Intro. [Recording date: January 28, 2026.]
Russ Roberts: Today is January 28th, 2026, and my guest is philosopher and author, Hanno Sauer. His latest book, and the subject of today’s episode, is The Invention of Good and Evil: A World History of Morality. Hanno, welcome to EconTalk.
Hanno Sauer: Thanks for the invite, Russ.
Russ Roberts: Now, your book opens–this is a sprawling book. It is full of interesting ideas, and it covers an enormous span of human history and human behavior. So, we’re going to do the best we can to get at some of the ideas in the book.
You open with a passage that reminded me very much of Adam Smith’s Theory of Moral Sentiments. The first sentence of Smith’s is: “How selfish soever man may be supposed, there are evidently some principles in his nature, which interest him in the fortune of others, and render their happiness necessary to him, though he derives nothing from it except the pleasure of seeing it.” And much of that book is trying to answer the question of: Why do we ever do anything that is not self-interested? I would not say unselfish. I don’t like that phrasing. But, Smith starts with the idea that we’re self-interested.
So, why do we ever do anything that even looks altruistic?
Your book starts with the following–or not starts, but early on:
The fact that cooperation is unlikely can be formulated as an explanatory problem in evolutionary theory: how did evolution manage to create altruistic or cooperative tendencies, even though–apparently, at any rate–these tendencies inevitably reduce our reproductive fitness? How could it ever be beneficial for me to help someone else? How could it ever be worth subordinating my self-interest for the well-being of the community? [italics in original]
So, take a crack at–your book in some sense is trying to answer that whole question, so it’s a long answer, but give us the short answer.
Hanno Sauer: Yeah, you’re right. Adam Smith, I greatly appreciate him. Fantastic writer, philosopher. More and more, he gets, I think, also recognized as a philosopher again. And, of course, his economic contribution is, without a doubt, first-rate. So, I enjoy that comparison.
And, you’re right: At the most abstract and general level, one way of putting it would be that I try to deal with a kind of reverse theodicy problem. So, when you think about it, in pre-Darwinian times, when you have a theistic framework, a theistic outlook on the world, you get a theodicy problem. You kind of need to explain how evil and suffering come into the world if you have a deity that is all-kind, and all-powerful, and knows everything. So, how do you square these three classical features of this divine entity with the fact that there are children dying early, and there are wars and genocides and torture and so on–
Russ Roberts: Earthquakes.
Hanno Sauer: That’s right. Yeah.
And, now you get the opposite problem when you move to a naturalistic Darwinian framework. All of a sudden, the default assumption seems to be that it’s ‘nature, red in tooth and claw.’ It’s dog-eat-dog, it’s elbows out. Everyone is selfish. Everyone is essentially sociopathic. Right?
And, now you get the problem: Okay, evidently there is friendship and heroism and love and altruism and sacrifice. But, where do those come from? It seems to not make any sense.
But people like Adam Smith noticed this tension as well, and many others did. Evidently, such things do exist. So, on the one hand, we do have selfish and antisocial and uncooperative instincts and drives and inclinations, but clearly we also have the opposite. We also have friendship and altruism and morality and nice and cooperative and generous dispositions. Where do they come from?
And I think interestingly, we haven’t really figured out a good, solid, and precise answer until well into the second half of the 20th century, where people started to understand–evolutionary biologists equipped with game theory, in tandem with economists indeed; sometimes these people were one and the same–figured out where do the returns from cooperative behavior come from? And, people started talking about inclusive fitness, and we started to understand reciprocity and how cooperation can get off the ground in a Darwinian world that in principle is morally indifferent and doesn’t follow any divine design or any plan.
Russ Roberts: And, those efforts, which include making the observation that, well, your genes are what–as Dawkins points out–your genes are what are driving this. If your brother shares your genes, it might be genetically valuable to make a sacrifice to save your brother, even though you might lose your own life. Or your cousin if the gains are large enough. Or multiple cousins; and so on. But, what I liked about your treatment of this–and much of the book has this flavor–that’s interesting, it’s helpful. It’s not quite the whole story, though. And, why not? And, what do you go on to posit as the fuller explanation?
Hanno Sauer: So, you’re right. So, it’s one thing to find out or to figure out why cooperation is an explanatory problem in the first place. But then, you get a whole, another set of problems that are specific to human beings, because we are indeed somewhat unusual in the animal kingdom, because we do cooperate and we thrive under conditions of cooperation. We are, in fact, especially good at cooperating, in our own way.
But, what we do is curious and unique, namely: we have managed over the past years, decades, centuries, millennia, hundreds of thousands, and even millions of years in a way–now we’re talking about proto-human, pre-human beings–we have managed to scale up our cooperation in a way that other animals don’t do.
So, we have chimpanzees, and they live in cooperative small groups of maybe a few dozen individuals, but they never build societies of thousands of individuals, or millions.
In fact, it would be a kind of science fiction horror story–it would be uncanny to see millions of chimpanzees cooperate, right? Because we know they don’t do that.
Now, there are some animals that engage in large-scale cooperation–for instance, certain insects, termites, and so on and so on. But also, they always do just that. So, they have a specific genetic programming, and they live one way and not any other way.
But, human beings have this malleability and plasticity and capability to live in all sorts of ways. We can, and sometimes occasionally we still do even nowadays, live in very, very small hunter-gatherer groups. And also, we live in the society that you and me live in: We talk across thousands of miles with modern technology, and we use trade that spans continents, and so on and so on. So, the society that we are part of essentially has billions of members–right?–who cooperate on the basis of institutions and norms and social practice, and so on and so on. And, only we can really do that.
And so, the book that I’ve written is essentially a story of how we manage to expand our institutional toolkit to scale up human cooperation from very small groups to contemporary times with billions of people cooperating, sometimes inadvertently or not even knowingly, but we do in fact cooperate with people from Egypt and Taiwan.
Russ Roberts: So, what are the evolutionary pressures that created that unusual result for human beings as opposed to other animals?
Hanno Sauer: I mean, obviously there isn’t just one, but I try to narrate this story of the various mechanisms that we use to scale up cooperation.
So, you can think of it as groups grow in size from very small, largely genetically-related people that interact face-to-face almost every day, know each other very well, and occasionally meet other groups that they are distinct from and that may be somewhat hostile, it may be unremarkable. But we’re talking about small groups. But, the larger these groups become, you need to introduce new mechanisms to stabilize cooperation, which becomes increasingly fragile, the more people you have. It becomes more entropic, so to speak, when you have more members in a group.
Just think about a camping trip that you do with six people in your family and compare that to a camping trip with 60 or 600 people. You see, you need complete different solutions regarding division of labor, enforcement of norms, who gets up in the morning and when, who takes care of the kids, who is cleaning the pots, who is doing the fishing, who is building the tents, who is repairing the tools, and so on and so on. And so, you need to figure all that out.
And, after a while, maybe the people who are supposed to clean the pots and repair the tools, they say, ‘I don’t feel like it today.’ And, cooperation tends to unravel when this happens. Right? So, you need enforcement mechanisms.
And so, I dedicate a whole chapter to this idea of enforcement, social sanctions from softer sanctions such as telling people off, gossip, all the way to capital punishment, and the way that this plays a role in the so-called self-domestication of humans.
So, we are the golden retriever in the primate kingdom. So, think about how golden retrievers relate to wolves, and that’s how we relate to chimpanzees and gorillas. Very docile, kind of peaceful, at least towards the in-group. Very norm-conformist, very eager to learn, to play, relatively low aggression–at least impulsive aggression–and so on and so on. And, we have become that way.
Well, the question is, how did we become this way? And, I think the story that is best supported by the evidence and also some theoretical considerations is that we just killed the most aggressive members of our tribes and bands for hundreds of thousands of years. So, you get a kind of very, very intense selection pressure on human groups where, if you take out the 10% of the most violent people each generation, you’re going to become less violent. Right? Because if you, I don’t know, if you take out all the wannabe bullies and tyrants before they get to reproduce, those genes tend to disappear from the population. And that sort of happened. And, that is another mechanism: the self-domestication mechanism is another way for us to stabilize cooperation and to do the next step in scaling up our group size.
Russ Roberts: I love that idea of the self-domestication and the role of punishment. Although it’s a little hard to believe. We’re not a docile species. And, I think about sometimes a different version of how we, through our own actions, changed our gene pool, which is: In early days, the alpha males had lots of wives–mates, I’ll call them as a better term. And those who couldn’t find a mate would launch off into territories unknown. Consider, the caricature would be a viking. You get in your boat with a group of other dissatisfied men, and you go conquer something, and you come back with treasure, and you earn some mates that way. And, we are in some dimension, I suppose, maybe the descendants of both of those groups: the people who remained after the bullies were killed, and the descendants of people who had the taste for risk, danger, and violence. Maybe they’re the ones who escaped the punishment and got out of town and came back later. I don’t know. What are your thoughts on that as another story?
Hanno Sauer: Well, I think this also happened. We know that humans can be very violent as well. They can coordinate on violence. They can form ad hoc coalitions. They can practice violent cooperation in hunting, which is a kind of–it’s very similar to ambushing and raiding a different tribe. I believe that’s what primordial warfare is mostly supposed to have been, is like raids. Nightly raids. So, you don’t really have: we’re going to meet war-like in an open field and engage in combat like that. It was mostly, yeah, and we take what we can get, and we take as many women as we can get. And so, that has left some mark on our genetic makeup as well.
And then the question would be like, which of these processes is, maybe swamped the other? And so, either way, this self-domestication dynamic would be something that applies more to the in-group, to the group itself. And, these violent tendencies would largely remain directed at the out-group. And, that is indeed another factor in our moral psychology is that we inherited this strong us-and them cognition.
Russ Roberts: But as you point out, a lot of times these urges are not–they’re not confined to the places they were first directed to. So, you have a great line, great two lines here. Quote:
One of humanity’s greatest moral developments was delighting in cruelty. It was all the more difficult to unlearn this lust for cruelty after it had fulfilled its purpose.
Explain those two sentences. Why would it be a great moral development to delight in cruelty?
Hanno Sauer: Well, if you have an increasing need to enforce social norms–and sometimes you need violence–it just helps to install a disposition for people to enjoy violence. Right? Just like, if you want to get people to reproduce, it makes sense to have people enjoy that.
And, that is not a justification for being violent today. It’s just under certain circumstances that can be an asset. And, later on, once you have solved the problem, as it were, that you were going to solve with this taste for violence, it can be that it becomes a kind of evolutionary hangover. So we still have it in our psyche, but now it’s impractical, and we are too violent and too harsh and too punitive.
And I think that is what we see sometimes that in contemporary societies and in modern societies, it is sometimes possible to need very little violence to enforce cooperation for various reasons, but it’s still possible for that instinct, for that atavistic instinct to flare up. And when an egregious crime happens that’s very, very salient perhaps in the public, people get this taste for punitive reaction. It’s understandable, but at the same time, it’s useful to know where that comes from and when it may or may not be appropriate.
Russ Roberts: But your claim–which is I think an unusual claim–is that the source of that–what you call a lust for cruelty, or we could also call it a psychopath has this lust for cruelty–you argue that it is the evolutionary hangover of a positive impulse. Which is, to really put it in economics language, a punishment for free riding: people who exploit for their own benefit whatever the group is doing and cheat on the agreement or on the norms has to be punished.
It’s interesting–you can punish them by expulsion, you can punish them by social exclusion, social isolation, but none of those are as effective as violence. And in fact, of course, as you point out, sometimes there’s a leader who doesn’t just free ride. That person becomes a tyrant over the group, in which case a group of people take the tyrant down–Brutus and his friends take down Julius Caesar. And, it’s that need for that punishment mechanism, that way to reduce free riding and tyranny, that you’re arguing has a negative externality in our behavior in other areas.
Hanno Sauer: That’s right. That’s right. I fully agree with that description. And, the reason I describe it as, well, some things can be moral improvements or moral changes that, at a certain point in time, constitute a moral improvement, even though looking at them from today’s perspective we would also view them as something unpleasant and at least ambivalent. And, I think the evolution of punishment and cruelty is a little bit like that, where it was like a ladder we needed to get on the roof, and now we want to kick the ladder away. It has that kind of thing about or kind of feel about it.
The very general point is that as human groups become larger, the need for enforcement of norms increases. We know from experimental studies, economics games, public goods games, for instance, that people start with in the first couple of rounds, one or two rounds, with cooperation, and then cooperation unravels. And when you do these experiments with a punishment option, you see that cooperation can be stabilized.
Now, that’s not the first-best way that we would like humans to be, but that’s the way it is. Out of the crooked timber of humanity, no straight thing was ever made, and social sanctions are one of the ways to straighten out the crooked timber of humanity.
And, it still works today. It makes sense to think about also, to engage in cost-benefit analysis of how punitive we need to be. And, I think we often overlook that because our psychology isn’t really cost-benefit psychology, as [?] to the greater chagrin[?] of economists who keep reminding people of trade-offs and hidden costs. It’s one of the big frustrations that economists have with the general public is that they only see benefits or they only see costs, and so on and so on, and they just completely omit and ignore one side of the equation.
So, we should engage in cost-benefit thinking when it comes to punishment and our punitive instincts and social sanctions. But they did play that role in the past, and they still continue to play that role today. And, I don’t think we could really do without any social sanctions. We tend to need some mix of incentives and sanctions, of course; and ideally more incentives and fewer sanctions, and ideally the sanctions aren’t too harsh. But in principle, well, collective action problems remain present and they always threaten to undermine social cooperation, and so you need some sort of enforcement.
Russ Roberts: Just to criticize economists a little bit, I think–
Russ Roberts: Well, yeah, easy for me, right? I’ll do the dirty work for you. Although you may not agree with this. Gary Becker, who happened to be my advisor, in his Economics of Crime and Punishment, if I remember correctly, advocated often for a large punishment with a small probability of being caught. Because, arguing that the expected value of the punishment is what would have an impact; and by having a low probability of being caught, you could reduce the cost of monitoring and enforcement. So, you could have a relatively small police force as long as sometimes, when people get caught, they will pay this huge price. And then the expected cost would be sufficient to deter future crime.
And, I’ve argued–somewhere on this program years ago–that morally that’s very disturbing to most people, because it means that a handful of the criminals–they’re all guilty in theory. Of course, that’s part of the problem, is that they’re not all guilty, the ones you impose the punishment on. And, even if they are all guilty, you’re imposing a large punishment way above probably the crime to make sure that the expected value deters the other criminals. And that offends our sense of justice, even if it’s, quote, “efficient.” It’s deeply disturbing, I think, to human beings.
Hanno Sauer: And, I think that’s a nice–so I like this way of thinking. I just disagree with this specific point. But, because I think–so I may not be 100% up to date on this topic, but the last time I looked at the evidence–
Russ Roberts: Hanno, I’m definitely not up to date, so go ahead. It’s okay. Go for it.
Hanno Sauer: So, then I’m just leaving it to the audience to look it up.
But, my read of the evidence right now is that, in terms of making deterrence and the threat of punishment actually effective in deterring unwanted antisocial behavior, it’s the opposite.
So, it seems to be motivationally the most effective when people are quite certain that they get caught, even when the punishment is not as harsh. So you can tell people–so, if you have a society where you rob someone and there is a 90% chance you’re going to get flogged–right?–people find that very motivationally disincentivizing. And, if they have an extremely harsh punishment as they’re going to get drawn and quartered in the town square, but it almost never happens, people think they’re going to get away with it. And, apparently, well, this would make sense because, I mean, there is a kind of selection effect of people who self-select into criminal careers. They tend to not be very, very prudent. Right?
Russ Roberts: Yeah, as good a probability, for sure. Maybe.
Hanno Sauer: Right. So, I suppose that you and I don’t tend to contemplate the pros and cons of bank robbery very often. It’s just not a thing that some–so the kind of person that contemplates seriously whether or not to rob a bank may not be the one that is most susceptible to the kind of cost-benefit analysis that Gary Becker recommended. So, it’s a little bit of–it’s the kind of punishment that would work among economists with an IQ [intelligence quotient] of 145, right? But not on the people that you want to deter.
But, I like that idea. I like that way you’re thinking. I remember that there is an example that Gordon Tullock gives at some point where he says, ‘If you want to increase road safety, don’t install airbags. You should install a sharp dagger into the steering wheel.’ Right? That’s going to increase safety. And, it’s just so counterintuitive, and as you said, is upsetting to most people to think that way. But I like that way you’re thinking, if only because it’s contrarian, and it cracks something open. I like that.
Russ Roberts: It’s a perfect Gordon Tullock example. He delighted in that kind of provocative example. But of course, the problem with that is that, yes, if there were a spike in the car or a dagger, you’d drive very slowly, very cautiously; but more likely you wouldn’t drive at all. But, that highlights what the costs are of excessive punishment.
Hanno Sauer: Exactly, exactly.
Russ Roberts: Before we leave this, I want to–I know my listeners are eager to hear me say that, quoting Smith, “Man naturally desires, not only to be loved, but to be lovely”–that is, we are hardwired to care about what other people think of us, but we also want to be actually good and praiseworthy. Not just praised, not just honored, but honorable.
But, you’re right: I love this:
The exchange of gossip and rumours has played an important part in our evolutionary history, especially for the evolution of language, whose original function may primarily have been social communication about other people’s behaviour.
And of course, that’s very Smithean–the way we care about our reputation, we care about what other people think about us and what they say about us. And, that’s an amazing thing, and it could be true about language.
Hanno Sauer: I mean, I think you’re right that we do not just desire to be loved, but to be worthy of love–to be lovely.
Now, I think in the vast majority of cases, these things are almost identical because the best way to be loved is to be lovely. Right? So, the best way to seem good is to just be good. If you want to just aim at seeming good straight, without the detour through being good, this is a little bit like the paradox of hedonism. If you aim for fun directly, you’re going to miss it. Right? Fun needs to happen–at your behavior.
So, there’s a paradox of being moral, which is that if you aim at seeming moral–if you aim at seeming moral directly–there’s also going to be–you’re going to come off as calculative and manipulative and just keeping up appearances. And, if you just go for just authentically caring about people, you’re going to seem moral, and that’s going to be something that people appreciate much more.
And, I think the process of evolution is smart in that way, in that it has equipped us with these abilities to be motivated in ways that feel authentic to us–that are authentic. So, we genuinely care about friends, we genuinely care about people we love, we genuinely care about our kids, or whatever group that we identify with: the tennis club, or the nation, or the tribe, or whatever. This is not fake, b. But of course, underlying it, distally speaking, there is a strategic rationale that led to the evolution of these tendencies that have become authentically experienced in our minds.
So, it’s sort of both: you have an underlying strategic rationale for the evolution of authentically-felt dispositions of virtue. So, it’s–the great primatologist, Frans de Waal, Dutch primatologist, once said that, he’s an advocate of this so called veneer theory of morality. He has this great line–I think it’s false, but it’s a great line–‘Scratch an altruist and watch a hypocrite bleed.’ And, the idea is that altruism is skin deep. But, I don’t think that’s really true. I think our altruistic and cooperative dispositions, they go pretty deep. Even though it is of course correct, there is a selfish-gene kind of rationale underlying them.
Russ Roberts: But, it’s like you suggest: if it’s authentic, it’s a much more effective signal than if it’s fake. I always like the example of Herb Kelleher when he was CEO [Chief Executive Officer] of Southwest Airlines. On Christmas and Thanksgiving, he would go work the baggage claim and to be with the workers, his workers. And, it was a tremendous stunt. And, he claimed, as far as I know, that they never publicized it much. Of course, we know about it, but they didn’t tell reporters to show up. It’s just something he did. And, I always wondered–it’s a very powerful cultural benefit for his company. So, his workers knew that he was on their side. And, I always ask, why didn’t other airline CEOs do this? And the answer would be–I thought it was pretty simple. They wouldn’t enjoy it, and they’d have to pretend they were enjoying it. I think Herb actually enjoyed it. He was a very down-to-earth guy. He was not a pretentious guy. And, the idea of–I’m sure there were times he’d rather be with his family, but doing manual labor with his workers was fun for him. It wasn’t a stunt.
Hanno Sauer: Yeah. It could also help to be a CEO that used to start at the very bottom in the company where you really know the ropes. So, if you actually used to do that 45 years ago, it’s going to be much more believable. You know the culture and how people talk, you know the movements, and so on and so on. And then, that’s going to be a huge plus for your credibility in the business.
I once heard a story about someone who was walking around London late at night, 11:00 or something. And, there was this famous, world-famous chef. There was many restaurants in London. And, he was getting out of a black SUV [Sport Utility Vehicle] in full kitchen garment only to walk through the back door and then greet the guests as if he had been in the kitchen. And then, he would go back into the SUV and go to the next restaurant.
And so, once you hear–that kind of story is really off-putting to people because people absolutely hate that kind of mimicry. And we are very, very sensitive towards deception and people trying to manipulate us with the signals that they send. And when they do it so strategically, people absolutely hate that. [More to come, 32:42]
