Beri:

Hello, Agnes, and hello.

Agnes:

Hey, Beri. Do you want to just introduce yourself?

Beri:

My name is Berislav Marušić, but I like to be called Beri. I teach philosophy at the University of Edinburgh, and I'm from Croatia, but I spend a lot of time in the States. I was in grad school at UC Berkeley together with Agnes.

Robin:

And so that means they're both very young. It was very recent, of course.

Beri:

Very kind of you to say.

Robin:

All right. Well, I thought we're going to be talking about a paper you wrote and topics near it on the rationality of disagreement. And I thought we might try talking about a slightly easier topic that's nearby, if you'll indulge me, which is imagine, say, you and I are car racers. We race cars. And we're about to do another race. And in the past, Uh, we've gone on different tracks, et cetera. And sometimes I win, sometimes you win in terms of who comes out ahead. And so we seem to be roughly similar quality car racers. Now we're about to go in this race and we've each, you know, done a few things to research this track and do some training and maybe tune the cars. And so now we each know some things. The other one doesn't about our car. And now the question is. How entitled could I be to think I'm going to win this time? Because, look, I tuned my car, I trained, and I think I'm going to win. But even though, from an object outside point of view, we look pretty similar, our track records are similar. I win, sometimes you win, sometimes. So obviously, I'm sure you see the analogy here, but what do you think? justifiably think I'm going to win.

Agnes:

Well, you've touched on the topic of Beri's other book.

Beri:

That's right. So if it were up to you to win, then sure, you could answer the question of whether you'll win by considering the question of whether to win. Can you say a little bit, you said it was obvious how the analogy bears, but could you say a little bit more how you see it translating to disagreement? Because part of my thought is that I don't think of disagreement in that way. I think of disagreement as a sort of thing one does with another in a way that It's a little bit like reasoning with another. And maybe racing is something one does with another as well. But in your picture, the thing that I'm really interested in, the interpersonal, is missing. So I was wondering how you see the analogy.

Robin:

So there's this literature on the rationality of disagreement in philosophy. And I have some issues with how that is usually framed. But I'll just summarize. The usual framing is we imagine two people considering some topic. And they have both been thinking about it and have analysis and evidence, et cetera. And they're similarly qualified and similarly smart, et cetera, and thoughtful. that is, and they both agree on those context features. And then, with regard to some particular topic, it's usually framed as they simultaneously produce opinions, and then they compare those opinions, and they see that they're different. And then the question usually posed is, how should each one respond to seeing that the other one has different opinions, They may have already shared some arguments, but they haven't fully entered each other's minds. So there's still remaining some uncertainty in each of their minds about what could the other person really be thinking fully. And then that's how it's usually framed. And as you say in your paper, the usual two stark positions are hold firm. You have reasons for your belief and you should stick with them or equal weight, give roughly equal weight to your and the other's opinion. and in which case you should adjust your opinion upon hearing that the other one disagrees with you. So that's the usual framing. And so the analogy to the car race would be, you know, the car race we're about to race and one of us will win. And in this analysis race, one of us will be more right than the other. And we're wondering who's more right.

Beri:

So the central idea of the paper is that the usual framing of the problem of disagreement is a kind of distortion and that it leaves out the way that disagreement really could be understood in a central case. So the way you've described it, we're supposed to imagine a case of disagreement without really knowing what the disagreement is about or the reasons that each person had for arriving at their view. And yeah, if you abstract from the fact what the reason, the disagreement is about, and if you abstract from the reasons for which they are held, and if you abstract from the people who are involved, then really there's no, as it were, reason to favor one over the other. But there's a lot of abstracting that's gone on, which I think hasn't, has kind of abstracted from the heart of disagreement. So sorry, I feel like I want to push back on the analogy. And the point of the paper is really to push back on a certain way of conceptualizing the problem.

Robin:

You could say those same objectives in the context of the car race. You could say my description of the car race also abstracted from all those details. Then would you similarly object to the car race hypothetical on the same grounds?

Beri:

Uh, and the question is, are you entitled to believe that you'll win? Are you entitled to say you will win?

Robin:

I mean, uh, I thought the topic was not car racing per se, but believing or saying, so in the context of the car race, are you entitled to believe that you will win and to say that you believe you will win.

Beri:

If I only know about my car, but not about your car.

Robin:

Well, you know some things about their car, but you know more about your car than the other person's car.

Beri:

I just don't know what to say. I'm sorry. It's not clear to me what one should say. It depends a bit on what it is that I know.

Agnes:

There's a way in which the car disagreement, it just reiterates the problem of disagreement, right? So you say, Robin, it's me and you, we have these cars and we disagree. I tend to think I'm going to win because I have all these grounds for thinking I'm going to win. I don't know yet what you think. I don't know that you think you're going to win yet, right? I haven't found out anything about you. I just have, you know, my car, I've tuned up my car really well. And then I hear you say you're going to win, right? So now, now we have, we don't have the analogy until I hear that part. Okay, so let's go there. I hear you say, now we have a disagreement. Beri, I'm going to confess something to you, which is that you're here partly as a therapist, because Robin and I have, it's like a Freudian repetition. We keep repeating this scene where Robin says to me, hey, there's two people, they discovered they disagree about something now, what should they do? And I say, they should talk to each other. And you say, no, no, no, I don't mean, That's not what I mean. I mean, how should they adjust their credences in the light of discovering that they have these different opinions? And I say, there isn't a thing you can do there. You can't exercise agency over your credences. You can exercise agency over who you talk to, whether or not you have conversations. So that's a decision someone can make. The other one isn't. Anyway, we've been through this territory. And I was just pointing out to Beri that he's partly coming in to adjudicate this kind of this kind of wall of opacity between us where we struggle to come to an understanding about this very question.

Beri:

Adjudicate. Wow. I mean, I think that I see things the way, Agnes, you described them. If I find that someone disagrees, then I'd like to have a chat with them. I mean, I find it really striking that the standard descriptions in the literature, the sort of examples that motivate the literature are ones where there is not much conversation. I mean, the most striking example is Tom Kelly's example of the conjecture.

Agnes:

Yeah, why don't you say that? I thought that was a great example in the paper.

Beri:

So we are, I don't know how familiar people are, but we're supposed to imagine that you prove this mathematical conjecture that's long been of interest. You actually prove it And then you go and you show it to others, and they disagree. And you show it to the first person, and then you show it to the next person. And you always make sure that they don't talk to each other so that each time when they receive the proof that it's an independent judgment that they have, and all of them disagree. And Kelly concludes, well, you should really lower your confidence. And I gather that this is the example that moved him to change his own view in the disagreement literature. But I find the example really bizarre because you prove this thing and you show it to others and they just say, no, I disagree, but they don't give you any reasons. And I mean, my line in the paper is they're very terse naysayers, you know, Imagine you actually have an argument, you're excited about it, it's supposed to be a momentous argument, and you go and you show it to your colleagues. People will not respond just by saying yes or no. People will give some sort of reasons, especially if they disagree. They will say, well, look, it's wrong here, or this is the bit that I don't understand, or here's the mistake you make. And so by excluding the bit of conversation, I feel like the standard examples completely mischaracterize the phenomenon.

Robin:

It seems to me this literature, the standard way it represents the question, is general enough to include all cases of prior conversations. They don't require that no one has ever talked before. They just require there's a point in time where you ask, what do you think? There can have been arbitrary amounts of conversation before that point. Nevertheless, they think the question is relevant. What should you think at this point in time? You're claiming that they are assuming nobody talks. And I say, I don't see that at all. The general question doesn't make that assumption.

Beri:

So I write up a proof and I show it to my colleagues and they express disagreement, but they don't give me reasons. I think it's very important that they don't give me reasons, because it's the mere fact of disagreement that's supposed to bear on the question of how I adjust my credences. It's not the reasons that are given in disagreement.

Robin:

I don't see that. I see the question is, you know a bunch of reasons. You've heard some reasons. Nevertheless, you still need to choose. I don't see the literature as assuming that you don't know of any reasons, or don't know of any of their reasons, or haven't talked. You can have talked and you can know of many reasons. All that's required is that you don't know everything. You don't know everything they know or everything they've thought. You haven't thought through every possible combination of things you've said and things they've said. There still remains thinking you could do. And therefore, you have to ask, what do I think at this moment, given that there's a bunch of things I don't know yet?

Agnes:

Right, but that means that there is some thinking that you want to do independently of conversing right so the conversation ends and they're still thinking left to do and.

Robin:

You might have an assume the conversation ends, we can ask this question in the middle of a conversation.

Agnes:

Right, but the point is there's some thinking that you want to do independently of talking.

Robin:

Why do we assume that? I don't understand. Where does that assumption come from?

Agnes:

Otherwise, the resolution should just be you keep talking to them until you guys both agree. So think about the way that you played the race. So the race is supposed to be one of the two of you wins, and that's supposed to be the analog to who's right. But in a conversation where you keep talking until you convince the person, there is no one winner. You're both winners, because at the end, you both have the right answer. Nobody lost, right? And as Socrates would say, you're more the winner if you were the one who got convinced around it the other way, because then look what you got. You got to get rid of a falsehood. But then you're both winners if the conversation persists all the way through to the change of mind. Your framing requires that we stop the conversation before we stop the thinking.

Robin:

There could be situations where the conversation reaches a resolution and then you don't disagree. The question is about when you do. So it certainly seems a relevant question to ask, what happens when you do disagree? What should you think then, even if it's possible to work through a process that might lead to the disagreement going away, when it hasn't happened yet? And as we know, it typically doesn't. Most people disagree on most things, and they don't, in fact, resolve those disagreements through arguments. So it seems like a valid question what to think in the usual case where you haven't come to an agreement.

Beri:

Well, you should talk some more.

Robin:

But what should you think now?

Agnes:

What should you think when you don't know what to think? There might not be a great answer to that question. You might need help by talking to someone. And you're right. The talking may not work. And so it may be that you're not brought to an ideal state no matter what. But it doesn't follow that there's some ideal thing that you should think. Like if you can't even resolve it by talking to them, how much the less so are you going to be able to resolve it by yourself?

Robin:

I guess I'm confused about whether you guys sort of accept the usual decision theory framing that when you make choices, you have to have beliefs and preferences of some sort to input into those decisions. And so we often face decisions constantly to make. And the usual decision theory framing says that You need to rely on your best estimate beliefs at those points in time, and so your life demands that you have opinions on things at inconvenient times. And the question often of rationality of belief is, what are the most appropriate beliefs to have? Is that not a standard framing in your mind?

Agnes:

I don't think we can choose what beliefs to have or take actions in relation to those beliefs, except by having conversations with people. So I don't ask myself the question, what beliefs should I have? I don't exercise voluntary control over my beliefs. That's my answer. But let me let Beri answer.

Beri:

I wasn't sure if the idea of voluntary control is part of the decision theoretic framework. I mean, this theory is a kind of idealization that doesn't have that much to do with how we actually think. But I presume that that's what you would accept as well, right?

Agnes:

Yeah, but I think Robin is kind of assimilating the question, what should I believe to the sort of the question, what should I do?

Beri:

Because I choose what to do, and I don't choose what to believe.

Agnes:

Right. But I mean, look, I think we can still ask, I mean, I don't know, suppose we want to praise one of these various beliefs. We want to say that's the good one for them to have. Whether or not the person can choose to have it, whether or not they can come to it by any means. Maybe we can frame Robin's question that way. Like, which belief am I more epistemically praiseworthy for having arrived at by whatever means? the one that, uh, the one that includes the, you know, that, that, that, that downgrades my belief on the basis of the fact that other people believe it or the one that doesn't.

Beri:

I mean, you've, you've introduced the idea of epistemic praiseworthy, you know, you might praise beliefs for all sorts of reasons, uh, as an example. So, so my teenager has this belief that, that smoking pot does incredibly bad for, uh, for health and, you know, We have been known to praise the belief. And perhaps for good decision theoretic reasons, even, have we done the praising. But praising is something you do. Now, the way you put it is you praise it on epistemic grounds. But then now we're back to the question, what is the epistemic assessment? Or what is the truth conducive assessment?

Agnes:

Yeah, but I agree that that gets at what Robin was asking about, but it's helpful to me just because when we say, what should you believe? I'm very inclined to hear that as, what should you decide to believe? And then I'm like, well, there is nothing you can decide to believe. So if we just say, what's the most epistemically praiseworthy thing you could believe, regardless of whether you can come to bring yourself to believe it, then I feel a better grip on the question.

Robin:

So if I may, in your paper, it seems that you put substantial weight on a certain sort of argument about which kind of evidence is relevant in what situation. I might have framed that as which evidence you choose to consider. If Agnes says, no, you can't choose what evidence you think is relevant, it still seems like in your paper you are accepting the idea that we can talk about which evidence is relevant or not in some situations. Is that at least a premise that we can start with? Right.

Beri:

Is it valid to discuss which evidence should be considered relevant for which calculations? Right. Which considerations bear on the question of what is the case? So let's say we disagree about free will. Which considerations bear on the question of whether we actually enjoy free will? Now, in the paper, Steve and I argue that the fact that you think we do have free will and the fact that I think we don't have free will, this is just an example. The facts about what each of us thinks don't bear on the question of what is true. They do bear on the question of whether to continue the conversation, but that's a different question.

Robin:

So that's a key claim I just wasn't persuaded by. So it's, I mean, it seems to me quite literally, it is relevant evidence. And so how is it that it's declared irrelevant? I'm puzzled by that.

Beri:

Right. Well, that's where the argument of the paper comes, which I feel like we haven't gotten to, which is that You know, we relate to other subjects as subjects and not as objects. And that means that the things people say to us, the import of what people say to us is not an evidential import, but it's a more interpersonal import. And one way to see this is if you just think about testimony, the fact that someone tells you something. Of course, you can treat that fact as evidence. But at least in some philosophical circles, typically the ones I hang out in, people think, no, the fact that someone tells you something is different in kind from evidence. It's a kind of assurance, or it's an invitation to trust, or something like that. They give you their word, and someone's giving you their word is a different kind of thing than the production of evidence. And then the idea in this paper here is that the point generalizes. When someone argues with you in a conversation, then their arguments are also not generations of evidence, but their arguments are presentations of reasons that you then should assess. And the right way to respond to a presentation of reasons in argument is not to the fact that it is a presentation of reasons, but the right response is to respond to the reasons.

Agnes:

Can I just add one thing from the paper that I really liked, which is that you say, it's not rational for me, when I'm thinking about whether something is true, to over and above my reasons for thinking it's true, also say, well, I formed the belief that it's true. So that's evidence of its truth.

Beri:

Right. I think that it's true.

Agnes:

I mean- Hold on, hold on. Let me, hold on, Robin. Just let me finish what I'm saying. And because, and that would be, that would typically be seen as a kind of, a kind of bootstrapping that you're not allowed to do. You're not allowed to add the fact that you believe something as a consideration in favor of believing it for yourself. And so similarly, you might think that also applies second personally, that you should treat another as you would treat yourself. And so you should engage with their first order reasons for believing the things that they believe. And not to use what Beri calls the psychological evidence, which is like the fact that they believe it. OK.

Robin:

So I'm confused here, and I'll try to ask questions to clarify. seems to me like whatever evidence is relevant, it's fair game to introduce. I think the fact, when you make an argument and say, here's my argument, I think the fact that you did make the argument is already implicit in the fact that you're making them. So you don't need to say it extra, but I think it is there and it is relevant and people do take it into account. But let's go back to the subjective objective thing. You say, if we're having a subjective relationship, It's inappropriate for me to introduce objective facts about either one of us into the conversation. So at the moment, I shouldn't talk about that you have brown hair or that you're a man or that you paused half a second before you said something. None of those are things I should mention. There's some sort of forbidden wall against such things. I'm confused about what the principle here is exactly.

Beri:

Well, I wish I knew what the principle was. I would be a better philosopher and a better person if I could give you a principle. I love principles. I just don't have one. But the flavor of the distinction is the distinction between what Strawson calls the participant stance and the objective stance, or the participant attitude and the objective attitude. So I think that there are certain considerations that we look upon, but only when engaging with another subject who is mature and capable of a kind of reciprocal engagement. And I guess the thought of the paper is that what Strossen says about the participant and the objective attitudes bears not only on his topics, which is resentment and gratitude, but also on the topic of disagreement. Maybe for those who are not familiar, one way to explain it is, suppose I find flowers on my doorstep. It makes a difference if someone brought them there or if the wind blew them there. In both cases, I might be happy that I have these beautiful flowers, but if someone brought them, it might be also fitting to be grateful. And gratitude is not the right kind of response to the wind. no matter how good the outcome of the wind is. And conversely, if there is damage in my apartment that's due to the wind, then I might be upset. But it doesn't make sense to resent the wind. But it doesn't make sense to resent someone who damaged my apartment. And then the thought is, well, what holds for gratitude and resentment also holds for reasoning. When someone presents me with reasons, the way that I think about this is in a different way than, there's a different register I can think about it. Namely, when someone presents me with reasons, I can reason with them. And I can't reason with the wind. I can't reason with the thermometer, but I can reason with another person. And disagreement is a moment in reasoning. It's something that happens as we reason with each other and find that we are drawn to different reasons. And so how do you respond to disagreement? Well, with reasoning, with more reasoning. And that's why I think Agnes earlier was pointing to the importance of continuing the conversation. And I think that that's something we only have as an option in response to other people who are in a position to reciprocate, so other people who are mature and so on. Now, is there a forbidden wall between people and non-people? No, that's not how I would put it. But I think that there's a very deep difference between persons you can reason with and the natural world, which can give you lots of information. very often better information that the people you're reasoning with, but there's a completely different way you can engage with people. And that's what I was concerned to articulate in this paper.

Robin:

So, but you can also look at people as physical objects.

Beri:

Yes.

Robin:

And you can use that in your reasoning about that. Is that forbidden or illicit somehow? Is it appropriate to look at people objectively as physical objects and as things that behaved in the world and use that as evidence in your discussions?

Beri:

You can, and that's great. And that's very important to me and to Steve as well, to us. I mean, the paper is called Disagreement and Alienation. And this is where you get to the alienation bit. I think that when you do that, when you take the objective attitude towards others in Strauss's terms, that is to say, when you look to the fact that someone disagrees with you as a psychological fact, and you take that fact into account to figure out what's true, then you are you're alienated from the other. There's a way in which this alienation precludes shared reasoning.

Robin:

Is it the same kind of alienation as talking about the color of your hair? I don't understand. What is it about mentioning the fact that you had an opinion that alienates us when discussing other things about your physical body or status don't alienate you? Where is the concept of alienation?

Beri:

So the fact that I believe that P is not a reason for which I can believe that P. It's a symmetric reason for both of you.

Robin:

It might not be a strong reason might be a reason you've already assimilated, but it is a reason.

Agnes:

So Robin, you're rejecting the bit of Beri's paper that I said I liked, where I thought it's obvious that that bootstrapping reason is inappropriate. You're just rejecting that claim. So you think when I'm like, why is 2 plus 2 4? Well, one reason is when I take two things and two things and put them together, I get 4. Another is that I believe that 2 plus 2 is 4. That's some evidence for the truth of 2 plus 2 is 4. And many people would think, no, that's not actually evidence for the truth of the belief. That's inappropriate bootstrapping. I don't get additional evidence by noticing what I believe, but you think I do get it that way.

Robin:

Yes, I think the fact that you had some thoughts, and that they came to a conclusion that's a fact about your life and that fact is evidence about not only other aspects of your life but about the things you thought about. And that's the whole point of this sort of analysis.

Agnes:

One way to translate this, Beri, is that one of your arguments, the argument that I found so persuasive, is you leverage off of our non-alienation from ourselves to say we should have that same non-alienated approach to other people. Robin's saying, I have that alienated approach to myself, so I'm fine. I can have the same approach. It can be symmetric. If you don't think it's alienated to take your own beliefs as evidence in that way, then you're also not going to see it as alienated to do the same thing to other people.

Beri:

I think that's right. Look, maybe alienation is pervasive. And I do feel like some people are really unmoved by these concerns about alienation. For some people, that's not a topic at all. But

Robin:

Say, say there's some issue in your life with respect to your marriage or something else. And you know that for the last few months you've been repeatedly thinking this issue through and coming to the same conclusion. And now you're asking yourself, should I think it through it again, it might, it's completely reasonable evidence to say. I did this 10 times already. I went through with the whole thing. I came to the same conclusion. At this point, I am justified in not thinking it through again and finally acting on this conclusion. That's a way we reason. It seems completely valid and useful.

Agnes:

That's a good example.

Robin:

we do this often, right? We have a difficult decision and we repeatedly go through it, right? And at some point we say, I've gone through it enough.

Agnes:

At some point we kind of give up and we kind of accept a certain kind of alienation from our, that's part of life, I think is right. And we give up on conversations with other people as well. But there's still something odd to me about framing your theory of both of rationality and of disagreement and placing at the very center of the theory these moments of giving up.

Robin:

When you call giving up, I mean like not having a complete answer to everything, which I think is the usual default state. I don't think it is a strange, sad state. I mean, you very rarely can just completely settle a question by thinking it through.

Agnes:

What I'm saying is you want to pose, you want to really sharpen this question, what should you think? But then you also want to frame that in terms of in a situation where you've given up and you don't know what to think. But what exactly should you think then? Maybe there is no answer to what exactly should you think then.

Beri:

Could you walk me through this example again. I failed to see how it connects with the thing I had in mind.

Agnes:

Can I state the example in my terms and you tell me whether I got it right, Robin? So it's basically like you could take the fact, say there's a problem in your life. You could take the fact that you have thought, tried to think through this problem a bunch of times and failed to come to a conclusion about it as evidence that you're probably not going to succeed if you try again. So it's psychological evidence about yourself.

Robin:

right and then decide on that basis say simply to act or whatever so i was thinking that you came to tentative conclusions each time but each time you didn't think you had a complete answer you weren't sure of the answer but but you had the same tentative conclusion each time and then you might decide to go with that tentative conclusion given that you tried a dozen times and that's as best you could do each time

Beri:

So I find it a little distracting that it's an example of decision making. Is that essential to it?

Robin:

So that I do think at some point you need to make a decision and therefore you need an opinion at some point, even though you might be able to go on for decades further thinking about it. But that is a common framing of belief.

Beri:

A decision is a practical conclusion. You make a decision in light of what is good. Whereas the topic here is belief. And we don't believe in light of what is good. We believe in light of what is true. And so I think that those are quite different. Maybe that's one reservation I have with decision theory, in that, at least on certain formulations, when you talk about betting behavior, it runs together, the practical and the theoretical. But would you be happy with a description of the example without the decision part, just in terms of credences, let's say?

Robin:

as long as you felt some pressure to have an opinion at some point. If you're going to deny that you ever need an opinion until you completely settle everything, then that's the point I resist. I don't need it to be connected to a decision, but I need you to feel some urgency of some sort to have an opinion at the moment.

Beri:

So let's say I go through the arguments against free will. And I go through, and I'm like, no, they just don't seem right. But then I go through them again and again, and then I'm like, well, no, they just don't seem right. And I go through them again and come to the view that they don't seem right. And now what? Can I be done?

Robin:

Now you can say they don't seem right. say that that's your opinion. And if you're asked about it in a conference or paper or conversation, you can say, this is my opinion. If somebody says, why do you think this? You could go through your specific reasons, but you could also just say, I tried this a dozen times in different parts of my life, thinking sort of different disciplinary backgrounds with different other conversation partners. And each time I came to the same conclusion. tentative as it was.

Beri:

I feel like we're getting somewhere. This is helpful. I agree. Once you've gone through it, you can say, yeah, it doesn't seem right. Now, someone asked you, why do you think this? I think there are two ways to hear this question. One is as a question for reasons. Why do you think this is true? And then I can't talk about myself. I have to give them the things I went over in my deliberation. Or you can hear the question, why do you think this is right, as a question about my biography, but that's a very different question. And that's largely uninteresting.

Robin:

Let me make an example to legal reasoning. So, if a detective goes out in the world and, you know, investigates a crime and collects evidence. they face this constraint that when they get to court, there's only a limited kind of evidence they're allowed to present. They can't present all possible evidence that they actually collected, like their emotional impressions of people when they met them and things like that. We have constraints on rules of evidence in court, and we say, for the detective to make their case in court, they'll have to limit their evidence to a certain set of categories. And then the detective, even once they're convinced they know who's guilty, they will keep working until they collect the evidence that's actually acceptable in court. And it sounds like you're making a similar sort of constraint here, you're saying, in conversation, there's a set of evidence to bring up and it's more limited than the set of all evidence I have available to me and our norms of conversation require that I limit my argument to that kind of evidence and so. That's what I'll have to do. There's some evidence that I have accessible to me that I, it is forbidden or inappropriate to consider in our, in my argument with you. And so I'm going to do the extra work of trying to find an argument that doesn't refer to the other things I know that I'm not allowed to mention. And we often do that extra work and that's how conversation goes.

Beri:

I guess the notion I wanted to use is not that it's forbidden, but that it gets bracketed. It does get set aside. Here I was drawing on the disagreement literature by Christensen. I think that's the sort of thing we do when we think we, we, we, we recognize certain things as considerations that bear on the question but we set them aside we don't take them to be significant.

Robin:

What could it just be for the same reason in court, some kinds of evidence is much more problematic in terms of being able to trust it and rely on it and what goes wrong if we give people incentives to be trusted. And so, even in our intellectual conversations we require people to limit their evidence to a certain sort of trustworthy more. robust things that will better withstand resistance and questioning, and less impugn people's intentions or morals, etc. We limit conversation for those purposes, and let's just accept that as a matter of ordinary conversation norms. I'm happy to do that, as long as we're clear. We're saying in a debate, you're only allowed to bring up certain sorts of things, and let's just have a debate and only bring those things. And we each know that this isn't all the evidence we have available to us, but we know that there's some value in having this kind of a structured interaction.

Beri:

I can't resist the thought that it's somehow disallowed, but I do find this conversation actually very helpful in that it clarifies where sort of some of the sticking points are. I mean, let me try to explain. So suppose I mean, this is an example from my first book. Suppose someone comes to me. I asked my readers to imagine a wayward lover, someone who has cheated on you. And they come to you and say, I'm sorry, it won't happen again. And they give you their word. And now my question is, in order for you to be rational to believe them, does it have to be the case? that you have to have conclusive evidence that what they're saying is true. And I argue that the answer is no, because when they tell you, believe me, I won't do it again, what they're doing is they're speaking to you, they're giving you their word, and the significance of their word is It's an epistemic reason. It's a reason that shows something to be true, but it's a non-evidential epistemic reason. And so you can accept that non-evidential epistemic reason. And in accepting it, you set aside the evidence. Now, the evidence from their past behavior and from statistics and so on, it is evidence. In a way, it bears on the question. But its bearing on the question precludes you from treating the reason they gave you when they gave you their word as the kind of reason that it is, which is a kind of interpersonal reason. So it's not that you're prohibited from using the evidence. engaging with them interpersonally, taking them at their word is inconsistent with engaging with them as an object, as an object of prediction where you would look to all the evidence. And so as it is in that example, I hope it was sort of clear. So it is in shared reasoning. When someone reasons with you, then it's not that you're prohibited from looking at the fact that they make a certain judgment, but it's that the practice of reasoning brings with it this need to bracket the psychological evidence. So this is not a matter of prohibitions or some prohibited wall. It's also not a matter of just conversational practice. I think it is a matter of what it is to reason with one another. What it is to reason with one another is to take certain considerations to be relevant in a certain way and not take other considerations to be relevant in another way.

Agnes:

Can I, I want to add something, which I can't tell whether it is of a piece with or intention with what you just said, Beri, because I would have given like a much, I'm realizing it would even much simpler answer, which is just that the psychological evidence is not a potential explanation of why the thing is true. So if we say, why is it true that 2 plus 2 is 4? You don't come to any understanding of why 2 plus 2 is 4 if you learn that I believe it. I don't come to any understanding, you don't either. And so it may be that these norms, where they're really coming from, is that we want to be introducing, like maybe Robin's right that on some level, things that don't give you any understanding could still constitute reasons for belief. But in the ideal case, which is what we're in when we're in a conversation, we want to introduce the sorts of reasons for belief that could also lead us to knowledge or that could be sources of understanding. And that's what the psychological evidence isn't.

Beri:

I just worry that this would preclude too much. So sometimes I can just give you some information and tell you that P. And that I told you that P is a reason to believe P, but it doesn't explain why P is true.

Agnes:

Oh yeah, I don't really think it's true. So we disagree about all the Moran stuff and telling people stuff. I agree with you about the interpersonal, but I wouldn't base it in any way on Strassen or on Moran or on telling. For me, all of that would be grounded in there's stuff you want to know and other people can help you know it.

Robin:

So that's a different- Before we get too far away from, you have a concrete example here of the person who come and says, I won't do this again. I think in real conversations, people often have said, you've done that three times before. Why should I believe you this time? It's a completely valid part of a conversation there to mention that fact, isn't it? It is. So how is it bracketed?

Beri:

Well, I'm not saying that whenever someone promises you to do something and they have a bad track record, you should believe them. But you might, in the conversation, bring it up.

Robin:

This isn't about your final conclusion. It's about a conversational move. You're saying things are bracketed there, so they're not supposed to be part of the conversation. And I'm saying, this sounds like a valid part of the conversation.

Beri:

Right. I agree with that. My thought is, if you actually were to take them at their word, and whether you can, I think, is a difficult question. Life is difficult. But if you were to take them at their word, you would be bracketing certain considerations. And that's the notion of bracketing that I was operating with rather than the idea of something impermissible.

Robin:

Say you said, OK, I want a penalty this time. If you do it again, you're going to suffer the following penalty. I want you to sign a contract like that. So that would be an explicit conversation where I am not fully believing them, but believing them enough given that they signed this contract. Is that in, have I bracketed the consideration there or not? No, now you've factored them in. But that's allowed, right? So now bracketing individual things or choices we make about each topic in a conversation, we say, oh, I want to bracket that. You have a good argument. Let me just bracket it so we don't consider it. Is that a loud conversational mood? Every time you have a good argument, I get to bracket it away?

Agnes:

Beri wasn't saying that this is the only possible response to the person who comes to you. He was just saying it's a possible response. that something that might happen, something that might happen is you might decide to bracket those things. Not that you will always decide to do it and if people are okay by contract they won't, but that's not how most people in fact conduct their relationship.

Robin:

Okay, but if we go back to the fundamental rationality of discommitment literature, it sounds like they say, you know, in this situation where you've had different, you know, you realize you have differing opinions, we wonder what to think, And you come in and say, oh, sure, if you were going to consider all actual relevant evidence, that might be the problem. But let me just notice there are situations in the world where we decide not to consider some evidence. And in those situations, of course, not all this evidence is relevant, and therefore, you wouldn't consider it. So you're just saying, look, we sometimes decide to bracket evidence, to set it aside for some other, yes, not fully explained social reason. There are contexts in which we don't consider all possible evidence in our conclusions. I mean, I think they would accept that. I mean, they said, sure, fine. If you're not considering all possible evidence, yes, of course you could draw different conclusions, but they usually are asking what would all relevant evidence indicate. That is, I think, the usual framing of the literature.

Beri:

Well, the argument is that we can bracket without irrationality. And, you know, even, I mean, partly I use this dialectically, even Christensen speaks of bracketing. Of course, Kelly then replies, no, it's going to be all of the evidence. I'm with Kelly. Now we can reason further, but maybe we're still stuck on this point. The point I wanted to make about the bracketing is to explain the way in which I envisioned this idea that we're setting something aside. It wasn't some sort of forbidden wall. I was basically provoked by your metaphor of a forbidden wall. And I tried to suggest why that metaphor is somewhat misleading.

Robin:

People often voluntarily accept limits and walls. People draw dotted chalk lines, and people don't cross that. It doesn't have to be a physical prevention. There's some line that you don't want to cross.

Beri:

It's not a matter of choice. I mean, the point of the example was to explain a certain structure of thought. When I think about a certain consideration in a certain way, like the fact that someone's made me a promise, then in thinking about the consideration in this way, that precludes me from thinking about something else in a different way. And so that other thing has to be bracketed. I haven't told you whether I have to respond to promises as promises. I certainly didn't mean to suggest that every time someone promises us, we should believe them. That would be ludicrous.

Robin:

Can I summarize bracketing as implicit conditionals? Because conditionals are a very familiar concept, that if we say, this following discussion is conditional on these assumptions, and we're just going to go forward with them, can I understand them that way? Certain conversations have implicit conditionals. Like, if you say something sincerely, I'm going to believe you. Maybe that's an implicit conditional of some kinds of personal conversations. Or the logical argument is conditional on, you know, we're all aware of the arguments, etc.

Beri:

That wasn't quite how I had envisioned it. I mean, my thought was, if I'm going to take someone at their word, then I shouldn't be predicting whether what they're saying is true. Because if I'm going to take them at their word, then that precludes.

Robin:

Right. So I could say prediction, I'm taking you at your word, that's an implicit conditional, I'm saying, you know, in this conversation, I am taking you at your word that we can just accept that the norm of many kinds of conversations are certain sort of conditional. or even you're not intentionally lying. That's often an implicit conditional of many conversations. We, unless we have substantial evidence of the contrary, we assume that people sincerely mean each thing they say. And we're reluctant to break that wall or at some point accuse someone of just consciously lying. That's a sharp boundary in a conversation when someone brings up such an accusation.

Agnes:

I mean, is this the same kind of idea as like, we, all of us right now, are talking about this question of disagreement. It wouldn't be okay, as tempted as I am to do it, for me to derail this conversation onto Emile Zola, who I'm currently obsessed with. And you'd be like, no, no, no, we're bracketing that, we're setting that aside. There's an implicit conditional inside this conversation that what we're talking about is disagreement, not your obsession with Zola. So like conversations of all kinds of, Constraints of that kind. And so you're asking Beri whether the I'm not going to use psychological evidence is a similar principle to we're not going to randomly start talking about Zola.

Robin:

similar in being a norm, but we might say there's norms about what conditions we're assuming. It's just a more smaller set of norms than all possible norms of conversation.

Agnes:

We couldn't inquire into anything unless we could organize each other in some way. Of course.

Beri:

it seems fine to me to put it that way. I mean, can I give you an example of the rationale for this, as it were, norm? I'm not sure if norm is the word I would go for. But suppose, you know, I read the standard example in our literature is David Lewis's paper. So I'll go with that. I read David Lewis's paper, and it presents some arguments. And then I think Wow, those arguments are good. But I stopped short from adopting the conclusion because I think, well, I should check whether David Lewis is an expert. I should check, you know, is he really in a position to make those arguments?

Agnes:

How famous is he? Pardon?

Beri:

How famous is he? Where do you go? I mean, to put it really polemically, where do you go to school? What was his SAT score? And I think that that is not relevant. The argument speaks for itself, and the person drops out. And if I do look to facts about the person, there's something I have misunderstood about the contribution. My Hegelian friends have said things like, the science speaks through me. I wouldn't dare. However, I think that there is a, there is a, I feel like I see something in this. When I give you arguments, they're not about me, they're about the thing that I'm talking about. And so if someone's going to respond to my arguments correctly, they shouldn't think about me and where I went to school and the fact that I'm making a judgment, they should just think about the arguments. Now, if you want to call that a norm of inquiry, so be it.

Robin:

An analogy came to mind. Recently I read an op ed about UFOs by a scientist. And the scientist said, there's all this things people keep calling evidence reports and you know. memories and things like that but they're not really scientific evidence and so we scientists should only consider scientific evidence here and scientific evidence needs to be set up with well calibrated instruments that have been preset to look at particular places and with respect to scientific evidence there's nothing here so scientists shouldn't believe in UFOs because scientific evidence hasn't been presented. They have a norm of in their conversation there's only certain kinds of evidence that are relevant and they shouldn't consider this other evidence that say a court might consider as evidence for their conversation. So it's an example of yet another sphere of conversation where people have norms of which evidence is allowed and which not and that seems analogous to your philosophy norm of only give the reasons, not sort of the meta evidence of who thought of which reasons, who supported which reasons.

Beri:

I think the rationale is very different. I've tried to explain the rationale, but Yeah, I see the analogy.

Agnes:

I have two questions I want to ask Beri that are sort of about the details of the way that he formulates his own view. So the first is, why don't we need to do even more bracketing than you're saying? Take the math case, right? So suppose that the other mathematicians are like, Well, the reason why I think your proof is wrong is that I've come up with a proof for the opposite or something. That's not irrelevant and it's not an inappropriate response, but it's not really the ideal response. The ideal response would be, where did I go wrong in my proof? And that really is what disagreement is about. And so disagreement actually requires everybody to bracket the fact of what everybody thinks, but also a structure where at least one person in a given moment is actually willing to bracket their positive reasons for believing their view in order to examine why the other person is going wrong. Why aren't our standards even higher than you say they are? There's something disrespectful if you give a case for P and I'm like, well, here's the case for not P. And I don't say anything about your case for P. I agree with that.

Beri:

I could. I could go with that. It does seem to me, if I have a proof and you come and you say, well, I have a proof to the contrary, I should hold off till I've seen yours. Now I know that there is something which, you know, I know there are considerations available that bear on the question that I should consider.

Robin:

But quite often say physicists won't even consider an argument for a perpetual motion machine because they say, look, basic physics says perpetual motion machines are impossible, so I'm just not going to look at your proposed proof. That is quite often we do just reject things and not consider the details of the proofs exactly because we think we see strong contrary arguments.

Agnes:

those might just not be the ideal cases of disagreement. So I'm saying, let's articulate the ideal case. We're not always, Beri's happy to acknowledge we're not always in the ideal case, but I think the ideal case requires more than, even stronger constraints, even stronger norms than, like in terms of what's excluded, a lot more is actually excluded than just the psychological states.

Beri:

I was also following the accepted premise of the dialectic that we're talking to epistemic peers. I don't have a story for a disagreement with non-peers, but presumably in your perpetual motion example, it's not going to be peers.

Robin:

Usually. But sometimes peers do make such arguments.

Beri:

Right.

Agnes:

If a person of sufficient status made the argument, I think people would listen. So it wouldn't be somehow impossible to get someone to listen to the argument. It would just be really unlikely.

Robin:

And that highlights often one of the implicit constraints here is status. That is. Sure. you know, we interact status with other things and say, unless you're sufficiently high status, then you're not allowed to make certain moves.

Agnes:

But we often don't get to the ideal case, except in the case where someone has status. But that's just a way to use status to help us articulate the ideal case. Wait, I want to ask Beri my other question. So throughout, like the sort of end part of your paper, you make an assumption that I wondered about, which is that sort of in the disagreement, in the act of disagreement, people are suspending judgment. And you take the fact of disagreement as a reason to suspend judgment. And I guess I wasn't sure why you thought this, and it didn't quite seem right to me. So I've argued about disagreement with Robin a lot of times now. We disagree about disagreement. And I don't feel like I'm suspending judgment when I'm arguing with him. I'm like, no, I'm right, and Robin's wrong, and he's missing this fact, this fundamental fact that's accessible from the first-person perspective itself. and I keep trying to fight it out with him. But I also think I'm like woefully unable to counter certain claims of Robbins. And that makes me feel like I need to come back and argue with him with this again. But I don't ever feel like I suspend judgment at any point. In fact, it's precisely because I don't suspend judgment that I'm in a position to defend my views as someone who like avows those views or thinks that they're true.

Beri:

The rationale was that the ideal is that we end up with something that we can describe as our view.

Agnes:

Yeah, absolutely. I agree with that. At some point, one of us will change their mind, or both of us will.

Beri:

Yeah, I see.

Agnes:

But I don't see why.

Beri:

You're holding out for the other to change their minds? I just feel like.

Agnes:

Or you'll change your mind. But when you do, it will be a surprise. It won't be like, I was changing it all along.

Beri:

Well, but the reason that. we wanted to characterize this as a kind of suspension is that in actually disagreeing with someone, I keep an open mind where we'll end up. Because if I didn't, if I already knew where we'd end up, then we're not really talking.

Robin:

But suspension of belief doesn't have to equal open mind, right? You can have an open mind even though you have a tentative judgment.

Beri:

That seems fine with me. So part of the difficulty is that I prefer to talk in terms of believing something where that means asserting that it is true and suspending judgment where that means neither asserting that it is true nor asserting that it is false. So suspension of judgment is compatible with holding very high credences.

Agnes:

But it's not compatible with asserting that it's true? Because don't you have to assert that it's true when you're arguing? Aren't I asserting a bunch of things are true? Yes. See, I think keeping an open mind just is disagreeing. It's not a psychological state that's slapped onto the disagreeing. The fact that I'm disagreeing with you, the fact that I'm willing to have a conversation, that's just what it is to have an open mind. There's nothing in my mind that marks that openness. It's the practical aspect that is the openness.

Robin:

But many people participate in debates without having an open mind.

Agnes:

Hold on. OK, Beri, you go first.

Beri:

Suppose I have my proof, yeah, and I'm about to show you. And you say, no, no, I disagree on step 13. I feel like as you're talking, I can't be convinced that what you're going to say is going to be false because I got to hear you out. But if I'm not convinced, But if I believe that I have a proof, if I entirely believe my proof, if I don't have some level of suspension, then I would have to be convinced that what you're about to say is going to be false. So it seems to me that in order to be receptive to what you're saying, there has to be a way in which I'm open. And this way of being open precludes all-out assertion that, look, don't tell me about premise 13, because I've proven it. Don't even bother. I know you disagree, but don't bother. I've proven it, so it must be true. There can't be a mistake in premise 13. So that would be an unreceptive stance. It would not be a disagreement. It wouldn't be listening. And so I think that disagreement involves a kind of listening. And this listening will require, I think, something that falls short of outright belief. Now, maybe the thing to say, and I think this is a very good point you made, that even in this kind of state of suspension, one can make assertions. So the fact that I suspend doesn't mean that I now no longer am prepared to assert premise 13 or something. I am prepared to assert it in a conversation with you where I want to invite you to tell me what's wrong with it.

Agnes:

What about if we just said you don't claim to know it? So you believe it, therefore you can assert it sincerely, but there's a gulf between what you claim to know and what you merely believe. And if you really, this is Kripke's paradox of dogmatism, if you really thought you knew it, why would you care what anyone else said? If you really thought you knew it, you wouldn't listen. And so, but isn't that enough that we don't need to, suspension of judgment is way stronger than just not claiming that you know something. So why not just say, yeah, it has to be something you don't claim that you know.

Beri:

Right. So the argument I was giving you is a version of the Kripke's paradox. I guess I like the knowledge norm of assertion. And I like the knowledge norm of belief. But you know, I think that the crucial issue is not here. I think you can change things around and have a weaker norm on belief. And then it could be that I believe but I'm still open minded. But you asked me why we did it that way. I think we did it that way because we wanted to capture the sense of open-mindedness that's required for the kind of reasoning together that we think disagreement is.

Robin:

So we're almost out of time. May we each summarize our position at the end of our conversation? I will say that this usual literature, which is about beliefs and whether you should change your beliefs in response to other things, Agnes basically doesn't think you can choose beliefs. So there's no point in talking about what beliefs you should choose. So she's rejecting literature that way. And I see you, Beri, as rejecting the usual framing that you should consider all the relevant evidence. And you're saying, no, in certain kinds of conversations, they should be bracketed and only limited evidence considered. And that seems valid, but doesn't address the question of the literature, which is, what should you believe when you consider all the evidence? Anyway, that's my summary position.

Beri:

And now you'd like me to summarize as well? Whatever final statements you'd like to make.

Agnes:

You can have the final word. If you want any final word, you can have it.

Beri:

I think you're right that I'm not answering the question as it is posed in the literature. And I think that's what's really hindered the reception of this paper. I think that I'd like to see a refocusing on this idea that disagreement is a moment in shared reasoning. So what we've done for the past hour is we've done a lot of disagreeing in reasoning with each other. We haven't talked about our status or the likelihood that we'd be right. We've talked about the substance. And I think that if the focus, if the problem of peer disagreement were to be understood as a problem about shared reasoning, about what it is to reason with each other, then you'd see that these questions about psychological facts really drop out. In fact, when we turn to the psychological fact that I believe that P and the psychological fact that you don't believe that P, then we are not reasoning with each other. And that's the moment of alienation.

Agnes:

I will say, I think that the more interesting and core questions about rationality are not where the philosophical literature is. So I really agree with Beri that the main problem is we're asking the wrong question, but it doesn't follow that there isn't a question there. It may not be the most interesting question, but there's a question. And I guess I think in response to the question, what should you do in response to the fact of disagreement? Well, Robin's right that mostly I think you can't do anything with your beliefs that you don't have agency over them. But we can ask, OK, but which belief, if you were to land on it, would we epistemically praise you for? And there, I just think there isn't going to be one answer. That is, it's going to just depend on the context. And so, for instance, how seriously should you take the fact that people disagree with you? It might be a function of, I don't know, how much responsibility have you been vested with, and how many lives are at stake? lots of other considerations, and so it might well be that you're epistemically praiseworthy sometimes for ignoring it, sometimes for taking it really seriously, sometimes for adopting the equal weight view, that epistemic praise is going to be tracking a bunch of other factors. Even epistemic praise about the truthfulness is going to be tracking all these other factors. If we really try to narrow down to this question, assuming you can't actually pursue the disagreement with someone, how should you react to it? Yeah, so I guess that's my concluding thought.

Robin:

It was nice to talk to both of you today.

Agnes:

Thank you so much, Beri. Thank you for joining us. Thanks for having me.

Note: transcript generated by audiotranscriber.ai.