And welcome to Non Trivial. I'm your host, Sean mcclure. In this episode, I'll be challenging the idea that human progress owes its advancement to the efforts of a few that society has progressed by riding the coat tails of giants. Those individuals that history tells us, made the greatest contributions. Well, the narrative that so called geniuses contributed far more others seems to make a lot of sense.
Does it actually hold up to what we know about how problems are solved, specifically how problems are solved among groups? While history paints the picture, the geniuses bring forth the momentous, the memorable, the profound, the things that are really important. Is it possible that this attribution to the few is far more narrative fallacy than reality? Let's go find out.
OK. So let's talk about this idea of, of the giants, the geniuses, the people throughout history that supposedly contributed a lot more than other people, right makes sense. That narrative makes sense. If we think about progress, this idea that society advances in, in some sense of that word.
Uh You know, we, we get inventions, we get new pieces of music, we get scientific discoveries, we might have uh you know, uh political leaders who contribute to our understanding, economists, perhaps um you know, philosophers, titans of industry activists, people of religion in all these different categories of life. If you look throughout history, we know that we have big names attached, right? So we're talking about science.
We got Galileo Isaac Newton, Albert Einstein, Charles Darwin Engineering, we got people like Nikola Tesla writing, we've got, you know, the William Shakespeare's music, Mozart Beethoven, art Leonardo da Vinci among others. Uh although he would fall under quite a few different categories, political leaders, Julius Caesar Alexander, the Great Napoleon mode, right?
We got George Washington, Abraham Lincoln, we got philosophers like Socrates, Plato Aristotle, you know, Karl Marx, we got the, the leaders of industry, the titans of industry like Ford Rockefeller, Carnegie Morgan. And we've got the new guys, the musk Bazil Zuckerberg's, we've got activists, you know, Martin Luther King under religion, Thomas Aquinas and so on. So so we can go on and on. We all know these big names. We've all heard of them.
And if, if you were to tell someone that look, you know, I think like human progress has relied largely on the contributions of very few people relative to the entire population that most people would agree with that. Like, yeah, that makes sense that so much of what we have today is because of the efforts of a few individuals who had, who were either really smart or they had some kind of insight that other people didn't have, you know, maybe they were just in the right place at the right time.
But look, if it wasn't for those people that these, these big challenges that had to be, that, that were solved got solved, it wasn't for them, they wouldn't have been solved or they wouldn't have been solved as early. And so we wouldn't be where we are today without them. That story seems to make sense. Human progress seems to owe itself to uh a very few select set of individuals throughout history that have gifted us with their contribution, gifted us with their genius and their insight.
Well, you remember in the logic episode, part one and part two, you know, I said that, look, if you believe in something, you sh you should probably be able to back it up. Right. Obviously, that's the case if you're debating another person.
But even just for yourself, if you hold a certain belief in something you should, you should be able to reason about it to yourself through introspection, you should be able to say, ok, well, here's why I think that's true and we, we tend to take a lot of things at face value through life. Uh I think a lot of times we have to, we don't have time to challenge everything.
And one of those things that most of us just seem to kind of take at face value is this idea of giants, this idea that we owe that we are where we are today because of uh you know, a very small number of people that made the biggest contributions and then everyone else is kind of in the background, right? Everybody else is kind of the backdrop to that and, and thanks to those few people, we are where we are, we tend to take that story at face value and it seems to make a lot of sense.
So you might not think to ever challenge it, but does it make sense? Does it hold up, does it line up with what we know about how problems are solved? Because that is what we should be asking and look, this is a big one. This is, this is this idea that very few people actually contribute to what we have today. It would affect how you think about yourself, right? Because how do you contribute to your job and to society? What will your life mean at the end of it in terms of its contribution?
Because if we think that only a few people are really, you know, contributing uh to, to what we have today, then you know, that might seem kind of depressing. Now, I'm not saying we should challenge the idea that giants made the biggest contribution because you might feel bad. I don't think nature really cares whether you, you feel good or not.
Uh in terms of what is true or not, but it might make you challenge it because if you are believing that story and that in turn affects kind of how you live your life and whether you're not, you think it's worth to take on a new skill. Remember we talked about in the ski episode, whatever the reason it doesn't even matter, the point is, is that is a big enough story.
This idea that very few people are contributing to the outcome of things that we call progress, that you should be able to back that up if you believe it. Ok? For whatever reason, for whatever reason, I think that's something that you should be able to back up. And so that's what I want to talk about in this episode. Does that story that a few individuals have made the biggest contribution hold up to what we know about problem solving? OK. Now, why am I framing this in terms of problem solving?
Well, if you think about progress, you know, you could really sum that up as you know, the human society solves problems through time, right? We we make scientific discoveries, that's a solution to a problem. We we engineer uh these new inventions, these are solutions to problems, writing, music and art. Those are, those can be framed as solutions to problems, right? Because somebody's trying to create a new narrative, a new story.
They want to to to solve the problem of entertaining or they want to solve the problem of trying to educate someone about a topic better, more effectively. They want to solve the problem of creating a musical composition that resonates with people that tells its own story with or without words. They want to solve the problem of communicating through art, right?
They want to solve some political issues, some societal issue, uh some deep philosophy, they wanna reconcile something that seems to be a contradiction. They want to make money. You know, we talk about the the titans of industry, they want to make a process more efficient, they want to get a product out into the market. Maybe they are activists who want to fight for a cause. These are all attempted solutions at problems, all of them, you can always frame it that way.
So when we think about the giants throughout history, you're really attaching names to people who have solved a major problem, right? Or at least push the needle if you will uh to, to a better solution if solved is too strong of a word. Uh you know, the Isaac Newton's and the Albert Einstein's and the Charles Darwin's, they solved at least partially.
They made progress by coming up with solutions to some outstanding problems with, with, you know, mathematics with calculus, with, you know, the geometry of space time, with this idea of uh you know, relativity with, with evolutionary theory. And again, of course, all those other topics that we talked about, it's a solution to an outstanding problem. So we are talking about solving problems.
So if you believe that these giants throughout history contributed the most, then that should line up with what we know about how problems are solved by groups of people because that's what we're talking about, right? We have billions of people on a planet and you have millions of people that are, let's say devoted to particular areas, whether it's music or science, engineering, and then you have subsets within there that are, that are based on which domain they're focused on.
But the point is these are solutions to problems. And every once in a while, a really big one pops out or hundreds of years later, it gets recognized uh you know, uh in history even maybe it wasn't recognized at the time. Right, Mozart buried in the Popper. Popper's grave wasn't as big as he became, yeah uh in history. But uh but eventually history recognizes it.
OK. So I want to talk about something called emergent problem solving because when we talk about giants that there's, there's few individuals throughout history that came up with the solutions uh to to our biggest outstanding problems. There's something wrong with that. It doesn't really line up with what we know about how really tough challenges are solved. OK. So this topic uh falls under the purview of emergent problem solving.
The basic idea is that the group can come up with a solution, but the individual cannot. So when groups of people and groups of individuals get together, and some of them might have good ideas, some of them might have bad ideas. Some of them just have neutral ideas that go nowhere in, in some sense, it doesn't actually matter. All of those contributions work together to produce a solution that only exists at the system level.
At the group level, the solution pops out from the group, it does not pop out from the individual. So you can see how I'm relating this to the, to the narrative that you know, a few giants contribute the most, but this is called emergent problem solving. So I want to talk about this.
Now, I talked about this previously in an episode when we looked at uh ant colonies and termites and how they're able to kind of find uh you know, the the shortest distance between paths and that's actually a very hard, difficult problem to solve. And in fact, no individual ant or termite could solve that problem. It really takes a group in order to explore all the many possibilities and to come up with a solution. So this is called emergent problem solving.
If something emerges, it's not a property that exists in the individual, it emerges in the group. You have to have many individuals in order to have the solution pop out. This is how solutions to truly complex problems are solved. So you can see how I'm going to relay this back to that narrative that, you know, only a few people throughout history contributed the most and that's where these solutions came from.
Well, it doesn't really make sense unless you want to try to argue that, you know, really, really tough challenges can be solved by individuals. And I think that's a tough narrative to stick to. But let's dig into this a little bit more because I may not have convinced you yet. So let's talk about this emergent problem solving. So we say that in emergent problem solving the system solves the problem. OK, the system solves the problem, not the individual.
We say that the global functionality so that the big aggregate, the the the the the thing that you see overall for the whole group cannot occur from the properties of the individual. It's not something about the individual that out of all those people made the biggest contribution or that led to the solution. In fact, it's not possible for that to happen for a truly complex problem. There are too many pathways with which you have to explore.
OK. The dimensionality we say of the problem is too high. There are too many features to consider too many attributes to take under consideration that the idea that one individual out of many individuals was able to kind of synthesize all that complex information together to come up with a solution. Uh doesn't really hold water. It's not what we see again, the system is what solves the problem. It's not the individual, right?
We don't see the properties of the solution present in the properties of an individual agent that makes up a group. So, so global functionality cannot occur from the properties of the individual. This is a type of self organizing knowledge and decision making. OK. Problem. This is important. The problem solving that you see in groups is greater than anything that could have been arrived at by the individual.
OK. So you might have a rock star in the group and you think they're really smart, but no matter how heavy their contribution might seem to be. And in fact, too heavy of contribution can turn out to be detrimental. It would never compare to the solution that a group could come up with. OK. Now there, there's, there's some debate here about what size of group this occurs for smaller groups. You could have an individual whose contribution outweighs the others and ends up being the best solution.
And that's not really that surprising because if you don't have a big enough group, you're not really exploring that many different paths. So in some sense, a small group is not that different than a really smart individual, right? The other individuals in the group might just be noise and the really smart individual might kind of just even if by luck have the best idea.
But if you start getting to reasonably sized groups that's where this really goes away, this idea that, that, that an individual or a rock star could have so much, you know, more intelligence or more insight than any other individual in the group. Just doesn't make sense. There, there's, there's no way that one of the agents could explore as many paths that you really need to try to explore than hundreds or thousands or millions of other people.
So, so in other words, there is a transition point from, from a pretty small group to a reasonably sized group where the group is definitely going to come up with a better solution. And that's just what we see and that is called emergent problem solving. So the problem solving of the system is greater than that of the individuals. Let's do some examples here.
So if you think about building development, right, so you're going to, uh, you know, you're, you're involved in creating uh uh a new building. It's gonna go somewhere in the city. Obviously, you got to know, you know, the landscape, you got to know the train, you got to know the area, you got to know the real estate, you gotta know all uh all these things to, to put a new building there. Well, one of the decisions you need to make is a walkway, ok?
You have to have walkways to and from the building, maybe around the building. And this could actually be a lot tougher to come up with. And you might think you might say, well, here's my building and here's the street and, you know, let's just, let's just put a walkway from here to here. You would think that that's something you could kind of pre design, right?
Something you could predetermine, it turns out to be a really, really tough challenge to get a walkway that works because what happens quite often is that people will decide that the building developers will say, ok, here's where we're gonna put the path and that's part of the design and we put it in place and then people take their own paths, right?
People start walking and they start be lining it to somewhere, you know, to, to the destination or you know, whether it's the train or security or weather or different modes of travel or different destinations, there's a lot to consider to actually try to come up with the best path. So you the the problem that we're looking for, the solution that we're looking for is a solution to the problem of what is a really good path, maybe even the best path to put as part of this building development.
Again, you think well, between the architects and the designers, I mean, surely you could come up with a reasonable kind of design uh to, to, to put into place a good path. But if people are just gonna trample over, you know, the lawn, wherever you put the path, you put lawn and other landscaping elsewhere. So they, if they end up trampling all over the other landscaping, then that wasn't a good design. So how do you come up with the best path in building development?
Well, what many building developers have discovered is, don't put the path in the design, do the building, do the landscaping and then let people figure out the path and once that path emerges from people just taking their natural path, we will uh eventually kind of lock that in and make that the path. And this is what a lot of building developers actually do. So you put the building there, you do the landscaping, you don't do any kind of pathing.
You allow people to basically, you know, wear out, right, the grass or we or, or, or the dirt or wherever they're going. And that will, that will be the path that emerges. Now that we, we it's hard to imagine that just the formation of path is like a solution to a complex problem. But again, different people take different paths, different destinations, there's different types of terrains, there's security, there's different weather conditions, people have different modes of travel, right?
You've got walkers, you've got people in wheelchairs, you know, some people are walking, some people are running. There's a lot to actually consider. So somewhere among all those possibilities is the best possible path you could make the most likely path. Uh you know, a human would actually take when going to and from your building. And it turns out that the best solution is to allow that to emerge. OK? That's the best approach to solving that problem. It's not to redesign it.
So some planners have learned that often the best solution is just let the system determine the path. OK. Why are we calling this a system? Because it could be over the course of time, thousands and thousands of people that are walking to and from that building and eventually the right path will emerge. And that is the best one. And there's no way to have known that that was the path that should have been laid out at the beginning of the building development project. Not possible.
You don't have insight, you don't have access to that kind of information, it's not accessible, it's not an opinion. OK? You do not have access to that kind of information. It's an opacity, it's a fundamental opacity of complex systems that you do not get to know the solution to that problem. It must be trial and error and it must have its possibility space explored through many, many, many individuals if you want a really good solution to that complex problem.
Another more familiar example could be Amazon book recommendations, right? So you go on Amazon and you look for books and you know, millions and millions of people are kind of just solving their own individual book finding problem, right? I want to find a good book, the the global system level solution that they're actually looking for at Amazon is how to recommend books to people. How do you recommend the right book? There's millions to choose from.
So how do you not just do that randomly or how do you not just take a a ridiculously simplistic approach, like just match the genre or something? How do you actually, you know, based on the content and based on a lot of kind of implicit features of the books recommend to people when there's millions of choices to choose from, you know, top 10 or top 20. How do you do that?
That's, that's a, that's a global level problem because you, you're, you're directing it at an individual but it, it's, it's involving uh the, the purchasing patterns of millions and millions of people, right? So again, every individual is trying to buy their own book. But the solution to recommending you the best book that you haven't read yet can't come from just you, right? It can't come from just you.
It's gotta come from the aggregate, it's a system level solution that must emerge from millions of purchases by millions of people. Let the purchase habits of individuals solve the problem for the collective. OK. And of course, there's, there's many examples like this, these non-trivial complex challenges are best solved via the collective, they have to emerge. OK. So why do we talk about emerging problem?
Solving again, going back to that narrative that there are giants or you're talking about a lot of people all working on the same thing, right? If it's, it's an area of physics or a type of music or whatever and in some sense, at some level of abstraction, they are working on the same thing, right? Just choose a domain, you know, the medical field, right? A certain area of engineering, maybe rockets or whatever, right?
It's, it's you, you choose a domain, there's always going to be lots of people working on that. And as individuals, you know, they are, they are trying to solve in their own way, that problem.
But in doing that effort, obviously, there are a lot of them talk to each other, they publish papers, there's communication, you're reading the works of other people, every individual agent in that system is even if they don't know each other personally, they're interacting to a great extent because you've got to read, you've got to promote, you've got to give presentations, you've got to consume information, you, you share knowledge to a great degree.
And over time, there was just an absolute ton of interaction between all the different people that comprise the group. And it's not possible to not be part of a group. There's no way it doesn't matter how much you isolate yourself. Because again, even if you didn't talk to anybody directly, you're not going to make any progress in science if you're not reading papers, you're not going to make any progress in science.
If you're not reading textbooks, reading the work of others and, and unlikely to make much progress if you're not promoting your own work, to some extent through publication, uh through online articles, whatever it is. So, so you're always in this communication no matter who you are. And that's been true all throughout history. Regardless of what the medium is.
people communicate, people talk and they, they, they get ideas even from areas, maybe even especially from areas outside the domain, right there, there's, there's the reasoning by analogy that happens where, you know, you might get inspired by a completely different unrelated or seemingly unrelated, you know, uh domain or aspect of your life and all of a sudden that motivated you to try this and there was some kind of conceptual overlap between the two and on and on and on.
So this idea that, you know, we're not even talking about emerging problem solving. If you just think about how embedded any individual is in the collective, without the choice to really separate yourself from that collective, then it's pretty hard to stand on this argument that yeah, the individual is, is, is really the person who contributed an overwhelming amount just because they're by default, relying on and using so much other information, so many other inputs.
And again, those inputs may or may not even be related to the problem that they're solving and yet they could lead to the next step, they could lead to the next piece that's done. And that's just the information. I mean, you're not even getting into the tooling that you're using, right, the background of things that had to be resolved for you to even have the language with which to communicate and the tools with which to use to make your measurements or to compose your music.
Uh you know, previous political theories that had to be in place in order to present the juxtaposition that you needed to come up with your own theories and on and on, just so unbelievably embedded in the collective that to take ownership to assign agency to an individual as though they were the major contributor to some true solution is pretty ridiculous on its face without even getting into emerging problem solving.
But then when you add in how complex problems are solved as we have talked about by groups of people, when you get an appreciable sized group, which of course is what we're talking about when we look at history and major problems that are solved. Because there's always, always, always going to be thousands and thousands of people working on any given problem, right? Any given problem.
And again, you could say, wow, no, this area, this particular domain was really, really, you know, esoteric and there was only a few people come on, give me a Rick, give me a break, there's too much overlap between things that idea that only a few people are actually working on one thing that it's just not true, that's not true. There's, there's too much conceptual overlap between different domains between different areas. There's always a ton of people.
If you really look in aggregate that are involved, forget contribution, but just involved in the communication of thought in, involved in the communication of ideas and in problem solving. So, so thinking about how embedded the individual is and then thinking about how complex problems are actually solved in aggregate at the level of the system, the assigning of agency to the individual starts to, to, to evaporate, it starts to disappear based on what we know.
So what I want to do now is uh is take a look at the properties of these system level solutions. What are the properties of the aggregate when we think about problems being solved here, one that we've already been talking about is how the performance of the collective can be better than any performance given by an individual that makes up that collective. OK? So that's something that we know about systems again, for really small groups, that's not necessarily the case.
But what we are talking about here is very large groups, always very, very large groups, right? Especially through time. So, so the collective outperforms any individual and it's it's not hard to understand why, right, a truly complex problems has this really, really big what you might call possibility space. There's so many possibilities with which you have to try and you don't have access to the information that would tell you which one is best.
So you have to have a lot of individuals attempting, most of which are failing at it. But they have to attempt. And those attempts, those failed attempts produced the most valuable information which eventually gets used by the winner by the person that we attach the name to throughout history to say that they were the giant. We also another property of these aggregates of these systems that solve problems is the reduced performance with less diversity.
And we'll talk a little bit more about this later. So if you have less diversity in the group, however, you want to define diversity but, but essentially people with different backgrounds, whatever that means, right? Different experiences, you know, they grew up somewhere different, uh they were raised by a different type of family, they got different ideas introduced and whatever that diversity, if you reduce that diversity in a group, right?
If it's more uniform, if more people are kind of the same, then you're going to get reduced performance. So if you're buying into the rock star narrative, this this idea that uh there are certain individuals that contribute a lot more, then how does that hold up to this property? Well, it doesn't, right? If you're reducing the performance with less diversity, if you have one major, major contribution. Well, that cannot be considered diverse, right? It cannot be considered diverse.
I mean, even if that individual is very, very different than the rest of individuals, that's not really an aggregate sense of diversity, that's still just one individual. OK. So, so for you to have, or you could almost think that that just kind of breaks down into a binary, then you've got one really different individual and the rest are kind of the same. Well, that's not really diversity.
Diversity is many individuals, all of which have a lot of different backgrounds, different world views, different experiences, different approaches, different methods, different ways of looking at things that's diversity, not having one individual that's really different, that's not diversity. Diversity is having many individuals all with differences between them. And if you decrease that diversity, if you make it more uniform, right?
Either because everybody is the same or because you only have one or a few rock stars in there and a against a backdrop where everybody is the same, either way that's not diversity. So if you decrease the diversity like that or, or let's say another way of thinking about this is you might have a rock star in the group, you think they're really smart and they have uh an overwhelming influence on the decisions that get made.
Ok, we, we're probably familiar with this in smaller groups and again, sometimes in smaller groups, like I said, before it could make sense. But as the group gets larger, you don't want to give uh a large amount of weight to any one individual. Because that by definition is decreasing diversity.
Even if you had a lot of diverse people in the group, they have to be allowed to interact and all contribute on equal footing if you don't allow for that, because you think some rock star should have a louder voice than everyone else. You're going to see reduced performance in the global solution. It's not an opinion. That's a property that's a property of complex systems. That's a property of many people getting together to solve problems.
OK. Another property is the ability to function in the presence of extreme noise and information loss. So if you have an aggregate group of people that are interacting and there's some global solution, let's say that emerges. And if you introduce noise into that system or you, which can mean uh you know, uh let, let's say taking individuals out or putting misinformation into the system, removing information from the system that it's extremely robust to those changes.
So you could randomly inject, let's say misinformation into a system and you would, you would assume that that would mess up the overall solution, right? All these agents are doing their thing and they're working away and the information that they're using to solve the problem, you know, is presumably reliable, right? But if you inject kind of uh you know, misinformation, wrong information or you kind of randomly take people in and out of that system.
The global solution is not being greatly affected. And this is true for the larger of the group. OK. And again, just go back to what we know about solving these complex problems. If you're, if you're attempting many different possible paths and you have lots of people doing those attempts. I mean, think about, you know, going back to that ant colony example from a few episodes back, if you're gonna step on the odd ant, that's not going to stop ants from solving the shortest path problem, right?
If you remove people that are walking in and out of uh the property after the building development, it's not like all of a sudden the path isn't going to be solved. I mean, it's going to be extremely robust because you have lots of agents constantly trying lots of things. OK?
You have a lot when I say agent, by the way, I mean, individual, that's just a, a term usually applied in complexity to, to mean the individuals, the agent could be the ant, an agent could be a human, 11 individual in the group. If you remove an individual or if you give, you know, a number of individuals, bad information to try to mess up the solution. It's like trying to kick something and it won't fall over.
That is a property of these types of situations where many people come together and a good solution emerges from it. It's actually hard to kill, it's hard to mess up. Ok? Again, because how would you do it? And if you can't know exactly where the solution comes from, which you can't because it's a mergen property, right? It comes from the group. But you don't know, you know, like how do the individuals come together to piece together that final solution?
Again, you don't really have a view into that, then you're not gonna know how to destroy it either. OK. And that's why a lot of these big, you know, when, when something takes off in aggregate, it can be very hard to stop, it can very be very hard to, to ruin. So they are very robust to change. So let's let's think about what that means again, for this idea that there's going to be an overwhelming contribution from a few individuals while that would be something that is quite fragile, right?
Because there would be this core source of information that is overwhelmingly leading to the best solution and, and if that person dies or gets removed or changes careers or whatever happens, then all of a sudden the whole system would collapse. But that's not what we see with these systems, these systems don't collapse.
So the discovery through the ages of, of, you know, new pieces of engineering, of new insights, of new compositions of, of art, of music, of new political theory, whatever it is, these things are extremely robust and these things have an inevitability to them, an inevitability to them and that they are going to come about. OK. So it's not like we needed Einstein, you could say to get relativity. Now, you probably needed him. I would say you definitely needed him to get relativity when we got it.
Right. But this idea that relativity was never going to be discovered is pretty ridiculous on its face. Because again, Einstein, like any other individual is embedded in a system with thousands and thousands of individuals, his work rested on all kinds of previous discoveries and mathematical frameworks that had to be preexisting for him to do what he did to have approaches and mathematical models and tooling.
Now that's not to take away from his great efforts and his insight and, and you know, the hard work that he obviously put into to, you know, sit in that patent office and go through those thought experiments and to spend hours and hours in the patent office or the cafes and taking those notes and going through those equations and, and you know, envisioning worlds in his mind.
But to have the fodder for that, to have the tooling for that, to have the foundation that you rest on for that is, is constantly being fed by the countless contributions of other people in that group that that group being the people alive at that time. The people that came before him, that group being however many individuals you could count up, that had some causal connection to finally arriving at what we call special relativity. And then ultimately general relativity.
And we can do that for the Mozarts and we can do that for the Rockefellers and the Carnegie and we can do that. The Teslas and the Galileo's and, and it's, it, it doesn't matter. The name that you attach through history is not a name attached to some core source of information that nobody else could have come up with, right? That is that, that must be very much a narrative.
And I think that this property of, of an aggregate able to function in the presence of extreme noise and information loss shows that that this the the signing of agency to discovery doesn't really hold up to what we know about properties of complex systems to, to, to properties of aggregates that solve problems. And so we can take this even further. Another property is that the collective functionality that occurs at the global level.
So the solution that emerges out of aggregate that can occur without any intentional problem solving on the part of the individual. OK. So try to get your head around that the the the the the a solution that pops out in aggregate the collective functionality, the global overall solution to the problem that emerges that is better than any individual can come up with that can occur without any intentional problem solving on the part of the individual.
So, so this means that no individual in that group has to be aware of the big problem that's necessarily even solving. In other words, the solutions can emerge from individuals that are doing activities that have nothing to do with the ultimate solution that popped out. And again, this makes sense. Now, we think about serendipity, we think about a lot of things through history.
The narrative fallacy is to step back and say, well, this person was, was obviously, you know, working on this problem the whole time. And that might be true for people like Einstein or, or Tesla, maybe the, you know, this was the domain but other areas they weren't even working in the right in, in that domain or they or they weren't even necessarily concerned with that type of problem.
Again, we, we, you know, I talked earlier about this, this kind of analog reason that happens all the time where you got this cross fertilization that happens between domains. This idea that that person must have been on that path. They must have been trying to solve the problem that history attached their name to when the solution popped up. That's that is not a prerequisite that does not need to exist. No individual in the group has to be solving the problem that gets solved in aggregate.
OK. Now I'm not saying this is always going to be the case. But again, this is another property of many, many, many individuals coming together, they're all operating in whatever kind of way they're doing what they're, they're solving their own particular problem, they're interacting the way they interact. And through time what pops out what emerges is a solution and it may not have been a AAA solution that anybody was actually looking for.
So again, we've got this major disconnect between the two scales, right? We've got the scale of the individual where if they're, they, they, they, they have global knowledge in the sense that everybody kind of is working towards the same goal. We already know that the problem that the sorry, the solution that pops out is is going to be better than any individual could have come up with.
But we also have this property where no individual in that group, you needs to be aware of the of the solution that ends up popping up, right? They could be all working on different things or you know, or, or, or they're all working on the same things. But the solution that pops out looks very, very different than any task that any individual was working on.
So, so again, and I know that's hard to get, you know, your head around if you, if you're not used to thinking in terms of complexity, but there is this absolute distinction and I say absolute because it's an abrupt transition between the two scales between the scale of the individual and the scale of the group, what happens is in, in the group is distinct, it has its own distinct properties.
It doesn't flow in, in, you know, via some logical chain of reasoning from the individual, the input to the output has this abrupt transition. The properties that happen in the group are fundamentally different. Then what we then then then what happens at the individual level?
So, so that property of collective functionality at a global level occurring without any intentional problem solving or problem solving that that's related to the solution on the part of the individual that show is just how abrupt and distinct that, that, that those two scales are OK. So no individual in that group has to even be trying to solve the problem that gets solved, which is probably it's really hard to get your head around.
But now again, go this is a property of these systems and let's go back to, to talking about giants, you know, history owes progress, it's progress to these giants, to these individuals who made this massive contribution. I mean, the narrative is, is is beginning to look more and more ridiculous. OK. We know that the group performance is greater than any individual could contribute. We know that group performance goes down.
When you, when you have less diversity, we know that it can function in the presence of extreme noise and information loss and misinformation no matter how much, you try to kick it, it gets back up. So this idea that some core root source of information was better than anything else that would have been too fragile. So that doesn't make any sense. And we've even got it that the collective functionality, the solution that emerges, no individual even had to be working on that.
No individual in the group even had to be trying to solve that problem in order to get that solution to pop out. Now again, the way that you're gonna hear about it in a history book or a science textbook is, is, is as though the person had been working on it the whole time. You remember in a previous episode, I talked about uh you know, we don't create the way that we consume, right? The way you consume information is a story is a narrative.
It sounds, this happened, then this happened, it goes by step and it builds up, builds up and then boom, there you go. You have the Opah. But that's not how it works in, in the real world.
The way we create is very much ad hoc and it's trial and error and you've got surprisal and the majority of what you end up telling in your story was as though you came up with, it was really revealed to you right through your constant uh you know, exploration and, and this, this almost near randomness approach to, to the way that we must, you know, create things.
OK. So, so history is going to tell you that story as though, you know, that's the problem the person was working on and they were just, you know, and I'm not saying individuals were never working on the same problem for a long time. But I'm saying that a property of these systems is, is, is such that the solution that pops out no individual even had to be working on that solution. So there is that distinction between the individual scale and, and what we see in aggregate.
OK. So, so you know, at the, at the beginning of this podcast, I said, how well does this narrative, the story that, that, you know, we all stand on the shoulders of giants that giants are responsible for, for progress. It doesn't hold up to what we know about how problems are solved by collections of people, which is of course, what this is is of course what this is everybody is embedded in a system with countless individuals in the past.
And today, you, you couldn't separate yourself from the collective. That's not possible. You know, you could be a hermit, but you'd still be reading, you'd still be studying, you would still be embedded. So how do we know that these are the properties of complex systems? I mean, obviously, you know, complexity science and, and computational complexity when it comes to problem solving and, and there's all kinds of rich uh you know, areas of study that you can look more into this.
But one of the approaches to really get insight into what large groups can and cannot do is through simulation. Because the computer is a window into phenomena that you would otherwise not have just through the kind of more traditional collection of empirical data.
Right, when we do things manually, traditionally, you know, when we use uh you know, the microscopes and we use a traditional scientific equipment to collect our data, we can only collect so much or or we collect so much that we have no ability to, to really analyze it or understand it.
We we we we have to have this computational approach to dealing with the data and to getting insight and and there's this asymptotic nature to computation, meaning that we can basically run things to infinity, right, or close to infinity, we can run simulations of many, many, many, many many individuals and kind of run that experiment through to see what happens in ways that you would never be able to, to, to do um you know, in the real setting, the closest you would get to do that in a real setting would be just looking at things uh through time.
So if you looked at, you know, for example, what survived through history, that would be kind of the real world example of, you know, many individuals over a long period of time doing things. But if you want to kind of get insight right here. And now you would do things like simulations, use the computer to get a window into phenomena that you wouldn't be able to otherwise.
So I just wanna uh make that point because there's one example that kind of extends from what I'm just talking about when we look at cooperation. OK. So we talk about the individuals not even being aware of the solution that emerges. So think about cooperation, you can do simulations of herd formation. So a bunch of people or animals or whatever getting together, and you look at how you know those dynamics play out herd formation.
And you can simulate those where the individual agents, again, those individuals are aggressive. Let's say they're only aggressive agents. So there's no, you know, built in mechanism for cooperation embodied in the individual. OK? We're not gonna allow for cooperation. It's just going to be aggressive agents. We can show that the collective behavior of the cooperation is still observed. The collective behavior of cooperation is still observed.
Cooperation can occur when nobody wants to cooperate. OK. So it means that cooperative behavior from essentially uncooperative individuals is a global structure that emerges. It's something that happens globally only at the scale of many, many individuals interacting, nobody wanted to cooperate. And yet cooperation can still occur again, that our minds are not used to, used to thinking like that because we want to assign agency, we want this logical nice neat narrative from input to output.
But when we look at complexity, when we look at many agents interacting, we get these properties, we get these solutions that pop out that don't look anything uh like like like what was happening at the individual level, right? You know, they like they say the water molecule is not wet, right? A water molecule is not wet, wetness is is something that only occurs when many, many water molecules are getting together to do what they do.
So cooperation in simulations is shown to be something that pops out even though that collective behavior was never imparted to the individual. OK, uncooperative individuals can lead to cooperation and and and this comes out of the dynamics of the system. OK. So, so we've run through a number of properties of, of, you know, the solving of complex problems by a large amount of people.
And I, I've relay that back to this, this narrative that, you know, doesn't make sense that giants are contributing the most and it just doesn't seem to align with what we know about problem solving with groups of people. Now, what I want to do is kind of touch on, you know, some people might be thinking, well, this is kind of depressing actually, right? This is kind of depressing in some sense.
I mean, at the beginning of the episode, I thought maybe it's depressing if you think that there are giants but you might think it's depressing if there are no giants because what does that say about your own, you know, potential to be known for something or, you know, is, is it, is the only way that I could be known for something kind of random?
Like if history decides to attach my name to something because I was in the right place at the right time or I almost my, my is my career going to depend on some kind of narrative fallacy is my recognition going to depend on that.
You, you could take this opposite side of things where, you know, uh you know, at the beginning again of the episode, I was kind of saying as though maybe this is a good thing to, to not think that there are giants, but maybe that's actually a bit depressing as well because what is the role of unique knowledge? You are a unique individual, you have your own experiences and, and you'd like to think that those maybe do in some circumstancess have an overwhelming contribution, right?
I'd like to think that my background is really going to shine in this situation. So what is the role of unique knowledge? Have we, have we kind of wiped out the notion of the individual here?
Have we have we, you know, taken away the chance for uh an individual's unique, uh you know, application of effort or their unique insights, their unique experience to play any kind of overwhelming, you know, role of contribution in the solving of problems because that could be kind of depressing what's important to realize here is that the group really, really, really needs your contribution. OK. Now, now that kind of sounds like a bit of a contradiction.
I thought you would, you know, I wasn't, I just saying that the individual contribution doesn't play an overwhelming role, it doesn't, but you're really, really, really needed. Ok. Now, to, to reconcile these two ideas, you have to understand that what matters in the group is diversity. So think about taking a really large group, you might say, well, if our, if our group just gets bigger and bigger and bigger, then that alone is gonna solve problems more effectively.
No, it won't actually because if you increase the size of the collective, that doesn't necessarily increase the diversity when the sampling, as we say is from a uniformly distributed population. Meaning if, if your really massive group is composed of all kind of the same people, then you're not going to solve problems effectively, you're not gonna get these aggregate, you know, uh emergent solutions that pop out that are better than the individual, you have to have the diversity in there.
But the important point here is that there is no strict definition of how you need to be different. The, the, the and, and, and I mean that in the most scientific systematic sense, you, there really is no proper definition of how you're supposed to be different. It's just, it needs to be different, the diversity alone, regardless of what that diversity is comprised of. It just has to be individual agents that are very different from each other.
And the reason is because again, the way these complex problems are solved is this is, is the attempting of many, many different paths. So the more diversity you have in the group, the more different paths you are going to attempt. And that's it. That's it. You can try to. Oh, but this person has relevant knowledge. There's no such thing as relevant knowledge when we're talking about very large groups, solving challenges.
This, this notion of really, really relevant knowledge starts to disappear is far more important to have people with different, maybe even seemingly unrelated experiences in that group because the diversity allows you to attempt more paths, period, period. So, so, so loading up your group with so-called experts or rock stars or people that you think have very relevant backgrounds is not the answer.
Now, you should have people with relevant backgrounds because again, that's going to increase diversity, right? If, if you said everybody in my group is gonna have irrelevant backgrounds and that's my diverse. Well, that's not necessarily diversity. Is it again, it's not, it's it, the the the notion of diversity is not, what is that difference supposed to be?
It's, it just has to be different because again, we don't have insight into what paths need to be sampled, what, what what paths need to be attempted. We don't, we don't know which direction we don't know which paths are supposed to be tried. We just know that we need to try as many as possible. So diversity, regardless of, of what that person's background actually is needs to be there to increase the effectiveness of the solution.
Let's use an example if you have a seasoned problem solver, and we think about false information. Well, a seasoned problem solver can actually be misled by false information. Whereas a novice problem solver can recover better from false information. And that's because the novice problem solver has experience and experience on how to return to a known path after being led astray.
In other words, the seasoned problem solver is usually very dedicated to a certain domain and so it constricts their problem solving ability. Now, if you land on that very specific problem, they're really, really good. OK? But if you don't land on that problem, you know, in other words, if it's a path that seems kind of unrelated, then the the the seasoned problem solver actually suffers from the misinformation. They can't, they can't work outside their narrow domain.
OK. Whereas the novice problem solver can actually recover better from the false information. Now, here's the 0.1 is not actually better than the other. What it, what this suggests is that diversity of novice and established problem solvers would actually be optimal in the presence of misinformation. So if you're gonna throw a bunch of misinformation at somebody, you want some so-called experts there, but you definitely want people that are not experts there.
You want the diversity again, going back to the robustness of systems diversity is going to give you the robustness. You're not gonna be able to kick it over. If you had only experts, it's gonna be fragile. And if you had only people that were, you know, selectively very different than what you thought the problem was, well, that could be fragile as well. You don't want that homogeneity, you want the heterogeneity in that system.
So again, I was talking about what is the role of unique knowledge, you know, what can you be proud of? You can be proud of your contribution because it's different, not because it's better, right? So I'm so I'm using this example of the seasoned and the novice problem solver because one isn't by definition better than the other. They are both needed. Your contribution is the fact that you're unique is the fact that you're different.
And I'm not saying this in some motivational sense to make you feel better because maybe you think you're different than most people. I'm just saying this is this is a rigorous scientific statement about understanding complex systemss. OK? It's important that you understand that this is a rigorous statement. This is not to make people feel better your true contribution is because you're different and it does not matter what the difference is. OK? That is a property of these systems.
That is a property. You have to understand how these complex problems are solved by groups of people. When we look at them, it's not what the difference is. Who cares? It's the fact that you're different and the bigger the group is, the more that this is the case and the more complex problem that you're trying to solve, the more that this is the case. So your, your pride and your contribution or whatever you wanna call, that should come from your unique experience.
You should be proud of, you know, I I I'm, I'm framing this as pride right now, but you should be proud of where you came from and, and, and what your unique uh you know, experiences are and the way that you look at life and, and the worldview that you have not because it, it maybe it's more relevant and not even because maybe it's more different, it's just diversity. You know, that in a group of people, a lot of people aren't gonna think exactly like you and that is what is needed.
So I think you can be proud of that because that's you and I'm saying the word proud, not because I'm, you know, trying to make you feel better, but because you, when we think about the role of unique knowledge again, if you think everything is gonna get kind of wiped out in the collective and, and that if there are no giants, then what is my contribution, your unique background is the contribution it is adding to the pool of variation to the, to the, to the pool of diversity that exists in the group.
And that's what's needed as a rigorous scientific statement. That is what's needed. OK. That's a property, it's a property of systems, That's that, that, that, that solve complex problems. OK. So we have these kind of self organizing dynamics that occur when many people get together. And it's not some simplistic linear superposition of information from individuals. OK?
It's not like, well, this person adds this and that person adds this and then it's all gonna aggregate together and the sum of all those contributions give you the output. No, no, no, no. The collective advantage appears via a complex interaction that requires diversity of performance. OK? Individuals can have wrong, quote unquote, wrong goals in mind that contribute to the beneficial outcome in the collective. OK. So, so it's better to think of differences almost in the random sense.
It's not what this person is better because it's more relevant or this person has better performance, maybe they do and that's good because that's a diverse contribution. But so is the person that's got a completely so called irrelevant experience. OK? And, and this is just a property of the system. And so the point is we, we cannot know what information is really needed to come up with a solution. Again, the way that story is going to get told.
Well, this is Einstein or Mozart or whatever they were influenced by this and then they composed that and then on and on and it builds up and builds up. And that's why you have this symphony. And this is why we have general relativity. This is how alternating current was discovered, you know. Yes, that's the narrative. We need to hear stories, but that's what they are, they're stories. Those are not things that align with known properties of how problems are solved. They're not.
So the point is we cannot know what information is needed to go from input to output. We cannot know the information that individuals are supposed to have those individual agents that, that, that will contribute only that there was a greater probability of finding the right information if more individuals of greater diversity are involved. Now, what about the role of heroes? I mean, am I kind of wiping out heroes with this? Right? You might agree. Well, yeah.
OK. Giants, maybe they didn't have an over, you know, it doesn't really align that narrative that, that Giants played an overwhelming contribution. But wait a second, I grew up being inspired by Einstein or Tesla or, or, or, you know, industry giants, maybe Elon Musk or, or Bezos or, you know, the, the Mozart or whatever it is I had these heroes and to say that they didn't actually play an overwhelming role in, in these big discoveries.
And it was actually, you know, more the collective of the group, you know, doesn't that kind of wipe out my heroes? Well, I think there's two ways to think about that. I think one is, first of all regard, I mean, narratives exist for a reason. We, we, we have to tell stories. If we didn't attach names throughout history, it would be hard to anchor the, the the story of progress, right?
I mean, we, we do this spatially as well in the sense that we do this with landmarks and, and you know, things that were built. You know, if you think about Paris, you think about the Eiffel Tower. If you think about Toronto, the Sean Tower, if you think about, you know, Rome, you've got the colosseum and we, we, we need to anchor our understanding, you know, geographically with these different, you know, landmarks.
Well, it's the same thing when we think about people through time, we wanna have heroes, right? We wanna have names attached to big discoveries, we want to anchor that. And so II, I don't think there's anything wrong with that per se. I think that, you know, it's ok to be inspired. But, but when you're inspired by someone's name, you really are just inspired by the name, not the person.
In other words, you don't really know the person, you don't really, you know, the, the, the, the idea that they were the ones that specifically contributed everything. You know, I don't think you should study the life and try to come up with that recipe again. I've talked about that in the past. It's not like they actually had a recipe to arrive at that problem because quite frankly they weren't really the ones by themselves that arrived at the solution. sorry. Right.
They didn't arrive at the solution by themselves. It was, it was something that happened in aggregate. Um And so I think the inspiration doesn't need to come from the life of the individual per se. But more just associating the name with the discovery. It's not about Albert Einstein, right? I mean, even though again, obviously he, he uh he did a lot of great work and had great insights and he should be, you know, recognized.
But it's, you know, Einstein as a name is much bigger than Albert Einstein, right. Einstein is really about special relativity. It's really about general relativity. OK. Mozart is really about the type of composition uh that, that can be created in music. Tesla is really about uh the the ability to, to harness alternating current and to use that for an amazing array of inventions. And, and we can go through that list of, of different giants.
The name is just an anchor to something that was discovered by many, many individuals contributing and we can assign agency, we can kind of make up these narratives, but really, you can be inspired by the fact that, you know, human society was able to come together and, and have those solutions pop out and, and the fact that everybody had that contribution via their diversity, right?
And again, it always sounds like, you know, I'm almost talking uh in a motivational way or, or sometimes, you know, you might even think politically, you know, you use the words like collective, it's really not what it's about. It's just looking at the properties of complex systems and, and, and that's the way it is. So I think you can be inspired by that when you know what the real story is.
And I think it's OK to use obviously names like Einstein and Mozart and to be inspired by the, the events and the discoveries attached to that name or, or to which that name is attached to but, but keep in mind, it's not that unique individual story that led to the discoveries. I know those stories are fun to hear and they themselves can be motivating, but it's really just a name attached to history. And I still think that can be inspiring.
And the other side to that is that I think it's inspiring to know that we all play a role. I think it's inspiring to, to take that realistic view and realize that my unique con is, it's not just AAA feel good thing to say, you know, your unique contribution is gonna be really helpful. It might be something you hear on like a getting hired to a new job or a job description or something, you know, we want your unique contribution and it just kind of sounds like lip service.
But from a rigorous scientific standpoint, you can say no, actually my unique background by definition really does help the overall solution and it might not be my name that gets attached in the history books to whatever great solution pops out. But I am actually playing AAA fundamental role in in the emergence of a of of a solution that will go down in history. My may might not be a attached, but I am actually playing that role and it's true and you are and so I think that is inspiring.
So you don't just have to have the heroes to look up to, you can say that look, you don't have to be famous, you don't have to be recognized to know that you know, you just have to contribute, right? You just have to contribute, you have to get in and you have to try to build and to create things and by definition that is globally solving complex challenges we all do by definition play a role. And so I think you can be inspired by that too. OK?
One more thing I want to touch on before we end this episode. Uh you know, I talked about how, you know, removing the Giants from the picture. Uh it might make you feel good, it might make you feel bad. It kind of depends on the perspective. But again, there's different ways of looking at it. One thing we didn't touch on is what about the bad guys?
So, on the good side of things, we have these giants as we call them the, the people throughout history that supposedly made these really good contributions. And so we anchor that uh in, in the kind of the positive sense, like here's all the good things that happen through history, these big discoveries, they, they, they improve things and they allowed progress to happen.
But the inverse of that is we have these atrocities that have happened throughout history and we've got names attached to those atrocities as well. And so if we take this stance of, well, look, it's not really about the giants, it's not really about these individuals or the or the major contributors.
Does that kind of wash away, you know, the quote unquote sins of these people whose names are attached to the atrocities is that kind of just, just, just to kind of blanket over that and say, well, look, you know, it just kind of came from the group. It's not really about that person. Uh you know, does that, does that just kind of remove almost the blaming the attribution of really bad things that, that really? Well, wait a second. That person needs to be kind of called out throughout history.
We need to know that, that they were kind of the, the quote unquote supposed architect of these really bad things and, and that's important to, to have that story as well. So, are we kind of washing away the bad guys as well? And maybe that's not good? Well, look, I think it's the same thing. Ok, I think that when you hear a bad, sorry, when you hear a name in history that is associated with a particular atrocity that that still is just a name.
It's because what's most important there is the atrocity itself. And it's important to realize that yes, we should use that name because that name is the anchor. That's the anchor to the underlying phenomenon to the underlying, in this case, atrocity that happened. But it's important to understand that that still arose out of a collective. If you just point at the individual, if you just assign agency to the bad thing that happened and say, well, it's because of that person.
And if we had stalk that person in history, if they were never born or if they were diverted to more productive means, if they were, you know, raised in a different environment, then that individual never would have come to architect the atrocity and therefore all would have been. Well, well, that's a very naive idea because you're not again, thinking about how things emerge and things arise through the collective actions of individuals.
Another way of saying this is that look, no country is immune to atrocities happening to atrocities, repeating. And we all do play a role. It's not necessarily a direct role. You know, again, the the individuals don't even have to believe or even be aware of what is ultimately going to pop out at the end in order for them to contribute to the thing that pops out at the end. So we're all responsible as a collective for not only the good things that happen, but also the bad things as well.
And if you don't think that you're not doing a service to, to the promotion of better ideas and better solutions and a better society of all overall, you're not uh pointing to individuals as though they were, the source of the evil is not productive. It goes against what we know about the properties of complex systems and it's not going to make the world a better place.
So I'm not saying don't use the names just as we have the giants, just as we have the Einsteins and the Teslas and the Bezos look those help anchor the story through history, but they are a label attached to the underlying phenomenon. They are a label attached to the underlying discovery in the case of good things and atrocities in the case of bad things. So use the labels, use the narratives but understand that these things arise through the collective actions of many individuals.
Those individuals do not have to be actively contributing to the bad thing in order for the bad thing to happen. So I think it's the same story. So it's not about washing out, you know, the the agency all together and saying, well, look, nobody's responsible. And so I guess sometimes good things happen, sometimes bad things happen. It's, it's understanding that we are acting as a collective and that good things can happen and that bad things can happen.
And if we have a better understanding of the properties of how groups act and how things emerge, uh regardless of, of what the individuals might individually be acting on or with, then that's, that's gonna better prepare us to, to realize that we are all susceptible to good things and to bad things. And there are actions that can be taken that are in that, that, that are done in a way that respect those properties.
So you might say, well, what can we do in the world to avoid atrocities and to, and to move towards better productive outputs?
And, and I think that's gonna be uh maybe another topic for another episode because getting into, you know, when it makes sense to intervene and when it makes sense not to intervene, you know, you got these ideas of social engineering and how those can be naive and those break apart and they don't understand how complex systems work and that, so, so what kind of decisions can you make?
Um when you don't have access to that kind of internal information that tells you how things you know, are kind of connected and, and now A leads to B leads to C if you can only kind of understand, I say only, but that's actually much more powerful if you can understand the properties of complex systems. And what does that mean for decision making? What does that mean for policy? What does that mean for, you know, government intervention or the lack thereof and all that? And now that's it.
So, so we can go on and uh I can go off on that. And that's, that's definitely a topic for I think another episode because that's a big topic in itself. But just the take home message there is that understanding how things emerge in collectives, understanding those properties is what gives us the tooling and the methodology to, to make better decisions, to not do naive things like saying, well, there's giants and that's why we have all these discoveries or there's evil people.
And that's why we have all those atrocities, these are not productive, they are, they, they do not align with what we know about how problems are solved in aggregate and how bad things arise in aggregate. If we actually look at how these things happen and we understand the properties, then we can take better actions and we can, we can move towards more positive outcomes. OK. So that's it for this episode. Thanks for listening.
I uh I know I took quite a bit of a break there between the uh the last kind of part two of the logic episode and this one here. So uh a bit of a busy spell, but I'm back at it. So I hope to uh hope to maintain a good pace from here on out. Um You can uh if you want to debate, debate me on some of these topics then feel free to jump on Twitter. Uh Just make sure that it is uh an actual debate uh that you, you are backing up what you say, you know, I'm a fan of people who back up what they say.
So uh you know, happy to have some fruitful conversations. OK? I think I'll start giving some suggested reading at the end of the episodes when it makes sense to do so. Um a really good paper that touches on a lot of things I talked about in this episode is called collective problem solving functionality beyond the individual. And it's by Norman L Johnson, I will put a link in the episode on the site. So go check that out, click, it can do a little bit of reading.
Get into some of the technical details if you want. Again, that's collective problem solving functionality beyond the individual by Norman L Johnson. So until next time. Thanks again. We'll see you soon.