Ben Weinstein-Raun

Ben Weinstein-Raun

And so my guess is that that fairness seems like a plausible candidate for something like an alien species might have… kind of universal, like given some sort of assumptions about how evolution worked or works.

Like there is something happening that results in us making choices. And if your philosophical determinism denies that, like you're wrong. And, I think it makes about as much sense to talk about will and free will in making choices as it does to talk about a glass of water.

Ben Weinstein-Raun’s Twitter and Website.

Timestamps

[00:02:00] - Meta-Ethical Stance

[00:11:00] - Contrasting with Moral Realism

[00:26:00] - What EA misses

[00:47:00] - Fanaticism and Outside of Distribution Sampling

[01:28:00] - Building new Tools for Forecasting and Thinking

[01:44:00] - Actually using Tooling for Better Decisions

[01:53:00] - Guided by the Aesthetics of our Distributed Systems

This transcript is machine generated and contains errors

Ben Goldhaber: [00:00:00] And hi I'm going to maybe make a quick introduction to our guest here, and then we'll just dive into some conversation. Ben - who has a great first name - Ben Weinstein-Raun has worked at a number of top tech companies including crews and council, number of innovative tech research organizations such as MIRI and Redwood Research, and Secure DNA.

Ben is currently building a tool for forecasting estimation, and dare I say, thinking. Having known Ben for some time now, I'd describe Ben as a careful, discerning thinker on issues of not just technology, but also philosophy.

And so, yes, welcome to the pod, Ben. Yeah, thanks so much. Yeah, we're excited to have you. I was saying to Divia, one of the reasons I was excited to invite you on[00:01:00] was because of a really great conversation that you and I had a few weeks ago about. Naturalism and ethics and how it's put into practice and like all good conversations about philosophy, it was like 3:00 AM around a kitchen table.

And I felt like you really captured something important about I don't know, naturalism and ethics, and I really just wanna see if we could recapture some of that magic in our conversation today. Cool.

Ben WR: Yeah, totally. Great. Yeah, I'm really excited to talk about it. Awesome.

Divia Eden: I'm excited. I haven't heard most of this yet, so I'll get to hear it for the first time.

Ben WR: Yeah. Nice.

Ben Goldhaber: Well, lemme just ask for the start. How would you describe your moral stance?

Ben WR: Yeah, I guess, I guess at the moment, I would describe my, like best I don't know, my, my best stab or something. My best stab at understanding ethics as more like a pretty strong stance [00:02:00] on meta ethics. And not as strong of a stance on like, sort of object level ethics.

Where meta ethics is sort of questions about like where ethics comes from. Like why do we say moral sentence or like sentences about morality, sentences about good and bad and should whereas like ethics, like object level ethics would be sort of like, what sentences do we say about or like should we say about good and bad?

And like sort of what's true on the, on the direct object level there? I guess the, yes, so the meta ethical stance that I really wanna take is basically derived from an observation, which is that MetaEthical questions are under I think some pretty reasonable assumptions entirely empirical.

So whatever the truth of the object level, ethical facts insofar as they exist. [00:03:00] There's some reason that people go around talking about shoulds and, and good and bad and so on. And, and that reason is like entirely inside of physics. It's not mysterious. It's the kind of thing that we could figure out by like, looking carefully at humans and at human societies and at maybe, maybe game theory and math.

We don't need fundamentally different tools to answer that kind of question. If we're interested in answering like, the straightforward question, why do people say sentences like that? What is sort of the driver for humans having these intuitions and having these sort of discussions and, and and so on.

And so that's sort of like the, the, the meta ethical stance. And then I think it has some kind of interesting implications for taking this stance has an interesting implications for the more object level stuff. So I think the most obvious one to me is that it, it does [00:04:00] not seem like it lends itself well to sort of very simplified rule-based systems.

So I think it, it pushes me pretty far away from like a sort of total utilitarian kind of a view where you're aiming to make your object level ethical system very simple and make that sort of like this very beautiful sort of object that you know, is unassailable.

Ben Goldhaber: I'd love to ask a question there about that because I could see some people actually like in intuition that I feel like I have, when I hear that from a meta ethical point of view, everything is within this same kind of system and I suppose all derived from physics on down it. Lends itself to thinking about things in a very maybe not very, but like I could see it applying in like a simplified manner.

Like thinking okay, well because it's physics and because it's something knowable, we can construct theories about [00:05:00] it that are simple and derive into this kind of utilitarian calculus. But it sounds like you're actually pushing in the opposite direction and saying, no, it's much more complicated.

Ben WR: Yeah, I think basically my sense that like you get something complicated is not like solely based on the idea that it's from that meta ethics is like empirically addressable. It's also based on sort of like observing what is going on when people talk about good and bad and should.

I think it seems to me that if you want to have, like, if you wanna have your meta ethical stance basically look correct or like consistent with the evidence I think you need it to include some things, which to me do not. Seem like they would come from an approximation to utilitarianism.

You need it to include things like I don't know. So, so Jonathan Hate has this [00:06:00] book oh shoot. I'm gonna forget what it's called cuz I'm being recorded. But it's the Jonathan Hate book. I feel like there's one Jonathan hate book people quote in this situation. No, it's definitely the out one.

Ben Goldhaber: There's,

Ben WR: let's look out and then we can, and then you can go back and say that. Yeah. Yeah. Ok. Is it the Righteous Mind? I think that is it. Yes. Ok. Okay. So, so Jonathan Ha has has this book, the Righteous Mind where he sort of goes into like doing, excuse me, a lot of the sort of like empirical like analysis of what is going in, what's going on in people's minds when they're having this sort of like set of ethical intuitions and conversations.

And he comes out with like several factors like I think, I think it has like five sort of like key key things that, that are sort of factors of people's morality. And one of them like, like an especially important one and one that like, I think is especially important to me is [00:07:00] sort of harm focused morality.

Which is I mean, it's quite widespread, but it's not like the only thing going on for almost anyone when they're thinking about sort of like what's ethical. So,

Ben Goldhaber: So there are like multiple different things going on when people are calculating, thinking about what's ethical and like a certain re reductionist point of view.

That's just looking at the harm's point of view is probably missing a lot of things. Is this one way to put this?

Ben WR:Yeah. Yeah. I think that's basically,

Divia Eden: yeah. Can I, I just, I just pulled up what the different ax axis, the five one s he has can, do you mind if I say them? Yeah, yeah, totally. Okay. So he has care, harm, fairness, cheating, loyalty, betrayal, authority, subversion and sanity degradation.

I have some, at least that's what I found when I looked it up. I have some imperfect memory that he added, like freedom in there later because he found that that was important to some people and it wasn't originally

Ben WR: covered. Yeah, and I think, I mean his I think he also has sort of generally this take that [00:08:00] there could be a, a bunch more, and these were just sort of like the most obvious when he was sort of like going through like the available, like the available evidence.

And so, yeah, I think, I think it's, it's not clear that that is like a complete list, but I think it is clear that like your meta ethics has to sort of explain all of that. And it's not impossible, I think to, to end up with an explanation, like from my current standpoint, it's not impossible that you would end up with an explanation that sort of like is more or less simple like utilitarianism but it doesn't feel like what you're gonna learn if you like, started off sort of like with a clean slate.

Like just sort of examining humans sort of a, as some, as like a species that has these sort of intuitions and and ways of functioning in a society. It just doesn't seem quite like, it doesn't seem like that's gonna be like near the top of my like, list of plausible [00:09:00] explanations. I want to kind of like almost pop back for a second or maybe double click on the word meta ethics. Cause one thing when I hear that I think about it is like selecting among some set of ethical philosophies. Is that how you mean

it? Not exactly. I think, so when I say meta ethics, which may not be quite what people like, typically mean by meta ethics when they're like professional philosophers.

I mean something like Like, what are the sources of our, like ethical intuitions and our like the sentences that we say about ethics? I think that like, some, like, I don't know. It's, it's the sort of like, it's the domain where like you might a wonder about like whether moral realism is true.

Like whether there [00:10:00] really is some kind of like, like objective moral truth. Whether, you know, is moral realism real? Yeah, that's a good question. I don't know. I mean, it's a, it's a real concept although I think it might actually be a lot of different real concepts. But yeah. Are you

up

Divia Eden: for saying what the different, what some of the different real concepts might be?

Ben WR: Oh yeah. I mean, so I think so in the like local social sphere, I guess like people use the term mo moral realism in a way that I think is not quite the same as the way that philosophers use the term. Excuse me. I think philosophers, when they say moral, moral realism, they mean something sort of broader.

Like is there any sense in which like, you know, morality is real or like good and bad or real or, or anything like that. And I think that that admits a lot more ways that things can be real than [00:11:00] often people are imagining when they're saying moral realism. So if you're, if you spend a lot of time around ea you might like, like.

Come across disagreements where people are sort of like talking about moral realism versus moral anti-real. And the anti-real are saying like, are are sort of, they're taking this view that like there's no supernatural, like the universe doesn't care. It's sort of like a combination of like, I dunno, they, they think it's like maybe morality is subjective.

They think it's like, maybe not like aliens might not have the same moral systems as we would. They think that like, you know, there's no sort of like underlying supernatural like thing that is that is mor morality or ethics. Whereas the, the realist think that there maybe is some kind of a thing like that.

Like the, the, I don't know you might get sentences like the arc of history bend toward justice or like, you know, [00:12:00] I only care about the worlds in which morality is real because like the other ones have no value according to me. And I think that like, it is pretty, I think this sort of like this sort of like splits too many things into those two categories.

And I think I, I would, my current sense is that like there is a sense in which morality is real. I, I don't think that it's supernatural. I don't think that the universe cares about morality. My guess is that some of it is kind of universal. Like if you were to like, go find some alien society, they would like share some of our moral intuitions and probably some of it is not.

Ben Goldhaber: And and you suspect those universal features are derived from shared evolutionary

Ben WR: patterns. Yeah. Shared evolutionary patterns, game theory basically the sort of [00:13:00] thing that like might be common between our culture and an alien culture. Can you give an example of some things that you, where you think the aliens would probably have the same moral intuitions?

Yeah. I think one thing that seems interesting to me is that like fairness seems like it's quite common as like a, a sort of, i, I don't know if it's quite fair to say moral intuition, but it's like, it's a common sort of motivator across lots of different animal species, like not just humans. It, it seems like sort of quite like an early thing to, to like evolve in terms of like Something like things that sort of look like morality.

And so my guess is that that seems like a plausible candidate for something like an alien species might have or something that would look sort of like how we would think of fairness I think, so fairness seems like a, a strong candidate for something that like, might be kind of universal, like given some sort of like assumptions about how evolution worked or works. I [00:14:00] guess it does seem to me like something like sort of like slightly weirder decision theories is also another plausible candidate where like, If you have a decision theory that isn't just c d t, sorry, caus causal decision theory like that is potentially going to help you coordinate with other people a lot better or other like members of your sort of species or people who you sort of like see as similar to yourself.

And so I pre predict that things like that are also gonna be quite common in, especially I guess like social animals or like, yeah, I guess, you know, generalized animals.

Ben Goldhaber: I'm still kind of thinking about something that you mentioned around like why utilitarianism doesn't seem like a good approximation of the meta ethics you might endorse.

[00:15:00] And I think you answered this in a way, but I didn't quite gro it around the point of like Jonathan Height and the like different ax axis. And I'm wondering if you can say a little bit more about it. The way I'm like kind of thinking about it is something like maybe you can't like, make trades between those axes in some way that all sums up to a number or something that the like stereotype view of utilitarianism might have.

Yeah. Is that somewhere in the ballpark? Or, or maybe just say more about that.

Ben WR: I do think that there's some, like, you probably could mathematically describe a way of making trade-offs because like, you know, obviously you, you sort of like can't Like, I, I mean, it's gonna be hard to construct a system like that where it's not, where you're, where you're not allowed to make any trade offs.

But I don't think that's necessarily gonna be like a simple sort of like add up all of the, the field and like, you know, with different weights or whatever. I think [00:16:00] some of it is gonna look not very consequentialist and look sort of more like, are you being honest? Like, is that like, you know things that don't have direct like, I don't know where the, where the morality of the thing does not like route through its effects.

Right. If that makes sense.

Divia Eden: I, yeah, I was trying to think of an example of this in my head. And so I guess one of the, like, I took one of those eight quizzes years ago and there's some question that, you know, I probably, a lot of people I know I'm somewhat low on, maybe a little higher than I used to be as I get older, which is something like sanctity or it's like, how bad is it to, I don't know, like play a game of cards in a graveyard or something like that, where like it might seem in, you know, there are no there are no concrete consequences that can be easily tracked at least.

But a lot of people have some sense that this is wrong. And maybe what you're saying is like, you know, if I'm imagining some pole that's like how many. I don't know, like how many, how many dollars would you have to give [00:17:00] to the against Malaria foundation to make up for playing a Rockus game in a graveyard?

That this, there's something Ill-conceived about this. Is that sort

Ben WR: of what you're getting at? Yeah, I think that's basically right. Where like, it's not, it's not quite clear that there's like no trade that you could make or, or anything like, quite that extreme. But I think it is sort of like the question is like not actually giving you the information that like, might be needed to answer the question.

For example like it might depend a lot on your state of mind, like when you were playing the game, it might like, you know, it may just be that there isn't really, like the thought experiment does not actually like, give you the relevant details. But that doesn't necessarily mean that there isn't like any kind of trade off that you could make at any given situation or that, like, there aren't details that could be given.

But I think it does mean, I don't know that a lot of these yeah, that, that it's like, it, it's not going to end up [00:18:00] being a very, very straightforward kind of a, like a function from like the state of the universe to like how good or bad it is. I think I especially expect it to not be a simple, a simple function of something like summing up all of the positive experiences.

So, so I think, I think I find total utilitarianism especially suspect as a, like object, the let system it's not crazy to me to have like a utility function. I think that's like maybe the kind of thing that you, you can still have. It's just that I, I don't expect it to be a very simple one. And I think I might expect it to sort of have like terms about what your mental state is and not just terms about like what's out in the world.

Ben Goldhaber: Right.

And how do you maybe personally or philosophically relate to this? Is it something where you try to hold like multiple points of view and like make decisions from that? [00:19:00]

Ben WR: Yeah, I mean, so since I had this thought a while back, I've been thinking more in terms of like, what would my, like, I mean, it, it's a little silly to say to me, but, but I've been thinking a lot in terms of like, what would my grandfather do?

Or like, what would my grandfather think is the right thing to do? Oh, that's nice.

Ben Goldhaber: I like, I just like that immediately, but please continue. Yeah.

Ben WR: And I guess maybe similarly thinking about like, What seemed to be kind of like universal features of moral systems. So I think it, it has caused me to sort of like up weight being honest, like because I feel like that is like quite a common, like admonition.

It's also like caused me to up weight, like the golden rule roughly. Like, you know, treat some kind of thing vaguely shaped, like treat people like you wanna be treated. Maybe I mean there's lots of different ways you can sort of like add epicycles to make it [00:20:00] better cuz there are obvious problems with it.

But but using that kind of thing where like, I, I just, I I have observed that like lots of different cultures and lots of different groups like seem to value those things. And that, that seems like if you, if you were to take a, the set of things that are like that, that would sort of be a minimal collection of like what might be considered like human morality, right?

Ben Goldhaber: Try to shoot for like a minimum rule pack of human morally. Yeah. And then layering on top of that, does that kind of come from your personal exper or maybe, maybe like, you're not supposed to almost like add more things on top of that and what do you

Ben WR: Yeah, I, I think I mean, so one question here is like why if I have this view, if I'm like, okay, where like ethics is all sort of like explainable by physics, whatever, like why should I find it compelling?

Like why would I want to be moral? Mm-hmm. Mm-hmm. And I think that in some ways I. Shouldn't want to be like, [00:21:00] like absolutely moral. However, I do care about the systems that I'm embedded in. Like I, I care about like the world that I'm in and like seeing it continue to, to thrive and continue to exist.

I like, you know, I care about my friends and, and all this stuff. And my sense is that like some of morality is going to be tied up with like the preservation or you know, creation of systems like that. And, and probably quite a lot of it. And some of that is gonna be directly helpful to me.

Like it's gonna be, it, it's gonna be derived from like patterns which like helped other people with those memes to like get more whatever, like basically propagate themselves or their memes. And some of it is gonna be like, it helped the societies that those people were in. And I think both of those can be like quite compelling reasons to want to be moral.

And so I [00:22:00] guess that's sort of another way that I like another sort of source of like maybe layering on top is like, can I sort of like figure out. How this is like, I don't know how this fits into that kind of picture. And like, like also sort of like helps to some extent like eliminate aspects of morality that I like, don't think are as important.

So for example, I think like believing in a particular God to me is not the kind of thing that I expect to like, come around to thinking I should do. Just because like, it, it doesn't seem like, I don't know, there are too many different choices there. It's like not, it's not like really well justified.

Any particular one. I might be like more open to the idea that like maybe I ought to pray or something or do some kind of like sacred ritual which seems sort of more common. But yeah, I think [00:23:00] my sense is that like, insofar as that would be helpful to me or the societies that I live in, like there are better things that I can do with that belief, like believ through things instead.

And so I'm not especially inclined toward that.

Ben Goldhaber: Yeah. It's funny that you started mentioning God and religion in this because in part of what you were saying, I feel like I was hearing real. Echoes and shades of Cs Lewis's concept of the Dao, like some set of rules or principles behind civilization that seems to be universal and that he tightly coupled with both morality and the Christian faith right.

And yeah, I, I, I certainly see like the way also in which, in you're bringing this into almost like a game theory kind of mode of like justifying it through like, all right, well these are the principles and rules, but with civilizations and people flourish. These are the kind of underpinnings of morality here, and I should look towards

Ben WR: those.

Yeah. Yeah, that feels really right to me.[00:24:00] I mean, yeah, I, I'm not sure if I've read the specific CS Lewis thing that you're re referencing, but I have read a, a bit, and I think basically I, I, I often feel like I'm with him like 70% of the way, and then like as soon as he starts being like very specific about why it's Christianity in particular, I'm like, Hmm that doesn't really make sense to me.

Mm-hmm. But I do, yeah, I do resonate with a lot of the stuff that he says about like, I mean, sort of like Yeah. The kind of thing that you're, that you're talking about.

Divia Eden: So I, I have a question about when you were saying, like, it almost sounded like what you were, when you were describing it before, that the grounding of why, to, why, like the motivation to follow these moral principles was to be helpful to you and the societies.

You're, the systems you're embedded in, but then that starts to sound more consequentialist in a way that I think you don't mean, and so my guess is that it's sort of hard to use language. So talk about this, and that's kind of what's up here, but I, I guess I wanted to like, if you could expand on

Ben WR: that point a little.

Yeah. I think [00:25:00] like, it's not so much when I say that, like, I, like why do this thing? I think it's actually, it is actually a bit consequentialist. It's just not, it's just not sort of like morally consequentialist. It's not that like, I think that I ought to do the thing which causes the best outcomes.

It's that like I separately like happen to want good outcomes. And along with a lot of other things that I want and like wanting good outcomes, like I think leads to wanting to be moral by sort of like observing that I'm like embedded in these systems. And that I like, you know, there are these apparent rules by which people seem to try to steer the systems to like whatever, like better, more productive aims.

Ben Goldhaber: Is there a moral principle or? Universal feature of some of the groups that you [00:26:00] think maybe is absent from one of the societies or groups that you're in, like something you think they should be doing more?.

Ben WR: So one of the things that I think feels really important to me about this is that I, I agree with effective altruism, like the movement on a lot of like, key points that most people disagree on. And I observe that ea seem to make sort of classes of mistakes that I think that they would not make if they understood this like this general sort of like direction of thinking at least a little bit better. So so like S B F I, I think like really is like a true believer. Totally utilitarian. And he's a smart guy. And, and I kind of think that, like, I, [00:27:00] my sense is that like you.

You can sort of only end up doing things which like, like the, the kind, you can only end up making the kind of mistakes that it seems like SPF F and others at FTX made. I think by sort of neglecting some of these sort of more traditional like ethical principles, like, like try to be honest, try to be scrupulous with people's money.

You know, things that I, I guess it does seem like like, I don't know. I mean I, I, I really respect his and, and also Caroline Allison's commitment to their, like, moral principles. And I, and I don't, you know, maybe controversially, I, I don't actually take Ft X's collapse and like fraud as that much evidence that they like abandoned those principles.

In fact, I, I, it seems quite [00:28:00] consistent with those principles to make that kind of mistake. And I think, I think it does feel really important to me insofar as this is like a correct observation to like point it out to the people who I think are trying to do as much good as possible so that they like, might not make mistakes like this in the future.

That makes

Ben Goldhaber: sense.

Divia Eden: Yeah. So can, like, can you be more specific about what it is that you wanna point out to people?

Ben WR: Yeah. I think something like, like your, my mainline guess, and I think a very reasonable mainline guess about sort of like the metaphysics of the world is that basically physics is what's going on. Like, you know, the universe is sort of more or less mechanical at, at the bottom layer.

And that everything that we say and do has perfect explanations inside of [00:29:00] physics and therefore like can be explained by looking at like by, by basically just being empirical. So I think a lot of people who I know have this sort of pseudo like sort of pseudo supernatural, like the people who I think of as, as, as thinking of themselves as moral realists in the ea sense.

They have a sort of pseudo supernatural view of like, like what it would mean for there to be like, like real morality in the world. And I guess it seems to me that there's a whole, there's like a category of belief that one can have that is not. Like almost inherently disconnected from the truth of the matter of that belief.

And there's a way in which if someone is like, ah, yes, I think that there is some sort of [00:30:00] supernatural like, goodness thing. And like, you know, it's gotta be the sort of beautiful, most beautiful, simple thing, probably utilitarianism. I think there's a way that that belief, if it were true, like there's no route for the truth of the belief to influence the belief.

And if it were false, there's no route for like, the falseness of the belief to, to, to influence the belief. So it's, you're saying it's unfalsifiable, it's not exactly that, it's unfalsifiable. It's that like there just isn't like if, so physics is probably causally closed, like in insofar as like anything can be like there, there everything that happens in physics, including all the things I'm saying and all the things I'm like experiencing and we're talking about like have perfect explanations in terms of physics.

And so if you're gonna pause it, something outside of physics that thing, like there's no route because I have perfect explanations for everything in terms of physics. [00:31:00] There's no route for anything outside of physics to influence my belief. I think. So it's different than falsifiable because like it might be the case that like, I have a belief about.

Like how evolution happened that in fact, I will never get enough information to like, know the truth of, and that's actually like a, I think that that does not fail this test, but does fail the falsifiability test. Where like, at least that belief, like there is some way that, that like could in principle, like be connected to the truth of the matter.

And yeah, I think that's, that feels kinda like an important distinction

Ben Goldhaber: and I'm still kind of puzzling on this. Like, so like maybe some folks have this kind of conception of some thing outside of the realm of physics that is the source of like moral truths. And your point is many people [00:32:00] in the EA space don't necessarily think that is God, but that they still treat it in some way like that.

And you'd wanna bring that back down into the realm of physics while still holding. Also hold, also believing that there is some source of truth that is beyond like, or that is like somewhat universal.

Ben WR: Yeah, I mean, I think that, like, that source of truth can be, and in fact, like I think more or less, the only source of truth as far as I can tell is like things that exist in the world.

And like, and that seems like a, like a totally reasonable place to me to look for like true morality is just like, yeah, try to figure out what it is we're talking about when we're saying these sentences. And so it's,

Divia Eden: it's part of where you're coming from that you think it's sort of, it's, it's appealing to people to create something elegant in their minds and give it, like, elevate it to a special place that's [00:33:00] ultimately ungrounded.

And is is part of what you're saying that with your meta ethical stance is that no, that's sort of unjustified and people actually need to do the work of looking at the world and grounding their ethical intuitions. Does that seem, is that closer?

Ben WR: Yeah, I think that's almost exactly right. And yeah, the, the only place where I guess maybe I would like slightly modify it is something like it's not, it's not clear that you like even have to ground it all the way out.

I think it just sort of like, it, it's going to be like, I think it's important to keep this in mind and like let it influence your like, Probability distribution over like what the truth is. Like if you are, if you are as I am, like, you know, relatively confident that like physicalism is roughly true that should impact your beliefs about why people say sentences about good and bad.

Imagine I sense imagine people. Yeah, yeah, yeah. Sorry. Yeah. Like if I imagine [00:34:00] my sense people first go, but yeah, go ahead.

Divia Eden: Yeah. If I imagine you say, you know, hypothetically you're running a bank, you have some customer funds, you're deciding what to do with them. Can you sort of walk me through how your physicalist meta ethical stance would infor, like, can you, can you sort of lay out the steps for that from then what you do with the customer money?

Ben WR: Yeah. I mean, I think for one thing like I think it, it pushes quite hard on. Things like honesty which I think as I sort of mentioned before, like my sense is that honesty is like like fairly universally seen as like, you know, like the ethical thing to do. Like all other things equal. And I guess it's not necessarily the case that all of the actions with the money will, will like be dic dictated by an ethical system.

Like it might be that I like, have a lot of leeway over like exactly what investments to make. Like maybe I would like to just [00:35:00] maximize my return. Like, that's probably ethical as at least assuming that I'm like being honest about that with with the people whose money it is. Maybe it's that I like want to invest in things that I think will make a positive difference in the world.

It's not super clear to me that like that like this is like sort of required by an ethical system. My guess is that it is probably like better to do things that are better, but I don't think, my sense is that like the true moral system is not extremely demanding. If that makes sense. Partly because in fact humans mostly are not fanatics.

And things seem to work out basically fine and in fact kind of better when they're not fanatics. And so I don't expect like to find that like the true moral system insofar as there is one, [00:36:00] or like, maybe there is one relative to me or relative to me in my situation or something like is going to make demands of me.

Like, you must find the most, like the, the best possible thing to do with this money. And like, I guess it seems to me like it's, it's totally, it's likely to be very compatible with like many options especially ones that are sort of like being what is recognized locally to be like like rather like fulfilling the role of a good bank manager or whatever.

Insofar as that's like what people are expecting from me. And that that may not be like I don't know, it may or may not be like involves sort of like maximizing profit. It may or may not involve sort of like, you know, being lenient with people on their loans or whatever. I dunno, I'm not sure if this is a good a good answer to your question.

[00:37:00] I

Ben Goldhaber: like it because it did help me grasp something much worldview here and then also kind of what I feel like is a common. Well, I'm not even sure failure mode, but like a tension point. I feel like with a lot of the philosophies that we talk about, which is like on the margin, what's the next action that you're gonna take and like how much should it be influenced by some other thing, like on the margin, are you going to like, donate the next dollar to a M f or Miri or some other charity?

And then there's always this question of like, well, why am I not donating the next dollar and the dollar after that? And I guess what I'm hearing you say is something like, no, we should kind of resist or at least be very skeptical of this idea that ethics can be a universal operating system for your choices.

Is that an accurate kind of statement?

Ben WR: Yeah, I think I think basically it, it, yeah.

I think I do wanna say something like, [00:38:00] I don't think that the, like insofar as there is gonna be a true ethical system, I don't expect it to make a prescription about every action. Hmm. I expect expected to make prescriptions about lots of actions and to like, you know, strongly push against some and strongly push in favor of others, but.

I don't think it's gonna be like, you know, like, I think, I think there is something kind of wrong with the sort of the like obligation framing, which I guess a lot of VAs sometimes talk about with respect to ethics, where I think like the obligation framing of a particular ethical system is like basically going to lead you into fanaticism for basically the reason that you're talking about.

Like, ah, yeah, well I spent my first, you know, 60% of my income on am m f I guess maybe I ought to spend the [00:39:00] next 30% too. And then, I don't know. And my guess that this just doesn't actually work that well for like, you know, building a society like, like actually in fact causing the most good outcomes if that really is what you want.

And I mean, in particular, I just, I just think it isn't probably good like making reference to what I expect to, to find about like what good and bad mean. Mm-hmm.

Divia Eden: I, yeah. I think the part that I'm most trying to clarify in my mind is something like the step from the sort of non supernatural, empirical view on morality to then you are sort of looking, there's both some question about what actually makes societies work.

And then there's some sort of additional sense that looking around at what seems to, which principles seem to people converge are convergent seems like a pretty good source [00:40:00] of information about what the true morality is. Mm-hmm. Am I, does

Ben WR: that seem right? Yeah, yeah, totally. Yeah, I think that, that seemed like it basically hit the nail in the head.

Like there sort of is, my sense is that there are, like, if you take this, I guess the way that I would phrase it or something is like there's sort of a, a. First step, which is like, you notice that you can answer these questions empirically, and then there's like a second step. And I've, like, I, I think it's easy to take the first step.

The second step is hard, which is like, okay, now we actually have to do the empiricism and like actually try to figure out what's going on. And I don't think that we've like, succeeded at that. I, I don't think that I can just like go crack open Jonathan Hate's book and like read out what the true morality is.

And like, I think it is in fact a quite difficult project. I actually think my sense is also, this is basically to some extent like the project of like the original [00:41:00] sociologists. And I think sociology in the meantime has sort of gotten like, redirected to other things. But my sense is that, that they were really interested in sort of like cultural universals and specifically around sort of like sacredness and and like morality and religion.

And so I think my sense is that like people have sort of like made these sort of like halting steps towards the second step in this, in this sort of program. Like enough that I, I feel like there, I can say more than nothing about like what I expect to find, but I don't think that I can say anything with all that much certainty.

Like I think like probably 80% of the things that I've said in this, in this like conversation so far I feel very un uncertain about really. And like I would be unsurprised to learn that they were basically wrong, but some of them Yeah, you go, yeah. And maybe that. Yeah, and [00:42:00] I think partly I, I'm even like open to the idea that in fact moral realism is false and like there is no sense in which morality is, is like a real thing and that like, you know, there's sort of this like non-cognitive view that like, we're basically just sort of making emotional expressions when we talk about good and bad or something like that.

And that there's like no sort of consistent, real thing there. It's, it's like pretty strongly not what my guess is. Like, I, I think there's like enough things that I can point to where I'm like, okay, but that thing definitely exists and like, it seems like it's part of what we're talking about. But but I don't know.

I, I, I'm open to being like wrong about even that.

Ben Goldhaber: It strikes me that some embrace of uncertainty or toleration of uncertainty is kind of central to your worldview

Ben WR: on this. Yeah, totally. I think there's a way in which, so some is talk about like like, like moral uncertainty like I guess especially like Will McCaskill has written some stuff [00:43:00] about it.

And when I first read that I was sort of like like, or read about moral uncertainty. I, I had this sort of like visceral yuck reaction to it where I'm like, ah, but like I feel like it's not really like he's describing it as sort of like, oh, well maybe like 30% on utilitarianism and 30% on like virtue ethics or something.

And I just kind of get the sense that that discussion is often not grounded in like, how you might actually answer the question. And like, like trying to constrain your predictions like so that they match as well as you can, like the evidence that's actually available. And yeah, I think it is, I think it is basically, it feels really key to me to have something like moral uncertainty but to in fact ground it in.

Like what, like, I don't know how you, how you would, how you would expect this kind of [00:44:00] exploration to, to, to play out.

This

Ben Goldhaber: kind of exploration. Could you say more there?

Ben WR: Yeah. So like step two, I guess, where like you're sort of going from the observation that probably like all of this stuff is empirically like sort of discoverable, right? And then in fact trying to go and discover it, so, right, right. Yeah. Like trying to figure out why it is that people have like fairness intuitions and like you know, things like this.

This makes a lot of,

Ben Goldhaber: I, I sense to me, or at least it's like one of the things where I really enjoyed talking with you about this was some sense of like, yeah, there's something here that feels like both very true and very humane, almost like kind of human-centric. While also as soon as I get to that word I'm kind of thinking like, oh wow, there's gonna be a, there'd be a lot of fighting.

There would be a lot of different points of view on this. And I imagine that's like some of the appeal of some other like, Modes where you can be like, no, we, we have the answer. [00:45:00]

Ben WR: Totally. And like I think, you know, I think there are a lot of like benefits to making simplifying assumptions. I'm not necessarily saying that people shouldn't like, use something like utilitarian calculus as like an input into their decision process.

But I don't think, I think that that, that people are sort of I don't quite wanna say diluted, but like something like diluted into like thinking that's please say diluted.

Ben Goldhaber: We need more good clips to be able to kinda like, get some spices in, you know, via

Ben WR: totally Twitter fans. Yeah. So like, I guess I have a sense that people are sort of diluted into thinking that that's all they need to do that that's like, in fact the whole deal.

And as long as they've done that, then like they're doing as much good as they can do. They're, they're, they're sort of unassailable morally. And I think that is not really true. I, [00:46:00] I mean, I think it's both. Yeah. I think that that point of view is both like too harsh and too, and like not harsh enough.

Mm-hmm. It like excludes a bunch of things that I think people sort of in fact ought to be paying attention to and aren't when they're sort of in that mindset. And also, yeah, it, it like does this sort of like pushing toward fanaticism thing and like pushing away from like sort of having a, a healthy life?

Divia Eden: Yeah. Can you say more about why you think fanaticism is not good?

Ben WR: Yeah. I mean, I guess my sense is like if I have a bunch of, if I think about like the central examples in my mind of like fanaticism from history they, like the fanatics often end up in the historical memory as being like a. Basically the bad guys is, I think one, kind of, one part of it.

Or the people who, like, were more fanatical of [00:47:00] the possible, like choices seem like they're sort of, they, they sort of end up seeming like the bad guys. That's like not itself that much evidence, I think. But I also observe that like, I don't know like the people around me who are the most fanatical are like, in fact not the most productive or like doing the best.

And in fact, I think there's probably like an anticorrelation between those like another, another I guess point that feels like it's on the scales there. But yeah, I mean, I also think about like times when like the more fanatical group, like clearly caused a lot of harm. Like I think I, 2, 2, 2 examples in my mind that that come up a lot are like the French Revolution and like the Russian Revolution, where like, I think both of these were sort of like more or less driven [00:48:00] by like a kind of fanaticism, which like ultimately resulted in like both directly a lot of bloodshed and, and suffering like during the revolutions.

And also like after the fact inst like installed regimes, which themselves were like, I think. Like quite clearly bad like, yeah. Napoleon and the U S S R both seemed to me like they caused a lot more harm than like many other possible, I guess regimes might have. I dunno. And I think another one would, another example would be like like China, like Maton and and like various like cultural revolution stuff.

I guess it just seems like many recent historical examples to me feel like they sort of have this correlation between like high fanaticism and like worse outcomes. Whereas [00:49:00] like in the American Revolution, my sense is that like people were much less fanatical. They were sort of much more pragmatic.

They like, I think sort of still saw themselves as like, you know, roughly British people and like, sort of were just like, they were pissed off, but they weren't like fewer, fewer people were sort of like calling for like the blood of the elites or something. And I think, I guess like it just seems to me like that resulted in a better outcome.

Divia Eden: Yeah. So that makes sense to me in a lot of ways it seems like sort of the, the term that comes to my mind is like a genre savviness thing. It's like, I'll read the baddies, like that gift,

Ben WR: whatever. Yeah, exactly.

Divia Eden: And then, so, and I'm sympathetic to your point, if I imagine myself coming from a more utilitarian.

Point, I would say something like, okay, sure, but if I were, I would simply do the utilitarian calculus and count it as super negative that, you know, that all the people would die of starvation and therefore I would [00:50:00] not do that. Yeah. Which, I mean, I don't think you would particularly, you know, I think you'd probably be happy to have them taking that into account, but like, I, I think I'd like you to hear, I'd like to hear a more direct response to like, okay, but why not double down on the fanaticism and get better at calculating the

Ben WR: consequences?

Yeah. I mean, I do think that like, there is some appeal to that, like, especially insofar as like the world changes over time and we're sort of like in a new regime where like maybe we actually could calculate all of the consequences. Like, I don't know, maybe a hundred years from now we'll have Jupyter brains and like be able to like actually figure out what all the consequences are.

And I think in that world I'm like a lot more sympathetic. For now, I, I don't think that's the world we're in. My sense is that like, I, I don't expect to be able to do those calculations well. If I were to try I basically expect to do better, like by my own lights. By sort of like internally [00:51:00] appealing to my sense of like what's wholesome and like what is sort of like morally good in the sort of more, more traditional kind of a sense.

Like you expect

Divia Eden: that sort of calculation to sort of fail on its own merits, you think it Yeah, totally. To give a