Making Sense - Sam Harris - #386 — Information & Social Order

00:00:00.000 Welcome to the Making Sense Podcast. This is Sam Harris. Just a note to say that if

00:00:11.640 you're hearing this, you're not currently on our subscriber feed, and we'll only be

00:00:15.580 hearing the first part of this conversation. In order to access full episodes of the Making

00:00:19.840 Sense Podcast, you'll need to subscribe at samharris.org. There you'll also find our

00:00:24.960 scholarship program, where we offer free accounts to anyone who can't afford one.

00:00:28.340 We don't run ads on the podcast, and therefore it's made possible entirely through the support

00:00:32.860 of our subscribers. So if you enjoy what we're doing here, please consider becoming one.

00:00:45.300 Okay, well, I'm about to go on a trip overseas, so I'm recording this housekeeping about a week

00:00:53.300 in advance of dropping this episode. This is just to say that if anything extraordinarily

00:00:59.560 important has happened in the world that you would expect me to comment on here, I do not

00:01:04.260 currently know about it, and no doubt we'll talk about it after I get back. Today I spoke to Yuval

00:01:11.240 Noah Harari, who's been on the podcast before. Yuval is a celebrated historian and author. He is almost

00:01:19.460 certainly the most popular historian on earth at the moment. He has written Sapiens and Homo Deus,

00:01:26.420 and his latest book is Nexus, a brief history of information networks from the Stone Age to AI.

00:01:33.540 And in the first part of the conversation, we talk about the book, and then we use it as a lens

00:01:38.220 through which to look at current events, of which there are far too many. On the day we recorded this,

00:01:44.120 Israel had just bombed Beirut in an operation that successfully killed the leader of Hezbollah,

00:01:50.960 Hassan Nasrallah. And based on much of the media coverage, you could be forgiven for thinking that

00:01:56.800 this was a pure act of aggression on the part of Israel. There was almost no mention of the fact

00:02:03.360 that Hezbollah has been launching rockets by the thousands, unprovoked, into civilian centers in

00:02:11.280 Israel since October 8th at the behest of Iran and in solidarity with Hamas. Also no mention of the

00:02:19.800 fact that Hezbollah's main contribution to human culture has been the perfection of the modern

00:02:26.220 tactic of suicide bombing, which they have used not only against Jews in Israel, but also against

00:02:32.540 Americans. They've even bombed Jewish centers in Argentina. This is a global terrorist organization,

00:02:40.260 but judging from the obituaries of Nasrallah, you would think Hezbollah was an entirely legitimate

00:02:47.240 resistance group attempting to throw off some colonialist power. Needless to say, there was

00:02:55.200 almost no mention of the fact that Nasrallah and Hezbollah helped Bashar al-Assad kill hundreds of

00:03:02.860 thousands of Muslims in Syria. The editors over the New York Times and the Guardian and the BBC

00:03:11.040 seem to think that their readers don't care much about these Muslim deaths, and they're probably

00:03:18.060 right. Reading this coverage, you're likely to form the impression that Nasrallah himself was quite a

00:03:25.500 grandfatherly figure with a winning smile. He is described as charismatic and revered and as a great

00:03:34.340 organizer and an astute strategist and pragmatic and idolized by his followers. The New York Times in

00:03:42.360 their obituary makes it seem like he was a man of peace committed to a two-state solution, when in fact

00:03:48.840 he was a religious lunatic running one of the world's most lethal terrorist organizations, and he had an

00:03:56.260 immense amount of innocent blood on his hands. Anyway, Yuval and I wade into this ethical morass, and

00:04:04.020 generally we talk about the consequences of having an information landscape that is so skewed. We talk a lot

00:04:10.480 about the connection between information and social order. We discuss how democracies and dictatorships are

00:04:16.880 different information networks representing points on a continuum. We talk about the tension between

00:04:22.920 truth and order, and the need to have institutions that self-correct for both. The advantages of fiction

00:04:29.620 over truth, what Yuval calls the naive view of information, social media and the breakdown in

00:04:36.060 democracy, what advice Yuval would give to Elon Musk and Mark Zuckerberg, the rise of populism and the

00:04:43.040 breakdown of trust in institutions, inefficiency as an occasional feature rather than a bug of

00:04:49.440 information networks, the banning of counterfeit humans. And then we take the framework that Yuval has

00:04:55.700 built and use it to talk about the U.S. election and Trump and the war in Ukraine and why so many

00:05:02.520 Americans seem confused about the ethics and politics there. And then Yuval and I have a very interesting

00:05:08.500 exchange about the exchange about the state of politics in Israel, the contributions of Jewish

00:05:14.720 religious fanaticism, the ongoing war with Hamas and Hezbollah, potential war with Iran, whether a two-state

00:05:22.820 solution is even conceivable at this point, and other topics. And now I bring you Yuval Noah Harari.

00:05:30.540 Yuval Noah Harari. Thanks for joining me again.

00:05:39.500 It's good to be here again in, you know, in a real space together.

00:05:43.160 Yeah, this is, you are my first interview in this studio, which is auspicious because we share so many

00:05:50.060 interests. We have struggled for years to talk about meditation and you think we might get there this

00:05:55.820 time. I'm skeptical.

00:05:56.760 I'm determined at least to say at least something about meditation this time.

00:06:00.520 Yeah. Well, it's, there's so much going on in the world that it's, uh, it's always a challenge to

00:06:05.060 get to that topic.

00:06:05.720 We can always connect it. I mean, we can talk about Nasrallah and meditation. I think there are

00:06:09.300 many links there worth exploring.

00:06:12.200 As counterintuitive as that sounds.

00:06:13.700 Yeah. If meditation is not related to the world and to reality, it's worthless.

00:06:20.880 That's true. Yeah. I guess I, well, I feel like I'm, I'm perhaps selling our audience short in

00:06:26.740 thinking that they don't have the bandwidth to think about it, given all the chaos and given,

00:06:32.180 given your expertise that is so relevant to so much of the chaos. You just mentioned Nasrallah. So

00:06:36.760 at the time we're recording this, it was just announced that Israel killed, uh, Hassan Nasrallah,

00:06:42.100 who's the, who was the head of Hezbollah, uh, which we'll talk about, uh, perhaps in the,

00:06:47.220 so I want to focus what you've done in your recent book, Nexus, which is wonderful and here,

00:06:53.980 and perhaps you can say how this relates to your two other big books, Sapiens and Homo Deus. So what,

00:07:02.220 what is, how do you view the project that you've?

00:07:04.500 So it's, it starts basically where Sapiens and Homo Deus ended, uh, in Sapiens, I, I covered how this

00:07:12.700 insignificant ape from a corner of Africa took over the world. And in Homo Deus, I explored what

00:07:20.420 could be the potential future of, of us and of our products and creations here on earth. And, uh,

00:07:28.740 Nexus starts with a key question about both the past and the future, which is if humans are so wise,

00:07:36.760 why are we so stupid? Like, you know, we've reached the moon and we split the atom and we can decipher

00:07:43.240 DNA, but we are on the verge of destroying ourselves in so many different ways. It could be ecological

00:07:51.900 catastrophe. It could be a world war, a nuclear war. Uh, we are producing the most powerful

00:07:58.680 technology in history, AI, which might quite easily get out of our control and enslave or,

00:08:05.840 or destroy us. And we know all that, and yet we keep doing it. So what's happening? And so many

00:08:13.760 mythologies and theologies throughout history said that the blame is that something is wrong in human

00:08:21.580 nature, that we are flowed, deeply flowed. And I don't think that this is the right answer.

00:08:26.860 I think that the problem is not in our nature. The problem is in our information. If you give good

00:08:34.560 people bad information, they make bad decisions. It's, it's as simple as that. And, uh, so then the

00:08:41.180 question becomes, why are we flooded with bad information? Why is it that after thousands of

00:08:48.420 years of, of, of history of developing, you know, sophisticated networks of information and, and,

00:08:55.560 and, and, uh, communication, our information is not getting any better. I mean, at the present

00:09:01.700 moment, you can say that humans have the best, the most sophisticated information technology in history,

00:09:07.580 and we are losing the ability to even talk with each other and to listen and to hold a reasonable

00:09:14.680 conversation. So, so what, what's happening that that's the key question of Nexus and it explores

00:09:21.400 the history of information and, and of information networks. It takes another look at history, but from

00:09:29.500 the viewpoint, not of, of humanity, but of information. And for instance, I look at, uh, or Nexus looks at

00:09:37.700 the history of democracies and dictatorships, not as we usually think about them as different ethical

00:09:44.720 systems that believe in different ideals, but as different information networks, how information

00:09:51.720 flows differently in a democracy, in a dictatorship, in a dictatorship, you know, there is one hub where,

00:09:59.580 where all the decisions are being made. So all the information flows to and from that single center

00:10:05.960 and democracy is a distributed information system in which decisions are being made in many different

00:10:14.300 places. And much of the information never passes through the center. Like if you think about United

00:10:20.380 States, so you have the center at Washington, in Washington, but so many decisions are made elsewhere,

00:10:25.880 like here in, in, in Hollywood, in Los Angeles, and most information never passes through the center.

00:10:32.740 And you can think about the historical struggle between democracy and dictatorship in terms of

00:10:39.680 different models of information flows.

00:10:42.740 Yeah. I want to pass over that ground again, because what you're saying is pretty counterintuitive and it's

00:10:47.860 very interesting. So it could, because people think about democracy and dictatorship as this kind of binary

00:10:55.340 that are just categorically distinct and you're placing them on a continuum of information flow.

00:11:01.380 Yeah. Yeah. And, and there, I mean, so let's add here this, what, what you call the naive view of

00:11:08.440 information because it's, there's a sense that more information is an intrinsic good, right? And we're

00:11:14.880 getting this now with the people who are running our, our social media regimes. It's just the idea that

00:11:20.280 if you could just let all ideas collide and remove every point of friction from the flow of information

00:11:28.660 information and amplify anything, however, a market or the dynamics of any internet business chooses to

00:11:35.340 amplify it, there's, it's just the principle, you know, people have these phrases in their minds,

00:11:40.760 you know, sunlight is the best disinfectant, right? So let's just expose everything and we're going to be

00:11:45.220 fine, right? And any effort to steer this information flow is by its very nature, sinister. It is edging us

00:11:54.180 toward the totalitarian side of this information continuum. So how do you react to that?

00:12:00.740 This is so naive. This is so disconnected from reality, from history. You think that you flood

00:12:07.220 the world with information and the truth will just rise to the surface? It won't. It will sink to the

00:12:13.100 bottom. Information isn't truth. Most information is junk. It's like thinking that more food is always

00:12:20.240 good for you. The more you eat, the more healthy you will be. That's the same thing. No. I mean,

00:12:24.660 yes, you need some food to survive, but if you just keep eating, it will not be good for you,

00:12:29.520 especially if you keep eating junk food. And the world basically needs an information diet.

00:12:35.400 And, you know, the truth is a subset of information and a very small subset because the truth,

00:12:44.240 first of all, it's very costly. If you want to write or to produce a truthful account of something

00:12:51.480 of the Roman Empire, of the economic crisis, whatever, you need to invest a lot of time and

00:12:58.000 effort and money in looking for evidence and fact-checking and analyzing. Like, I don't know,

00:13:04.400 if you want to know something that happened in the Roman Empire. So, you know, historians,

00:13:08.300 they go to university to study for at least 10 years before they become professional historians.

00:13:14.040 You learn Latin and Greek and how to read ancient handwriting and how to do these archaeological

00:13:20.440 excavations. And even if you found a document from the Roman Empire and you know Latin and you can

00:13:25.940 read it, maybe it's just propaganda. Just because Caesar says that the enemy had 100,000 soldiers,

00:13:32.760 it doesn't mean they actually had 100,000 soldiers. So how do you evaluate information?

00:13:37.500 So the truth is very costly. Fiction, on the other hand, is very cheap. You just write the

00:13:44.360 first thing that comes to your mind. The truth is also very complicated or often it's complicated

00:13:50.780 because reality is complicated. Whereas fiction can make it as simple as you would like it to be.

00:13:56.980 And people tend to prefer, in most cases, simplicity over complexity. So this is another disadvantage.

00:14:03.700 So you're pointing out there an asymmetrical relationship between truth and fiction.

00:14:09.060 Absolutely.

00:14:09.520 Which redounds to the advantage of fiction in this friction-free environment.

00:14:14.240 And with the third advantage of fiction, that the truth is sometimes, not always, but the truth is

00:14:20.480 sometimes painful. You know, from the personal relationships that we often don't want to know

00:14:26.400 the truth about how we treat the other people in our lives. This is why we need to go to therapy

00:14:31.840 for many years to acknowledge the reality. You know, all the way to entire nations or entire cultures,

00:14:38.360 if you run for elections in the U.S. or in Israel or anywhere else, and you just tell people the truth,

00:14:44.660 the whole truth, and nothing but the truth. I mean, an Israeli politician who would just tell

00:14:50.320 the truth about the Israeli-Palestinian conflict is not likely to gain many votes that way. You need at least

00:14:58.080 some dose of fiction, of mythology to make it more attractive, more pleasant for the voters.

00:15:05.620 Well, it's already so unpleasant. I've shudder to think what the truth is, but we'll get there.

00:15:09.260 So again, the truth is it's costly, it's complicated, it's sometimes painful, fiction is cheap, it's

00:15:18.760 simple, you can make it as attractive as you'd like it to be. So in a completely free market of

00:15:24.820 information, truth will lose. You have to tilt the balance in favor of truth by building institutions

00:15:32.400 like courts, like newspapers, like universities, like research centers, that make the effort to produce

00:15:42.340 and to protect the truth. And when people attack these institutions, they often claim that they are

00:15:50.320 liberating people from the yoke of these elite institutions and conspiracy and so forth. But no, when you

00:15:58.540 destroy all trust in these institutions, you're paving the way for dictatorship. Society needs

00:16:05.720 institutions. And democracy works on trust. But if you destroy all trust, the only alternative left

00:16:13.820 to hold society together is with terror, which is what dictatorships do. So this is the game of many

00:16:20.320 would-be dictators. They systematically destroy all trust in the institutions that are our main

00:16:28.400 access to truth and knowledge. And then when all these institutions are gone, then the only

00:16:35.860 alternative left is a dictatorship. So what would you say to someone who says that the institutions

00:16:41.000 have proven themselves to be untrustworthy, right? So we have the capture of the most elite academic

00:16:48.060 institutions, certainly in the United States, by a kind of woke moral panic, right? You have Hamas

00:16:54.560 supporters, not only among the students, but among the faculty. I'll be happy to talk about that.

00:16:57.980 And I have a lot of criticism of my own kind of disciplines and institutions that I'm sometimes,

00:17:04.180 you know, I'm, you hear things from people who went to study history for 10 years, and then they come

00:17:09.780 up with the most simplistic views of reality.

00:17:13.400 Okay. So what, so, but, so let's take, let's stay at the 30,000 foot level. The experts in many

00:17:19.520 institutions have heaped shame upon their own heads in recent years.

00:17:24.440 But the reaction is not to destroy the institutions. I mean, this is why we need two things. First of

00:17:28.980 all, we need several institutions, not just one. So they keep each other in check. I mean, the basic

00:17:34.900 assumption is humans are fallible. Institutions are composed of humans. So all institutions are fallible.

00:17:41.340 They can make mistakes. They can be captured. They can be corrupted. And therefore you need

00:17:46.760 several institutions to keep each other in check. So if one institution is really corrupted,

00:17:52.000 you can go to the courts or you can expose it in newspapers or in other media or whatever.

00:17:57.000 And secondly, every institution needs a self-correcting mechanism. This is the sign of a good

00:18:02.320 institution, that it has mechanisms inside the institution to identify and correct its own

00:18:08.680 mistakes. This is again, a key difference between democracy and dictatorship. Dictatorship has no

00:18:14.600 self-correcting institution. There is no mechanism in Russia that can expose and correct Putin's

00:18:21.760 mistakes. But democracy is all about self-correction. That, you know, the basic mechanism of elections

00:18:28.400 is that every four years or so, you can say, oh, we made a mistake. Let's try something else.

00:18:34.780 But of course, every mechanism like this is itself fallible. Elections can be rigged. Like we just

00:18:41.080 had elections in Venezuela and the Venezuelan people said, okay, we made a mistake with Chavez and

00:18:46.400 Maduro. Let's try something else. But because Maduro is in power, he rigged the election. He said,

00:18:52.340 no, no, no, I won. And this is also, of course, very, very relevant to what's happening in the

00:18:57.320 upcoming elections in the United States. Because the greatest danger for a democracy,

00:19:02.240 in a democracy, you give power to someone for four years on condition that they give it back.

00:19:08.560 And there is always the danger, what if they don't give it back? So giving power to somebody that you

00:19:13.960 have good reasons to suspect that he will not give it back, very dangerous. So again, elections are not

00:19:21.360 enough. You also need free media. You also need free courts. Now people ask, okay, so what if all

00:19:27.120 these institutions are corrupted, then bad luck? I mean, nothing is perfect. If all the institutions

00:19:34.540 of your society has been corrupted and taken over and all the self-correcting mechanisms are

00:19:40.920 disrupted, dysfunctional, very bad news, society collapses. Hopefully we don't reach that point.

00:19:49.140 And it's very important to try. And the solution is not to lose trust. One of the key problems I see

00:20:00.360 today in the world is that you have an extremely cynical view of humans and of human societies

00:20:07.980 spreading both on the right, but also on the left. This is something that the extreme left and the

00:20:15.760 extreme right agree about. They have an extremely cynical view of humans and of reality. They say

00:20:22.760 that the only reality is power, that all social interactions, all human interactions are actually

00:20:29.920 power struggles, that all human institutions are just conspiracies to gain power, that journalists,

00:20:38.760 scientists, historians, judges, politicians, these are just conspiracies to gain power.

00:20:45.100 Whenever somebody tells you something, you shouldn't ask, is it true? Because nobody cares about the

00:20:51.660 truth. This is naive. They would tell you, no, this is a power play.

00:20:57.100 Who benefits.

00:20:58.040 Who benefits.

00:20:58.840 Whose privileges are being served? Whose interests are being served? This is something you hear from

00:21:03.940 Marxists and from Trumpists. This is something that Donald Trump agree with Karl Marx, at least on that,

00:21:10.380 that everything is just a power struggle. And if you think like that, all trust collapses. And the only

00:21:18.380 thing that is left standing, that can remain standing, is a dictatorship, which indeed assumes that

00:21:24.780 everything is just power. Now, the important thing to realize is this is not just extremely cynical, this is

00:21:32.440 just wrong. People are not these power crazy demons that care only about power. Even powerful people

00:21:42.880 really care about the truth, at least some of them.

00:21:46.440 Well, even, I would just add, I totally agree with you, but I would add as a footnote to that cynicism,

00:21:52.340 even if the cynical take were true, people's incentives are not perfectly aligned. So even in a rivalry of people

00:21:59.740 seeking power, the kinds of conspiracies and collaborations and Orwellian, you know, star chambers

00:22:07.620 rarely exist the way populists imagine, right? I mean, it's just, you can't get, you take...

00:22:13.140 You take Elon Musk and Donald Trump, you know, two people with a very high opinion of themselves

00:22:19.660 and with not necessarily the same goals in life and in the world. Even if they can ally themselves

00:22:28.340 for some time around a certain common interest, in the long run, it will be very, very difficult

00:22:34.480 to keep this alliance.

00:22:36.120 Yeah. Yeah. To say nothing of the people who are not aligned with them. There's a fascinating

00:22:41.340 tension between the self-correcting mechanism that would deliver truth and the self-correcting

00:22:47.780 mechanisms that would deliver order. There's this trade-off between truth and order that you

00:22:51.460 describe in the book. Yeah.

00:22:52.740 Let's cycle on that for a minute.

00:22:54.540 Okay. That's very important because, and if we think about human societies really as information

00:23:01.980 networks, so the question is, what do these networks produce? And to function, they need to produce

00:23:10.040 two different things, not just one. They need to produce both truth and order. A system that just

00:23:18.660 ignores the truth completely will collapse. It will sooner or later encounter reality and will collapse.

00:23:26.200 But also, a system which is only interested in the truth will not be able to maintain order

00:23:32.880 because it's easier to maintain order with fiction and with fantasy and with mass delusions.

00:23:41.000 So we just spoke about the tension between truth and fiction as though fiction were by definition

00:23:46.400 invidious and something to be canceled. What you're saying now is that we need certain fiction. It can't

00:23:53.920 be purely truth-seeking, right?

00:23:55.880 Fiction is very efficient in creating order. And the main thing is that it's complicated

00:24:02.820 to maintain a human society. It's complicated because you need to balance two things that

00:24:09.200 are pulling in different directions. You need to balance truth with order.

00:24:14.000 There's lots of trade-offs.

00:24:15.060 Yes. And I'll give one example of how it works. Think, for instance, that you want to produce an

00:24:24.740 atom bomb. So let's say that you are Iran and you want to produce an atom bomb. You need to know some

00:24:31.820 facts about the world. It's essential. If you just ignore the facts of nuclear physics, you will not be

00:24:38.340 able to produce an atom bomb. It's as simple as that. On the other hand, to produce an atom bomb,

00:24:44.960 just knowing the facts of physics is not enough. You need millions of people to cooperate on the project.

00:24:50.960 If you have a single physicist, she is the most brilliant physicist in history. And she knows

00:24:57.020 that E equals MC squared and all the secrets of quantum mechanics and whatever, she cannot produce

00:25:03.140 an atom bomb by herself, working in her garage or something. You need people to mine uranium,

00:25:09.900 you know, thousands of kilometers away. You need engineers and workers to build the nuclear reactor.

00:25:15.420 You need people to produce food so that all these workers and physicists have something to eat.

00:25:21.580 How do you get millions of people to cooperate on the project? If you just tell them the facts of

00:25:26.240 physics, E equals MC squared, now get on with it, it doesn't work. So what? So just because E equals MC

00:25:33.860 squared, we should now work on building this atom bomb? No. This is where ideology and mythology

00:25:39.660 and fictions come into the picture. You usually convince millions of people to work on a project

00:25:45.920 together by telling them some ideology or some mythology. And here the facts don't matter so much.

00:25:52.800 If you try to build a very powerful ideology and you ignore the facts, your ideology is still likely to

00:25:59.760 explode with a very, very big bang. And in most cases in history, the people who are experts

00:26:07.280 in nuclear physics, for instance, get their orders from the people who are experts in Shiite theology

00:26:14.880 or in Jewish theology or in communist ideology or in Nazi ideology. It's the people who are experts in

00:26:23.040 the truth usually get orders from the people who are experts in order. And this is something that

00:26:30.800 very often scientists and engineers don't understand. That they work, I don't know, on AI and they think

00:26:39.720 that the scientists and the engineers will decide what to do with it. But no. Once you produce this

00:26:46.660 very powerful technology because you know the facts about the world, then you will get the people who

00:26:53.500 are experts in mythology and theology coming in and telling you, thank you very much for producing this

00:26:59.160 very powerful technology. Now we will decide what to do with it.

00:27:03.140 Is there a truly benign and wise version of this? Because what you seem to be describing is a yet

00:27:11.740 another reason for cynicism and distrust.

00:27:14.320 So this is important. Now, it's very hard to create large-scale societies without any fictions.

00:27:22.420 Even money is a fiction. Even, you know, think about what is the last thing that still holds American

00:27:27.100 society together. What is the last thing that Republicans and Democrats still agree on?

00:27:32.240 It's the dollar. And even this is under attack, you know, from cryptocurrencies and so forth.

00:27:36.980 But almost the last story that still holds the place together is that everybody agrees on the

00:27:41.960 value of a dollar, which is just a fiction.

00:27:44.260 I think we hit this in a previous conversation. There's something a little confusing about

00:27:47.780 your use of the word fiction here. Because fiction, in any kind of context where we're talking about

00:27:52.100 the truth sounds intrinsically pejorative, right? So like this is, there's truth and there's fiction.

00:27:57.920 It's not pejorative.

00:27:58.700 You're talking about conventions. Something that's conventionally constructed or socially constructed.

00:28:01.660 Something that comes out of the human imagination and not from reality.

00:28:04.700 Yeah.

00:28:04.940 I mean, the value of the dollar is purely an imaginary reality.

00:28:09.200 Exactly, yeah.

00:28:09.800 The paper, the dirty paper in your pocket.

00:28:12.040 And most dollars are not even paper. They're just digital tokens in computers.

00:28:15.840 Right.

00:28:15.960 They have no objective value.

00:28:18.240 Yeah.

00:28:18.520 They have value only in our imagination. In this sense, the dollar is a fiction.

00:28:22.080 It's a socially constructed reality.

00:28:23.600 Yes. And now the big question comes, so is everything just a conspiracy? Is everything

00:28:30.400 just a fiction? And the key thing is that fictions can be extremely valuable and positive provided

00:28:38.380 you acknowledge they are fictions. I don't think that the dollar is a bad thing. I don't think

00:28:44.560 that the fictions holding society together are a bad thing as long as you acknowledge the

00:28:49.860 reality that this is a man-made imaginary thing. And this is important because then you can correct

00:28:57.280 it. Then you can make amendments in it. Let's compare two texts that are foundation texts for

00:29:04.400 holding human society together. The Ten Commandments and the U.S. Constitution.

00:29:09.660 I have a preference.

00:29:10.280 Both are fictional in the sense that they came out of the human imagination. But one text refuses

00:29:18.740 to acknowledge this reality, and the other text is fully honest about it. So the Ten Commandments,

00:29:26.360 they start with, I am your Lord God. This text, it claims to be written by God, not by any human,

00:29:35.660 which is why it contains no mechanism for correcting its errors. And if some of our listeners are

00:29:44.120 outraged, errors in the Ten Commandments, what could possibly be wrong with don't kill and don't

00:29:49.460 steal?

00:29:49.900 I don't think I have those listeners. You give me too much credit.

00:29:52.920 Yes, but maybe for the one or two are still left out there. Notice if you read, for instance,

00:29:57.480 the Ten Commandment, that it endorses slavery. The Ten Commandment says that you should not covet your

00:30:05.940 neighbor's house or field or slaves, which implies that God is perfectly okay with people holding

00:30:14.000 slaves. It's just God doesn't like it when you covet the slaves of your neighbors. No, these are his

00:30:21.080 slaves. Don't covet them. Now, because the text doesn't recognize that this is the creation of

00:30:27.740 the human imagination, there is no 11th Commandment, which says, well, if you discover something wrong

00:30:34.500 in the previous Ten Commandments, by a two-third majority, you can vote on changing Commandment

00:30:40.740 number 10. No mechanism. So we still have the same text from the Iron Age until today. Now,

00:30:48.520 the U.S. Constitution, it's also a foundational text which gives instructions for people to how

00:30:54.800 to manage their society. It also came out of the human imagination. It's not an objective reality.

00:31:02.740 It's also, in this sense, a fiction. But it is honest. It starts with we the people. We the people

00:31:09.760 wrote this text. And because we are fallible human beings, maybe we made some mistakes, like endorsing

00:31:16.660 slavery. So we also include in this text a mechanism to amend it, to identify and correct

00:31:23.220 its own mistakes. It's not easy, but it can be done. And it has been done. So fictions can be

00:31:30.180 extremely valuable. We need them. We cannot have a large-scale society without them. But they should

00:31:35.920 be honest about their own fictional nature, which gives us the ability to identify and correct

00:31:44.820 our mistakes or the mistakes of our ancestors.

00:31:47.880 But there's something intrinsically conservative about this picture. Because admitting that something

00:31:54.080 is a fiction or a convention is not to say that you should want to revise it impetuously, right?

00:32:01.660 Absolutely not.

00:32:02.380 It's very good that very few people try to rethink the convention about whether to drive on the left

00:32:07.840 side or the right side of the road on a daily basis, right? I mean, it's clearly arbitrary,

00:32:11.620 but it's crucially important that we all, once we've decided, that we not keep rethinking it.

00:32:17.000 Yeah. And there are many things like that. And so this distrust in institutions, that what has

00:32:21.660 grown so corrosive is the sense that the appropriate response to each of the errors, however embarrassing

00:32:29.300 they have been of late, is to break fully with the institutions and in some sense reinvent

00:32:36.200 civilization on your own. You find some place to stand where you can reboot from. And that actually

00:32:42.260 seems to be, I mean, obviously there's a populist version of this, which we can talk about. I think

00:32:47.580 we should talk about populism. But this message seems to be coming from on high. I mean, you've

00:32:53.340 mentioned Donald Trump and Elon Musk as prime offenders. They're prime offenders on this very point.

00:32:59.020 They sow distrust in our institutions on such a fundamental level and at such scale.

00:33:05.480 It's coming from the right and from the left. It's the populist position. It's the Marxist

00:33:09.220 position. Let's destroy the old world and create a new world in its place. This is in the international,

00:33:16.380 in the heim, how do you call it, of communism. And it's also on the populist right. And, you know,

00:33:23.320 as a historian, I tend to be conservative in the deep sense of the word, in the Burkean sense of

00:33:28.980 the word. I mean, what you saw recently all over the world is the suicide of conservative parties

00:33:33.880 that they abandoned the conservative values and became revolutionary parties. Like the Republican

00:33:39.720 party today in the United States is a revolutionary party. Revolutionary in the sense-

00:33:44.780 It's a personality cult. Yeah.

00:33:45.420 Not just that. It says that, you know, all the institutions are rotten. They cannot be reformed.

00:33:50.340 We just need to destroy all of it and start again. Which, I mean, people can say, this is true.

00:33:55.640 This is false. Let's leave that aside. But just look at the structure of it. It's, this is what

00:34:01.840 revolutionary parties look like. They say things are so corrupted. Things are so out of order that

00:34:08.620 the only thing left to do is to destroy all the existing institutions and start from scratch,

00:34:14.040 which was the Leninist position, the Bolshevik position a hundred years ago. And now it's the

00:34:19.140 position of many so-called conservative parties. That the, the, the traditional insight of

00:34:25.180 conservatism is that yes, institutions are flawed. Institutions can be corrupted, but it takes

00:34:31.500 generations to build a functioning society, a functioning institution. Humans don't really

00:34:38.780 have the capacity to understand the full complexity of reality and to invent a perfect society from

00:34:46.880 scratch. It just can't be done. Every time we try to do it, it leads to disaster. Even worse than the

00:34:54.300 things that we try to, to, to, to, to, to correct. You cannot create a perfect society. So move more

00:35:01.220 slowly, be more respectful of the existing institutions and traditions. They need correction. They need

00:35:09.340 amendment. They need improvement, but just destroying them completely and starting from scratch.

00:35:14.880 I mean, you have hundreds of years of kind of previous corrections from real things that happened

00:35:22.120 in history that are baked into the system. Be very careful before you throw all of it out

00:35:28.560 and try to start from scratch. Okay. So what do we do about social media? Given that picture,

00:35:34.920 what, what advice do you have for correcting the obvious pathology we see here? I mean,

00:35:39.900 if you could give advice to Elon Musk or Mark Zuckerberg, I mean, one thing, two things,

00:35:45.800 corporations should be liable for the actions of their algorithms and only humans have freedom of

00:35:54.740 speech. Bots and algorithms do not have freedom of speech and they should never masquerade as humans.

00:36:03.140 So these are the two main things that, that are needed to, to, to correct social media.

00:36:07.900 So they're publishers. These platforms should be viewed as publishers and their algorithm,

00:36:12.960 the tuning of the algorithm is an editorial choice.

00:36:15.960 Absolutely. But this is strange to think about it, but one of the first jobs that was fully

00:36:22.680 automated was not taxi drivers. It was not textile workers. It was editors, news editors. It's amazing

00:36:29.320 to think about it. It was automated. The job that once belonged to Lenin and Mussolini is now being

00:36:35.820 done by algorithms. I mean, Lenin, before he was Soviet dictator, he was editor of a newspaper,

00:36:41.760 Iskra. And Mussolini also, he rose to power from being editor of the newspaper Avanti. So this was

00:36:48.800 like the promotion scale, the promotion ladder, editor of a newspaper, dictator of the country.

00:36:53.320 And this is for a good reason. I mean, the editors of, of news, they sit at one of the most important

00:36:59.840 junctions in society. They shape the conversation. They decide what people would be aware of, what

00:37:06.400 people would be discussing. And this is what some, one of the most important positions in society. And

00:37:14.240 now it's been taken over by algorithms because again, in, in Iskra, it was Lenin who decided

00:37:21.580 what would be the top story of the day, what would be the main headline. And on Facebook,

00:37:27.100 it's an algorithm deciding what is at the top of your feedback of your, of your news feed.

00:37:33.240 But both of those sound bad. So you're, if you're going to give people a choice between

00:37:36.440 Lenin and an algorithm, they're going to take the algorithm.

00:37:38.480 So the thing is that in, in, not in a dictatorship, in a democracy, newspapers and other news outlets

00:37:46.120 are liable for their decisions. Like if the editor of the New York times decides to publish

00:37:51.960 some conspiracy theory or fake news at the front page of the New York times, he cannot hide or she

00:38:00.300 cannot hide behind the argument, but free speech, there are some people who believe it's true.

00:38:05.900 So I put it on the front page of the New York times. No, your job as editor is not just to put

00:38:11.860 something random there or something that would please people. Your job is to fact check and to

00:38:18.680 take responsibility for these decisions and to make sure that if you publish something on the front page

00:38:25.280 of the New York times, you better be sure that this is accurate and this is responsible. And if you

00:38:31.680 don't know how to do it, then you're in the wrong job. And what, again, I will tell to, you know,

00:38:37.340 Facebook, to Twitter, to TikTok, be very, very careful before you censor human users. I mean, if a human

00:38:45.440 user decided to publish even a lie, even a fake news, even a conspiracy theory, I would be extremely

00:38:52.280 careful before I shut down their account or send those in. But if the, if my algorithm as the

00:39:00.580 corporation then decides to promote this particular piece of fake news or this particular conspiracy

00:39:08.080 theory, this is on me. This is on the company, not on the user.

00:39:12.280 The promotion, it's the amplification, not the fact that it exists on the network in the first

00:39:16.720 place because they can't possibly prevent, and they have billions of pieces of content arriving

00:39:22.200 every day, right? So they can't, they can't guarantee that there'll be no malicious lies or even child

00:39:27.740 pornography on their, on their network. And they need, it's the same way that, you know, people send

00:39:31.280 letters to the New York times every day. So it's not the position of the job of the New York times to

00:39:36.260 send them, but don't publish them on your front page unless you're sure that you did, you know,

00:39:42.160 and sometimes, okay, sometimes you would make mistakes, but still your job is to fact check

00:39:48.060 and to, to think about the, the, the, the implications of this story and then take a very

00:39:54.440 responsible decision about what you choose to promote. The other thing is that social media

00:40:01.480 should reserve freedom of speech to human beings, not to bots and to algorithms. And in particular,

00:40:09.040 we should ban fake humans, counterfeit humans. If a bot pretends to be a human, we should know about

00:40:17.520 it. Like if you see that some story on Twitter gains a lot of traction, a lot of traffic, and you,

00:40:25.660 you think to yourself, oh, lots of humans are interested in this story. This must be important.

00:40:32.400 I also want to know what everybody's talking about. And you also start, you click on it,

00:40:37.540 you also start commenting on it, but actually the entities that at least originally pushed this

00:40:46.120 story to the top of the conversation, they were not humans. They were bots working in the service of,

00:40:53.380 of Putin or whoever. This is wrong. We should not have fake humans shaping the human conversation.

00:41:02.060 You know, democracy is a conversation. Imagine it as a group of people standing in a circle,

00:41:08.360 talking with each other. Suddenly a group of robots join the circle and talk very loudly,

00:41:16.100 pretending to be humans.

00:41:16.880 Very persuasively. They pretend to be humans and you don't know. You can't tell who are the humans

00:41:22.520 and who are the robots. The conversation breaks down.

00:41:25.880 To be clear, you're not against bots of various kinds. You just think they should be declared as bots.

00:41:31.040 Absolutely. Again, if you have a medical bot and you want to consult with that bot about some

00:41:36.840 medical condition, I mean, soon we'll have AI doctors with capabilities far beyond human doctors.

00:41:42.620 I'm not against that. They can improve healthcare dramatically. They can help provide better healthcare

00:41:48.880 for billions of people. But when I talk with a non-human entity, I want to know that this is a

00:41:54.920 non-human entity, that I'm not talking with a human being. They are welcome to join the conversation

00:41:59.620 on condition that they don't masquerade as humans.

00:42:04.240 What you're arguing for, essentially, and I think this is a phrase you use in the book,

00:42:08.620 that what we need are benevolent networks that have a fiduciary responsibility to their users.

00:42:15.420 Yeah, and it's a very old principle. I mean, we don't need to invent anything new in this respect.

00:42:19.380 Like, if you think about your doctor or your therapist or your accountant or your lawyer,

00:42:25.640 for centuries, we already had these regulations and understanding that they have access to extremely

00:42:33.280 private information, to potentially explosive information about us that could maybe ruin our

00:42:40.240 life. And they have a fiduciary duty to use that information only for our interests,

00:42:46.920 except in very extreme circumstances when there is a crime or something. But our doctor,

00:42:52.380 for instance, cannot take my personal information and sell it to third parties for profits. And the

00:42:59.420 same principle should hold with our relationship with the high-tech giants. I mean, they should have

00:43:05.260 the same responsibilities.

00:43:06.920 How do you think about this trade-off between efficiency and inefficiency? Inefficiency sounds like

00:43:13.080 it's a bug. But as you point out in the book, there are places where it's a feature because it's a

00:43:17.340 bulwark against totalitarianism. And yet we want a certain kind of efficiency so as to be able to

00:43:23.440 find malicious actors and terrorists, et cetera. So how do you view that in a reasonably well-functioning

00:43:30.880 democracy that has institutions that are error-correcting both with respect to truth and with respect to

00:43:37.440 order? How would you, if you could get your hands on the dial of efficiency, how would you tune it?

00:43:44.100 I mean, that's the democratic conversation. We avoid the extremes and find the middle path,

00:43:48.900 and you're bound to make mistakes. So keep correcting your mistakes. It's not like there is a magic bullet

00:43:55.420 that solves it once and for all. So, you know, what is the right level of surveillance? What is the

00:44:02.020 right level of immigration? You know, this is what we have the democratic debate for. If you go for an

00:44:07.260 extreme position that, you know, humans have a right to immigrate to anywhere they like in as huge

00:44:16.380 as numbers as they like, this is completely unfeasible. Open borders, yeah. So again, how many

00:44:23.000 immigrants a country want to absorb and under what conditions, let's discuss. Different people have

00:44:29.580 different views. I don't think that people who want a more strict immigration policy, this immediately

00:44:35.280 turns them into fascists and Nazis. And similarly, people who want more lenient immigration policies,

00:44:41.720 less restrictive, that doesn't turn them immediately into traitors who want to destroy the country.

00:44:48.400 Let's have a conversation and try this policy and try that policy. It should not be a kind of all-out

00:44:55.640 war between good and evil. And the same goes for the level of surveillance and the same goes for the

00:45:01.500 level, again, of free speech. I mean, in all these cases, we need to find the middle path and it's

00:45:07.680 difficult. And we need to start with the assumption that we are not infallible and that other people

00:45:14.800 might have good ideas about these questions. Okay. So let's take this general framework that you've

00:45:20.420 sketched in your book and look at a few current events. I mean, there really is too much to talk

00:45:25.400 about, but we have the U.S. election. We have the ongoing war in Ukraine. We have the ongoing war

00:45:31.500 between Israel now on at least two fronts, Israel and her enemies. Let's start with the U.S. election.

00:45:39.220 How do you view our circumstance here? I mean, there has been, we really are the poster child for a lot

00:45:46.140 of the dysfunction you describe more generically in your book. I mean, there's just a pervasive sense

00:45:52.220 that I think social media is, is, doesn't fully explain it, but it certainly has amplified the

00:45:57.120 problem. There's a pervasive sense that we have, we've lost the capacity to speak to one another

00:46:02.740 about rather fundamental issues. And we're just hurtling towards some political catastrophe here.

00:46:10.260 So we have an, we have an election, which however it goes, it's quite plausible to imagine that half

00:46:17.380 the country won't believe the results, right? Given, given what has happened in recent years. So how do

00:46:23.400 we pull back from the brink here? So historically there are two big dangers for the survival of

00:46:28.640 democracies. And you can see both of them now in the U.S. One big danger is what we discussed earlier.

00:46:37.020 Democracy is this system when you give power to a person, to a party, to try out some policies

00:46:45.180 on condition that they give the power.

00:46:48.500 If you'd like to continue listening to this conversation, you'll need to subscribe at

00:46:52.900 samharris.org. Once you do, you'll get access to all full-length episodes of the Making Sense

00:46:58.200 podcast. The podcast is available to everyone through our scholarship program. So if you can't

00:47:03.400 afford a subscription, please request a free account on the website. The Making Sense podcast

00:47:08.440 is ad-free and relies entirely on listener support. And you can subscribe now at samharris.org.

Making Sense - Sam Harris - October 07, 2024

#386 — Information & Social Order

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript