 Hello my friends and welcome to the hundred and second episode of Patterson in Pursuit. Got a really interesting episode for you today. So here's a question. Does scientific progress look like a very straightforward accumulation of knowledge? There's a big old pile of knowledge and year after year we just keep adding more and more knowledge over time and eventually we have this wonderful mountain of knowledge. Or does a science and scientific progress go through ebbs and flows? Do we gain knowledge and then potentially lose knowledge? And is there any difference between gaining theoretical knowledge to improve our understanding of the world versus gaining technical knowledge or engineering knowledge, the ability to manipulate the world in ways that we prefer? Are these the same types of knowledge or are they different areas? If you guys have been listening to my show for a while or following my work you know I tend to be rather skeptical when it comes to the contemporary paradigms in which we live. I think a lot of ideas really aren't resolved at the foundational level. This is true in philosophy, seems to be true in a bunch of areas of science, maybe even engineering, and to my shock and horror even something like mathematics. So I tend to see our contemporary academic system as not producing particularly high quality knowledge, but there's this interesting thing here which is we seem to still be making engineering progress in some type of scientific knowledge even if maybe the fundamentals aren't sorted out we still seem to be progressing along some dimensions. So how is that possible? How could it be that both the fundamentals of various disciplines are not sorted out and are maybe completely wrong and the experts in those domains don't even understand the fundamentals of their disciplines and yet we still seem to make scientific and engineering progress. These are the types of questions I'm discussing with Dr. Jeff Anders today. Jeff is an excellent philosopher and a research program designer who's the founder and executive director of Leverage Research which is an independent research organization located in the Bay Area. He's currently studying early stage science. In this conversation, Jeff brings up the concept of intellectual shelling points which is a really neat idea to explain how researchers can coordinate on a particular topic even if maybe the fundamentals of that topic aren't quite sorted out in a perfectly logically rigorous way. Shelling points are an idea that comes from game theory and recently I've been hearing a lot of people talking about shelling points in the Bitcoin world which is unfortunate because generally when bit coiners talk about shelling points they kind of abuse the idea and they don't port the concept smoothly from game theory into Bitcoin but this is actually a great application of it and I'm sure you guys will enjoy it. Though I tend to be more skeptical and pessimistic with regards to contemporary and maybe historical scientific knowledge, Jeff makes a bunch of good points and is significantly less pessimistic than I am and argues that things are in fact trending the right direction over time. So I hope you guys really enjoy my conversation with Dr. Jeff Anders. Alright, Mr. Jeff Anders, thanks so much for coming on Patterson in Pursuit Today. Can't wait to talk with you. Thanks for having me, Steve. I'm excited for the conversation as well. So I just want to jump right in and I know you've been doing a lot of research on the philosophy of science and the history of science and I want to give you a bad idea that I used to hold, I think a lot of people used to believe and tell me what you think about it and tell me maybe why you think this is incorrect. So the story maybe that people inherit growing up and thinking about science is that scientific progress is linear, that there's over a period of time there's just pure accumulation of knowledge. We didn't know something and then we know a little bit more and then we build on it and we know a little bit more and we know a little bit more and it just keeps going that way indefinitely and I remember thinking that and just kind of gathering that maybe from our culture and maybe explicit education but the more I've learned myself I'm thinking okay that story is definitely incorrect for a bunch of reasons I have suspicions about but what do you think? What do you think about that story? I think that as far as a really simplified story it's like not completely terrible because it's absolutely the case that one would have to figure out how to quantify across disciplines and so forth but I think sort of on any fair measure we're at the absolute height of scientific knowledge and technological sort of skill and so I think there's definitely a sense in which there has been sort of like really solid progress then there are a couple things that I think are sort of wrong about this story so the first the most basic thing is that if so sometimes we lose knowledge this is a thing that happens there's a number of different interesting cases one of my favorite cases is the Anarchithra mechanism you may be familiar with this but essentially the around 1900 some archaeologists found people as a shipwreck off the coast of the Greek island of Anarchithra they found a bunch of regular sort of ancient Greek artifacts and then this crusted gearbox and they sort of left it on the shelf didn't really worry about it and then you know a number of years later someone was like wait a second gears hold on and so they went through and they x-rayed the thing and they found out that it was actually really substantially advanced that it was essentially a geared mechanism with tons of gears and you put in a crank and you turn it and then it would keep track of the relative positions of the sun moon the planets as well as the timing for the Olympic games and the whole thing had an essentially an instruction sort of thing sort of etched onto a plate and then there's nothing of this level of technological sophistication for like many hundreds of years like this is you know you know over a thousand years and so this is an example where you think well that's crazy there's this super advanced artifact relative to the time and then more than a thousand year stretch where we have nothing like it okay this means that it's not the case that there's simply straightforward accumulation okay so that's like the sort of first thing that I point out there's lots more to say about the Anarchithra mechanism it's a super cool example another thing that is wrong about the really simple model of knowledge accumulation you can see this sort of embedded in a lot of the histories where when you go back and you look at something like the history of our knowledge of infectious disease or the history of electricity for example what frequently when you look at a history you have essentially like the simplest version of a history is like a timeline and then there's the first time anyone ever noticed that something that we now know to be true was true well the problem with this is that in a lot of cases people figure things out and then those things just don't get adopted right where they get adopted by a couple people and then the relevant sort of pocket of knowledge dies out and so then this is but so then in addition to there being sort of things being lost you also have plenty of cases where people are making discoveries and then the discoveries don't end up get getting added to the total pool of knowledge until substantially later right it seems to me that there's a domain of knowledge which is like engineering and technical knowledge and then there's the thing I'm really interested in which is more theoretical knowledge and sure and and these tracks also seem to diverge that sometimes you get engineering progress when maybe you don't get theoretical progress maybe sometimes you get theoretical progress you don't get engineering progress and so when I think of the Greek mechanism this is this really fascinating case where you have engineering that was lost knowledge that was lost sure like that that seems remarkable because you think there would be more incentives aligned to preserve that type of knowledge you could understand some esoteric theory via being forgotten and then it gets recovered but think that we have concrete mechanical technical knowledge that just gets destroyed is amazing yeah well let's see so there's a number of questions there so right now I think that in a lot of cases where you see societies losing some of their engineering capacity there's in some cases you have civilizational disruption as the sort of cause in other cases you have not so much civilizational disruption as you have society changing its interests so I wrote a piece recently that talks about the rocket dine f1 engine I completely love the example essentially you have the air force once a huge rocket a contract rocket dine air force actually loses interest but NASA comes in and then rocket dine ends up building these f1 engines which these absolutely enormous engines that end up being used on the Saturn 5 so these are the engines that took us to the moon so then the there's you know if you look online you'll find there's a bunch of videos and articles and discussion and so forth on the topic of can we still build the f1 rocket or the f1 engine and if we can't build the f1 engine maybe this means that we've lost we've lost you know important knowledge and so forth and and how could that be the case you know given the fact that this is so recently people are like completely shocked at the proposal that we couldn't just rebuild the thing that took us to the moon and then if you like dig deeper into the actual details of the case I think what you find is that you know so first of all the crazy thing about the f1 is that they were essentially handmade so you have a very large number of very skilled welders or engineers doing all of these complex individual welds and essentially hand crafting the rockets in the intervening time there's all these technological advances there is then much less incentive to produce nearly as many skilled engineers and a lot of the people who worked on the original f1s were tired and so what you end up with is sort of technology moves on this changes the sort of which skilled people get produced and so then it becomes hard to reproduce the original engine in this case that that's not exactly a sort of real loss of knowledge because we now you know the company Rocketdyne made designs for a more advanced a more advanced engine it's never actually been produced but it would be the sort of thing that we could make using laser welding and machine tools and all the stuff that's been developed over the last you know 50 years the but it's interesting because you can imagine what happens if we as a civilization like there's a way in which we're less interested in space now than we were in the past it's not that we're opposed to space exploration we have Elon Musk and Jeff Bezos and we just really hope that they succeed at what they're trying to do but you just have so much less sort of a much smaller share of the GDP there's just much less total societal support and you could imagine that if you know Elon and Bezos go pick up different hobbies or something like that and then the industry continues to change for a while you might imagine that then we lose the ability to even be able to manufacture the things and so this this happened with Brom and Roman concrete where the Romans had this really interesting form of concrete that gets stronger as you leave it in the saltwater modern concrete is nothing like this when you think of things like the pantheon or the Coliseum you've got these concrete structures that are they've been around for a long time and they look really solid if you imagine making anything like that out of modern concrete you imagine the thing starts to crumble and okay so this what's happening is they're making it out of this special form of concrete but then it's you know it's interesting the you had a bunch of these different um so you a bunch of these uh sort of capacities for manufacturing concrete and then people stopped building the relevant structures like you had fall of Rome fall of you know central support people still use concrete in some ways but essentially over the course of many hundreds of years you just end up with less and less and less support for large centralized projects at which point people then stop understanding how to uh like basically how to construct the large buildings you could imagine the same sort of thing happening with rockets so another interesting fact about this is when you're talking about the the rocket uh scenario that wasn't that long ago i mean good heavens that was what 60 70 years ago they're not not even that long yeah late 60s early 70s yeah like when they were building that so yeah to think that that kind of knowledge degradation can happen in a short period of time is remarkable to think over long periods of time if if those engineers died out you might legitimately forever lose um those relevant pieces of knowledge it also makes me think that something i'm continually learning is that there are different there are dramatically different types of knowledge you're talking about uh theoretical knowledge that can be captured in a book or an instruction manual that's one type of knowledge that can be made explicit and and maybe transferred to other people but there's this other category of knowledge which is the implicit knowledge the stuff that you it's it's it's the engineers know they have a bunch of knowledge that they never put into the blueprints that they just kind of poured around that they can share amongst the the others in their class yes um but it it never gets made explicit and that seems like that's the type of knowledge that can be extremely easily lost is when there's a some group of people that has tacit knowledge and then for whatever reason they aren't building the thing anymore or they die out or funding dries up that kind of thing yeah so i mean i think it's interesting i think you can measure what a civilization cares about by looking at its words um so like for example the romans had all of these different words for power and the greeks had all of these different words for knowledge um we don't have as many words for knowledge um we do have an almost unlimited number of words for sending short messages so it's like slacking DMing texting you know etc um and you could just list tons of those and that's because i think as a civilization we are completely amazing with regard to sending short messages you could imagine someone you know of course you wouldn't send them a slack message you should send them a text message in this circumstance uh so we've got sophistication as a civilization on a particular type of communication but then the greeks really cared about knowledge and so then they had a whole bunch of different words for knowledge like they had so there's episteme which refers to third person transferable knowledge the sort of explicit objective etc then they had tecne which refers to technical knowledge or know how we have a word for that and then they had sofia which refers to wisdom or particular type of theoretical wisdom there's also nosis there's furnesis furnesis is practical wisdom and there's no natural translation for that word in english um but yeah it's interesting when you think about the preservation and decay of knowledge um the easiest thing to preserve is the episteme the explicit uh easier to you know you can write it down sort of knowledge whereas i think in a lot of cases knowledge traditions are based around tecne you have a whole bunch of people who know how to do a particular thing and then they lose the ability to it's very hard to completely communicate that in fact in a lot of cases and this is something that i think relates to the history of science in general and something i think is underappreciated nowadays when people think about the acquisition of scientific knowledge when you said you're really interested in the theoretical knowledge well the theoretical knowledge itself frequently historically actually comes after you've developed a whole bunch of tecne so you end up with an engineering capacity first before you end up with a lot of knowledge of what's going on so like a good example here is going to be chemistry where prior to chemistry you have the alchemists and a bunch of the alchemists are doing something really weird but a bunch of the other alchemists are essentially working on figuring out how to break down rocks and materials and so forth into the relevant elements as they're doing this they're getting a whole bunch of knowledge of how to work with materials and then that knowledge ends up being really helpful but this is something that you get before you get to theoretical knowledge right this is there's an interesting parallel here just in philosophy which is the discussion between experience and theory so what comes first do you have a theory and then you go out and experience the world through that theory or do you gain a bunch of experiences and then after the fact you think about all those experiences and you construct a theory to try to explain them yeah i mean i well i think that it's it's a delicate and complex matter it's extremely difficult to look at the world without having a whole bunch of concepts in your lens so to speak when you look at the world as the idea of theory ladenness of observation you will see the world as behaving in a way where you'll see it with a layer of a sort of interpretation added my favorite example of this comes from philosophy it's the idea of causation so you know if you see one billiard ball roll along and hit another billiard ball and the other billiard ball rolls away you think okay i just saw causation but that's not actually true if you pay close attention to what you see you see one motion here a noise you see another motion you see a sequence you don't actually see the causation you don't perceive it this is a point made by david hume then one of the few things that's largely accepted by philosophers of course some philosophers disagree but it's really interesting because then whenever you look at things and think that you see one thing is causing another what's actually happening is you're adding interpretation okay but this is true generally yeah and so and so then in it's in the way i think about this is that there's a relation between observation instrument and theory where normally people the sort of simple view is that you have a sort of experience you know maybe people don't take into account theory lateness they think they just have the experiences and then they observe a bunch of things and then they build a theory and then they go and they the theory makes various predictions and then they go and they test it by making further observations i think that's certainly happens that's only one really small part of the picture actually i think what there is is a bidirectional relation between theory instrument and observation so you get the you start off with a theory you think that you don't you're wrong you've got a theory it's in the background you go you make a bunch of observations the observations then inform the development of the theory but if you're really serious about making observations then in a lot of cases you'll develop instruments to help you to make observations and notably the well the instruments help you to make more observations but also the observations help you to refine your instruments you're you're there trying to observe some phenomenon and your instrument is giving you weird readings in some area well okay the weird reading is an observation but then you fiddle with your instrument to make it better and so the refinement of the instrument in response to observation that happens and then your theories tell you to build certain instruments but also sometimes you end up with instruments that you don't know why they work and then you have to build theories in order to help with that right okay and so i want to unpack this a little bit yeah yeah okay so just to digress a little bit um so i've got my premium ginger ale here right this is a it's not a non-alcoholic ginger beer and i always like to use props whenever anything that's close by i always grab so just to highlight the point about this relationship between theory and experience because it's essential sure most people i think prior to examination operate within the realm of their own theories and they they their theories are so strong that they they almost lack access to the the real texture of their own experience so people would say oh what's happening right now well there is a there is a bottle of premium ginger brew that is being held in my hand well how do i know it's right there i see it they don't they that's like one that they're mistaking their theory for their experience if i'm actually telling you about my experience well there is a green color color blob right here with a lot like the yellow color blobs and this red color blob and here's this peach color blob i call all of these well this is my arm this is my hand this is a bottle i construct a theory not i'm not rationally sitting down and thinking like i'm not explicitly writing out whenever i experience a green color blob that is this way it is a bottle it's just this theory that gets constructed the background that sometimes you can even lose access to because you get it's such a compelling theory you don't even realize that it's a theory anymore yeah i completely agree with this in fact it's actually really hard to there's a first of all a technical question about whether the mind is capable at all of accessing experience without interpreting that experience through a concept that's the thing the next thing i was going to ask you well yeah so i mean con says you can't do it and i'm in cons camp in this regard um i my current models of the human mind um include that it's not possible to uh have um it's not possible to cognitively access experience that you're not conceptualizing um now you know is that actually true um it was a part of you guys are like pure awareness if i would say if i'm trying to be honest and i cease the naming and labeling function pure awareness i mean that's that's that's a further matter because you know there's a question of do you mean pure awareness or do you mean awareness of an experience or awareness of a sensation right so you've got your bottle yeah and you were talking in a sort of sense data type of language where you're like there's a green color blob and so forth yeah okay but does pure awareness involve a green color blob like you know so it's something like socrates would say well you know steve is it not you know is it not the case that the um the awareness is a different thing than the green color blob or should we say that they are the same okay how about so awareness is too hard how about just the the pure color blobness presenting itself yes um is that does that come packaged with a concept um well so there's a question and so one meta question here is how could we know any of these different things i i happen to think that there's a way to investigate them philosophically i also think you can investigate them empirically uh but i can tell you my you know the state of my current theories on the thing um i currently happen to believe that the sensations themselves or the experience the green color blob doesn't come with concepts um but it's always the case that the mind has to generate concepts or it has to apply pre-existing concepts so that the experience is conceptualized in some way so it's you know and this is this is an interesting sort of you know question that you know covered by you know rationalists and empiricists and so forth of whether all of your concepts arise from experience um but anyway it sounds like you want to take this somewhere well yeah i want to keep going down the radical but i actually can import this back into the philosophy of science okay so one of the areas in which i think from my investigation into the theory that i find lacking uh in contemporary times is interpretations of quantum mechanics oh there's particular types of experiences that people are having in the laboratory and then they explain them in a well they sometimes they explain them and sometimes their explanation is that there is no explanation uh right and then they come up with concepts like well okay maybe reality can be in no state at all or it's in a superposition of states whatever that means and i think okay some smells a little funky from the theoretical angle though the though the practical knowledge the technique of quantum mechanics is amazing and beautiful it seems to be uh profoundly lacking in the uh theoretical dimension so do you have i don't know how deeply you've researched this do you have any thoughts on that that area in particular and in general that that maybe the theory lags behind these experiences um well let's see yeah sure so i have some thoughts on that it is really funny when you see especially science news reporting where you know people talk about quantum mechanics and say things that seem like they just forgot to be straightforward contradictions science my favorite one is definitely scientists in australia prove that reality doesn't exist oh yeah there's a million of those probably not that's that's shocking um uh but so i think that there's there's actually a further concept that's really helpful and also part of i think how it's important to think about the progress towards truth um as part of the scientific investigative sort of scientific enterprise or the intellectual enterprise generally um a lot of people think that the intellectual progress is it's like there's there's sort of two different sides um on one side you have people who say that the thing that drives intellectual progress is the pursuit of the truth and the things people are trying to do is they're trying to find true true propositions and when they find true propositions those are the ones that people adopt and so there's this accumulation of knowledge of true propositions then you have people like thomas coon book structure of scientific revolutions i think is excellent um where he goes through and catalogs all of these different ways that there seem to be things that aren't precisely aimed at the truth that are affecting the way that science works and so you'll have people adopt a paradigm even though the paradigm conflicts with some known existing evidence and so then it's like the evidence seems to indicate that the relevant theories are not true but nevertheless it ends up getting adopted and this is you know and ends up being very fruitful and then scientific progress occurs so then you have on one side people who think that oh so this this can lead you to think that well maybe the entire enterprise isn't related to the truth um you have coon you know concluding some pretty radical things at the end of his book you've got paradigm and commensurability and anti realism and so forth basically maybe there's not actually a reality and we're not actually getting close to it to simplify a bit um i think that there is a middle way between these where i think that the um and i'm about to publish an essay on this topic um the uh but there's this concept uh which i'm uh been using which is the idea of an intellectual shelling point so shelling points you may be familiar with those um this is basically a concept from game theory where uh you have if people can't talk to each other could they still arrive arrive at the same conclusions there's like a if you're in new york and you can't contact the other parties and you know that you need to meet where do you go um and then there's going to be maybe it's time square because time square is the most sort of evocative um available possibility and no one has any way to find any other possibility and so then maybe you can have people um end up with the same answer that's the idea of a shelling point i think there's a pipe of shelling point that takes that is uh relevant in the history of knowledge and in the history of science uh which is the one that pertains to the pursuit of the truth so i've been calling these intellectual shelling points and the idea is that intellectual shelling points don't necessarily need to be true in fact they don't even need to be propositions you could have an observation or a set of observations you could have instruments you could have theories um and which would then be made out of propositions the the key thing is not whether they're true but whether they bring in a whole bunch of whether they attract researcher attention in a way that causes the researchers to all then be oriented on the mutual pursuit of the truth yeah so the i mean so like for example if you look at euclid right so like yay euclid amazing um was it true so you're you're a stickler for truth and you use the word truth in a particular way so tell me okay euclidean uh euclidean geometry true or false false emphatically false spaces discrete yeah spaces wow okay you're going to space is discrete root yeah um and you you know if you even if you didn't believe that space is discrete you could maybe say that um euclidean geometry is been empirically refuted as the correct theory if we at least if we accept general relativity probably if you're going for discreteness then you're going to reject a whole bunch of different things but okay well here's a puzzle for you then okay given your you know statement that euclidean geometry is emphatically false okay well first was it useful in the development of knowledge and definitely okay definitely um and did that did that have nothing to do with the truth so this is a fascinating question and here's my attempt at answering this in particular because euclid is fascinating me it's like it's one of the it's arguably the prettiest prettiest axiomatic deductive structure of knowledge that's ever been created and it's wrong because the axioms are wrong and this is really this has puzzled me for a long time here's what i think here's my attempt at answering that for a few thousand years there was not sophisticated enough or yeah sophisticated enough mathematics to deal with discrete uh space okay it was too hard that and uh to to assume continuity in mathematics is a fundamental assumption that goes back a really long period of time however recently like there's a guy there's a researcher i'm a big fan of in the university of new south wales dr norman wilder who has discovered a way to reinterpret uh continuous mathematics through the discrete lens and that he wouldn't phrase it this way because this is going to be a great example keep going yeah yeah so so now i would say the though the formulas are are approximately correct that come out of ideas from euclidean geometry uh there is a much more logically precise and concrete way to reinterpret them through the lens of discrete mathematics so it's something like it's something like you uh you asked is is the progress based on truth it's it's something like a they were almost right for a few thousand years but they didn't have quite the right level of logical rigor in order to get it all correct well let me press you on so i think something like this is correct but but if you look at the details i think there's something weird happening okay where you say they were almost right right do you mean almost in the sense that the you just have to change the theory like in one small place and then it's right well it depends on what you mean by one small place so if the if the structure is is hierarchical right yeah you might just have to change one little thing but the one little thing's pretty damn near the bottom which is okay discrete yeah okay so so there's at least some sense then on your view in which euclidean geometry was really quite far from the truth yeah well from a theoretical standpoint yes if you're looking if you're just trying to analyze the pure logic of the theory it was very far away from the truth yes okay well so and then this is perfect because then it's like i think we we we recognize that there's something truth oriented about it even if you think that the thing is not true in like a strict and philosophical sense yeah so this is where i think the idea of intellectual shelling points is really really helpful the thing that euclidean geometry does despite its flaws is it causes there to be a common place for people's minds to go that make it so that even if they disagree with each other they can all still work together so you can you don't have to be a euclidean and you could still make an advance in your euclidean geometry right and it's and so the if you normally think about it like the like when you think about trying to get researchers to work together this is something that's actually really quite difficult like imagine you've got a bunch of theories you spend a bunch of time trying to like get people to like adopt theories and so forth and you'll know from your experience that it's super hard to do this right okay well and then this is if you think it's like well one of the problems is just that everybody disagrees right i mean if we were in an area where we'd already figured out all the answers and we wouldn't need to you know then there wouldn't be a problem and so then there's this weird puzzle which is how do you go about causing people to all work together on the same or similar problems even though they disagree with each other and so then i think that there are certain things so i think this is what's happening with euclidean geometry what's happening is it has a well-defined enough terminology a you know explicit enough set of axioms it doesn't even have all the axioms like the people were disagreeing with euclidean geometry really early on in fact the first construction where you create an equilateral triangle um there's just you people were disagreeing that i mean there's the assumption that the um lines don't intersect for example before they you know hit the uh that they intersect only once for example um and so there's all of these tacit assumptions in even embedded in euclid's axioms this doesn't get fixed until helbert but that doesn't matter i mean it does matter in terms of actually getting the truth but as far as being good enough to sync up researchers in a way that causes them to all focus on the truth okay but so but so then just to finish this point the i think i mean there's an interesting question you talked about this mathematician who you really like it might be really really hard to come up with the relevant mathematics had a whole bunch of the original mathematics not been developed and so i think the correct attitude to have towards the truth is that the discovery of the truth goes through many different points and then each of the sort of waypoints or the you know points you hit along the way aren't necessarily themselves literally the final truth right but they serve a coordination function and one that helps to orient the people's minds on the discovery of the truth yeah i think that's a true and exciting point and i would just say the when you're talking about uh you said the this is a progress of arriving at the truth and at no point necessarily have we arrived at the full truth i just wanted to make the point that all of these points along the way are discreet they are not continuous i have a few real things to say um um like i'm finding this isn't a very exciting idea um and i'm finding its application in religion and so i grew up as a christian evangelical and uh and that was very important to my parents to have that kind of training and from my now perspective i would call indoctrination and fell away from it and thought all these ideas are a bunch of nonsense well recently the past few years i've been coming to it through a different lens which is to say okay if we just change one little thing then suddenly religion makes a whole lot of sense and what if when people are talking about god they're actually just talking about the universe like the whole thing all together and suddenly when you think about well there's a particular relationship you have with god you are a part of god and god is a part of you well that's actually kind of true and you think god is omnipresent well yeah the universe is everywhere and god is omnipresent so it's what so i'm approaching religion through this lens and finding all kinds of certainly truth ideas that are profound in there and then i look back on the religious indoctrination i got and i thought okay you guys are in a sense a kind of getting at true conclusions it's just you're from the theoretical interpretation it's all tortured and preposterous where where then they tell very elaborate stories about the about people god is a person and he came down and he had you know and he was nailed on a cross and they tell all these very concrete things and it's like well you may act you may be close but you're missing this one critical thing that if you get it right actually the whole structure makes a whole lot of sense yeah so i mean there's certainly so spinoza you may be familiar with spinoza has you know his book the ethics where he tries to prove the existence of god gives a proof i gives multiple proofs and then attempts to and then derives a whole bunch of the attributes and so forth and you think okay well you know this is spinoza believing in god except that then a number of people think that spinoza didn't really believe in god and that the word god was merely being applied in this case to the universe or something pretty close to the universe i think that it's a really good impulse to try to preserve or reclaim as much of value as you can find in different traditions i think especially for people the traditions that they grew up in so i think that that's super important and then i think that the they're going to be some things that you'd be able to preserve by means of this but i i think that it doesn't it's probably not going to get at the total core of the thing because i think that yeah well because i think that and i think that religions i mean their religions are super complicated they have all these different parts but one part of it pertains to the way that you should relate to the rest of the world and the way that you should relate to other people and the way that you should relate to various experiences you have concepts and so forth and then if you do the sort of transposition where you're like oh i see but by god we're referring here to the universe i would say there should just be an independent check on each of because what's going to happen is a lot of the practitioners at least of the religions are going to the thing they're really going to care about is whether you're preserving the correct relations to everything else right so you might you might then say so so here's a question if you see yourself as part of the universe does that mean that you should have an attitude of piety towards the universe should parts have what relation should parts bear to the whole and so i think that you know with your focus on inference and getting the logic exactly right i would propose that there is a further inference that needs to be made from the claim i am part of the universe to the claim i should relate to the universe in a particular regard so i think i think that especially yeah as you go farther down the specifics too they'll make prescriptions about culinary intake so it's like yeah oh you have to relate to the universe with piety which means you can't eat shellfish in this particular way and i think they go off the rails but there was a little there was one word that i think is critical here you said the practitioners will often focus and now this is really fascinating and i and i'm gonna make a hierarchical claim that also seems to me that there's some well this sounds impolite but here we go it seems to me that practitioners of of religion tend to get things grossly incorrect in very important ways and miss the the many kernels of truth to be found in their religion so that they might focus on trivialities that don't matter and are probably wrong however it also seems to me that that the practitioners theoretical interpretation of their religion is different from maybe the people at the top level so when i think about the philosophy of Jesus and he's talking about don't you know that you know i i'm in you as i'm in my father or what i think he's talking i don't think he was speaking literally i don't think Jesus as the philosopher was thinking i am literally inside of you right and that's that would be kind of weird i think he's talking metaphorically and then there's a whole maybe even the majority of people over long periods of time who then suddenly get this weird literal theoretical interpretation of what jesus was saying and then they develop rather ornate theories around the bastardization of the of what what some of the leaders maybe were saying and what maybe what some of the higher level people were saying from my elementary research into early christianity i do get the impression that there's maybe a little bit more focus on metaphor and embracing of metaphor and the non-literalism that we get in in it and maybe like i don't know the 1800s or something like that so it seems to be rather hierarchical you get the high quality theoretical stuff coming out of some a small amount of people and then maybe the practitioners in the general public get things embarrassingly wrong well i mean at least in sort of there's like there's a sort of division of labor i mean you can think about you know you could you know accuse the adherence of either science or any religion and so forth of not really understanding the fundamental principles and so forth and then you have to go to the higher ranks in order to get the you know the real the sort of best arguments and the sort of best defenses of the relevant thing so i think that that's you know obviously we want to strive for a world where people know everything or specialists and everything or something like that but until we are even remotely close to that i think it's something i tend to you know encourage people when they argue about religion or argue about science to not worry so much about what your average person on the street says about the things you want to go to the source and you want to go to the highest quality defenders of the relevant things now it's an interesting question i mean we we could dig in more deeply and so forth you talked about over literalism and so forth the and then i mean all things considered the Protestants have had substantially less time to develop their theology than the Catholicism has but you know in my knowledge of Catholic theology is limited but i from what i understand i think there are you know metaphor is something that is you know included in the proper ways to interpret the bible in accord you know so i think that at least some religious traditions admit admit of metaphorical interpretations of various things yeah yeah absolutely it seems to me that the thing that i'm looking at having grown up in this you know extreme Protestantism is it's a mixture of anger and frustration because because it's like we were talking about before where you can be a little bit wrong but if you're a little bit wrong at the fundamentals your your whole thing is wrong and i think some of these truths are so important and they're so profound that to focus to focus on the you know the historical fact that was there was a literally a serpent that made vocalizations in a garden and that's important for you to believe it's not just that that happened it's that in terms of how you relate to the universe you need to have this belief right that's preposterous when you could just there's so much truth that you can discover just about talking metaphorically i do i do want to return if i can to uh um so take it back to science here sure okay so another reason i found what you said exciting about uh the shelling the intellectual shelling point idea yeah is it also seems to me that there is there seem there is a if you were to examine the shelling point let's call it a particular theoretical narrative and it could be about Euclidean geometry i from my conversations i've also discovered something delightful which is at high levels it seems like people are aware that that shelling point is wrong yes so i think about like in mathematics right i i when i was so i i started researching mathematics through logic and through philosophy because people making all kinds of ridiculous claims that might even be logically contradictory which brought me on this long journey and then i dive into the history of mathematics and the philosophy of mathematics and start talking to some high quality mathematicians then they go well look the the formal story is that mathematics is built on top of set theory as there are 10 axioms and zerno from you know zfc set theory okay but then they say yeah but it's kind of a religion it's not really some it's it's a story that we tell that gives structure to the mathematics that we're doing but you don't necessarily believe all of the axioms not like everybody has concluded that these axioms are true which is not something you learn when you're just when you're a student when you're a phd student and you're learning the theory because you actually think it's true yeah well i think that this is and it's it's interesting i think and this is something that i'm thinking a lot about which is the relation between the authoritative use of knowledge and the or the function of authority versus the epistemic function where there's an interesting thing about school where when you're in school you learn a bunch of things and they tell you the things are true and then maybe in the back of everybody's mind they know that in 30 years it'll be totally different okay well that's really interesting because the things are being taught as if they're true but the professors and the students maybe at least a lot of people recognize that the epistemic status is not we know that this is the case the epistemic status may be something much more like these are our current best guesses and you know they're not exactly guesses we have some evidence but we recognize that the tools are really early on and it could be that you know all this stuff ends up being changed after 20 or 30 years or something like that but so i think that a lot of times when people are interacting with intellectual material they a lot of you have to distinguish the authoritative use from the use that's designed to set you up as an epistemically equipped agent so imagine you know you've got a bunch of kids and they're learning math you know and they're like all right the uh four times zero is zero and it's like what's zero you're like well uh it's like nothing it's like if you have like one but you take it away like it's like so it's nothing yes it's nothing like how can you multiply a thing by nothing if you multiply a thing by nothing is it not the case that you are not multiplying um uh how can there be a relation between multiplication um uh you know a revelation of multiplication that itself has you know a lack of a rilatum if i may use the latin word um the child says to the the student teacher at that point it's like look just do your multiplication tables okay um and notably if the child doesn't do the multiplication tables they may have this cool philosophical insight but they're not actually going to be able to do the relevant math so there's certainly a role for the authoritative employment of these sorts of statements where they're like look this is just how it is or they say well there's a proof somewhere i don't know you know some important mathematician said so that that though doesn't mean that when you actually go talk to the relevant people who have the best command of the things that they're going to in fact the the people who merely receive the things authoritatively are going to have a lot more difficulty making really substantial breakthroughs i think that if you accept a whole bunch of things authoritatively you can make incremental advances but you might receive you might be in favor of challenging your assumptions but it's too hard you know how are you going to figure out which assumptions to challenge you received a thousand propositions you're going to challenge all of them at the same time well then it could be anything you don't know well you just challenge a couple which ones well you don't know right well so then i think that the people who and and so then i think that there's this difficulty figuring out when things are being advanced authoritatively versus when they're being advanced in a way that notes the actual epistemic status and one more thing on mathematics so for example the it's an interesting question what actually keeps mathematics together as a field you could say you know so first you know naive you might be well they all believe the axioms they don't believe the axioms okay we might say well well they at least all agree that you can infer these things from these other things except that you know there you might then be tempted to say well what exactly are the are the claims about i mean some of the mathematicians are gonna be formalists you know or they're just gonna say look you know so some you'll have our Platonists and say we're talking about mathematical objects here and then some you'll have believe that we're not talking about mathematical objects we're talking about concepts and you'll have some who say well actually math is you have things written down and then after we write down some things then we write down other things so the mathematicians don't have agreement at that level but there is a cohesion inside of the field of mathematics that keeps it together it doesn't actually split into the Platonists on one side and the formalists on another side so that means the thing that's actually holding it together isn't agreement on the propositions some of them are going to say like look just look at the you know we have practices that we after we write these things we write these other things okay so then that then it's a really interesting question what is the thing that's holding the whole thing together okay a few points first i think that's generally correct but in mathematics there do seem to be structures of math that are built out that are unique to the conceptualist school versus the Platonist school versus the formalist school so for example if you have an assumption about processes existing outside of the mind you might be more favorable to the axiom of choice which results in something like the Benaktarsky paradox which i would say doesn't make any sense but if you have a different philosophical assumption then you don't actually get some of those concrete conclusions in mathematics well except for the fact that then you'll notice that only philosophers will say the thing that you just said the mathematicians themselves will say oh well if you're working in you know ZFC then sure you get the Benaktarsky paradox you know Benaktarsky paradox thing and but they're not fighting about whether the things are true well that's not true though i mean so this this wild burger guy i'm talking about sorry sorry sorry let me let me go meta on my own comment okay the peep and this actually sort of this this so this actually illustrates the idea of intellectual shelling points sort of really nicely it's there's something in math that's causing the people to be able to still work on the same problems or work on closely related in the problems even though they disagree so you could that was i'd say that was a sort of more operationalist gloss where when i say they don't you know they're they're not making claims about what's actually true it's so that's that's not right so i think actually many of them will be making claims about what's true many of them will not be making claims about what's true many of them who make the claims about what's true will be making different claims about it and the fact that mathematics can exist as a discipline is with that taken into account is thus proof that mathematics is not staying together and mathematical progress is not happening because of that agreement mathematical progress is happening for a different reason and it's because the axioms serve this intellectual centralization function yes i would continue to add some small asterisk because it does seem like there are there are minority schools of heretical mathematics which would say that the progress that's made that's built on some of these assumptions that are dubious aren't even something they can work with it's like the you they there's it's garbage from the fundamental starting axioms so there's a few areas in which there's nothing you can save yeah and and and luckily the coordination mechanism the intellectual coordination mechanisms that hold math together are sufficiently strong that you can have people saying that inside the discipline and they still can get along with each other okay so there's two other things on this i really want to talk to you about um one is when you so there's when we're talking about intellectual progress one could look at the coordination of the practitioners yes what are the mathematicians doing or the mathematicians are kept together by some intellectual shelling point there's another approach which is to say the actual progress of the ideas so if so just to illustrate the principle if i were to talk about the progress that astrologers are making because they are unified around a particular shelling point and they're creating their various structures of knowledge they're built on well is that if well but is it are there such intellectual shelling points for the astrologers are they i don't know anything about astrology well so but the thing is this this there should be real examples here right because well that's i mean that sometimes that's what this is exactly my point which is that if we don't yet know the truth or even if you use special philosophical methods i know you're a fan of in order to figure out the truth in advance then you figured it out sort of well in advance of everybody else nevertheless then the group hasn't figured the thing out and there's this question of what produces the sequence of steps that then results in the people actually getting the answer and so then i would say that you know in order um it would be a very interesting phenomenon if there were intellectual shelling points inside of astrology that enabled the astrologers by their own lights to make progress um so my current view is that if you look through and look at functional versus dysfunctional fields the thing to look at is intellectual shelling points and so you'll you'll have and this is why you say well the thing to look at is the truth so i think you're on the one side and i'll say that it's not the case that that's exactly the thing to look at the thing to look at is related to the truth but it's not the same thing as the truth the so you you'll grant that there was progress in mathematics from euclid yes okay okay despite your thinking that that's not literally the ultimate final answer as revealed by philosophy okay okay and the practitioners themselves had all sorts of different views about um you know what really was progress and what really wasn't and so then my claim is this is the sort of the thing that's really puzzling about um advances in knowledge we don't actually have the answers in a lot of cases and so we can't use the answer as the way to tell whether we're making progress right right and that there's unique to mathematics and in some areas of philosophy because by virtue of the domain you're talking about a subject matter in which you have all of the tools to know the truth if applied correctly so like if you're talking about logic for example whether or not there are logical contradictions is something i think we can already it's unique in that we can know we kind of once and for all yeah so i mean i'm you know myself uh very partial to the idea that we can use logic and special philosophical methods like i think there's this like particular approach to philosophy that does yield knowledge um but the thing that's notable is that like compare math and philosophy for example um so whereas both math and philosophy use logic okay but math is very palpably making progress and philosophy is making substantially less progress well why is this okay if it's if they're both using logic as the sort of means for making progress right then there's this sort of important and interesting surprise and that the thing is working in mathematics and working substantially less well in philosophy well i think that this has to do with this is this intellectual shelling point idea where there the you need something to cause the researchers to all work together enough to be able to error check each other and give each other feedback and all of the like benefits that come from collaboration to get the views to the point of transferability and i think that the intellectual shelling points like are substantially better in mathematics than they are in philosophy so i don't think philosophy has made no progress people who have specific you know philosophical views will look back and say ah this one philosopher set a really good thing in this one text and then if you like look through the history of philosophy you're like progress is being made because more good arguments are being hidden in the massive sort of you know mound of philosophical writing um but i think that in looking at the accumulation of knowledge you want to look at what the group ends up adopting and i think there actually are a few examples in philosophy of things being widely adopted so one of them i mentioned before this is the hume point about us not perceiving causation and thus causation being something we conclude not merely from our experiences but we have to try to infer to it or something like that another example which also relates to induction i mean we're familiar with goodman's new riddle of induction so good so the uh all right well so imagine so let's let's go through this this is a lot of fun so imagine that you know you believe in induction and you have this bag of gems so you open up the bag and you don't know what colored gems there are and you pull one out and it's green oh is this the example yeah i didn't know the name of it yeah okay yeah this is goodman's thing um so but what why don't you go through it actually because it's a good case yeah it's really fun so you know so you pull out a gem and it's green so you're like okay green okay and you pick out another gem and it's green and you keep picking out gems you pick out many many gems and they're all green clock strikes 12 now you're going to pick out another gem and you want to use induction on induction the idea that the future will resemble the past but the superset resembles the subset or something like that the um and so because they've all been green so far you infer inductively that the next one will be green then some aliens show up and you're about to draw the gem out of the bag and they're like wait we've got a prediction here and we predict that the gem is going to be blue you're like well that's obviously you know contraindicated by evidence i have this long series of green gems that i've selected the aliens say well we'll hold on here green blue these aren't our native concepts we use this other concept which we call grew um grew means uh we've managed to figure out how to translate it it means in your language green if before 12 o'clock and blue after now all the gems that you have pulled out are both green and before 12 o'clock and so they're grew and so if all the gems have thus far been grew then and you know using induction it would stand to reason that the next gem pulled out would be also grew but there's you know the clock has struck 12 and so now a grew gem will be blue and so that means you should infer that the next gem you pull out should be blue not green and we used induction also okay so then you think okay that's a problem we're both using induction and if we're both using induction and we get different answers what do we do about that you might say well look i have normal predicates though i have using green and blue you're using these weird things grew and you know grew this is like obviously this made up thing um my predicate green is more basic um you have to define grew in terms of green and blue in this particular way and the aliens say whoa whoa whoa i mean we were going to say the same thing to you you have really weird unnatural predicates green and blue we have the natural predicates grew and bling of course in your language meaning green up until 12 o'clock blue after bling meaning blue up until 12 o'clock green after and so then we can define green as being grew until 12 o'clock and bling after and we can define blue as being bling until 12 o'clock and grew after so we're the one using the natural predicates and you're the one using these strange gerrymandered predicates okay so that's the so then that's the new riddle of induction the upshot from that is even if you just grant as an assumption that induction works you uh you can have opposing predicates such that if you do induction on the same set you get different answers and so then you know maybe you conclude induction doesn't work maybe you can include that there is some special way of picking out what the correct predicates are that you can do induction on but at least it's you know not a not trivial to figure out what follows even if you grant that you're allowed to use induction okay cool well so then to um you know bring this back the Goodman's new riddle of induction I think is broadly accepted by philosophers not the answer but that there is an uncertainty and then if you try to figure out like once you get into this perspective of intellectual shelling points what's causing people to sync up then there's this interesting question of what are the intellectual shelling points inside of philosophy I think that one uh at least a partial answer to that is that people can recognize where they don't have answers at least some people can recognize and there's all sorts of things you can do to cause people to recognize that they don't actually have answers and if you imagine all of these people who are trying to find the places where the answers are the least solid then that I think is one of the things that attracts into philosophy which which means that that's not like a full account of what philosophy is but I think that's an important part of it then you can look at like I think this helps to explain philosophical progress in the sense like in terms of what things get adopted and what things don't get adopted I think that Hume's observation about causation or his claim about causation not being perceived was largely adopted I think that Goodman's view was largely adopted very very little is largely adopted in philosophy but both of these things are examples of things that we don't know they're saying you thought you knew this but actually you don't know this here's how you can tell that you don't know the relevant thing so then I think that philosophers actually have a much more substantial ability to sync up with regard to knowledge of what we don't know well it appears to me we don't know a whole lot right yes you're a philosopher you've been attracted by that fact I suppose but it seems like a hard pitch to tell people I guess I suppose if you're at the set of people you expect to coalesce around these shelling points is very small I guess that's not a concern but the more I learn the more I'm going okay well we just there's all kinds of reasons why we just don't know very much almost but see it would actually be easier to say we don't know anything that would be the nice position like oh okay finally I understand we don't know anything that unfortunately we can't even take that position here because we do know a few things or exactly well so then this is this is I mean I think this is when you think what leads there to be successful fields versus not it has to be that the fields allow people to come together and think together despite disagreement because that's the whole point of the field is to work the thing out if you there wasn't any disagreement you know you would maybe you wouldn't even need to do that but so then I think that the in mathematics you have you know common language methods for checking whether something follows from something else you have commonly accepted axioms of course you don't actually need to accept them they happen to be fruitful if you follow these methods that we can show you how to follow or we can show enough people how to follow then philosophy doesn't have that there might be things in philosophy that have been discovered but this wouldn't be a case of knowledge accumulation because for it to get to the level of it in order for it to accumulate there have to be enough people all able to look at and interact with the same phenomenon which is why I think in philosophy knowledge of what we don't know has more of a tendency to accumulate than for example like suppose it's the same that space is discrete is actually true that claim has been being made for a while there's a bunch of different philosophers who make it and yet it's not really sticking exactly and that's because philosophy is this area that draws in all the people who have you know who are recognizing the sort of higher levels of confusions yeah well I mean I would also play lay part of the blame here on the mathematicians who's fascinated by the history of philosophy because I'm sorry the history of mathematics while the history of philosophy as well but there has been a consistent story which is that the mathematicians come up with a useful and fruitful formula yeah that that is logically ungrounded the the contemporary mathematicians at the time of that invention go no no it's fine it's good enough come on look at all these great things we can do with it the the philosophers and strict logicians went out to make better arguments and then like a hundred years later there's some update to the fundamentals and they go oh well actually yeah those criticisms right all along but now we know that calculus is on sound foundations until the next you know 50 years from now people will realize well actually calculus isn't still isn't on sound foundations well right right but I think what's happening is you're you're you're tracking you're you're tracking the question of whether the people have gotten the truth and the thing I would say is that despite the sort of repeated recognition that the thing isn't on solid foundations yeah progress is still occurring okay so my so my challenge to you is how would you explain progress occurring despite the fact that the foundations are not set up properly how would I explain it uh yeah because because you you could imagine and you know an argument that said well in order to make progress you have to have truths and be building on truths but you think the people don't have truths and are not build it you don't think they're building on yeah so I think you should conclude that they're not making right to put this more you know I don't know if this is actually represents your position but you know to argue it it seems like you should you should say that they're not making progress and I think that they are making progress well it depends on what the goal is so the goal is is okay so yeah yeah and then and then I think there's a way in which you suspect that they are getting closer to the actual truth okay here's here's my best guess right now yeah they are mining valuable gems that will eventually be polished and refined into the truth by future thinkers yes yes that yes that and and you'll notice that that's what happened in every successful science is that the case I don't I mean it's a claim yeah yeah so think so you know you could look at me you know we're looking at the history of electricity right now people will know more about the history of physics right but so you know the so in physics for example are we at the final physics no how do we know that because general relativity and quantum mechanics have not been reconciled okay so then quantum even quantum mechanics and general relativity are themselves in a way these not yet fully polished gems there's still something in there right and so I think the you know noting where the things are logically inconsistent or haven't been reconciled or the foundations aren't set up help to remind people that we're not yet there but I think that what's happening is we're accumulating value and getting closer along the way okay so a personal question and then I want to ask you one more critical question about from a systemic standpoint so a personal question is how how do you deal with the state of believing that the foundations are not set up and there are internal contradictions within our greatest theories of physics and and not get bothered by that I'm still in the state I'm still in this horrible state of thinking okay I'm some dude on the internet who happens to be curious about various things and I think that general relativity and quantum mechanics are wrong and I think part of the reason they're wrong is because space is discreet and that actually solves a bunch of problems but yeah but but I listen I consume content and I listen and I'm curious ago this is I know this is wrong I know this is wrong and I know nobody will listen to me because it sounds preposterous but I still have a I think a rather justified belief that in a whole bunch of areas of thought the foundations are poor which means a lot some of the conclusions are poor and backwards how do you deal with that on a personal level how do you not how are you not angry by that well is do you feel frustration um no no I I mean trying to remember I definitely like many years ago had many many arguments with people on many topics and found it to be sort of sort of shocking how hard it was to communicate certain types of things but since then I've spent a lot of time thinking about what the overall process towards knowledge looks like as well as things that can help to make it faster and then and and I definitely and so something like I've come to understand why a lot of the things are the way they are while at the same time and I'm still think that there are ways we can do things much better but so for example the take this idea about the discreteness of space and a sort of a particular logical approach to answering particular philosophical questions and imagine that we're in the world where that's actually right well we're not in a world where it's right and lots of people can tell right now that it's right okay what would we do about that on a practical level I like to shake your fist at the sky approach that's what I've been doing no but that that that's from the perspective of like an individual citizen now now occupy occupy the perspective of like a sort of responsible sort of governing person entity or something like that where imagine you're in charge of knowledge in the world here's the thing that I'm doing which I'm sure is not the correct thing to do it is to try to articulate the failures of the contemporary paradigms to say oh actually still these stories that are told about calculus do not pass logical muster yes so and so I think that that's obviously a useful function but now again I want to invite you out of like the sort of citizen role for a moment okay and imagine that you're actually designing a system of knowledge for the world okay like this is an important like very important point because uh you'd say that knowledge is powerful sure knowledge is very valuable yeah okay lots of people should have knowledge everyone everyone to have knowledge yeah okay um but it's difficult for people to tell what's true even in many cases after the things have been explained and then this you know there's a whole bunch of different reasons for this so then there's a question of what what do you do about this so should you for example you could say well I you know as Steve Patterson in charge of knowledge distribution for the western world let's say recognize that the discreteness of space is correct therefore promulgation and so you press the button and now everybody believes it right or at least you teach it authoritatively in schools it's true that most of the people can't understand why that's the case in fact only a very very small number can recognize this okay but they're told any way that they should accept this does that sound like the right answer uh well so it depends I guess on how closely connected you you see humans so like I running theory about the world is is there are there are a few isolated minds that get it and that understand things maybe at a higher level and just get to sit back and bask in their greater understanding of the world and that's the end of the story there is no button pushing there is no there there's no mechanism for getting those ideas out in a way that people understand them yeah I mean so I think that if you think about how you would actually set up a system to deal with the uh like the there's a whole bunch of different challenges but so I think for example um if you have a type of knowledge that is only accessible to a very small number of people and it can't be more widely replicated or checked then you're gonna have to answer the question well why should we listen to these people now of course those people will be saying well it's because we're right okay but other people can't tell that that's true right and so you could say well we should listen because we will imbue those people we will give them military force those people will have you know we should listen to them because they are in control of uh you know worldwide military you know something like that okay you could say well no we don't want that we want to um what we're gonna do is we're gonna have a system of general deference to you know uh the thinkers um and these thinkers even though people can't tell that they figured out a thing we're gonna have people defer to them I mean that I think that's gonna rub you the wrong way yeah sure okay okay so then um because they're wrong by the way because like historically the people would be in the positions of power and they would be wrong that's the reason about it but what if the people were were right and you had to have so this is this is I think this may be you know it's an interesting question because I know you have at least some libertarian sympathies um there's an interesting question here where I think that the thing that's happening in physics for example or mathematics where you have people coalescing around much more shareable and recognizable um and shelling points that uh essentially allow them to check each other and etc um I think that that system is in a way substantially more accessible than a system that had uh you know had it set up so that we would listen to people who could um work out the truth but we couldn't tell whether or not they'd actually worked out the tree does that make sense yeah I also I I guess in thinking through this thought experiment yeah I don't have a clear understanding of the value of other people discovering the truth like it's extremely important for me to try to discover the truth and it and it direct directs how well I live but I'm not actually sure how far that extends to other people well if that's true why are you frustrated um when you find that other people don't have the relevant truths it seems like maybe you do have some concern for whether they have I do have so it's twofold one is sometimes they get in my way right if they have bad political ideas or even bad cultural ideas I navigate the world I don't want to be stepping on porcupines and the other is yes some type of empathy I think people cause a massive amount of uh pain to themselves that's ultimately based in confusion and I don't like that yeah I mean this is so I I completely agree with you that there's tons of unnecessary pain that comes about as a result of confusion and then there's just a question of how do you structure a you know on a smaller scale there's how do you structure research programs and then on a larger scale there's how do you structure societies to be able to engage in this process of intellectual investigation and discovery and so I think that the um as like a sort of you know more like a sole a solo investigator um uh it makes sense that you know you would look into methods that only a very small number even if the methods work only a small number of people have ever been able to use them okay and so then but then in terms of thinking about what's happening in these other disciplines I think uh a thing to look at is like look at the size of the disciplines look at the total number of people involved right if you have you know a hundred people and maybe if you have two people or three you know given the current state of like take your favorite philosophical method maybe you could get two or three people synced up on it okay but if I give you you know 10 you know smart able honest you know etc philosophers it's gonna be really hard for you to get them all to be synced up on the same method right and so then what one way to think about this intellectual shelling point idea is to look at the number of people it's able to sync up into a research program and so I would say that the research program size that can be supported by your favorite philosophical methods TM is going to be actually really small whereas the size of the research program that can be supported by Euclid's elements is actually really large and that's true even though Euclid didn't have a full axiomidization that you know you know it's non-discreet and you know maybe empirically disconfirmed by uh you know modern physics okay so the other question I wanted to ask you about the intellectual shelling point idea is uh you have spent your time um thinking about a lot of things and engaging with various people at high levels um more in the academy that I have right my engagement with people in our contemporary academic paradigm as a lot of interviews on their show some personal conversations throughout the years but I imagine in terms of gross numbers you probably engage with more than I have and uh my observation I wonder if this is correct or not is that if you were to line up 100 PhDs who've gotten their PhD not PhD students because this number would be worse with them 100 people who are certified doctors of x yeah and ask them about let's say the axioms of their discipline yes a large percentage would be so early in their intellectual research as to believe the axioms are true because that's what they were taught in school and however a small amount of really high quality minds would get it they would totally understand that you know these are the current paradigms are definitely wrong when we're talking about sure things like so so if I would have put numbers on it I would say just in my you know you go to a faculty like I actually had to do this when I was traveling around interviewing people as I would go to the faculty pages of universities that I was in proximity to and try to find high quality researchers so I would go through and like do a little bit of reading and look at their CV and be like oh this person's guaranteed to be terrible this was terrible and I would try to find the really good ones and I would still sometimes be grossly underwhelmed so I would look at it something like nine out of 10 eight out of 10 maybe more than that don't even grasp that their the the paradigm and the shareling point that they're working around is definitely wrong do you think that's an overestimation um I think that if you we have to get the sort of the claim exactly right but I definitely think that a very large proportion of people in the fields don't recognize importantly recognize the epistemic quality of the material in the fields so I put it like that so to make it more concrete like in mathematics the majority of mathematicians that one encounters actually maybe even more so in math than other disciplines I'm not sure but the majority of your average mathematicians are not going to have deeply thought about the epistemic status of ZFC that set theory they're not even they might not even entertain the idea that those things might be completely wrong um yeah so I think that most of them will not have looked into it deeply it's an interesting question how many will have had one day you know I wonder if the axioms are you know in fact not fruitful or in fact you know incorrect do they match reality are their Platonist objects that match them etc I'm not really sure about that percentage but it's it's notable that I actually think that it's good that most of the people aren't especially fixated on that point like if I could I would want to have many or most of the people clued into the epistemic foundations question and so forth a bit but it's sort of like you know there's currently an intellectual division of labor and so you know and you want to have people making progress on the fronts that where you actually can make progress in some cases the um you're going to end up getting progress from attempts to push the cutting edge of the functional research program rather than critiquing the relevant research program so like a good example is Newton with um the perihelion shift of Mercury so Newtonian physics has this nice counter example which is the way that the orbit of Mercury around the sun shifts on its axis over the course of time Newtonian physics doesn't explain it this was known well before we got relativity relativity explains it better but you don't want to have all the physicists drop everything and then try to figure out what to do with the anomaly you want some of them doing that and then you want the others pushing regular Newtonian physics forward and so the thing that I would say for your particular like research program or angle on this is I would suggest that you want there to be enough people trying to examine the foundations enough people trying to figure out whether the relevant philosophical methods are right and so forth rather than everyone so like the the focus on the percentage I think is not the correct focus I think the thing you want to ask is is there a high quality probably small which is the right size for it research group that's working on this right now and if not then there's a question why not maybe it's too hard to form one of those um but that's that that would be the place to look rather than causing everyone to be concerned about this right now okay how do you deal with the following observation that there are various domains in which as a result I think of the lack of rigorous foundations people end up profoundly getting harmed and I think of nutrition research as an example like my wife has has I think come into some pretty high quality theoretical concepts about human health and nutrition and if if her theory is correct then the contemporary paradigm in medical establishment included is way off base and probably harming people a lot so so I look at that I think okay well that's because their their philosophy is totally mistaken they think I would have input it in terms of uh like metaphysics right they're they're they're not recognizing that objects are not isolated from one another they come bundled with relations and this part of the system the body system affects this part of the system which affects this part of the system so when you intervene you actually can cause a bunch of problems that you didn't see so so how do you deal with that objection to say we need more rigorous foundations look at this example of nutrition science yeah well I mean this is I think that this is why these things are so important I think that we absolutely need to make a bunch of different changes in order to improve the quality of the research that's happening and then the question is and and from the perspective of the final completed science of nutrition I think that a bunch of the things wrong in the foundations are going to like will have been responsible for people giving bad recommendations and people not knowing what to do with their health and I think so but then there's a question of what do you do about this where most of the researchers just aren't aren't equipped to challenge the foundations themselves um and so thinking that that's the sort of point of interventions it's like I agree with the problem and then the question is what's the correct point of intervention I think the the correct point of intervention here and nutrition you know so I study early stage science there's this question of how do you how do successful sciences get started and I think that nutrition or I'd say even more generally health is a really good place to look the my guess is that the problem is going to not be as much a failure to build up from rigorous foundations as opposed to a different problem which is not having as many researchers build a generalist models in a way where they can all then check against each other and improve the generalist models so my guess I don't know but my guess is that in you know the case of your wife and nutrition like trying to figure out what's happening there it wasn't an attempt to actually derive everything out that's true principles but it but what it was was a substantially more generalist approach that tried to then take so it's like okay why do we have in a bunch of cases like less than stellar advice coming to us from the medical establishment well it's because in a lot of cases they don't have the answers okay why do they not have the answers well a lot of people would love to get the answers you got a lot of well motivated people and you know if somebody who figures out the true theory of health will also get status and prestige and all of that and so people like to say well it's incentives there's like more to be said there but I think the thing I'd say is that there really aren't well established methods for approaching things in a generalist way so this is something that we see as we examine the history of science early on people are much more there's a transition that happens from the early stages where you have substantially more generalism through to the late stages where the thing is specialized and precise with an established terminology with an established set of methods that you can use and so I suspect and there was an absolutely enormous increase to the total amount of funding that went into academia and scientific research over you know since world war two the last 75 years and I think that if you add so many people like that it's a pretty difficult thing to ask for to ask for all of them to invent new methods and etc but with so many people it's really likely that the thing that will happen is that people will try to find methods that are much more established much easier to agree upon and so then the my current guess as to what's happening here this is something I expect to research over the next little bit but my guess is that you have a very large number of people doing research using established methods and you have substantially fewer people doing research in a way that matches what happens at the early stages of the development of sciences where people are you know developing theories with less standardized terminology doing things that are qualitative rather than quantitative it actually takes a long time for a field to get to the point of being able to be quantitative developing instruments that relate to the theories like we've got a whole bunch of really advanced instrumentation to help us understand what's happening in people's bodies but we don't have like a health meter we can't like point at the person and be like okay your health score is 72 and that's because we don't have theories about what that is and and so then I would suggest that the real problem isn't building up from the foundations the real problem is not having basically having not having enough research effort going into the development of early instruments development of generalist models and things like that that would specifically aim to have us understand things like health I think that's largely correct though I guess I would say when I'm thinking of building things up from sound foundations I'm not necessarily proposing that we come up with a full and complete metaphysics of health before we discover anything it's more like it it's more of a methodological point that if you're the correct approach to health research is one that recognizes the value of the generalist model so it would be the one that recognizes the complexity of the discipline and how when trying to gain knowledge about health we have to rely on anecdotal experience and we have amazing tools about getting higher quality data you know on the internet at a low level so it's like a methodological point about approaching it from the right mindset yeah yeah and and so there I think this is the sort of place where you have to separate the there's the question of how do you acquire knowledge and there's a question of how do you fix bureaucratic systems these are or two separate things but like if you just all of a sudden had you know practitioners sort of open the floodgates so that they would now listen to a lot of anecdotal evidence then well then you're going to end up with a lot of variation in recommendations that comes out of that and then how does that fit with our legal system somebody gives a recommendation on the basis okay well so you could say well we need to fix that well yes definitely we need to fix the thing but let's separate the problems right so one this is why I think that I'm most optimistic about there being special research groups and teams that make progress in uh sort of like they would make progress with a sort of generalist type model and get it to a point that it was then that it it then became possible to spread more widely it's something like again this has to do with looking at I guess there's a simple model on which you'd want every researcher to be looking at every question but I think actually the researchers it's something like you want to have the right design for research programs so so take mathematics again the before Euclid the check this I'm pretty sure Pythagoras predates Euclid the you got a whole bunch of different methods for people coming up with mathematics but the it's not something that's nearly as shareable right so you had Pythagoras's group they were all on a boat they were all you know special you know methods for eating a special method of life etc etc and that you know there you have a small group that's more tightly coordinated that then is going to be able to share information internally better whereas and that's because you don't have the axioms as soon as you get to Euclid's axioms and postulates and so forth then it becomes possible to loop in way more people and so I think like this this the same thing is you know true for and we said we're studying the history of electricity it's like pretty hard to study to study electricity early on you've got like weird objects occasionally attracting each other yeah like a whole bunch of like you know people saying that diamond attracts this and amber attracts that and so forth and the lodestone attracts in this way and the lists are frequently wrong and nobody and then you get to a point where William Gilbert publishes this book in 1600 where he talks about this device the versorium which is essentially this needle that you can put down and allows you to pick out electric attraction before you have that it's at least like makes sense that it would be substantially harder to sync up on things after you have that instrument it's easier to sync up so you could imagine in the case of health like a very late stage device as we've got our healthometer where we you know scan the person we see how healthy they are and you're like how does that make sense it's healthy even the sort of thing that admits of a well according to the advanced theory that we'd have at the time health would have four components and you scan them in the following way but it's obvious we don't have that and we're not going to produce that next the thing that would happen way earlier than that is that we would have a less good instrument and before that an even less good instrument but the development of instruments help to make the discoveries and theories and so forth much more transferable so I guess what I'm saying is the thing that we should expect to see in the development of any new field anytime you look at a field and you're like this isn't working properly you want to look to see is this the sort of thing that in the true story of how it came to work properly it makes sense to imagine that it would go through a sequence of people develop initial theories there's lots of disagreements someone comes up with an instrument a lot of people coalesce on the instrument it refutes a bunch of theories someone develops a new theory there's lots of testing a bunch of anomalies are discovered you get a new instrument and then eventually we have the healthometer okay so uh come into a close here we've already gone a little bit over time so I appreciate you being generous here when you are looking at the big picture of scientific progress or knowledge production at present yes do you think in general you know we're at the peak right there hasn't been a higher point that we've we've been before and I'm generally optimistic and thinking okay well we've done a pretty good job or do you look at it and say okay though we may be at the peak like if you're looking at the exponential curve of how these things should accumulate we're like right at the bottom and things are really terrible right now my of course my orientation is I generally look at the state of knowledge production thank god it's it's terrible but but you've done more research than I have here um I would say that it's like overall very good in terms of the you know terms of world history uh and then I would go on a field by field basis basically um and then I would look at um whether the field is itself an early field or a late field I would look at whether the people are using the sorts of methods that should be used at the relevant stages I'd look at whether incremental advances are the right things versus whether we need paradigm shifting revolutions um and and then I think that my scorecard comes out pretty similar to what I think a lot of people believe where it's like the hard sciences have done really well the soft sciences have not done nearly as well one place where I'm I'm very optimistic is that I think that in a lot of different places where there hasn't been much knowledge accumulation I think that in those cases there have been advances in the past that just haven't managed to become part of a standing tradition um and so there have been discoveries without accumulation that's one thing um and then I'm just also optimistic that we can make a lot more progress yeah one area that I just recently discovered and and flattering my curiosity that I've been surprised where I look at the structure of knowledge I go damn this this is actually pretty good the only one I felt that way is chemistry as part of the reason I've just been consuming yeah chemistry content and I think okay there's this really interesting combination where you don't have to have the fundamental theory like yep at the base level we don't have physics sorted out so we can't have chemistry sorted out but there's a there's enough of a of a low barrier to entry where anybody can order the right chemicals online sometimes in the legal gray area they can literally in their garage or basement whatever do these experiments themselves to confirm the results and that happens there are these basement chemists out there that are doing that I'm thinking wow that's really that seems really unique and awesome to me it's not so with physics right you can't just build your large hadron collider you know and like experiment things in your in your basement right now to that level yeah it sounds like something you're really excited about is the democratization of the process where people have the ability to check the things themselves I like that I think that I mean it's an interesting question because you have to ask how much you really want people building things in their basements like to some degree um like you know there's new synthetic biology capabilities and do we want people to have the ability to synthesize absolutely anything sounds like no um but I do think having people have the ability to check things seems really good one idea I really liked uh was you could imagine having some sort of substantially more accessible to the public it's not exactly like a museum but it'd be more like a cross between a museum and a lab or something like that where you could go in and look at something like reconstructions or replications of all of the progress that occurred across history so that you can see like here's the original experiment that helped us to distinguish magnetism from electricity and here's the experiment that was done that showed that Franklin's theory of electricity was wrong and here's the and and sort of do the whole thing from the perspective of uh the people who are making the discoveries along the way so that not only would you have the ability to check but you would understand why it was that each of the changes had occurred I think that that would the checking function is good but I think we also want people to understand where all the things came from I think that's a that's a really cool idea I would pitch that to somebody and I would definitely attend I would be an early attendee of that museum awesome all right hey this has been a great conversation Jeff I appreciate your time yeah yep I really enjoyed it all right bye bye