Gentlemen's Disagreements
What are 'Gentlemen's agreements'? How binding are they? What happens if they are broken? Why do the hold sometimes, but not others?
For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email podcast@alephinsights.com
Transcript
Hello and welcome to the Cognitive Engineering Podcast produced by me, Fraser McGruer, for Aleph Insights. In this series of podcasts we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Peter Coghill and Nick Hare of Aleph Insights and this week we're discussing gentlemen's agreements.
Speaker B:So Nick, why are we talking about gentlemen's agreements? Well in this particular case, it's inspired by a recent event in the House of Commons, the British, obviously the British Parliament, and one of the votes on Brexit, on the government's Brexit, it was all quite convoluted but there's a, the Parliament needed to both agree to a, agree to the proposal that Theresa May is going to put to the EU and also to vote on some amendments that were put forward at the last minute and it was all very convoluted as usual. But anyway, the thing of interest here is that there is a long-standing convention in the House of Commons where people on, in different parties will agree to be paired so that they can either both be present for a vote or both be absent for a vote, right? In case one of them needs to be absent, the other one simply doesn't turn up and it's just, you know, it means that the vote is not, the outcome of the vote is unaffected and, you know, it's basically a way of both, both sides benefit, you know, if they can both, if they both agree they're not going to be there then they just don't turn up. Anyway, so during one of these votes the, Brandon Lewis, who's the Conservative Chairman, I think, was paired with a Liberal Democrat MP, Jo Swinson, who was on maternity leave and the sly old dog turned up and voted for the, with the government and indeed and the motion, I think, passed but obviously this was judged fairly scandalous and yeah, so it was a situation where, well, there was no real obligation, there's no, certainly no legal obligation for them to do it, so the question is why do they do it anyway? So what's, you know, and what motivates them to break that agreement when they do? I should say, in his defence, that he claimed he got muddled up and, you know, did not, had not realised that it was being applied and, you know. And to be fair to
Speaker C:the British press, I think the Whip's office is getting much more flack than he is individually. I think the attention is on the Whip's office for being, for bullying him into voting against
Speaker A:these unwritten rules. And yeah, he was probably distracted by too many WhatsApp messages and chats and emails and so on, things like that, maybe that's what happened. Okay, but the question isn't, no, why, I think you said the question is why does this exist or why, you know, I think it's clear why it exists, it's just a nice... I think we're interested in why did gentlemen's
Speaker C:agreements exist? And why sort of social norms more generally exist? And why do they stick in some circumstances and then but not in others? Well, it's not clear how it exists, actually.
Speaker B:I mean, it seems like, yeah, it's obviously a good thing. But what's to stop you breaking it? Nothing. So why don't people break it the first chance they get? You know, and then hey,
Speaker C:presto, a free vote, right? Well, there is stuff breaking it because there's value in having a
Speaker A:system like that, right? I mean, I thought also, putting into context of quite a long and stable democracy. And there's the weight of convention can grow during that time. It's very much like someone
Speaker B:saying, well, I really want to give up smoking. But I'll just have this one last cigarette, right? There's always a temptation to break that agreement. Because it's always in your interest to do that.
Speaker C:Like what, what difference would it make? Well, I think, well, I mean, I'm gonna speak for the economist. What is going on? Well, because it really comes down to a game theory situation, doesn't it? So you have, there'll be a certain cost and benefit associated with breaking that social rule, there and then. So the micro benefit might be that your party wins that vote. And the cost will be you are, you're not taking it seriously, you're not trusted as much, but also the social norm which exists and gives you benefit in the longer term is diminished.
Speaker B:Yeah, but whose interest is it to enforce the social norm? That's what I'm saying.
Speaker C:Well, it's a collective thing over a sort of macro period.
Speaker B:Well, this is the part, I mean, this is the puzzle, right? It's puzzling why we, why people do things against their interest. And, you know, so that they continue a social norm, which again is only in the future going to potentially work against them then, you know. Yeah, I mean, you're right. The answer is there is an answer in game theory, which is, you know, looking at how people play in a repeated prisoner's dilemma situation. Now, we haven't, I don't think, bizarrely, I can't remember that we talked about the prisoner's dilemma before. Yeah, not in any detail, no. But it's really universal. It's a universal situation. There's no point, it's so well covered on the internet, people can look it up. But the gist is that you have a situation, the thought experiment is you have a situation where there's two people have been arrested for potentially committing a crime. And, you know, let's say they both actually are actually in fact guilty of that crime. And they can either keep Sturm and not say anything, or they can dob in their mate. Now, if they both keep Sturm, they'll get convicted on some sort of technicality, like carrying a knife. And they'll both go to prison for three months. If they both rat on each other, well, great, they'll both go to prison for 10 years for murder. But if one of them rats on the other one, the one who ratted on his mate gets let off scot-free, right? Whereas his mate gets, let's say, you know, five years or something. So the question you have is, well, should we keep Sturm or rat on our mate now? So here's the two choices. First of all, you say, well, what's he going to do? What if he keeps quiet? He doesn't say anything. What's better for us? Well, if we dob him in, we get off free instead of having to go to prison for three months. So it's better to dob him in if he's going to be quiet. If he's going to rat on me, well, I should rat on him because then we'll both be better off than if I get, you know, sent to prison for... I've explained it wrong. I need to do the whole thing
Speaker C:again. Whoa, okay. Well, given that you said that it's all well covered, I would suggest listeners go and look up game theory on Wikipedia. So the prisoner's dilemma is pretty well covered.
Speaker B:But it's essentially a situation where you can get one over on someone in a particular situation and they can get one over on you, right? They can get an advantage at your expense. But if you both do that, you're both worse off than if neither of you had done that. Now, the problem is, so it's in your individual interest to try and exploit the situation you're in. But if you both do that, you'll both be worse off. And this is a situation which crops up all the time, crops up in everyday life, in social interactions, in business, in the environment, particularly. You know, you could see climate change as the result of lots of people. Nobody wants the earth to be destroyed. But if everyone else is behaving themselves, then you get a little sneaky advantage by polluting a bit more, by not paying to try and clamp down on your own pollution.
Speaker C:The game has changed though when it's repeated, because the prisoner's dilemma is just a one-off event. Yes. And when the game is repeated, and also when the balance between the cost and the benefit are closer, so there's not much difference in it, then you get the state where you can have a Nash equilibrium. And that's the point at which you can get a, where social norms exist. So if you're, if on aggregate, over like your entire lifetime, it's worthwhile taking your shoes off when you go into people's houses, versus not doing that. Because the cost of being ridiculed and being ostracized because you're not taking your shoes off, for example, is outweighed by the benefit of taking your shoes off, then you will just leave your shoes on.
Speaker B: first talked about it in the: Speaker C:numerically demonstrated, there's a lot of numerical evidence that the tit for tat is the optimal.
Speaker B: he, I think I want to say the: Speaker A:Yes. Look, I don't want to, I don't want to, I don't want to delve too much into game theory, because it's discussed so much elsewhere. I just want to bring us back on track a little bit, if you're okay with that. Yeah. I mean, is there any, before we, I want to talk a little bit more about convention and written, unwritten stuff, but before we do. It needs to be explained. Yeah.
Speaker B:Right. But before we do, go on. Yeah, no, it's just that the, he basically did a competition where he got people to submit strategies where they would, you know, see what the other person did, and then they'd play again. And then they could make their decision based on what the
Speaker C:person did before. And it was tournament based, so all these different strategies were little bits of code. They're all fighting each other. They're all fighting each other. And the tit for tat was, was, was strangely the simplest and the best. But it wasn't the simplest. I mean, always, always. Okay. Well, it's one of the, compared to some of them, which had very long memories and did all sorts of clever computation stuff. Tit for tat, basically, you just retaliate if somebody, if somebody does the bad thing, but then you instantly forgive them. Yeah. And you, if as long as they cooperate, you keep cooperating. Yeah, that makes sense. So, yeah. And that's what
Speaker B:this is. Well, that brings us right back round to convention. One of the examples. Convention develops effectively. One of the examples he talks about is the sort of development of those of emergent ceasefires in the First World War. So the necessary conditions are, you need to play against the same people over and over again. If you randomise it, it doesn't work. Yeah. So you've got to be playing against the same people. Well, this parliamentary pairing situation is obviously that. You know, you can verify who they are. You can verify what they've done because voting records aren't secret. So, you know, and, and, and that's where you get these, these conventions are allowed to emerge. They can emerge in situations where you have those conditions. Okay. And also, I mean,
Speaker A:and nothing needs to be written down, right? It's sort of learned on a sort of almost a corporate
Speaker B:level or systemic level. No, it needs to, it does, you need to be able to verify. Does it need to be the same person? And you have to know what they did. If you don't know what they did, then you've got a whole new, more complex class, class of problem. Okay. So can we extend this out
Speaker C:into other areas on a society level, for example? Yeah. So I, the gentleman's agreement is a form of social norms, sort of a rule that exists, but it doesn't, isn't written down. It's not a piece of policies, not a piece of law. It's a thing that a convention that, that people of a certain group subscribe to because generally it's good for the group. So, and social norms are everywhere, totally prevalent. So just for, just as a sort of an example, if you, if you Google the list of social rules on Buzzfeed, you'll, you'll find lists of 20, 30, 50, 100 different social rules that you go, yes, I totally recognize that. And it, I find that interesting, not because of the rules themselves, but just because there are so many rules out there that until you see them written down, you don't realize you just do them. So as, for example? So for example, holding a door for somebody, or, you know, saying hello to people or just things that you just do without thinking about it, that you've been in some way programmed to do by, by society,
Speaker B:by your family, by your friends. Well, although what we're saying is that it's a strategic optimum. Is that, is that, you know, you, you are bearing a cost, but it's better that you bear that cost and, and have the benefit in future than that you've always got to open your own flipping door.
Speaker C:Exactly. Yeah. Yeah. So, so there's, so that, yeah, so, so the, for some, it's, it's at an equilibrium where everyone benefits a bit on aggregate. Can I just say, actually, I think with
Speaker B:the door holding example, I think there's a really important piece of physics here, which is that if a door is already open, it's easier to hold it open than it is to let it shut and then to have someone open again. Absolutely, yeah. And that's vital, right? Now, we haven't evolved a norm to open doors for other people. We've evolved a norm to hold doors open if they're already open when someone's coming along. And I, it wouldn't work. You couldn't get a social convention that people undertake a cost that is, you know, no smaller than, than the cost that, that individual would bear if they were, if they were going to have to do it themselves.
Speaker C:And if you designed doors to be easier to, to open than they were to close, they'd always be open. Like automatic doors. No one holds automatic doors open for each other. Yeah, they'd always be open and it would, the doors would not fulfill their main purpose.
Speaker A:Actually, as an aside, one thing that I lament is the automation of locks on, central locking on cars, right? Because it means I've had to forego the opportunity to behave gentlemanly towards ladies. Because in the past, I would prefer to lock them in and only be a child lock and then only let them out. Yeah, because in the past, I would always go around to the passenger door, open it for the lady. Central locking's got nothing to do with this whatsoever. Of course it does. Because now it's, I mean, I guess I'm talking about unlocking a door. Because if I've gone around to unlock it, I might as well open it.
Speaker B:Yeah. Okay. You'd have to unlock the doors. Wait, hang on. Are we saying that you used to have to go and unlock each car door separately? I'm afraid so. Yeah. It's amazing we made it this far. I can't believe it. You're right. You're absolutely right. That's what central locking is.
Speaker A:Sorry, I've slightly distracted us. What were we talking about? Well, I've distracted myself anyway.
Speaker C:We were talking about social norms. Talking about doors. Talking about doors in the context of social norms. But anthropologists and sociologists will maintain that social norms are absolutely crucial in terms of the stitching together of groups of whatever size. So even pairs of people. So a couple will have their own particular peculiar social norms to that couple that is part of the social cohesion, which keeps them together. And any size of group, their bigger will also have their own social norms.
Speaker B:Yeah, but that's typical anthropologists, right? To observe something and say, oh, social norms are important. I mean, the point about the game theory analysis is that it explains why it is. Why it is that people will bear... Okay, I mean, look, it is puzzling. Someone bears a cost, albeit a small one, to hold the door open for someone they don't know and who they're very unlikely to ever meet again, right? So they're undertaking a cost for somebody else's benefit. Now, whether you like it or you might like it, that we'd have evolved to do that, but we would have been wiped out as a species if we did, right? So you have to explain why that phenomenon exists. Why it is that we do things that are beneficial to other people at a small cost to ourselves. And why we're entitled to expect that that will continue to happen to us. So we'll all benefit, right? Because it isn't an instantaneous trade. I mean, the crucial point about the prisoner's dilemma is that you don't know. You don't get to see what the other person did first. So it is a prima facie puzzle. Now, we're all humans, so we're so used to doing it that we think it's normal. We think it's just sort of obviously being nice to people is your default setting. But you have to realise that the urge to be nice to people is the result of quite a lot of combination of evolution and then evolution's chance to express what it is we've evolved to do
Speaker C:in the form of these specific social norms. Well, I mean, it's not an explanation, but I'm a big fan of the Buddhist concept of karma, where it makes this kind of, quid pro quo, like a currency. So the more good stuff you do for people, the more kind of capital you build up as karma, which the more capital you have, the more likely or the deeper you go into Shangri-La when you ultimately
Speaker A:die. I think that's not, I mean, that's not exactly how karma works, but it's kind of it. It's not, though. I mean, that's an element of it. I think another element that's probably closer to the idea of it is that is by doing something good that has a knock on effect for others of they will also do good things. It's not about building up the personal karma. That's a causal analysis.
Speaker B:Well, it might be. And all discussion of this kind of thing always gets bogged down in the same old crap, which is people saying, oh, you know, obviously it's better to be nice. That's why we're nice because it's obviously better. Well, that doesn't explain it because it's not in our interest to be nice. When you see someone... Well, that may or may not be the case. When you see someone holding open a door for someone, it is pure self-interest because they are playing they're playing a match equilibrium, which is in their interest. If it wasn't in their interest to do it, then those people, those door holders, holder openers would die out.
Speaker A:God, you're such an economist.
Speaker B:Well, I'm not. I'm just, you know, this is how I am, actually. But that's irrelevant. This belief, what I'm saying is simply happens to be true.
Speaker C:Yeah. And we rationalize it as having a moral compulsion to do that.
Speaker B:I mean, you can see morality, social morality, praise, blame, punishment as a way of solving this prisoner's dilemma problem. And if you read Steven Pinker's book, The Better Angels of Our Nature, where he tracks the sort of decline of violence, you know, over the long term in human societies, his analysis of it at the end is that basically, you know, the better we have got as a society at having methods to enforce the good outcome for the prisoner's dilemma versus the bad outcome is the is ultimately the explanation for why we have, you know, got, we've all become better off.
Speaker C:And we, when we're teaching our children about the right way to behave, we'll dress it up as being just, we'll just say it's just the right thing to do, or God says that's the right thing to do. Because that's an easier concept to grasp than, oh, well, you need to understand game theory, and you talk about Nash equilibriums, especially when those things hadn't been worked out yet, thousands of years ago.
Speaker B:I mean, when I discovered the prisoner's dilemma, when I was about 15, and I bought a book called Games of Life, I can't remember the author, you can look it up on the internet. And this really, it blew me away. I just thought this is, this is, you know, absolutely right. This is, you know, actually, you're being shown something which is universal, but completely baffling. And it happens all the time, everywhere you look in society, why we don't drop litter, why we don't, you know, why we don't, you know, just steal things when we're in a shop, if we could get away with it, all of these things, and then the evolution of the institutions to enforce that, you know, why we have, why we, I mean, if you think about, why do, well, why do we follow laws, you and me follow laws, maybe because we're scared of being punished by someone, we might get caught by the police, why do the police bother enforcing laws? Why do they put any effort into enforcing laws? You know, it's not in their individual interest at that point to do that. You can see the whole edifice of, you know, our political institutions, our political and social institutions is all founded to solve this one problem. Of, you know, which is very nicely expressed by this, by this pairing.
Speaker A:Right, which brings us back round nicely to that convention. We need to wrap up. But just before we do, I've got a question. Conventions, do you have a particular social convention that you're particularly fond of? Because I do. I'll kick off. I really, I'm a big fan of queues and queuing. Very specifically...
Speaker C:Very efficient way of doing things.
Speaker A:Yeah, it is. And, you know, when I'm in other countries where, you know, other, you know, benighted places less fortunate than us in the UK, I get nervous, you know, when there's a bit of a scrum, you know, and I get very disgruntled. And just what I like about the queue is that I can relax. I know it's going to be okay. It's all gonna be fair. You'll get your turn. Yeah, exactly. But particularly...
Speaker B:Prisoner's dilemma right there.
Speaker A:Okay. You could push in. But at what cost? Right. Very particularly, I like British pub queues. I like the fact that there is a queue that is not physical. It's distributed. Yeah. But everyone kind of knows it. And you can tell whether another fellow is a fellow gentleman or not.
Speaker C:Although I find it does break down quite often, particularly in busy London pubs, Friday, Saturday night. It does. There is no honour in London.
Speaker A:Well, I pride myself, no matter how busy the pub. So do I.
Speaker B:No one is the guy who pushes in in a pub queue. No one stands up and goes, I'm the guy who takes advantage of the fact that pub's busy.
Speaker A:Even as tempting as it can be in when as busy as it can get. So that's my favourite. What about yourselves? It might be the same for all I know.
Speaker C:Well, I have a distant military background. And so the wearing of any kind of denim was more or less frowned upon at any time. But we still do quite a bit of business with military people now. And now that I'm out of the system, I love that rule because it gives me an opportunity to dissent and to break the rule because I know how much it offends people. I will do it deliberately just for fun.
Speaker A:I like that. I like that. That it's a convention you like to break. Nick?
Speaker B:Yeah. So mine would be, it's really straightforward and it's bloody obvious why it's a good rule. But it's not dissimilar to queuing, but it's letting people off the tube first. It's possibly the thing that riles me the most. This is why I'm thinking about, when I think about, OK, adherence to social norm, which social norm do I like best, is almost the same as saying, which social norm do I hate it when it's broken most. And not letting people off the tube. Before getting on drives me up the wall to the extent that I will resort to minor assault by barging people with my shoulder if I can, if they try to get on. And they always have the same glassy eyed expression when they're getting on like zombies. And you think, well, what is the point? I've got to get off, right? What is the point of you not letting me off first? So I'm afraid that's a bit negative, really, because I'm being a bit shouty. But that's one I think I hold sacred, because there's such a good rationale for it. Yeah, I agree with you, though I don't
Speaker A:resort to physical violence. But I tell you what, I've given a few people some stern stares in my time. I bet you left them hurting. I bet I did. Yeah, no, the other social convention I like is offering people cups of tea when they come to your house or you go to their house. Yes, no,
Speaker C:that's a good one. I think any visitor, no matter who they are, even if they're an adversary, you should always ask them, what would they like to drink? So that's why I still manage to get a cup. Stop stealing my silver. Also, is it white with two? Okay, so let's wrap up there. You've been listening to the Cognitive Engineering podcast. I'm Fraser McGruer.
Speaker A:We've been here with Nick Hare and Peter Coghill of Aleph Insights. That's all for now. Until next time. Goodbye.
