Dear reader, I used the Eleven Labs AI platform to turn the text of yesterday’s article into a short podcast to make the content more accessible and to sort of bring it to life.

Let me know in the comments if you like this additional approach to our content and if you have any other thoughts at all.

NeoNarrative is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Transcript lightly edited.

Infovores: Today is March 14th, 2023, and my guest is University of Utah Dean of Humanities Hollis Robbins. Hollis writes frequently on the topic of higher education, drawing from a wealth of first-hand experience as a student, professor, and administrator that spans eight states, four academic disciplines, and a diverse range of institutions both public and private. We will focus our discussion today on the American University and the state of campus life. Hollis, thank you for joining me!

Hollis Robbins: Glad to be here. Eight states, goodness gracious! I hadn't counted them up in a while, so thank you.

Infovores: Thank you for sharing your expertise today. I'm really excited to hear what you have to say.

Let's start by talking about academic freedom. Many believe it is on the decline. As someone tasked with handling a lot of the relevant policies and disputes, how do you assess the health of academic freedom at the moment and what are the biggest challenges you face personally?

Hollis Robbins: First of all, I'm really delighted to have this conversation, and one of the distinctions I'd like to make is between academic freedom and free speech. I was fortunate to participate in that Stanford Academic Freedom Conference last November, and there were some really interesting guests there, some people who have made headlines for academic freedom disputes. Jordan Peterson was there, Amy Wax was there from Penn, Greg Lukianoff from FIRE was there.

One of my critiques of the discussion is that there hadn't been a differentiation made clear between academic freedom, which is the freedom for faculty to both pursue a subject and teach it in class. And this is an important bedrock of universities across the globe. The idea that you give researchers the freedom to pursue whatever line of inquiry they think will turn into gold, turn into innovations that will allow humanity to progress. So academic freedom has been upheld at American universities, at global universities for a very long time. Free speech is a right, a first amendment right to say what you'd like, but you can't say it everywhere, and it isn't tied to universities, and much of the disputes that we've seen bubble to the surface in terms of cancellations or violations are really free speech disputes. They are not academic freedom disputes. 

And I as a dean can talk about an academic freedom dispute, which is to say, you know, I've got faculty members who want to have a textbook that costs $300. But we have policies of limiting costs, for example. That's not an academic freedom dispute. But it's more of an academic freedom dispute than a first amendment dispute. Or let's just say a faculty member might want to teach his or her own book right and accrue some profits there. And I might raise my eyebrows, but that's perfectly within the realm of academic freedom. A faculty member came to me actually this morning wanting to screen a movie that's been sort of controversial. And his right to screen this film is within his academic freedom. Now if he does this and students protest, that's their free speech to protest.

And at the Stanford Conference, there was a great deal of “deans need to step in and stop mobs.” Well, I'm not going to stop a student's right to protest something. I'm going to defend the faculty member’s right to research something or show something in class. So that's a distinction that I think should continue to be made clear.

Infovores: That's very important background to have. I know personally as someone who is not administrating this, not often getting into the nitty-gritty details, I don't always make this distinction between a general culture or kind of a feeling that people have that they are more or less free to speak or that they might not be safe bringing up certain topics and the actual letter of the law that universities are tasked with carrying out.

How much though, do you think the cultural elements that worry people might flow out of the campus environment in some way? I know that a common theory I hear is that cancel culture started on campus and then gradually kind of spread out to the rest of society. Basically there's a bit of a lag period, but any new decrease in free speech or increase in censoriousness in broader society can ultimately trace its way back to the university. Do you give any credence to that?

Hollis Robbins: Well it’s interesting, and again I think it's important to make a distinction on campus between the classroom space and everything else because if you think of the campus and if you think of schools generally as sort of this combination of educational space and social space, what does that mean? There are some spaces like the cafeteria, the dining hall, the dorm, the quad, where a certain kind of rule applies. And then there's the classroom where a different kind of rule applies. And I think one of the challenges on campuses right now, or in this kind of conversation about free speech, is that distinctions aren't made for what's allowable, or what is defendable, or what I am going to defend as dean. I'm going to defend academic freedom and the freedom to query in a classroom, but the quad is something different.

I've written about Mario Savio from the free speech movement in 1964 at Berkeley, and this free speech movement was not in the classroom at all. It was the right to pamphlet about the war, about civil rights in the plaza, right? So they were literally protesting on campus for the freedom to do political speech outside the classroom. And I think that distinction is really, really important.

Now, is there a relationship? If some young student is out in the quad and then they walk into the classroom, there isn't sort of magic dust that says you're in a different place now, but there is a difference. And part of my duty as an administrator and part of the duty of faculty members, but I also think of the larger discourse is to make that clear.

On Trauma

Infovores: One of the insightful points you made in your op-ed piece about the Stanford Conference was that a lot of free speech related grievances coming from professors can trace to expressions of someone else's free speech. One place this often plays out is on social media, where people have pretty broad rights under the first amendment to criticize you in a lot of different ways which could reduce your comfort speaking without violating any enforceable law. And it wouldn't even necessarily be desirable to enforce such a law.

Hollis Robbins: I think this discourse of comfort is so interesting. One of the things that I really enjoyed experiencing at the Stanford Conference is that the discourse of trauma and discomfort was being articulated by people there who are usually people that poke fun at a different groups of people articulating discomfort and trauma.

As a scholar of of black history and black literature, most of my scholarly work addresses African Americans who have felt like outsiders in classrooms and in certain spaces and have felt inability to speak up, discomfort, and certainly trauma. And you hear articulations of this from a host of individuals on the campus, right now of different gender, ethnicity or orientation, so to hear those same exact evocations of trauma and feeling suppressed from the right, was of real interest. It was interesting to me, because on the one hand people say “why should we assume belonging?” and then the next moment it's “I felt so traumatized I didn't belong”, so perhaps we can help all of us try to avoid traumatic experiences in the classroom. One of the things that I didn't have in my essay, but I spoke a lot about was was just kindness and civility to one another. And I think that goes a long way.

Infovores: I think that point is really well taken actually. Over the past several years, I think I've seen certain tables turn and you see—this happens all the time in politics actually, if you're paying attention—But you see people on the left take arguments from the right, and you see people on the right take arguments from the left, and it's all moving back and forth. In a way, it's really good for getting some clarity about the merit behind some of these these principles.

One thing I think back to is Christine Blasey Ford testifying, and a lot of people's perception of that, probably myself included to some extent, was that she didn't always come across as the most credible witness at times. And I think that you could take that immediately and say therefore there's nothing to this allegation. But another interpretation could be that perhaps she's genuinely traumatized in some way that impedes her from being her own best advocate. And that argument has kind of a left wing flavor to it, but I think it applies in all kinds of situations where I think people are inclined to dismiss, you know, right wing professors who are under siege as like “Oh, they're just a baby”, or they have no validity to what they're experiencing, and that's not true either. Like you can't just say that a priori. And when you really look into some of these cases, there are genuine things to sympathize with and understand there.

Hollis Robbins: I think it's quite clear that Jordan Peterson has been traumatized by what he has been through. One would hope that would encourage an empathy with others’ trauma, we will just see.

Infovores: Stepping back from the trauma point a little bit, I've also seen more and more publications like First Things or National Review make arguments along the lines of “well, now the culture is really being dominated by women in higher education”, for example, because they're a majority of of degree winners, or a majority of certain leadership positions. And you know it is interesting. It does speak to there being a validity that was overlooked in opposite side debates in the past.

On the Stanford Law Protests

Hollis Robbins: Well, it's a funny time in higher ed, and I didn't know if you were going to bring up Florida or some of what's happening…

Infovores: I will, actually I was just going to get into that. 

Hollis Robbins: Well, I'll let you pose the question then because I'm interested in that.

Infovores: So to press deeper into some of these things and make it a little bit more concrete, I have basically one recent event from each side of the aisle to look at and get your thoughts. 

The first one actually happened just this week with this incident at Stanford Law, where an appellate judge was invited to speak. I don't know if you saw this? 

Hollis Robbins: Oh, I saw everything about it. 

Infovores: Yes, basically he got shouted down. And then the law school dean apologized. And then a lot of activists showed up to the dean’s classroom while she was teaching, and basically intimidated her and the few students who didn't join the protest. What steps do you think the dean should take in these types of situations? It seems like this clearly crosses a line between the right to protest and disruption and intimidation of others’ speech. What are your thoughts there? 

Hollis Robbins: It was an interesting… and I don't know if I saw all of the videos, I saw a video of Judge Duncan sort of having a rough encounter with some students who clearly had some issues with his rulings. I didn't know whether this was an actual class that they were required to be there or were there in order to protest. And again, this matters back to this question of whether something is in academic space or not, right? If somebody is giving a guest lecture right in a class where the assumption is, this is a learning space, and you're required to be there, otherwise you lose points or something, then that's one kind of space. If it is an optional guest lecture of a person being invited by the Federalist Society, I believe you know it's a kind of quasi-academic space, you're not required to be there, and so you could just not show up. So that's the first sort of analysis that I would put in. 

Second of all, clearly there are strong feelings about his rulings. The second video I saw was some shouting back and forth, and then a dean of students or a diversity dean, I'm forgetting her name, came in and read a prepared statement. As an academic dean, I thought that was odd. I thought that was not… clearly there were things that were prepared, clearly she was taking sides with the students, and clearly she was not doing what an administrator should do, which is actually, let us talk about the rules of the space. Is it an academic classroom or was it an extracurricular classroom? And then I saw the apology letter signed by the law school dean, which is an academic dean and the president of the university to the judge. And that seemed to be fine, acknowledging that certain things shouldn't have happened. 

But I think the central point to get back to what I was saying earlier is, what was this space? Because if this space was the kind of space that Mario Savio was fighting for, a kind of space to have political opinions or some sort of extracurricular exchange, then everybody's got a right. It's a free for all. Everybody has a right to do what they want to do. You can protest, you can counter protest, even if it's held in a classroom. But if it's an educational space where it's part of the curriculum, a different set of rules apply and I think that would have made a lot of things clearer.

Infovores: I take your response as speaking primarily to the initial speaking event that was protested, and I take your point that the protesters have a lot of rights to voice their opposition to  particular viewpoints as well.

Hollis Robbins: If it's not a classroom, right? Let's define what it was. Was this like the quad or was this an actual curricular space with curricular goals in mind? And that was not clear to me.

Infovores: Yeah, that makes sense and I think I'm mostly with you there, but let me clarify. I was actually mostly thinking of the very most recent events where there was a reaction to the apology given. And so the law school dean also teaches her own constitutional law class and apparently student activists arrived in her classroom and were masked. And they were glaring at her and the other students and posting things all over the the classroom to disrupt, and then when she finished teaching basically a large percentage of the broader law class, not just that one classroom, was there and they formed a human corridor basically shepherding them from the exit of the classroom to the exit of the building, forcing them to do like a walk of shame through a lot of protesters. What would you add given that follow-up?

Hollis Robbins: I was going to say I did not see this. So protesting an actual curricular classroom, I would not stand for that. But you know, what are you going to do if you can't say look, you're breaking the rules here? People are going to break the rules. So that I would find out of line because part of attending university or law school is that it's a contract, right? And these contracts are upheld by the Department of Education if you're taking any Federal money, which means you have to teach your class. It has to be this many weeks. You have to actually deliver what the student has paid for. In an emergency, in floods and fires and things, you can't shut down a institution because you have to continue to deliver instruction somehow, because students have paid for it. So to disrupt the delivery of a service is a bad thing. You can call it a kind of theft of somebody's educational opportunity. So that's one thing. 

The walk of shame, you know once you step outside the classroom, if nobody's physically hurt it's probably very annoying, but I think you know it's tough being a dean, and at some point in time things happen. There's the President of Connecticut College, I believe is under fire right now with students calling for her to step down for holding a fundraiser in a place in Florida with a segregation history. These things happen. Is she okay?

Infovores: So far as I've heard, I don’t think she was physically harmed. I'm not sure that she's had any official response to it yet.

Hollis Robbins: I'll take a look, but again I think it's really important that we make distinctions between the classroom space and the extracurricular space, because there's free speech in one, but there's a paid delivery of curriculum in the other.

On the New College Takeover

Infovores: Gotcha. So considering an incident coming from the other side of the aisle, we could talk about the New College takeover in Florida. In January, as you know, Governor Ron DeSantis replaced six of the 13 members on the school’s board of trustees in an effort to transform the institution into a “Hillsdale of the South”, which I don't think were his words but one of his representative’s words. What troubles you about this event, and what does it portend for the future?

Hollis Robbins: Well it's a complicated story. So I was asked to apply for that presidency, and I did. I wrote a letter and submitted my application materials. I didn't get an interview. That's fine, these things happen. But in my letter there was a really big paragraph about what it would mean to be the president of New College of Florida with this administration. What it would mean to be a very well respected public liberal arts college, that is paid for by taxpayers in this political climate. Because if you're a public institution, as I am right now and also my last institution in the Cal State system was a public institution, you should always be cognizant that taxpayers are paying for this, and that you are a government employee.

I happen to have seen Patricia Okker's letter, because if you apply for things then you get access to others’ letters, and she mentions nothing about the State capital, or the legislature, or the governor. She mentions nothing about how this is a taxpayer-funded college, and I'm not saying that the loss of her position was deserved. But she had an opportunity to engage, everybody saw this coming, to talk about the value of a liberal arts education, to talk about the value of being part of the public system to engage at a pretty high level. And she didn't.

On UATX

Infovores: Why do you think that is? You have no inside knowledge as to why she didn't do that, but why do you think she may have declined to take that opportunity?

Hollis Robbins: Well, I think if I have one complaint about some recent coverage, like in the New Yorker piece or other pieces that are going around about the decline in the humanities, or the problem of the liberal arts, or problems with climate on campus… let's talk about the Hillsdale part, or let's talk about the creation of the University of Austin.

There's this sort of produced narrative that liberal arts is so messed up. We need a place that is going to have old-fashioned fundamental values, and enough of this sort of woke liberal humanities education. And if you look at the history of New College—I've had a number of friends who have gone there—it's obviously very left. But that doesn't mean the education isn't good. There's a creation of this sort of discourse, that being left and having a classical, fundamental education are somehow at odds. People say I'm going to get away from X, and I'm going to go back to the classics. Well frankly, there's no better place to discuss LGBTQ issues than Greek literature.

Infovores: Yeah, I can kind of see that.

Hollis Robbins: And there is no better place to discuss gender fluidity than in a Shakespeare class.

Infovores: Historically, there are antecedents to some degree.

Hollis Robbins: Right. And so I'm launching a great books program here at the University of Utah, in our College of Humanities, and it's a fantastic course. We've got Kafka, we've got Baldwin, we've got some Virginia Woolf, and it's an opportunity for our students to study great works of literature. But because we're studying great works of literature, we're also dealing with the humanities questions of the day about race, gender, and belonging. These are not easy topics to tackle, and it's not an either/or situation. Obviously I dodged a bullet not being at New College of Florida. But that's what I was hoping right would be said, is that we don't have to think about this as either/or and I think some of the discourse coming out from the new board members about turning it into a Hillsdale… well, what does it actually mean What does that mean about the quality of instruction that you're delivering? What does that mean about the knowledge base of your faculty? This is not a left/right question. It's just a framing question.

Infovores: Yeah, it's interesting to see how these debates that start out ostensibly about maximizing the student experience and improving the generation of knowledge sometimes kind of dovetail into efforts that almost seem geared towards sort of taking down education.

Hollis Robbins: Well I was just going to say, James Weldon Johnson's great anthem, “Lift Every Voice and Sing”, was sung at the Super Bowl this past year, right? And you know some critiques were like why was this song played, which is sometimes called the Black National Anthem. Well, James Weldon Johnson spent a great deal of time in Florida learning Greek and Latin, this was in 1888, he learned from a shoemaker by the light of an oil lamp. There is this idea that diverse literature is somehow not classical, that fundamental humanities is not caught up in those same concerns about learning and timeless texts.

That's a matter of people not understanding, not knowing who James Weldon Johnson was, and why he wrote that particular song, and his educational background of thinking that it was going to be very important to him to find a tutor in Florida to teach Greek and Latin when he was home quarantined from some illness.

There is this idea that Peter Thiel talks about all the time from the 1980s, the sort of polluting the core curriculum, the Western Civ with multicultural texts, is frankly a ridiculous charge. When some of the texts that he was opposed to like "I, Rigoberta Menchú." It's actually a very Burkean, in some ways a very right-wing text. If you read it and you take it seriously, that text is about American freedom, about liberty, about equality, about striving, about having a place of one's own to farm. These ideas are not simply in Western texts. They appear everywhere.

Infovores: Yeah, I totally buy into a lot of that. I mean just fundamentally, when talking about concepts of freedom and rights, that discussion is very helpfully informed by African-American history and a lot of the things that you've written about and studied with respect to the abolition movement.

Hollis Robbins: Yes and so there are ways, especially in Florida, to say… look the anthology I put out in 2017 with Henry Louis Gates Jr., almost every one of these nineteenth century African American women writers, most of the texts these women wrote that we compiled in that book, I would consider very conservative today in terms of what they wanted. Education, family, freedom, autonomy, justice, rights. These are the same things that the nation's founders wanted.

The idea that a text that I wrote with women who are basically wanting what everybody wants could somehow be banned is crazy to me. It's absolutely crazy. And you think about Tocqueville’s great book, “Democracy in America”, he famously never spoke to a black woman. In his tour around the United States, he talked to black men, he talked to women, but no black women. And so part of what I suggest is that if you're going to teach Democracy in America, you should teach this anthology, and you'll see resonances between these two texts.

Why Conservatives Distrust the Academy

Infovores: I take a lot of what you're saying as very true and valuable. If I were to push back just a little bit, I feel like you framed it almost as a matter of directly accessing these texts and people are just ignorant of the value that can come out of these approaches. And I think there's probably something to that. The whole reason why you go to college is because there are things that you don't know, and you can learn new things of course.

But I feel like the fundamental issue for a lot of people is not so much the content the way that you're talking about it, as it is a matter of trust. Gradually over time, the people that are teaching and propounding their own viewpoints and interpreting these texts… they're not helping you to see how Burke was actually a really great thinker, and that there's merit to conservative ideas. They're approaching it in a tendentious way, and it's perceived that there's really no space for someone to explore conservative interpretations or non-woke interpretations to different things, or to challenge the orthodoxy based on the curriculum. So it's really almost more a matter of personnel. What would you say to that?

Hollis Robbins: Well, you know you raise a good point. That's a good pushback, in fact. And there's a way to say to go back to the porous boundaries between the classroom and everything else right, because at a certain point I can say all these things, and believe all these things in an academic space right? But then I can walk outside where there's yet another police shooting of a black man. Like with George Floyd, or I was in Baltimore at Johns Hopkins when Freddie Gray was murdered, and the protests that erupted thereafter because at a certain point, racism and police violence occur right? These are also facts of American life, and it puts pressure on having certain conversations in the classroom.

As much as I'm sitting here talking about distinctions between a classroom space and what happens outside, there are classroom spaces that I think easily lend themselves to the politics of what's going on outside because you're already discussing things like lynching. And it's not going to happen in a chemistry class, perhaps it's not going to happen in the history of Chinese Empire, but it will happen in an American history class or literature class, and it affects the teaching of texts such that they are politicized. And they're politicized forward to 2020, or 2021, or 2022, or 2023 right? They're not looking backwards to Burke, even though they might have been written in an evocation of Burke. They're pressed into service about today's problems, and that will happen.

And the answer to that is wishing perhaps, or wondering what can be done to address racism and violence in America.

Infovores: It makes a lot of sense what you're saying. Do you think…

Hollis Robbins: If you still want to push back though, you can still push back!

Infovores: In some ways what you're saying deepens the problem potentially of having kind of a monolithic culture, or an agenda setting power that’s not really being shared in the university. It’s perceived as being completely controlled by whatever MSNBC is talking about and not admitting other angles. And of course there have been various people in the economics profession that have challenged certain narratives about racial profiling or about police violence and that have a very hard time creating any sort of space for a discussion there, which seems like all the more important discussion to have given the controversies happening in the present as you describe.

Hollis Robbins: Well, I mean you can go back to this question of the ivory tower. It is both sort of an insult to say you're not paying attention to what's going on in the streets, but then you can also make the complaint about a university that is not ivory tower enough, where all it is doing is reacting to what's happening in the streets. And I think this has been a challenge for universities forever. When you read some Lionel Trilling for example, the great critic, and you realize the things that he's writing about were happening at the same time as when Martin Luther King was agitating you think to yourself, “were you not noticing that when you talk about American striving for freedom?” Or how is it that if you read the Atlantic Monthly from the 1890s right after Plessy v. Ferguson, after a really huge rise in lynchings, and it publishes all sorts of things about Italian freedom fighters and isn't it glorious that people are fighting for their freedom. And you think, “are you paying no attention?”

On Academic Excellence

Infovores: Well taken.

So you mentioned the University of Austin and you made various conceptual critiques of the way that they frame their approach. I wonder what you think about the broader enterprise of creating new universities to compete with the existing lot, and if you were to embark on a project like that, how would it differ from what Pano Kanelos and others are building?

Hollis Robbins: Well I guess the real question is, what is excellence? And they’re saying that something is too left or too right or too this or too that, but left out of that conversation is excellence. If I were starting a new university, which I have no interest in doing, but if I did the question Isn't who's been tossed out at previous institutions. The question is, who's really, really good? And what do we mean by really good? Who is changing the conversation? What faculty members are pushing knowledge in new directions?

That to me, even as a dean, is my primary responsibility. When I grant a faculty member a sabbatical to go finish a book, or I give a faculty member funding to go visit an archive somewhere, I want to know what are you doing to really push knowledge forward? What are you doing that nobody else is thinking of?

That’s its own sort of thing in the humanities. Certainly this makes more sense in some ways to your listeners from the sciences, but there is such a thing as new knowledge.

I don't know if you've seen Paul Reeve’s new book, “Let's Talk about Race and Priesthood”?

Infovores: Oh, at first I was thinking of Richard Reeves. But I think there was an article you tweeted about this.

Hollis Robbins: Yeah, so he's the chair of the history department here, and this is Deseret Book, it's not an academic press. But he wanted to to write about it, and they asked him do you want to write about race and the LDS Church? And he said, sure but i'd like to make this speech by Brigham Young from 1852 kind of the center of the book. And I'm not telling this story completely right, but they said something like, what speech? And he’s like, well let me tell you about this page, and it had been taken down in Pitman shorthand, lots of things in the in the nineteenth century were taken down that way…

Infovores: A faster way of taking notes basically.

Hollis Robbins: Yes, exactly. Certainly better than any note taking software now. But he had turned the speech into sentences and paragraphs, and he understood that the history of race in the LDS Church if you go back to Joseph Smith—and I don't need to be telling you this—was that God doesn't see individuals or doesn't see distinctions between people. Everybody's welcome.

Infovores: Joseph Smith ordained black members.

Hollis Robbins: Right. So when did this change happen that banned black members from the priesthood? Well it basically happened with Brigham Young, and it was put forward in this speech that changed everything until the revelations of 1978. And you know the amount of work required to do this, and patience and working with librarians, and having the trust of keepers of knowledge and writing a very emotional and compelling book about the history of basically solidifying racist views that were not in accordance with the original principles. Setting it out chapter by chapter and saying, look we've got to undo this right. And here's the history, and here's what what it shows. That takes a certain amount of training, a certain amount of understanding your larger conceptual framework of race, understanding history and book publishing. This wasn't going to be an academic press, but I bring this up to say that this is a book that will change knowledge, it may change a heck of a lot of things about this particular community going forward. What you want in a faculty are professors who are going to do that.

Infovores: I definitely see the point that merit could be more directly emphasized rather than just diversity of opinion for its own sake of course, because you could pick random people off the street, and they would have diverse opinions in some sense, but wouldn't necessarily have the the ability to forward human knowledge. 

That said, I feel that there are a lot of pretty interesting and accomplished thinkers that are involved with University of Austin. Are there any that stand out to you as being people that you'd put in that category of really forwarding knowledge?

Hollis Robbins: I'm not sure I know enough to say whether they're the leading lights in their field. I have been very lucky over the last 20 years to work with Henry Louis Gates, Jr. who is the leading light of African-American studies pretty much. There are many, many wonderful scholars. but in terms of creating an academic discipline, in terms of the the number of fellowships, scholars, students, and generations that he has supported, that he has created the field, the number of books with his name on it…

So my standard of excellence is admittedly very, very high. Is anybody close to that height at UATX? I don't think so.

Outsider Perspective

Infovores: Henry Louis Gates himself has of course faced cancellation pressure, so it's interesting you mentioned him. 

Maybe I can connect this to something you said earlier I found interesting. You were talking about an outsider perspective, and I think in many ways you kind of embody that outsider perspective because you are a leading scholar in the area of African American studies, and you're not African American yourself. You're a white woman, which at least in some settings I feel is sort of pushed against, or it feels like there might be an extra barrier there. How would you react to that assessment?

Hollis Robbins: Well it has been a bit of a challenge, and I had some worries early on in my career. I was in my fourth year of a 5 year program PhD program at Princeton, when I met Gates and started working with him on a narrative he found by an enslaved woman named Hannah Crafts and completely changed fields, which let me just say to any young person in a Phd program, do not change fields! 

Infovores: Hard to do under any circumstances. 

Hollis Robbins: Yeah no, it's just… it's no good.

[laughter]

But my particular expertise is in the Western literature that black writers in the nineteenth century were reading. And so to understand, how I got to know Gates is that he had found this narrative and this writer Hannah Crafts, who a scholar named Gregg Hecimovich identified in 2013, was cribbing from Charles Dickens. At one point in her discussion of her experience she was describing slave huts, the really squalid conditions in which she was living, and she borrows and sort of signifies with, changing some words of Dickens's description of the London slums from his novel, Bleak House. And part of the story of this wonderful writer is how she came to know Dickens, which she had overheard some recitations it turns out enough times because you know that's what you used to do in school, you had to recite long, long passages, and recognizing something in Dickens's telling that described her reality.

That's interesting, right. And so I started working with him on that book. But then, we did an edition of Uncle Tom's Cabin together, looking at how that very complicated novel operated and the reactions to it. We have collaborated on several books. Right now, we're working on a book about Phillis Wheatley who was an eighteenth-century poet. She quotes Dryden, Pope, and Milton so if you're not really steeped in that tradition, you can't really understand those works of writing.

And so there's not a lot of people who do this particular kind of work. So I have a little niche.

Infovores: I think that's a great example of a piece of knowledge that's valuable and interesting. I mean just recognizing that there's this very deep connection between African American tradition in literature and Charles Dickens, is a great contribution that you wouldn't have been able to make if you had been shut out from entering into that area of study. If you were starting out today, do you think that it would be possible to make that kind of transition?

Hollis Robbins: No, and I think that the thing where I am an outsider, and I'm not sure I would call myself that, is that I see myself at this pivot point between these two conversations. Because on the one hand I'm fairly old-fashioned in both my education and my training. My PhD at Princeton is very much in nineteenth century British and American literature, and you can't get more traditional than that.

But when I moved into the field of African American literature I saw how it wasn't very different. I mean there are things that are very different, but passionate fights for freedom and autonomy and bodily autonomy and rights, when you read Frederick Douglass, you realize this is one of the most American works of literature that can be imagined. And yet it is left off so many Western tradition lists. And to me that is just…

Infovores: It’s a tragedy. It’s great writing, it’s extremely historically significant…

Hollis Robbins: Right! So what’s that about? Henry Louis Gates had a piece in the New York Times recently about the effort to keep the curriculum white for a long time. and so efforts to push back against that are no different. You started out by saying that the discourse of the left and the discourse of the right begin to you know…

Infovores: Often they kind of meld and switch places.

Hollis Robbins: Yeah and so you know, one wishes that one could go back to Peter Thiel in the 1980s at Stanford and say read a little Douglass. You'd like him.

Infovores: Standing up for Thiel a little bit, he did speak to Jerry Bowyer about this topic to some extent. He was pushing back a lot against the idea that because we're in this dispute, this culture war, that therefore we're going to almost forget about important lessons from the holocaust, or we're going to forget about important lessons from African American history, or we're going to deny that racism exists, when of course racism exists, and we shouldn't deny that that's the case.

But I think that a lot of people just feel that the education system is so at odds with their values and viewpoints that it's hard for them not to see it as the enemy. And I think that's where a lot of it's coming from. 

It was interesting in your panel discussion during the free speech conference, Greg Lukianoff was in that discussion with you, and at one point— I mean a lot of the discussion was more directly to the point of what FIRE is doing, and how FIRE helps professors in certain circumstances to overcome violations of their scholarship—but at one point there is a little bit of a tangent where he started talking about education signaling theory, and the idea that education is actually not at the end of the day contributing very much of value which I thought was kind of an interesting digression because it's neither here nor there within the free speech discussion to an extent.

But I feel like you see, as time has gone on, more and more people are kind of… I don't know if it's coping to an environment where they don't feel welcome in higher education, or they don't feel like education is trying to help them specifically, and that makes them more receptive to the idea that education is just not adding a lot of value in general, or that it's not a noble enterprise. But I do think it was kind of revealing of what a lot of people are thinking at this moment.

On ChatGPT

Hollis Robbins: Well in some ways, I think about ChatGPT or GPT 4 now, we’ve got these machines that have just an incredible amount of knowledge. An incredible amount of knowledge. I don’t need to dig something up, I’ve got Wikipedia, I can ask GPT, I don't need to memorize my state capitals anymore, or what have you, right? 

But what is it that higher ed is doing? What is the value added that we're having, and that we're offering, what difference are we making right now? You're in a PhD program, do you have to do a dissertation?

Infovores: Yeah.

Hollis Robbins: And what is your dissertation going to be on, do you think?

Infovores: It's not determined yet. I have some ideas. 

Hollis Robbins: It's all good right. But is the pressure on you in your field to do original work?

Infovores: Yeah, definitely.

Hollis Robbins: Right. So this question of what is actual, original work comes down to what are you doing that nobody has done before? What has nobody thought of before, or maybe you have a data set that is new. This question of newness, of actually pushing the edges of knowledge, is what you are being trained for.

And good faculty will keep doing that. I can't even imagine writing a book that does the same thing. If somebody's going to write a book on Shakespeare right now, it better darn well say something… which is really hard when you're talking about somebody like Shakespeare, because so much has been written on him.

A lot of the work that I've done with Skip Gates is on newly found text that nobody has ever done anything on. So the bar is a little lower, because we're the first people that say something about it.

I think the people that critique higher ed don’t generally have a real grasp about what originality is, and what progress is, and why it matters to always be marching to the end and beyond of what is known. 

I will also say, trying to be optimistic and pleasant here, that there's a fair number of faculty members teaching in some excellent universities as we speak who are not as interested in originality as they were when they were finishing their PhD program. The job of a dean is to be like, “Hello! Can you please go back to being original?” Can you have that  spark of an idea that's been sitting there because you've got a high teaching load and you're tired and you've got too much advising? Can you go there, please? Can you  really think of yourself as cutting edge again? And that's what I try to do.

Infovores: How does that last point relate to discussions about tenure reform and peer review?

Hollis Robbins: Well, again, I think that the issue is, what counts as new and what counts as peer review? I do a lot of reading of applications for various fellowships, prestigious fellowships, and you know I read 10 or 15 applications every semester. I would say it's definitely a minority of people who even position themselves as doing new work. There's a lot of faculty that put themselves forward as doing important social justice work, and there's nothing wrong with that; social justice work is great! But can it be new and focused on social justice? Can we focus on the newness part of it? What are you doing that other people haven't done yet? So that's what I’m pushing. Tenure should be about newness and continued focus on that kind of excellence.

Infovores: What lessons should the academy draw from ChatGPT going forward and how can we harness it in a way that creates more new and original research in contrast to regurgitated but grammatically correct text?

Hollis Robbins: Well, it's so funny. I asked it once I said, “how influential is Hollis Robbins on the Emperor's New Clothes?”

Given my background and my scholarship, it said that I wrote an important piece on race and the Emperor's New Clothes and that I was talking about things like royal privilege, white privilege, and those people who don't have access to that kind of fine tapestry, and that the final scene was the young boy calling out the privilege of the king. Now I never wrote anything like that, but I could have!

Infovores: Maybe you will.

Hollis Robbins: Maybe I will now.

[laughter]

It was just funny to see what kinds of assumptions it makes, a kind of scholarship “Minority Report.” What will I write next? So maybe I can sort of put some of my faculty members’ previous works into ChatGPT or GPT  4, and say, “what should this person write next?” Maybe it will tell you what your dissertation topic should be.

Infovores: I'll ask it every day.

[laughter]

Hollis Robbins: It's really excellent. I’m so supportive, and I've been working with some of the folks at Anthropic and got access to their model called Claude a couple of months ago, and compared what the guard rails on are what it allows you to ask and where it doesn't want to go. And you know, I don't know if I've been as helpful to them as they are to me, but I have tried to articulate the position which I think I'm in a minority of humanities deans in not being worried.

Infovores: You mentioned the Emperor's New Clothes, so this might be a fun way to to wrap up. I love your Straussian reading of the Emperor’s New Clothes, I actually attended your seminar with interintellect on the subject, and I was wondering what connections or interpretations you might be able to draw from that fable with respect to free speech?

Hollis Robbins: Oh with respect to free speech, that's a good question to ask.

You know one of the reasons that one says that there isn't free speech in the classroom is that you can't get up and say anything, you can't do anything, it's got to be within the sort of bounds of polite academic inquiry. And years ago there was this story of some absent-minded physics professor, who would wipe chalk on the blackboard, and just by the end of it he would be covered in white chalk dust, and students never said anything. They would just make fun of him behind his back, but they never said anything during class like, “you know you've kind of got chalk everywhere.”

Infovores: You're a mess.

Hollis Robbins: Right. Years later, the professor said, “I did that on purpose because they're not speaking.” You get in habits of not speaking. You get in habits of listening, of deciding, you know what in another setting I might say, “Buddy, you're kind of a mess.” But now, I'm just here to learn physics.

And that self-silencing was actually good for classroom learning. Apparently he was a very successful teacher, and students remembered a heck of a lot that was going on, and I think about that deliberate non-speech, which is different from where we started. We started on sort of self-suppression and suppressing speech as a bad thing, as a testimony to how things are bad in higher ed. People don't feel like they have free speech in the classroom. Well in my reading of Emperor’s New Clothes, “time passed merrily in the land”—things were going pretty good. And if all that mattered to get transparent government to actually work was not pointing out the transparency of his clothes, that may not be such a bad thing. Speech isn't always the single most important thing.

Infovores: Hollis Robbins, thank you so much for taking the time to talk with me. I'd like to recommend that people go to your Twitter account. You always have interesting things to say, you keep up with all the latest discussions on topics from tech to literature to race relations, and I really appreciate you being on the show today.

Hollis Robbins: Thank you. This has been really fun.

Thanks for reading! If you enjoyed this, check out my other writing at Infovores Newsletter or follow @ageofinfovores to keep up with me. On related topics, “A Grain of Salt for FIRE Free Speech Rankings” and “Is the Book of Mormon Straussian?” may be of particular interest.

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Transcript edited for clarity, Spotify link here.

Infovores  00:03

Hello and welcome to Time Well Spent, a place where the most brilliant minds in the world take on the toughest questions in science, politics, technology, and much more. My guest today is Robin Hanson—economist, futurist, and author of the influential blog Overcoming Bias, which as of this week is available on Substack. Robin, thank you for joining me! 

Robin Hanson  00:24

Thanks for having me, I'm not sure we can live up to this most brilliant minds thing, but let's give it a try. 

Infovores  00:29

Let’s give it our best shot.

So let’s talk about the recent developments in AI. At least in my experience, I see these sorts of things emerging as a trendy set of buzzwords getting used more and more in the last 5 to 10 years. And then it really only breaks through into the mainstream, "current thing”-level status in the last few months. But you have been researching and pondering these things much longer than most people talking about AI now. What would you most like to change about the current conversation coming from that perspective?

Robin Hanson 1:04

Well, historical perspective I guess.

So we've seen similar bursts of interest and concern about AI, going back a long way. We're now maybe near the peak of the current burst, but we had another one in say, the late 1980s/1990s. We had one in the 1960s, plausibly one in the 1930s as well although they didn't call it AI, they called it automation. And even plausibly one in the 1820s with the rise of the Industrial Revolution and concern about automation there. And we've just consistently seen over this whole time, the scenario is that some new demo and technology appears, and it starts to have impact. And then people are immediately trying to envision how far this will go, how fast, and they're saying, gee, this looks unprecedented. This could soon take over all the jobs and replace humans. And that's what people have said every time over this long period. In previous things, we've had presidential commissions and big news and media coverage about the unprecedented new abilities that have just appeared. And that's what we have again. So I'd like people to know about how that's played out each time.

Typically there are some concrete developments and some actual useful applications. And then there's more investment in terms of new firms and existing firms making new projects and students going into interrelated fields. And of course usually the fears are overblown, or the expectations are too high. And what you have is a new set of technologies that diffuses out into practice, but not a revolution. And that should be placed in the context of just say, the last seven years of computer science and computer applications, where we've had relatively steady progress. But every once in a while, we have a  local breakthrough and that gets attention. But the distribution of innovation in terms of their size is almost all innovations in small things, relatively few in medium-sized things, relatively little in big things. And that's just the distribution of innovation in general, which seems to apply to most fields. But AI has this unusual potential to help people envision a much bigger impact of the thing they see in front of them. Everybody's saying that someday there's going to be machines to do everything. Could this thing I see right in front of me be the thing that I hope for or fear? And that's what's different about AI is people project these huge potential advances on the thing they see in front of them. 

Infovores  04:16

Makes sense. A lot of what you're saying I think applies really to any sort of technological fad.

Robin Hanson  04:29

Well if it inspired such big visions, right? So most technological innovations don't make people think of rampaging robots. Whereas AI does and so that's the difference here. I mean, just in terms of actuality most innovation is lots of small things, and every once in a while, a big thing. But different kinds of innovations just project different images people have about what could happen.

Infovores  04:52

So the next big breakthrough in building a better mousetrap is not going to inspire the same vision because, sure it could revolutionize the field of mouse-trapping and eliminate all mice problems completely. But it's not going to threaten anybody's sense of the world being run by robots instead of humans. The interest is especially susceptible there.

Part of the motivation for asking this question was a post you wrote a few years back about how telecommuting was a really neglected trend. I think that telecommuting was kind of the last big technology breakthrough that everyone got really excited about, and now I wonder if we compare those two things side by side, are we still sleeping on remote work to some extent?

Robin Hanson  05:48

Certainly remote work has had a much bigger impact on society in the last 10 years than AI has. That’s hands-down true. But I think I still think we haven't realized most of the potential of remote work. It's still mostly there to be realized. So I think over the next 30 years, we will see it realized, but so far, we have just seen the simple thing of people working from home instead of the office, without that much changing how things are done.

Infovores  06:21

We haven’t restructured institutions.

Robin Hanson  06:23

There's a famous story about electrification, which was that originally, people had factories with non-electric power, and then they had electricity but they basically plugged the electricity into the same sort of factories they had before. And that didn't give a very large productivity boost.

But then people spent 30 years to reorganize factories to take advantage of electricity. And the main thing was, you could have a lot smaller motors, many smaller motors instead of a few big motors which you drove power around through some big bands. And then there was huge benefits from electricity once you reorganized factories to take advantage of it, but that was what was required. And so similarly, I would say for remote work, it'll be the reorganization of work around remote work that will make the big productivity gains. 

Infovores  07:18

And what do you see as kind of the time horizon for that? I know you mentioned that you wouldn't be surprised if 30 years from now, remote work were the most important trend that people weren't really keyed into fully. Do you think that's still true? Even with AI, if we go out 30 years from now and compare them side by side, will remote work have more impact than AI? 

Robin Hanson  07:49

Possibly yeah. You have to envision…what I have in mind for remote work is… so imagine, today a plumber comes to your office, and they are human who come with their bag of tools. In the future, a plumber might be an avatar that sits downstairs in an apartment complex that anybody could use whenever they need it.

And it shows up with a standard set of tools, and then a plumber uses it from long-distance. But then each little step of the plumbing job a different person swaps into the avatar to do so we ended up with as much specialization as you might have as an automobile factory. And as you know, automobile factories are vastly more productive than independent small team of four people putting together a car, because of how much they can specialize in each particular task. And so, if there's firms of 1000 plumber specialists that each would do one little piece of a plumbing job, they can each swap in at their moment when the avatar needs to do that next thing, and they swap into the avatar, and they have to do that task. And now you're getting sort of car factory levels of specialization, in something that today is done by a generalist. That's the kind of productivity gains that I'm seeing as huge. It's the opportunity for specialization. Basically a city’s scale of specialization even in remote places.

Infovores  09:21

Gotcha… specialization and trade is determined by the extent of the market and if you have lots of people that are essentially within the same market, then you can have lots more specialization, lots more gains from trade. I'm not sure I fully understand the plumber example though, because that seems to me like something that is really restricted by physical space. What am I missing there?

Robin Hanson  09:50

Once they have a good enough avatar, they can do it from elsewhere.

Infovores  10:02

Would the avatar have physical character?

Robin Hanson  10:05

The avatar will be a physical robot, basically with arms and actuators and claws. And if it's good enough for the person from long distance to control it with the video and audio, then they can swap in and just do the next step or the physical job that needs to be done. So yeah, you would need a good enough haptics and interface, fast enough internet so that you could just naturally directly control it. But those are coming.

Infovores 10:29

How soon do you think that's coming? This is remote work paired with a lot of hardware advancements as well.

Robin Hanson  10:34

Right. But if there's a large enough industry demand for it, then I think this is well within the reach of current abilities. It's just a matter of doing it. So the AI is much harder job than you're talking here of sort of just physically good enough haptics.

Infovores  10:56

That's fascinating. I guess if we think about it in terms of the advances in self-driving cars for example, it seems like they've come a really long way and some of the biggest obstacles remaining are more regulatory than anything else. Probably something like this with robotic avatars is closer than I was thinking before.

Robin Hanson  11:25

Well obviously regulation could get in the way of this plumber. If local plumbers make really high regulatory requirements and what this avatar has to be capable of, then they could prevent it. But it seems to me enough places might allow it to let it take off. And then other places would be shamed into copying or be left behind being much less productive. With a self-driving car you can imagine, basically the problem is it can help but if you want to have the driver turn off their attention and not be paying attention to the driving, this thing needs to meet really high standards. But we don't have the person at home doing the plumbing and somebody looking over their shoulder replacing it. And so the nice thing about the remote work thing is that it does help automation. That is if you break any plumbing job into 30 different subtasks, you can automate each subtask, one by one, you could figure out which subtasks an automation could do. And then at that point, instead of swapping in a human to do that, you swap in the automation to do it. Because you've broken it down into lots of little things, as opposed to the self-driving car, if it's going to be that it has to do all the tasks, right? It’s not enough for it to do 80% or 95% of them. It has to be ready to do all the tasks. And that's the hard part about that problem.

Infovores  12:42

That makes sense. The plumber is not at high risk of causing a really bad outcome in the tail and killing someone just by repairing a pipe. And so getting it to the stage of good enough to start rolling out and then learning from user feedback and more kind of outside the training world of creating the thing, you could see it take off by capturing one local market at a time maybe and then eventually barriers to adoption in other places just kind of evaporate.

Robin Hanson  13:16

The plumber is just a stand-in for lots of industries where we don't have much specialization because we need somebody pretty close to the final place where it's done. So services that happen in the city actually aren't very specialized. And you're imagining factory-level specializations from all these little local tasks. That's the sort of thing I see a huge payoff from. And automation can swap in there. But I don't think automation is the main part of it, it’s just a nice add-on. 

Writing Futurism

Infovores  13:48

Makes sense. This idea of robotic plumbers and robots doing various other tasks gives a good segue into your 2016 book, which is unlike any other book that I've ever read. It's called “The Age of Em”, and you give this incredibly rich and detailed prediction of what work, love,  and life will be like when robots rule the earth.

We will get into some of the specific specific details of how that relates to current things going on in AI. But first, what inspired you to write a book like that so far out into the future 100 years or more and what are the main benefits of thinking through far off future scenarios that inevitably are going to differ from what we predict?

Robin Hanson  14:35 

The main goal of that book was to show people that it's possible to do much more detailed futurism than I'd seen done before. So a lot of people sort of say, you just can't do futurism, it's not possible. And so they say, a science fiction story is kind of the best you can do, you might as well be inspired and speculate on the basis of things like that. And I was trying to say, No, you can take a specific scenario and work out a lot of detail carefully if you follow my method, which is a pretty standard, simple method, but it takes somebody knowing a lot to do. And if you believe this claim, then it would be worth having hundreds of books like this, each taking a scenario that has maybe a one in 1000 chance of happening. And by having all the books, you'll cover  a lot of scenarios. So each book doesn't need to convince you of a high chance that that scenario will be true. A one in 1000 or a 1% chance is plenty. We can afford to have 1000 books, because we've got lots more than 1000 books in the library even about the future. And so the claim is we can actually study the future, it's quite possible to do so.

And the key thing to do is you define a scenario, let’s say a particular technology that might appear with certain characteristics. And then you just turn the crank and figure out how that plays out in the world. And so you assume the world is much like ours, except it has this new technology. And then you go through all the main fields that we understand applying your standard results in those fields to how would the world change with this technology. So my example of that is brain emulation, i.e. a computer model of a particular brain that would substitute for that brain and have similar behavior. And I assume such a thing could exist and be cheap. And I assume a few things about its features, mainly, that they are simple and not very complicated, and therefore it forces me to have a relatively simple scenario describing them that is, these simulations are black boxes, and you don't know how to take them apart or mix them and match them with others. So all you can do is turn them on, run them fast or slow, copy them, that’s it. And then I say, “okay well, what will the world be like as a result of that change?” And if you apply economics, electrical engineering, structural engineering, urban economics, sociology, you just take every field you can think of, and say, what does the basic 101 version of that field say about this scenario? And just pull those all together to make a consistent combination of them all and ta-da, that's a description of what such a world might look like.

And you might think, yeah but you couldn't do that. And I'm gonna say, but I did it! I did this trying to prove to you that it can be done, it was done, and I hope to inspire other people to do similar efforts.

Infovores  17:44

I can definitely attest that you achieve your aim of constructing something that is very internally consistent, that puts together all these pieces that we've learned over time, and still makes sense even though it's kind of projected onto an imaginary or simulated scenario. I also feel like writing a book like this—apart from it being such low hanging fruit, since you have millions of books about history in the past, and there's pretty severe diminishing returns to another book like that, in comparison to the very few books that look at the future with a similar level of almost historian-level detail—but there's also some interesting kind of advantages to this kind of style or art form that I was curious to ask you about.

So it seems like in a lot of situations, it's very difficult to get an audience interested in understanding basic facts about how human life is right now. If it's not new stuff, if it's not urgent like a brand new breakthrough, it doesn't catch people's attention in the way that kind of the gradual build up of human knowledge across a lot of different fields does, and so framing it as “I'm writing a book about the future”, but then including all these things that are, in my mind, super relevant to the world that we already live in that I hadn't really thought of before, it’s kind of a cool way to market understanding that's already relevant now.

A related point is that I think a lot of people are too enmeshed in the details of their social life and there are things that they take for granted intuitively in interacting with other people, that they're actually not very receptive to being told that they're a particular way, even if that could be very well supported. A lot of the theories that you have about how people behave make a lot of sense, but people don't like listening to them. But then in this future scenario, all of that resistance kind of evaporates because it's like, “he's not really talking about me and saying that I'm like this, I'm signaling or I'm status-conscious, or whatever. He's saying that these robots are like that. Did that figure into your thought writing this at all? Is this an opportunity to tell people to some extent how they already are?

Robin Hanson  20:31

I mean, I didn't plan it. It didn't influence my plans. But I certainly noticed and other people notice that this could be more of a tutorial about how the world works. I've certainly noticed that people are just too interested in changes relative to the constant. That is, a lot of people are eager to look at news about trends and how things might be different a little bit in the future. And they're less interested to know that you don't understand the present, you don't understand the basic world you're in now very well. Why are you so interested in how it might change in five years? You should be more interested in just knowing about this thing you're in that you don't understand. So I think people do tend to assume that because they can navigate the world effectively that they understand it. I mean it’s also a thing in history that good history analyses will focus on theories about how things work that you might realize apply to you too. And you can learn about your world from history as well.

Infovores  21:33

If you had to put a number on this, what percentage of Age of Em would you say pretty much directly translates to social life today?

Robin Hanson  21:45

I guess I'd have to go through and do a survey, but I would think at least 30%, maybe 30-70%? I guess I'd have to look more carefully. There's certainly some things that are different. I mean, I tried to find the things where there were differences and trace those out. But even the differences highlight things about today by contrast.

What “Ems” Are Like

Infovores  22:07

Of course, that makes sense. And so right in line with that, the robots in Age of Em retain some very human features. They're based on high fidelity brain scans from actual humans, and so they still socialize and have desires for things we care about in terms of romance, belonging, success, all of these things. It's very  familiar in that sense. But I'm wondering, in general terms, what criteria you looked for to determine whether certain human features get kept versus others that get abandoned. And based on that framework, whether there are any major human features you see as more or less likely to change in a possible AGI future, if that were to happen before, or instead of, the Em scenario that you envision.

Robin Hanson  23:07

So you're right that I need some sense of which features are how pliable or plastic in order to project the future. That is, some features are just so hard to change that I kind of assume they stay constant. Maybe their expression or application changes. And then other features are more ready to change if the world wants it.

So part of that is just looking at the history. Which are the features that have been most constant across our past, and which things have changed a lot across space and time? Whatever has changed a lot in the past is a candidate for something that could also change in the future. And then I'm also using our theories about which things are deep and hard to change versus on the surface and ready to change. If we could pick concrete examples, I guess we could talk about, I don't know, work hours or something. And clearly, work hours have changed over history. So ems could work more than we do today because our ancestors worked more than we did at least in certain time periods. But pretty much all work in the past people needed breaks, breaks every few hours, breaks at lunch, breaks in the evening, breaks on the weekend. That seems robust. So then I presume ems will need breaks. They may do different things with them perhaps, but one big clue is just what has been pretty consistent in the past.

Infovores  24:37

You say that ems will be good advice givers, to give a concrete example of a human... actually I'm not sure if that would be considered a good human feature or not… But I'm wondering, if the ems are good advice givers, will AI necessarily be a good advice giver? And what are your thoughts on AI to human mentorship possibilities? Even just in the immediate future with GPT.

Robin Hanson  25:06

Well, the whole idea of ems is that they are really close to being humans. So we know a lot about humans, and we can use all the things we know about humans to talk about ems. Ems are only a modest modification of a human.

AI is a name for a vast category of possible computing and machine systems. And you have to pause to realize just how vast is the space of possible systems. If you started out with a human and say, AI must be something like a human but not that different, you'll just be really wrong. The space of possible AI is really, really large. And that's one of the things you have to learn to understand about AI—AI is just a name for this vast space of possibilities. So that's one of the biggest problems and reasoning about AI is to try to categorize it somehow, to make some shape of the space of possibilities to draw some lines through it or playing through it so that you can distinguish, like what are the main types of them. So that's why if you ask me, are AIs good at advice, I go well, some are and some aren’t. It depends on what assumptions we are making about the kind of AIs.

So in that case, I might just look more to the human. If we're talking about AIs advising humans, the constant there is the human. So I might say, when do humans want advice? When are humans willing to listen to advice? That would be the analysis base I could go on. And now we might say,  when could AI fill the slot that humans want for advice? If we look at humans who take advice, we see that often, humans asking for advice or taking advice is just a way to suck up to people basically. So this anecdote from a book I once saw, struck me, which was, there's this high-level manager at some firm. And one of the things that happens to high-level managers, people are constantly asking for meetings with them coming in, asking them for advice. That's the thing high-level managers do all the time. And so this person announced that they were retiring in six months. And immediately, nobody wanted their advice anymore. They're the same person knowing the same things. But nobody wanted advice, because in fact, asking for advice is really kind of a way to ask for social support. They wanted his alliance, they wanted him to support them, you see, and all of a sudden he was no longer a valuable supporter, because he was going to retire and therefore they didn't want his advice. So that theory of advice says that AIs aren't going to be very useful as advice givers until they have power, so that you can pretend to solicit their support by asking for their advice. Because it's not really about the advice.

And then we also noticed that humans often try to give advice as a way to exercise dominance, which is often a reason why people resist advice. People are like, usually pushing advice on other people who don't want it. That's just a really common thing in our private interactions. And so it's not about me begging you to give you to give me advice, and you like reluctantly turning your head to me and maybe helping me. It's about people constantly pushing advice on each other. And people trying to be polite about it, trying to pretend they're listening, and then also trying to show their independence and that they aren't taking the advice from everybody who pushes it on them. So that scenario says that AIs will have to have some social power in order to push their advice. And then maybe they could push people to take it because of the same reason humans take other people's advice, but it's not about actually wanting the information. It's more about a social dominance display of, people are dominant if they can get other people to take their advice. So, for example, managers and politicians often resist taking advice in public even if they take it in private. That is, they'll often want advisors to advise them in private but if somebody gives public advice they go out of their way to not do that thing. Because that would make them look submissive to this other person who gave the advice. So now if you put all this together with AI we say, AI doesn't count for advice until it has some social power. And otherwise, people aren't going to want it.

Infovores  29:28

So it seems like one question is, to your point in the book, are ems good advice givers or are they just better advice receivers? Because in the situation that you describe in the book, a lot of what's going on is ems are receiving advice from other copies of themselves that have different life experiences, but are fundamentally linked in that way. And so they have a lot to gain from asking for advice, because it's very applicable to their specific idiosyncrasies.

Robin Hanson 30:14

But they’re not competing, having dominance contests with other copies of themselves, that's more the presumption here. Say you're a dentist and your twin brother's a dentist and you live in different counties right? Now you might be willing to call up your twin brother and get his private advice about being a dentist, especially if his practice is a lot like yours. If you saw him as your ally, and the one person in the world you most loved and were closest to, then you could be asking him for advice. And he could be giving you advice, if his world is actually a lot like yours, and he's actually a lot like you, that's the situation of the ems. They're giving advice to very closely similar people in very closely similar circumstances, they're giving it privately, and they're not in a dominance competition with each other.

Infovores 31:04 

And it's also much easier just at an informational level to interpret a request for advice, because I feel like if I sincerely really do want advice, and I want to apply it in order to improve, I know that, but the person I'm asking for advice is thinking, are they trying to ingratiate themselves with me? Are they trying to present themselves in a particular way? What's kind of the ulterior motive? Whereas it's much easier to to interpret and understand the intentions of an em copy.

Robin Hanson  31:38

Actually what I say in the book is that people will look at the experiences of people close to them, which is a little different than asking for advice. So imagine you have a marriage with your spouse, and there's a copy of you who is married to another copy of your spouse. Now, if they did something that got them in trouble with their spouse, that's information to you about what might get you in trouble with your spouse. They don't need to actually give you advice you see, but they're in just a very similar situation. So you'll just be very interested in what happens when they do things, because that'll be informative about what happens when you do things.

Infovores  32:16

That seems related to a paper you wrote with Tyler Cowen about… I remember one of the points in this article about why people disagree. And basically, one thing that stuck with me from that is that if you if you know that someone else has the same priors as you and that they also hold a particular belief, you should update to believe something similar to what they believe, even in the absence of any sort of explanation or seeing the evidence yourself.

Robin Hanson  32:53

Sure. At least, on average, you should move to their position, sometimes less, sometimes far. But on average, you should be moving to where they are.

Infovores  33:01

And it seems like among among a tribe of ems that are all copies of of each other, but different experiences, it's very easy to, without having to ask explicitly, you just see what they're doing and learn a lot of things about what they believe based on their context, and you can update very quickly.

Robin Hanson  33:23

So humans today have many kinds of social groups that they’re a part of. And then this set of all the copies of the same original, which I call a clan, is a new social unit in the em world. And so we're somewhat speculating about this new unit. So the client itself will try to make this unit bond as closely as you would bond with yourself at different times, or you with your twin brother or something. The question is, how well they'll succeed at that. And, the more that you will compete with other people in your clan, the more that you'll start to see them as rivals and maybe distrust them and maybe not be entirely honest with them. And if there's play in leadership positions, maybe that's unavoidable there will be some conflict, but I think the clan would go a long way to try to suppress that. So for example, if you can hear your clan members in your head as a voice, but you don't like see them as a body in front of you, that could help because we often trust ourselves as a voice in our own head, and a body in front of us we might more see as a rival. And they would try to avoid being in situations where they compete with each other for the same roles. And that would make them feel more like they were just part of the same person, basically. But it's an open question how far they'll succeed. That's something we don't know, because we haven't actually seen that. But there's a sense in which clans have just a lot more ability to be close than most of our other units, because they are actually very similar to each other. And they do actually share a lot of interests. And so it can be structured that their interests are really quite well liked. But is that good enough? I don't know. 

Infovores  35:05

When I discussed the notion of em copies with a friend, something that came to his mind is how in science fiction movies, sometimes you have plotlines were a group of clones kills another group of clones. They're the same genetic material, you would think that they should be grouping together as a clan as you talk about, but actually they turn very violent. And I wonder wonder why it is that you don’t put more weight on that possibility in the book?

Or to give a related example, why should there be any em vacations or em leisure in equilibrium? Because the  person who has the same capabilities that has a copy of you could they exploit your leisure time. Any leisure time that you have, they have an incentive to exploit and kind of cut you out in some way by taking your position. Or what am I missing there?

Robin Hanson  36:15

Well fiction is always going to be highlighting conflict in the most dramatic forms. That's just a generic formula for science fiction. So even if you look at the Black Mirror TV series that explores various future technologies, it is pretty much always trying to find the worst possible scenario with the worst possible conflict for any tech, right? It's not a very good guide to whether such tech would actually be good or not because it's basically always trying to find the worst case.

So ask yourself how much conflict you have internally to yourself across time. So we do have some conflicts, you might rather spend money now and leave it to your future self to make up the difference. Or take advantage of some fun now that you could have that your future self will suffer from. And we mostly limit that sort of interpersonal conflict by sort of a shared sense of identity— you see that future person as you and you see that if you hurt that future person, you're hurting yourself. And that doesn't seem so fun. And so that shows there this possibility of producing more coordination and cooperation if you see these other ems as yourself. And so we can predict that they will try to do that. And the question is just how far can they go? Obviously, they won't always succeed. But it seems clear that if they can succeed, then the ones that succeed will win. The clones that are killing each other and exploiting each other all the time, well that's just not a very winning formula for being a successful clan. They will die out.

So basically, the em world does select for personalities to get along with themselves. There are sometimes drama people where there can only be one drama person in the room, right? They're the drama person, and the drama always has to be about them. And their story has to be the center of attention. Those people don't get along well with themselves. If you have them in the room, they're fighting over who is the center of attention in the room, right? That doesn't work so well in the em world, as the world needs a lot of coordination with people like yourself. So it's more rewarding people who get along with themselves, people who can cooperate with themselves. But just like our world doesn't really that much reward people who can't save for the future, who can't delay gratification, the people who exploit the future self by taking all the advantage right now and spending every penny in their bank account. They don't survive in our world very well. You could just ask, why does anybody ever work because they could have fun now and let their future self work.

Infovores  39:06

Yeah, so it definitely seems like the em world is selecting much more strongly on being able to delay gratification, patience, and doing what makes sense in the long term versus the immediate short term benefit. And this is part of a broader argument in your book that we've been living in a sort of dreamlike state where there actually hasn't been very much selection pressure among humans for the last several centuries. At least in recent times, it seems like part of the reason for that is, as liberal societies develop, and as richer societies emerge, we seem to be more concerned about making sure that people don't get selected out, making sure that welfare is provided to people who maybe are not very productive or are even clearly causing problems, negative externalities. But we still feel like it's a basic sort of human right to ensure some level of preservation regardless of people's choices, except in some rare exceptions.

What makes you think that ems won't feel that sort of impulse? Is it just a function of the ems being paid at basically subsistence levels in your scenario or what is the main reason why selection pressure will return with a force?

Robin Hanson  40:51

So in our world today, we still have pretty strong selection pressure on businesses. Businesses you can create pretty quickly, they can grow quickly, they can die quickly. And we don't actually prevent them from dying by saying, “Oh, we're so sorry for you, we will subsidize you in order to prevent you from dying.” So there are big parts of our world where we allow fierce and severe competition. It's also true in say, democratic elections. We allow candidates to be voted out of office, and we don't feel very sorry for them and make sure they get an office to stay in indefinitely. We allow severe competition among candidates for office. And similarly, in music or movies, we just have pretty severe competition. For artists or music, we don't feel sorry for musicians, and make sure they all have enough people hearing them in order to be respected as a musician. If people don't like your music, we are allowed to have nobody listen to your music and even make fun of you. That's part of what we allow.

So it's only a limited range of things where we don't allow as much competition and in the em world, there's just naturally more competition for population, because they can just make population much easier and they can also make population go away. But it's just mechanically much easier to make ems, and mechanically much easier to end them or at least put them in a retirement stage. So for example, you might feel sorry for ems enough that they don't die. But if you can just make them retire at a really slow speed, they're not dying. And so now, maybe you don't feel so sorry about them retiring, right? So I can imagine people say, let's make sure nobody ever has to die, let's make sure the worst that ever happened to anybody is they retired to say 1/1000 of a human’s life speed. That's really cheap. And so now, okay fine, that's the worst it can get. But we still allow enormous competition for who gets to run faster. And then it's all about how fast you live. And then we'd allow that, right? If you had a sympathy, “no, everybody needs to run at the same speed, we would all like people running at different speeds.” Now, you're just gonna have a lot more regulation in society. And so a fundamental dynamic now and then is really just how many elements of society are we going to allow competition on? How much are we going to prevent competition? And that's in part sort of how much we are embedded in a larger competition.

So up until recently, the world was embedded in competition between nations and societies. And so if one nation was very kind to its members in a way that was not very productive, it would lose out over time to the other nations. And that was a discipline that limited how bad things could get in various dimensions. Today, we have more of an integrated world where, in fact, we don't have so much competition between nations and we more have a unified world community that if it creates a rule that is adopted everywhere in the world, then it could overturn competition and we have gone that way for some forms of regulation. But  I'm basically assuming in this analysis that doesn't happen in the em world. But there substantial competition is allowed and it has these effects. Now, partly that's because of my method of analysis. One of the main reasons I'm able to say so much about the world of the Age of Em is that I assume it's competitive. And competitive worlds just tell you what happens. In a non-competitive world, the world of rich people who are full of leisure and can do whatever they want, it's not enough to just know how rich they are, what technologies they have to know what happens, you have to know what they want. But in a competitive world, you don't actually have to know what people want to know what they do. Because they do what they have to do to win the competition.

So if you ask what kind of movies would Hollywood make? If you have a world where Hollywood has a government monopoly, where they are the only ones allowed to make movies and they can make whatever movies they want, then the movies Hollywood makes is whatever the executives want them to make, right? But in a world where the public decides which movies get sold, and if you keep making movies that don't sell you go out of business, then we can predict the movies that are made are the movies that the public wants. And we don't really need to know what the executives want. Their preferences are not very relevant. It's the public that decides what movies get made. And I'm using that method of analysis in the world of ems, I'm saying this is a competitive world, and therefore I can figure out what they do, and I don't need to know what they want.

Infovores  45:31

That makes sense. And even in our world today, which is much less competitive in comparison to the age of Em, there's only a relatively small number of dimensions where people are actually really concerned about alleviating inequalities. So a lot of focus is put on income inequality, but there are lots of inequalities that are arguably even larger that are pretty much never talked about.

Robin Hanson 46:06

So even income inequality we don't do that much about.

Frozen Brains and Em Gender Dynamics

Infovores 46:10

Right. So to mix things up a little bit, let's try playing a game I call “how to change my mind”. I'll throw out something relating to views that you have espoused and you can tell me the conditions that would make you open to reconsidering your position. How does that sound?

Robin Hanson  46:30

Ok, I may have to pause.

Infovores  46:32

Sure, no rush. So the first one: you plan to have your brain frozen after you die, so that it can be emulated into a future version of you. This would be once the ems arrived. What, if anything, would convince you to change that plan?

Robin Hanson  46:48

Well it's based on a cost benefit trade-off. Certainly, if the price of freezing myself went way up that might tip me over, or if the benefit went way down. So the two main risks are that the kind of information that's preserved in a current freezing of a brain is just not sufficient. If you could show me that, contrary to my current impressions, the actual information is coded on a very fine spatial scale with very fragile structures that are quickly destroyed in a freezing process, then I'd say, I guess it doesn't work, the structure is not there. The other risk is just that the organizations won't survive. So if you can convince me that they're going to face some hostile opposition that’s just gonna try to destroy them and probably will, then okay I guess the organization can't last. So then I wouldn't last. 

Infovores  47:37

People often use the phrase, “the future is female.” But Age of Em actually implies that the future is largely male.

Robin Hanson 47:45

Does it?

Infovores  47:50

In your section on business, I think you talk about people whose qualities tend to be most competitive today, who tend to be the most productive, and work the most hours. And on kind of right tail of the distribution, a lot of those people are male.

Robin Hanson  48:11

So I describe two contrary factors, and I'm not sure which one dominates. So one factor is certainly that men tend to have higher variance in productivity. So the most productive people in each area tend to be men. That's one factor. But the other factor is, as I say in the book, that women tend to do better on raw sort of grit and determination and persistence in difficult situations, where men will often lie down and die, or quit, or just be lazy, and slovenly and unmotivated. Women will just keep plowing through. And the question is, how much that matters.

We will select the highest performance, but you see men often have this highest performance because they get a lot of social benefits and celebration from it, right? Men often get more girlfriends and wives and praise for being the best. But the em world will select the best, and then they'll be ordinary in that world. So can they be motivated to be the best when they aren't distinguished and they aren't celebrated? They are just existing because they're the best. But that's it. That's all they get: just existence. If men are demotivated then if they say, what's the point of working so hard to be the best if I'm just average? Then if em women are more willing to just keep working hard even if they're not celebrated, then maybe women will have an edge there. That's the question.

Infovores  49:59

What do you think the ideal gender mix is to have in an em clan, if you had to guess in that case, because I'm wondering whether all women might band together better in some ways, because of..

Robin Hanson  50:15

Well, the idea of a clan is all copies of the same original person. So they would all be one gender, unless they do some sort of gender transformation within the clan.

Infovores  50:24

So do you think male clones will get along better or worse than female clones? Or will there be any mixing at all?

Robin Hanson  50:31

I would think that the more similar are clan members to each other, the better they get along. Because we're going to what we're going to try to do is make them feel like they're just each other in their head. It's like I'm just hearing myself talking to myself, basically. If that's what the image we're going for, then they should be as similar in as many ways as possible. So if you have yourself talking in your head, except it's the other gender, that might be a little disorienting, sort of break the illusion that it's you talking to yourself.

But I do think probably most people want to do a male female pair bond. That's just a pretty common human preference. So that would be my assumption is most people just for deep human emotional reasons want to be part of a male-female pair bond. And so they will do what it takes to make that happen. And we will pay substantial costs in order to make that happen, so that's probably how it happens.

I tried to discuss how you can have that be true, even with an unequal demand for male and female labor, by basically having different numbers of copies of each side of the relation or different speeds so that they can have the same sort of equal relationship with each other, even if they have an unequal relationship with the labor market. But, obviously, the simplest way is just a nearly equal domestic labor demand. But I think you can accommodate maybe a factor of two or four in the ratio difference through some other means. 

AI Sentience, Em Religion, and Human Rights

Infovores  52:04

Gotcha. What do you think are the odds that we look back on Blake Lemoine and in some way, kind of see him as being one of the first people who really understood something about AI?

Robin Hanson 52:20

I’m sorry, Blake who?

Infovores 52:25

Blake Lemoine was a Google developer who basically was telling everyone that AI was sentient. He got fired about a year or so ago, before the excitement that we're seeing about AI today. Do you think there's any chance that people look more favorably on him in the future, rather than just as a crazy person?

Robin Hanson 52:46

So far, people have just made this sharp distinction between humans and machines and animals. And when they make that sharp distinction, they're willing to just be much more harsh in their treatment of non-humans than humans, right? And so simply focus on animals, right? The more you start to see animals as continuous with humans, say chimpanzees, then the less willing you are to tolerate harsh treatment of chimpanzees. And the more that you're going to insist that people be kind in various ways to the chimpanzees. Presumably, we're eventually going to do that for machines, too. And so the question is, how will we draw the lines?

That is, well, we just have some graded treatment where you can slowly treat them more harshly as they get different from humans, or we draw some sharp lines is a past this line, you can ignore it because that's nothing. And that's just going to be hard. Especially with AI because again, there's this much larger space of possible AI than there is for animals even right the space of animals and compared to humans is relatively limited in their the space of possible AI. So, the nature of morality and the sacred is such that you tend to want to have relatively simple clean rules that everybody can observe and enforce. And the question is, well, what can those be?

So, there's two kinds of sets of rules, one set of rules is how not to be mean and cruel. And maybe another set of rules is how much support and help. So even in the past, if you don't have any sort of welfare or obligation to save somebody else who's like dying in the desert or something, you can still have norms about not mistreating them when you come across them, so it will be easier to just set up norms of avoiding mistreatment than it will be the norms of help. So a basic problem with ems and AI is it's just so easy to create them that if you have an obligation to help them once they're created, this is just unsupportable. It's just too big a demand. It's just easy to create things that if there was an obligation to help them then they would just quickly suck up all available resources to help them because they're so easy to create. So either you prevent their creation, or you don't have such a strong obligation to help.

But you can still enforce mistreatment rules, where if you create it, you don't do the following things. That's more feasible as a set of rules in a world where ems are easy to create. And now the question is what counts as mistreatment? And one standard we have is some sort of voluntary standard, like if they object, then you're mistreating them. But if they say it's okay, maybe you're not depending on what the rest of us think. And you could certainly imagine in the age of em, one simple rule you could imagine that applies to a very wide set of cases, is just a right to suicide. You say look, pretty much any em and maybe any AI should just be given the right to suicide. If you understand the concept of suicide, and you decide that's what you want right now, okay, it's yours. And that can also be a way to define torture, right? What's torture? Well, if they'd rather commit suicide than continue existing, you must be treating them pretty bad.

We're searching for simple rules that we can all agree on to apply in this really vast space of possible creatures, that's the hard problem here. And we haven't really explored that space for which we don't have much of a sense of how big it is, and what all the features in it are, and what they could be like. I mean, obviously, some of them could even like torture, some of them could enjoy suicide. How do you apply rules to them?

Infovores 57:01

For a very long time, in Western societies, the Bible has been a really foundational, the foundational book I would say, and that affects so much of how people think about things in ways both conscious and unconscious. Will the ems still have any connection to the Bible? Or will they perhaps develop an alternative bible of some sort? Will there be like a Good Samaritan story, except the good guy is the one who walks by the em on the road to Jericho without mistreating them? What do you think that looks like?

Robin Hanson  57:46

So the two big facts to notice, or maybe three, like one, clearly, religion was very functional in the farming era. I mean, the modern religion showed up halfway through the farming era, and before that, they had some pretty different religions. But, for the last few thousand years, we've had the modern religions, and they have been pretty functional. And the industrial revolution happened, and they survived it all. The major religions are still the major religions in the world, even after this enormous transformation in the industrial revolution. So if the Em revolution isn't any bigger of a transformation than the Industrial Revolution, you’ve got to predict whether they can survive that too. Now, what we have seen is a great decline in the interest of religion in general, over the last few centuries, but we still have the same religions being the major religions. I mean, if you're going to do religion, you're going to do it the old way, in our world. And so I would predict that if the Ems want to do religion, they'll do it the same old religions and those religions have already shown that great ability to adapt to changing circumstances because that's what they did over the last few hundred years, so they'll probably be fine at adapting to these new changes. Then the question is really just how much demand for religion would Ems have? And for that, you need to try to understand, well, why have we had a declining interest in religion in the last few centuries? Because we have pretty strong interests before. And for that, I have this sort of forager to farmer to foragers story as my explanation of the trends in the last few centuries. And that theory predicts that we would want to return to religion with animals, because they would be much poorer and need a lot more self control and dealing with hard circumstances. And religion would be there for them for that. And so my prediction is okay, they go back to emphasizing religion more, and they use the traditional existing religions, because those seem pretty robust. And that's the simple prediction. Now, you can already predict, you'll need a few modifications to traditional religious dogma in order to accommodate the Ems, but they're pretty obvious and easy. So of course, yeah, they just switch, Ems, they say, ems are children of God, ems have souls, ems can go to heaven. Now the question is, if you make a five minute copy of an em does it go to heaven?

Infovores  1:00:13

Right, so these are like the theological debates in the age of ems.

Robin Hanson  1:00:15

Or if you make a copy and it sins, is that your sin? Or is that the copy’s sin? Because the law will actually have to make a decision there that if you make a short term copy, and it sins, it breaks the law, that's your fault. That is you will be held responsible for the legal violations of short-term copies. And I think if the law does that, the religion will probably go along and call that a sin too. So if you make a copy that commits a sin, that's your sin. That's my guess, because it fits with the law.

But those are the questions that we'll have to answer in order to adapt religion to it, but again religion has adapted so well to the modern world. It's really pretty flexible, so I expect it will continue to be flexible. Logical consistency has never been the main reason people got into religion anyway in the first place. And theologians have long been able to tell twisted stories to try to make things that appear inconsistent be consistent, so there's gonna be plenty more of them. So they'll just do that.

Infovores  1:01:14

The thing I was hung up on there for a second is, without physical bodies, in principle, ems could go on indefinitely. And so if ems are already immortal, does religion like… is that really such an easy hole to plug? Or how do you…

Robin Hanson  1:01:35

That's a framing question right. So you can imagine that every five minutes, a copy of you goes off to a heaven even if you continue. Or you can imagine only some end point of you that never continues would go to heaven. Those are two different religious choices that theology can have right? Or something in between.

That's a key question. So I don't know. I guess the conservative thing is to say, only a branch that ends goes to heaven. But maybe does a five minute branch that ends go to heaven? Does it go to a slow heaven? You get into fast heaven if you live longer? Those are the tough choices there, right?

Infovores  1:02:17

You could imagine roping in maybe a theology of suicide since suicide is not frowned upon in the same way in em society, maybe ems commit suicide occasionally on faith that they will be transcended into something beyond, into heaven. 

Robin Hanson  1:02:37

Right, but if they commit suicide with one copy that was branched off and then the mainline goes on, does that count as suicide?

Infovores  1:02:43

That debate still remains as well. Maybe you're not showing enough faith if you if it's only a short-term copy.

Robin Hanson  1:02:55

So maybe you need to have sacrificed and suffered for a long enough time to qualify for heaven or hell or whatever, I don't know. But the point is, religion will just search in the space of possible answers and they will compete with each other and some will win the competition because they more motivate people and bond people together and we can understand the underlying function of religion as staying constant while the details vary.

Infovores  1:03:27

Fascinating. So I guess I'd like to close with just a couple of miscellaneous questions. So just very practically, how do you use GPT? Like, what have you found to be the most valuable applications of it for what you do and what insights would you have to share with others?

Robin Hanson  1:04:00

So at the higher meta level, my attitude is that we've seen rapid improvement in abilities over the last few years and that we’re plausibly going to hit a knee of the curve in the near future, where they won’t improve as fast–they’ll run out of data, run out of hardware. And so we're seeing rapid improvements, but they'll reach some plateau sometime soon, and then they’ll be stuck near that plateau for a while. If that's true, then the time to really pay close attention to their abilities is when they hit that knee and go into the plateau. If they're rapidly changing in abilities, then looking at the limitations of any current version isn't very informative about the future, because they're going to change. You can look at what they can do, but looking at what they can't do won't tell you that much.

Then the thing to ask today is what can they do in a robust way? So what I did is I did some tests to see like, what kinds of reasoning can they do or not? In closer to my area, what kind of exam questions they can answer for my students, because maybe my students will try to use it to answer my exams. And I saw that it was impressive in terms of what it can do compared to previous systems, but it's just not good enough to do a lot of the ordinary tasks I gave it. And I can see how to sort of assign my exam questions so as to avoid its ability so that I'm safe from people using it to take exams. I can't see that many other ways I could personally use it. But maybe in the future, it will be better. But the point is to get a rough guess now and wait and do the extra effort when it reaches the plateau.

The other thing I should be doing is just listening and saying, what do I hear about who can use it to do what? And the main thing I hear is that programmers use Copilot to program and I know some experienced programmers who are saying they're getting like a 20% boost in productivity from using Copilot and I go, okay I believe them. And that's plausibly the biggest valuable application for it. If I try to think about what applications that could work for, that just makes sense as an application that would be high-value. So at the moment, that's my best guess about what’s going to be the highest value application. And if now it’s 20%, maybe it'll go up to 40%, probably it won't get that much higher. That's roughly the boost that you're gonna see out of it. That's the main group of users and how they'll use it. And that's my tentative conclusion. But again, I'm going to wait till I see them hit the knee of the curve. And I'll try all my suite of questions again, I'll see okay, have you figured out how to do the following questions and what can you do? And maybe it'll be better enough then that I'll think more about what more can it do but at the moment is just not good enough to do the range of things I’ve tried to have it do. But apparently, it's good enough for doing code so great for the coders. I'm not a coder.

[laughter]

Infovores  1:07:10

Gotcha. Yeah, I've definitely been dabbling with copilot. Unfortunately, it doesn't do Stata, which is the main thing that I do.

Robin Hanson  1:07:22

Not yet, but maybe in a few months they will. 

Infovores  1:07:25

And I think it'll make it a lot easier for me to use my knowledge of Python, even though it's not my strongest language, because I know enough to take advice from AI. Yeah, that makes sense.

So last question… I mentioned at the beginning of the interview that you recently moved Overcoming Bias to Substack. What all went into that decision, and what do you what to you makes Substack a superior option over blogging as it existed before?

Robin Hanson  1:08:08

I think the main thing that happens when people move support systems or computing environments, is just that support declines for an old thing and rises for a new thing. That is, there's just a lot of value in using a supported system, a system that other people are using where there's people who are building or just fixing the old ones when they break. The more that you're stuck with old computing systems, the more  your hardware options are limited and your feature options are limited. And then when things break, it's hard to fix them. That's just the nature of using old things.

So there's a sense in which, for many kinds of computing applications, many people would just be better off if they just kept the old systems and didn't keep touching the new systems. Because there's all these people who just have to keep switching every time old systems go away, and new things come and they aren't actually getting better service from the new systems, they’re just getting the basics of having a system that’s supported and works. So that's the main thing I'm getting from switching to Substack is like, the thing I've been doing before hasn't been so supported, because it was old. And lots of people are using sub specs, if I use the same thing lots of other people are using, I'll get support, right? I can talk to other people about what they're doing, and they can give you recommendations and somebody's fixing the bugs and upgrading the hardware. That's just the sort of thing you get with a supportive system. I know many people who have very happily used old systems, and then they gradually have to switch, not because the new system has any better features or things just because it's hard to keep using old unsupported systems.

Infovores  1:09:59

Bryan Caplan has pretty much been evangelizing Substack since he moved from his old setup to there. And he just says that so many more people are reading his posts on a regular basis than before.

Robin Hanson  1:10:16

That's part of the infrastructure, right? They have an infrastructure to help people find it and read things in the same environment. And that’s of course, what they're designing. But that's part of being in a supported system. People find you more easily there because people go to supported systems to look for things.

Infovores  1:10:33

If you had to predict a year from now, what multiplier would you would you expect to see in your existing readership?

Robin Hanson  1:10:50

Readership is another part of the losing support because I lost my stats on how many readers I had. That's part of the features I was no longer getting with the old system is even knowing how many readers. So at least now I'll know how many readers I have.

Infovores  1:11:06

Well, I'm sure it will be a huge success and I look forward to continuing to read your posts. Thank you so much for sharing some of your insights today, and I encourage everybody to check out “The Age of Em”.

Robin Hanson  1:11:23

Well, thank you. Thanks for discussing it. So last thing I’ll say, I just went to University of Chicago where I did three events. One was a Night Owls thing where I was interviewed on the topic of the sacred, that had like 175 people there.

Then I did a talk and discussion about "Elephant in the Brain" with about 25 people. And then I did another one with “Age of Em” that had about five people. So not so many people were drawn to the topic of Age of Em there compared to the other topics. But because I gave that talk two weeks ago, I reviewed Age of Em. And so I was all prepared to talk to you about it, because I had gone through it in my head. So thank you for revising a topic here that not so many people wanted to come to a talk on.

Infovores  1:12:10

Well, I think it's definitely underrated in that case. That's… I think I've only scratched the surface of fully internalizing some of the some of the things in that book. And I suspect it'll continue to be very relevant.

Robin Hanson  1:12:24 

One thing you could say is that it gives you a more validated mark of how weird the future might possibly be. That is, usually the world doesn't seem so strange and we get used to it. And science fiction tries to make these really strange futures in order to make it dramatic. And you might think, yeah but how strange would the future be, really? And so this is a certain degree of strangeness, you have to admit, right? And it's plausibly about roughly how strange you should expect the future to be, even if my scenario doesn't apply.

Infovores  1:12:55

Absolutely, and a lot of insights into things that are going on all around us right now as well. I think the present is maybe more strange than people realize, and maybe when they imagine the future it's not grounded enough in the things that are likely to persist.

Robin Hanson  1:13:13

And they don't know all the things we don't know about our world today.

Infovores  1:13:17

Absolutely.

Robin Hanson  1:13:19

All right. Well, nice talking.

Infovores  1:13:21

Likewise, thank you for your time.

Thanks for reading! If you enjoyed this, check out my other writing at Infovores Newsletter. You can also follow @ageofinfovores to keep up with me.

And make sure to subscribe to Robin’s Overcoming Bias, now on Substack!

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Transcript edited for clarity, Spotify link here.

Infovores 0:04

Hello and welcome to Time Well Spent—a place where the most brilliant minds in the world take on the toughest questions in Science, Politics, Technology, and much more.

My guest today is visionary economist and author Dan Klein of George Mason University. Dan has written cogently on many important topics, including groupthink in academia and the tendency for innovation to outrun government regulation. He also has a YouTube channel, which offers wisdom and insights from intellectual history that listeners may appreciate. Most recently, Dan has released the book “Smithian Morals”, which will be the focus of our discussion today. Dan, thank you for joining me!

Dan Klein 0:43

My pleasure. Thanks for having me.

Infovores 0:46

Let's start with a classic quote from Wealth of Nations—

It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest. We address ourselves, not to their humanity, but to their self love.

Without additional context, it sounds like what really matters to Smith are financial incentives, and that cultivating personal virtue is not all that important. How would you reconcile that impression with your reading of Smith?

Dan Klein 1:12

Well, he's talking about a particular context there. When you go into 7-Eleven and buy some bread, there's not that lively feeling of benevolence between the clerk and the customer. They might say hello or something, but that's more just common decency.

And so he's looking at particular sorts of interactions there, which are impersonal commercial interactions, and talking about how that realm works. But that is by no means all that he treated. His first book spoke very little directly about commerce, and talked much more about a person in society in a community context, in a social context, where all of the virtues are explored, not just the sort of prudence that you might glean from the quote you just read.

Infovores 2:10

You made reference to his first book, The Theory of Moral Sentiments. That book serves almost as a foundation for some of his thoughts on economics, how would you relate the two books, or where do they connect?

Dan Klein 2:23

Yes, it's broader. I sometimes speak of it as the larger umbrella under which the Wealth of Nations sits. So the Wealth of Nations is almost like an extension, I would say of TMS. The Wealth of Nations treats just sentiments in matters of commerce, trade, pursuing honest income, and of government policy, with respect to such things, and indeed government policy in general.

So he's exploring there the proper beliefs of those activities, and ideas and policies about those activities. So those are other objects for moral choice. And so he's teaching us how to be moral with respect to that part of our life and world, which is an important part of our culture. Because commercial life and earning a living is a big part of your life. And because we care about the whole of society, it's one of the things that draws us together in common interests. And we need to think about those big issues properly. So he's teaching us propriety, kind of a moral concept, in thinking of those big questions and issues.

Infovores 3:57

Thank you, that's an excellent introduction. Individual liberty is of central importance as one of the virtues in Smith's moral philosophy. Yet he also acknowledges exceptions to this principle when he writes, “those exertions of the natural liberty of a few individuals, which might endanger the security of the whole society, are and ought to be restrained by the laws of all governments.” What lessons should libertarians in particular take from this insight?

Dan Klein 4:24

That the liberty principle is not an axiom. That it's defeasible, which means to say that the burden of proof can be overcome. There's a presumption, I would say in Smith toward individual liberty, just as in a courtroom we have a presumption of innocence. But that doesn't mean that everybody is innocent. It means that the burden of proof is on the prosecution. And that's in a sense what Smith is doing. Someone who's prosecuting a case for government intervention has the burden of proof to bear and Smith does make exceptions like he does there in that passage. That's about the small denomination notes restrictions that he endorsed, kind of oddly, I'd say. Sometimes his exceptions puzzle me to put it mildly. But he does make exceptions, and I endorse some of his exceptions. And I would make exceptions.

And so you get in Smith, very clearly I would say, the idea that exceptions don't destroy principlehood, if you will, like a principle is not undone because there are exceptions to it. We need to develop principles that admit of exceptions. Because if you try to do things that are 100%, without exceptions, you find you can't find very useful ones. Or you force fit certain things which aren't really sound, and then they're brittle. Because if you're claiming they're 100%, and then someone throws a sound exception at you it shatters, right?

So really your principles are more robust by allowing and acknowledging exceptions, they are presumptions. They are maxims, not axioms.

Infovores 6:22

This makes a lot of sense. And I think this is something I find really resonant in your philosophy of economics compared to either the rigid, mathematized version that you see in a lot of economic journals. And also, as you pointed out in some of your writings, Austrian economists can be very axiomatic as well. So it all makes a lot of sense, and it resonates, but sometimes in practice it's not as clear how to actually pull this off.

So I think COVID-19 is a really good example of a situation where there is this really extraordinary circumstance, and there are a lot of potential justifications for making amendments to individual liberty, at least for some amount of time. But as we saw over the course of the pandemic, there were a lot of abuses of those exceptions. How would you speak to that concern in the context of this specific example?

Dan Klein 7:23

I became very concerned about those very issues right at the onset of the pandemic, with all the news and panic about the pandemic. And I was pretty panicked frankly, I was kind of the worrywart in our little household. My daughter was going out to cheerleading practice—this was in Sweden, which was not closed down—and I was saying, “no I don't want you to go.” I was in that state for some weeks, and then I changed my thinking quite dramatically shortly thereafter.

I wrote some things during that time, I wrote a piece asking how much are we ready to compromise? One thing I said is that there are no absolute libertarians in foxholes. So I do think that the precautionary principle has some merit, that there is something sound to it, when you're puzzled and you're worried about really catastrophic possibilities, as we very much were in March of 2020. You maybe should take precautions, and what was it called The Diamond Princess, the name of that ship? Maybe if I were the mayor I wouldn't let those people off that ship. I'd be ready to make some exceptions to the liberty principle.

But I do think that the COVID case turned out to be overblown actually, and that a lot of the restrictions on individual liberty turned out to be completely unsound, counterproductive, and highly destructive. So it's an interesting lesson. In some ways, it teaches us the importance of serious presumptions, for liberty, because panics do come around, opportunism and oppression comes around. And we need these principles partly as checks on expansive intrusive government and the ratchet effect, which has clearly taken place in a dramatic way. Both sides have their moments in the unfolding of these affairs.

Infovores 10:00

So if I were to push you just a little bit further on this, we've learned all these things from the experience of this pandemic. But if we were thrown into an equally uncertain situation tomorrow, and it was unclear just how contagious things were and just how bad things could get at their worst, what would your approach be? Keeping in mind to not limit liberty more than what would be just or appropriate.

Dan Klein 10:30

Well, one thing that's important in these moment by moment deliberations and actions is how much trust you have in the powers that be, in the information and narrative providers. That trust has been shot to hell in the last three years.

So right now, if anything comes along I don't believe a word they say essentially. I need to see kind of like an asteroid darkening our sky at this point for me to start screaming about how we need to clamp down on liberty dramatically.

On Technological Change

Infovores 11:09

Moving on to the next question. In Theory of Moral Sentiments, Smith suggests that man “can subsist only in society” and “was fitted by nature to that situation for which he was made”. What implications does this idea have in the context of rapid and at times alienating technological change?

Dan Klein 11:32

I take your question to be that given that man's meaning is very socially immersed, or socially embedded if you like, if the social conditions are very rapidly changing, doesn't that jeopardize or upset man's tranquility and happiness. Smith saw tranquility as a very important element of happiness. Is that kind of what you're suggesting?

Infovores 11:59

Yeah. Let me give a little bit more texture from my own experience. So I have very unusual interests perhaps, but I see this happening with all of my peers to some extent, where the local circumstances of my neighborhood and community are becoming less and less important, and it's easier to silo into very narrow groups. I don't feel as attached or connected to the people who are physically around me the way that I used to. And I wonder sometimes whether such a rapid change to the way that we experience social life is detrimental to us. What in your view would Smith make of that?

Dan Klein 12:53

He may well feel that it's a bad development. And it may be a bad development in at least some respects, if not bad overall. These are very, very serious and real issues. Technology is kind of oversold by a lot of people, and this effect that it has on instincts about propinquity, about geography if you like, and the kind of duties that would be attached to some of those considerations as well as other traditional things like family and past history together even outside of the family, your old neighborhood friends and so on.

It's a real issue. And of course, technology also facilitates talking to your family members regularly and meeting and seeing their faces. My wife and daughter are Swedish and live in Sweden. I talk to my wife every day on FaceTime, and that's a kind of communion at a distance. So it's a very real issue, and there's no easy answer.

I don't think it does much to undermine the support for classical liberalism, because I don't think the government has anything good to do about it. I don't think restrictions on liberty are going to help; I think the toothpaste is out of the tube, and I think Smith saw things that way. The toothpaste is coming out of the tube, and there's no way to kind of force it back in, or at least not in any kind of wholesome manner. And so it's more about going with the flow and managing all these crazy developments.

My own view of of humankind is that our genes and our basic instincts, and our mentalities, penchants, and bents come from the hunter gatherer band, which is wildly different than the modern world. And we are constantly having to adjust, and temper, and channel, and subdue all sorts of things to cope in the modern world. And I think Smith actually had similar notions. He didn't make it altogether explicit, but so much of what he says dovetails with that view of man, that I'm prepared to kind of suggest that.

Infovores 14:45

I think you made this point in one of your YouTube lectures, that Smith saw all of the ways in which economic and technological advancement could upset society and create cultural churn that was uncomfortable in certain ways, but he still firmly believed that this was a benefit on net to human prosperity. It would be interesting to hear his take on that today, especially being able to talk maybe more openly.

Dan Klein 16:30

Yeah, and there's two "on nets" here. There's on net in the sense of new technological capabilities arising and how historically these have often been net beneficial developments. And then there's on net in the sense of, could we do something to roll them back through government? Would governmentalization of social affairs somehow improve on net how things will otherwise go, given that these developments have arrived?

As a liberal, I'm more interested in the latter. I'm more concerned with getting it right there, because we're not going to reverse history. And I'm not wedded to saying that these developments are so wonderful on net. I'm very worried and upset. I've written against designer babies, for example. I got into a dispute with Bryan Caplan and David Henderson over that, where I said, you know, designer babies is not necessarily a blessing at all.

Infovores 17:26

And why do you say that?

Dan Klein 17:29

It's these kinds of issues where it could throw off so much sort of contextualization of life, meaning of life, bearings, groundings, bases of comparison, connections to the past, connections to history, learning from previous human experience, what to aspire for, what to feel good and bad about. You can think of it in terms of sports and steroids—designer babies is like steroids on steroids!

Infovores 18:21

It seems like you could potentially go far enough in the direction of programming new human beings like machines to the point where you devalue the autonomy of that new individual, or you devalue the rights that such a person has, either relative to their parents or to the government. I'm sure government would be heavily involved in trying to regulate some of these things.

Dan Klein 18:48

Yes, I follow you. I don't think about designer babies to the point that somehow you're kind of undoing the basic nature of a soul, if you will. So I think that even if I designed my baby, and it came into life, it would still sort of have its own will, its own soul and everything, but you have kind of and presumably will know that you have chosen certain things and edited in certain directions in some sense. And so all of that might feel like a big incursion on autonomy.

Infovores 19:31

And Smith, as well as Hayek, offer a warning about overestimating what you're able to design and how much wisdom you have in doing that.

Dan Klein 19:40

Yes, yes indeed.

Promoting Freedom Around the World

Infovores 19:47

As the editor of Econ Journal Watch, you have published articles investigating liberty in a wide variety of countries and cultures. What are the major lessons you've taken from that experience for promoting liberty in different contexts?

Dan Klein 20:03

Okay, good question. So you're referring to our series, which has covered about 23 going on 24 countries—classical liberalism in economics by country. It's really more just classical liberalism by country; we ask for some emphasis on econ, but that's not really that central. And so we've had authors from all these very many countries tell us about the history, progress, current status, current organizations, figures, activities of the classical liberal movements in those countries. And there's been a number of lessons.

One thing I learned was that it's harder to get papers about countries with huge histories like the United States or England, because it's too much to write about. So we don't have some of the prime countries you might hope to read about. Most of the countries are a little bit smaller, or just have less classical liberalism to write about. But anyhow, that's one thing.

One thing I learned also is the significance of something analogous to the brain drain, and that's the "liberal drain". That’s what I call it. The liberals from India, from Venezuela, from Eastern Europe, what have you, they get out. They try to get out, just as anyone would. And that's a very significant factor in a lot of the stories because a lot of the people then make their careers in Britain or in the United States, let's say they're from India, and in a sense you've kind of deprived the fight at home of somebody that could help, in a way analogous to the brain drain. So I think the liberal drain is actually something I learned from doing this project.

Now, I can answer your question a little bit more broadly, and a little more deeply going into political theory. Specifically it's about classical liberalism, and liberalism is generally this presumption of liberty allowing every man to pursue his own interest in his own way, as Adam Smith put it, in policymaking. And that's all well and good. And you can wonderfully cite and learn from Bastiat, about mutual gains from trade, and explain that there's mutual gains, so let people trade, and that's kind of like the major thing to say—freedom.

Now, in a lot of these countries, and even in every country, politics is much more complicated than that. And politics is directional from a status quo. And the Liberals want to move in a more liberal, less governmentalized direction. Politics is also about the lesser evil. It's about trying to keep the government from getting terribly bad. And so I think it's part of civic virtue for them to consider the reality of politics on the ground in their country. And so being part of politics and making all the compromises, getting the votes, and everything else gets very much mixed in with all of these liberal movements, including the intellectual efforts.

So that's something I've learned in a significant way. I mean, it's not really new to me, but you see it so deeply. And when some countries, as opposed to America which we might have as a reference point, some of their politics is so bad and frightful that this becomes a rather dominating matter. Now, even when we talk about the lesser evil in terms of partisan politics within the context of a country, all that presupposes a certain level of political stability, of basic polity.

And sometimes the countries don't even have that presupposition in place. It's about being a principality of Russia, for example. In the case of Finland, it's about foreign wars, it's about civil war, it's about all sorts of cultural tensions, and old traditions, and so on and so forth. And so the Liberals sometimes get lost. Bastiat's basic advice about letting people make their voluntary trades... it's like you don't even know who you're saying that to. You don't even know who it would be who is letting people. And I don't mean just this party or that party, I mean the whole structure of government. Like in Venezuela, I mean, it's chaos now. We have a piece on Venezuela, and it's mainly a lament. So that's something else I learned, just these deeper kind of polity issues that we tend to presuppose in this country. I think we're coming to presuppose them less as things go more crazy all the time, but those are some of the things I've learned from doing that.

Infovores 25:49

Part of what you said struck me as interesting in context of recent things that have been written and said by libertarians about politicians.

A lot of times libertarians have the tendency to kind of devalue the entire process of politics itself, and they kind of don't... it feels too messy to them, or morally compromised. Bryan Caplan has a book along the lines of all politicians are evil. And it sounds like part of what you're saying is that by getting a closer look at some of these countries and seeing just how messy things are on the ground, that there's a real need for some of these institutional details or for some of these activists and political actors that maybe do things in ways that are not along the lines of what you might read in Bastiat or in very clean libertarian philosophy.

Dan Klein 26:49

Yeah, I think that's right. When Smith referred to “that insidious and crafty animal” commonly known as the statesman or politician, I think he meant that even of the ones he liked, like his friend Edmund Burke. That as a politician or statesman, he is an insidious and crafty animal. And it's an insidious and crafty trade or line of work, or you could say calling even.

I hope there's more virtue rather than less virtue in politics. What are we supposed to say about the people who have done so much good, like a Margaret Thatcher? I think it's... I mean, just to throw her into the same bucket as all the regular politicians whom we justly dislike is I think a terrible, terrible injustice and ingratitude, actually.

Politics is part of civic virtue. In this famous lecture called “The Liberty of the Ancients Compared with that of Moderns”, Benjamin Constant says modern liberty is mainly about individual liberty, allowing every man to pursue his own interest his own way. He said ancient liberty, which had more to do with participation in political life, is still a major duty. And it's almost a danger that people are too focused on their modern liberty so that they neglect checking and improving the ancient liberty, which still calls to everyone I would say.

Infovores 28:52

It sounds like you're pushing back a little bit against the common economist argument that people really shouldn't bother to vote, because from an individual standpoint one vote is not going to matter. And so you're better off not really investing in politics, even to that minimal degree, unless you can really find an avenue that has clear obvious success. Would you push back against that and say that it's important to vote anyway, or what are your thoughts there?

Dan Klein 29:22

I would push back against it if somebody was really making an argument against voting along the lines you just suggested. I'm not sure that I want to urge people to vote or tell people that it's terribly important and a major part of their civic duty to vote. I used to not vote, and now I do find myself voting in the regular elections. I think the Republicans are the lesser evil. We have a two party system. And I believe that third parties are damaging to their own cause in a two party system, so I vote Republican. But I'm not urging people to do it.

To kind of take your question and change it a little bit, something that goes with the argument against voting is sort of this "pox on both their houses" attitude that you get from a lot of libertarians. And part of the rationale for that attitude is that there's not a dimes worth of difference. And I don't believe that, I do think there's a dimes worth of difference. And I think that difference has been growing of late. So like I said, politics is about the lesser evil. And so this is a part of our responsibilities I would say. And the size of the part varies person by person, I'm not saying it ought to be like a big thing, or you need to vote or you need to even think about these things necessarily, because…

Infovores 31:07

You always have competing responsibilities.

Dan Klein 31:09

You have competing responsibilities, and some people in some sense have comparative advantages to focus on other responsibilities and so on and so forth. But if the lesser evil isn't supported, we get the greater evil. I don't think this idea is really news to people, but some of the argumentation tends to kind of elide this basic intuition.

An Overrated Game

Infovores 31:49

Gotcha. Let's try playing a game of overrated/underrated, starting with various strains of libertarianism that have been popular during my lifetime. I'll just do these kind of rapid fire and then you can give your short form case for conservative liberalism, which is, in your view, the best of them all.

Dan Klein 32:14

Can you tell me overrated/underrated by whom?

Infovores 32:18

Let's think of it from the perspective of people who could conceivably be attracted to some form of libertarianism. And so I was gonna start with the Chicago School of Economics, and you could say whether that's overrated or underrated among libertarians or libertarian adjacent people.

Dan Klein 32:45

How about this: Let's just go through the topics and I'll tell you something, I think about it.

Infovores 32:50

Sure, that works. What do you think of the Chicago School?

Dan Klein 33:00

I love Ronald Coase. I love Milton Friedman. I love a number of other people who are associated with Chicago. I am pretty high on Gary Becker, George Stigler I have a much bigger problem with... It's significant, and there is a cluster there. There's also the older Chicago cohort, you've got Knight and Viner and Simons. It's a mixed bag. I don't know, I'm not good at this... I don't like this kind of glib, quick overrated/underrated thing.

Frankly I can tell you quite a bit of what I think about Coase, but let's not get into that. I want to write more about Coase, but let's give me your next one.

Infovores 34:01

What are your thoughts on Silicon Valley libertarianism?

Dan Klein 34:08

I'm not even sure I know what that means.

Infovores 34:12

Like Balaji Srinivasan, Peter Thiel, Marc Andreessen.

Dan Klein 34:17

I don't know much about those guys. I watched some of Peter Thiel on YouTube. I think he's quite good. I don't know what else to say. I don't get why all this excitement about René Girard. I mean, he doesn't seem bad, but I don't really... I can't say, I haven't delved into it. But anyhow, next?

Infovores 34:45

Maybe this will be the last one—“State Capacity Libertarianism.”

Dan Klein 34:52

That's like an expression Tyler Cowen brought up. I don't take it seriously. I don't know what it means really, I don't know why it would be a good word for whatever it means. I don't really like the expression state capacity to begin with. I don't know why if you want to try to get to some better outlook, you want to particularly use the word libertarianism. So on the whole, not great I don't see much there that's of merit.

Infovores 35:24

What academic, activist, or public intellectual is having the most positive influence for conservative liberalism over and beyond what is commonly recognized?

Dan Klein 35:40

Which academic or public intellectual is doing the best things for conservative liberalism? That's easy.

Infovores 35:48

Oh, who is it?

Dan Klein 35:48

Jordan Peterson.

Infovores 35:50

And why is that?

Dan Klein 35:53

He fits your description. He's got a huge impact. And I think basically what he says is very good. And even great. I don't... yeah, there's minor things I could carp on perhaps. But basically I think he's a phenomenon.

Infovores 36:13

And what do you think makes him so great? Why is he able to do what so many other academics are not?

Dan Klein 36:19

I don't know. I mean, he's really quite extraordinarily gifted. First of all, he's got an extraordinary range of knowledge. I've been listening to some of his interviews, and I'm very impressed with how much he knows about the fields that his guests work in often.

I have to say though Jordan, if you're listening, when you're an interviewer you have got to realize you're not being interviewed. So you have got to let your guests be more like the guest. But apart from that, you know, that's fine because it's almost like a charming foible, because he's so vital in his thinking, and you feel that he's a man who is thinking. You can kind of feel him thinking, he's not just replaying. I mean, sometimes he does repeat himself certainly. But still, we all repeat ourselves of course, especially if you've got social media going and all that. But anyway, I'm very impressed with Jordan Peterson.

Infovores 37:23

I think that you key into something really big about his appeal there when you say that his foibles are almost endearing. I think that's a huge part of what he does to the point where even when he had major health challenges, related even to some mistakes that he made, it was very relatable to things that I had experienced or that people that I know have experienced. And there's a certain kind of authenticity in it that I think is hard to fake.

Dan Klein 37:55

Yeah I think he's quite open and frank in his communications, his discourse.

Infovores 38:05

So it's clear to me now that the concept of doing overrated/underrated is really overrated to you.

[laughter]

Dan Klein 38:13

Yeah it's overrated, at least by some people.

Infovores 38:17

But humor me with this one—was Tyler Cowen overrated or underrated as a roommate? How did he differ from what you expected, since you even knew him beforehand.

Dan Klein 38:28

Again, it's by whom? We got on great as roommates. We were roommates when we were undergraduates at George Mason, and those were great, great years. And we certainly got along fine as roommates and friends and everything. So there was no problem at all, just room-mating.

Infovores 38:57

What's something you learned about him that would be hard to learn without living with him?

Dan Klein 39:03

Just the reading habits. Like the tremendous amount of material he deals with. I mean, it's obvious from the blog, but in a sense seeing it in the living room somehow still makes a big impression.

Infovores 39:35

I remember reading somewhere that you were actually not a particularly big reader when you first met Tyler Cowen. What influence did he have on the reading habits that you took up? What would you look like in a world where you hadn't met him, just in that one narrow aspect?

Dan Klein 39:55

Yeah I guess a little bit. I had met him earlier, because we were in the same eighth grade class. We had met incidentally before that and became friends in eighth grade. But he didn't get me into the free market stuff and politics stuff until maybe sophomore year in high school. I went away to boarding school for my junior year of high school, and that also made me a little bit more oriented towards reading, perhaps. But then I kind of clicked in with all of this libertarian type of stuff starting with Tyler. I just fell into it and got into it.

Infovores 40:55

I had a college teacher and he would have all of the students suggest a book, and he would read all of the students' suggestions within the semester for all of the sections that he was teaching. And before that happened, I actually had kind of fallen out of reading. So I kind of was resonating with that, and maybe projecting my own experience onto some of your interactions with Tyler from that perspective.

A Straussian Reading

Infovores 41:45

So I'd like to close by delving a little deeper into one section that really stood out to me from Theory of Moral Sentiments while I was preparing for this interview. It's from Part Two about halfway through that section, and Smith suggests that there are some crimes that are difficult to punish because they don't immediately or directly hurt anyone. He tells a story about a guard who falls asleep during his watch and is sentenced to death for putting the entire army at risk. Are you familiar with that passage?

Dan Klein 41:59

Sure.

Infovores 42:01

And so the logic of the penalty is clear, and it seems like Smith basically accepts that logic. But then he says something interesting. He says,

Though such carelessness appears very blamable… a man of humanity must recollect himself, must make an effort, and exert his whole firmness and resolution, before he can bring himself either to inflict it, or to go along with it when it is inflicted by others.

If you take this statement at face value, he seems to be suggesting that to carry out justice here you have to go against the impartial spectator, which is Smith's idea of a moral guide throughout the book. How do you interpret what Smith is arguing here?

Dan Klein 42:44

I think what he is saying is that you see the centinel falls asleep during his watch. And there's a standing rule in the military that if you're caught falling asleep on your watch, you are put to death. And so the guy fell asleep, he was caught, and nothing happened while he was sleeping. No harm in the moment came to anybody. And consequences play a very significant role for Smith. He says our sentiments are greatly affected by perceived consequences of things and we see no ill consequences of his having fallen asleep. And so it kind of seems like “no fault no foul”, why should we kill this guy?

But in fact, Smith favors killing the guy. And I would say Smith feels he has the impartial spectator’s agreement or approval in supporting the death penalty, because you’ve got to see the larger consequences of letting the discipline go flabby. So it's about kind of first impressions and thinking deeper, in this case. It’s kind of a thinking fast and thinking slow distinction, where certain immediate feelings of propriety make it feel like the death penalty is not in order and is overly severe and all that. But this is the responsibility of a serious, disciplined army is to maintain such disciplines. So it's something along those lines.

Infovores 44:51

That's interesting. I think that certainly seems to be the surface reading. But I'm not sure that I agree.

[laughter]

Dan Klein 44:59

Okay, give me your esoteric reading.

Infovores 45:02

Right so if I can be just a little bit Straussian, I see Smith suggesting that the impartial spectator is correct here. There are instances throughout TMS where, I haven't read all of it yet, but there are instances where he makes exceptions to the spectator being correct as you’re saying. But it seems to me generally when those exceptions are made, the impartial spectator is wrong because of some personal defect in the individual sort of simulating or accessing this spectator. A defect in their character. And when Smith says, "a man of humanity" has to recollect himself and basically force himself to go against the impartial spectator here, I think he's suggesting that this is not one of those cases where the impartial spectator is wrong.

I think he's suggesting that there's an important counterbalance to justice that may have some weight in this case. And that even if there may be cases where nothing observable is done to harm an individual person but still require a punishment that those cases should be very few and far between. That we should be hesitant to go against the enlightened intuition represented by the impartial spectator here.

I think this is strengthened by a passage just a few paragraphs later. In the earlier editions of TMS this was even more the case, but he makes kind of a digression where he starts talking about the mercy of God, and the limits of justice, and the need for an atonement. And it seems to me like he may be suggesting that there's a role for mercy to escape the bounds of justice in this situation. What do you make of that esoteric reading?

Dan Klein 47:02

Interesting, and a lot about it I think I like. One thing at issue here is the multiple meanings of the expression "impartial spectator". I'm not sure what you're thinking is quite what I was thinking. It seems like maybe what you were thinking was someone's conscience, and that is one meaning. But there's also the supreme impartial spectator, and the conscience is not necessarily a faithful or accurate representative of that supreme sense of the impartial spectator. That supreme sense you could think of along the lines as God. So certainly when you're saying the impartial spectator could be wrong, Smith would certainly say that your conscience could be wrong, but he would never say that God could be wrong. And I'm not sure if you were implying that, but that God is kind of right by definition. The supreme impartial spectator is right, by definition, that's just part of the way things are organized in his ethics and in benevolent monotheism generally.

But then what you're saying, that there are exceptions, and leniencies, and mercy. I don't deny all that. But I never read the centinel passages as him subtly suggesting that maybe the death penalty there was improper. Are you suggesting that?

Infovores 48:51

I don't think he's suggesting it's generally improper. He makes a clear distinction between putting the guard to death who has fallen asleep through human frailty, on the one hand. And in contrast, he says it's natural for us to be vengeful towards someone who has committed outright an unjust murder. If the latter person is set free from prison without truly being punished, that is a great injustice according to Smith.

And so he's taking a very nuanced position here in my view. He would affirm the death penalty in certain cases, maybe even many cases. But in this particular case, he seems to be carving out a little bit of space for what I would call the mercy of God, and along the lines of what you're saying, Smith doesn't talk of God as being incorrect when he invokes him throughout the book.

And so a lot of the things that he has God saying in this long passage that the section ends with, that was cut out just before the seventh edition I think, excuse me the sixth, seem right in line with classic insights from Adam Smith. For example he says, "the doctrines of revelation teach us how little we can depend upon the imperfection of our own virtue", which is related to this idea that we need the invisible hand to direct our imperfect acts, our selfish acts, or our acts that are not really motivated by benevolence most of the time to serve the greater good of the whole.

And you make a point in “Smithian Morals”, in the essay about the invisible hand, how Adam Smith places that reference dead in the center of his works to highlight it in a way that was maybe not entirely obvious to people who weren't reading closely. And I wonder whether he does kind of a similar thing here, because this passage on the atonement is in the middle of part two. It's not in the middle of the overall work, but it's kind of right in the middle of that section. And it seems to suggest some deep points that he may not have felt totally comfortable going into in other places in the work.

Dan Klein 51:39

I can see you've got a taste for esoteric reading.

Infovores 51:41

Yeah, I've written several posts about it.

Dan Klein 51:44

Oh, really? Good. It's kind of rare, but I'm glad to see some young people getting into it.

Infovores 51:53

Yeah, so I really appreciate your willingness to have this very kind of nerdy and wide ranging conversation about morality, and economics, and religion. I really appreciate the book you wrote, because it helped me to get deeper into Theory of Moral Sentiments and connect some of the things that maybe I wouldn't have understood as well trying it on my own. And so I hope that people who listen to this will take a look, it's available as a free download via the Fraser Institute, and it's the first of three volumes that you're working on. Is there anything you'd like to to add about the forthcoming volumes or other things you're working on?

Dan Klein 52:43

No, you covered it pretty well. I'm really grateful for this conversation, I've enjoyed it. And I look forward to hearing your future esoteric readings.

Infovores 52:56

Thank you, and likewise. Appreciate you and hope you have a great day.

Dan Klein 53:02

Likewise, thanks a lot.

Thanks for reading! If you enjoyed this, check out my other writing at Infovores Newsletter. You can also follow @ageofinfovores to keep up with me.

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Transcript edited for clarity, Spotify link here.

Infovores  00:04

Hello and welcome to Time Well Spent—A place where the most brilliant minds in the world take on the toughest questions in Science, Politics, Technology, and much more.

Today I have the pleasure of speaking with someone I greatly admire, economist and author Arnold Kling. Arnold writes frequently about the economics of business and tech at “In My Tribe”, often drawing on a wealth of hands-on experience he obtained outside the ivory tower. We will focus our discussion on his book “Under the Radar: Starting Your Net Business Without Venture Capital”, with an emphasis on how its advice can be adapted to the current environment. Arnold, thank you for joining me!

Arnold Kling 00:30

Thanks. So I'm also curious to see how it can be adapted to the current environment. It was written shortly after the original dot-com crash, so there have been a few changes since then.

[laughter]

Infovores 00:40

Well I just read it a few days ago, and I actually was surprised how many of my questions it answered, so I'm excited to get more of your thoughts.

So you started your business at a time when no one was really doing anything commercially online—before you corrected them, some of your early clients even assumed you had personally created the internet! Can you give a little more background on what things were like at the time, and how you came to realize you had an opportunity to build something no one had fully conceived of before?

Arnold Kling  01:33

Okay, so I think part of what sold me on the internet was the work of Hal Varian, who later became the chief economist at Google. It was fortunate that I happened to latch on to him, because he obviously had a lot of insights into where things were going. So just from a technical perspective, he distinguished between a circuit switch network and a packet switch network. The telephone system at the time was a circuit switch network, so when you made a connection it created a circuit between you and me. If we were doing this conversation on a circuit switch network, we would have a durable connection that would only drop after we shut down. With a packet switch network, little packets of data could be sent over any number of routes by these routers, and they're sort of sent from end to end. So right now I assume that what's going on is that we're sending this conversation to each other in these little packets, and then they're being reassembled where I am and where you are and somehow put in the right order. So it sort of makes it look like we have a durable connection. But we don't have a durable connection, we have this intermittent connection.

Anyway, Varian looked at the economics of that process and saw that packet switching was getting cheaper and cheaper because it was following Moore's law while circuit switching was not. And so there was a big future for packet switch networks. In fact, other people were able to sort of pinpoint the date at which telephone calls would be cheaper to send over the internet than over the old AT&T telephone network, and that prediction proved to be accurate. So for a technical reason, I was aware that things were going to move to the internet.

And then some of it was just kind of romance. I was at Freddie Mac, I always say, before it became famous. And some time in late 93, we went to a GE research site, I think it was, and we were looking into their methods of doing underwriting using computers until at some point the meeting came down to the higher ups talking to each other and I was no longer welcome in the room. So one of the engineers took me downstairs and showed me Mosaic, which was the first graphical web browser, it came out of the University of Illinois. And I just fell in love with it. I said “this is the future.” I was like this character in a book called “The Wind in the Willows” who just gets fascinated by cars.

Infovores 03:00

Was that the toad?

Arnold Kling 03:05

Yeah, I was the toad.

So I had that view, and I was in a very low point in my career at Freddie Mac. So I was willing to go off and take chances, and I rationalized to myself that “well, worst case is that I'll learn something about this internet.” I did think the internet was gonna be a big deal, so in the worst case I thought I could be some kind of internet consultant when all was said and done. So I conceived the business idea in one sleepless night of setting up a site that would have everything for people who are interested in buying a home—including home listings and mortgage opportunities—because there's so much fat in the home bank system.

There still is a lot of fat, and I've learned since then that you can't cut a lot of it out because the rent-seekers are able to hold on to their rents. But if you'd asked me in 1994 what the home buying experience was gonna look like in 1999, I would have said that we would have almost no transaction costs, consumers would have lots of knowledge about mortgages, there are not going to be real estate agents anymore. I'm not the only one who ever came up with that idea, but that was in my mind and so it took awhile for reality to set in with me.

Infovores  07:13

So it sounds like you had a tremendous amount of expertise in housing, and then you got exposed to this new technology, and basically just ran with it and said, this can be applied to something that I know a lot about.

Arnold Kling  07:27

Yeah. And fortunately I hadn't gone to business school because later that summer I talked to a couple people who had, and one of them pointed out that hardly anyone really had access to the worldwide web. He said, “you know, you really should just should use Gopher”, which was kind of a weaker protocol, less user-friendly, that no one uses anymore. And another one I talked to sort of grilled me on who could really use the internet, because ordinary civilians like you or me—including me when I was running my business—had to access it through a very low bandwidth modem. So if you were getting just one picture over the internet it could take two minutes to download. Whereas the people who really had connections that were high speed were people on universities, those were the only places that had it.

But there was one entrepreneur that I talked to about the state of the business about four or five months into it, and his eyes just lit up and he said, “This is an apartment market.” College students aren't going to buy homes, but when they graduate they go looking for apartments in new locations. So he said I should put apartment relocation service on, and that was actually the first thing that actually worked. Although it's not the most lucrative part of the real estate business by any means.

Infovores  09:27

So to make it really concrete before we move on, what were the core services that you provided? You had like a mortgage calculator on your website, you had…

Arnold Kling  09:36

Okay, so I had various articles informing people about mortgages because most people don't know much about them. And the mortgage calculator, which is now a dime a dozen, but back then it was sort of a novel thing that you could stick in an interest rate and a home price and get back what the monthly payment would be. So people thought that was cute.

We had very early on the New Homes Guide for Washington DC. There were about 24 partly independent, but sort of mutually linked New Homes Guides for different cities. Washington and Philadelphia were the ones that were most interesting to me. There was a guy from Philadelphia who said he wouldn't bother putting his New Homes Guide on there because he said it's an apartment market. But people liked to look at the DC New Homes Guide, they could browse through the the pages of the New Homes Guide and look at homes for sale, and they could sort by various criteria. And even though it was very little certainly by today's standards, it was very popular with people because there wasn't much on the internet that was of real practical use.

This is the web in April of 94, there were probably fewer than 1000 websites total. So people liked just playing with it to play with it, just to see what they could do, but that wasn't great at generating revenue. One comment that somebody left on the site was, “Congratulations, you've started your lemonade stand on the moon, you just have to wait for the astronauts to get there.” That was kind of the situation as of that time.

If I can say one more thing which was really crucial, it’s that it was really, really hard for an ordinary civilian to get on the internet. And I think what the comparable thing today is that it's really, really hard for a non technically skilled person to invest in crypto. And that's why these third party exchanges have kind of cropped up. So Windows didn't have a way for you to get on the Internet, it didn't have what's called a TCP/IP stack. So I had a disk, a little diskette. It was a floppy disk that had a TCP/IP stack, but I just could not figure out all of the procedures. There was this complex set of procedures, and I couldn't get it to work. So I had to call the internet service provider, which is a small company, and the president of company drives over at night in a driving rainstorm and helps me install the TCP/IP stack. So I really should have known like a week after I'd started the business, I should have realized that boy, you're in a lot of trouble.

So that situation lasted from April of 94 to August of 95, and that was a big reason I had no competition because ordinary civilians couldn't get on the web—why bother putting up something related to home buying or mortgages? And it was only in August of 95 that two things happened: Microsoft finally put out Windows 95, which had a TCP/IP stack built and so all sorts of people could get on the internet. And America Online gave its users access to the web, which they've been holding off and holding off doing because they thought it was a threat to their business, but ultimately they had to do it or else they would lose their customers. So when AOL came on and Microsoft had Windows 95 then all of a sudden the web was exposed to the general public at that point, and there was actually some potential for the business.

Infovores  14:08

My understanding is that up until 1993, it was even illegal to do most commercial activity on the web. And so you're hopping in right at the beginning in early 1994 before all the entrepreneurs have come in and really put a lot of effort into making this accessible to the public so they can make money off of it. And so you really didn't take off until a few years later, which I'm sure we'll touch on later in the interview.

So in the book you teach many things an aspiring business founder wants to know—what makes an idea worth pursuing, how to build productive relationships with team members and clients, and how to finance it all. But you also describe the process of starting a successful business as “a series of miscalculations”.

How does that tie into this discussion and how is that different from how people conceive of the process of entrepreneurship?

Arnold Kling  15:03

Well I think a lot of people, especially venture capitalists, have come to understand that founders make a ton of mistakes. Imagine a kickoff returner in football trying to find a hole where they can run for a touchdown, and they keep running back and forth and side to side and trying different things. That's kind of what this is like. I think most people familiar with startups would say that it's actually rare for something that you plan before it starts to work out exactly according to plan along the way.

I think that what was amazing in my case was the tailwind that the growth of the web provided. It’s like nothing that you've ever seen in history and probably won't see again.

Infovores 16:10

You never know.

Arnold Kling 16:15

Well you know, I would have said you won't see it again, but I think the smartphone acceleration did that. People who were early to figuring out what would be useful apps were in the same position with a strong tailwind. So once you had that tailwind, you could afford to do a lot of stupid things and do a lot of trial and error. The big first mover advantage I had was that by the time competitors came in, I already knew that what they were trying wasn't going to work because I had already tried it myself. So that's part of the first mover advantage that maybe you don't appreciate—you make your mistakes early, and then you just sit and watch other people make those same mistakes and throw even more money at it.

Infovores  17:11

So I think that in the book you put a pretty encouraging spin on this idea. Basically trial and error is so important that you can get a lot of things wrong and still end up with a really great business if you have a good idea that's well matched to your talents.

I wonder how that may have changed more recently. If you look at a company like Amazon, I get the sense that they're essentially in the business of doing trial and error at scale. And they're super optimized, they have a bunch of small teams that experimenting with all kinds of possible innovations, they run user testing, and they iterate over and over again from there. Does the existence of companies like this make you less optimistic about “under the radar” style entrepreneurs?

Arnold Kling  17:58

I'm not sure what to say on that. For example, at some point around 1997 or 1998 Microsoft jumped right into our space—they put out Microsoft HomeAdvisor. The first version of that even had some things that were better than ours. But I think it just wasn't important to Microsoft, it didn't draw enough attention from their corporation. If they had it as a separate business with an all out drive to make it work, they could have buried us.

There was nothing that we had that was so proprietary that they couldn't have buried us, but they just kind of let it languish. It's was like oh yeah we know we can do this, and we will do it well, and probably the executive who was in charge of it cared about it. But it was just a lack of will to compete. So I think the thing about Amazon is, if they have the will to compete in whatever you're doing, God help you. Because you're right, they've got all this capability.

I would say in general one of the things that surprised me is how corporate the Internet became, and how successful some of the bigger businesses became, and how much more difficult it became for a smaller business. Part of what attracted me to the internet in the first place was this idea that the little guys can come in and because the cost of a web server is tiny compared to what the cost of building a store would be, it appeared to be just the ideal thing for the little guy. I remember Daniel Pink sometime in the 1990s wrote something called “Free Agent Nation”, which was another one of these sort of manifestos for the little guy. That’s in some sense what “Under the Radar” is—I positioned myself against venture capitalists. I thought the venture capitalists are looking for these homeruns, but the internet's gonna be a bunch of singles and doubles. And I wasn't really right. The venture capitalists and the big companies, or the companies that became big, ended up being much more powerful than I ever expected.

Infovores  20:40

What's maybe the best hope for avoiding notice of one of these companies and starting a business today?

Arnold Kling  20:50

That's a good question. I think you'd have to know the domain, you have to be pretty confident that you have something that they won't go after, or that you can have some particular proprietary advantage. So I see in your Zoom box, you have “Time Well Spent.” That’s Razib Khan's phrase, right?

Infovores  21:20

Oh, yeah. He has like a reading list and I think that he uses a similar phrase.

Arnold Kling  21:25

Yeah so you can imagine his domain knowledge in genetics being something that could be a real competitive advantage that somebody else couldn't replicate necessarily. But even so, I think in biotech in particular the small companies that have domain knowledge, their exit often has to be purchased by a major pharmaceutical company, as opposed to being able to go all the way to doing an IPO themselves.

Infovores  22:08

Definitely an interesting topic, and you talk about potential exit strategies into venture capital in the book as well, which people may be interested in looking at.

So my next question is related to the interviews that you do for your book. You interviewed 25 other founders, and I was wondering what additional lessons or cautionary tales have you learned from these individuals since the time you published?

Arnold Kling  22:31

The most striking thing I learned was the problem that I call the early divorce. Venture capitalists don't look for individual founders, they look for teams. So I was an individual founder for the first year and then kind of met my beshert, as they say in Hebrew, which means your love match. But a lot of people start out, and it turns out, their partner really doesn't want to put full time in it. Or it turns out there are a lot of things that you have to work out with your founding partner, or founding partnership if its three people or four people. And if you have to go through a divorce that just really sets sets people back.

I just hadn't thought about that, but when I did interviews I found people who'd gone through that. And that's just a very tough thing. The moral of the story is if you're going to found something with somebody, don't make it on the basis of assumptions, or half a conversation. You really have to work out a bunch of contingencies in advance and really know that you can work with somebody and that their assumptions connect with your assumptions.

Infovores  24:14

And you should have maybe a trial period or some plan for what to do if something doesn't quite work out.

Arnold Kling  24:20

You definitely don't want to be in a situation where you haven't agreed how to split things up if you have a split up. So if you've done 80% of the work, and the other person says that they're entitled to 50% of the company that's just brutal.

Infovores  24:39

Yeah it was striking to me how many of the arrangements you made as founding partners, even as partners getting along really well, were pretty much just even-steven in order to avoid conflict.

Arnold Kling  24:50

Yeah but then there's an assumption that it's going to be even-steven in terms of contribution as well. When that gets out of line, that's a real problem.

Infovores  25:03

Gotcha. Let's play a game of overrated/underrated, entrepreneurship edition.

We can start with an easy one—Elon Musk, overrated or underrated?

Arnold Kling  25:13

That's an easy one?

[laughter]

Infovores  25:14

I was being facetious a little bit.

Arnold Kling  25:17

[pause]

People have thrown a lot of banana peels at him, and he hasn't quite hit the ground yet. So I guess I'll go for underrated.

Infovores  25:40

What causes you to hesitate?

Arnold Kling  25:44

Well, because he's so highly rated to begin with. So he can't be underrated by much.

Infovores  25:51

Make sense. The show “Shark Tank”, overrated or underrated?

Arnold Kling  25:57

Overrated. It's just entertainment. I think it's a lot of baloney.

Infovores  26:03

Do you think it actively steers people wrong in any specific ways that bother you?

Arnold Kling  26:09

Hmm I guess I don't… I'm not a devotee of it. When I watch it, it’s because I'm trapped with people who watch it, and they enjoy it more than I do.

I think to the extent that it focuses on the idea and not the execution of it—and maybe I'm not even wrong about that—but to me ideas are easy to come by, execution is hard. I don't know that you can explain your execution in a five minute pitch.

Infovores  27:05

This next one predates the internet by a little bit, but I'll ask it anyway—is Phil Knight overrated or underrated? That’s the founder of Nike.

Arnold Kling  27:18

Yeah if you hadn’t added that I wouldn't have gotten it. I'm gonna say I don't know because I'm not familiar with the business history of Nike.

Infovores  27:31

I liked his book “Shoe Dog” quite a bit about his founding experience.

How about marketing and sales relative to other things that entrepreneurs tend to focus on?

Arnold Kling  27:42

Oh, as you know from reading the book, I'd say it's very underrated.

The whole point of the book is that the time that founders spend fundraising, they should spend selling. There are creative ways to sell before you really have a product. If nothing else, you can pre-sell, you can say what would it take for you to buy this, or how much would you pay for this? I'm a big fan of selling. So I'd say underrated, especially by the person who thinks in engineering terms or in buzzword terms. Sell something.

Infovores  28:26

You suggest in the book that personal charm is an underrated quality. How can one learn to express charm over email and text communications?

Arnold Kling  28:36

That's an interesting one. I think just respecting people's time is important, and early on you should be trying to hear from them as opposed to sell them. So sending somebody a long description of your product is probably not as good as finding out where their pain is.

That's really a key in selling—you want to know the customer's pain point. That's much more important than the features of your product. That's a lesson I've learned. I learned both in person but also from reading a book called “Strategic Selling”, which has been around a while but has survived and been revised. It's probably more valuable than my book, but I’ll leave that to others.

Infovores  29:48

It seems like charm under your definition is is highly related to empathy or putting yourself in their shoes to the point where you’re able to see things the way that they themselves see it.

This is a broader topic of interest in your later work on politics and the conflict that arises between individuals due to asymmetric insight. What connection would you maybe expand on there?

Arnold Kling  30:18

I think that it's hard to beat something into people's heads. Meeting them where they are is the skill which involves sort of this cognitive empathy. And I think it's very important in sales, because you have got to understand the other person's pain point.

One of the things that helped a lot in my web businesses, one of the first things I put up was a comment block, and I read every comment within 24 hours. So people could comment on the site and just say whatever they wanted to let me know what they liked and what they didn't like. So I was constantly making tweaks, or maybe even major changes in response to those comments. Having feedback from your consumers is really important, and I think a lot of people realize that.

Infovores  31:34

And trial and error is really important here too. Because when you put out a product sometimes they really like the feature that you didn't think much about and they don't care about the thing that you spend all your time on. You’ve just been hyping it up in your mind.

Arnold Kling  31:48

Yes and especially on the internet, you just have to keep changing. Sometimes you see something that a competitor does and say we should be able to do something similar to that, or you wait and say that doesn't look like that's gonna work and you watch it. Hopefully you're right and it doesn't work, but then sometimes you have to change your mind.

Infovores  32:16

Right after you published the book, you came out with a Q&A. And in that Q&A, you answered the question, “what business would you try to start today?”.

I'm curious to hear what potential areas you would be most excited about right now for starting a business.

Arnold Kling  32:35

Okay, so where's the tailwind? I think what we've seen in the last couple of weeks is that there’s a tremendous tailwind behind machine learning and artificial intelligence. I think Open AI is just begging you to come up with businesses that are based on that. So that looks like a strong tailwind.

I think that genomic medicine or biotech in general is interesting but a little bit tougher, because I think you you have to have a lot of domain knowledge. And you've got to incur a lot of fixed costs and a lot of risk because in the five years it might take you to have a viable product you could be superseded by some technological change. So I think I go with the machine learning as something that has the tailwind.

I'm trying to think what would I do if I were to start a business, because I don't think I could do machine learning. I think the field of education even though a lot of people are trying things there. I think there's a ton of potential there because I think there's a lot of dissatisfaction with the way public schools are going or even the way private schools are going. There’s a lot of dissatisfaction with higher ed. So I think that's another another field to look at, with the downside being that there's tons of competition.

And the thing you have to get past there too is that people are very change averse. They don't like what they see maybe, but there's a lot of status quo bias. People just think “this is what a K-12 classroom looks like.” They think “this is what a college looks like” or “this is what a credible degree looks like.” So probably that’s the thing you most have to think about in the education space. You have to think, “there's this status quo bias to begin with, how am I going to get get past that?”.

And again you're not going to talk people out of it, you're going to have to meet them where they are, give them trial periods, and hopefully kind of lure people in gradually to a new model.

Infovores  35:25

It seems like homeschooling is one area where this could really take off. If you did it right it could spread through word of mouth as you just deliver a really good product, it wouldn’t even need to be a huge number of parents necessarily in a metropolitan area.

Arnold Kling  35:42

Yeah, I agree. I think that for an ordinary civilian who's not super techie or not an expert in the biotech field, education probably has more potential.

It'd be fun to try to marry the education with the AI. If you could figure out a way to turn kids loose on the Chatbot and find things out for themselves and kind of monitor them and mentor them, that might be a cool thing to do.

Infovores  36:22

And maybe you could even start to kind of whittle away at education’s monopoly by starting with kind of the side arms to education.

So there’s music education, for example. I was just playing with ChatGPT the other day, and I was saying write me a song about this in the style of this artist and give me chords. And it was giving me chords. It was simple stuff, but for a beginner guitar player, like I would have loved to have that. And it'll tell you the strum pattern in more detail then you usually get even if you talk to a teacher. Like maybe the teacher knows how to play guitar, maybe they even have some experience teaching, but it's just so intuitive to them that they don't explain it in the terms that you're really looking for it.

Arnold Kling  37:05

Yeah that's a really cool application. Awesome.

Infovores  37:16

You often say that price discrimination is a great way to be profitable. Can you explain what you mean by that and give a few examples for how a new entrepreneur could apply that insight?

Arnold Kling  37:27

A lot of this was sort of informally laid out by John Perry Barlow, whose claim to fame was that he was supposedly a Grateful Dead lyricist although I don't know that he wrote any of the songs that I most liked from the Grateful Dead. And then he became a founder of the Electronic Frontier Foundation, kind of an early rebellious, free speech oriented group. Who knows what they're like now. But he wrote something called “The Economy of Ideas”, which pointed out that ideas are not only costless to distribute but that their value goes up the more distributed they are. And that's a real difference. So Hal Varian along with Carl Shapiro wrote this book “Information Rules”, and the lesson you get from that is that when you go from distributing something in the form of atoms to the form of bits, it really changes things.

So in 1990, music was distributed physically on CD-ROMs. Once you get distributed on the internet with bits, the marginal cost of distribution is zero. If anything you want things to be more distributed, because it'll be more popular. And that's been true of a lot of businesses, right? For Facebook, the marginal cost of an additional customer is negative—the more customers they have, the better off they are. Developing Facebook requires a huge fixed cost, just as with developing pharmaceuticals. There’s a huge fixed cost, but manufacturing the pill itself is low marginal cost. And that's not just a normal economic good, it's almost like a quasi public good. So the economic problem isn't about trying to set marginal revenue equal to marginal cost, because the marginal cost is close to zero. If you do that you're you get you get nothing but losses. So the the challenge is to recover overhead.

That is the main economic challenge of pretty much any internet-based business and also some non-internet based businesses like pharmaceuticals. And so what follows from that is that if you're trying to maximize revenue, you look for ways to price discriminate—you look for ways you can charge the lowest possible price to the marginal user. That’s the user who has what we call elastic demand, the one who is just barely interested in what you have, because even charging a low price, you get something for them. And then you want to charge the highest possible price to the person who really really wants it. So my example with Twitter is that people who have very few followers probably don't get a whole lot of value out of Twitter, but people who have a lot of followers do. So the thinking is that you charge a monthly fee to people with more than, let's say 1000 followers or 500 followers, and charge less to other people. And that would be a useful way to price discriminate. If you charge the same fee to the person with 100 followers, who is not getting that much value on Twitter, they’ll just say forget it. So you have to charge them a low fee. But the person with high followers says this is really important to me, so I’ll pay.

And you can also find other features of Twitter that people like to use, so a lot of people like to use direct messaging. There you can say, well if you're not a subscriber, you can't use direct messaging. People like to use quote tweets and retweets, and you say, well to do more than five of those a month you're gonna have to be a subscriber. You can have different tiers, with high paying subscribers, low pay, and no pay. Those are some practical examples of price discrimination, and we see some of it in non-Internet businesses, but it's got to be all over for internet business.

Infovores  42:09

I'd like to close with some questions kind of in the mold of Tyler Cowen's “Production Function”—basically trying to get a sense for your more personal background and how all of these things have informed your success.

What features of your personality or background most enabled you to make such a bold shift into the business world? You are an MIT trained economist, you had worked at these prestigious institutions at the Fed and Freddie Mac. I imagine that many of your colleagues must have thought that you were crazy to leave that environment…

Arnold Kling 42:30

Oh, absolutely.

Infovores 42:35

They probably looked down on entrepreneurship to some extent as well.

Arnold Kling  42:48

I wouldn't say look down on it, but yeah they thought it was kind of crazy.

Part of it is contrarian personality, and part of it is… I'd say probably the number one thing is creativity, that there are some people who are compulsively creative. I think I'm one of those people. And large organizations are very misaligned with a compulsively creative person. Because the bureaucracy is set up to say no. I think that's actually rational, I once wrote an essay saying that it's basically rational because if you're a middle manager at a big organization, you don't have enough skin in the game in the organization to justify them taking one of your wild ideas and running with it. If they did it frequently, it would be bad because the expected value is negative and they've got valuable franchises. So it's quite rational for them to set up a vetocracy, or a veto bureaucracy. So there's a fundamental misfit with these large organizations, where if you're compulsively creative you just have to deal with it.

So it was just time for me to go. And I was just lucky that the internet was sitting there as a kind of safety net, oddly enough, that I could jump to.

Infovores  44:31

What role has religious community played in your entrepreneurial endeavors, if any?

Arnold Kling  44:36

I would say not much. I’m part of a Jewish synagogue, and I think of the religion as there for you primarily at the main points of life. So Judaism has a ritual at birth, it has a ritual around the time of puberty with the bar mitzvah, it has a ritual when you get married, and it has a ritual when you die. And so that's not too closely related to the business world.

Infovores  45:17

I thought perhaps it would be a little easier socially to part ways with some of your work colleagues if like… at least in my life as a religious person, I feel almost insulated from what other people think because I'm mainly tied into the religious community.

Arnold Kling  45:35

Yeah, I don't know how I was able to part from colleagues… that was maybe again just another part of my personality. There are people that I've been engaged with really strongly I can remember from elementary school, high school. Not as much in college, somewhat in grad school, then at the Fed and at Freddie Mac, but then… you're thinking at the time that you’ve got such close relationships with these people, and yet it just doesn't take me that long to forget about most of them, which is maybe sad, but I guess I've not had a problem disengaging. I don't think the religion has anything to do with it, I think it’s just either a feature or a bug in my personality.

Infovores  46:41

Very interesting. You started Homefair when you were already 40 years old, what advice would you give to those who might be inclined to think that the entrepreneurship window is closed for them?

Arnold Kling  46:53

Because they're too old? I mean at my age I think it's true, I think that window is closed for me. And I've said that to people.

The evidence seems to be that 40 is close to the optimal age in terms of the chance of success. Younger people have more energy and perhaps more risk tolerance, but they don't have much domain experience. So as you get older, you really ought to be very dedicated to a business. In my book I say that it's okay to start something part-time, but at some point if it gets going you really have to put in a ton of energy. It's not just calendar time, it's really doing things that you've never done before. Like I never had networked before, and I'm still terrible at keeping track of people. In theory I could have thousands of LinkedIn people because over the years I've interacted with so many, but I let them go. It's just sort of my story about not staying so much in touch with the people in my life. So intrinsically I'm a terrible networker, but networking was absolutely the key to success for my business.

I think maybe that just comes down to being super motivated. Maybe once you're past age 40 or a little older, the ability to be super motivated kind of goes away because you have family ties, you have hobbies and interests that matter, you may have a more diversified portfolio of things to do with your time and that might make it harder to do the entrepreneur thing.

Infovores  49:15

You mentioned networking as not being an inherent strength of yours. Motivation and really caring about making a particular project work definitely helps to overcome that, but another piece is bringing in people who have strengths that offset some of your weaknesses. What advice would you give for seeking out people to do that?

Arnold Kling  49:38

That it's important and that it tends to introduce personality differences.

The key to my business was that I partnered with this guy in Arizona who had no technical knowledge and a high school equivalency degree. He had a temper. I mean, we could barely get along with each other. But the synergy was incredible. He was a super salesman, a very passionate salesman, and I never could have done it without him. I couldn't have done it without him, he couldn't done it without me, and we kind of recognized that. But even so, it’s hard to get along.

So he had this philosophy called “The Integrity Agreement”. The agreement was that if you've got a problem with me, you come straight to me. And if we're going to have a hard time dealing with it, we'll bring in a third person that we trust to observe and monitor us arguing through it. And we had to use that a few times. So you might need a formal mechanism like that when you’re working with somebody whose skills and mindset are quite different from yours.

Infovores  51:07

Last question. How do you continue to express your entrepreneurial side in the things that you do now, whether it's writing for “In My Tribe” or on other projects you may have in the works?

Arnold Kling  51:17

Well as I say, I'm sort of compulsively creative. I put ideas out and work on them to some extent, and if it feels like it's being rewarding I'll go deeper into it. If it's not, I'll back off.

For example, we talked about education and I tried to do an education startup. I tried to develop sort of an all purpose app for teachers to share quizzes and lesson plans, and didn't take me long to realize it wasn't gonna go anywhere. So I dropped that.

The Substack thing was almost like Mosaic, and I said oh, this is a great thing. I was very romantic about it, and I still am. And so I pursued that.

I've tossed out this idea of a network based university where you replace the grades and credentials with a mentor-student relationship, where the mentor gets to know the student and provides letters of introduction or letters of recommendation to employers. Again I feel like I'm too old and not sufficiently networked to carry it out myself, but I keep pitching the idea to different people and somebody somewhere is going to hopefully take some variation of that idea and put it into practice. So those are some examples I guess.

Infovores  52:49

What's the missing piece for having a university like that?

Arnold Kling  52:55

I think it would take someone who is very well connected in the business world to take it and run with it. Somebody who has a great Rolodex.

Imagine if Patrick Collison called up anyone in the business world and said, "Would you like to be a faculty member? I'm going to be a faculty member, would you like to do this?" I'm sure that 90% of people would accept. Once you have 500 people willing to be faculty members, you have tremendous network coverage and the ability to cover a lot of subjects. People in business often know a lot more than just their business, they know things like Shakespeare and about Classical Greece. With that, you could recruit high school students to take a year to learn from their mentors. Hopefully, a fair number of them would be placed in jobs by their mentors and continue to learn. That's the vision.

Infovores  54:44

It seems like Tyler Cowen's Emergent Ventures is not too far off from being able to implement this. He'd probably want to delegate to someone else to really take charge of it. But just with that network of grant recipients alone, you have a lot of people who owe a lot to Tyler and I'm sure would be willing to mentor young people in a wide variety of different areas.

Arnold Kling  55:05

Yeah, I think getting hold of young people is solvable. The bigger challenge is putting together a network of mentors, primarily from business, some of whom might be adjunct professors or people like Brett Weinstein and Heather Heying. These are people who really like teaching and think that universities hold them back. People like that plus business people would be the mentors and faculty, and I think that's the challenge. There's also the status quo bias of parents, but I think you can push past that.

The other question is how many young people are ready for something like this. Is it just a few thousand autodidacts out there, or are there 200,000 to a million? I think if you look around the world, not just the United States, you can find a critical mass of people who could be self-starters or autodidacts who could be the ideal students for something like this.

Infovores  56:30

Definitely an exciting idea. I hope to see it come to fruition someday.

Arnold Kling 56:35

So do I.

Infovores 56:40

Thank you so much for taking the time to talk with me today, it's been a pleasure.

Arnold Kling  56:42

Thanks, I enjoyed your questions.

Know someone interested in starting a business? Consider sharing this post!

If you liked this post, check out my other writing at Infovores Newsletter. Over there I write reviews, conduct interviews, and generally try my best to navigate a world of infinite knowledge. You can also follow @ageofinfovores.

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Transcript edited for clarity.

Infovores  00:04

Hello and welcome to Time Well Spent—a place where the most brilliant minds in the world take on the toughest questions in science, politics, technology, and much more.

My guest today often appears under the heading “The Wisdom of Garett Jones” and he is one of the very best people to follow on Twitter for insight into public choice and macroeconomics. Garett is the author of three excellent books, all of which address the biggest question in economics—why some nations are rich, while others are poor. Today we will focus our discussion on his latest release—“The Culture Transplant: How Migrants Make the Economies They Move to a Lot Like the Ones They Left”. Garett, thank you for joining me!

Garett Jones 00:35

Thanks for having me, great to be here.

Infovores 00:40

A central claim in your book is that populations have “deep roots”, a cultural heritage that continues to meaningfully influence attitudes and behaviors long after migrants relocate. To make things concrete for the audience, what kinds of attitudes are we talking about here? And why are they so important for national prosperity?

Garett Jones  01:39

Well, the attitudes that have been best measured by economists would be attitudes toward trust, towards savings rates and frugality, and attitudes toward the proper role of government in life. Those are three big areas that have big economic implications where economists know from theory and other research that these traits matter a lot.

And they've been able to find, not just for the US, but for countries in Western Europe, Australia, and Canada that for these traits migrants from the second through fourth generation seem to carry maybe half or a little bit less than that of their attitudes from their old country to their new country. So Italian Americans are in terms of attitudes a lot like Italians in Italy, German Americans are a lot like Germans in Germany. This is economically important, and it's underappreciated. It's been sitting in the academic journals for a long time. And I decided it was time to write a book that brings these well documented facts to the masses.

Infovores  02:13

So a lot of times when I tune into these debates, it seems like even people who are maybe a little bit skeptical of the open borders position mostly emphasize that “well, it would be great to have lots of immigration, really open borders, but you're going to have nativist backlash.” The problem is not so much the attitudes that the immigrants bring themselves, it's the reaction to those attitudes.

How much is that a part of the story or are you kind of diverging from that and emphasizing that really it's the cultural attitudes themselves that matter?

Garett Jones  02:56

Well I'm a big believer that the nativist backlash is important, and it's a huge negative we can't ignore. I wrote an essay about this for the Independent Institute a few years ago. And so Donald Trump should be counted as one of the costs of a migration of American migration policy, unfortunately. We probably have to count Brexit as a cost of EU migration policy.

Indirect effects are still effects. This is something that economists are trained on. Effects are not moral choices, they're not morality. They are effects that people should be taking account of when they're deciding what policy path to go down. But that's not the whole story. I do think that ethnic diversity is associated with conflict in rich countries, poor countries, middle income countries. We all know about wars that have been caused by ethnic conflicts. It's just a routine, horrifying finding of international politics and domestic politics.

But I'm saying something beyond that. I'm saying that basically the fact that migrants carry cultural attitudes from their home country to the new country and make the places they come from a lot like the place they left means that if you're an Effective Altruist, if you're a long-termist, you should be thinking about the long-run effects of your migration policy. “Past is prologue”, or past is at least half of the prologue. And so migration policies that focus on having migrants come from countries that are more successful than your own is a good idea. That's why I kick off the book by saying that poor and middle income countries should really consider a long-run pro-Chinese immigration policy.

Infovores  04:42

Gotcha. So speaking of long-term consequences, one of the things you point out is just how dependent the world really is on the innovation of just seven major innovating countries and you say that because these countries are so important, there's a really strong case to be made for being extra careful about immigrating a lot of people with attitudes who might disrupt that climate in the long-run.

But I wonder if given fertility rates in these nations, whether we pretty much have to find some way to make substantial immigration work. Otherwise, these places will become depopulated and it'll be bad anyway.

Garett Jones  05:14

Oh, this is an interesting point worth thinking about right? So there's seven countries, I call them the I-7, and they're the most innovative seven countries in the world. They tend to create the vast majority of patentable ideas and scientific research. They do the vast majority of the world's R&D. And so I argue for a kind of conservatism is sort of cautiousness about choosing migration policies that would really hurt the institutional quality there.

But you're right, if they shrink enough that's a huge cost of the world. But that at least means we need to think through that trade-off. I'm open to the idea that basically, taking some risks with the most innovative countries in order to keep their populations from going to zero is a risk worth taking. But quantifying that risk is well worth thinking about. Really what I want to do here is start a debate by getting people to realize that there's a huge globally important factor that the whole world should be thinking about when it comes to migration policy to the rich countries. And if we lose these geese that are laying golden eggs for the whole planet, that'd be a real loss for everyone.

The whole world has an interest in America's migration policy, and the whole world has an interest in Japanese migration policy. And changing that debate helps us get away from these sterile debates about whether pre-existing Americans are helped or hurt by low-skilled migration to the US. That's an issue that's been debated a lot, and I tend to think they're not hurt very much in the short-run at all. But the whole world will lose if the I-7 substantially lose their level of innovation.

Infovores  07:23

So let's focus in on America specifically for a second. They're one of the seven most innovative nations, probably they're either first or second depending on how you measure it. And yet America is a nation of immigrants. It's an extremely diverse place. When people have tried to measure the contribution to American innovation from various groups, a lot of it comes from immigrants, either directly or indirectly through spillover effects.

Is America just a huge outlier here? Or does that push against your theory at all?

Garett Jones  08:37

America has been a huge outlier in a lot of ways, right?

So in the old days for instance, a lot of progressives used to tell this story where you couldn't compare America to Europe when it comes to say, tax policy. You couldn’t say America has a smaller welfare state, and they're richer, so maybe the smaller welfare state is causing it. They just would say, America is just kind of exogenously richer. We're not going to explain why. But it just seems like it's just exogenously richer. And so America is a big outlier.

But I'll stick to a quote that I have from Barack Obama where he says something to the effect of “America is running this experiment in having an ethnically diverse democracy, and we don't really know if that's going to work.” So he treated our experiment as new and part of the reason he's right to treat that as new is because until about the 1960s, America was a multiethnic democracy on paper only right? We know that American institutions were unwelcoming to widespread political participation from non-whites right? And so to treat America as a big outlier institutionally when it basically governed itself with the electorate of Western Europe for the first two centuries, and for about 60 years now we've been making halting steps toward a true multiethnic democracy, that means that we do not have a century of multiethnic democracy success to look back on and say, “Oh yeah, this will work just fine.”

Social conflict between ethnically diverse groups, including that backlash effect, is very real and probably is putting some sand in the gears of our institutional quality.

Infovores  10:05

How well do you think that experiment is going more recently? Do you think that it was kind of inevitable that moving so far in the direction of the Barack Obama ideal of progress was going to cause institutional challenges, and now we have to back off or maybe insulate our institutions in some way from diversity or from a purely democratic process?

Garett Jones  10:37

I think when it comes to cultural questions, nothing is necessarily so. I do think that culture has a lot of focal points as we call them in game theory, where there's just multiple equilibria. And it's poorly understood how cultures choose one equilibrium rather than another, whether it's about fashion or vastly more important things like politics or economic and cultural norms. But I do think the risk of bad outcomes has grown. Part of the reason is because of this white fragility backlash, which is institution weakening, and I think that deserves real attention as a cost of America's migration policy over the last 60 years.

Infovores  11:39

Gotcha. So maybe thinking back to some of the 90s era conceptions of what defines America, how true do you think it is this idea of the “melting pot”, that maybe at some point assimilation was uniquely good in America in a way that it wasn't in other countries? Do you think that was ever true? And if so is that becoming less true over time?

Garett Jones  12:03

Well, you know, I think the the attitude persistence literature, which the late Alberto Alesina was a key figure in, shows that a lot of the melting pot stuff was never happening that much in the first place. It's just the mere fact that most Americans sort of set aside some of their detectors for noticing the difference between say Italians and Germans and English people and Swedes, the fact that a lot of those detectors faded away in cultural importance does not mean that Swedish Americans became the same as Italian Americans. They're still statistically detectable! They're still distinctly discernible from each other. So a lot of that melting pot was a metaphor that was used for European American integration, which fortunately did not lead to large levels of ethnic conflict.

Ultimately in the 19th century, the Know Nothing Party was anti-German and anti-Irish, so going after my ancestors. That sort of conflict has faded among those groups, but the melting pot we can tell from the attitude persistence literature maybe worked half-way. But not just that it worked halfway, instead I think it was a two-way street. And this is something where I really think further research is needed. A lot of what happened probably, when we say migrants have become a lot like other Americans is that maybe the Americans became a lot like the migrants.

I call this the “Spaghetti Theory” of cultural change. If you were a fancy statistician and you said, “I'm going to check and see whether Italian Americans assimilated to becoming more like other Americans” and said “here’s my statistical test—I’m going to check and see what foods they eat.” If you checked the foods Italian Americans and other Americans ate, you could make a list and say they ate pizza and hamburgers and spaghetti. And you could say, “yeah everybody eats pizza and hamburgers and spaghetti, so I guess the Italians assimilated.” But of course, part of what happens is that the Italians assimilated us. Italian Americans assimilated other Americans. So meeting in the middle is part of assimilation.

There's this belief that a lot of libertarian open borders optimists have that migrants will move to a new country and overwhelmingly move their norms in the direction of the country they moved to. And they never really consider the possibility that people who were there originally will move a bit in the direction of the migrants. But it's probable that a lot of life is a two-way street and norms are negotiated. I would love to see formal statistical tests of this, there are ways one could do it. But Spaghetti Theory is I think an important way of explaining why it is that attitudes converge about half the way in the ways we can statistically detect. But what's the other half? The other half is that a lot of Swedes move to America, Swedes are really trusting, and we became more like the Swedes, thank goodness. I'm not Swedish, but I think I get some of the benefit of their high trust attitudes because it's probably affected me other people I've known.

Infovores  15:46

Yeah, I think that makes a lot of sense. It's kind of hard to reconcile that image of assimilation moving only one way when if you spend any time in Los Angeles for example, there are a lot of billboards in Spanish and the food that people eat is very different from traditional American food whether they're originally from Latin America or native born American.

Garett Jones  16:11

Right. Migrants are assimilating us to some degree and many of these features are fantastic, right? And we should try to see if all of them are fantastic.

Infovores  16:24

Even if there is a lot of two-way assimilation going on, that still seems broadly consistent with the idea of the melting pot. Because a melting pot kind of implies that everything is melting together and meeting halfway. With the evidence you provide in your book, looking at the persistence of cultural attitudes based on where people originally migrated from in America and comparing different counties that have different amounts of various ethnic groups, it's still not really getting at… it’s hard to do a cross-country comparison and see how much more two-way assimilation is going on in America than other places. Do you have maybe at least some sense whether America was doing a lot more of that and if maybe that's why they were able to bring in a lot of people without a ton of explicit conflict relative to other places?

Garett Jones  17:03

This is a case where I don't have econometrics to back this up. But I will say that just based on the experience of many migrants to the US, they felt like America was welcoming to them. And they felt like if they assimilated to a moderate degree, they could be treated as part of the group to a moderate degree. If that's happening for half of the migrants, that's huge. And it seems as though the norms of people who are careful, non-statistical observers is that America was just better. That's right. I think it's a side effect of American openness. I think that's a lot of what it is, that Americans historically had a relatively high degree of openness.

If I wanted to point to one study that is quite consistent with that openness story, it's actually one of the unconventional “deep roots” papers. The Financial Times discussed it years ago, and it’s about Swedes who migrated to the US. It turns out that the Swedes who migrated to the US tended to have more unusual names. And so this is one of our statistical indicators of individualism and openness to try new things. The Swedes who stayed behind were different statistically than the Swedes who left, and this is one argument for why Swedes chose the welfare state—because a lot of the individualists who would tend to be high in openness left. So I don't want to say that this is conclusive proof that all migrants to the US, at least all voluntary migrants, had the same bundle of traits but it's at least suggestive.

I think that fortunately Americans have had a little bit higher openness to treating newcomers well than a lot of other countries in the world. But you know, what's important for my story is not just this diversity question. I see that as a smaller thing. And I emphasize that the conflict created by diversity is a smaller thing. I think the more important question is, if a lot of folks moved to your country who support bad policies and if those attitudes are relatively persistent, then things are probably going to get worse. And this is why I kicked off chapter one with a story about Argentina.

Argentine migrants, according to mainstream histories of Argentina, were disproportionately folks who were interested in strong trade unions, socialism, political radicalism, and anarchism—which was a left-wing movement at that time. And within a couple of decades, you had first and second generation migrants who were pushing hard for progressive politics. Not just mild welfare state stuff, but full-on direct action. And this is part of what landed us in the end with “Don't Cry For Me, Argentina.” It's a complicated history, I'm just giving a potted version of it. But mainstream Argentine history says they imported attitudes by bringing in new workers who carried those attitudes with them. Traditional progressive historians are proud of that. I think those of us who care about institutional quality and prosperity can learn from those lessons, even if we draw different policy lessons than they did.

Infovores  20:51

Do you think there are any meaningful differences between legal and illegal immigration with respect to assimilation and the attitudes that migrants bring with them?

Garett Jones  21:03

If I wanted to just speculate without having tested it, it would be that illegal immigrants are probably more likely to want to assimilate. Because your goal is to stay below the radar so that you're not likely to draw the attention of authorities. So that would be my uninformed hunch, without having econometric data that I can put to this. I would suspect that illegal immigrants are probably more interested in assimilating… on the condition that they are planning to stick around, conditional on expected long-term duration. If some folks are overstaying their visas or coming over to work for a few years to make money and then go back, I'd say that one would expect those folks to be less interested in assimilation. And maybe they would teach their kids to be less interested because they'd be thinking, I'm gonna raise my kids back home and I want to teach my kids in the cultural norms of where they're gonna live. So I think the bigger distinction is long-term versus short-term. But because people are longer term there, they're probably going to push in that direction.

What I'm really interested in is the second and third generation because I think of it as a book about the long-run, not about the short-run. And so if we're thinking about long-run immigration, I would expect undocumented, or as they say in Europe, irregular migrants to be more interested in assimilating.

On 10% Less Democracy

Infovores  22:41

I'd like to ask a few questions that more explicitly tie in this discussion to your previous books.

On balance, who is more right about immigration, voters or elites?

Garett Jones  23:00

Oh, that's a great question.

I think of our political elites as mostly trying to imitate, or trying to read the views of what they think the voters really want. And I think that voters hold very confused views on immigration. People on the right are happy to show you polls saying people want less migration across many rich countries, and especially in the US where I see more data. But it also appears strongly that those same voters want lower immigration only as long as there are no bad headlines that will make them feel uncomfortable in any way. So I think politicians have to weigh not just what voters say they want, but whether voters will want the consequences of the thing they say they want.

So you can find many voters saying that Americans should tighten the border and have lower rates of immigration. But are you actually in favor of having a dramatically militarized border that hurts a lot of people? Are you really in favor of deporting large numbers of people who've been here five, ten, twenty years? And all of a sudden voters are like, “No, no, I didn't mean that.” So politicians have to be aggregators, and in that respect they have to think through the consequences of what will happen if they give the voters what they say they want.

Now on the broader question of non-political elites, I do think that these elites overall are too optimistic about the costs of ethnic diversity because they're living in bubbles where they can screen out the costs and only get the benefits. They're living in affluent neighborhoods, where poor and lower skilled first generation migrants have to drive half an hour, forty-five minutes, maybe more than an hour just to get there. So they're getting all of what they think of as the benefits of diversity, which is like a lot of ethnic food and people to take care of their homes and engage in what they think of as menial low skilled labor, which is often quite grueling. And they're not paying the costs. That's over in somebody else's neighborhood. Those are somebody else's school districts. I don't have to think about that. So I think that's a big problem.

There's a salience element of the costs of ethnic diversity that even if they're not overwhelming as I emphasize in the book, there's something still that elites do not see. They live in a filtered world, and so they screen out the costs to get the benefits and that makes their views unfortunately biased. I think they would do better to actually read the academic literature.

Infovores  25:54

I think the core question I was kind of wrestling back and forth with in my mind is, how would “10% Less Democracy” affect these issues?

It seems like on the one hand, a lot of elites who would presumably have more power in a 10% less democracy society are kind of out of touch from some of these downsides of immigration that they don't internalize. But on the other hand, if you insulate your well-functioning institutions from the will of the people, then that could potentially reduce a lot of the biggest costs that you point to in your book where the populist attitudes of the people, the anti-institutional attitudes, are going to have less influence.

Garett Jones  26:49

So I talked a little about this in 10%. The European Union had this major migration crisis in 2015, some of which involved refugees. And the Western European saw for the first time just large numbers of people coming across, often in boats, and often large numbers walking through mountain passes to get to countries where they can claim asylum. And for the first couple of months, treating the migrants well polled very well with the voters. And then after about six months it started polling really negatively with the average European voter. So if I thought that the out of touch elites could just ignore what the masses wanted, I would expect that migration would just continue. But once the once the European politicians saw what voters wanted, they actually created an effective border control agency that reduced irregular flows of migrants by 90%. And it stayed that way for years.

They created Frontex, which had been sort of in an embryonic state beforehand, and they snapped it up dramatically. EU leaders started holding press conferences basically celebrating how effectively they were protecting the border. I tend to think that European elites created a border control policy that Donald Trump could have only dreamed of. But because European elites speak in the idiom of Brussels, in this sort of neutral technocratic way that doesn't have the rage of a Donald Trump they accomplished what Donald Trump wanted without actually…

Infovores 27:45

Without the friction.

Garett Jones 27:50

Exactly. So the populist parties of course hold many very bad views, but they got many things that were on their wish list. And they won't celebrate it because the people who accomplished their goals didn't talk the way they wanted them to on Twitter.

So European elites both respond to voters and are effective at giving the voters what they want. So this is something where I think I am one of the few people in the world interested in talking about this to people because the right doesn't want to give Brussels credit for implementing largely a right-wing policy. And the left doesn't want to say “Oh, we did something really right-wing, look at how good we are at this.”

Infovores  29:21

So mood affiliation aside, you basically have a situation where if you have too much democracy, too much populism, then you can't even effectively implement the parts of the immigration skeptical worldview that is correct. And by giving just a little bit less power to the people, you're allowing technocrats to be just a little bit more effective at doing the things that actually kind of make sense on the margin.

Garett Jones  29:54

I think that's a fair summary as I understood it, yes. The technocrats can implement things without giving the voters the mood affiliation they crave. You could say that's the downside of technocracy, right?

Voters crave mood affiliation. They crave seeing the memes that they love. But maybe they should just take their 90% policy win and move on to the next issue to get mad about.

On Hive Mind

Infovores  30:26

So your other book in the Singapore trilogy is The Hive Mind. And there you emphasize national IQ as distinct from the impact of culture. How are the two things related in your view?

Garett Jones  30:38

I would actually tweak that a little bit. I tend to think of my book as describing what a high IQ culture looks like or what a culture with high cognitive test scores tends to look like.

Intelligence tests tend to predict individual levels of trustingness and trustworthiness. And they tend to predict higher levels of group cooperation in an important cooperation game called the repeated prisoner’s dilemma. And when you think about what high trust cultures are like, they're distinct from lower trust cultures. And I think it's not that IQ is the only thing that is an important causal factor here, but it certainly predicts it and it certainly has a higher effect than many other traits people have tried to look at. So I tend to think both books are in a way doing what economists call looking under the lamppost for your keys, even if your keys are somewhere else, because the light is better under the lamppost.

So Hive Mind is my short to medium run book and Culture Transplant is my medium to long run book. But while you can say a lot from economic theory and experimental research about where IQ takes you, it's a lot harder to say where IQ comes from— there just hasn't been enough research that I find to be high enough quality to be confident about that. So I can be pretty confident about forward causation, a little iffy about how much reverse causation is going on.

This is kind of the way monetary policy works, right? I can say don't do this monetary policy and do that monetary policy, but if you asked me why some countries have different monetary policies than others, then I don't know. I just want to tell you what to do, I don't know whether you're gonna listen to me or not. So I tend to think that these cultural traits are certainly going to correlate with IQ to some degree—it's going to be far, far, far from perfect—but what I can say about culture is that because this intergenerational persistence has been so well documented, I can say something that's much more relevant to migration policy. Hive Mind isn't built for that, it's just not a book that lets me address those questions in a credible way.

So Hive Mind is more about forward causation and the precise channels of forward causation. Why is it that higher IQ people are more patient or better at saving or better at cooperation? The Culture Transplant is a book that says, if you want to get trusting people I know how to do it: bring in a lot of people from Sweden.

Infovores  33:19

Bryan Caplan had a big review of your book, and one thing that made him really upset is basically that you didn't just throw IQ into the regressions. And my reading of your response to that was “well, we can't just throw IQ into the regression, because quite plausibly culture is actually causing IQ to develop in the long run.” Is that an accurate assessment?

Garett Jones  33:44

Yes and that is something that even his own comic book would suggest, right?

So if you think of culture as not quite an ultimate cause, but a long-run cause of prosperity and if you think that IQ is one of multiple medium-run causes of prosperity, one of the rules of good statistics is that you don't control for medium-run causes if you're trying to learn about long-run causes. If you're trying to understand what makes a car go faster, you could look at how powerful the engine is. And then somebody else could say, well why would you want to look at how powerful the engine is when you can just look at how fast the wheels turn? If you look at how fast the wheels turn, that'll tell you how fast the car goes. But it would be a statistical mistake to control for wheel speed if you're trying to find out whether an engine is important for making the car go faster.

So I think that's part of why statistically it's a mistake to control for IQ when you're trying to find out whether these deep roots or other long-run cultural factors matter. In multivariate regressions, you don't control for mediators. That's the jargony version of what I just said. Controlling for mediators is a mistake, and you shouldn't make that mistake. So it's unfortunate to see Caplan making a well known mistake in that essay, but he can always correct it I guess. You don't control for mediators.

Infovores  35:21

Gotcha. And so, if culture is its own thing, distinct from genetics in some sense, how much do you think adoptive culture should matter? So for instance, in New Zealand many more people identify as Maori than is likely to actually be the case biologically. But maybe identifying with an ethnicity is important and meaningful in and of itself?

Garett Jones  35:53

That's a great question that is really outside the scope of my book. And I'm adopted myself. My parents adopted me at six weeks, and I know that changed my life in many ways.

The thing I was just thinking about the other day, one of the things I'm most sure of about how adoption affected me is that it's the reason I drive pickup trucks. My dad was a pipefitter, an industrial construction worker. And he always drove real big pickup trucks. And so I drive a pickup truck. I'm sure that as a college professor, there's no way I would have done that.

So this question of how much adoption matters and how much that long run experience of being put into a new environment matters is a great question. The normal result from behavioral genetics, as you know, from adoption and twin studies is that on most things adoption doesn't matter much in the long run. Being adopted into a smart family doesn't seem to make you that much smarter in the long run. But one place it does matter is education. It turns out that the effect of adoption on your education level, being being adopted into a high education versus a low education family has a moderate correlation. Usually I think people say .3, with your adult level of education.

And we know that culture gets transmitted through education. Most famously, the best evidence is that cultural attitudes—not political attitudes, but cultural attitudes—towards social tolerance for instance seem to be transmitted through education. The one thing that it seems you can say is just that college makes students more left-wing, not on economics, but probably does make them more left-wing in terms of social tolerance and cosmopolitanism, which is a form of adoption effect I totally welcome.

So you can call those soft effects. I mean, if one wanted to call it a soft effect, I'd say okay, fine. But a lot of those things can add up and matter for a lot of people over time. Again, some of this is outside of my area of expertise, but modern behavioral genetics as summed up by Caplan's very good book, “Selfish Reasons to Have More Kids” is a good place to learn about how being adopted into a certain family affects a few things, but not as many as you think.

Infovores  38:19

So it's interesting you mentioned the adoptive effect of education, the nurturing effect, because I actually thought about that in relation to one of the most interesting facts you mentioned in passing in your book. You note that apparently, at least according to the one study that really closely looked at this, mothers pass on much more of their cultural heritage than fathers do. And I believe that's also true in the education literature, that mothers’ education is much more closely related to child education than that of fathers. It seems like both of those things suggest that there's some sort of not exactly genetic effect there.

Garett Jones  39:01

Exactly. There are things that we think of as the softer side of life that might be transmitted, especially through our mothers, at least for the average person, right? And do those things matter a lot? Without further information, it's hard to say what what counts as a lot. But that one study that looked at the effects on trust of your mother versus your father's level of trust, or the ancestral level of trust, that was really one of the striking things where I hope an enterprising grad student goes and does like five studies of this in seven different countries because it's not something where you need new data. The data are already sitting there in these datasets and someone who did this, who just pressed a button and did it for 20 countries and made that a project, that would certainly be publishable in a good journal.

Infovores  39:58

So maybe to try and punch holes in this just a little bit, I think one thing I've heard economists say is that of course mothers have more influence than fathers just because the father is much more likely to not be around at all. Is that the same thing that’s going on here in this other study? Or do they have pretty uniform involvement from both parents?

Garett Jones  40:24

No, this is something where there's no detail that lets you know this. These folks are very likely using World Value Survey measures and something like the General Social Survey, perhaps Eurobarometer, something like that. And so the detail level just probably isn't going to be high enough for this. I mean, I'm glad to be wrong. The US GSS has a lot of questions, but then you're still just relying on survey responses, which are always a noisy indicator.

It is worth noting though that if you're using noisy indicators and you get a small effect, that's a sign that the true effect is probably bigger, not smaller. A lot of people think that if your measure is noisy and you find something it's just a fluke. It's not impossible, but best guess is that if you get a small effect with a crummy statistical proxy, you'd get a bigger effect with a better one. That's known as the errors-in-variables finding.

Infovores  41:17

It seems like that might be another reason why you would never ever want to control for IQ against these cultural measures, because IQ is just measured much more precisely than a lot of cultural things that we care about.

Garett Jones  41:32

That's right yeah. It's not as important as the mediator channel that I suggested before, but you're right. When you have a choice between the thing that's measured better and the thing that's measured worse, the thing that's measured better is just going to win naturally, even if it's intrinsically less important. Even if you knew hypothetically that it was causally less important.

We're always using mediocre proxies. As a monetary economist by training, that's something you know with money, right? All of our measures of money are terrible, and somehow we can still say something useful.

Infovores  42:06

I was thinking about this when you find that there are very strong positive effects of bringing in immigrants with Chinese heritage. And you find actually negative effects for Protestantism, despite a lot of things in the more general sociological history…

Garett Jones 42:45

Max Weber, right?

Infovores

Yeah I just wonder whether maybe the Chinese indicator is just a much better measure of culture. And if you had a really good way to distinguish between like, “true Protestantism” as opposed to a lot of different things you might see something different. I get the sense that there's a lot more cross-migration throughout all of the Protestant countries that kind of mixes things in a way that makes it hard to really identify what Protestantism might be doing.

Garett Jones  43:02

I'm open to that possibility. It might just be that the particular study that I know of that I mentioned that discusses this is looking at 1962-1996. At that point, I don't think updating the numbers have changed that noticeably. But in that period Protestantism was a different thing than it was when Max Weber was writing, and certainly Protestantism was a different thing than the period he was writing about. He's thinking about the 1600s and 1700s. So what we call Protestantism is a different creature than it was when Weber wrote and in turn it's different than the day he was writing about.

The Garett Jones Cultural Production Function

Infovores  43:48

Make sense, yeah. So I'd like to close with just one or two more questions.

You alluded to this previously a little bit, but what did growing up in the LDS church teach you about the importance and persistence of culture?

Garett Jones  44:01

Oh, yeah. So my Mormon ancestors came to Utah, and at some point Brigham Young told most of them to go to Wyoming actually. So my ancestors are from Wyoming.

And you know Mormons brought a lot of English Northeastern culture to what we now call Utah. So when one is doing genetic studies and wants to learn what founding era Americans were like, one would want to actually do some studies in Utah. Because a lot of English folks moved there, Joseph Smith and all his family were English.

Infovores  44:44

I believe Utah is still the most English state in the nation.

Garett Jones  44:49

I'm not certain of that, but I'm open to that because I know it's between them and basically the far northeast—Maine, upper New England.

So you've got the English, you've got the Swedes, you've got the Welsh, which is where I come in as a Jones. And it does really seem like they imported the cultures of their homelands to a sizable extent out to Utah. So you know, I would not be surprised if one could statistically distinguish fifth generation English Mormons from fifth generation Swedish Mormons statistically. But at the same time Mormon culture seems to be its own independent force.

Though it has changed a lot. I can use Mormonism to make both points because it captures this diversity of cultural persistence and change. Mormons used to be a theocratic, polygamous, communist cult essentially right in the mountains. And they became pro-democracy, pro-monogamy, and pro-capitalism. I mean all it took was one invasion by the United States government, and they just completely changed their story, right? So that's a reminder that a culture can change in what all of us would think are important ways, and yet still keep other attitudes, other traits of cultural persistence that are there for anyone to see. So it does capture, as I've mentioned to someone on Twitter a couple of months ago, how I could have called my book “The 40% Culture Transplant” and that would have captured a lot of my story. There's a lot of both, and when it comes to persistence and change my Mormon ancestors are obviously carrying the traits of their original countries and then they're able through the repeated environment of an LDS ward, an LDS congregation, by meeting so regularly and spending a few hours a week with the same folks, week after week and year after year, they're conveying the culture of Mormonism and transmitting it kind of persistently from generation to generation.

It's a really effective culture. I think Mormons have created something fantastic that the world should study more. I was lucky to work for an LDS politician, Orrin Hatch. And they are a sign of how culture can be transplanted, and they give evidence that some of this really is just preaching norms, implementing them, and practicing them on a regular basis.

Infovores  47:31

Don't Mormons also exemplify this two-way street of assimilation because they really actively proselytize? And so you have this persistence on the one hand because there are a lot of selection mechanisms that filter who is going to join the church and who's going to stay in the church and that creates persistence, but at the same time you do have a lot of new people flowing in all the time, and that's naturally going to change the culture of the church.

Garett Jones  47:59

Yeah so the economics of religion literature, has thought through this a lot, and if you're going to be a religion that is successful according to Stark and Iannaccone, they say you have to stay in medium tension with the society around you. So the Mormons have had to learn to adapt all the time. And because a lot of it is top down, the leaders have to keep an eye out on cultural changes that they don't want to get too far away from right there. I mean, in a way their goal is always to stay about a generation and a half behind the times probably. And so the leaders are steering to the extent they can the culture of the faith, so that it doesn't get too out of touch with the surrounding society.

On the question of whether their acculturation is changing, the people who joined that's tougher, right? I mean obviously in outward ways, they're going to change. But there's so much selection going on, right? This is one of the standard Scott Alexander points, which is that, in so many areas selection must be driving everything. So the people who are joining Mormonism are the people who think, yeah, I can hack it. I can give up smoking, I can give 10% of my money to these folks, I can go to church every Sunday, right? Mormons lose the vast majority of their converts within a year. And so that's all got to be selection right there, 80% just selection. So basically having a tough filter is one of the ways to make sure that you get people who are going to stick to your cultural norms. So I hope that we encourage the Mormons to keep their tough cultural filter so they can continue to have their fantastic cultural contribution to the world.

Infovores  49:48

Looking at George Mason Economics more generally, quite a few of the professors have strong ties to a religious tradition of some kind. Usually they're not practicing themselves, but within one or two generations there's a pretty clear, significant link there for some of the most influential people that come to mind. What influence do you think that has on department culture and the ideas that come out of it?

Garett Jones  50:14

I do think it means that we want to talk to people. It means we want to share our message with folks. I think George Mason University economists have a proselytizing, even prophetic bent that a lot of academic economists lack. A lot of academic economists are happy to toil away at their journal articles, get them published, and get accolades from their elite peers. And those are great things to do—I've been happy to have some success there myself.

But what GMU professors want to do, and it may be related to their ancestral exposure to strong religious traditions, is that we really want to explain things to others, both in and outside the profession, and bring them around to our views. So the need to persuade, the need to sort of sincerely have people look at your stuff and say I'm actually going to come around to your point of view, that seems to be a passion that many of us share.

And I wish more academic economists had it, because academic econ has so much to bring to the world. And I hope we can find ways to get more of the great ideas of economics, both those that are GMU friendly and those that are not, shared with the world as a whole.

Infovores  51:34

I wonder if that's related at all to the number of Jewish Nobel Prize winners there are. Everyone fixates on high average IQ, but it seems to me like a really important factor in winning a Nobel Prize is just really pounding home again and again a really important idea and doing everything you possibly can to thoroughly convince the entire community of it. Is that related at all? Or do you not think so?

Garett Jones  52:09

This is something that I don't know enough about to say, because I don't think of…

Infovores  52:16

You don't think of it as the same evangelism.

Garett Jones  52:18

Yeah, it's not quite the same evangelism. And I actually don't think of 20th century Judaism as having this strong prophetic bent as much as Judaism inheriting a historic prophetic bent. So that's something where I'm open to learning more about that, but I probably have to say I don't know if that's true at all.

Infovores  52:45

But of course if you have really deep roots, and they're well preserved, it could be that you're tapping into something that goes back to a time when Judaism was actually maybe more evangelical in some ways.

Garett Jones  52:59

It’s a prophetic religion, right? So I mean, I think that the book of Isaiah still deserves to rise in status. It’s just a fantastic book, and not just because it has so much of an influence on Joseph Smith but because it has great messages of what a moral society looks like and what morality has to wrestle with.

You know, every culture that seriously wrestles with the prophetic message of the Old Testament I think ends up coming away the better for it. I wish we were in a culture where a greater portion of the American population were being exposed to the prophetic voices, especially the Old Testament. Because it would elevate our national discourse in a really important way.

Infovores  53:41

Garett Jones, thank you so much for joining me. It was a wonderful opportunity to meet you.

Garett Jones  53:49

Again thanks for having me.

If you liked this post, check out my other writing at Infovores Newsletter. Over there I write reviews, conduct interviews, and generally try my best to navigate a world of infinite knowledge.

Time Well Spent is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

Get full access to NeoNarrative at www.neonarrative.us/subscribe

After being knocked out of commission for several months by covid we’re building back better or whatever like never before!

Get full access to NeoNarrative at www.neonarrative.us/subscribe

Daniel Ek is the cofounder and CEO of Spotify, one of the most dominant music streaming and music discovery platforms in the world today, with 400 million active users worldwide. Daniel was named the most powerful person in the music industry in 2017 by billboard, which I think is a pretty apt description. Previously Daniel held a senior position at Tradera (acquired by Ebay), was the founder of Advertigo (acquired by TradeDoubler), and was CEO of µTorrent (acquired by BitTorrent). Daniel is singularly responsible for saving the music industry and anyone who has a different opinion here is simply wrong, and more importantly he’s also just an overall great guy and this becomes immediately clear the second you hear him express his ideas. To keep up with Daniel, follow him on Twitter here. 

(Edited for clarity) The structure for today will be a traditional rapid fire back and forth Q&A. The questions will come fast, and I'll jump into the first one now: Spotify saved the music industry (which lost half of its revenue between the turn of the century and 2015) by making on demand music streaming more persuasive to listeners than piracy, and from this fact alone it's easy to decide that the future of music has finally arrived. But what might music distribution and consumption look like in 100 or 1000 years? And how does Spotify see its role in shaping that future with tools like discovery? There has really never been more potential for user centric innovation in music than right now. And my first instinct is to imagine that Spotify’s slogan, the soundtrack to your life, may become increasingly literal. Will algorithms collate user data and curate music according to life events? Will Spotify integrate with one of the many excellent facial recognition platforms today and become a melodic mood ring? What data does Spotify have now that may tell us what listeners in 2121 might want?

I don’t think anyone has ever asked me what things might look like in 100 or 1000 years. I think the longest time-frame I’ve been asked about has probably been only a decade or two, but that's a fun and engaging topic to talk about. To start I want to first think about the basics of music. It's something that's been present with us really throughout the entire existence of humanity, right? Music used to be how we shared knowledge back in the day; our elders and tribes-folk used to share wisdom and stories through music. And so I’d like to believe that in the next 100 or 1000 years music will still be a very important part of our society and how we express culture and belonging. But as you said, I think the real fascinating thing is that music is becoming more and more omnipresent. It's become easier than ever for today’s creators to express themselves, and it's becoming easier than ever for anyone in the world to enjoy and be inspired by that. 

I think there are two ways things may change over the next 100 or 1,000 years, both of which are sort of the most fascinating things I can imagine. The first has to do with the fact that the bar of creation will continue to drop lower. A few hundred years back, if you wanted to write a symphony you basically had to be a genius. You had to hear all the notes in your head of every instrument, and if you wanted a live sample of your ideas you’d of course be restricted to playing one instrument at a time despite needing to hear them all in concert. And the greatest of that era never heard their music being played in full up until the very moments prior to a performance at an opera, there’d be only a few rehearsals before a concert event and so time also limited the ability to actually make changes to the work. The sequence of music creation then was from 1) hearing an entire piece in your head (which is pretty insane), 2) then writing it down on a piece of paper and hardly ever hearing it, 3) and then performing.

Fast forward to the 1950s, or even a bit earlier with jazz in the 20s and 30s, where it was the age of instrumentalism, a time similar to the classical era where you basically had to be a virtuoso at the instrument you were playing in order to express yourself (except with the added benefit that you could more freely play with other people and freelance and freebase a musical response). And then again if you fast forward and contrast all of this with today, you can see that many of the greatest musicians that we have now don't know how to play a single instrument. But what they do know is that they're almost like computer programmers, they're expressing their vision with software and are able to bend sound and bend our reality.

And so I like to believe that with respect to overall creativity, in the next 100 years it will be even easier to express yourself than ever before. The barrier to be creative is just going to become lower so that more people can become creatives or musicians, but that’s not to say that the bar isn’t still high. Of course you have to learn the software, you have to learn all the plugins, the notes you have to use, composing, and stretching, and beat. And that's something that takes a lot of time for people to learn today even though there's a lot of knowledge on how to do it. And I think that's a huge barrier for people creating today. But I like to believe that even that's going to get lower. 

The second aspect to the changes I imagine involve just the consumption side of music itself. To start I like to describe where music consumption is now. It's quite obvious to avid music consumers that you can create a better playlist yourself than could an algorithm. If you really wanted to sit down to create that perfect playlist for your day, you'd probably be able to do a pretty good job by yourself doing it. Spotify will do an okay job at it, but it's pretty clear that you, with just 20-30 minutes of work, can do a much better job. For now Spotify is like your friend who knows music in general, but we don’t understand you that well. In the next decade or so Spotify will become that friend that really knows you and also knows music. You might still be able to do a slightly better job yourself in expressing what you want in that moment, but given more time I'm not so sure you will be able to do a better job than we can. It's very possible that Spotify will be able to create a much better Soundtrack To Your Life than you could potentially even imagine doing yourself. And this is kind of a crazy thought for people.

Next I think a good comparative example for potential changes I see in music is video conferencing. I love going back in time to 10 or 15 years ago when we were first talking about remote conferencing, and I remember using an early service which was insanely expensive and difficult to use. You had to have a separate ISDN connection, you had to put in the IP address of whoever you were calling, and they had to have a similar, very expensive, very difficult system. It wasn't a great experience. And so people then said, well, it's pretty clear that physical meetings that humans organize are always going to be superior to digital ones, but now in 2021 we can have these video conferences on whatever equipment we want, wherever we are in the world, and communicate easily.

I don't have to know your IP address, you don't have to know mine, we're just communicating and we can interact with 50 or 100+ people at the same time. But we still think somehow that meeting virtually can’t be as good as meeting physically, but I actually believe that the virtual meeting might even supersede the physical meeting in the future and we will have even better experiences virtually than we are having physically, which is kind of insane to think. But if you believe this is all possible, then it's not a stretch to imagine that future musicians will create songs that can do things like alter based on who you are.

So there'd be a song and the general structure of it, but it would adjust and adapt based on you to become something perfect. I had this thesis for some of our folks that are working on better music software that went, “What if we could create a song that always updates itself and tries to become the number one song all the time, a song that just constantly updates?” And even if it sounds pretty crazy, I do think that something like that will be possible and transformative sometime in the future, I just have no idea when and where. And I think we'll have some amazing musicians and artists and songwriters that will imagine-up some crazy way to express that in the future.

I'm going to jump straight into the next question and I'm really excited to hear your thoughts since this is one everyone is more or less anxious to know: 

Spotify has never been more essential for stress management, personal growth, work motivation and leisure than since the beginning of the pandemic. We've all had to find some way to maintain our bearings as the world hit turbulent and uncharted waters, and some of us, separated from friends and family, had to go it all alone. But Spotify led the way back to each other and back to ourselves through music sharing and discovery, seeing a whopping 50% increase in TV and game console usage in 2020 alone. With Spotify becoming something like a center of gravity in everyday life, and with the new shipment of Greenroom, Spotify’s live streaming platform, is Spotify’s emergence as a full scale social media platform in the works?

Well, we're certainly trying to become the soundtrack for every moment, whatever that may be. And that obviously means that we're trying to make it a lot easier for people to consume music and audio in more places and in more ways that fit their lives. I don't believe we're trying to become a social media platform, but in the sense that a social media platform is just a means to have lots of users interacting with each other, and that being the center of the experience, what we are currently trying to do is allow music artists and consumers to interact in more ways than we have in the past.

So when I think about Greenroom, I think about the ability for you, as a creator or a host, not just to host the room and invite people but actually moderate and steer the event or conversation. And that's a very different social experience than what you'll have on Facebook or WhatsApp or any of these other social platforms that exist. Our view is we want to make it so that artists and creators have more ways to express themselves, more ways to interact with fans, and ultimately more ways of monetizing their art so that they can sustain themselves as artists. And that medium of interaction between artists and their fan base is something that Spotify hasn't done up until now, but it's certainly one of the areas we want to develop and offer since we're now at the point where we have about 400 million people on the platform. So there's a huge audience there, and we just want to give artists more tools to interact with their fan base.

So is Spotify considering adding a like button and a comment section? And will these features be worth transforming Spotify into a platform which is no longer opinion free?

I think we'll add something like it. But it won't be what most people probably imagine when they think of a like button or a comment button. So again, the most important thing for us is that we put creators in control. One of the big things we started doing just a few months ago, and you can see it already in the product, is that if you're a podcaster, you can actually use some of these features already. You can have polls where users respond to some of the content you're putting out and ask questions, and you can also, as a creator, choose whether you want to share what the answers were from the audience. There are also other things like surveys and so on that we're working on as well as ways for creators to express themselves, but also get instruction from the fan base.

But I personally think that many of the things that we talked about, like buttons, comments, etc, are things that are very much old social media norms. And I think the newer social platforms like TikTok and others are innovating past that in a new way. They still have the same type of functionality, but it doesn't work in the same way as many people perceive something like social media should work. And we are sort of the same way, we're definitely creator-control centric rather than a usual social media network.

I want to go back for a second and touch on something you mentioned before about online space approaching a point where meeting a person over your phone can supersede the quality which you might experience physically. And my question is: Tech innovation is no longer confined to physical places because of the mass adoption of things like zoom, which exploded ever since COVID arrived on the scene, so how can we make Europe a place of innovation?

This is a matter that's near and dear to my heart. I think the good news is that in some ways, it’s already happening. When I started as an entrepreneur, the reality was that many of us didn't know how things like venture capital worked. There was no norm about how to work with venture capital, what a Series A round was, what a series B round was, how to pitch a venture capitalist, none of that stuff existed.

The good news is that, today, if you're a budding entrepreneur, wherever you are, you practically can find all of that stuff online in five minutes, how to do all of it. You can even reach out directly on Twitter or LinkedIn to venture capitalists and get a meeting set up, and there are many VCs now in Europe, there are many all over the world that are happy to look at pitches, there are even angel investors and other helpful people in the business community, a lot of these things just didn't exist when I was starting. And that's led to a lot of growth.

So I think when you look at entrepreneurship, the crazy stat is this: You’re right to assume that Europe is behind the US. But what's fascinating to me is even when you look at the US, and if you take Silicon Valley and Seattle out of the equation, the US would actually do worse than Europe. So really, up until now, innovation has been confined mostly to two places, Silicon Valley and Seattle. All the tech innovation that you think about, all of the Apples and Googles and Facebooks and Microsofts and so on have been in a very confined, small space. I think that barrier has been broken now. You're starting to see more companies pop up all over the place. You have Snap in LA, you have companies in New York, you have Klarna and Spotify in Stockholm, you have UI Path out of Romania. There are companies all over now that are breaking ground and serving as inspiration for those spread-out clusters of people. So I think to some extent it's already happening. But I think the primary way for it to continue to happen is just the spread of knowledge that's happening, with podcasts being a prime example of this, It’s so amazing how much knowledge is available to you just by searching for it, you can now find more about almost any topic that you want.

And so many people are paying it forward and sharing their experience and expertise and I wish I had the same amount of resources when I started out. So I think that means that we're going to see more innovation spread out in more places now. Of course the rest of the ecosystem needs to get there too, with venture capital and so on, and I'm hoping that that will catch up eventually.

Returning back to music and music’s future for a second, my next question for you here is: Most developments in the state of any art emerge from combinations of popular culture and niche cultural forms that are hardly ever considered, with the history of rock being a great example. Rhythm and Blues eventually met with country and folk to effect the huge change of combining what had initially been exclusively black and white forms. And now something like the syncretism singularity may be on the verge of happening with all the world's music being made available everywhere by Spotify. Is Spotify seeing anything like a blending of music across cultures? is Swedish music becoming more South African? Is South African music becoming more French? Is everyone becoming more American? That's sort of what I'm wondering about.

This is something so fascinating to see, because, if you look at the history of pop music, it's very much been the case that American and British artists have ruled the world. But what we're seeing for the first time now is a sort of new world order (if you'd like) where you have BTS from Korea, making it big, you have several Latin acts. I mean, right now as an example, one of the popular songs that my kid loves is the song My Universe, which is Coldplay and BTS. They're singing along to Korean lyrics and loving it as one of these crossovers.

So I think you can see that already. And it's happening in the ecosystems where streaming has matured. I don't think it’s a coincidence that South Korean and Latin music are making it since both are heavily streamed areas. For me, I'm more excited about what's next. So if you think about the sort of next regions, you have India, and Bollywood music I'm sure will have a big moment on the global stage in the next few years, and then past that, I'm super excited about African music, where we're still much more nascent, but West African beats are already influencing a lot of hip hop, Kanye West and Drake, amongst many others. But I think that's just the beginning. Because right now, no one in the western world understands that that's actually very much inspired from Africa.

So I think you're going to see for sure in the next decade a lot more African songwriters and artists also making it on the global stage and more collaborations between artists across the world. And what's cool for me is that a lot of them are writing these songs not even together, they are literally on opposite sides of the world speaking via technology and FaceTiming some of their work and then recording it in their own studio and then sending over their files. And I believe in the future, you might have a live recording across two rooms, working in real time on a single song. Right now it's kind of a tedious process where if I do the vocals on your song, I have to record it separately, send you the file, you have to incorporate it into mixing, you then have to export it out, which is a tedious process, and then send it back to me, and we have to bounce it back and forth like that. But obviously in the future it should work like Google Docs, we should be able to do it at the same time, somehow, and Spotify actually has a product called Soundtrap, where we're trying to do exactly that.

My next question is probably the most important for young people today, and it's that in my interview with Andreessen Horowitz co founder Ben Horowitz, I asked whether or not college is still necessary to achieve the American dream, considering that technology now makes it possible to access the same learning material and communities that have mostly been confined to campuses for the last few decades, and his answer was that unless you want to be at the leading edge of a field like physics, the value of college may not be worth the current cost. Is that sentiment true in Europe in general, and Sweden in particular? I mean, one thing which really makes you so interesting to me is that you left college early as have a lot of the most brilliant founders and artists in the world today, and Spotify makes it possible for millions of people to do the same thing and make careers doing what they love without any formal certification hurdles. Will the internet end up eating the university?

I think the value of education, obviously, is immensely important. But there are many ways to educate yourself, as you rightly pointed out, and I do think education needs a fundamental reform in order to keep up with where the world's going, both the wealth of knowledge that's now available online and the ability for us to realize that as individuals we all learn in different ways.

Some of us may prefer just to hear the content, some of us want the visual interaction, some of us actually want to write it down, some of us want to teach it to another person in order to understand it. And so my view is we have to do a fundamental educational reform, because the reality is, going through a formal education system just once in our lifetime probably doesn't make sense anymore, we have to continuously educate ourselves.

With that said, I do think when you're dealing with a base level of knowledge and a breadth of knowledge, something like a college or a university still makes a ton of sense. It's not for the faint-hearted to drop out of college and have nothing to lean back on. And it's not something, when people ask me, that I would highly recommend unless you're absolutely certain what you want to accomplish in life. So actually, for those that aren't sure what they want to do, having an education to fall back on could be very important and valuable.

But It's ironic when we talk about the American dream because there are two things that are fundamental for it to happen which have gotten insane, the first is that education has gotten so expensive, and the second is that healthcare has gone the same way, while almost everything else in society has gotten cheaper, like electronics, food, all of that stuff. Both of them, for me, are the foundational aspects of an equal society and equal opportunity. So I think it's essential that we, as a society, and broadly speaking, all societies, spend a lot more effort and resources both on reforming health care, but in particular, reforming education too so that more people can participate in it and have success.

And I think many of the founders that you're talking about, including myself, we just always knew what we wanted to do, and it just so happened that that was the bleeding edge anyway and wasn't being taught in schools, so there was no knowledge to be found there. So we ended up going out in the world and gathering it in a different way. 

My next question is that I always wonder how extremely successful founders managed to do what they do, and then also have kids. So I'm wondering, how do you do it all?

I think most successful people I know are just very good at managing their time. And it really starts with that. It's not that we have more hours in the day, even though it might seem so sometimes. What I've observed is that, and certainly in my case, I start with a very different principle in mind when I allocate my time, which is, most people I've found allocate their time on a first come first serve basis. They basically try to fill up their day with as much as they potentially can, and whoever comes first gets the time slot, and then they just keep building on that.

I actually start off with a pretty much blank slate, and then ask myself what I want to accomplish during the day. And then I start there and say no to everything else. So sometimes my days, surprisingly, have almost nothing on them. And sometimes my days are very full. But it's not unusual that they are almost blank.

And obviously of course when speaking about family and kids, this is something that I prioritize very, very high up on my list. I make sure that comes number one, whether it's putting my kids to bed, like I just did, or showing up at a recital or something else, that has the highest priority. Unless the house is burning down I'm just simply going to prioritize that above and beyond everything else. I think it starts there and most people haven't thought out what their clear priorities are.

And as I said, they already start with a calendar that's pretty full, and then it's very hard to prioritize. But if you just clean it all up and say this is what I want to accomplish for the day, and then if you're ruthless about prioritizing, It'll do wonders for you. But this is the thing, time is the most precious commodity we have, we just can't get more of it, right? So, for me, it is way more important to prioritize and be ruthless about that than any other thing in the world.

So going a bit into prioritizing time and doing things which are meaningful, I'm wondering if you listen to any podcasts?

Generally I should probably say I do listen to podcasts at about two times the speed which has become a good time hack, and it's strange for me to listen to a lecture or anything like that in real life because I keep wondering when I can turn up the pace to 2x the speed.

And my final question for you is: why does it seem like the people in tech and the people in art are so similar? My background is almost exclusively in art, but the people I wind up connecting to most are all in tech, and every single one of these guys is virtually indistinguishable from a writer or painter or composer, and it's really the strangest thing. Since you live in the nexus between the highest levels of both spheres, art and tech, and you're also an engineer, I'm wondering, is tech a sort of veiled art?

This is a good question. I'm not sure we can say this universally about all tech entrepreneurs, but I certainly agree with you that there's quite a few of them that honestly could have been artists as well if they just applied themselves to this different field, but the reverse is true with many artists, too.

Many artists I know, if they had just chosen to apply themselves on a different path, they could have been amazing entrepreneurs, and a lot of them actually are amazing entrepreneurs as well in their own fields. So I certainly think that there is a lot of skill-overlap and some of that is sort of just seeing the world in a different way and being willing to be misunderstood. This is an interesting observation, I haven't thought too much about it, but now that you say it, yeah, I guess it's kind of true. I do hang out with both a lot of artists and founders and there are a lot of similarities between the two.

Well, it's been an absolute pleasure of a lifetime because I consider you, without a doubt, one of the top 10 most important people walking on this planet today, so it's a total honor to have been able to ask you a few questions, and I appreciate you coming in to talk to me today.

That's all too kind Sotonye. And by the way, I'm a huge fan of the show so I really appreciate you doing this with me.

To keep up with Daniel, follow him on Twitter now!

Get full access to NeoNarrative at www.neonarrative.us/subscribe