Robin:

Hello, Agnes.

Agnes:

Hi, Robin.

Robin:

This time, let’s talk about rot.

Agnes:

OK.

Robin:

So as you know, creatures like us, we are born, we grow, we develop, we improve, but then later on, we decay, decline, we die. We rot. So the principle or the idea is that many kinds of systems either grow and improve or decay and decline.

Agnes:

Or first grow and improve then decay and decline.

Robin:

Right. That’s a very common pattern. And so first, we want to think about which kind of systems do which in our overall experience. And then ask the question, which kind of systems that we might care a lot of about might switch from one to the other, i.e., from growth and improvement to decay and decline?

Agnes:

Can I just ask a question here? Isn’t always going to be the case that you first get growth and improvement and then decay and decline? Is there going to be something that only ever decay or something that only ever grows and improves?

Robin:

Well, if we draw the largest circles around say, the entire planet earth, you might say, the planet so far has been improving and growing in terms of biosphere, innovations are accumulating. So we have developed eyes, we developed ears, we developed brains. Overall, the biosphere has been improving. And life has been expanding into more kind of niches. Recently, it has left the planet. And similarly, we might say for civilization and human societies, they have spread around the planet. They have become larger in number. They have each developed more sophistication and complexity, more innovations have accumulated. We now have social media, surely, the highest form of all. Overall, human society looks like it has improved and grown as a larger collectible. Though individual human civilizations have risen and then decayed and declined. So it’s about the scale of aggregation that we put things together.

Agnes:

And I mean in terms of the biosphere, I guess I’m not sure how do we assess – are you assessing just by the leading edge of evolution being us or are you assessing overall like numbers of species, climate change, killing off certain species, or whatever? What is the measure?

Robin:

I think most measures will agree on growth. You could just look at sort of the total biomass or the total metabolism. You could look at particular innovations and whether they had – how many innovations are represented. You can even look at the length of genomes because that expresses the complexity of organisms by all these measures and overall improvement.

Agnes:

OK. So the decay hasn’t come yet.

Robin:

Right. Well, that’s what we want to ask and we are worried about, is the decay coming? Because that’s this big important future question. Similarly for human civilization, we could say, individual human civilizations, those have grown and declined like the rise and fall of Rome or of China or other kinds of human civilization in the past, rise and fall of Egypt. And we might worry that we are today a civilization that will rise and then fall and that sounds like a pretty important thing to figure out. So are we part of a current civilization that will decline – decay and decline in the future? And so in order to ask that question, my mental strategy is to go look at other examples of rot and decay and try to see what pattern we can about when things rot and decay. So if we look – start with biology, then we could say there are these individual organisms clearly, usually grow and improve and then decline and decay. We look at species, we can say, “Well, they look like they are not decaying.” Overall, the whole biosphere is improving but it does seem like individual species typically decay. That is, it’s a small fraction of species who are the source of most future species. And most species actually don’t give rise to descendants and decay. And so, we might want to ask, what’s different about the species that tend to give rise more descendants versus the ones that go away?

Agnes:

So, does every species presumably does give rise to descendants?

Robin:

No. In fact …

Agnes:

How could there be a species that has no descendants?

Robin:

Well, descendant species.

Agnes:

Oh! Well, well, well.

Robin:

No. No. So every species will have organisms of which it’s composed across time. But most species eventually become extinct. They will become extinct. And then – without even giving rise other species.

Agnes:

Right. But so, like elephants.

Robin:

Right.

Agnes:

Right. They have not given rise to any other species, right?

Robin:

Well, there might be like several kinds of elephants now and they might be seen as different species of elephants and maybe only one of them will give rise to more in the future or something and the other species of elephants will go away.

Agnes:

OK. But your thought would be like, what would be success for the elephant species would be to give rise to another known elephant species.

Robin:

That would – I mean …

Agnes:

That seems an odd way to measure success for species. Like we – the success for humans would be to generate another nonhuman species like what – isn’t the idea that those species wants to survive as its species, not to give rise to some other species.

Robin:

So there’s something of an arbitrary choice of when we say two things were different enough to be in different species. But there is some coherence to the concept. But if species almost never last a long time unchanged then we would have to talk about the changed versions and some of the passing also. So somebody like you and I might want to live forever but more realistically, it’s our descendants through children that will forever and so they are different from us. But still, they are our heritage.

Agnes:

Right. But it somehow seems rude to me to iterate that at the species level like the species is the inherited …

Robin:

Well, species do change. And so, a species might change enough over time that it wouldn’t match well with its initial version for example. And then you might think it had changed but that’s not central to this discussion.

Agnes:

I mean you might think it has improved or something.

Robin:

Right. But in our discussion, that’s not very important.

Agnes:

OK.

Robin:

The main thing is just, what do we know about species that give rise to more descendants? And it turns out they tend to be more general. So if we take a kind of beetles let’s say in an environment that’s changing a lot and very varied then it would be kind of a generalist beetle. And then if one of those beetles went to a very stable environment, the desert say, then it would become adapted to that desert and then very specific – deal much better with the specifics to that environment but then it would not give rise to as many descendants because of those specific adaptations, say, if this desert beetle gets dropped into a jungle, it will find it harder to adapt to the jungle than it would with the original generalist beetle. So that’s one thing we understand about rot is that becoming specialized tends to produce rot. That is, adapting to very specific circumstances tends to mean it will be harder to change and readapt to some other circumstance.

Agnes:

So like I just want – I want to get some sense of what your sort of markers of like success are, because rot is a sort of a negative term like value term whereas growth is the success term. So like ferns, OK. I guess I heard at some point like there are more species of ferns than of anything else. OK. I have no idea if that’s true but someone said it. Let’s suppose it’s true. So does that make on your view, does that make ferns super successful because they have all these?

Robin:

Yeah. So – well, there are different ways to measure this. You could count the number …

Agnes:

And they are very old too.

Robin:

Right. You could count the number of species. You could count the total biomass of all these ferns. You could count the total metabolism that all the ferns produced through themselves. These would be different measures of success but for our purposes here, they don’t make much difference.

Agnes:

Right. It’s I’m trying to say growth in like …

Robin:

I would more want to focus on accumulation of innovations maybe.

Agnes:

Right. It’s what I mean. It’s like ferns are not very impressive.

Robin:

Well, they are better than they when started out. So ferns have been getting …

Agnes:

And they’ve stuck around apparently a really long time.

Robin:

Right.

Agnes:

Right? So by some metric – that’s what I’m saying, by some metrics ferns are a giant success.

Robin:

Sure.

Agnes:

But I wouldn’t want to be a fern. So …

Robin:

OK. But that’s – again, rot is such a ubiquitous thing that we don’t need to settle all possible questions of value to just decide which things are rotting.

Agnes:

OK. But I mean, but you want to use it in this genetic sense and I know want to know whether ferns are rotting or surviving.

Robin:

And there are some – I mean maybe if we have to answer that question, we have to go back to it. But my bet is, we don’t need to answer that question to discuss rot usefully. So another example here, my best example actually where we understand rot the best is software. So I used to be a software engineer for nine years and in software, you make big systems. And one of the main mechanisms for making systems is modularity. That is, you make different parts and you try to make the parts not dependent on each other very much. When you make the system modular, then when you add or change any one of them, you need to make minimal changes to the other parts in order to change one part. And so, that’s a key feature of many of our engineering systems in our world like your car has an engine and it has a brake and it has windshield and you’re allowed to sort of redesign windshield without making matching changes to the brake and the engine. And that’s a key thing that allows improvements to things like cars, etc. There are different parts and you can change each one without changing the other ones that much. And so, modularity is a key thing that allows improvement in many different parts of the system – our world. And software makes heavy use of modularity, these different parts being sort of independent of each other. You can change one without changing the other parts. And a key thing we noticed that as people make systems bigger and older, they continually have to change them to adapt to either changing computer language, changing computer hardware, changing customer environment, or changing demands of customers for new features. Just continually people are changing software, and as they change it consistently, it reduces the modularity. You go in and you make a change in one thing and then you have to make changes elsewhere and that connects things together now and makes things dependent on each other that didn’t used to be. And those dependencies are what make software rot. And software does rot. And so if you don’t know what’s an amazing fact that software itself literally doesn’t change, right? The computer memory can be so reliable that no bits would be able to change and you can have exactly the same file from 100 years ago and you know no bits have changed. However, even so, the software rots in the sense that it becomes harder to modify, harder to usefully change, and eventually, pretty much all software is thrown away and people start over from scratch. And that’s true of the largest software systems. And in fact, Google today basically rewrites all of their software every three years.

Agnes:

So, can I just ask because I didn’t fully follow the explanation here? So the idea is that as I make changes to one part of the software, that require – and I make changes to another part, I’m making those two parts less independent of one another?

Robin:

Right. Exactly.

Agnes:

Just from the fact that I’m making changes.

Robin:

So it might not be obvious but this is just the key fact that we want to build on here. That is, it’s just a fact about when you make a system separate, when a new part of the environment comes in, it turns out you usually need to make matching changes in multiple parts of the system in order to adapt to this new thing.

Agnes:

But supposed you do make matching changes, why does making matching changes make them interdependent?

Robin:

Because the – in order to maintain that match, when you change one part, you’ll have to change the other part to keep the match.

Agnes:

I see. So it’s not just that you’re making the same change in two places but you’re making a change that is itself going to rely upon that match.

Robin:

Right. Exactly.

Agnes:

And why are you doing that? That is, why don’t you make a change in such a way that it doesn’t rely on the match?

Robin:

It’s a deep question and I don’t think I can give you a good answer and I don’t think I’m necessarily know the right way to explain it. But it’s just intuitively clear when you write lots of software that this is just really how it happens.

Agnes:

Is it also true of other kinds of engineering? I mean is it true …

Robin:

Yes, it is true of many kinds of engineering. Indeed. So for example, it turns out that your brain over time starts with what’s called fluid intelligence and ends up with what’s called crystallized intelligence. Crystallized intelligence knows more but an old dog finds it hard to learn new tricks. And so in fact, the human brain is more flexible when young. It’s more able to learn new flexible concepts, new changes when it’s young, and then overall it becomes inflexible and harder to change. So this rot process happens with human brains across time.

Agnes:

I just came up with an example. You tell me if this is an example but it’s something. So I had an Apple laptop computer that I wrote my dissertation on. So this is in the early 2000s. And it used to have a thing where if you could put a nickel in the back of it and twisted it, you could take the battery out, which was wonderful because I would work in coffee shops and have a spare battery.

Robin:

Right. Swap it in.

Agnes:

And swap it in and out.

Robin:

Right.

Agnes:

And then somewhere around 2008 or something …

Robin:

They said no more.

Agnes:

You can’t do that like on the new Mac, you can’t take out the battery. It’s integrated now.

Robin:

Yup.

Agnes:

Right? Which is annoying to people like me who like the – but it’s also much smaller, right? So I guess they improved. They made it smaller but they could only do that but also, making it less separable.

Robin:

Right. Less modular.

Agnes:

Less modular.

Robin:

Right. Right. So this was I think was the difference between IBM PC and the Macintosh from the very beginning. The IBM PC was this modular system. You could buy the chip from one company and the memory from another company and the speakers from another company. You just put them all together in the box and then it would all work. Whereas Apple said, “No, you buy them all from us and then we make all these parts well-matched to each other but we’re not going to let you mix and match.”

Agnes:

And so that has been very successful for Apple. So why not just think, what about just embracing non-modularity right from the start? Would that be just a nightmare for software if you made it all interdependent?

Robin:

It just means it rots faster. So the signature of rot is it’s harder to change. That is, it’s harder to usefully change. And so if you just make it hard to change from the beginning then it’s just hard to change right from the beginning.

Agnes:

And so like if we talk – Apple is pretty successful, right? But the idea is that though it’s not modular from the customer end, still inside the Apple it is still modular.

Robin:

Right. Exactly. They are still. But they need to like have it work in fewer environments. So I told you that one of the things that produces rot is having to have your system work in new environments you didn’t anticipate. So the more different things you’re trying to deal with, one straightforward way to do this is make all these incremental changes that slowly make it rot. A better approach if you could do it is to pick a general structure that sort of added a level of generality that encompasses things. So one standard solution to avoiding rot is generality abstraction, but it’s hard. It’s hard to be general and abstract. And so typically, we don’t bother to search in the space of abstractions to find the most general things. But one thing that happens in software is called refactoring. As systems start to rot, one thing people do is they try to work out the higher-level structure and simplify it and clean it up in a way that doesn’t require them to change most of the code but then makes the whole thing more flexible and then allows it to last longer but it’s still doesn’t prevent the eventual rot.

Agnes:

And it can’t just do the refactoring again?

Robin:

Yeah, it can. But it’s not as effective. So that’s a scary fact because with your brain when we think, “Oh, my brain cells are decaying, that’s why I’m getting more inflexible or I’m old.” But the software or hardware is not decaying. The explanation is in the software itself, not in some other system. And that’s probably also true in your brain. That is the reason your brain is getting inflexible is probably because of the nature of your learning process makes more and more interconnections. So now, we would move to say, social systems. And we might say social systems, do they rot? So language is – it doesn’t seem like languages rot, although it’s harder to tell. But they don’t necessarily have many parts that have to match each other. But legal systems, more plausibly rot.

Agnes:

What are the oldest languages? Like I wonder – I mean there’s a problem about individuation, right? It’s the same as species.

Robin:

So old languages accumulate more vocabularies so it’s a known feature that older languages have more words and especially languages that were used by more people end up with more words. But that’s a – but still, law or systems of rules, so they’re more like software, and systems of rules can get large and complicated in ways that when you change one part of the rules then you need to change the other parts of the rules to match. So if you have say, a rule about salary and compensation and another rule about travel reimbursement, it might be that these two rules need to be made to match each other.

Agnes:

So like one thing is, you can imagine that there’s an area of human life that was not engaged in very systematically. So like there weren’t like rules or whatever. It’s like free law periods, right?

Robin:

Right.

Agnes:

And then it’s important we make laws, right? And then from the point at which the laws are made as a system, from that point, the decay begins.

Robin:

Possibly.

Agnes:

And so, you can also ask sort of at what point does the thing becomes such as to decay? And I wonder – so the reason I’m wondering – if you didn’t like universities, and this thing about the bureaucratization of the university, you could think of that as a kind of rot.

Robin:

It might certainly feel that way. Yes.

Agnes:

And that in effect there are all these things that we are trying to make consistent now with one another, hiring practices in different departments. There’s a lot of consistency desire and admissions and whatever and that there’s more and more like committees and oversight required to make all these things consistent with one another and then we would go back. And was there some time, maybe not even that long ago like maybe like 1950 or something when in effect, a lot of this stuff was just unregulated.

Robin:

Right.

Agnes:

That there were no laws and there was no system. The people just did whatever they felt like doing and lots of injustice or whatever.

Robin:

But that could be flexible and therefore, not rot. And you simply say, norms like informal social norms because they’re not very formal or precise then they can flexibly change as you change environments. So you might have a bunch of people about words, about hitting people or being close and as you move from the jungle to the savanna like the meaning of the words will change. And then the norm can stay flexible because it was never very precisely penned whereas words on paper for a university policy might not be so flexible as to be able to change with such circumstances.

Agnes:

OK. That’s really cool. I think it suggests an answer, a different sort of answer to a question that we were considering two podcasts ago about why don’t people just say like when you go to France, you should ask three times for something when someone says no if you want a library card or whatever?

Robin:

Right.

Agnes:

Right? Why don’t people make explicit these little social norms and rules that we have like where …

Robin:

Because they would be different rules if they are explicit and then might they change with context …

Agnes:

But once you – or I mean it’s not even just that they will be different. It’s that once you start making these things explicit, you put a clock on it, right? It’s good now.

Robin:

Right. Sure.

Agnes:

But this is going to expire. It’s got an expiration date once we made it explicit. That’s super interesting. The idea – because people, people like you and me actually are often complaining that social interactions can be sort of difficult if people are not willing to be explicit about what the rules are, what they want, or whatever. And it’s like, but here’s a really good incentive to not be explicit because then we are going to have to in effect, systematize and then we have to manage that system. And if we avoid that point of systemization, we can prolong so to speak, our infancy.

Robin:

So another set of data here is about firms or some other kinds of organizations. Firms consistently rot in the sense that they grow faster when young and they grow slower when they’re old. And of course, they also die with a high rate. And so, firms rot. And one of the ways that firms rot is that initially, small firms don’t need very many rules and they don’t need very many formal processes. And then as they get bigger, they tend to accumulate more rules and formal procedures and ways to evaluate people and to rate people and to make changes, etc. And interestingly, most firms don’t like take some standard package of all these things from elsewhere. They kind of invent them all themselves organically as their firm is growing. And then later on, firms become stable but harder to adapt to circumstances. And so, old firms are famously like having trouble competing with young firms because they are stuck in their ways just like an old dog has trouble competing with a young dog learning the new tricks. And so, old firms are consisting of such rules but also just procedures and sort of dependencies between different parts of the organization in terms of how you do marketing, how you do manufacturing, how you do design. You end up with them having gotten a corporate culture that assumes certain kinds of features of the rest of the organization which then they become hard – resistant to changing.

Agnes:

So – but I’m really interested specifically in the case of the university. I think – I mean sorry, the case of the firms is interesting to me and I read Peter Thiel’s book, Zero to One, and there was almost a kind of obsessive neurotic focus on like – it’s always day one of your business, like maintaining that mentality of like never being complacent, always being alert. Like in a way, that struck me as like …

Robin:

So there’s a famous book by the founder of Intel about only the paranoid survive. So like in a rapidly changing world, you have to act as if you were a very young, flexible firm or you will lose out to them.

Agnes:

Right. Right. So – but like, I’m less interested in that in that it’s OK – I think it’s OK if firms kind of come and go and that seems OK. But I’m interested in the case of universities because I mean universities are some of our oldest institutions, right?

Robin:

They are.

Agnes:

And so in that way, you could see them from your point of view of like let’s say, longevity or success as being very successful. And I at least have this feeling but I think – I mean I don’t know if this is true but like it does seem like sort of like sometime around, I don’t know, 1970 or something, this sort of process of bureaucratization begin and it is kind of proliferating and proliferates itself. And so like, was there some specific point at which sort of like universities began to rot? And what caused that tipping point?

Robin:

So there is this larger phenomenon in society of increasing regulation and increasing government agencies. And that larger pattern produces government agencies that are very long-lived. That is, it’s very unusual for regulation ever to get rid of that agency or even to substantially change its structure. And so …

Agnes:

So it seems immune to the rot problem.

Robin:

Well, but the question is, it might be slowly decaying in performance but still retained. And so, in the world of say, firms or organisms and biology, they are facing this strong competition. So the older animals die out because they can’t keep up with the younger ones in say, fights or running away from predators or things like that. And similarly, firm, older firms …

Agnes:

Well, that doesn’t seem true about animals like even – I mean animal – humans and other animals just decay and die and even without competition. I’m not going to die because of competition. I’m going to die because my body is just going to decay.

Robin:

Well, I mean quite often, what happens is your body is captured by predator before it would fully decay. That is …

Agnes:

OK. But even if there was no predators …

Robin:

Most animals don’t die of old age. Most animals die of say, a predator or …

Agnes:

But they would die of old age. It’s not like they would just keep going like the government regulation.

Robin:

OK. But – so it’s – but what is happening as they decay is that they become less functional, right? Both internal and external functions, that is, they are less able to run, less able to chase, less able to climb, less able to digest, less able to hold themselves up with bones. So their bodies have all these different functions to perform and pretty much all the bodily functions are decaying. And that will show itself either in an internal failure, i.e. the heart fails or the bones breaks or it will show itself in an external failure. They can’t chase – they can’t get enough food to sustain themselves or they can’t escape their predator. That’s how organisms decay and rot is there is a steady rot process that then produces these various events like starving or failing to swim across the river and drowning or all these other sorts of things that will make an animal go bad. And firms, similarly, as they get older and less flexible, they are less able to come up with products customers like or less able to keep their customers happy, less able to lower the cost of their products and so they slowly get weeded out. Sometimes it might be a pandemic like just happened that suddenly makes a bunch of them go away. But it will be because they are becoming weaker over time that produces death at that point. So, this gives us – a question like about things like universities or government agencies. Could it be that they are decaying, i.e. becoming less functional but still because they are not in a competitive – as competitive an environment, they are just preserved?

Agnes:

And sort of propped up.

Robin:

Yes.

Agnes:

Like a stuffed lion or something.

Robin:

Exactly. So I mean the standard observation for example about wars and militaries has been militaries who were a long time between wars, when they start a new war, they are often pretty bad at it. And the recommendation even since the Roman times was just always be in a war somewhere so that your soldiers were in practice and learning about all these techniques, etc. so that you could be ready to fight whichever wars you need to fight. And so, in an era of longer peace, it might be that your militaries are decaying and you don’t notice it because there hasn’t been a war yet. And many other kinds of government agencies are there to deal with the problems. Say for example, the CDC supposedly is there to deal with the pandemic but it has been a long time since we saw them deal with the pandemic. And now, here’s the pandemic and can we really tell if they are doing a bad or good job? Maybe we just like keep them in the role anyway because they say they are doing a good job and we don’t know the difference.

Agnes:

Yeah. I mean I actually find that really hard like a lot of people say they are doing a bad job and maybe that the people I trust say they are doing a bad job so I tend to think, OK, I guess so. But I wonder like what do we compare them to? What ultimate conceptions of the CDC do we have? But anyways …

Robin:

Right. So that would be the reason why government agencies do continue, right? It’s hard to create a coalition to get rid of them because you can’t point anything very concrete to say they are doing a bad job.

Agnes:

Right. So – but like let’s say we can measure rot, I don’t know how we measure it with the government. But it seems like with universities, potentially, we could just measure it by the ration of the bureaucracy to the faculty or something, right?

Robin:

Well, it has been getting worst by that figure. We could talk about how much students are learning compared to the cost of what they are paying but it turns out, they don’t actually learn that much anyway. So …

Agnes:

But I think that – I think the cost issue could be a different – could be a distinct problem, right?

Robin:

Sure.

Agnes:

Like if you think, well, maybe the cost of university is rising because the cost of everything else is falling but we haven’t found a way to be – teach at a bigger scale or whatever. That could be – it doesn’t necessarily need to be rot. I mean you could think of that as a kind of rot but I think it’s slightly different.

Robin:

Right.

Agnes:

But I’m interested specifically because the bureaucratization really seems to me to match your picture of rot because it basically is this thing where you’re trying to make things consistent and I see this happening in my own department, where now we need to figure out like oversight for this or a way of choosing the oversight for thing, right? We need a fair method for choosing the oversight committee for it, right?

Robin:

Right. We could just go choose and not know if it’s fair but then we’d be done quick and we would have a choice.

Agnes:

Right. And – well, I mean – but I don’t think like it seems to me that it’s not that like there are people who as a fetish just love to create more bureaucracy. That’s a story that I hear but I don’t think it’s true. I think everyone is sort of dislikes it but we are driven into it I think by exactly the thing you’re pointing to, there’s a desire for consistency. And as – in effect, the desire for consistency itself adds things that then have to be made consistent and then reduce more …

Robin:

You could say a desire for procedure. That is, if you say – when people say, “Well, how do you make that decision?” you just say, “I just do.” They say, “Well, that’s not good enough. We need to see the criteria for the decisions so we can expect that and approve the criteria for decision.” People often want there to be an explicit procedure with explicit criteria and choice moments and deadlines, etc. so that they can think that it’s an OK process.

Agnes:

Right. So it’s sort of like – I think it’s really a lot like the France case, where it’s like – you’re like, “Why don’t people just tell you that you have to ask three times?” Like there is this idea of the value of transparency and fairness and explicitness, right?

Robin:

Right.

Agnes:

And that you really do get with having and explicit …

Robin:

But if they wrote down that you need to ask three times, and now somebody only requires two times and they give it to you, now, people might complain to that person. Now, they need, “Oh, I need to ask three times.” Now, they go, well, which counts as a time? Like this was a small thing. Is that big enough to be a thing?

Agnes:

Right. Right. Exactly. So that’s my point, is that it seems to me that the desire for bureaucracy is born out of this desire for something like explicitness.

Robin:

Right.

Agnes:

And fairness and a procedure and then you’re going to want like consistency just because part of what it is to have a procedure is to have consistency.

Robin:

So it turns out that our legal system has had the same long evolution in the sense that say, 150 years ago, most people could go to court without a lawyer and then they could just make their case informally because the law wasn’t very detailed or formal and mostly things were the decided by a jury. So it’s kind of up to a set jurors just to decide what was and they didn’t need much instruction from judges. And over that 150 years, the law acquired all these rules and distinctions and principles and legal facets such that now the jury can’t just decide. They need the judge to tell them, “Now, on this case, you need to be applying this principle, which has there three tests. You need to go through these three test, to see which one applies to this case.” And in some sense, this has been supposedly able to adapt the law to have more consistency and formality about each particular case at the expense of meaning, you have to hire a lawyer now to go to court. And so, you don’t go to court unless you could – it’s worth it enough to bother to hire the lawyer. And you don’t really understand the process even as a juror what you were doing and why because you are following all these rules somebody gave you.

Agnes:

See, I really get how that would happen with law because of the idea of precedent as being so important. If the precedent is going to be the thing that like is determinative then of course over time, you are just getting more and more and more precedence.

Robin:

But most of the world doesn’t use precedent as we do and they still have the same problem.

Agnes:

I see. OK. So then the precedent isn’t a big driver. Really the driver is this procedure …

Robin:

It’s one of many drivers.

Agnes:

OK. So right. So then the question is like, is there – I still want to know about the university. If we measure rot in the way that I suggested, not by cost, OK? But by ratio of bureaucracy to faculty at a given school, college, university, whatever, then like it seems like that just picked up at a certain point relatively recently.

Robin:

Right.

Agnes:

We are like for kind of thousands of years before that …

Robin:

So you picked the year 1970 and that turns out to be the pivotal point from any other kinds of regulations as well.

Agnes:

Oh, OK. Good.

Robin:

And so in fact, in 1970, there are a lot of things, trends that changed around then including a lot of increases in regulation in many different ways including universities. So we still could ask, what’s the fundamental social cause of that change? But it does seem like that that showed up.

Agnes:

Right. And so, I mean I think what’s interesting is that you could think that the rot like in the software case, it looks very sort of in some way, endogenous in the sense that there’s almost like this life cycle or something because it’s slowly adapting and modifying and trying to maintain its modularity than lose the modularity and maybe there’s that reboot thing that you described.

Robin:

Refactoring.

Agnes:

Refactoring, yeah. And then – but like eventually it’s going to go out and you can fight this to some degree just like with that aging. But with the universities, it almost sort of seems like they were going along for like a really long time and then something happened and then it’s like the process sets in or something. I mean of course, you can think more generally about the university and the government that there is this danger. Maybe you think it’s actualized of the dead carcass stuff that happened, right?

Robin:

Right.

Agnes:

But I guess that with university, we have this way of almost measuring it, right? And so it’s like what happen such that the thing began to rot?

Robin:

Certainly, one of the things that happened is just expanding the list of things that universities are supposed to do for faculty and students. That is, you can imagine a very sparse university that basically just sends students to classes and assigns which teachers teach which classes and then tells them, “Go figure out your own research.” And then if there’s not much structure, there’s not much to rot. And so rot is about large complicated structures. And so in that world, there is no human resources department. There is no student entertainment department. There is just – there is the classroom and there’s the teacher and for students, we tell you, “Show up at this classroom,” and for teachers we say, “Show up at this classroom,” and that’s all we have to do. And then there’s not much to rot. But if we now want to like have a fair procedure for picking the students and a fair procedure for picking the professors and for which professors go to which classes and how they are graded, the more we add on desires to oversee and to – and more services we are providing like we have a pool hall and we have dining halls and we have …

Agnes:

Looks like schools – schools like Harvard had dining halls and student theater and all this stuff for a long time. It’s not as though before 1970, it was just classrooms and teachers.

Robin:

But they might have just – they had a student here, they just put somebody in charge and say, “You can do it however you want.” And they didn’t have much oversight to tell them how they had to pick their people or how they pick their place and how they assign which actors for which and what days they could do performances on. They are just like, “You are in charge. You just do it.”

Agnes:

Yeah. So this is super interesting. So it’s sort of like what we would want – what would one would want to solve the problem of rot would be a mode of becoming explicitly conscious of a certain domain that somehow did not lend itself to this problem. Because it almost seems like it’s a problem with consciousness itself or something. It’s about – that is, it’s a problem with something becoming conscious or explicit.

Robin:

Well, so I would tend to say that’s a problem when you are trying to oversee too much. So I tend to advocate like trying to set up the right sort of competition and then just leaving it alone and letting people like work it out. So for example, if you have sporting teams and you say, “These are the rules of that game and anybody who wants to make a team, they can make a team. And if you win the game then you go on to the tournament. If you win the tournament, you get the trophy.” And you don’t tell them how they have to train their students or how they have to pick them or what they do between games. You just tell them, “This is the game and you show up and this is how you play. These are the rules of the game.” Then that process doesn’t necessarily need a lot of oversight of the other elements. But if you start to add rules about who you are allowed to hire and what gender they are ought to be or what age or what race or how you have to treat them in between, now, you will see accumulating more rules. This is in essence, the issue of regulation. That is, when the economists are staying, look, you need a very minimal regulation. You just need to find the things that are most important problems and deal with those and not be having all the rest of this stuff because all the rest of this stuff is going to produce this big complicated mess that’s going to rot.

Agnes:

OK. Fine. I get that. Like that’s the kind of very unambitious support to this problem, which is to say, sort of like with the software thing, it’s sort of like doing that say it again, reboot, it’s another word.

Robin:

Refactoring.

Agnes:

Refactoring. Like it’s refactor this to its basics and let’s not have all these rules. But it’s just going to – it’s just going to happen again. That just seems to me like not really solving the problem. It’s sort of pushing back against the problem in a reasonable way that we should probably do some of that. But the really interesting thing would be to try to say, OK, like there is this human predicament where we kind of get into habits, right? And those habits have patterns and we learn the patterns from one another. And then at a certain point, we start to speak them and make them explicit and we write them down.

Robin:

And punish deviations.

Agnes:

Right. We legislate, right? And we do this with language and grammar. We have books of grammar or grammatical rules that like – I don’t know when the first like book of Greek grammar is that we have a record of but it’s not going to be like in my Homer’s style or something, right? There’s going to be some point at which someone is like, “We have to make the rules for Greek. People have been speaking Greek for a long time before somebody told them what the rules were.” Right? So there is this thing of making the rules. The rules always come after the activity.

Robin:

Right.

Agnes:

So the really interesting question to me would be is there a way to make rules to be explicit that somehow doesn’t lend itself to rot or lends itself far less to rot.

Robin:

Again, I think you just have to pick a very limited set of things about which you’re making rules and limit how much rule-making you make. So for example in rot, there are software libraries that say, handle matrices. And those have not rotted. That is, we have a sort of standard definition of what a matrix is and what the matrix operations are. That’s sort of well-embedded in math. And the mathematicians aren’t tempted to change those definitions. And so, if you make code that does those operations then you can just leave that code alone and come back to it decades later and it still works basically because it was in such a stable environment that rot wasn’t the problem. So the rot is a problem of having a lot of different things you are trying to match and having them changed. And so for example, I would recommend that we do something very minimal for say, even regulating medicine in the sense of saying, making sure they have transparent – telling you what their prices are or their qualifications are but then we might say, yes, if somebody convinces you they can do something but they can’t, you make it hard. And what we want is this process of reputation and people talking about what doctors are good to be the process by which we slowly over time produce quality because that’s a sort of informal process that doesn’t rot. And so, there are these more flexible, more organic processes that we have available to us to solve many problems but we often would rather have a formal process instead.

Agnes:

One thing that’s interesting is that you could actually imagine two different directions to push back on rot. One of them would be the one you suggested, which is you have more barebones instead of rules. I still think you’re being too unambitious about trying to think about different ways of making rules that’s less acceptable to this problem. But – so that’s one way. Another way would be stop change, right?

Robin:

Indeed.

Agnes:

I think it’s actually a popular solution to this problem.

Robin:

I agree. It’s emotionally popular, yes.

Agnes:

There’s this question, why do schools like Harvard and the University of Chicago double or triple their size, right? And I think that one answer, you could say, OK, it’s prestige. But still, these places would be very prestigious if they double but they still have really low admissions rates. I think that – you might think, well, think about how bad a rot problem they would have if they have to build those new buildings. They have to do all this new stuff, right? And so the resistant to change maybe because they feel it would – they would rot more.

Robin:

At a larger scale, that is the main cause of rot. That is if you look at old company, probably within that old company, each part has been saying to the other parts, “You want to propose a change over there. I don’t think that’s a good idea. Let’s just stay with what we’ve been doing.” Because that helps them keep their existing structures perpetuating but then as the overall firm, they are innovating less. And that’s why they lose out to competition firm. So you could imagine an entire nation that decided not to change very much that said, “We don’t need this innovation. That’s very disruptive. We just want to keep things the way they are.” And then it might decay relative to outside competitors. And so, if you face outside competition of other things that are say, rotting and dying and being replaced by young things and you try to stay old and stable then you might just lose. And so, as we talked about earlier, one scary solution is that people will form a world government which then has all these world agencies and then sees the problem of decay if they have to adapt to changes and say, “No, no more changes.”

Agnes:

To me, it’s not obvious why that’s scary. I was trying to bring that out to you, right? So like if the problem is, that other people are changing but you’re not, that’s going to be a problem for you. But supposed we all just decide not to change, then we can keep all our institutions and there’s less rot like internally, right? It seems great. And now you’re like, “Yeah, but we are not going to send out these colonists and then they’re not going to take over a big swath of space and then the other guys are going to come and they’re going to eat or whatever like eventually.” And it’s like, sure, they were the – they were the grabby ones and we were the non-grabby ones. So there’s more of them now and there could have been more of us, though the us that there would have been more of would be like the colonists who became super different from us and came back and conquered us. And like it’s just not obvious to me that that was like the success scenario.

Robin:

OK. So this is the myth of say, Shangri-la. Some isolated community that has founded some ideal array of arrangements and it comes at the cost of stasis. And as long as they can stay isolated, they enjoy their stasis in harmony and then eventually outside contact, they’re overwhelmed because they did not keep up.

Agnes:

Eventually, everything dies I think is a good lesson here. And so eventually, we are all dead.

Robin:

But not everything dies and fails to produce descendants, right? So the Shangri-la of the scenario doesn’t create descendants, doesn’t make a tradition that grows and accumulates with time such that it can become a big influence.

Agnes:

I think it’s less obvious. So supposed that we don’t send out any colonists, right? So we just stay like the way we are and we don’t advance down the same point. But we have like culture and literature and we’ve got lots of peace and good stuff.

Robin:

Right.

Agnes:

Now, they come back and they just kill all of us. But they still have like all of our culture artifacts so they would be interested in us. They are going to be interested in our way of life and in the fact that we were able to like give contentment for let’s say 100 years or something. And then they might – like for instance, they might adapt some of our practices, read our books. So we might be influential and that in that way, perpetuate ourselves. It’s not clear to me that we would get a more substantive perpetuation of ourselves through the half descendants of us that come back and take us over.

Robin:

I am content to just merely point out that there are these two major strategy options that have these dramatically different consequences. I think most people who see these consequences, they all know how they feel about them. I don’t need to argue them into it.

Agnes:

I don’t think you’re actually content. I just think you retreat to that when I argued against because when you start …

Robin:

But you’re not having an argument. You’re just saying maybe it’s not so bad. But that’s – I could say the same about the other one, right?

Agnes:

Well, I’m asking you to give me an argument that it is bad because it’s not obvious to me. You are presenting the thing as though it were bad, right? And I’m like, not only do I not see it as bad. It seems good and it seems like potentially a legitimate way of reproducing our culture too. And so, you’re not just – your “scary scenario” as you introduced it is – I don’t see why we should think it’s scary.

Robin:

So when the invaders come into Shangri-la and they meet happy natives before they slaughter them and slay them, and they see the pretty spires and then – and they are pretty dressed as they walked around in the village square, one of their main framing is, “These are losers. These people – we are the winners. They are the losers.” This is the behavior that results in losing. That doesn’t necessarily inspire people to copy it. It’s an object lesson and the failure to avoid. And this has been a long-standing story in history of the successful and then decaying decadent culture or decadent elites even, which often were realized in our actual story in history where outsiders did come in and deride and blame the locals for having allowed the decadence to make them weak and allow them to be displaced and to no longer have descendants.

Agnes:

I mean there are many people like throughout history who – let’s take Socrates, right?

Robin:

Kind of random I guess.

Agnes:

So Socrates – so first of all, he was – the period of like Athens being a democracy in the sort of cultural height of Athens, that’s sort of when he was living, right? And we could say it decayed, right?

Robin:

Afterwards.

Agnes:

It decayed afterwards. And yet, people to this day are obsessively interested in it, read about it, and it’s hugely influential even though it decayed. That’s the fact number one. Fact number two, Socrates himself proved to be a “failure” in your sense being in a competitive system of people wanting to put him to death and he didn’t have enough friends and influential people to convince people not to put him to death or whatever, right? And so, you could watch him fail, being – his Shangri-la being taken over, right? But neither of these things actually prevented people from learning from him, imitating him, admiring him. People that are taken over, individuals that die heroically, those aren’t necessarily considered to be losers. So like I guess I’m not sure that you’re right about that generally. But also, I’m not sure that what we should care about is like, will they think we are losers? We should care about will we actually be losers? That is, by our own lights. And they might make all sorts of – those aliens will make all sort of false judgments about us. So neither seems obvious to me that they will think of us as losers nor does it seem obvious that even if they did, that would be a sign that we had failed.

Robin:

So I would present the choice as like you have some scope of influence and existence now and you can just focus on making your current existence the way you wanted or you can focus on having future influence. There are many channels to future influence but nevertheless, if you are focused on future influence that would be a different set of actions you would choose compared to just enjoying things now. And so, that’s sort of the primary distinction. And then in this context, I would say it’s a very dramatic example. So basically, here we are right now as a newborn civilization and we will either become say, the grabby aliens that go out and meet the others in a billion years after which we would have filled a volume of roughly a billion light years across with people like us for – indefinitely.

Agnes:

I mean not much like us.

Robin:

No, somewhat like us for indefinitely, right. Or we could be one of these non-grabby civilizations that stays in one place, lasts for a little while, and they may dies off and leave some ruins. And then one of these grabby civilizations in their vast expansion across their billion light years of space comes across her ruins. And they read something about us, maybe something about us touches their fancy and they remember it a bit or copy it a bit. Now …

Agnes:

Or like they developed a whole theory and ideology of the base of it and it becomes like the Paul Principle.

Robin:

OK. But I mean you got to – like they’re going to come …

Agnes:

It’s pretty influential.

Robin:

OK. But you’re looking at the tail of distribution, right? I mean most other thinkers from thousands of years ago, you feel lucky, punk.

Agnes:

What we are going to give them is like the totality of all. And so our best stops. So they are going to take …

Robin:

OK. But one of the reasons that we remember Socrates and not somebody from Bangladesh is that Socrates was part of a very successful civilization.

Agnes:

Right.

Robin:

Right? And so …

Agnes:

That decayed.

Robin:

But was still much more successful than many other competitors. So being part of a successful species, a successful civilization, a successful empire, those are all things that help you spread your ideas, etc. into the universe. So again, you should think about yes – I mean one of the reasons you might want us for example to expand the universe is that you think we already have many valuable things like Socrates. And if we don’t expand out there, other people wouldn’t be as interested in hearing about Socrates. In some sense, our choice to expand in the future will make a big difference to whether anybody will ever care about Socrates in the future.

Agnes:

Like I can get here with this. I mean I supposed I think that it seems to me we ought to – the reason we ought to like want to – of course, the main reason we want to expand is to meet some of these other creatures because it would be cool to meet them.

Robin:

Then we just be more. I mean there can just be more of us.

Agnes:

I’m just not sure that the being more is that valuable. But I mean I get that there’s the utilitarian argument. OK.

Robin:

Then there could just be more Socrates. I mean the more there are, the more chances there are for future Socrates, right?

Agnes:

I mean I’m not sure that that’s right. Maybe it’s right. But in any case, like it seems to me that we should want to expand and to meet other civilizations, sort of to the degree to which we actually think we have something to contribute like that is not merely to like …

Robin:

Well, the way we hope – so I would express it that way. I would say, “We should hope to expand and develop and then when we meet the aliens, we should hope to have something worthy of emulation.” And we shouldn’t assume that. We should think about that long and hard and try to find things that would be worthy of emulation and admiration by them. That is, we should try to win their approval honestly by being admirable and good at something. Not necessarily at everything. There should be many other things out there that they are better at us and we should emulate them and that we should expect this future where each civilization looks at the others and ask honestly which things about them are better than what we have and we should copy those and which are our things that are best of anybody and that we should try to spread and show to them.

Agnes:

Right. And like one of the big advantages I guess the other side and I think you’re right is that like if we need these aliens, there are things we could learn from them that can’t learn by ourselves.

Robin:

Right, all their Socrates.

Agnes:

Exactly. And so, those seem to me to be good reasons to want to both stick around and go far. Not the reason of wanting to take up more space or wanting to not be the losers, I find less compelling. But the wanting to have something real to contribute to these other aliens, something of genuine value that they would want and to be able to take the things of value that they have. And that those opportunities for growth are not really available like if we think that they are not really available to us, absentees encounters or this kind of long-term strategy. I mean you might even just think, “Well, humanity has a lot of improving to do and it’s not going to be able to do that improving within the shorter time.”

Robin:

We don’t know. So it could be that within the next million years, we will basically learn everything worth learning. And so will they say. And then we will meet in a billion years. We would not have much to tell each other because long ago, we all just learned everything worth learning, right? That’s a logical possibility.

Agnes:

I think in that case, it really just doesn’t matter whether we are the grabby aliens or they are.

Robin:

OK. But there are other possibilities in which there are things to keep learning over the extended long term even after a billion years of a billion light years, you are still learning more interesting things. And in that case then we could have things to offer and they could have things to offer us.

Agnes:

A really interesting thing is that Socrates in the Crito, he is offered a chance to escape prison, right? And he is – and you might have thought, well, he could learn more. He could keep doing more philosophizing if he escapes. And he refuses to escape because he thinks he would be breaking the law and he thinks he was tried and he has agreed to these principles and it would be wrong to – even though he thinks that the verdict was wrong, he doesn’t – he says, “I wasn’t wronged by the laws but I would be breaking the laws.” And so, I think that you’re right that it’s worth having this ambition of having something really big to contribute. But it’s also like I guess important to keep in mind that there are ways we could conduct ourselves that would sort of undermine the value of what we have to contribute. That’s what I think Socrates is saying about escaping. He is like, “I just wouldn’t be worth anything. In fact, I wouldn’t be worth emulating later.”

Robin:

OK.

Agnes:

So one of – the stasis people, maybe one of the things driving them is this fear that like in effect will become corrupt or bad in some way.

Robin:

Right. And many people think competition is the source of that corruption. So this is a common sort of dispute between economists and other thinkers. What many people think about the future and what can go wrong and how it would go wrong, they put that at the feet of competition. They imagine that if there are bad products in the future, it’s because of competition forced companies to offer bad products or if there’s bad university it’s because competition forced people to try to offer bad universities as a way to compete in the competition for universities or nations have produced militaristic attitude because that’s what has helped them win the military competition with other nations.

Agnes:

Well, I have to say, I used to not be very sympathetic to that and I was always just a really big fan of competition that exists viscerally innately. I enjoy competition and I’ve always seen it as a positive. But the thing that I see – I see actually really damaging effects of it in my own field where in effect like the competition for getting into grad school, for getting a job, for getting the paper accepted as a journal, you would think all of this – my old theory – according to my old theory, Agnes will just produce amazing, excellent philosophy, right? And it hasn’t – I don’t think it has had that effect. It actually has like I think a bunch of negative. There’s some great stuff but what it has really done, what competition above anything else is create a lot of uniformity. So people – candidates applying to grad school kind of all the same as one another, papers sent to journals all look the same as one another.

Robin:

And so that’s why this whole subject of rot is relevant. That is – so clearly, say biological competition has enormous cost, right? All these animals out there hiding from each other, running from each other, attacking each other, all this energy spent on like camouflaging yourself or finding ways to sneak up on something, etc. So there is biological competition and it covers enormous cost and enormous hardship. I mean predation is a pretty rude and mean thing. And many animals literally suffer as did our ancestors from predation. So it’s definitely not an ideal world but it at least, it doesn’t rot. And then if we pose me some other world and you could say, “This world that we all talk about things, we vote on them together, and approve the rules. We are a community and we value each other, etc, but we rot.” And now if we look in the long run, we say, “Well, how does this game end?” And the competing ways to worlds that are improving will displace the peaceful, cooperative rotting world that go away. And so, this is a fundamental hard question is, how can we have the minimal competition to prevent rot so that we can have a long-term future and minimize the damage but still not rot?

Agnes:

Yeah. I still think that the real conclusion for me is, we just got to find a way to create non-rotting explicit rules.

Robin:

Well, I am happy to endorse someone’s effort to try. And I guess we should end now.

Agnes:

Yeah.

Robin:

Thank you for talking.