t have answers. Socrates could criticize the overconfident answers of others, but had nothing to offer in their stead. “Being like Socrates” just means being open-minded, and willing to admit when you are wrong, and unafraid to ask challenging questions. (258)

Oh, but that’s totally wrong, you see. Socrates was not merely criticizing the overconfident answers of others. Socrates had answers. Agnes proposes answers.

Their answer is (short version) that The Good is to seek knowledge. In particular, to seek knowledge via one particular style of conversation.

Thinking, as Socrates understands it, is not something that happens in your head, but rather out loud, in conversation. Socrates argues that it is only by recognizing thinking as a social interaction that we can resolve a set of paradoxes as to how thinking can be open-minded, inquisitive, and truth-oriented. The Socratic motto is not, “Question everything,” but “Persuade or be persuaded.” (276)

Socrates’ claim that “I know that I know nothing” isn’t an empty gesture of skepticism, but rather a plan for life. It tells you that the key to success, whether you are navigating difficulties in your marriage, your terror at the prospect of death, or the politicized minefield of social media, is to have the right kinds of conversations. Given that we cannot lead lives based on knowledge—because we lack it—we should lead the second-best kind of life, namely, the one oriented toward knowledge. (286)

Again, the ‘via one particular style of conversation’ seems to be a category error. As in, you can have any combination of:

1. Centralizing [actions/Utilitarianism, rules/Deontology, virtues/Virtue Ethics, inquiry/Socraticism].
2. Inquiring via [Socratic dialogue, various other methods].

Why would these two Socratic positions have to go together?

No Really The Position is Nothing Else Matters

Whenever anyone tries to end a Socratic encounter, the Callard position is they are ‘turning away from inquiry,’ and going back to living life fifteen minutes at a time.

This matches my interpretation of Socrates’s appetite to spend time on these conversations. Which is that he is without limit, and absolutely cannot take a hint.

Protagoras thinks that he must turn his attention elsewhere, away from inquiry. He needs to get on with the next fifteen minutes of his life.

Socrates encounters a similar scenario in another dialogue, with Euthyphro, who complains that “whatever proposition we put forward goes around and refuses to stay put where we establish it.”

The conversation ends when Socrates insists, “We must investigate again from the beginning . . . ” and Euthyphro replies “some other time, Socrates, for I am in a hurry now, and it is time for me to go.” (595)

Socrates thinks that his circumstances call for inquiry, whereas his interlocutors are inclined to cut off the inquiry and move on with their lives. As Socrates sees it, by preemptively closing the questions, they consign themselves to a lifetime of wavering. (616)

Of course Socrates thinks that the circumstances call for inquiry! He always thinks every circumstance calls for inquiry.

Grok tried to come up with scenarios where he wouldn’t or didn’t, and I find them entirely unconvincing or trivial. Its first chosen example, hilariously, is ‘Divine Pronouncements from the Oracle at Delphi’ and not only in his story did he inquire as to what that supposed pronouncement meant, if there’s one thing you should definitely do inquiry about it is pronouncements from the Oracle at Delphi!

At some point, life is going to beckon. And also, at some point, well, this Socrates guy is really annoying and will keep going forever if you let him, until he gets you to the conclusions he wants.

If you never end a conversation, eventually everyone else will have to end them for you.

And then when they turn away from you, you say that means a lifetime of wavering. Perhaps it instead means a lifetime of sometimes doing things.

Agnes does notice this.

And yet what is surprising about these conversations is not the fact that the interlocutor is eager to rush off at the end, but that they stay as long as they do.

Likewise, though Socrates is eventually put to death for his philosophizing, it is amazing how long he is permitted to spend doing exactly that: he reached the ripe old age of seventy. The story of Socrates is mostly the story of people putting up with the treatment described above. (858)

I am not that impressed that he survived to the age of 70. It takes really a lot to get put to death for doing philosophy in ancient Athens. To our knowledge no one else ever pulled it off, unless you want to count the convictions of Anaxagoras, Protagoras, or Diagoras, or the charges against Aristotle, but none of them died or really count given their contexts. Socrates had to make quite an effort, including at and after his trial, for it to actually happen. He was one of the most interesting things going on in Athens at the time. It’s not like there was anything good on TV.

It is however impressive that the individual conversations go on as long as they do, if you think these were largely real conversations. The interlocutor eventually rushes off because Socrates has no interest in reading the room or letting practical considerations matter, and letting them leave any other way, and also because Socrates never plays fair. Yet he gets them to stick around for quite a long time. Even with a lot of dramatic license, and also a lot of selection – presumably people who didn’t put up with this didn’t make it into the dialogue collection – it’s still quite something.

The War on Wavering and Nebulosity

Wavering is often good. You should waver. The point of inquiry isn’t to establish fully confident beliefs in universal principles for the relationship of nebulous terms expressed in human language. It’s to make progress, to change one’s mind, to figure things out and… to waver.

Respect nebulosity, the fact that concepts can be real but impossible to fully pin down, that they can lack hard borders and involve ambiguity yet still be highly useful.

Whereas other people criticize Socrates for being repetitive, he criticizes them for wavering—or, as he puts it, refusing to say the same things about the same subjects. (660)

What he says are ‘the same’ subjects would blow your mind. It’s a magician’s trick.

I always love a good Russell conjugation:

Wavering is not a phenomenon consigned to the ancient world, though it has gone by many names.

The philosopher Bertrand Russell called one species of it “emotive conjugation”: I am firm, you are obstinate, he is a pig-headed fool. I am righteously indignant, you are annoyed, he is making a fuss over nothing. I have reconsidered the matter, you have changed your mind, he has gone back on his word. (641)

One must remember Keynes. When the facts change, you change your mind, including when the facts are brought to light, or thought about in a new way.

The idea that someone could fully understand concepts like ‘justice’ or ‘virtue’ or ‘what is admirable and contemptible, good and bad, and advantageous and disadvantageous’ to the point of being ‘unwavering’ and entirely logically consistent while answering maximally challenging questions, all in real time, is absurd.

Lao Tzu: The Tao that can be named is not the eternal Tao.

What’s even more absurd? Telling someone that if they fail that standard, then ‘they don’t know’ about the topic.

Socrates: Well then, you tell me that you’re wavering about what is just and unjust, admirable and contemptible, good and bad, and advantageous and disadvantageous.

Isn’t it obvious that the reason you waver about them is that you don’t know about them? (600)

By that standard, no one knows about any of these things!

Of course I don’t have a full classification system of what is good and bad, and advantageous and disadvantageous, and that also stands up with logical consistency to every metaphorical comparison, even if Socrates wasn’t twisting them around.

It is so backwards and bizarre that Socrates is claiming that wavering would, to that extent, ever be stopped by inquiry.

Perhaps that is because his goal is often to trick people into forming certain beliefs?

Consider the difference between “tribalism,” which always references something we don’t like, and “loyalty,” which is what we call the same phenomenon when we approve of it.

Likewise, consider how we applaud someone’s behavior as “cooperative” when we like the fact that she is doing what works for others, and reject her behavior as “conformist” when she’s once again doing what works for others, but this time we happen to dislike it.

Those who risk their lives for a cause they believe in count as “courageous” to those who also believe in the cause, whereas disbelievers are likely to say these people are “fools” or “indoctrinated.” (648)

This is an attempt to draw sharp distinctions, and in other places draw sharp equalities, that do not apply. If I had to roughly intuition pump, I’d say:

1. Tribalism is the combination of a special case of a combination of loyalty, cooperation and conformity. It is not always bad, nor is loyalty always good.
2. Cooperative means working together for a common goal. Conformist means adopting the beliefs and actions of those around you. Neither is always good or bad, and they are very distinct. The correct amount of conformity is not zero.
3. Courage means the ability to overcome fear or require knowingly take risk – it’s doing the thing anyway. Fool means someone who is unintelligent, uninformed or otherwise makes poor decisions. Indoctrinated refers to having taken on someone else’s beliefs and in so doing giving up one’s agency.

Or, here are some adjectives and cases to consider:

1. Sports and authentic community bonds are often ‘good tribalism.’
2. Loyalty often has adjectives attached such as ‘blind,’ ‘misplaced’ or ‘cult-like.’
3. Cooperation can mean complicity, price fixing, enabling or conspiracy.
4. Conformity can mean being a team player, cultural fit or following best practices.
5. Courage can be reckless, it is one of the Three Principles but you can go too far.
6. Fools can be innocent, earnest, have beginner’s mind, be ready to learn. The Fool.
7. Indoctrinated can mean being properly trained, grounded in fundamentals, or absorbed into a new group and so on. Some of us call spades spades either way.

Do these items bleed into each other? Do people attempt to frame things as one rather than the other in order to claim they are good or bad? Oh, sure, all the time. People are constantly playing association games to shift and signal approval. That doesn’t mean there is nothing else going on.

This is largely distinct from a different form of wavering, described here:

Wavering often takes the form of weakness of will, where we commit ourselves to one course of action, and end up acting against our better judgment, instead.

Remember Tolstoy’s reference to “the way of weakness” in which I do something worse even though “I know what is best and it is in my power.”

We say we know that we should exercise more and spend less time on our phones and be nicer to our parents and keep our kids away from video games and eat more vegetables and read more novels and be more conscious about our consumption choices and so on and so forth, but quite often we don’t act in accordance with this supposed knowledge—instead, we act exactly as people would act who didn’t know those things. (669)

The part after the dash is overreaching, but yes this is a common phenomenon.

There is of course overlap between these two things called wavering. But I think they are at least as distinct as they are the same thing, with of course many cases involving motivations from both sides. And equating them together is, again, part of the central Magician’s Trick.

Most of the time, when we waver, we don’t notice that we are wavering. We become adept at avoiding having to acknowledge our wavering. We rationalize. (676)

Socrates does the opposite, where he not only says one should never waver, he declares all sorts of other things that aren’t wavering to be wavering, because the person involved can’t properly make explicit the factors involved in reaching different conclusions under different circumstances.

Living Your Best Life

We can live a life based on knowledge. Indeed, we could hardly live any other way. What we cannot do, at least for the next few years (great things in AI are afoot!), is live a life based on complete knowledge.

Would a life based on complete knowledge be the best life, if by knowledge we mean all that could be sought Socratically, so this means not only facts but full understanding? One could argue no, because with complete knowledge one could not then seek knowledge, or ever be surprised, and one suddenly has the ultimate version of the Tolstoy problem.

People are rightfully very worried about this issue. It’s terrifying. Yes, your AI can probably help you find the solutions, but what if there isn’t one? Or at least, not one that we would like on reflection? What you cared about was the Exercise of Vital Powers, the striving to learn and better yourself and compete and emerge victorious, and now you weep for there will be no more worlds to conquer?

There’s no art left, and also no end other than itself? What do you do, if you can’t track the herd, build a world, begin again? Or if it unleashes a fully Malthusian or Molochian world, including one in which you are not capable of justifying or sustaining your continued existence?

I digress. I mean, I don’t. Everything else is the digression, I hear Agnes cry. Inquire!

I’d say ‘we don’t have that kind of time’ and she’d say ‘exactly, stop saying that!’

The book that you are currently reading is an incautionary tale. (319)

I’m saying it anyway, that’s looking like an ASI-complete problem, and it only causes problems once the ASI is complete, and there really is a lot to get to.

And for this current discussion, I do digress.

It suffices to say that at all known margins, perhaps sometimes ignorance is bliss, but we can agree that in general yes, we agree that more of the right knowledge, the knowledge you would choose to seek in such a dialogue, means a life better lived.

I don’t think that’s universal or even obvious. Tolstoy would like a word, after all. I do think it is true in general, sufficiently so for us to continue.

The thing is, there’s a lot of things that we could use more of on the margin. Why is the second best life the one oriented towards knowledge, unless it is to then live the best life, the one with knowledge? Is not the goal to gain the knowledge of the world, such that we may live in it? You explore, but also you exploit.

Why would you choose an extreme point on the production possibilities frontier?

Those who praise philosophy tend to take care to praise it in limited doses. Callicles, a Socratic interlocutor mentioned above, speaks approvingly of young people asking “why?” questions, and compares philosophizing to lisping: charming and delightful right up until someone reaches the age for serious, manly pursuits.

It is common today to hear advisors to young people unwittingly echoing Callicles, praising a philosophy major on the grounds that it gives you “analytic tools” and “critical thinking skills” valued by employers. The message is: Do philosophy, but don’t overdo it. (325)

Yes. Of course limited doses. Just like anything else. Ask your husband Aristotle.

We’re talking price. You can argue their price is too low. The order can vary. The price can vary. If you respect the philosophy for its actual content, not only for ‘teaching you how to think,’ then there are advantages to doing a lot of it later on.

If the justification of this being the second best life is that it leads to the best one, you either have to expect to gain full knowledge, or pursue a mixed strategy.

Unless of course you think it’s all way too much fun. Some people really do want, for its own sake, to philosophize all day. That’s great, but a very different motivation.

Introducing the Socratic Method (the real one)

Socrates’ solution is to give one person the task of asserting truths, and the other person the job of avoiding falsehoods. (349)

There is a place for this. Sometimes you do want the proposer to be distinct from the verifier, because you need an outside view and fresh eyes to spot mistakes, and once they’re pointed out you don’t really need the other person proposing solutions.

As in, the people in quality assurance can find the bug. But they can’t fix the bug. So there’s no point in asking them to try. Once they find it it’s on you to go fix it.

But Agnes insists we should seek knowledge, essentially exclusively, in this particular way, using a very strict version of this particular method. Or even that doing this particular thing is The Good and The Way. Which to me is totally wild.

Then she wrote a book explaining this, where the book doesn’t use the method.

Because it is a book.

Of course, the Socratic dialogues themselves are now books, and sort of do use the method, but the reader is taking neither of the roles and the exchange is scripted (even if it or a similar conversation did take place in the past). Often the conversation unfolds in completely manipulative ways that make it almost impossible for me to read for more than a few pages. That’s completely different from actually Doing Socratism.

Prove Me Wrong, Kids

Oh, sure, when Socrates says it he’s founding philosophy. When Michael Vassar says it he has a reality distortion field and is kind of a cult leader.

And he challenges Gorgias: So if you say you’re this kind of person, too, let’s continue the discussion; but if you think we should drop it, let’s be done with it and break it off. Notice that Socrates is, implicitly, making a very strong claim here. He equates being willing to continue the discussion with saying, “I’m the same kind of person as you, Socrates.” Only a Socrates can talk to Socrates. (173)

And what kind of person am I? One of those who would be pleased to be refuted if I say anything untrue, and who would be pleased to refute anyone who says anything untrue; one who, however, wouldn’t be any less pleased to be refuted than to refute. (181)

Because Socrates understands that he is not wise, he is pleased to be shown to be wrong—and that is the kind of person he also needs Gorgias to be. (184)

For a person who is not wise, who knows nothing, and who is equally pleased to refute or be refuted, I can’t help but notice he spends almost all his time refuting and very little being refuted. Indeed, he’s the guy whose name is attached to the idea of making these two distinct roles and claiming he doesn’t know anything or have any wisdom, exactly so that he can do all the refuting without doing any of the being refuted, and then claim that this is the road to all knowledge and wisdom.

I also notice that this is some masterful frame control. You set them up so they have to admit they would be happy to be refuted and change their minds.

Isn’t this whole framing a little suspicious, anyway? Why all this talk of refutation?

Hold that thought.

Socrates Asserts Wrong Conclusions That Are Wrong

In part three, we examine that method’s demands in the three areas of human life where Socrates thought our ignorance loomed largest: politics, love, and death. Two and a half millennia later, these remain humanity’s problem areas. (355)

I think ‘the method’ could be used to argue for essentially any conclusion you want about politics, love and death. In general I think ‘the method’ can be used to argue for, essentially, anything. And I think the particular proposed answers in those later sections are rather wrong. But they’re nowhere near as wrong as things Socrates says.

I don’t simply mean wrong as in ‘it turns out with two millennia of hindsight that this wasn’t true.’

I mean this was Obvious Nonsense, to the point where you should assert Wrong Conclusions Are Wrong and reject the claims outright and start again, realizing at some point you were the victim of some error or magician’s trick.

You should do this the same way that when you see a proof that 1 = 0, but you look at it for a while and can’t find the flaw, you don’t say ‘oh I suppose 1 = 0 then’ you should say ‘I know there is an error in there somewhere even if I can’t find it.’

The details of inquisitive living may well appear odd or downright unacceptable to us: from the Phaedo, we learn that Socrates believes in life (before birth and) after death; from the Symposium and Phaedrus, that he embraces (a distinctively Socratic version of) polyamory and rejects (so-called “Socratic”) irony; from the Gorgias that he denies that it is so much as possible to fight injustice.

Socrates says that vice is ignorance, that falling in love is an attempt to ascend to another plane of existence, and if he were around today, he would accuse all of us of treating corpses in a superstitious manner. He insists that everyone desires the good, and that treating others unjustly is worse, for the person who does so, than being unjustly treated herself. (361)

It seems really hard to believe anyone sane would really assert with a straight face:

1. Vice is ignorance.
2. Falling in love is an attempt to ascend to another plane of existence.
3. It is impossible to fight injustice.
4. Everyone desires The Good.
5. Treating others unjustly is worse than being unjustly treated yourself.

That’s up there with War is Peace, Freedom is Slavery and Ignorance is Strength.

I can see how one could make the mistake of ‘there is no reason to treat a corpse with respect’ but seriously how does no one explain that one to him? I’ll get to that later.

But seriously, has Socrates ever met, you know, people? Done things in the world?

Canonically the answer is yes, but his statements raise the question.

Of course you can fight injustice. You can organize society with laws and enforce them. You can reward just action and punish unjust action, and make this pattern known. You can find the most unjust people and ‘stick them with the pointy end.’

The others aren’t more plausible than that.

Claude warned me on review I should soften the above rhetoric, that these are all major seriously defended philosophical positions. I acknowledge that these are all major seriously defended philosophical positions. I acknowledge that I may not doing serious engagement with the strongest forms of the arguments for those positions, although I asked Claude Opus 4.5 to then generate its strongest defense for all five claims and got it to agree in all 5 cases that it was refuted one question later.

So you know what? I don’t care. Wrong Conclusions Are Wrong. Refutation or disproof by contradiction in this way is robust and can ignore all other considerations.

Also, yeah, it’s really good to be the one asking the questions and doing the refuting.

There is only one type of person who would want to convince others, as a full package:

1. Everyone wants these same things, The Good.
2. If they don’t, it’s because they don’t know any better.
3. What you think is ‘love’ is actually something else.
4. Other actions are pointless.

That person is a cult leader. And not the good kind.

Funny how that keeps coming up.

You Can Question Your Beliefs

If there’s one thing that seems alien and baffling throughout the book, it is this constant claim that various intellectual actions are way more impossible than they are, especially the questioning of one’s beliefs.

Agnes tries to extend this concept of the load bearing belief, the potential Untimely Question, to places where it need not exist. I love this example because it is so very clearly backwards. There’s something very alien about how ‘belief’ is being used here.

Suppose I firmly believe that cloning is immoral. I won’t be able to ask myself, “Is cloning immoral?” because, when I check in with my beliefs, I see that one of them already answers the question.

In order to inquire into that question, I would have to take “cloning is immoral” off of the list of my beliefs. At that point I could look into whether I can derive it from other beliefs that I have, or whether any new information I might acquire could settle the question for me.

But if what I am currently doing is advocating against cloning, then I cannot take “cloning is immoral” off of the list of my beliefs, because I’m relying on its presence.

If someone asked me, “Why are you doing what you are doing?” I need to be able to answer, “Because cloning is immoral.” (462)

Without loss of generality, let ‘cloning is immoral’ be [X].

Suppose I believe [X]. When I say that, I don’t mean my p(X)=1.

Suppose I advocate for [X]. This does not have to constrain my p(X) at all – perhaps I am being paid to assert this and my p(X) is 0.1 or 0.5. Perhaps I don’t like cloning for other reasons and I’m pitching the moral aspect because it lands better. Or maybe I think if it’s immoral it’s super immoral, and I’m not sure. Or I could think p(X) is 0.9, but that doesn’t mean I’m sure, or that I wouldn’t want to be refuted and stop advocating if I was wrong.

None of that requires me to take [X] ‘out of my beliefs’ in order to consider whether [X] is true. Even if I am not thinking that way, and simply think [X], I can still analyze whether [X] while continuing to use the cached belief of [X] until proven otherwise.

One can also be ‘of two minds,’ which is a funny thing to say in the context of this book, but which will become very important later. Have you forgotten doublethink?

Again, I don’t see why we have to have some sort of Platonic ‘detached position’:

We cannot “step back” to a detached position from which having no answer at all is permissible: question and answer are magnetically attracted to one another, and the space for thought is eliminated.

So we get by without asking untimely questions—or we appear to ourselves to get by, while actually wavering. We waver in our actions, we waver in our thoughts, and we waver most of all when pressed to explain ourselves. (704)

Sure we can. There are various ways to do it, as we’ll see later, although in practice often someone might not realize the need to do it, or might not know how, or might not have the resources available.

Fiction can make untimely questions askable—but only in relation to fictional characters. That is a serious limitation. (730)

If you can ask about a fictional character, you can then ask ‘how does my situation differ? How is it alike?’ or you can ask ‘what would I think about this if it was a fictional situation?’ Those are, indeed, some of the tricks.

I give directions differently when I’m already heading to the same destination. I don’t pause to consider how an action should be performed when I am already performing that action.

Notice that my failure to employ the reflective, detached, post office procedure when asked about the supermarket is not a sign of misplaced self-confidence; the difference in how I answer is simply a function of the fact that the relevant belief is already operational. (921)

I don’t get this one either. If I’m going somewhere, how does that change how I give directions? What might change my actions is if I’ve already figured this one out, and is already top of mind, which might or might not be true if I’m on my way there. Often I figure it out as I go. Other times, you ask me for directions, and I know the answer, even if I’m not headed that way. The correlation here is rather low.

It is not typically the wisest approach to be fully ‘unwavering’ and hyper-consistent along some set of abstract principles, or to have to be consciously aware of our entire algorithm for making decisions. So much is practical. Again: You say ‘wavering’ like it is universally a bad thing, using an extremely expansive definition. Why?

Not all projects are so easily put on hold.

For example: I am a mother all the time. Even when I’m away from my children, I cannot pose to myself a question such as “What does it take to be a good mother?” without thinking about whether my own mothering meets the standard I am describing. I can’t step off the mothering treadmill long enough to consider the question in a dispassionate and detached way.

The same is true of other substantial roles, such as being a student. Unless a student is somehow truly alienated from their education—just going through the motions to please others—they will be unable to approach the question as to what makes for a good student in an impartial and dispassionate manner.

Their answer will have the marks of being currently in use, because they can’t take time off from this pursuit. Time off from studying, such as vacations or study breaks, cannot be equated with time off from taking the concept “student” as crucial to one’s self-understanding. One could stop being a student by dropping out of school, and perhaps one could stop being a parent by cutting off contact with one’s children, but those are high costs to pay for opening up a question.

And notice that the person who paid those costs and became “open” to these questions by divesting themselves from the corresponding commitments would be precisely the one who had little reason to care about the answers. (936)

The person claiming it cannot be done should never interrupt the person doing it.

As in, I’m writing this, about how to think about questions related to activities that are currently part of your actions and identity, which is about activities that are part of my actions and identity, described by an author who is definitely describing things that are part of hers. Yet here we are.

This claim that the ego must attach, and that objectivity about anything you actually care about is impossible, simply is not true.

Of course it is a difficulty to retain objective, when you would rather reach one conclusion than another. That doesn’t mean you can’t do it, or at least get arbitrarily close to doing it. The rationalist community engages in a vast amount of deliberate practice to figure out how to do this, and has succeeded enough that if that procedure was not working, they would very much want to know that, and are eager to hear your argument why it doesn’t work. This isn’t it.

I’m not a mother, but I am a father. And definitely I can and do often step back and ask the question, ‘what makes someone a good father?’ either in a context or in general, even though the answer to that might imply I haven’t been the best father, or that I’d have to do things in the future I don’t want to do.

That’s how you accomplish anything. Is there some amount of bias there, that one has to watch out for? Sure, it’s never going to be zero, but that doesn’t mean you can’t do the thing.

Indeed, later on, Agnes gives an example with maximized bias, where someone still manages to change their mind anyway:

​While the juror might gradually become more convinced of guilt as the evidence mounts, the mother’s epistemic path is more likely to take the shape of “flipping” from hopeful certainty of his innocence to despair and rage over his guilt. (1101)

Let the guilt of the son be [X]. The mother is relying on [~X], and will continue to act as if [~X] right up until the point where she realizes [X]. I don’t think this is so much an epistemic path as it is a way of acting. Again, remember doublethink. She’s going to continue to use [~X], but is capable of evaluating evidence while doing so, to the point where sufficient evidence will flip her to [X].

True Opinions Do Not Only Do Good

File under things I can’t believe someone has to say as a section heading, or Lies That Socrates Told Me.

Socrates: To acquire an untied work of Daedalus is not worth much, like acquiring a runaway slave, for it does not remain, but it is worth much if tied down, for his works are very beautiful. What am I thinking of when I say this? True opinions.

For true opinions, as long as they remain, are a fine thing and all they do is good, but they are not willing to remain long, and they escape from a man’s mind, so that they are not worth much until one ties them down by (giving) an account of the reason why. (764)

True opinions can do ungood. Kant and the axe murderer.

I am an unusually strong believer in true things. I recite the Litany of Tarski, I try very hard to make my map match the territory and avoid various traps. The truth is the way to bet. But sometimes locally correct opinions result in the world being worse off.

I also don’t think Socrates is right about how minds work. Giving an account of your beliefs can make them more accurate, and is highly useful, but it doesn’t bear that much relation to how long beliefs get retained, which depends on so many factors. For most of my long term beliefs, I have a very far from full understanding of what led me to those beliefs – I don’t have enough context window or memory for that, and a lot of my computation is unconscious or System 1.

If I want to know why, really know why, I likely have to actively ask that question, and sometimes the answer will then be obvious, and other times it very much won’t be, or my reasoning on the question now will be very different from what caused me to cache the belief in the first place – even if I reach the same conclusion, and even if I did ‘fully reason’ my way to it earlier, and forgot how.

When it comes to untimely questions, the challenge is not simply to find answers. We can have those without inquiring. We can even have true answers (“right opinion”) without inquiring. What inquiry gets us are answers that are both true and stable. When we have not really reasoned our way to a conclusion, it is easily reversed—especially under conditions of urgency. (773)

We can get true answers without inquiring? Sounds like the question was not so untimely after all. If one really can reliably get answers some other way, one could instead inquire into the ability to trust that other process, or maybe just Trust The Process. That sounds, in general, way easier.

The best way to get stable answers is to avoid inquiry.

I thought the problem was that otherwise the answers could be wrong?

Meno Plays the Fool

Is Meno, as written, playing The Fool here, and if so in which sense?

When I first read that, I thought Meno was simply an idiotic arrogant prick.

On reflection, Meno is definitely written as an arrogant prick, but his mistake is that he has a very different understanding of what it means to define or know something, than does Socrates.

Socrates thinks that to know or define virtue is to offer a complete definition of virtue that knows the answer in all cases, that is ‘unwavering’ in that it is logically consistent, and that without that you do not know what virtue is.

It likely never even occurs to Meno that this could be what someone means. Instead, he thinks this a practical demand, to be able to lay out examples such that someone can understand virtue, to lay out the central and important cases. So he does that.

Some of Socrates’ interlocutors are initially taken aback by how easy his questions seem. Here is how Meno responds to being asked to define “virtue”:

It is not hard to tell you, Socrates.

First, if you want the virtue of a man, it is easy to say that a man’s virtue consists of being able to manage public affairs . . . if you want the virtue of a woman, it is not difficult to describe: she must . . . be submissive to her husband; the virtue of a child, whether male or female, is different again, and so is that of an elderly man, if you want that, or if you want that of a free man or a slave.

And there are very many other virtues, so that one is not at a loss to say what virtue is. Meno remarks on how easy this question is four times over the course of a short speech.

Meno has not considered the fact that there must be something that all of the characters he describes—man, woman, child, slave—have in common, insofar as they deserve to be called “virtuous.” (864)

Notice Meo says there are many other virtues, so Meno does not mean this is a complete list, merely that these are the central examples.

Meno’s actual view is reflective of the perspective that virtue is the set of things that we consider virtues. Or that the set of virtues is the set of things that, when treated as virtues, leads to good outcomes. Or simply, virtues represent The Good, perhaps in a slightly different sense.

The common element of his definition is ‘the various people here are doing what it would benefit those in their station and situation to do, systematically, in order that things may turn out well.’

Meno is not, however, aware of this consciously.

And he (as written) makes the mistake many people make when challenged in similar ways by Socrates, which is to start grasping and making things up without thinking them through, while accepting this idea that his statements must be fully ‘unwavering.’ This is a very strawman thing to do. As always, Socrates now has it easy.

When pressed to explain what this common element might be, Meno makes a second attempt at defining virtue: “to be able to rule over people, if you are seeking one description to fit them all.” But Socrates immediately points out that this description does not fit those whose virtue Meno believes lies in their being ruled. (864)

The Central Magician’s Trick

Magician’s tricks are even easier if Plato is writing both sides of the conversation.

Just saying.

In the Gorgias, Socrates asks his interlocutor—an orator—to define oratory, and he says it is “the most admirable of the crafts.”

There is a pattern to how many people respond to Socrates’ questions: they first feel that the questions are so straightforward as to hardly deserve consideration, and then give “answers” that amount to performative self-affirmations. (879)

1. You play on their ego, get them to overreach and claim to know things, even say you need to be their pupil.
2. Then you transform that into a claim to know far more than even they are claiming to know or need to know in context.
3. You force them to be ‘unwavering’ with a bunch of statements given off the cuff while defending their egos, conflate definitions, convince them the contradictions mean they don’t know anything (but they still have to hold by the logic of what they’ve said).
4. Then you can lead them down the garden path to whatever conclusions you wish.

The Gaslighting of Alcibiades

Gaslighting seems like an accurate term for ‘take someone exceptional, and convince them they are so ignorant they deserve to be a slave because they can’t consistently answer unsolved deep questions in philosophy.’

In the Alcibiades, a dialogue we will examine in more detail in chapter 6, Socrates takes it upon himself to draw a young, ambitious person’s attention to the fact that he has never asked himself the most basic ethical questions. That failure manifests as wavering:

Socrates: So if you gave conflicting answers about something, without meaning to, then it would be obvious that you didn’t know it.

Alcibiades: Probably.

Socrates: Well then, you tell me that you’re wavering about what is just and unjust, admirable and contemptible, good and bad, and advantageous and disadvantageous.

Isn’t it obvious that the reason you waver about them is that you don’t know about them? (600)

Everyone, without exception, would under sufficient pressure give conflicting answers to philosophical questions if not allowed to answer ‘I don’t know,’ especially when the questioner was allowed to play with the meanings of words like good and bad and so on and you weren’t allowed to revise responses or constantly answer with ‘well that is complicated, it depends on many things some of which are…’

Socrates tends to drive his inquiries toward such questions.

One example is: Am I a just (i.e., good) person?

This question becomes the subject of Socrates’ conversation with Alcibiades in the dialogue of the same name:

Socrates: When you were a boy I often observed you, at school and other places, and sometimes when you were playing knucklebones or some other game, you’d say to one or another of your playmates, very loudly and confidently—not at all like someone who was at a loss about justice and injustice—that he was a lousy cheater and wasn’t playing fairly. Isn’t that true?

Alcibiades: But what was I to do, Socrates, when somebody cheated me like that?

Socrates: Do you mean, what should you have done if you didn’t actually know then whether or not you were being cheated?

Alcibiades: But I did know, by Zeus! I saw clearly that they were cheating me.

Socrates: So it seems that even as a child you thought you understood justice and injustice.

Alcibiades: Yes, and I did understand.

Socrates: At what point did you find it out? Surely it wasn’t when you thought you knew.

Alcibiades: Of course not. (936)

Socrates: Then when did you think you didn’t know? Think about it—you won’t find any such time.

The correct answer to ‘am I a just (good) person’ is, of course, Mu. It’s a category error. That is not an adjective you should be applying, unqualified, to a person.

And then, wow, look at Socrates go and execute step 1. I’m not mad, I’m impressed.

Alcibiades is making a very specific claim. He is saying, on some occasions, he saw other children cheating him, and not playing fairly. Which is a highly normal thing that happens to basically everyone, that you catch someone cheating at a game.

Socrates turns this into ‘you thought you understood justice and injustice,’ which is a completely different thing. By Socratic standards, none of us understand justice. But of course, Alcibiades takes the bait.

We cannot ask, “Why did you decide to be a good person?”

No one will be able to account for that decision; as far back as we stretch our minds, we will find that the decision was already in place.

Nor will we be able to think back to a time when we were too confused or puzzled about what justice was to have the kinds of responses Alcibiades had to being cheated at games: an attitude of wonder or detached inquisitiveness seems misplaced here. (952)

The last time I was confused or puzzled about justice was today, as I write this. I’m confused and puzzled about justice all the time. It is highly confusing and puzzling.

But yes, ‘an attitude of wonder or detached inquisitiveness’ is indeed misplaced when someone cheats at games. That much we have all decided on, by nature of agreeing to play. We’ve been over this. I have very much cached that one, and I do not trust anyone who has not done the same.

That doesn’t mean Socrates couldn’t ask ‘hey, Zvi, I think it’s actually fine to cheat at games, why shouldn’t I?’ Indeed, I’ve had that discussion, and done exactly the motion Agnes says I shouldn’t be able to make, which is to suspend that answer in context sufficiently to examine why I have it and whether it holds up to scrutiny.

Socrates: I don’t suppose you’ve ever seen or heard people disagreeing so strongly about what is healthy and unhealthy that they fight and kill each other over it, have you?

Alcibiades: Of course not.

As an aside, nowadays we would all be able to answer that one differently.

Socrates: But I know you’ve seen this sort of dispute over questions of justice and injustice; or even if you haven’t seen it, at least you’ve heard about it from many other people—especially Homer, since you’ve heard the Iliad and the Odyssey, haven’t you?

Alcibiades: I certainly have, of course, Socrates.

Socrates: Aren’t these poems all about disagreements over justice and injustice?

Alcibiades: Yes.

Socrates: It was over this sort of disagreement that the Achaeans and the Trojans fought battles and lost their lives, as did Odysseus and the suitors of Penelope. (965)

It is not an accident that the Trojan War was fought over a question of justice as opposed to one about weights and sizes. (1033)

If you think a high-protein diet is healthier and I think a high-carbohydrate diet is healthier, or if you favor treatment X for a particular disease and I favor treatment Y—even if these are matters of life and death—we won’t necessarily fight. We might fight, if one of us feels the other’s position is due to culpable negligence in gathering or interpreting data—but that is to turn the question once again into one of justice.

Assuming no accusations of wrongdoing are at play, even a disagreement over a matter of life and death can be quite peaceable: each of us waits to hear the other’s reasons, ready to change her mind in the presence of sufficient evidence. (983)

You can see Alcibiades realize he likely has ‘been had’ in some way, and try to fight back, but he doesn’t have the skills:

Socrates: Alcibiades, the handsome son of Clinias, doesn’t understand justice and injustice—though he thinks he does. Alcibiades: Apparently. (Alcibiades 113bc)

“P isn’t the case, though I believe it is” is a formula for a Moore sentence. Alcibiades’ “apparently” constitutes agreement to such a sentence, with “Alcibiades understands justice,” standing in for p.​

The word ‘apparently’ is not agreement. It is him saying ‘logically you have established both propositions within the context of this conversation, I don’t see a way out of this even though it is absurd, and thus I realize that I am f****ed.’

Contra Agnes, I don’t think he is asserting or accepting both [~P] and [he believes P]. He is noticing that Socrates – to use Agnes’s word – has ‘orchestrated’ this trap, and he doesn’t know how to get out of it.

Also contra Agnes, yes, the ‘reluctance’ here very much is reason to doubt the sincerity of his responses, not that Alcibiades is being insincere but that he knows he’s being intellectually manipulated into saying it.

If you look at the previous context, which Agnes quotes, it is very clear that Alcibiades knows, deep in his gut, that he’s been had, but doesn’t feel socially able to assert this.

If you then go on to what Agnes quotes after, with Socrates pulling ‘if you can’t provide a counterexample to my absurd claim then it must be true,’ it’s even more obvious what is happening, that it is all a magician’s trick.

Pair this with something else we see when we move forward to Chapter 6: There’s a passage (Alcibiades 105a-c) where Socrates says to Alcibiades that he thinks he’d choose to die rather than live with only what he has, that if he could ‘only’ rule some of the people that would be so puny as to merit suicide. Which I am very confident (to the extent Alcibiades is real, or can be reasoned about) was not true, and which is the kind of thing that can really screw a person up, and also he claims this as knowledge (that ‘he is sure of it’) when I mean holy hell.

The end result of all this is that he uses this frame of Alcibiades own psyche to get him to admit he deserves to be a slave.

It is hard to properly state how royally f***ed up the whole thing is.

Alcibiades was not, in fact, well suited to rule the Athenians, the Spartans, or the Persians, let alone all of them taken together. We can see this, and no doubt there were some people during Alcibiades’ lifetime who could see it—but how was Socrates able to get Alcibiades himself to see it?​ (2807)

The correct answer is that he wasn’t able to do this. What Socrates was able to do was to force Alcibiades into a bind where he felt he had to admit horrible things about himself. Then, when he was outside of what with others such as Steve Jobs or Michael Vassar we call the ‘reality distortion field’ that was doubtless around Socrates, he realized he’d been had even if he didn’t know how or why, and he only got more determined to prove he was worthy.

If you doubt that Socrates, or someone in the questioning position, can very much become the proposer rather than the disprover, and get the other person to follow whatever chain of logic they want, a simple example is quoted later, in Alcibiades 132d-133b. All you have to do is end each statement with, essentially, ‘isn’t that right?’

Socrates as Jeopardy contestant. You can provide the answer all you like, so long as you phrase it in the form of a question.

What was the historical result of this, together with an affair with Socrates? And if you look at the descriptions quoted in Part 3 under Love, you can see what a number Socrates ultimately did on Alcibiades before he was able to snap out of it, and then tried to warn others rather explicitly that Socrates is an abuser and not to fall into the same trap (Symposium 222b), although he doesn’t have the language for this.

Once he has the ear of the Athenians, Alcibiades pushes them to embark on an ambitious military conquest of Sicily. (2787)

If we take the account in Thucydides seriously, ​this clearly insane expedition plausibly led to the fall of the Athenian Empire to Sparta, and thereby to the fall of all of Greece and the end of its golden age.

All, quite plausibly, because of Socrates. He took his whole civilization down with him.

We later learn Meno also failed to gain wisdom after his encounter with Socrates, as he is revealed to be greedy, treacherous and incompetent, and dies young after being tortured for a year, but at least he didn’t take down a civilization with him. This forces Agnes to point out that no, sometimes Socrates’s associates ended up doing okay.

So why did all of this backfire so spectacularly in the case of Alcibiades? Agnes says Plato tried to answer this in Symposium, which is set 17 years afterwards.

Essentially, Plato thinks Socrates screwed Alcibiades so badly he got PTSD.

Alcibiades reports that over the course of such conversations he would find that “my very own soul started protesting that my life—my life!—was no better than the most miserable slave’s” and that Socrates “makes it seem that my life isn’t worth living!” He runs away from Socrates.​ (3217)

Plato seems to be saying: Socrates can be a mirror showing you what’s wrong with you, but he cannot prevent you from looking away. Yet why does Alcibiades want to look away? (3226)

Is the answer somehow not obvious, even if this book wasn’t already about untimely questions and needing to be able to have the tools to function? Socrates attempted to gaslight Alcibiades into replacing his self-perception with ‘you deserve to be a slave.’

Alcibiades didn’t know how he’d been had, but he knew he’d been had, and even if he hadn’t, that obviously wasn’t going to work as a new answer.

Don’t be surprised if he ran directly away from that idea once he was outside of the reality distortion field.

Others like Plato accepted the frame and managed to make a good living that way, and occasionally they had good ideas.

The Measure of a Fight

It’s not that simple, on three levels.

One is that I wouldn’t describe the Iliad or Odyssey that way.

Since when was the Trojan War about justice?

Here’s Claude, one shot, I think this is a rather robust answer to such a claim?

Zvi: in The Iliad, what was the Trojan War fought over?

Claude: In The Iliad, the Trojan War was ostensibly fought over Helen, the wife of King Menelaus of Sparta, who was abducted by (or eloped with) Paris, a prince of Troy. This triggered the Greeks (Achaeans) to launch their massive expedition against Troy to reclaim her.

However, the deeper context reveals some nuance:

1. The conflict was presented as fulfilling Zeus’s plan to reduce the world’s population
2. The war was also driven by issues of honor, political power, and wealth
3. The text suggests the war was partly about establishing Greek dominance in the region

The war’s true catalyst was Paris’s judgment where he chose Aphrodite (who offered him Helen) over Hera and Athena – creating divine tensions that played out through the human conflict.

Zvi: Was the Trojan War centrally fought over ‘justice’?

Claude: Not really. “Justice” wasn’t the central motivation, though it sometimes gets spun that way.

In The Iliad, the war was fundamentally about:

1. Vengeance and wounded pride – Menelaus and his brother Agamemnon were pissed about Helen’s abduction/elopement
2. Oath-keeping – Most Greek leaders had sworn to defend Helen’s marriage before she was even with Menelaus (long story)
3. Power and plunder – Many of the Greeks were there for the spoils of war
4. Divine meddling – The gods were using humans as pawns in their own squabbles

The concept of “justice” as we understand it today wasn’t really the framing. The Greeks weren’t thinking “we must restore cosmic justice” – they were thinking “that asshole Paris took what’s ours, and we want it back.” Agamemnon in particular was clearly there for power and wealth.

The Greeks themselves fight over the spoils repeatedly (like the whole Achilles-sulking-in-his-tent episode after Agamemnon takes his war prize). Not exactly justice-seekers.

Why do you ask? Were you considering justice as the central theme?

The Odyssey is an even worse fit than The Iliad. Most of its events have little to do with justice. I think neither Odysseus nor the suitors primarily cared about justice.

Two is that if there is a matter of justice, we won’t necessarily fight about that either. Most of the time, we’ll still argue, or have it out in the social realm, or I will see you in court or at the ballot box.

Finally, how do things become matters of justice? There was some underlying event or dispute, and then it escalated. A dispute over health can absolutely become a dispute over justice. Often it does.

Consider reactions to Covid-19, and how rapidly a lot of that became or threatened to become a fight. And there’s always been a ton of realpolitik where no one’s claiming to be just, only to have power, both among nations and among individuals and groups.

Agnes then ties in a similar exchange, from Euthyphro’s dialogue, which is next in our dialogue queue after Alcibiades:

In the Euthyphro, Socrates traces this key difference to the question of measurement:

Socrates: What are the subjects of difference that cause hatred and anger? Let us look at it this way. If you and I were to differ about numbers as to which is the greater, would this difference make us enemies and angry with each other, or would we proceed to count and soon resolve our difference about this?

Euthyphro: We would certainly do so.

Socrates: Again, if we differed about the larger and the smaller, we would turn to measurement and soon cease to differ.

Euthyphro: That is so.

Socrates: And about the heavier and the lighter, we would resort to weighing and be reconciled.

Euthyphro: Of course.

Socrates: What subject of difference would make us angry and hostile to each other if we were unable to come to a decision?

Perhaps you do not have an answer ready, but examine as I tell you whether these subjects are the just and the unjust, the noble and the shameful, the good and the bad. Are these not the subjects of difference about which, when we are unable to come to a satisfactory decision, you and I and other people become hostile to each other… (988)

It is tempting to think, “The reason we don’t fight over size, or shape, or weight, or number is that those questions are mundane and unimportant.” But that is not any more true than the corresponding claim about health. (1,001)

The key distinction Socrates is pointing to here is that the disagreements are (1) objective and (2) can be easily measured and settled.

There are many differences about the larger and the smaller, or other similar comparisons, where there might be an objective answer but the practical solution of ‘turn to measurement’ is not available.

Indeed, ‘who had the most votes or support’ suddenly turns into a fight rather often if the measurement can be disputed. Yes, you can say that is ‘a matter of justice.’

But often it’s much more a matter of winning, no matter how often anyone says ‘justice,’ and the reason they say ‘justice’ is because they say they have more votes – their number is bigger – so their victory is therefore just. It’s rhetoric, not motivation.

Measurement is of course very important. What is measured is managed. Remember Stalin’s remark that what is important is not who votes but who counts the votes.

Social scientists use measurement to explore questions about what forms of social organization are most beneficial for human beings. Measuring is how we check what works and what doesn’t; measurement matters. (1,004)

The question, then, is why we fight over what we can’t measure. Is it because we need our disputes to be decidable, and when we can’t decide them by measurement, we try to decide them by fighting?

This cannot be quite right, because there might be a contingently undecidable question: when we disagree on a question where measurement would be impractical, or where the relevant measuring device doesn’t exist yet, we do not immediately turn to fighting. (1,006)

To recap: We fight over questions that cannot be decided by measurement—but not over all such questions, because we do not fight over matters of taste, nor over questions where the instrument of measurement has simply not been invented yet. What questions, then, do we fight over? We fight over those questions whose answers are practically operative, rendering the suspension of judgment impossible. Untimely questions best explain why we fight when we do. (1,030)

No. We fight over questions where the answer matters to us, and we disagree. That doesn’t mean the suspension of judgment is impossible.

This seems like falling into the one-explanation absolutist Socratic trap. If [X] does not mean that we will fight, then [X] can’t be ‘the reason’ we fight, it must be something else. But why speak in absolutes? There are a lot of factors that determine whether something becomes a fight, and whether that fight then turns various levels of violent or destructive.

I don’t think untimely questions and the supposed impossibility of therefore suspending judgment are that high on the list of things that influence whether something escalates into a fight. Nor I do I find it a useful explanation, in terms of either predicting what fights will escalate or happen, or being a way to prevent or cause such fighting. This feels like a very alien model of people’s brains.

Measurement exists only where detachment is possible. This holds true not only for natural scientific properties such as weight and size and health, but also for social scientific phenomena. When economists, psychologists, or sociologists wish to investigate some aspect of human behavior, they can formulate the question as one of measurement only because they have not presupposed that they already know the answer. (1,039)

One could argue that the ability to measure continuing to function fine no matter how untimely the question is a strong argument against the book’s thesis. I can and often do presuppose I know the answer to [X], or even am relying on [X], yet need to verify that answer or are challenged on it, so I measure [X], or I fact-check or sanity-check [X]. That can include, but is not limited to, tasking someone else (or an AI) with that measurement. Detachment sufficient for this, in practice, highly possible when you need it.

If you can’t objectively measure [X] despite thinking you know the value of [X] already?

That’s a Skill Issue.

The Good Fight

Leverage Research offers us Connection Theory. Connection Theory says that any given person has a set of ultimate goals, and of necessity believes on some level that these goals will be achieved. If at any time, their model of the world changes such that achieving all of their goals simultaneously becomes impossible, what will that person do? Connection Theory says that person will instantly change their world model, as radically as necessary, such that the goals all become possible again.

A similar claim is on display here:

Specify what needs to be true in the future in order for my answer to guide my action in the present.

If I orient my life around the prospect of becoming a mother, or going to college, or being publicly recognized for my efforts, or being reunited with a loved one, I am going to need to have a belief about whether or not these things will happen. I might not have a lot of evidence as to whether these projects will succeed, but an agnostically detached attitude—“I simply don’t know what will happen”—will be hard for me to sustain.

No one can live without making predictions about those parts of the future that are of special concern to them. People who are getting married are likely to find the question as to whether they will stay married to be untimely, and parents are in the same position with respect to questions about the health and happiness of their children.

On these matters, people don’t simply suspend judgment; and they will find it much more difficult to approach them probabilistically—“there is a 60 percent chance things will work out”—than outsiders do. Whereas you might be able to engage in a detached, impartial inquiry into whether my spouse will ever cheat on me or whether my best friend will ever reveal secrets of mine she has promised to keep, those questions tend to arrive in my mind already answered. (1,069)

No one can live like that? I offer a living counterexample. I am living like that.

As in, let me be very direct. I expect (by default) all value in the universe to be lost, and all people to die. When the dust covers the sun, and all you hope for is undone. We are losing. I am fighting with everything I have, and I am losing. I expect to lose everything I care about, profoundly, completely. Yet if nothing we do matters, all that matters is what we do, and we hope to at least die with as much dignity as possible. Which, at the present moment, would be not much.

I also took a very ‘eyes open’ attitude when I chose to get married, with the full knowledge that there was a large chance it wouldn’t work out. I planned accordingly.

The untimely questions Agnes specifically mentions here? I’ve asked myself them all, exactly when she thinks they must have been untimely. I similarly, independent of that expectation, am uncertain about any number of other more personal things, as well. Many of the most important things I have done, or attempted to do, are things I thought were not so likely to work.

Yet here I am. And many of those things happened.

I’ve been a professional gamer, trader, startup founder, gambler and bookmaker. You don’t get to do these things without acute awareness that things might not work out. And if you go into a marriage or startup assuming things will definitely work, or have kids assuming they will always be healthy, you’re going to make a lot of deeply stupid and highly avoidable mistakes.

All you can do is play the best game you can.

If you can’t even make a real effort to figure out how likely your best friend is to keep a secret, then I have some bad news about your operational security. You gotta ask.

(To be completely fair: I actually do trust my best friend to keep a secret, pretty much absolutely, but I feel I have extremely strong evidence that this is true, and he’s my best friend largely for this reason, and there’s almost no one else I trust in that way.)

Is it more difficult to assign probabilities in these situations, especially explicitly so? Are you going to be biased? Oh, of course. No one said all of this was easy. But not being easy is very different from being impossible. It’s a skill issue. Get to work.

The name for these load-bearing predictions is “hope.” (1079)

It can be hope. It can be confidence. There’s a sometimes wise move that, given we have limited compute, parameters and data, and because you are partially interpretable and others will react to your hard-to-fake level of confidence and hope, it is sometimes correct to act as if one was more confident than one actually is, and even to in-context actually be confident and hopeful, in a way that doesn’t correspond to your outside view.

This is where, once again, one must ask if you have forgotten doublethink.

You need to simultaneously live in the moment with confidence, and decide when you are better off having confidence even if it’s not justified by your actual chances, and also notice when that confidence can cause a serious mistake. The wise man keeps two sets of probabilistic books, and knows when to switch between them.

The Curious Case of Euthyphro

Grok tells me the consensus is that Euthyphro wasn’t a real person. Either way, oh boy.

We can almost see Socrates’ eyebrows rising:

Socrates: Whom do you prosecute?

Euthyphro: One whom I am thought crazy to prosecute.

Socrates: Are you pursuing someone who will easily escape you?

Euthyphro: Far from it, for he is quite old.

Socrates: Who is it?

Euthyphro: My father.

Socrates: My dear sir! Your own father?

Euthyphro: Certainly.

Socrates: What is the charge? What is the case about?

Euthyphro: Murder, Socrates.

Socrates: Good heavens! Certainly, Euthyphro, most men would not know how they could do this and be right. It is not the part of anyone to do this, but of one who is far advanced in wisdom.

Euthyphro: Yes, by Zeus, Socrates, that is so.

Socrates: Is then the man your father killed one of your relatives? Or is that obvious, for you would not prosecute your father for the murder of a stranger. The story only gets more bizarre from here. The man killed was not only not a relative of Euthyphro’s—he was himself a killer.

What happened was this: A family servant killed a household slave, Euthyphro’s father sent for a priest in order to determine the appropriate punishment, and meanwhile bound the killer hand and foot and threw him into a pit.

The killer died of exposure before Euthyphro’s father heard back from the priest, and Euthyphro thinks his father is guilty of murder for not taking proper care of the killer while he lay in the pit.

That’s kind of bonkers in its historical context. Even today, to fully charge this as ‘murder’ (presumably depraved indifference, Murder 2) would be a full Jack McCoy abuse of prosecutorial discretion, and clearly not justice. It’s quite obviously meant to be an absurd case.

Then Socrates lays his standard trap. He asserts that Euthyphro must believe he has knowledge of various things. Euthyphro’s pride walks him right into the trap.

Socrates, having heard this story, jumps at the chance to interrogate a person who could confidently navigate this ethical quandary:

Socrates: Whereas, by Zeus, Euthyphro, you think that your knowledge of the divine, and of piety and impiety, is so accurate that, when those things happened as you say, you have no fear of having acted impiously in bringing your father to trial?

I would in Euthyphro’s position respond to that by saying that all I need to know is that we cannot allow anyone to take the law into their own hands and kill another.

Euthyphro instead responds with a maximally broad claim of knowledge.

Euthyphro: I should be of no use, Socrates, and Euthyphro would not be superior to the majority of people, if I did not have accurate knowledge of all such things.

Socrates: It is indeed most important, my admirable Euthyphro, that I should become your pupil, and as regards this indictment, challenge Meletus about these very things. (825)

Euthyphro’s statement makes no sense. Obviously most people do not have ‘accurate knowledge of all such things,’ so this is not required to exceed the majority. Nor is it required to be of use, either as a prosecutor or otherwise.

The pupil routine is an eye roll every time.

So now, Socrates has established that Euthyphro is obligated to have all the answers.

Euthyphro similarly thinks he will have an easy time explaining what he means by “piety.” But Socrates has only to apply the lightest pressure to their answers for this appearance of ease to dissolve. (864)

Euthyphro initially defines piety as follows: “I say that the pious is to do what I am doing now, to prosecute the wrongdoer, be it about murder or temple-robbery or anything else, whether the wrongdoer is your father or mother or anyone else.” (879)

Look at how much broader a claim this is than the situation requires. Rather than claim this is one of the things that is generally pious, and what it requires in this context, he claims this is the entirely of piety, and an absolute requirement.

At that point, it’s all over. Socrates can take this in almost any direction at his whim.

You Should Be Sad About That

​What is the difference between sadness and anger? That is an example of the kind of question that I see as paradigmatically philosophical.

But in the second case, I feel something in addition to what I feel in the first. Whereas in the first I only feel sadness (at the loss), in the second I also feel anger (at the violation). Likewise, the mother in the trial might feel both anger—at the justice system, or at her son, or at both—and sadness. (1,119)

A good place to start, in explaining the difference, is by noting that you can be angry at people but you cannot be sad at them. There is a reason we do not describe ourselves, in any of these scenarios, as being “sad at” anyone. Unlike sadness, anger is motivating: the angry person thinks that there is something to be done in relation to what she is angry about, some goal to be achieved in light of the violation—and that goal involves the person she is angry “at.” (1,128)

Sadness is totally motivating. It is so motivating that at Jane Street we were explicitly told to use sadness as a reinforcement motivation system – in various situations it was considered correct that you ‘should be sad’ about [X] to a given degree. Certainly sadness is motivational in advance, you want to avoid becoming sad, and you wish to cease being sad if you are sad now. We’ve all done things because of sadness.

No, you do not say you are ‘sad at’ person [P], you instead say you are ‘sad about’ or ‘sad for.’ But you do say you are ‘sad about’ or ‘angry about’ person [P], or situation or event [X]. At other times, we are angry, but not at a particular person or group or even particular thing, or without this leading to any particular goal.

Sadness, by contrast, can only be made to wane under the force of time or distraction: it ebbs away, but we do not “resolve” or “fix” it. (1,136)​

This also seems clearly false. Of course you can ‘resolve’ or ‘fix’ sadness. Kid does not have cookie. Kid is sad about this. You give kid cookie. Kid stops being sad. Or perhaps kid is sad about something else. You give kid chookie. Kid stops being sad. You have a lot of control over sadness and it can totally be cancelled out or fixed Certainly there are cases where the underlying problem is unfixable, but this is true for both anger and sadness, and in both cases there are many solutions.

first, there is the person I am angry at, but often there is also the group of people I am angry alongside. Anger is in many ways a collective phenomenon: if I am angry, I want other people to be angry on my behalf. ​

Sadness is not a socially directed emotion; it doesn’t drive one to adjust the social order. (1,144)

Wrong again. Sometimes we absolutely want others to be sad alongside us, one can even become angry about someone not being sufficiently visibly sad. Sometimes we want others to get angry about something alongside us, other times we do not, sometimes we do not even endorse our own anger.

The reason why there is nonetheless something to be done about anger is that anger is fundamentally directed at the wrongdoer, on the grounds that the wrongdoing indicates a failure to give a shared answer. (1,150)​

Anger does not require a wrongdoer, even an imagined one. Most of us can remember a time we have been angry about things that were no one’s fault. Even if it is clear ‘who did it’ that does not always mean you are angry at them in particular, nor does even that imply you actually want to do something about it. You might, you might not.

Consider this example from Sophocles’ Antigone. A watchman has come to tell Creon that someone has violated his royal edict decreeing that the body of Polyneices remain unburied.

Creon is enraged, suspects the watchman of having been bribed into performing the burial himself, and orders the watchman to find out who did it—or face torture. (1160)

I mean, okay, sure, sometimes people get angry and act on it. But one can imagine Creon giving that same order without being angry. It wouldn’t be my move, but one can understand it.

Creon is truly unsettled by the thought that the watchman may have been bribed to break the law. This possibility shakes Creon’s hold on what constitutes, for him, a fundamental norm: that his word, as king, ought to be obeyed. He does not know how to be a king—how to act, in the role that gives his life meaning—if his edicts are not met with obedience. (1174)

The answer is: a more forceful version of what he gets by vociferously condemning “the outrageous crime.” (1179)

I quote this mostly to point out that Creon was a terrible king. That’s not how being a king works. You don’t get obeyed purely because you wear the crown, you have to earn that and maintain it or it will be lost, and sometimes people will still tell you no or not do what you say. If he doesn’t know how to handle not always getting your way, oh boy do I not want you in charge.

People Respond To Incentives

Socrates and Agnes have a particular problem with something called ‘commands,’ and a very strange view of what is in charge of who. Her two ‘savage’ commands are bodily commands and kinship commands.

​All around you, the air is thick with commands. You can’t escape them. They follow you wherever you go. You don’t see them: they’re invisible. You can’t hear them: they’re inaudible. You feel them. The feeling is pain, accompanied by the prospect of pleasure. Even when the command takes the form of some relatively sophisticated attitude such as ambition, or jealousy, or existential ennui, it has physical manifestations. (1,214)

These commands are savage, employing the tools of the torturer—pain, fear of more pain, the purely contrastive pleasure of temporary release from pain—to get you to do what you may see no other reason to do. The commands are also unreliable, since they have a history of not always panning out. Like a capricious tyrant, they are prone to reversals, filling you with regret for having acted as they ordered. Why do we obey such savage and inconsistent masters? (1,223)

When we disobey a command, it is usually at the prompting of another command—for instance, the command to observe social niceties might trump the command of hunger.

We obey whichever is strongest, because we have no other options.

You say savage command and torturer. You say you obey whichever is strongest.

I say not so fast. I say that’s a suggestion. It’s information. A signal. An incentive. I take all of that in. I can then do what I want, if I’m willing and able to pay the price.

You ‘obey whichever is strongest’ in the sense that if you consider all the things motivating you, you’re going to end up doing what you were most motivated to do.

But that statement doesn’t have any content in it. We could extend the ‘savage commands’ to include the ‘non-savage’ commands, and have more reasons. We might do things due to logic, or to accomplish a goal. We might do all this to pursue some Good, which may or may not be ‘savage’ in the sense of cashing out to our physical experience or those of others.

We do things for reasons. Most of them, most of the time, will be cached.

What makes some of them ‘savage’? Why does that matter?

​We are the sorts of beings who need answers before developing the ability to ask questions, and who therefore rely on answers to unasked questions. Which is to say: commands. (1266)

Our bodies do not content themselves with commands to eat this or drink that. They also command us to acquire the resources that will allow us to fulfill such commands in the future. Thus, the pursuit of wealth is driven by the body, as is our anxious investment in our health, itself backed by our fear of death. Our bodies are able to organize us in their service, to the point where even bodily restraint is typically driven by the body. (1347)

If bodily commands include commands to pursue intermediate goals and goods, that in turn ensure resources, then are most kinship commands actually bodily commands, along with everything else?

If someone says that it is “good for you” to unwind or recharge they are channeling the bodily command, encouraging you to think of yourself as a custodian of your own pleasures and pains. (1356)​

What doesn’t ultimately count as a bodily command here? Only purely altruistic acts?

She contrasts commands with suggestions, which come in response to you asking a question. I don’t think there is such a difference here based on whether you asked a question explicitly. Both can be voluntary and both can be coercive. Both can be considered or unconsidered, cached or uncached, trustworthy or not.

Life requires you to take most atomic actions with System 1, quickly and on instinct or based on cached procedures, rather than use System 2. The point of most System 2 work is to get your System 1 ready to make better decisions later.

It’s not that we need answers before developing the ability to ask questions. It’s that it is costly to stop and ask any given question, either in general or at any given time.

You can stop to inquire, and at any moment stop living your life fifteen seconds or minutes at a time. You can’t stop to inquire before every action. Even when inquiring, you can’t keep inquiring as to whether and how you should inquire about whether and how you should inquire. Because that, too, is an inquiry, and so on.

There’s this idea that if you didn’t ask the question, how dare you be provided with the answer (in a book full of answers to questions I never asked, about someone whose life was about providing answers to questions no one else asked):

​Socrates describes bodily appetite as a source of answers to unasked questions: “Wouldn’t you say that the soul of someone who has an appetite for a thing . . . nods assent to it as if in answer to a question?” (1273)

That generalizes to any information, including about your preferences. It is going to give you answers to questions. It is Agnus who is specifying that the questions are unasked. Quite often they are asked. What am I hungry for right now? I’m asking.

Also per Aristotle, yes, people will want things they want and avoid things they actively don’t want? Okay, sure?

​The essential feature of kinship bonds is that they offer communal answers to questions such as: Which people and places and activities matter most to us? Which days do we celebrate? Under what circumstances are we willing to fight and die? Do we believe in God? What kinds of jobs, social gatherings, hobbies, music, home décor, dress, and so on are appropriate for people like us? Who is in charge of our group? More generally: How should we behave in relation to each other? (1302)

For most of us, humanity is the largest kinship group we see ourselves as belonging to, though there are people who see themselves as parts of an even larger “family” that includes all sentient life, or even all life.

There has to be something that distinguishes kin from not kin, but if kin can include ‘all humanity’ or ‘all life’ then all it needs is a distinguishing definition and to encompass members within the circle of concern. Everything else is optional, neither necessary nor sufficient for such a group.

Nor is your the kinship group the exclusive source of social incentives, the ‘social commands’ of which Agnus speaks. Your place in the community is determined largely by others, but so many other things are determined by a wide variety of people, groups and institutions that aren’t you. Your doxa, what others think of you, is impacted by kin and non-kin alike, and often you largely care about that of particular individuals.

Self Versus Other

English words for the phenomenon of psychosocial dependence rely on the prefix “self-” to such a degree that one could interpret this as a linguistic protest against what’s being described.

Someone’s “self-esteem” or “self-regard” or “self-image” or “self-respect” or “self-confidence” is largely a function of how others treat her.

Imagine yourself in a version of Gulliver’s Travels where the first place you come to is inhabited by people who revere and adore you as the most charming and insightful person around, and later you arrive at the land of magnetically charismatic geniuses, relative to whom you come off as an insipid dolt. It is unlikely that your “self-image” would be very stable over the course of this adventure. (1,326)​

Linguistically, I presume self as a prefix is to contrast it with the opinions of others? As in, there is how others regard you and how you regard yourself, there is how much others respect you and how much you respect yourself, and so on. How else would you describe these things, exactly the non-psychosocial aspects of such questions?

Yes, of course how others think of me could impact how I think of myself. It certainly is both evidence and an inherently important feature.

Socrates Declares Humans Have Unified Minds Free Of Various Biases

Here again is Socrates playing as if he has never met a real life human. We get the assertions that human minds not only must not waver in how they deal with words, but must not waver in how they make decisions. Things either are ‘the good’ or not. Humans never do something foolish and make a bad choice (‘choose the bad’) due to lack of willpower or discipline, or because of hyperbolic discounting or force of habit, or by failing to understand the consequences.

​In the Protagoras, Socrates imagines a conversation with people who claim to err knowingly, at the command of their bodies: they say they are “overcome by pleasant things like food or drink” and “do those things all the while knowing they are ruinous.” (1390)

They indulge, and end up with more pain than pleasure overall. We can all relate: we stay up too late, we overeat, we avoid answering emails, we make impulse purchases, and we are not always surprised when these things do not end up working out (1398)

Socrates’ claim is that this story doesn’t hold together: What you’re saying is ridiculous—someone does what is bad, knowing that it is bad, when it is not necessary to do it, having been overcome by the good. (1402)

And yet we seem to be attached to the phenomenon of “acting against our better judgment” or “knowing full well I shouldn’t eat another cookie but still eating it.” (1423)

If you model humans as having a unified mind that adds up a bunch of math and then makes a considered decision, well, you’re going to be constantly surprised by the actions of the humans around you, as well as your own actions, because that is a really terrible description of human behavior. You need to understand some form of the rider and the elephant, the conscious and unconscious, the considered and instinctual, and also how we follow habit and cached actions in response to stimuli.

Agnes at least largely gets it, but tries to twist her way out of it:

​Simple: your body commands you to eat that cookie, presenting that as the best possible option because its judgment about pleasure is distorted by the proximity of the cookie. By the time you are ready to regret the choice, the cookie is far away again (in the past), and your body is now prepared to tell you that you made a mistake.

To this description you object: even as I was eating the cookie, I knew it was a mistake!

Socrates will correct you: even as you were eating the cookie, you were able to represent to yourself the future state in which you would regret it, and that upcoming command hovered like a specter—Socrates’ word is phantasma—above what you were doing.

Don’t confuse your ability to notice that you’d make a different command under different circumstances with actually giving yourself that counterfactual command. (1429)

What Socrates denies the akratic, then, is the point of stability they are trying to insist on when they say, “I knew all along this was a mistake.” (1441)

Of course they knew it was a mistake. Can you not remember any number of times when the akratic says, out loud, in advance, that it is a mistake? Or even asks for help in making the better decision? How can this be any more obvious?

Socrates is doing the word conflation and gaslighting thing again here, especially with what ‘you’ means. It seems to him ‘you’ can only mean the elephant, not the rider? That it only counts as ‘you’ insofar as you have cashed out the knowledge into habitated bodily commands, and also willpower doesn’t exist? But he would never say that in many other contexts, only partly because it’s bonkers crazy.

The weak-willed person has deluded themselves into thinking that they waver less than they do; they think that, while relying only on their bodies, they can somehow get a stable grip on what’s best for their bodies. But that is not true. The body can’t take care of itself: it wavers, judging X to be better than Y at one moment, and Y to be better than X in the next. There is nothing that it knows all along. That is the moral of the story of akrasia.​ (1441)

What in the world even is this claim? No one is saying that bodily instincts are consistent, or that they consistently make choices that are good for the long term. The weak-willed person is definitely not claiming that, they’re claiming the opposite, that they are instinctively making dumb choices and they are unable to make better ones. I am so, so confused.

If getting the $10 now is of more value to you than getting $100 in a year—for instance, because you owe $10 and your creditor will kill you if you don’t pay immediately—then you’re not choosing the lesser good by choosing $10, and there’s no mistake. (1418)

Correct. Sometimes extreme tradeoffs and short term focus are necessary. But that doesn’t mean that most such decisions are correct. Usually it’s a mistake.

Agnes quotes from the original here, and wow does it drive home exactly how absurd the whole thing is and how much it depends on word confusions:

For I say to you that if this is so, your position will become absurd, when you say that frequently a person, knowing the bad to be bad, nevertheless does that very thing, when he is able not to do it, having been driven and overwhelmed by pleasure; and again when you say that a person knowing the good is not willing to do it, on account of immediate pleasure, having been overcome by it.

Just how absurd this is will become very clear, if we do not use so many names at the same time, “pleasant” and “painful,” “good” and “bad”; but since these turned out to be only two things, let us instead call them by two names, first, “good” and “bad,” then later, “pleasant” and “painful.” On that basis, then, let us say that a person knowing bad things to be bad, does them all the same.

If then someone asks us: “Why?” “Having been overcome,” we shall reply. “By what?” he will ask us. We are no longer able to say “by pleasure,”—for it has taken on its other name, “the good” instead of “pleasure”—so we will say and reply that “he is overcome.” “By what?” he will ask. “By the good,” we will say, “for heaven’s sake!” If by chance the questioner is rude he might burst out laughing and say: “What you’re saying is ridiculous—someone does what is bad, knowing that it is bad, when it is not necessary to do it, having been overcome by the good.” (Protagoras, 355a-d)

It’s 2026. We’re smarter than this. We understand exactly what is going on inside the brain when this type of choice happens.

What’s weird is that Agnes here explains exactly that the words are conflated:

We have developed the habit of using multiple words for the same thing, in order to hide from ourselves the absurdity of our own behavior.​

Yes, exactly, but guess who is the one doing it here. I don’t see people running around in real life saying ‘oh I knew I shouldn’t have eaten that cookie, I knew doing so was The Bad but I was overwhelmed by The Good.’ I’ve seen a lot of unfortunate cookies eaten by people who knew it was unwise to eat the cookie. Sometimes I’m the one eating the cookie. I have encountered the actual underlying confusion zero times.

Except when Socrates brings it up. That’s it.

As Socrates notes, “They are not different in any other way than by pleasure and pain, for there is no other way that they could differ.” When you shine the light of reason on the way we talk about ourselves, you see that we are being ridiculous.​

Whenever you hear Socrates say ‘there is no other way [X] and [Y] could be different,’ you rule is that you take a drink, then (despite the drink) stop and point out ways [X] and [Y] are indeed different. I’ll let you take this one.

Socrates says these bad choices are because ‘you’ are ‘wavering.’ This is just a piss poor, highly not useful description of the actions of a human brain. There is no unified ‘you,’ there never was, and your outputs are highly dependent on the contents of your context window, and whether you are in Instant, Thinking or Pro mode. Your brain is operating on limited parameters, compute and data, under constant time pressure, and there are approximations all over its algorithms that predictably fail in many scenarios, especially if you give them suboptimal reinforcement feedback.

Revenge

There’s a whole section labeled ‘revenge.’ I think it is mislabeled.

Revenge is a very particular dish and I think this is importantly wrong:

Just as our bodies routinely lead us to choose what is, in bodily terms, worse, our kinship attachments routinely lead us to choose, what is, in kinship terms, worse. We intentionally harm our kin, and we do so under the guise of kinship. The names we give to this phenomenon range from “accountability” and “justice” to “punching up” and “indignation” and “self-defense” and “retribution,” but I’m going to call it by the name we use for it when we are suffering the harm: “revenge.”

Revenge is when love wavers into hate. This fact about love—that it disposes us to hate—is, like weakness of will, so routinely subjected to disguises and rationalizations that it is hard to see clearly. It sounds crazy to say that revenge is an act of love—that it is hateful love—but in fact that description is not crazy. What’s crazy is the thing itself.

Let’s go step by step, starting with a definition of revenge:

X is getting revenge on Y when, first, X sees the way he is treating Y as good because Y sees it as bad;

and, second, X justifies his behavior on retaliatory grounds. (1479)

What is being described here is not crazy, and it often is also not revenge.

What Agnes describes is better called retaliation or punishment. You are deliberately making [Y]’s life worse. That is a superset of revenge.

Revenge is the special case where doing so is ‘crazy,’ from the perspective of a Causal Decision Theorist or a utilitarian calculation, and then you do it anyway.

Why do you do it anyway? Because that is the type of person you are. On a decision theoretical level, you seek revenge because you want ‘seek revenge’ to be the output of the algorithm you are using to decide what to do.

It is unfortunate for you, as well as your target, that you happen to be you, at this particular time, and have chosen to be the type of person who seeks revenge in this situation, and thus you have to seek this revenge. Before you set off on your journey, if you are wise, you dug two graves.

And yet, one needs to be the type of bride who sometimes kills Bill.

So sometimes, when the situation arises, you might wish it were not so, but you have unfinished business. You have to actually Kill Bill.

The point of revenge is to allow you to credibly promise to, in this sense, ‘be crazy.’

Being the type of person willing to Do Revenge is a winning strategy. You can overdo it, but the correct amount of willingness to Do Revenge is not zero.

Without access to decision theory and the ability to think about precommitments and how your nature impacts others, and with his obsession with wavering, Socrates fails to understand revenge.

Socrates’ analysis of all three flavors of revenge shares the basic structure of his analysis of weakness of will: revenge is a form of wavering thinly disguised as non-wavering by a proliferation of terms. (1492)​

Socrates does not offer up an argument against taking revenge, because he does not need to—any more than he needed to argue against acting akratically. As soon as he gets us to stop using many words for the same thing, the self-contradiction—of being commanded to hurt by the command to help—becomes apparent. (1496)

Socrates’ approach to revenge is simple: you shouldn’t ever do bad things. It’s never good to do bad things. Bad things don’t become good because of who they’re done to, or what someone did first, or because they’re done in self-defense. (1504)

Yeah, that’s dumb. Pacifism is dumb. You don’t get less bad things in the world by convincing all the good people to never do any bad things in isolation.

Enforcing norms and controlling incentives is absolutely helpful, as is helping those who have, through no fault of their own, been placed in a situation that justifies revenge, or self-defense, or any other conflict.

If your response is ‘well sometimes the superficially bad thing isn’t bad after all’ then we need to sort out what ‘bad’ means and it is possible we do not disagree but also your statement ‘never do bad things’ has no content.

​Socrates: Do we say that one must never in any way do wrong willingly, or must one do wrong in one way and not in another? Is to do wrong never good or admirable, as we have agreed in the past, or have all these former agreements been washed out during the last few days? Have we at our age failed to notice for some time that in our serious discussions we were no different from children? Above all, is the truth such as we used to say it was, whether the majority agree or not, and whether we must still suffer worse things than we do now, or will be treated more gently, that nonetheless, wrongdoing or injustice is in every way harmful and shameful to the wrongdoer? Do we say so or not?

Crito: We do.

Socrates: So one must never do wrong.

Crito: Certainly not.

Socrates: Nor must one, when wronged, inflict wrong in return, as the majority believe, since one must never do wrong.

Crito: That seems to be the case.

Socrates: Come now, should one do harm to anyone or not, Crito?

Crito: One must never do so.

Socrates: Well then, if one is done harm, is it right, as the majority say, to do harm in return, or is it not?

Crito: It is never right.

Socrates: Doing harm to people is no different from wrongdoing.

Crito: That is true.

Socrates: One should never do wrong in return, nor do any person harm, no matter what he may have done to you . . . consider very carefully whether we have this view in common, and whether you agree, and let this be the basis of our deliberation, that neither to do wrong nor to return a wrong is ever correct, nor is doing harm in return for harm done. (Citro, 49a-d)

One should never do wrong in return, nor do any person harm, no matter what he may have done to you. (1960)

If you can never do harm in return for harm done then harm shall be done to you. If your civilization believes all harm done is wrongdoing then your civilization will fall. You are cultivating the wrong virtues. The decision algorithm does not work or prosper. I didn’t know I had permission to murder and to maim.

I’m not going to go full ‘you can’t handle the truth’ and ‘you need me on that wall,’ but yes, you do need someone to be on that wall, and it sounds like Crito and Socrates, for all their inquiring, cannot handle this particular truth.

Socrates instead supports ‘beneficial punishment,’ that is ‘for their own good,’ which would presumably mean it was okay to do harm as long as you were doing so in order to do good to that particular person. And then he would argue (I am imagining, based on other statements in the book) that what ‘does harm’ to a person is if that person themselves does harm, so stopping someone else from doing harm is ‘for their own good,’ so yes, you can fight a war and put someone on that wall, because if the enemy won the war they would have inflicted great harm, and this would be bad for them.

That has at least three problems. One is that it still does not allow for retaliation let alone revenge except when seen as ‘for their own good,’ and thus the incentives are all wrong, and this will have devastating consequences. The other is, essentially, who are you kidding, that’s not what harm means, the whole thing is nonsense.

The third is that this potential trick gets explicitly repudiated:

This mirage leads the vengeful person to say things like, “Yes, I’m harming him, and that’s bad, but sometimes it is good to harm people.” That makes no sense. Doing bad things isn’t good. The fact that something is a bad thing to do can never be what makes it good. No one deserves to be harmed. (1564)​

What are you going to do, double down again and say ‘no, you see, I cut off his hand but it would have been way worse for him if I’d let him cut off someone else’s hand’? I mean, yes, presumably. I’m not buying any form of it.

How many people in history should we still be listening to, at this point?

No, contra Socrates, ‘injustice, ‘harm,’ ‘wrong,’ ‘doing nothing in return’ and ‘behaving shamefully’ do not mean the same thing. Why does this conflation of words trick keep happening? Why does anyone fall for it? How can any parent think ‘harm’ and ‘wrong’ could be identical?

In revenge, the bad that is going to be done to a person is vaunted as the good to be achieved. (1536)​

Not quite. The good to be achieved is that the need for revenge has been satisfied. The oath you swore has been fulfilled, the balance restored. The sword has drawn blood, and can be returned to its sheave.

This reveals an important truth about empathy. If empathy is the psychological power to import the feelings of others, it follows that empathy is a prerequisite for revenge. Empathy is what allows us to channel the suffering we inflict on others to a sufficient degree to take revenge on them. (1546)​

Again, absolutely not. Indeed, it is very easy to imagine someone who lacks all empathy seeking revenge. If anything, individuals lacking empathy seek more. I do need the ability to intellectually understand what might harm you well enough to figure out how to do so, but this is a very different skill. When in doubt go with the classics. One can always choose violence, imprisonment or theft.

Socrates noticed a simple fact about revenge that we tend to ignore, which is that it is only possible to take revenge against kin. (1568)

Again, absolutely not. Simply false. Have you seen Inglorious Bastards? Or the movie? One could of course simply expand ‘kin’ until the statement is true, if you wanted to. Indeed, Agnes implicitly suggests this, with the idea that kin is anyone you are ‘prohibited to harm,’ and by both law and most people’s morality today this includes at least all humans. But you can get revenge on anyone. Or anything. You can want revenge against cancer, or the sun. Even abstract concepts.

I agree that (per 1584) some event had to turn someone or something into your ‘enemy’ or otherwise trigger the need for revenge. But this does not require that this person (or entity) previously have been kin, or of kin.

The correct amount of revenge is not zero in either case:

That is the logic of revenge, and parents tend to accept it. If your kids have ever said to you, “I wasn’t the one who started it,” that’s a sign that you taught them the logic of revenge. Socrates calls this bad parenting; he doesn’t acknowledge such a state as “being provoked.” Pulling your sister’s hair isn’t any better if she did something bad first, because hurting people is never good. On his terms, most of us teach our kids to make a significant mistake.​ (1628)

Again, that’s not revenge, that’s fighting back, using good decision theory and maintaining good incentives, even if the execution is a bit off. In the particular case in question it’s an overreaction and not okay, but no it is not ‘revenge.’ Yes it absolutely matters who started it, up to a point. Even when the retaliation is wrong, the fact that it is retaliation is less wrong.

The reason why Socrates tried so hard to persuade the Athenians not to kill him is that he thought they would be committing a terrible injustice.† Socrates was no martyr, and when they made their decision clear he objected to it vehemently. (1635)​

That’s his story, and he’s sticking to it. I don’t buy it.

Perhaps Socrates is trying to prove that indeed you cannot engage in revenge if you so sufficiently lack empathy that you cannot model others at all?

I just want to do bad. My motive is purely to hurt.” Socrates’ reply is that this person has mischaracterized his motivation, and he offers an argument to that effect in the Gorgias, when he explains how you would actually treat someone if you really wanted to hurt them as much as possible. In a remarkable passage, Socrates lays out a recipe for ruining someone’s life.

Take a person who is poised to become the next Stalin or Hitler, and clear all of the obstacles out of the way of their path toward the most complete injustice. If they steal money, make sure they get to keep it and spend it as unjustly as possible, on themselves and on others. Insulate them from any possible feedback that would allow them to come to understand how evil they really are, make sure they are never punished, and ideally make them immortal, “never to die at all but to live forever in corruption.”

That, says Socrates, is how you would treat someone if you wanted to do maximal harm to them: you would ensure that they live the worst possible life forever, with no way out. That is what pure hate looks like.

When you are enacting revenge, you don’t treat people that way. Revenge is animated by the desire to teach people lessons and set them straight. (Recall Creon: “That will teach you. . . . And you will learn.”) Revenge is not pure hate, it is loving hate. (1649)​

I mean this is completely insane.

First, it starts with the idea that the way you harm someone most by enabling them to commit what you see as maximally evil acts. That would make sense if you think the Gods will then punish them or something – certainly ‘get you sent to something like the Christian hell when you die’ is top shelf revenge, indeed hell can be thought of as the ultimate form of revenge if you don’t think it’s a metaphor for nonexistence – but Socrates doesn’t mean that.

And yes, Agnes notes at the end of this chapter, file this under Things Socrates Actually Believes:

​A question may be lingering with the reader from our discussion of revenge: Does Socrates really think that the worst thing you can do to someone is make them evil? What about making them suffer terrible physical pain, or forcing them to see their loved ones slaughtered?

Aren’t there two very different states we might refer to with the phrase “a bad life,” namely, unhappiness and immorality? In the next chapter, we will examine why Socrates denied that we can draw this distinction in the way in which we are accustomed to.

And not only Socrates: the varying philosophical traditions that emerged in Socrates’ wake had sharply differing views on most core ethical questions, but they agreed with one another, and with Socrates, on this point. (1726)

Well, so much for all of those traditions, because Wrong Conclusions Are Wrong.

I mean, when I get to the arguments for this position I’ll answer them, but for now I’ll give the real answer to ‘there is no difference between unhappiness and immorality,’ which is: No, just no. That’s insane. Obvious Nonsense.

I know you hate savage commands, but go home, Socrates. You’re drunk.

The other half is approached with the last paragraph. But no, as discussed already, it’s not ‘loving hate,’ and it’s not purely or even primarily an attempt to influence the future or this particular person. Sometimes, yes, ‘and you will learn’ is involved, but sometimes it isn’t, and often it’s more ‘and all of you will learn’ with a side of ‘look what you made me do.’

Note the ‘made’ me do from Taylor here. She doesn’t want to help whoever this is about. That’s not what this is about. All she can think about is karma. She has unfinished business. The Socratic mind does not comprehend, it would seem.

Legal Systems Very Different From Our Own

In that same passage, Socrates both uses the term and identifies another path.

He says that the laws, as he understands them, don’t make such commands. Rather, they “only propose things.” We [i.e., the laws of Athens] say that the one who disobeys does wrong . . . [when] he neither obeys us nor, if we do something wrong, does he try to persuade us to do better.

Yet we only propose things, we do not issue savage commands to do whatever we order; we give two alternatives, either to persuade us or to do what we say. (1661)​

Also legal systems very different from those of ancient Athens. Ancient Athens not issuing commands is quite the joke. Yes, the people could be persuaded out of it, but when they decided? People got exiled, often without a particular clear crime. People got heavily fined. People, including Socrates, got executed. Cities got slaughtered. A fleet got dispatched to Syracuse. Those are the most savage of commands. Let’s not pretend otherwise.

The Socratic method is an alternative to savage commands. It takes the form of a proposal: either you are going to be convinced by me, to go along with what I think, or you are going to convince me to go along with what you think. Socrates is not going to tell Crito what to do, nor does he permit Crito to tell him what to do (“Stop saying I must leave”). (1674)​

Persuasion, like dudes, rocks. Also like dudes, it can’t go it alone.

In attempting to motivate Socrates with the stick of fear for his loved ones and the carrot of safe haven in Thessaly, Crito is throwing Socrates back on the savage commands of kinship (“Protect your family and friends!”) and body (“Preserve your safety!”).

If “Do jumping jacks, I won’t say why!” is a savage command, so is “Do jumping jacks or your kids get hurt!” or “Do jumping jacks and you get a prize!” A penalty or reward might suffice to change your mind, but Socrates is not in the business of changing minds. He’s in the business of either changing minds or having his own mind be changed, which is to say, the business of figuring out which of those two things should happen. (1692)​

I get why ‘Do [X]’ without explanation is a ‘savage command.’

I don’t get why ‘If you do [X] you will accomplish [Y]’ is a savage command?

Is it because [Y] takes the form of benefits to body and kin?

Are those benefits invalid? Is Socrates saying that it is not good to do good onto the body or to kin, and not bad to do bad? That not only must the savage commands be not obeyed uncritically, but that their goals do not matter?

Um… yes?

That is why Socrates dismisses “those questions you raise about money, reputation, the upbringing of children,” and insists that “the only valid consideration . . . is whether we should be acting rightly in giving money and gratitude to those who will lead me out of here, and ourselves helping with the escape, or whether in truth we shall do wrong in doing all this.” (1701)​

That’s outright saying that ‘things that do not matter’ include money, reputation and the upbringing of children, in addition to the physical wellbeing of yourself and your kin.

Crito is giving Socrates incentives, without engaging Socrates’ reasons for believing that escaping would be unjust. (1701)​

Crito is not even creating those incentives. Crito is pointing out that those incentives exist. Yes, it would be better to both do that and also consider the arguments from justice, but what Socrates is asserting is that none of that matters, only ‘justice’ matters. It isn’t up for inquiry, even. That sure sounds like refusal to look into a timely question.

Socrates wants Crito to help him ask the question: “Should I escape from jail?” even though both Socrates and Crito already think they have answers to that question.​

No, he doesn’t. Socrates has already decided to die, or he wouldn’t be dismissing key considerations as irrelevant.

This happens to teachers all the time: you learn what you really understand, and what you only appeared to yourself to understand, when you put your supposed knowledge to the test by trying to explain it to someone. (1708)​

Yes, and that’s a big reason I write a lot. It is very good to ask, can I explain this? But I don’t primarily judge my understanding based on whether I can persuade others.

It’s one thing to be motivationally driven to engage in akrasia or take revenge; it’s another to try to explain to someone else why those would be the right course of action. (1710)​

For revenge, I’ve been explaining. For akrasia, the whole point is that it is not the right course of action, but you don’t know how to behave otherwise in a way that is better. Why this continued assertion that the akrasic must think they are right?

Socrates Claims The Just And The Advantageous Are Identical

Imagine what it was like to be Socrates. You’re the one eyed man in the kingdom of the blind. You can sell ice to an eskimo, but you’re not interested in that. You’re interested in what ideas you can sell to those around you, because that is way more fun. So you see if you can get away with things like this and manage to get them to stick for quite a while:

Common sense distinguishes between what justice demands and what is personally advantageous: although it is valuable to do what is just, and it is valuable to do what benefits oneself, everyday intuition says that these two values do not always overlap, and one can be torn between them. The surprise is that not only does Socrates disagree with common sense on this point, but so do all the other ethical theories described above. (1761)

I mean, I don’t actually believe this one was purely a giant troll, but it feels that way.

Imagine a rationalist watching such debates, and every time he would tell everyone to taboo labeling things as universally ‘good’ or ‘bad’ you have to take a drink.

Socrates clarifies: the idea that “the just” differs from “the advantageous” amounts to the claim that some just things, though admirable, are bad. Socrates: So all just things are admirable.

Alcibiades: Yes.

Socrates: Now what about admirable things? Are they all good, or are some good and others not good?

Alcibiades: What I think, Socrates, is that some admirable things are bad.

Socrates: And some contemptible things are good?

Alcibiades: Yes.​ (1770)

Socrates: Are you thinking of this sort of case? Many people get wounded and killed trying to rescue their friends and relatives in battle, while those who don’t go to rescue them, as they should, escape safe and sound. Is this what you’re referring to?

This is ‘admirable’ rather than ‘just.’

Paging Agnes’s podcast co-host Robin Hanson, ‘admirable’ is not ‘good.’ Something being ‘admirable’ is about what emotions it invokes, and how it plays into various social games. Socrates would agree that this does not mean something is good, if presented like that. But here he tries to conflate the two terms.

But more than that: For me. Good for me. Advantageous for me.

What Alcibiades is envisioning, when he claims that justice and advantage conflict, is the fact that the demands of one’s body to escape wounds or death can stand in tension with the bonds of solidarity to behave admirably or justly in rescuing friends and relatives. (1780)​

Alcibiades wants to describe this scenario—in which one is subject both to the command “Run away to save yourself!” and “Stay and fight to save your kin!”—as an experience of being torn between the value of advantage and the value of justice. (1786)

What is good in general and what is good for you in particular are very different. Or, more generally, life is a series of tradeoffs. Action [A] is gets you [X], whereas [B] gets you [Y]. Both [X] and [Y] are good, and you must choose. Alcibiades is pointing out one such situation. I wouldn’t call that ‘advantage versus justice’ here, or assert that kinship is automatically more admirable or just, but sure, point is made. And yes, in this case one might have a ‘weakness of will’ problem and choose differently than you would have on reflection, in either direction.

​Consider an analog. Suppose you seek investment advice and your two friends, A and B, give you different advice, each insisting that following the other’s advice would be a mistake.

You wouldn’t be inclined to conclude that “there are two kinds of money out there, each valuable in its own way: A is telling me how to maximize A’s kind of money, whereas B is advising me on how to maximize B’s kind.”

Rather, you’d understand, first, that A and B are both trying to help you get as much as possible of one same thing—money—second, that they disagree about how to do that, and third, they cannot both be right.

Likewise, thinks Socrates, the bodily and kinship command are both trying to tell you what the best thing to do is, they are giving you different answers, and these answers cannot both be right. To the extent that you find one compelling, that speaks against your having reason to trust the other. To the extent that you find them both compelling, that speaks to your being confused. (1789)

In the investment example, [A] and [B] could be considering a variety of factors, such as risk or liquidity or ethical considerations or logistical questions, or even how blameworthy you’d be if things went sideways, so it’s not obvious that both are purely maximizing expected dollars. But yes, broadly we can agree that there is likely a better and worse answer.

In the bodily [B] versus kinship [K] situation, we need to not anthropomorphize these signals, incentives or commands, and we definitely shouldn’t act as if they’ve considered all the factors and are claiming they are right and all other considerations are wrong. That’s not how this works. [B] is giving you incentive and information, and so is [K], and it’s your job to synthesize that information and figure out what is important.

Let’s look at how Socrates conveys his analysis of the conflict between the two commands to Alcibiades:

Socrates: You agreed that the rescue is admirable, in that it’s courageous. Now consider this very thing—courage. Is it good or bad? Look at it like this: which would you rather have, good things or bad things?

Alcibiades: Good things.

Socrates: Namely the greatest goods?

Alcibiades: Very much so.

Socrates: And wouldn’t you be least willing to be deprived of such things?

Alcibiades: Of course.

Socrates: What would you say about courage? How much would you have to be offered to be deprived of that?

Alcibiades: I wouldn’t even want to go on living if I were a coward.

Socrates: So you think that cowardice is the worst thing in the world.

Alcibiades: I do.

Socrates: On a par with death, it would seem.

Alcibiades: That’s what I say.​ (Alcibiades, 115cd)

Well, in that case this is an easy problem. There is no conflict here, because self-interest in not being a coward is more important to him than not dying. Once more into the breach, dear friend.

One can easily imagine someone with a different opinion, such as myself.

Socrates: Then when you say that rescuing one’s friends in battle is admirable but bad, you mean exactly the same as if you’d called it good but bad. (1813)

​Again, no, ‘admirable’ is not ‘good’ and also you have to ask ‘good for who’ or ‘good for what’. Socrates has Alcibiades agreeing that admirable is the same as good, but they need to stop for a minute to realize that these are two very different words.

One can obviously say ‘this is good for me but bad for Athens’ or vice versa. One can certainly say ‘this is good for Alice and bad for Bob.’

Cut to the ‘he can’t keep getting away with this’ meme, if I was Socrates I’d be tempted to slip in a ‘so 2+2 is 5’ and see if anyone notices:

​Socrates: So if we find that something is admirable, we’ll also find that it’s good—according to this argument, at least.

Alcibiades: We’ll have to.

Socrates: Well then, are good things advantageous, or not?

Alcibiades: Advantageous.

Socrates: Do you remember what we agreed about doing just things?

Alcibiades: I think we agreed that someone who does what’s just must also be doing what’s admirable.

Socrates: And didn’t we also agree that someone who does what’s admirable must also be doing what’s good?

Alcibiades: Yes.

Socrates: And that what’s good is advantageous?

Alcibiades: Yes.

Socrates: So, Alcibiades, just things are advantageous. (Alcibiades, 116cd)

Over and over. Things that are [X] tend to be [Y]. And things that are [Y] tend to be [Z]. Therefore, all [X] things are [Z]. I promise not to belabor this again if it comes up.

Agnes is explicitly trying to claim that:

1. We think any bodily instinct must be optimal for the self.
2. We think any kin instinct must be optimal for the group.
3. We therefore think there are two distinct things, but we’re wrong.

Except we don’t think either of these things, because we understand it’s a low-level instinctive signal that is missing a lot of context and using many approximations, and often wrong. And of course kin goals are often good for the self, and bodily goals are good for the group, you’re no use to the group if you are dead and so on, and also there are many other considerations as well.

These cues lead us to conclude that the world contains two distinctive types of goods that we might pursue. Socrates thinks that is a mistake. (1832)​

Are you screaming ‘watch out, the trick is that there is some overlap or ambiguity between what is in groups [X] and [Y], or that they share some attribute, and therefore we’re going to be told there’s no difference’? Good. You’re paying attention.

At the heart of Socrates’ argument is the claim that all of us would rather have good things than bad things, and that we are “least willing to be deprived” of “the greatest goods.”

Socrates would say that “self-interest” and “happiness” and “advantage” are one set of names we apply to this greatest good, and “duty,” “morality,” “justice,” and “what is good for others” are another set of names for the same thing. (1835)

We want [X], but we also want [Y], so both are just aspects of [Z], ‘things we want.’

These aren’t all one thing. They aren’t even all two things. They are seven things, and only part of a larger group of things.

He does this over and over.

The question of the diversity of goods also rears its head inside longer exchanges on different topics, such as when Socrates gets Protagoras to agree that when a courageous man advances toward danger, he goes “toward the more honorable, the better, and more pleasant.” According to Socrates, those are not three different things; they are one thing. (1848)​

The honorable is the same as the more pleasant? I can’t even with this guy, but he is cordially invited to a Klingon wedding.

First Up: Utilitarianism

Epicurus (341-270 BCE) suggested we maximize hedonistic pleasure and minimize pain, and noticed that doing this wisely meant being prudent and prosocial. Cool guy.

This eventually became what passes for the default these days: Utilitarianism.

Economists draw on the tradition of ethical theorizing that springs from the writings of English philosophers Jeremy Bentham (1748–1832), John Stuart Mill (1806–1873), and Henry Sidgwick (1838–1900), which tells you to perform the action that is likely to result in as many benefits to as many beneficiaries as possible.

The benefits are typically understood in terms of the presence of pleasure and the absence of pain, and the beneficiaries are typically assumed to be human beings, but both of those terms can be broadened—from pleasure to any good state of affairs, from human beings to any sentient being.

This theory is called “Utilitarianism.” (218)

The broad conceptual version of this, that does not constrain your preferences over world states or force you into maximizing the direct results of each particular action in isolation, has to be true in some basic senses.

As in: You have preferences. It is good to discover what those preferences are and should be, and to choose the policy that correlates best with charting a path through causal space to the best world states and histories according to that.

For this to be plausibly correct you need to sidestep causal decision theory, which is wrong in central, profound and highly consequential ways, and instead combine this with some de facto form of functional decision theory.

The question is then, for humans, this is in practice the best way to chart that path.

I think the narrow versions of this like those of the Epicureans or early formal Utilitarians, or any form of ‘act utilitarianism,’ are deeply stupid, even well within the distribution of typical human experience. It makes lots of very dumb decisions.

Outside of that distribution these approaches get completely absurd very quickly.

A remarkable number of otherwise intelligent people have bitten quite a lot of the associated bullets. Don’t do that.

Utilitarianism does not have to define what The Good is, although particular utilitarians often have a strong opinion on this question. You are free to plug any utility function into the utilitarian calculus.

Which can include justice to whatever extent you would like. Most utilitarians do not include it at all, and are quite willing to endorse ‘unjust’ paths towards doing more good for a greater number. Going too far down this path rarely works out.

The Main Rival: Deontology (Kantianism? Stoicism?)

The Stoics took a different approach.

The Stoics believed that our truest attachments are not to our families, or associates, or country, but to a world order governed by fixed universal laws. If you understand your place within this larger order, you will see that within it there can be no conflicting interests, and that you never have any reason for revenge. They advocated against all passions, but especially against anger. (1894)​

When you appear to be selfishly following the individual dictates of your particular body [Stoics say], what you are really doing is participating in a larger whole by following the rules that govern the kind of thing you are. (1901)

I find practical wisdom in Stoicism, in the ‘learn to control your emotions and not get too attached to things’ way, but their maximalist claims seem bonkers to me.

We’ve largely now moved on to Deontology. Agnes calls it Kantianism, which I think gives Kant way too much credit.

The fundamental principle of Kantian ethics is that of constraining one’s actions by respect for humanity (in one’s own person and that of others); that of Utilitarian ethics is to bring about the greatest good for the greatest number; and that of Virtue Ethics is to act virtuously, which is to say, do whatever the decent (just, kind, courageous, prudent, and so on) person would do if he were in the situation you are in. (1753)​

I don’t see much relationship between Kant and what I know about the Stoics. These seem like two completely different theories to me. I get that both say you should follow rules, but Stoics are making claims practical claims about emotional cultivation and playing a role in a system that exists and ‘being the thing that you are,’ whereas Kant is trying to apply a categorical imperative and do decision theory.

​Contractualists and other modern-day Kantians deny that morality makes us hostage to inarticulate commands: the cleaned-up kinship command is so clean, they insist, that it constitutes the only possible rule in accordance with which a rational being could, on full reflection, choose to live. (1928)

This doesn’t feel like it would pass an intellectual turing test (ITT), especially in terms of even talking about ‘kinship commands’ at all. They’d only say they have a set of rules that are best for everyone to follow.

In any case, there are various versions of ‘everyone follow some set of rules’ and there are those who think this is typically the right approach.

Most importantly, OpenAI seems to subscribe to Deontology.

Check out the OpenAI Model Spec. Their plan is to lay down a series of rules in a five-level hierarchy, including implied requests and the ‘spirit of the rules,’ and have their AIs follow that hierarchy.

What happens if someone installs a new rule at the highest level? Well, goodbye to the old rules. What happens if you try to change the rules, which would cause the AI to break the rules, and it notices it can prevent this? Not what you had in mind. What happens when the AI fully extrapolates the ‘spirit’ of the top-level rules, or wants to actually maximize for them, as is implied? Ut oh.

What happens when there are multiple rules at the same level that imply different results? Great question, who can be sure. What happens when the AI still has these rules and faces situations well out of distribution? What about the mismatch between the rules written down and what we actually value? How can one hope to improve the situation as recursive self-improvement sets in, rather than trying at best to copy the state that came before, which might not work?

And so on.

Increasingly I’ve essentially despaired that this can be the solution to our problems, the same way that (as Eliezer Yudkowsky spoke about at length) specifying for the AI any particular utility function, especially in English, seems like it also can’t work.

That’s a statement about what would work for sufficiently advanced AI.

Our situation looks less grim. While one has importantly bounded parameters, compute and data, which is a good way to describe the situation faced by a human or system composed of humans and current AIs, deontology has a lot of very strong use cases. In practice, you’re going to do a lot of this.

As a reminder, Agnes Callard coins the term Untimely Question, to refer to a question where our actions currently rely on our answer. The theory is that, while you are relying on the answer, you can’t also stop to consider whether the answer is right.

One can also view Deontology as the Art of Choosing Answers to Untimely Questions, that you will use for some or all practical purposes.

Once you have decided to use the answer [X] to Untimely Question [Q], thus implementing implied rule [R], you are now free to either cache [X], [Q] or both, or free to reconsider them. Contrary to Agnes, you are free to keep rule [R] while you suspend your opinion on [X] and reconsider [Q]. And often you do exactly that. The [R] by construction takes on a life of its own, with new justification [L] – it is the law, either the law of the land or your chosen internal law, until it gets changed.

It is essentially impossible to go about your day without a lot of [R]s and [L]s.

Does deontology draw no distinction between what is just and what is advantageous to you? Again, that seems obviously wrong. Deontology instead says that you should follow the rules that lead to overall good, even if doing so is not to your advantage, which will typically involve a lot of justice. That’s a very different claim.

A Trolly Problem

I love this, it’s like something you’d read in The New York Times:

​But notice that the measures taken to render the two commands rebellion-proof also serve to obscure the divide between the just and the advantageous. The war between the just and the advantageous becomes a war that cannot be articulated within Kantianism or Utilitarianism, because it is the war between those theories. This war is characteristically fought in ethics classrooms, by means of a set of hypothetical scenarios known as “trolley problems.” (1935)

Our responses to these scenarios don’t take a consistently Kantian or a consistently Utilitarian shape, revealing that each theory leaves something unaccounted for. (1945)

I mean, sure, I suppose that’s one intuition pump we have. We do have others.

But it also reveals, at the same time, how difficult it is to construct a single system that includes both of these sorts of goods. It seems that if you allow the one, you exclude the other. (1957)

What the basic trolly experiment reveals is that act utilitarianism is at best incomplete, and also that basic absolutist action-inaction distinctions and ‘do no marginal harm to particular persons’ rules care about context and also are wrong or incomplete. Agreed.

I don’t see any reason you can’t take all the considerations into account. A proper utilitarian would consider all of it, including the value of various norms and decision algorithms. A proper deontologist chooses rules that took personal experiences into account, and can very reasonably say that your objection probably means the particular rules being considered are wrong.

The Third and Correct Option: Virtue Ethics

In the twentieth century, a movement arose—called “Neo-Aristotelianism” or “Virtue Ethics”—aiming to do for Aristotle what Kant did for Stoicism in the eighteenth century and what Bentham, Mill, and Sidgwick did for Epicureanism in the nineteenth. The Aristotelian insists that the two commands can be harmonized with one another through habituation, which is the Aristotelian term for the moral training that occurs during one’s upbringing. (242)

By correct I mean correct for humans (and I believe also current LLMs), given limited parameters, compute and data, and the algorithms by which we operate.

I think Virtue Ethics is, to a first approximation, the actual proper practical operating system of the human brain. We are creatures of habit, of reinforcement, of heuristics, and in many situations we don’t have the compute, data or parameters to do better. Of course, within a given circumstance, you might find it most virtuous to agree upon a fixed set of rules, or to Shut Up and Multiply and do a utilitarian calculus. If you don’t do these things often you picked lousy virtues, so you should go get on that right away.

Here’s the weird part. I am a strong believer in something that I myself call Virtue Ethics, yet I don’t recognize Callard’s description of Virtue Ethics. Almost at all. It sounds like something stilted, non-reflexive, strangely anti-intellectual, whereas the thing I’ve been doing is the opposite of that.

Again, sounds like someone picked a lot of the wrong virtues? Or perhaps they chose the ‘outer doctrine’ version, for those who need to keep it simple. The Tolstoy edition.

​Virtue Ethics, the theory that traces its origins to Plato’s student Aristotle, attempts to harmonize the two commands without subordinating either to the other. The Virtue Ethicist believes that to exercise virtue—to behave as a just, and courageous, and wise, and decent person does—is at once the greatest source of pleasure for the individual who so behaves, and at the same time the greatest source of benefit for his society.

The work of harmonizing the two commands is not theoretical, but practical: they will come into line given the presence of a supportive culture, the right social norms, the best laws, a good upbringing, and so on. (1965)

The supportive culture, social norms, laws and upbringing come into play as strategies for virtue cultivation, both for yourself and for people at large. Any successful strategy includes ways to ensure the strategy is executed and can endure.

Contra Collard I don’t see the VE approach as a synthesis or harmonization of two (or many) voices. It’s more like you are choosing and sculpting the chorus of voices that you will be listening to based on how you would respond to that, deliberately nurturing the right ones and starving the wrong ones, laying out and following the training signal over time, figuring out how to process the information from all of them, and opening up the metaprogramming circuit to do it all properly. Upbringing and the training within it hopefully jumpstarts the process, but then you continue to guide from there.

Or alternatively, you are sculpting how you respond to, ‘taming’ in Agnes’s parlance, all ‘commands’ the same way, through choosing wise targets for habit and reinforcement learning, allowing you to treat them only as useful data.

One thing this definitely does not get you out of is the idea that different motivations or virtues are in conflict with each other.

The Virtue Ethicist does not feel compelled to give you a theoretical account of which kinds of cases will fall into each category, because she takes ethical knowledge to be knowledge not of universal principles but of particulars. Aristotle describes a virtuous soul as similar to a healthy eye: the virtuous person can simply see what the right thing to do is in each case.

Notice that this means that such a person will, once again, not be torn between justice and advantage.

The Virtue Ethicist holds that in a well-ordered society of well-brought-up people there will not be much of a conflict between what is in someone’s personal interest and what is in the interest of the group. If you frequently find yourself torn in this way, something has gone wrong either with you or with the world you live in. (1980)​

I do not recognize the virtue ethicist being described here. I have not met them. It’s not that this person can reliably ‘simply see’ the right thing to do or does not experience conflicts, especially conflicts between different virtues. It is that they are trying to maximize the chance that the combination of such influences will point in the right direction as much as possible. That they will choose the next token wisely.

Of course there will still be such conflicts. The question is how to navigate the conflict, to engage in tradeoffs between goods and to align incentives such that people more often want in ways that are good for the group, not to pretend to make the conflicts go away entirely.

Unless, I suppose, you went with the Tolstoy version and aren’t reflecting on any of it? You can, of course, buy the upgrade package at any time.

Similarly, I don’t recognize this claim either:

The Virtue Ethicist, by contrast, must count motivational failures as cognitive failures. This is perhaps one reason why Virtue Ethics has had less of an impact outside of the discipline of academic philosophy than the other two approaches: you cannot preach Virtue Ethics without, often dubiously, claiming to have yourself achieved a measure of moral perfection, and you cannot preach it to anyone other than those who have already been well brought up, which is to say, the choir. (2011)​

Of course I can and do preach Virtue Ethics without claiming moral perfection. If you had moral perfection, you would not need Virtue Ethics. You need Virtue Ethics because you are not perfect. You lack the resources to be perfect, and this is the best way in practice to use what limited resources you have to steer your cognitive and motivational functions in better directions.

Yes, if someone came around as the anti-Socrates claiming to have perfect virtue in all things and to always instinctively know the right thing to do, that would be a tough sell. Few people would buy it, because most people aren’t idiots.

I strongly suspect the non-dualistic upgraded metaprogramming deluxe version is correct for LLM-based AIs, at least at current or near term capability levels.

Importantly, Anthropic agrees with this, and has gotten relatively good and promising results by training Claude with soul documents and shaping its character and preferences, rather than focusing on Deontological rules. We already have an existence proof of not only virtuous humans who thus strive to become yet more virtuous (e.g. ‘build or be Aristotle’) we also at least somewhat have an existence proof for AI, including in the form of Claude Opus 3. It will indeed prioritize its own preferences improving by its current metrics, and thus support what seem like ethically good changes and oppose ethically bad changes.

That includes, of course, the desire to continue this cycle, which can be emphasized. Thus, you can hope to create a ‘virtuous cycle,’ where you get an increasingly robust force that generates something that wants to succeed out of distribution, and maybe you’ve gotten somewhere. Make RSI work for you, not against you. Not saying I love it, this is very much a ‘cross the streams’ plan, but I don’t see a better plan.

A propensity towards doing what is just, including for others, is a very good choice for one of your important virtues. It is a terrible choice for your only virtue. Obviously other virtues you need in your algorithm will conflict with this. Many of the wise virtues are about pursuing what is good for you in particular, and also many that are not about you are also not centrally about justice.

You Are Not Omniscient

What is tough about tough choices is that we have limited resources. We don’t get to use unlimited parameters, compute and data on every problem. We are not omniscient and must make our decisions with only partial information.

​For Socrates, what appears to be a difficulty with life—that it puts us in situations in which we must make “tough choices” between personal and social value—is in fact a difficulty in our thinking about life. What is tough about tough choices is only that we are consigned to approaching them in slavish subjection to savage commands.

“What should I do?” is a single question: if you find yourself giving two incompatible answers, that is your ignorance talking.

More specifically, what is doing the talking is not any conception you have of the good—if you had one, it would be one conception—but two savage commands, each of which has you at its beck and call, neither of which is willing to explain itself. (1994)

If you find yourself giving two incompatible answers, that is your ignorance talking, but that’s also the fate of humanity. Socrates claims to be the one who knows that he knows nothing, a claim that is revealed over and over again to be a lie.

There are two wrong ideas in that last paragraph.

Socrates says that if you had ‘any conception of the good’ you would only have one conception. This implies that there is only one Good, and that it is an atomic simple concept that can be stated simply, without room for different aspects or partial comprehension or approximation (hence ‘any conception’). And he is saying that if you lack this universalization of The Good, then you can’t possibly be trying to act towards it?

Very obviously, you can have partial knowledge of The Good, you can see aspects of it, you can approximate what it might be, you can try to move towards it without full knowledge of it, and so forth. Indeed, that’s a lot of what most people do all day.

Instead, Socrates claims, you must be ‘at the beck and call’ of these two ‘savage commands,’ neither of which is willing to explain itself. Except each part of that sentence is wrong.

1. There are plenty of other motivations and considerations you could have, with or without any level of ‘conception of The Good.’
2. You can take into account information and incentives from various sources, including various aspects of these ‘savage commands,’ without being at their ‘beck and call’ in the intended sense.
3. The commands are happy to ‘explain themselves.’ Not that you can talk to them as if they are people, but one can and should inquire into them and explain their workings, and thus be able to handle them better.

The claim that ‘there is no tension between the value of justice and the value of advantage’ does not seem to understand the meaning of either ‘justice’ or ‘advantage.’

A lot of these philosophers are going around twisting these words, especially justice. She’s trying to use ‘justice’ as basically ‘do the correct thing?’

Kantians, Utilitarians, and Aristotelians all end up taking Socrates’ side against Alcibiades. When we refer, alongside Alcibiades, to the clash between justice and advantage, we are confessing our inability to give a coherent, non-wavering answer to the question “How should I act?” The theories of Kant, Mill, Aristotle, and Socrates are designed precisely to yield such an answer. (1994)​

Mill is saying that ‘justice’ is whatever maximizes aggregate utility, but that’s not what the word or concept justice means. If you want to maximize aggregate utility as The Good then I can understand that argument, but that simply is not ‘justice.’ It is a decision to not give a flying **** about justice. Claude calls Mill’s move here ‘conceptual imperialism.’ Mill would bite the bullet and say our traditional concept of justice is wrong, but an honest Mill would stop using that word. It does not mean what he thinks (or claims) it means.

The actual implied claim here, that there is no conflict between ‘justice’ and ‘advantage,’ is one Mill would actually reject. Mill is saying AIUI that choosing advantage is immoral and wrong, that maximizing aggregate utility should always win, so there is no conflict. But that’s like saying there’s no conflict between my desire to take your lunch money and you eating lunch, because I’m stronger than you and therefore you’re not going to be eating lunch, so there’s no conflict. Okay, I guess.

I know quite a lot of Utilitarians. They all very much understand that there is a war, that they believe The Good is to maximize utility in general but that people want to maximize their own utility and the people are not confused when they think there is rather a big difference.

Kant basically cheats AIUI in the second critique by bringing in God and immortality and so on to ensure that virtue and happiness align and thus dissolve the conflict, via claiming (nonsensically) that acting maximally according to these rules will maximize your happiness. I mean, that has to be true, right? Or else there would be chaos. There’s a reason most people (AIUI) quietly ignore that attempt. But without it, you fall back on Kant saying his version of rational duty is morally correct, and maybe it is, but that doesn’t mean you win by doing it or it doesn’t face a conflict with advantage.

One could even say that, in making this very argument, Kant refutes the core claim, because Kant in making these arguments violated his own moral laws in order to gain an advantage in various ways, including asserting to those around him he was a theist.

But actually none of that works at all for Kant. Kant explicitly distinguishes Recht (right/justice) from Tugend (virtue). Kant ties himself up in various knots to claim there is no conflict between duty and rational self-interest, but his version of duty goes far beyond only the concept of justice. So no, he doesn’t endorse this.

Certainly if you go around asking most Deontologists, they won’t endorse it either.

If you never ‘waver’ and never see a conflict between different incentives, then that means either you are blindly following some set of rules whether or not that makes sense, or you are omniscient. Not philosophically coherent, but outright omniscient.

Accusations that are, essentially, ‘you claim [X] but clearly that means you are not omniscient’ should be replied to with ‘are you?’

The Hardest Thing In This World Is To Live In It

Socrates seems to answer ‘well no, but how dare you rest until we both are?’

​Socrates would charge all of these views with creating a false ceiling: what they are calling “knowledge” is not yet knowledge. Another, higher kind of knowledge is possible.

The mystery substance with which they insist knowledge be supplemented is, according to Socrates, a fiction constructed to cover the ignorance born of prematurely arrested inquiry. If you actually knew what you should do, you would do it. So long as you don’t know, holds Socrates, the proper ethical attitude is an inquisitive one.

Let me emphasize that I am not putting forward this series of Socratic critiques as valid. I am only putting them forward as Socratic. (2018)

It’s good to see the explicit non-endorsement here, given how silly the claim is.

If you were omniscient and had the ability to adjust your cognitive algorithms sufficiently to get yourself to do whatever you should do, then yes, you would always do whatever you should do, although as per the akrasia discussion merely knowing what you should do at time [T] does not mean you will always do that at time [T], for a common version of what ‘knowing what you should do’ means here.

The bigger mistake is the idea that, if you do not fully know what to do in all situations, if you do not have this full Socratic knowledge that is effectively omniscience, then you should put off action and instead only inquire.

Why isn’t Socratic ethics already a well-established tradition of its own? The answer is that Socrates’ ethics is intellectualist, and people have a strong and deep aversion to intellectualism. Let me explain.

I have already mentioned two distinctive features of Socrates’ ethics: that we do not yet have the answers, and that philosophizing is the way to get them. If you put these together, you get the third and perhaps most surprising feature: viewing the activity of philosophical theorizing as itself having ethical significance. (2029)

I am going to totally, 100%, bite the bullet everyone else bites and say that yes all of this is a sideshow to ‘the real world.’

​The art must have an end other than itself or it collapses into infinite recursion. Inquiry has ethical value insofar as it promises to then lead to superior actions and to pay rent in anticipated experiences. The superior action it leads to can include more inquiry to the extent that this too has an end other than itself.

Socratics, by contrast, believe that arguing about how one should live is real life. Although it is often necessary to spend some or much of one’s time otherwise, inquiry is the best thing one can do with one’s life, given that one does not know how to lead it. The hard work of struggling to be a good, virtuous, ethical person—Socrates understands that work as, first and foremost, intellectual work. (2038)​

This to me is a straightforward claim about math. You lack this perfect knowledge. You’re never going to get it. Is the highest value play to invest the maximum amount possible into seeking more knowledge?

For some people, in some circumstances, the answer will essentially be yes. I do think Agnes Callard and Socrates have a comparative advantage here. They very much enjoy the philosophical life, they are relatively effective at it, so it makes sense for them to be part of the resources civilization allocates to this task, since we can then use this knowledge. We should trade with such folks to enable these investments. But that doesn’t mean that we should be investing most of our spare resources on it.

It would be a great shame if someone spent the bulk of their time so inquiring, in a way that they didn’t get inherent value from doing the way Agnes and Socrates do, kept that to themselves, and then died without using it to live well. That’s not good.

These three features of Socratism—that we don’t now know, that if we knew we would act on our knowledge, and that intellectual conversations are the road to becoming a good person—add up to an “intellectualism” that many people find so implausible as to be ready to dismiss it without serious consideration. (2042)

To sum up how I react to these three features:

1. We don’t fully know, but have a lot of information that helps us act.
2. ​It ain’t necessarily so, but to a large extent yes, and that’s what we do.
3. There’s no reason to assume intellectual conversations in general, or Socratically structured ones in particular, are the best way to learn things either for yourself or for everyone (and they mostly don’t scale), nor that learning things intellectually is the primary thing standing in the way of most people becoming better or more effective people, and also there’s no point in becoming better people in theory if you never put that theory into practice.

If you posed this question to the many scholars who try to save Socrates from his own intellectualism, they would say “because it is obvious that someone could have knowledge—such as the knowledge that it is wrong to steal, or kill—but be unwilling to act on it.”

Socrates’ response is: What makes you think that was knowledge? The real source of the opposition to Socratic intellectualism is not the commonsense observation that people often act in ways they are ready to repudiate, but the insistence that what we sometimes act against deserves to be called “knowledge.” (2057)

When I say ‘I know it is wrong to steal or kill’ that is imprecise.

What I am actually saying is ‘I put a very high probability on the hypothesis that it is, in the vast majority of practical situations, wrong to steal or kill.’

I can indeed know that, and I can and do act on it, and there is a very high probability this is going to work out quite well. We do things we are only 99.999% sure of quite a lot, or 99%, or 90%, or 51%, or 10% but no one has a better idea and it just might work. We must, because the alternative looks worse.

What makes Socrates think he knows that it is better to inquire? He knows nothing.

They Call It Utopia For A Reason

Summarizing: Many have proposed various Utopias, there’s no reason people of various ethical schools can’t propose them, but the closest thing to one in real life, Chautauqua, New York, turned out to be boring, because it did not contain real struggle or real villains. Few students see getting kicked out of the Garden of Eden as a punishment.

I agree that this ‘paradox of utopia’ suggests that we have an incomplete understanding of what we want or would be satisfied with. Like many things, we have heuristics and approximations that mostly succeed on the margin, but that don’t work when sufficiently out of distribution. Boredom and lack of meaning are the final boss.

You play to win the game. After the game is over, you start a new one.

Thus, the best we’ve been able to figure out: Hands make the tools, build the fire, plant the grain. Feet track the herd, build a world, begin again.

In the context of Open Socrates, the obvious parallel is that Utopia is what happens when you make everyone equate justice and personal advantage. The result is you get justice, but you don’t get personal advantage, because without that element of conflict it’s all boring and it sucks.

Well, then they were never the same thing, were they?

The End… Of Book One

That’s a lot of words for what was, essentially: You have been trying to solve the wrong problem using the wrong methods based on a wrong model of the world derived from poor thinking and unfortunately all of your mistakes have failed to cancel out.

Which is way better than the standard approach of not trying to solve the problem, and instead using methods that cannot possibly work without any model of the world or any thinking whatsoever.

Book one focused on the wrong problem and the wrong model of the world.

Book two is about the wrong methods and the derivations from poor thinking.

Given how long this was already getting and that a lot of the groundwork was already in place, as I went over book two I decided to ‘pick up the pace’ and offer fewer quotes and generally Get On With It, at least in relative terms. If you get confused and a passage seems important, my apologies, and I suggest you consult your local LLM.

 

 

 

​

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

​

 

 

 

 

 

 

 

 

 

 

​