The following is a rough transcript which has not been revised by The Jim Rutt Show or by Simon DeDeo. Please check with us before using any quotations from this transcript. Thank you.
Jim: This is The Jim Rutt Show. I’m Jim Rutt, your host. Today’s guest is Simon DeDeo. Simon is Assistant Professor at Carnegie Mellon University, Department of Social and Decision Sciences. He directs the laboratory for social minds. He is also external faculty at the Santa Fe Institute. Simon, can you give us a brief overview of the department and your lab?
Simon: All right. Thanks, Jim. Social and Decision Sciences is a department here at CMU where we mix together the economic and the psychological. We’re interested in how people buy and sell, obviously. We’re interested in how businesses work, how markets work. We’re also interested in how the mind works. In particular, how the unusual workings of the mind intersect with the markets in ways you might not expect. We tend to think of ourselves at least when we operate economically as rational people. Money from the line in solution behave well.
Simon: We mostly don’t behave well particularly when we’re confronted with situations that are somewhat counter to the things that are cultural both to handle. Out of this comes behavior economics. Out of this comes thinking fast and slow, behavioral heuristics. We’re interested in the ways that those things go right. In fact, a lot of the snap decisions we make actually are pretty good. We’re also interested in the way those things go wrong. I come out of the Santa Fe Institute. When I was a post doc there, I wanted to study the ways in which the way we think scales up to the way our cultures work. At the time, we did not have access to a psychological lab.
Simon: We couldn’t do tests. Mechanical Turk was pretty young. I ended up turning to large scale archives. Things taking around in places that historians had to look as opposed to psychologists. From there, we ended up working mostly in texts because we don’t usually keep very structured records of stuff that happened in the past. We tend to write down things that people say. The lab today, we work on a range of projects. Some of these are contemporary. We look at online behavior. We have a big project now on the men’s seduction movement with Chloe Perry. We have a project on the evolution of mathematical theorems.
Simon: In fact, now mathematicians are able to prove theorems essentially by running a computer program so we can look at the syntax rates. Out of that, we understand something more about the proof that the square root of two is irrational. We work with some really big and deep data sets. A big paper that we had last year was on the French Revolution. People kept records and transcripts of the debates that were held in that parliament. We’ve looked at the long time evolution of the criminal court system. The bureaucracy of the court is much younger than the need to actually solve our problems to punish people who do wrong and to readdress things in a way that keeps society going.
Simon: In fact, most of the way we did justice certainly before let’s say the 20th century was by talking them through. In each of these cases, we’re looking not at a spreadsheet so much as a set of manuscripts. Those manuscripts can get very, very big. In the case of online behavior, we’re running a model right now that has 100 million bulletin board coasts and close to half a billion words. French Revolution, that long ago, 40,000 speeches, hundreds of thousands of words. You might say we do data science but a lot of what we do actually is much more focused on the cognition of that process. The way people think and how the way people think scales up into the way all through things.
Jim: Extremely interesting. My professional life back before I got involved in science, most of my projects were involved in texts also. We were always in a say back in the ’80s, in a small corner of technology. People dealing with texts. I wrote a couple of full text search engines back in the day. I wrote an automated spell checker in the early days, really crude automatic tagging systems for my topic, etc. Only in the last 15 years has text processing really come into its own.
Simon: Yeah. I think 2010 maybe we had algorithms that came out and the early 2000s. Famously something called topic modeling. Just the ability to use a computer to process a large archive when it’s sitting on your desk really change things. You can interact directly with some of the objects that come out of those algorithms. It’s funny. In the end, when we want to get something done, we usually come and talk it over. There’s a reason there’s the financial district. It’s not just the light speed travel time of a market trade. We think of ourselves as computers but I think we’re more talkers than calculators.
Jim: That’s right. We’re not very good calculators because we know $1.95 calculator can beat our ass.
Simon: Exactly.
Jim: So far, nobody has been able to automate this course in a way good enough to fool another person.
Simon: I think machine has trouble making a restaurant reservation.
Jim: Yeah. The other one not quite linguistic but it’s procedural is the Wozniak test or artificial general intelligence. Steve Wozniak, the Apple guy proposed, which I like, which is to be able to take your robot. Plunk it down in any random kitchen in America and have a relatively seamless, figure out how to make and actually make a cup of coffee. It’s something most humans could do without any great distress. It’s the kind of multi-variant pruning rule problem that if a robot could do that, we’d say, “Pretty damn smart.”
Jim: Quick question on the text process. Do you work with the linguistics technology people? Carnegie Mellon has probably the best or arguably the best or number two department.
Simon: I would say the best, Jim. LTI, Language Technologies Institute is a fantastic place. You might say they build the jets and we fly them. We’re much less interested in the high tech side. A lot of our algorithms that we use are maybe even 10 or 15 years old. That’s something you want. If you run a chunk of ration from 1890 through a machine, you want to have a pretty good understanding of how that algorithm works. When I watch what comes out of those places, super exciting. At the same time, untested. If something comes straight off the assembly line at the Skunk Works, you’re not going to take that into battle just yet. We work with a B52 much more often.
Jim: That makes a lot of sense. Of course, that is the problem with a lot of state of the art machine learning. So much of it is now neural nets. Hence, the black box. It doesn’t have the visibility that the old fashion AI and linguistic processing, which frankly, at the end of the day was a bazillion if then else rules. We no longer have the ability to even look at the code and say what it’s going to do.
Simon: This is a big problem. It’s problem for science. For example, if you run Russian through one of these algorithms, Russian’s a highly inflected language. We don’t really know what its representation is internally when it’s splitting up the other side instead of semantic claim. It’s a problem for science. It’s also a problem for policy. Machine learning is incredibly powerful. It can help us make better decisions. In the end, we want to be able to ask the people who govern us the why questions. Why didn’t I get a mortgage? What’s your problem? Why did you pull me over? Our machine learning algorithm, it can’t give us a straight answer.
Jim: At least on neural net machine learning algorithm.
Simon: Yeah.
Jim: I always want to keep this clear. At the moment, the field is highly concentrated on neural nets. Personally, I believe it’s over concentrated in the same way that particle physics was over concentrated on string theory 10 years ago. I’m already starting to see in my conversations with AI people that there is a renewed interest in things outside of deep learning, deep neural net, neural net reinforcement learning paradigms. A lot of people when they say machine learning now are automatically linking it one to one with black box neural net type systems.
Simon: Yeah. Gary Marcus is a great proponent, I guess we still call it good old fashioned AI, symbolic representations. We’ll probably get to this in our conversation, Jim. This is perhaps the difference between biological evolution and cultural evolution. We’re really good, we’re getting better now at building machines that can do things that we evolved to do hundreds or maybe millions of years ago. Recognize objects, avoid obstacles. We’re not so good at the kinds of things that we think are hallmarks of our species. We’re not so good at programming machine to reflect upon itself for example.
Simon: We’re not so good at programming machines to communicate to another machine for example. In a way that doesn’t require a set of protocols and a really precise specification of the inputs and the outputs.
Jim: I like that. If I heard you right, if so, it’s the first time I’ve heard this suggestion. That perhaps neural AI is analogous to biological evolution. Symbolic AI might be analogous to cultural evolution. Did I get that right?
Simon: Yeah. It’s pretty hard to learn to drive a car. It’s something we learn maybe in our teens or in our 20s. There are things that children do well before they can drive a car very well that a machine still can’t.
Jim: The thing is though I’m riding a bicycle.
Simon: That’s an interesting one. It’s a set of feedback mechanisms much stronger than the mechanisms you’ll find in design system like a car.
Jim: Interesting. Interesting. Yeah. I’ve also been involved with some symbolic AI of late for the last several years. I’ve been affiliated to a greater or lesser degree with the OpenCog project. Ben Goertzel and his crew and they’ve been working for many, many years on a symbolic form of AI. It’s open and they can also include neural net as inputs. Essentially, the symbolic components can orchestrate non-symbolic component.
Simon: Some scientists what we’re trained to do when we grow up, we have instincts. We have things that work very well in a world that is no longer around. We go to school. They teach us a couple of symbolic representations that help us get through.
Jim: Exactly. In fact, the model I’ve been using, think about how this intersects your cultural versus biological. The neural net components approximately congruent with perception. That symbolic components are more or less analogous to tenement’s system two type thinking. The system one type thinking is something in between, which is a little bit less clear to me.
Simon: We’ve learned a lot about system line from deep learning. I think that’s right. One of the things that’s quite amazing is when you pipe an image or a video through one of these neural networks, the kinds of things that does on the way to producing an output is very similar to the kinds of things that happen in the visual system and the layers that take a retinal image and pump it all the way up into an interpretable one.
Jim: I’m very happy to see just the last year and a half more and more people expanding their perception of what machine learning can be and starting to put a little bit more emphasis on some of the neural net components. At the end of the day, my hypothesis is it will be both as always. What’s the answer in social science? Both, right?
Simon: Both and everything.
Jim: Yeah, nurture and nature. People get so tied up in nurture versus nature. Say, you idiots. The answer is obviously both.
Simon: We wouldn’t have the concepts, Jim if they weren’t constantly relevant.
Jim: That’s probably correct. All right. That’s a wonderful introduction. Simon’s a very interesting guy if you can tell. Next topic we’re going to go on to is something that I think I’m going to make a feature of this show for the guests that have a relevant background. In addition, too is very interesting work in data and society and cognition way back yonder. Simon was an astrophysicist. Guilty, right? I’m going to have him talk a little bit about The Fermi Paradox. For those of you who don’t know about The Fermi Paradox, the name Fermi goes back to Los Alamos where they’re making a nuclear weapon, first bomb during World War II.
Jim: A bunch of really smart, famous guys were sitting around the lunch table speculating on how many intelligent species there were in the universe. Fermi, what the hell was his first name?
Simon: Enrico.
Jim: Enrico Fermi comes up to him and says, “Okay. Where are they?” Ever since, this perspective has been called The Fermi Paradox, which is under some reasonable sets of assumption. You would assume there are many, many intelligent species probably more intelligent than humans because they could be around a lot longer out in the universe. Yet despite our looking for the last 60, 70 years with growing levels of sophistication, we have seen absolutely no sign of them. There’s a long list of thoughts and they fall basically into two big categories.
Jim: One is they’re not out there for some reason, that’s not obvious or maybe is obvious if we think about it. On the other side, they’re out there but they’re not showing themselves or they exist in a way that’s not easy for us to detect. I’ve been fascinated with this question even before I knew the name of it since I was probably 12 years old. I read everything I can on the topic and try to keep up on it. I thought Simon would be a great guy to give his thoughts on The Fermi Paradox. Where are the aliens? Aren’t there any?
Simon: I think you set me up pretty well here, Jim. Because there are two large schools of thought. For those who think there’s no one out there, Anders Sandberg has a really terrific paper where he takes essentially our best after physical knowledge, sticks it through a machine. The other side claims there’s a 15% chance given our best knowledge that there is nobody out there, at least not in our galaxy. Anders takes the Drake Equation and just tries to factor in our uncertainties. Our uncertainties are extremely large. They’re not a bell curve distribution. It’s not that we know these facts plus or minus one.
Simon: We probably know them plus or minus a factor of 10. That really changes. Once you understand how weak our knowledge is of even something like planet formation, you can get a lot more skeptical about the idea that they should be out there. I think they are out there, Jim. I think they’re actually here in a funny sense. We know that if you stick an organism in an environment and have it evolve, we’re pretty confident that it’s going to look reasonably similar to the stuff that we have around here. The visual system let’s say of a bird is at least in its mathematical properties very similar to the visual system of a human being.
Simon: We were just talking about very similar to the visual system in a self-driving car. At that level, I think that they’re around. If they were around, we would be able to spot them pretty clearly. Once you go to the symbolic level, once you ask who are we as thinking species, you can’t build a faster and drive with a deep neural network. I think you and I will probably agree on that. Once you evolve the symbolic representations that you need to get things done, all bets are off. When a machine or a human being or any kind of creature can reflect upon itself, what comes out the other side is completely unpredictable.
Simon: The minds that we would encounter would be to my mind, things that we would have a great deal of trouble recognizing. I think we would have a great deal of difficulty recognizing the very things that we’re looking for. If we could go to a planet with life on it, I think we’d spot it pretty quickly. We have an [inaudible 00:15:28] equilibrium system. There would be something like oxygen. There’d be a set of batteries probably coming out of their nearest star. If we look for the symbolic products, which is what we do when we turn on a radio telescope and we run something like the search for extraterrestrial intelligence, we look for radio signals.
Simon: We’re looking for signals that would be created by something similar to a human mind. If the monkeys as David Gritten would describe, if the monkeys got uplifted, if the dolphins got uplifted, we would probably be able to spot it. I think if we saw a communication from a living organism that had evolved beyond our planet, I think it would probably drive us mad. My guess is we’re running some pretty good filters to keep us outside the realm of being able to see them. Psychologists wonder a little bit about this. Why do we have such a theme of alien abductions? You go back further and further, these are essentially religious experiences that people have.
Simon: Those experiences tend to drive people a little bit around the fence. Like I said, I wouldn’t be surprised. When I was at Arecibo, we had SERENDIP, the Search for Extraterrestrial Developed Industrial Populations. This is a very ambitious project. My guess is we’re getting a lot of signals from the guys out there. We just don’t quite know what to look for.
Jim: Interesting. I’m surprised that there aren’t mathematical techniques to be able to distinguish noise from signal irrespective of how it’s encoded.
Simon: This is a great one. Cris Moore at the Santa Fe Institute, this was a paper I think he wrote when he was quite young. The most efficient encoding, the best way to get data from one place to the other given a particular transmission power is also the maximum entropy distribution. What that means essentially is that if somebody’s really good at communicating, it’s going to look like a warn label. There’s a lot of inefficiencies in the way in which we transmit information right now. If we’re really good at it, if we’re able to compress what we want to say efficiently, out the other side, it’s actually going to look purely random.
Jim: Interesting. Just thinking out loud here. Suppose a society is a billion years old. It essentially codified huge amounts of structures of thought. Essentially, communication is really sending codes to a look up table on this vast tree structure of thought and knowledge. Codes do look exactly like noise if they’re well-designed.
Simon: Exactly. If there’s any pattern of communication, you should be able to use the pattern to make the communication shorter. If you have a repetition, you don’t repeat. You just make a notation.
Jim: Yeah, particularly if the society’s been old. Think about if we were able to codify everything we knew over a billion years. Everyone in the galaxy shared this semantic net. Probably not too much because it hadn’t been explored in a billion years other than the re-combinations of things in the trades.
Simon: This is a great point. One of the things that comes up when people talk about The Fermi Paradox is the incredibly short time scale of cultural evolution compared to biological evolution. It took us a long time to get from the sponges, this is something David Krakauer talks about. We spend a long time as sponges before we really got our multi-cellular act together. We went from a lifespan of 30 years, no written language. No cities, no culture up to where we are today, mastery of quantum mechanics, mastery of the electron. Not even the blink of an eye, it’s a time scale on which biological evolution has almost no handle. Maybe lactose digestion.
Simon: The Fermi Paradox says, “Look, if you pick a civilization at random, it’s really unlikely that they’re going to be in this very early childbirth phase.” Chances are, they are millions of years old. We have trouble predicting the next 10 years. I think on the one hand, people say, “They’re so old that they’re flying around the galaxy and work drives.” That may be true but the thing that’s just as old is not their technology but also their culture and their minds. We have a lot of thermal sources sitting out there in the galaxy. We have a lot of things that look like warm stuff. Maybe some of those are broadcasting.
Jim: Interesting. As a typical 12-year old nerd, I of course assume they had to be out there. All along, I always assume they had to be out there. Until I started seriously researching The Fermi Paradox and the arguments on both sides. I’ve refined this idea in about a four-hour conversation with Stuart Kauffman one time, which is obviously familiar with the error catastrophes in evolution. When error rates are too high for our audience. In an evolutionary system, you copy things into the next generation. You copy them with a higher or lower level of finality. Our current DNA system is very good. It’s only a few errors per billion.
Jim: Until you get to that rate or something close to that rate, evolution doesn’t work very fast. You can’t evolve a very complex structure with a high error rate. It’s called the error catastrophe. I don’t remember the exact directional level where you have to get to before evolution to take off. It’s pretty high. This conversation with Kauffman was, “All right, Stuart. I buy autocatalytic networks,” which is Stuart’s theory on the pre-evolution of life. That network can replicate themselves and become an outstanding chemical wave essentially.
Jim: I said, “How do you go from low fidelity copying to this massive machinery we have in our DNA to have high fidelity copying, which then allowed evolution to really ratchet itself relatively rapidly?” As far as we know, all the life that we know about today, all the way back to Archaea, the most primitive bacteria. They all use high fidelity copying DNA system. How did we jump from autocatalytic chemistry, chemistry in a warm pond to this very complicated machinery? Maybe it’s exceedingly rare.
Simon: There is filters we talk about. There are big, lucky accidents. In the modern era, some of our big, lucky accidents are of course cultural. The Cuban Missile Crisis ended with a whimper, not a bang. I think of evolution as a conversation among a bunch of organisms. We’re all sitting around the table. We’re talking some of the ideas are better than others. Maybe one way to think about the error catastrophe is how do you keep that conversation sufficiently on track? If you keep it too much on track, if the ideas aren’t coming, if people are simply rehearsing and repeating what’s come before and nothing happens.
Simon: If that group starts moving too quickly, it becomes chaotic. You leave the room at the end of the day and you’ve got nothing done. There is a balance. There’s a balance between change and comprehensibility. There’s a balance between making sure your body plan looks roughly like my body plan. The old joke is if a cow gives birth to a calf and you’ve never seen a calf that looks like that before, it’s not going well for that calf. We need some stability but we also need evolvability. We need the chance for things to change. That change is always stressful.
Simon: Most mutations are horrific events for us. Once in a while, someone takes a hit for the good of the species so to speak. At the other side, we get something new.
Jim: Great. I think that’s it for The Fermi Paradox. We’re not going to solve it today but very interesting topic. Let’s pick up this idea of noise versus coherence. Let’s talk a little bit about social media. I know you’ve done some work on blocklists. If you could tell us what blocklists are and how they relate to social media. Maybe some speculation about they’re good or they’re not so good.
Simon: We’re 99% speculation lap and that maybe once every 100 days, we actually get something done. Social media just on the biggest scale, Jim is evolution sped up by factors of hundreds, tens of thousands. That’s happened very, very quickly. Wired Magazine, I was a junkie for this stuff in the ’90s. Wired Magazine, it puts out an idea. William Gibson writes an article and you read it and it blows your mind. You do something with it. In order to get the feedback into that system, if I want to take Gibson’s idea and do something with it, let’s say I read a letter back to Wired, the magazine comes out. I go to the store. I buy it. I write the letter. I mail it in.
Simon: Wired decides to pick it up. They write something. It takes about a month, let’s say. It’s time scaling publication. We’re now down to time scales a minute. If William Gibson’s on Twitter and he says something, I can hit right back as soon as I see it, 53 seconds after he’s written it. If you look at what the time scale factor, by how many factors have we sped up cultural evolution to go from a time scale of a month, to go to a time scale of a second or a minute? If we had sped up biological evolution by the same amount, we would have gotten from the dinosaurs to us in about 5,000 years. I don’t think we’re ready for what’s unfortunately or fortunately already happening.
Simon: We’re well beyond the kind of world that we were used to in the 1990s. Social science has changed. We can look in the past and what we see is a movie on incredible slow motion. When I study social media, I feel a little bit like somebody studying the DNA replication mechanism. In order for an organism to make a copy of itself, it has to do all sorts of fun things and all sorts of unpleasant things. That happens nine months, six months, 12 months depending on the species. Nowadays, not only does it happen faster but we’re developing new mechanisms. We’re inventing new ways to make those copies, new ways to keep those copies faithful, new ways to make those copies to break them and reform them.
Jim: A comment here. When we go back and compare Wired to Twitter, let’s say. You write your letter back to Wired, something is very different in that model than the current model, which is there’s a curator in the loop.
Simon: There’s a filter, yeah.
Jim: There could be 100, 200 letters sent to Wired. They might publish one. On Twitter, everybody gets their say, good, badder and different. There is much less curation going on at least in the raw sense. Perhaps what we’re searching for are emergent bottoms up curation tools that allow us to mold the social media environment. It is more positive than it’s been today.
Simon: That’s right. You might think of Wired as an incredibly well-engineered multi-cellular organism. There’s a head. There’s a stomach and accounting. In some sense, we’ve devolved. We’ve broken apart some of those institutions. We’re much more like a soup. That soup ran from let’s say the Usenet era probably into the Reddit era. This gets to your question both directly after a long detour. What we’re seeing now I think is people rediscovering mechanisms to produce stable institutions that last 10 years and perhaps even 100 years. For a while, for example, we worked on Wikipedia. Wikipedia is as probably with the financial markets.
Simon: Wikipedia is something that works in practice but not in theory. We don’t have a good theory about how they created an incredibly stable and largely functional institution. It has the same problems that any government has. It has unfairness. It has rebels. It has protests. Wikipedia probably was the first example of the post soc era for social media, for social interaction online. We had maybe early ’90s, people imported the structures from the institutions around them, which were mostly university institutions. People are online because they were at a university. They got their account there. AOL comes along and the system completely decomposes.
Simon: Those of us of a certain age talk about eternal September. The day when the new people came and they never stopped coming. Wikipedia was able to handle that new era. We’re very curious about essentially what we think of as a common law system that emerged. It’s not a Napoleonic, top down “civil”. It’s a much more informal way of resolving disputes but not just resolving disputes. Keeping the system running, finding ways to provide restorative as opposed to retributive justice. Okay. Let’s go on to Twitter. Twitter is in some sense, even more difficult to run than Wikipedia. Wikipedia always had a goal. No one was quite sure what that goal was.
Simon: They all knew they had a goal. Sometimes, they fight over it. It’s not quite clear what Twitter’s goal is. Out the other side, people have produced mechanisms for forming societies, for forming groups that are reasonably stable. Even in the absence of an idea where this is all going. One of the things that we noticed and this is partly just being online but it’s also partly talking to people what we call qualitative research. What we noticed is this new phenomenon of bottom up censorship. We’re used to let’s say Facebook or Twitter cutting people out of a conversation. They’re corporations. They have to do it.
Simon: You can’t have a free speech zone online if that company also has to answer to shareholders and to the people who could sue them if they have to answer to the advertisers. It’s just not possible. It’s not the Agora. It’s something far more like a café. If you go into a café and start saying some weird stuff, you’ll find yourself kicked out. We’re used to the top down censorship. What we started to notice was the creation of what are called blocklists. On Twitter, if you and I really get into it, at some point, you might decide to block me. Blocking is the individual level version of disappearing.
Simon: You’d tell Twitter erase Simon completely from my experience. I never want to see him. I never want to see people reply to him. I never want to see him re-tweeted. I don’t want to see commentary on anything he’s saying. By the way, if he wants to talk about me, game over, too. He can’t see me as well as I can’t see him. This is something Pascal, a Twitter friend of mine talks about. He’s a classicist. The Romans did this. You do something bad enough, they scrub you. You can’t mention their name. Any document, any statue of them is done. You could do that on an individual level.
Simon: If you have people who think the way you do and who are equally irritated by me, they might also choose to block me. What we saw merging were the shared lists, these shared blocklists that people were creating. Let’s say I am a member of a particular community. I don’t know. Let’s say I’m a hardcore democratic socialist or America guy. Let’s say I’m a hardcore Jordan Peterson, Joe Rogan type. Instead of going through the trouble of eliminating from my life the people who disagree with me in a way that I find upsetting, I can look at what my friends are doing. I can look at the people who are blocking them. I can essentially say, “Look, give me your list. I’ll just copy it over.”
Simon: Now what you have is not an individual level decision. In fact, some distributed curation of a filter bubble. We knew Facebook was creating these. We knew that Facebook in order to keep you online was presenting you with information that you wanted to encounter, the friends you agreed with. We didn’t like it. We thought this was a problem. What we see now is people creating their own bubbles. They’re creating their own bubbles by consensus. They’re not just banning people. They’re giving other people the opportunity to ban exactly the same group. They’re forming in some sense a negative community. Negative in the sense that they don’t know who’s in it.
Simon: They certainly know who’s not in it. I don’t think we know yet what those blocklists are going to do. We have science fiction stories of this. We have cases where a community shuns someone. Puts a Scarlet A and kicks them out into the forest. Here we go. Let’s see what happens, Jim. Let’s see what these distributed village style mechanisms will do when they happen not on the scale of a Hawthorne novel but on second time scales.
Jim: Interesting. First, I’d like to remind people that the online world didn’t start in the ’90s.
Simon: Thanks, Jim.
Jim: I actually worked for a company called The Source in 1980, which was the very first consumer online service. We had essentially everything that’s on the web today. A 300 baud character mode only, $10 an hour.
Simon: Just about as fast as you can type.
Jim: Yeah. We quickly had tens of thousands of customers and eventually hundreds of thousands because there was nothing else like it in the world. Soon, Copy Serve came along. It’s just useful to remember. Someone should write a book on this. The prehistory of online because there’s actually several generations before AOL even. Protégé.
Simon: Delfi I think I remember.
Jim: Delfi.
Simon: Yeah, Delfi. Yeah.
Jim: Yeah, yeah.
Simon: Jim, this is great. It’s a great story. I remember I gave a lecture on Wikipedia. I had a young guy come to me who had collected a lot of the old bulletin board systems. In fact, these were things you dialed up. They were the local things we remember from the war games era. They’d even developed a packet switching mechanism. All the systems would dial into each other and share things. A little bit like Usenet, no centralized server. One thing I’ll say and this is for you and your colleagues who were in that world. We’re losing all those archives.
Simon: That stuff is disappearing because when a company shuts down, there’s really no incentive for them to spend any money, creating an archive, saving it to disk in a sustainable way that we can look at. We’re doing work right now even in the ’90s looking at the early Usenet postings just to get a handle on the origin of some of these online seduction communities, these men’s seduction communities. We can’t get stuff from let’s say 1993, 1994. Some of it’s held by Google. We have some guys who maybe shouldn’t be scraping Google the way they’re scraping it. This stuff is disappearing in a way that stuff didn’t disappear in the 18th century.
Jim: That’s very interesting. I have a very interesting example.
Simon: Very good. Hit me, yeah.
Jim: I’ve been an active member of probably the oldest surviving commercial online community called The WELL.
Simon: Yes, The WELL.
Jim: Started in 1985. It was eventually a spinoff of the whole earth reviews. Stewart Brand, Larry Brilliant, that crew. I’ve been a member since ’89. It’ll be my 30th anniversary this year.
Simon: Wow.
Jim: Until recently, I was on the board helping them keep their shit somewhat straight. Anyway, somebody recently suggested that I talk to the British museum who was saving these old online services. Unfortunately, I reached out to the archivist and he said, “We only have a remit for British sites.”
Simon: Yup.
Jim: I need now to think about probably Library of Congress. Because the world is continuing to dwindle. It’s becoming the colonial Williamsburg of the internet. It’s still a very interesting place. It has the highest quality discourse but it’s only a few thousand people, a couple thousand people now. Eventually, actuarial science will take its toll. It may well close someday. It’d be a horrible shame if The WELL archive is not preserved.
Jim: There’s some very interesting things that came out of it including the EFF, Wired Magazine, Craigslist or all ideas that were brooded on The WELL back in the very early ’90s when it was almost the reactor. Where ideas about what the internet was going to be, ’90, ’91, ’92 were happening.
Simon: Actually, an old colleague of mine, Jamie Murdoch, we work together as academics. Jamie spent some time with the internet archives. There are some non-profits that run this stuff. A big chunk is you put something on a CD-ROM and it’s gone in 15 years. Internet archives has the servers has the reliability that you need. Even the curation is hard. It’s really difficult to scan through a million things threaded together with some adhoc database system that was invented by a guy in between hanging out and the mission. I was on the East Coast in the ’90s. We had the dirtier, meaner, grittier, grungier version of The WELL called MindVox.
Simon: MindVox was a local system to the New York City area. Around about 1996, I logged on to MindVox and there is just a message on the screen that said, “We’re shutting down. Good luck. We’ll refund any stuff that you’ve sent us pro rata.” As far as I could tell, they got into some pretty heavy drugs. The discussion boards we had there, we had the iron cross symbol of tyranny, which is what happened when you posted something that was too insane or too crazy even for MindVox. As far as I know, that world and these were some of the most interesting intellectuals in New York but also some of the craziest drugged up people in New York.
Simon: I think we’ve probably lost that forever. I’ll just tell you a story about this, Jim. On my wall, I have a pamphlet from The French Revolution. This is a piece of paper that stuck around since 1792. Picked it up on the banks of the Seine. Paper sticks around. I can look at it. I don’t need a special interface to read that thing. We don’t have the kind of interpretability and readability that we got from paper. A colleague of mine, Tim Hitchcock says, “Look, 20 years, 30 years go by, we’ll have a better way to preserve this. These are the Babylonian tablets. They’re getting broken.”
Jim: Yes. It’s something worth thinking about. I’m going to give some more thought on reaching out, see who might preserve The WELL. Interestingly, the reason I thought about The WELL is we had a precursor of blocklists. It’s got a web interface but the underlying mechanism is a command line unit system. About 20% of the users still the command line. I’ve almost thought it’d be funny to offer a channel on twitch to show kids today people actually doing work on a command line.
Simon: Logging into pine.
Jim: Using a thing called PICO span, which is what The WELL actually runs on. The web thing is a magic fingers beast on stop of PICO span. Anyway, because it was written in this craziest way, users could augment and add new functionality. One of the things that was developed probably in the early ’90s, ’92, ’93, I can look it up and find out when exactly is called bozo filtering. You could create a file in your directory called dot bozo, I think it was and you put the username. Everybody has up to eight character username, old fashioned thing. No spaces allowed. My username is memetic on The WELL, M-E-M-E-T-I-C.
Jim: You put somebody’s name in the bozo filter. When you’re reading essentially the forms, it will say blah, blah, memetic data, time, day, etc. In line, it will say, this comment has been bozo filtered. It’s not a complete make you disappear but it makes your content disappear in the context of the discussions. You can un-bozo people. You can bozo them as we would say. It was actually quite funny. I will say I took the position from early on. I will never bozo anybody. I never did.
Simon: All right. This goes all the way back to basic psychological drives. When we talk, we want to be heard. You look online. When people discover that they’re blocked, it’s a painful thing. If you’re a purely rational guy, somebody hates me so much they don’t want to hear from me. I probably don’t want to hear from them either. Screw you, too, buddy. There’s something really deep in us, the need for recognition. This is the Hegelian story, the story of history is the story of us figuring out when it is to recognize each other in increasingly sophisticated and sustainable ways. We had the bozo filter at MindVox. It had a darker name. It’s called the Kill File.
Simon: I wonder, Jim. The thing about the blocklist is that they can circulate.
Jim: Yeah. That’s new. Interestingly, even though I follow this stuff fairly closely, I did not know about blocklists until I looked on your site and saw that you had done some research about it. This is a whole new level of evolution. That these things are replicatable. They’re memetic. What does that mean?
Simon: If blocklists keep going, we said this about by laws revolution. Most ideas die. There’s a reason they didn’t do it that way. The idea of blocklists especially since you tell me that this has been running in The WELL since the ’80s, this might end up being a permanent feature of the online world. If the online world is where culture is happening now, this may be a driver and maybe something that fundamentally influences the way in which we create ideas and the way in which our culture evolves.
Jim: I will say my initial reaction to it was I find it preferable to the top down censorship. You said yes, we have to have or we will have censorship on the platforms. Yes, at the limit but as was my own gut level reaction that we ought to encourage it to be as little as possible. Again, this is a huge tension right now. I’m in the middle of a number of Facebook arguments about this. My own list is we should ban calls to violence against anybody, either personal or by group. Platforms should ban discourse that is in bad faith. Now this is unfortunately difficult. If a person posts something they know to be untrue but they do it anyway.
Jim: They don’t do it satirically. The Pelosi thinking fallen into bad faith. There are those who re-tweeted it or posted a link on Facebook may not have been bad faith. It’s very complicated. Third is misrepresentation of who you are. This gets into the vaster question of should there be anonymous posting or not? Twitter being a cesspool of anonymity. Facebook, at least attempting to produce an emergent network identity of some reasonable finality. Of course, they’re always fighting infiltrators. The fourth, this is again tricky.
Jim: Popper’s paradox of tolerance, which is perhaps we should encourage the banning of people who want to suppress free speech or to reduce the civil rights of identifiable groups. Other than that, I would be in favor of tolerating pretty out there stuff.
Simon: It’s an interesting one, right? The American legal tradition, Sullivan vs. New York Times. We have a sense that militias wise are grounds for a suit. We have sometimes a good faith exception. I spent time in Britain. Britain does not have a robust free speech tradition. You can get arrested for saying things that are racist. You can get arrested for saying things that are homophobic.
Jim: The official secrets act. There’s a whole lot of things.
Simon: Yeah. You sign them on a piece of paper. At this point in time, I find Britain to be an increasingly uncomfortable place to think and to speak. It hasn’t stopped anything. It hasn’t prevented the emergence of really extreme views and really extreme expressions of those views. What it does do and maybe this is just me being too much of an American. It has what we call a killing effect. Even the knowledge that somebody out there is evaluating what you’re saying is something that prevents you from thinking clearly. It’s something that prevents you from exploring ideas right out to the edge. The distributed nature is tricky, too.
Simon: This is something that happens much more in Britain and the United States. Although it happens in the most British places like New England liberal arts colleges is Deplatforming or no platforming. This is something where people show up and scream and bang drums and chant to prevent a speaker from giving a lecture. People that you and I would think of as radicals but reasonably innocuous radicals in the sense that no one’s going to kill anybody because of what Camille Paglia says. No one’s going to kill anybody because of what Germaine Greer says. Yet both of these people have been no platform or at least in the case of Paglia, people have attempted to no platforming.
Simon: You can’t stop an army with a law, very well at least. If there is going to be some consensus at least within groups of the limits of speech, it’s going to have to be something that it self-evolves. The paradox of tolerance is something that sits out there. Even just getting the tolerance is a huge achievement for a group.
Jim: I must say I am very disturbed by the Deplatforming movement. The high, high irony that perhaps the hottest Deplatforming center is Berkeley. The home of the free speech-
Simon: Free speech movement, yeah.
Jim: Free speech movement. Again, these people you might disagree with them but-
Simon: It’s funny. It’s left and right. It’s like getting the flu. It’s unclear what you have to do wrong to end up in a situation where you can’t speak on the campus of a state university. How bad it has to be? I think people are usually surprised. Jon Ronson had a great book, So You’ve Been Publicly Shamed. We have not just the big figures, not just the people who’ve written books. Paglia by the way came out. She was on Salon for many years. That came out of The WELL. It was one of your spinoffs. People who are far less important in the are you going to make the New York Times sense, also fall victim to this.
Simon: It’s hard to say where that impulse comes from. Maybe the better way to say it is we have that impulse. It’s hard to say why it’s been weaponized and magnified in the way we see online.
Jim: I think it’s extraordinarily dangerous and gets us inching towards a possible civil war. Literally, you start trying to shout down someone like Ben Shapiro or Charles Murray, whether you agree with them or disagree with them, they’re within the context of reasonably serious speech. They’re not threatening anybody. They’re not operating in bad faith as far as I know. The other side starts to apply the same rules. Where does that end up? It ends up with blood in the streets. In fact, I wrote a short story called Blood in the Streets. You can find it online.
Jim: Maybe in the account under Jim Rutt. It’s actually two linked short stories about a series of backs and forth between Antifa and Trump supporters that essentially ends up in the beginnings of a civil war. It could happen.
Simon: Back in the day, we had a pretty clear dividing line between north and south. Here in Pittsburgh, I used to teach at Indiana University. The geographical separation of the people who would start shooting each other is on the order of yards as opposed to hundreds of miles.
Jim: It’d be more like the Syrian or Spanish civil war. It would be like the US civil war. The Spanish civil war had at least 10 factions. The Syrian civil war, many more than that.
Simon: You like to think that Americans are too sensible. We’re as a country, I think ideologically pretty weak. Of course, it’s what drives the people on the far left and the far right nuts. We’re not so quick to sign on to ideas. We’re pretty quick to sign on to ways to exploit people economically. Those go pretty far. Of course, they go back to the founding. I think this one is hopefully going to be more comedy than tragedy.
Jim: That would be very hopeful and very good. All right. Let’s move on to our next topic, someone Simon and I both know reasonably well. Murray Gell-Mann, one of the great scientific minds of the 20th century recently died. I think it was last week. Just an amazing character. I showed up at the Santa Fe Institute in 2002. Murray at that time was probably 70, something like that and was still probably the second smartest person I had ever met.
Simon: You’re killing the game. Who beats Murray?
Jim: Eric Schmidt. Eric Schmidt.
Simon: All right.
Jim: This is at age 70. I’m not talking about weighted for age. I’m talking about absolute smart. If you know Eric Schmidt, I tell people, “You want to know what Humans 2.0 might look like? Go talk to Eric.”
Simon: Eric, if he’s hearing this, he’s squirming in his seat. Because these advanced mental technologies also come with a high level of humility in his case. A lot of us, we were very sad to see Murray go. I showed up at the institute, Murray was sharp as you ever wanted somebody who not just won a Nobel Prize. If there was a Nobel Prize for Nobel Prize, I think Murray would have had a pretty strong shot at it. There’s the classic Apollonian and Dionysian split. We have Richard Feynman on one side. We have Murray Gell-Mann on the other.
Simon: In terms of people vying for the image of the particle physics, the physics style, hard science personality, they were both very larger than that. I think we underrate Murray as just in terms of pure personality. Feynman, he played the bongos. He picked up women. He was everything that you wanted out of a mad man character. Murray was an extremely cultured guy. Feynman was somebody who essentially thought culture was literature, art, music, of any level of refinement was fundamentally feminine occupation. He played dance with the daisies. He couldn’t handle it.
Simon: Murray was somebody whose ability to think through and appreciate art and literature, the products of the human mind was I would suggest as much off the scale. Murray was somebody who was involved in the MacArthur Genius Grant for many years. He was one of the first members of the board. You and I both know him, Jim. Cormac McCarthy was one of the first recipients of an award that changed his life. He had written Suttree. His greatest work was in the future. Without question, Murray was involved in that decision. We never know. These guys are pretty secretive about it. To be able to spot a talent like Cormac McCarthy that early, Richard Feynman couldn’t have done that.
Simon: Murray had a vision that along artistic lines, along human lines, that I think I’ve never seen a scientist with that level of ability. We love the wacky Feynman character. We love the fact that he had a really thick Brooklyn accent. Murray can rub people the wrong way. He used to rub people the wrong way because he also feels a great deal about your mind as well. Murray I think is due for reevaluation in the coming years.
Jim: Yeah. I was always amazed at his breadth. Yeah. As a dinner companion, he could talk on almost any topic, I should say both cultured and erudite at the same time. To the degree there’s a difference between the two. The other interesting thing about him unlike some other characters like that, he was happy to admit he was wrong. On three occasions that I was having a discussion with Murray, he somehow wandered into one of my patches and made some strong statement. I took him through the thinking and the logic and the evidence. He said, “You’re right. I was wrong,” three times.
Jim: Murray’s sort to do that and he really was intellectually honest in a deep and fundamental way as well as unbelievably bright. There’s almost no topic you could bring up that Murray didn’t have a reasonably informed perspective on. Surprisingly large body of information-
Simon: Murray got defined lines. Forget it. You’re done.
Jim: I wouldn’t waste my time on that.
Simon: I think in the modern era, you look at people online. You look at people and their favorites. I think the basic split here is between debaters and thinkers. It’s really fun to watch a really good debater just crush something. If you are good at debating, you’re good at defending a set of ideas that may be wrong. The way to lose a debate is to say, “Well, Christ, I hadn’t thought of that. You might be right.” That’s not a debate move that’s going to keep you on a podcast. It’s not a debate move that’s going to keep you on a TV show. Murray was not a debater. He was not somebody who’d come in with an idea and walk all over anybody. He was a thinker.
Simon: He was somebody who is great to have at a dinner party, great to have on a long hike. He’s not somebody who would have the 50 million follower account on social media.
Jim: I think that’s probably correct. I did have one item that I was aware of that he was very doctrinaire about, which is hatred of mathematicians. That came from I don’t know because obviously, the [00:53:59] was an extraordinarily talented man mathematically. In any kind of discussion about mathematicians, he’d throw a fit.
Simon: That’s one thing he and Feynman had in common. He developed the mathematics behind the structure of a proton. He discovered, invented, whatever you want to say. My understanding was he used to check on mathematics, Lie algebras it’s called. My understanding is Murray just had to dig that out of scrolls of ancient wisdom. The mathematicians, the way they deal with the objects that physicists love is at a profoundly bizarre and pleasant way. I was a physics major in college. Being a physics major, you think you know everything. A bunch of us took the most advanced mathematics class that we’re allowed to take.
Simon: Math 55 and we studied what’s called real analysis. It’s just stuff you draw. It’s curvy lines, straight lines with an X and a Y. As a physicist, you understand those things very well. I was like, “How hard can this be?” We had a textbook. We called it Rooted after the author. You would not believe the bizarre and unpleasant things mathematicians will do to a function. Completely useless for anything that we cared about. Completely useless for the structure of matter, completely useless for the function of economics, the function of a market.
Simon: Completely useless for writing a computer program. I guess they have fun. They’re talking about alien species that will drive you mad. They’re one of them.
Jim: I love it. A mini Murray.
Simon: There’s a famous theorem that mathematicians discovered, which says that you can take a sphere, cut it into pieces and make two spheres of exactly the same size as the old one. What? That doesn’t make any sense. How do they do it? They do some crazy construction that involves trans finite numbers and finite. Spheres are made of atoms. I count the atoms. I tell you, unless some very high energy things happen, that numbers takes constant. Murray’s distaste for that gag is not something restricted to just one or two.
Jim: I love it. Murray, farewell. One of the greatest people I’ll probably ever meet. Some time very well spent with Murray over the years. Let’s go on to cultural evolution, broadly construed over a very long time period.
Simon: Yes. Biological evolution, what does it do? One of the great one liners is that your body is a really great model of the world that you would encounter. You have feet that you walk on because we have a gravitational field that we sit in. You have a bilateral symmetry as opposed to radio symmetry because you don’t live underwater. You have a left hand and the right hand but not an up hand and a down hand. You can think of yourself as a story about the world. In particular, story about the physical world, the chemical world and to a great extent now, as a model of the biological world as well. Your body has a really good theory about small things that make you sick as well.
Simon: That’s biology. That’s the physical stuff that you really can’t change very well. You can replace parts of it, Jim. By and large, your body plan is fixed. What is cultural evolution? What does it do for us? The underlying mechanism is the same Darwin spotted this really early on. The underlying mechanism is we talk, we listen. We repeat. We change. We alter. The DNA mechanism for us is the human brain, far more complicated. Your DNA has these ribosomes. They make a copy. They translate the internal representation into proteins. Your brain is the ribosome for culture. It translates internal representations into actions, behaviors and speech.
Simon: It’s far more complicated. Underlying all of that is the sense that most of what we see around us is the product of we know not who. Who made this language? Who made the idea of a conversation? Who made the idea of a building with windows? With mortises and handles? To what extent they’re reasonably adaptive as we say, windows keep the heat in.
Jim: The rain out.
Simon: The rain out and let the light through. They’d kick it too big or the room will get too hot. In some sense, a lot of our culture’s adaptive. There’s some really fun stories of the ways in which even things as basic as our language are adaptive to the climates in which that language evolved. Italian has a different pattern than Swedish in part because Italians are speaking in a country that’s much warmer. You don’t have to be in such close borders. You could stand further apart. It’s also probably why we gesture a lot. Being half Italian American because when you’re that far away from somebody on the patio, you need to show what you’re saying.
Simon: Some of this is adaptations of the physical world. Some of it is adaptation to the very nature of knowledge. I secretly as an ex-physicist, I’m a Platonist. I think there are underlying structures. Thanks, Jim. Yeah. People can’t see here your large Italian gestures right now. When I look at the development of let’s say a political system. Let’s say working with Alexander Baron and a group in Indiana when we were studying the evolution of discourse in the French Revolution. How they’re working out, how they run a country. I see them as doing two things. One is responding to the necessities on the ground.
Simon: Responding to the fact that once you decide to hold a revolution, somebody’s got to run the pension system. If you dismantle the church, someone’s got to keep records of that all the gold doesn’t disappear into someone’s pocket. One of the people on this paper is Rebecca Spang, a historian. Her view on the French revolution is in part the idea that we’ve neglected some of that stuff.
Jim: Think about the stupid shit that we did in Iraq. Literally, the money did disappear. The pensions were gone.
Simon: Back to the city of London mostly.
Jim: Exactly.
Simon: Yeah. Some of that stuff is adaptation to circumstances. You hold a revolution. Everyone declares war. You’ve got to run it. There’s got some problems. I also see the story of the French Revolution, the story of the Soviet era, something we have under study right now through people’s diaries, the evolution of theater. I see these things as in part humankind’s contact with basic objects of sense and knowledge. That there is such a thing as a Lie algebra out there. It’s not simply a social construction that’s propagated by the fact that Murray was super charging it and had a twinkle in his eye. That what we see is our adaptation to the invisible.
Simon: We see in the French Revolution for example, people encountering some of the most basic ideas of ethical behavior. Unlike the Americans, the French hold a revolution and one of the first things they did is free the slaves. To us today, this is something that we expect people to understand. We don’t think of slavery as something that one context is fine. Another context, not so fine. When we look at what people are doing, what I’m most interested in is the way that they deform their minds in response to the invisible.
Jim: Expand on that a little bit.
Simon: Okay. All right.
Jim: A lovely statement.
Simon: Okay. All right. Here’s one. This is an unusual project we have with Ken Chang. Ken joined us and I said, “What do you want to work on?” Ken said, “I want to work on theater. I want to work on the fact that theater is famously a endeavor that gay people are drawn to. In particular, gay men.” I’m interested in the idea that let’s say take someone like Oscar Wilde, as gay as they come. Oscar Wilde writes plays that are heterosexual romances. Famously, The Importance of Being Earnest is on the surface, about two men fell in love with two women. They’re all nuts. They’re all eating cucumber sandwiches with the edges cut off.
Simon: What we ended up looking for is the ways in which Oscar Wilde is encoding human relationships that are fundamentally different from the surface that he’s describing. On the surface, these are really normative-
Jim: Highly conventional.
Simon: Highly conventional.
Jim: I saw it recently, yeah. We have a great Shakespeare theater in Staunton, Virginia, the American Shakespeare Center. They also occasionally put on other plays. They happen to put on The Importance of Being Earnest about two years ago. I’ve never seen it. All right. This is a beautiful clockwork piece of machinery.
Simon: It’s an unbelievable play. Clockwork is a nice poem. You look at a clock face and you know there’s something behind it that looks nothing like numbers and hands. What is Oscar Wilde doing? The simple thing to say and both Ken and I think this is not the case. The simple thing to say is there’s a lot of subtle codes sitting in here. There are some bits of slang that if you were a certain type of guy and you were in a certain social world, you understood that he was saying a couple of things that were not exactly the straightest things in the world. We think it’s pretty limiting to view something like The Importance of Being Earnest as simply an encryption of something else.
Simon: What we think he’s doing and we see this in not just in Earnest but in a long tradition that includes some people like Tennessee Williams. We see Wilde imagining new ways for people to be together. Not necessarily sexual. We see Oscar Wilde as working out the terms of a male friendship for example. Wilde when he began was somebody who thought mostly about how people relate to each other. What we see him doing is finding ways or describing and implicitly projecting onto the stage most of being that today, we might even take for granted. For example, the idea that people’s sexualities are more fluid than you would expect is something that we see there.
Simon: We see both examples in the two main male characters. Wilde’s discovering something. He’s discovering in those plays. He’s discovering facts about how humans can be together, how humans can form relationships. These are possibilities that he both creates and runs across. Daniel Boorstin his labor into Congress has written a couple of books-
Jim: Great historian.
Simon: Great historian. He has a book called The Discoverers. The creator’s the artist and the discoverers are the scientists. Oscar Wilde in one sense, he’s a creator. At first, he’s got to write. In another sense, I think we can view artists also as discoverers. They learn things about our minds and tell us about our minds in ways that have massive downstream consequences. Kurt Cobain didn’t invent wearing flannel. That’s something that disappeared. He also created cognitive patterns that enlarge our ability as a species. In order for that to happen, they’re had to be some contact with the real, something beyond just the idea that we all agree to say the thing that we all agree to say.
Jim: Interesting. You’re talking about an unfolding it seems to me that culture is at any given time X. There’s an opportunity and it happens constantly to unfold into new areas. One of the things I look back at our era I find amazingly interesting is how rapidly point of view about homosexuality has changed. Stone walls was not that long ago. Now, gay marriage is accepted everywhere. Frankly, by almost everybody under the age of 35. I would have never guessed it in 1985 that this could happen so quickly. It’s a broader unfolding because it actually fits perfectly well with the Declaration of Independence. All men of course, created equal. Women, let’s add them, too.
Jim: When you think about it, it’s an unfolding but it is grounded in something we already had.
Simon: Emerson has this line. You sound like Cavell. Our new yet undiscovered America. We are I think building things whose consequences we don’t completely understand. Pythagoras discovers the fact that the square root of two can’t be composed out of a bunch of rods that you divide into pieces. I don’t think that Pythagoras would have anticipated where that would end up. Cecilia Heyes at Oxford is somebody that we’ve been strongly influenced by. She has a great book that just came out, which is called Cognitive Gadgets. There’s a trend right now that’s a politically popular trend to over estimate the power of biology in influence in human behavior.
Simon: Evolutionary psychology is only partly a science. It gets some things right. It misses a lot. Heyes is wonderful. In some sense, square is the circle. Culture has lifted off from biology at least 5,000 years ago. What you and I do is in no way sitting inside our genes. It’s more than just the adoption of a particular language. It’s more than just the adoption of a particular set of cord sequences and a piece of music. It’s also the creation of the conditions of possibility for those. We don’t just invent music. We invent the idea of music. Heyes I think has a very good account of how the conditions of possibility of culture are themselves a product of culture.
Simon: Something that is not hard coded in us. This debate goes. If you want to get deep on this, it goes back to Chomsky in the universal grammar. Chomsky looks at language that there’s no way that this is something that we invented. This has to be hard coded. I think we’re coming around and partly simply through empirical investigation, I think we’re coming around to the idea that Chomsky was probably wrong about universal grammar.
Jim: I had a long, long conversation with John Holland about just this item. He was very strongly of the school that Chomsky grossly underestimated the amount of data that the child is exposed to. Because he was thinking in terms of formal instruction. Formal instruction is part of it. That’s by no means the whole thing. That was probably his number one interest right before he died was trying to figure out how to demonstrate Chomsky was wrong.
Simon: This is a great point. Chomsky came at language essentially as a computer scientist. He wasn’t a computer scientist. If you look at the early stuff he’s doing in the ’60s, it’s yanked right out of the theory of computer programming languages. If you look in the modern era and this is something that certainly, a colleague Ricard Solé has worked on. It’s also stuff that we see in Josh Tenenbaum’s lab at MIT. Holland I think was more right than wrong. We can learn. We can learn from a set of data that in the past I think we would have seen as too impoverished, as too small. Children learn language. Solé discovers these explosive bursts in how language fits together.
Simon: He’s not the first person noticing. He’s produced incredibly interesting mathematical model of it though. We go from words to word pairs. At some point, there appears to be a little face transition. There’s a discreet jump. All of a sudden, we’re producing the kinds of sentences where there’s a continuum from that to a Henry James novel. How that happens is mysterious. Someone like Heyes would say and I would tend to agree that the very possibility of that happening was something that we developed independent of our genes.
Jim: Yeah. I don’t know if you’re familiar at all with a very iconic classis scholar, Daniel Everett.
Simon: No, I don’t know Daniel.
Jim: Check him out. He takes a very strong argument that language is a social construct.
Simon: Yeah.
Jim: I’m still working my way through his books. As I said earlier on, my answer to all these questions is always both. I do suspect there is something genetic that allows us to be way more linguistic than chimps who we are very close to. On the other hand, my instincts are Chomsky’s wrong. We don’t have a strong universal grammar with 17 switches, which determines our language. For no other reason that Chomsky’s assertion about recursion strikes made us overwrought. There’s at least one language without recursion. Everett was the anthropologist that studied that language.
Simon: I see.
Jim: Which is interesting. Further, if we look at language in the raw, we never see more than four levels of recursion and seldom see more than three. Being a computer dude, it’s a big difference between universal recursion and a series of nested statements through your four-level state.
Simon: It’s a great story. The cat who chased the rat who ate the cheese. This is tail recursion, Jim. It’s an easier one to process. Years ago when I was a student in middle school, we studied Latin. We read Julius Caesar. We tried our hand in Virgil. At some point, our Latin master said, “No one actually spoke like this.” These incredibly baroque, intricate, inflected sentences where the verbs are showing up in all sorts of places, this was not how they actually got things done day to day. Virgil constructed a language that in some sense is sitting on top of Latin. If Latin’s the assembly code, he invented less.
Simon: The stuff that Chomsky was responding to in many ways is something that we built through our symbolic systems. The idea of recursion, how often do we use it, Jim? We use it in part for social signaling. I read a lot of stuff. I was educated by people who read a lot of stuff. Probably, my parents read a lot of stuff, too. The transcript of our day to day getting things done and including really high stakes things like going to the doctor or making a trade on the market. My guess is we’re roughly working with a regular grammar here. We’re working with essentially a finite state machine to get things done.
Jim: All right. Let’s move on to our next topic, which is consciousness.
Simon: This is a fun story for us. Consciousness is one of the things that we think distinguishes us from the animals. We can go back and forth on the extent to which animals have a consciousness. Of course, the word itself decomposes in lots of different ways. Is consciousness simply the ability to have a phenomenal experience of pain, for example? If I do something wrong in my computer, it might crash. We don’t think it feels pain. We do wonder, do Simon feel pain? Is it okay to either sum it? Was it having the experience that you and I would associate with being poked through the mouth and bashed on the head?
Simon: That’s one notion of consciousness. Another one that we get from in many ways a computer science perspective but also from a philosophical perspective is consciousness is the ability to reflect on our actions and our own minds. Most of these definitions have gotten us into some really tight squeezes. We don’t really have a good story about what the philosophers call qualia. The idea that if we’re sitting around and looking at a big New Mexico flag here and it’s yellow. I can tell you it’s yellow. I can tell you the configuration of the lines on it. There are some things we think that we can’t quite communicate to each other. You have to be there.
Jim: In fact, I remember in second grade asking my teacher, is your sense of red the same as my sense of red?
Simon: Yup.
Jim: Unfortunately, she just said, “Shut up, kid.”
Simon: Shut up. Right. This is Hume. We’ve been worried about those things for a long time. Dan Dennett says, “You know what? Qualia, it doesn’t exist.” Dan is one of the great demystifiers. It’s a bit like a Darwinian Lichtenstein. He sits around telling you that all the things that you really wanted to think about don’t actually exist.
Jim: Yeah, Lichtenstein I always describe as a very long journey to nowhere.
Simon: Yeah. He’s the guy who ruins the party.
Jim: I think he’s right actually. Another discussion for another day.
Simon: Yeah. It’s what happens when you get essentially European and stick them in Britain for too long. He gets a bit too practical for the world. That’s one thing that’s sitting out there. I think qualia are real. During difficult times in my life, I stopped believing in qualia. I think it’s actually a measure of how well things are going for me. I think a mentally healthy person does indeed reliably have experiences that are first person centered. Experiences that rely fundamentally on the existence of a subject, to proceed in a subject.
Jim: Let me interrupt you if you don’t mind.
Simon: Yeah, yeah, sure.
Jim: I have a day job, which I don’t. It would be in this area. I’ve built computer simulations. A very, very rudimentary consciousness, etc. I’ve read 50 books, 200 papers. I have some thoughts at least in this area. First, you’re using consciousness in a way that seems to distinguish human consciousness fairly cleanly from sub human consciousness. We started doing down the road but then you said there’s problems. I suppose I come from the other view. Gerald Edelman, he distinguishes primary consciousness, which is the same consciousness we share with a dog approximately. The dog is alive in a scene of some sort.
Jim: The same dog every time it wakes up, it has similar attributes, similar tastes, similar likes and dislikes. He also posits something called extended consciousness, which is this we’ll be much clear on what that is. This ability to perhaps reflect on our own consciousness. I also look pretty strongly at John Searle. I really like Searle and his down and dirty definitions of consciousness. He says consciousness is a lot like the digestive system. You can point out at it, too. This is the digestive system. It’s the lips. It’s the throat. It’s the liver. It’s the stomach. The Rutt coral area there is consciousness is like the digestive system.
Jim: It often has the same output. Under those scenarios, human consciousness is only a bit different than some human consciousness. It maybe has one or two new tricks. Perhaps Terrence Deacon’s symbolic thinking maybe. There’s others who think that it’s a very bright line. Human consciousness is nothing like animal consciousness.
Simon: You bring up some great minds here. When I first started thinking about this seriously maybe back in college but then this obsession kept going for many, many years. The default mode for thinking about these issues is the computational analogy. We are thinking machines. Whatever we consider fundamental to our experiences relates to the performance of a function. That has taken us pretty far. Famously, this theory of consciousness called IIT, integrated information. The idea that what you need is a sufficiently couple cent of reasoning devices. Once they’re tightly enough together, like FWAM frame out the other side comes what we like to think of as consciousness.
Simon: It’s some emergent property like a market. Market’s full of fools. The market’s pretty smart. At what point do you go from a ship of fools to something that reliably gets us coffee on the table everyday? It’s certainly a scientific theory. It doesn’t have to say that there’s a spirit sitting in your head kicking your pineal gland around. There’s also something still unsatisfying about it for me, Jim. I think consciousness is one of those problems that gets us to the limits of science as we do it. We think of ourselves as evolved beings. There’s no reason we should have gotten it all right. There’s no reason that the cognitive gadget of science is answerable to the world as it is.
Simon: It’s at least partially so. We build things. We discover things. Murray discovered the quartz. It works pretty well but there’s no reason to think that we’re a universal translator. That our reasoning processes, our symbolic representations can solve every problem.
Jim: Even if they could, there’s by no means a guarantee that they have. Our modern model is no older than 1625. Science plus axiomatic systems join together. Big symbols so that’s very rare reasoning. We talk about 5,000 years. 400 years ago is almost no time at all. Even if our tools are universal problem solving machines, they might be 400 years. Nowhere near enough to run the universe through the engine and come up with results. Those are two different points of view. They can both leave us at a current state where there is a ship load yet still to be discovered about the nature of reality.
Simon: If you look at some of the smartest people in the history of the human species, for me, Socrates, Plato, these are up there. You read something like the symposium where Plato, through the voice of Socrates is laying out a theory of knowledge you see in the republic, too. These guys are not dumb. They have a sense of symbolic systems. They called it geometry and music. I don’t think that the stuff we do today at least at the abstract level would be incomprehensible to these guys. Put it another way, they’re completely comprehensible to ask.
Simon: The further on that story goes, the further on the symposium goes, the stranger and stranger things get. At some point, Socrates is hanging out with this mystical spirit. He’s climbing ladders. He’s looking at boys, that whole thing. Something goes weird here.
Jim: They’re not that weird. As you said, it doesn’t strike us even as strange as science fiction.
Simon: That’s true. It would get you kicked out of even CMU if you did this for too long. He seems to go beyond what we informally think of as our best mechanisms. What if we took that seriously? Take Murray seriously. He’s not always right. You can go pretty far if you take a genius seriously. What’s that all about then? He’s not a fool. He’s not someone who believes in the little reality of angels. What’s going on? What if we missed in our computational metaphors? What if we missed even in our evolutionary metaphors? That’s something that we didn’t get 1625. We didn’t have that.
Simon: That took us literally until Darwin or maybe his grandfather had to blow it up. Consciousness may be one of those things that we don’t yet have the gadget to understand and explain.
Jim: That’s possible. I think because there is so much un-clarity about the topic, there’s a lot of bad discourse about the topic. It’s amazing. One of the ones I like to add to this is to my mind wasted argument on can a computer be conscious? I lived pretty much full from Searle and then extended a little bit John Searle’s perspective. The answer is no because the consciousness is what a human body does with the what, where that it has in the same way it does digestion. However, I then extend that to say in the pharmaceutical industry and in the food processing industry, we have things called Digestors, which are used to grow all kinds of biochemical feed stocks and food products, etc.
Jim: We can say by analogy that these things are Digestors used in the pharmaceutical industry. They’re analogous to how digestion works in the humans. In the same way, we can create in computers something we will call consciousness, which is analogous to the consciousness that is the emergent phenomena from various bits of meat and neuron. It won’t be the same thing. Searle stopped by just saying it’s not the same thing. I go one step further but say we can do things that are analogous, too. We go back to Tononi and IT. Having done a lot of reading about Tononi and anti Tononi.
Jim: I come tentatively that the inclusion that perhaps integrate information is necessary but not sufficient. That if we measure II on analogous machine that has something we’ve analogously called consciousness, it will show a high II but so all other things that aren’t conscious. All the human, if we could measure it. I split the difference and say, we can talk about machine consciousness. We just have to realize that it is not exactly the same thing as meat consciousness, as human consciousness. In the same way that a Digestor use in the food industry isn’t the same as our digestive system but they both do similar things.
Simon: Here’s a way to come at this problem. What do we want consciousness to do? If we look at let’s say the cultural effects of consciousness, which is an odd thing to say, you go back in the anthropological record. Karl Jaspers invents this phrase. He calls it the axiom era.
Jim: 800 BC approximately.
Simon: Exactly. He’s got to stretch it. At some point in time in the history of the species and this happens throughout the globe. It’s almost like these little fires lighting up. You have the Upanishads and the Indian tradition. You have some of what I think of as the modern books in the Torah. You have my favorites in Greece. You have the dialogs. You have amenities a little bit before. You have the emergence of Buddhism. Each of these things is a dramatic shift in the nature of culture. Before the axial age, the priests were also the kings or the priests were fundamentally allied with earthly power.
Simon: Out the other side of the axial age, out the other side of the Old Testament, Socrates and Plato, Buddhism, the Upanishads, all of a sudden, you have people discovering this notion of the transcendental. People deciding that there was something beyond the purely human biological realm. Jaspers thinks of that as a consciousness that emerges at the level of a society. Maybe the right thing to say is you know what? When will we know what consciousness were having this conversation and this conversation is happening in the context of people who because they had conversations like this, builds a recording device in a university.
Simon: Our computer’s conscious. One thing I’d say is look, I might run one of these consciousness assays on the iPhone. Maybe it is or it isn’t. I want to say where’s the great religious work of the computers? What have they done recently? Have they had an axial age? Have the machines written the Upanishads? Have they had the Socratic dialogs? Answer I think, no. Under that task, we’re the only ones.
Jim: Using my analogy, the computers aren’t even there yet. The smartest computer in the world, maybe they’re at the shrew stage. We don’t expect shrews to write Shakespeare. Give it 30 years and then your questions may be more valid. Something these analogous consciousnesses may well come up with things that we think of as on par with our best creations. It’s too early to tell yet.
Simon: This gets back all the way to the beginning of our conversation with The Fermi Paradox. In the end, the reason we’re really keen about aliens is we want to hang out with them because we’re bored with ourselves. Maybe the question if that does happen, if these machines do have their axial age, if their consciousness does actually bubble up to the visible level. Maybe the question is, will they be fun to hang out with? We have new playmates on the earth.
Jim: Again, every nerd believe yes that there were aliens out there. Every 12-year old nerd boy, at least in my era believe we could eventually build AIs that we could talk to. Absolutely. When we’re talking about The Fermi Paradox, it’s not just they’d be fun to talk to, which they would be if they didn’t kill us, the silent forest theory. The reason everybody’s quiet out there is because there’s the certain number of predators. If you speak up, you will get eaten. There’s another much deeper issue. If we are indeed alone, we have a gigantic moral responsibility.
Jim: If we are the only general intelligence in the universe, we have a greater duty to preserve ourselves and if they’re relatively routine. Because we may be a point of the sphere of evolution. It is perhaps our destiny to bring life to an alive universe. If we knew that to be a fact, which unfortunately, proving a negative is extraordinarily difficult. If we did know that for a fact somehow, I suspect it would make us take everything we do a shit load more seriously.
Simon: This is exactly the thing I would expect from a good self-replicating organism, Jim. Do we have a moral duty? Only if moral duties are beyond their species because if we disappear and morality disappears with us, then we’re done. That’s fine. Maybe there’s a way into this intuition. When we study, we look at the big archives. Sometimes, we get famous people. Here’s the phrase we sometimes use. Sometimes, we get ordinary people. This always rubs me the wrong way, this phrase ordinary. Because we’re absolutely extraordinary. There’s nothing like a human mind. The differences between two human minds are so negotiable compared to the difference between a human mind and a non-human mind.
Simon: If you’re a cognitive scientist or you study social behavior from a cognitive point of view, it can be an overwhelming experience to walk into a crowd. Not because the crowd is irrational but because you’re surrounded by this unbelievable mechanisms, unbelievable creatures. It’s not implausible that there’s nothing like it elsewhere in the world or elsewhere in the universe. I think neither of us will be surprised if a self-replicating organism worked out if they could use solar energy to make copies of themselves. That’s not a surprise. We wouldn’t be blown away. We’d be thrilled if we found life on Mars. We would be shocked.
Simon: It would be a shocking event to find things with consciousness, things that we could hang out with or at least we thought one day we could hang out with. If we discovered that, I think we would be shocked. There’s something miraculous about it.
Jim: It would be the biggest news in the history of the world.
Simon: It would be in some sense the end of news as we know it.
Jim: Yeah, exactly. Let’s go to our last topic and then let’s wrap up.
Simon: Wonderful.
Jim: Current science fiction. I used to be a true science fiction aficionado. In fact, growing up, my main source of science fiction is the library. I prided myself on reading every single science fiction book that was printed in hard copy, which wasn’t that large of a percentage but it was the better stuff of it. Unfortunately over the years, getting distracted into many other areas and fiction and non-fiction, particularly literally fiction for my fiction reading. I have drifted away from sci-fi. I’d love to hear what is cool in sci-fi today.
Simon: In one sense, I think I’m pretty much in tune with what everyone’s reading. [Da Chang 01:30:34] obviously blew up. Ken Liu, translator of Cixin Liu. Chinese science fiction has reached the modern era. That’s super, super exciting.
Jim: The Three-Body Problem?
Simon: Three-Body Problem. Ken is the translator for Cixin. He’s also a science fiction writer. We have British Victorian steampunk. He has Chinese empire silkpunk. Re-imagining the legends of the culture in a way that’s not quite fantasy I think because it’s a story about the technology and not simply the magic. Anybody who’s bumped into science fiction is going to bump into this. One of the things that I do with science fiction and this is partly because I’m interested in the cultural evolution of the last 10 minutes of evolution. The fact that science fiction drives culture. It tells us stories. People just can’t help but make it real.
Simon: The Star Trek communicators is a great example. Here it is. Apple Health given another 10 years and it will die. It’d tell you if you have something deeply wrong-
Jim: A little bit of a tricorder.
Simon: The thing it’s just smaller than the tricorder. We did better.
Jim: Yeah.
Simon: Science fiction helps us imagine. Neal Stephenson, there’s a great project that Arizona State called the hieroglyph project. This is a bit politically loaded. Science fiction got too Dystopian. It forgotten the idea that Arthur C. Clarke invented geostationary satellite. Somebody has to tell a story to motivate the engineers to build it. Actually, it’s really boring building a spaceship. We are in an era where our most interesting technologies are social. I started asking people not what are you reading but what are the books that people are reading that account for the cultures they’ve created?
Simon: One of the groups I’m completely fascinated by are the people associated not with the university system but with all the tech money that floats around out there. In part because the things like Deplatforming, some of the wealthiest and most interesting people on the West Coast have just given up on the university. Never given up on the tech schools. CMU is fine. MIT is fine but they have given up on liberal arts university. They’re developing parallel systems. Peter Thiel says dropped out of college. Things like the Singularity Institute, now the Machine Intelligence Research Institute.
Simon: These are groups that exist from one sense in the point of view in academic, they exist on the margins. The future of the institute if you’re a professor at Oxford looks like your rented room with an ugly chair. If you flip that around, Oxford is the thing that’s a rented room with an ugly chair. What we see is people creating a very different idea of what learning, teaching and research could be. I know some of these people. They’re fascinating people. I could give you all their names but they’ll probably be embarrassed to be associated with this. Sometimes, people call this the rationalists. Maybe after live nets.
Simon: There are some famous characters that surround that group. Elle is one. I’m sure you’ve encountered him.
Jim: Yeah. He and I have had a few chats.
Simon: I could imagine immovable objects and unstoppable force. Okay. Yeah, exactly. These guys are smart, fine. You write your code in Haskell. You build a block chain. Good job. What are your fantasies? What’s your Star Trek? What is your account of what things could be? How do you know if you’re falling short of yourself? I know that something’s gone wrong in my life if I think George Elliott would not like me. I ask this and one of the answers is and this one’s crazy, Vernor Vinge. We think of Vernor Vinge as the guy who invented the idea of the singularity. One of the books that a friend of mine pointed me to was Part 2 of a trilogy is a book called The Children of the Sky.
Jim: By Vinge?
Simon: By Vinge. This is not a new book. This is maybe a couple of years ago. Children of the Sky I think encapsulates the fantasies and the mythology of that group. It’s a story about they’re coming for us. The silent forest was right. We screwed it up. They’re coming for us. A small group of people have crash landed on the planet that’s reasonably backward. What they have to do is bootstrap themselves into a sufficiently post industrial civilization that they stand a chance of surviving. Children of Sky has encoded in there is an extreme version of The Power of the Mind. The idea that if you’re smart enough from a grain of sand, you infer the Sahara.
Simon: If you’re smart enough from a cup of water, you know Niagara. It’s something that comes up in quasi science fictional forum for David Deutsch, The Fabric of Reality and The Beginning of Infinity. Children of Sky is the dramatic version of that. It has some sex. It has aliens that are actually a bunch of minds all linked together and they look like pets. It has a bunch of tropes that we’re familiar with. The overall structure is this idea that the universe in one sense is fundamentally inhospitable that’s trying to kill us. In another sense, the laws of physics drive us inexorably to a horizon of infinite possibility.
Simon: If you’re sitting in a polyamorous nerd commune in Berkeley right now, The Children of the Sky answers I think to your fantasies. I use that word in a nice way. I don’t mean that in a psychologically condescending way. It’s a mythology that you can get your teeth into and tells you if you’re doing it right.
Jim: Any other examples?
Simon: That’s a good question. I brought up David Deutsch. Deutsch is one of these unbelievable super geniuses.
Jim: I don’t think he prefer to have his stuff thought of as science fiction at this point.
Simon: The two books, Fabric of Reality, Beginning of Infinity, they’re incredibly compelling books. If you read them, he has one chapter. I think it’s in Beginning of Infinity where he battles his way out of a brain in the vac. He explains exactly how he’s going to show you that you’re not living in a simulation. It takes on a science fictional angle. One of the things I found really amusing, I think this is in Fabric of Reality, the earlier book. He gives this proof from first principles that the matter density of the universe is below critical. Omega’s less than one. Absolutely right. The inexorable force of reason is such that physics must come to know itself.
Simon: The universe must have this property and that property. I’m reading this. Absolutely right. I remember actually, it turns out the matter density of the universe is not less than one. Something’s gone wrong. It is precisely one unfortunately as far as we can tell.
Jim: As far as we can tell, what is dark matter? What is dark energy? Is it bullshit?
Simon: I had to check my dissertation. These were great problems. David has a centrifugal intellect. You can learn a lot about the many worlds hypothesis or the many worlds theory. You can also follow it as a narrative, almost like Olaf Stapledon. It’s an update of star maker where Deutsch goes on a journey that takes him in the end to an account of human nature that suggests that one day, we will manipulate indeed the very fabric of reality. Maybe 10,000 years? If we don’t hit one of those great filters.
Jim: Yeah, great filters. Thank you, Simon.
Simon: Yeah, it’s super fun.
Jim: This is unbelievable. Really interesting I thought it would be and it was.
Simon: Yeah.
Jim: This is very, very good.
Simon: Thank you, Jim.
Jim: Thanks.