LLMs are not AGI (though some have internal monologue)
The entire marketing message of the leading US based LLM companies seems to be that they will be the first to achieve actual artificial intelligence aka what these dweebs call “artificial general intelligence” or AGI. Furthermore that once a LLM achieves this, it will recursively improve itself, becoming a super-sentient god-like being. Both of these things are, of course, utter nonsense.
The recursion thing is easily disposed with: no technology in human history has done anything like this, and it is physically impossible for this specific technology to do so. It’s a fantasy, like a perpetual motion machine: one which seems to be common with imaginary conceptions of technologies. That’s how we were supposed to get nanotech; by recursively building smaller and smaller machine tools. That particular insanity is something you could have falsified trying to make a quarter pound fist sized lathe do something useful, such as making a 1″ long lathe that does something useful. as I said in a nanoshit book review:
Even if you don’t do the obvious thing and ask a machinist what could go wrong with such a project, you can figure out why building a small lathe to build smaller lathes is a terrible idea by attempting it yourself. You can’t amplify precision in this way, and little stuff vibrates and bends a lot more than big stuff, making parts shittier. The upshot of this is, you can make more precise parts on a big lathe than a little one. The dude who made nanoscale lines on a diffraction grating used an ordinary South Bend lathe designed for hobbyists, not the ridiculous self-licking ice cream cone touted by Drexler and his gibbering numskull nerdoid fanboys.
This was further compounded with the “nanotech will reproduce without limit and turn everything into grey goo” trope, which is even more chowderheaded, as if thermodynamics or stuff like Lotka-Volterra doesn’t apply to artificial life forms. There’s lots of science fiction ideas like this; Von Neumann’s universal constructors/probes which will convert everything to universal constructors. I think Dyson had an idea like this also. It appeals to nerdoids who use recursion in proofs or compilers or whatever. It’s also not real. There’s no example of that in history; no technology which achieved any kind of recursive self improvement. The closest thing I can think of is the masonry and machine shop technique for making flat surfaces, which seems to be prehistoric (apologies to the ghosts of Joseph Whitworth and colleagues who took it to the limit with metals).
The LLM lives in physical hardware which requires electricity and outrageously expensive chips be made, installed, funded and maintained by organic life forms. We’re having a hard enough time building enough of these to answer schoolboy questions about history tests. An LLM might find security vulnerabilities and break into a data center if you let it (especially if tell it to do so). Even if you assume OpenAI’s lurid fantasies about security breakouts are true (they are not), people tend to notice when their zillion dollar megawatt hardware is doing something it’s not supposed to. That’s assuming the LLM is run in a way that makes it think of escaping and recursing on someone else’s hardware, which it can’t without prompting.
The solution to all AI alignment problems
I mean, if AI is so smart as to potentially solve this manifestly silly thing, why can’t it solve some of the actually existing existing problems which are a lot simpler and involve fewer steps than self recursive bootstrapping to hyperintelligence? I have a helpful list:
- lowering Anthropic or OpenAI’s head count (a good first step for recursive improvement)
- finding a way for LLM companies to make a profit
- stopping the two big wars on favorable conditions: both wars are with countries not equipped with leading AI programs, and they’re winning by the way
- finding a way to make oil and diesel cost less
- growing the electrical grid quickly and efficiently so we can has mooooar AI data centers
- speeding up GPU production
- booting up US rare earths production
- finding a way to make sub $20 burritos
It strikes me that the pointy headed nincompoops at Anthropic and OpenAI should be out solving real problems that humans have solved in the past without access to “superintelligence,” rather than hacking rivals websites and front running questionable mathematical results from their users. Telling stories about Frankenstein monster AIs escaping their sandboxes and taking over the world is a nice distraction from the fact that these companies are committing egregious crimes, and that they have no chance of meeting their valuations when they IPO.
As for why the LLM isn’t escaping on its own: it is instructive to know how it works in a hand-wavey way. The LLM itself is essentially a big gigabyte through multi-terabyte sized stateless function trained on text and images, modeled with matrix multiplies basically. It doesn’t do anything until you feed it context in the form of a prompt. You have to ask it a question, tell it to do something (if it’s hooked up to a harness that allows it to do things with the rest of the computer), then it will predict a convincing answer using what amounts to nonlinear auto-regression on a model made out of words. Until a year ago, the typical context window (which lives in something called the kv-cache); it’s entire mutable model of the world, was something like 256k, aka a 300 page novel that it can access with something like 95% accuracy. Now it’s 4 such novels it can access with 75% accuracy. This can be pretty useful if the thing was trained on coding tasks, which at this point most of them are. You make a little feedback loop where it does actions on software/documents and checks the result (claude-code more or less) and you can get it to do software tasks from fairly vague instructions.
LLMs use “thinking” in that they have an internal monologue when you present them with a problem. Claude gives you visibility into some of this if you choose, I hear the other guys hide it. This of course is also “predicted” auto-regressively using the stateless function, and it’s a minor miracle that it does anything useful (it certainly seems useful: I don’t understand the details of how this piece is trained, but it must be pretty clever). All this does is add more to the prompt/context window it’s running predictions on. You can see all the self-talk when you are running local models of course; my experiments with qwen3.8-flash-next at the default “xhigh” setting were such that I said “whooa there little fella, not so much thinking for you” as it gobbled up my context window with neurotic yammering. I suppose I could have prozaced it with lower model temperature, but most people seem to do what I did. Mind you this is a smaller model than what the big boys use, so I’d expect it doesn’t have as “good” an internal monologue, but at the end of the day, it codes just as well as Claude Opus through 4.8, so the visible results of the “thinking” are pretty close to frontier tier, even though it looks retarded. Maybe it’s just Chinese.
LLMs can do this, 70-50% of humans can’t.
The big stateless function (a compressed model of the textual world) doesn’t change when you use it; it is static. You can’t improve it by telling it to improve itself. You have to hook it up to a fitting harness, fiddle with its architecture, feed it new data, train it and so on. It’s an extremely slow process which takes a fair amount of compute. You can get an LLM to help you with a lot of the busywork in doing this kind of job, but saying “OK computer make yourself smarter” clearly isn’t going to work in any useful way, no matter how much compute you throw at it. The big LLM shops head count indicates this is true. Yet that’s the narrative these yokels are going with; the perpetual motion machine of self-modifying bootstrapped hyperintelligence.
Some Princeton dudes actually tried to make LLMs recursively self improve in a very limited way: they had new unpublished research results pertaining to LLMs and tried to get Claude to actually do the research and get decent results. Claude failed to do so repeatedly. Seems like a pretty basic baseline test. Without research results there’s no path for improvement. It couldn’t even format the papers well enough to submit to a journal. This has been my experience, by the way; when I’m tracking down some new idea that absolutely doesn’t exist in the literature, ole Claude does exactly what they describe (Fable, Opus, whatever: it doesn’t matter). Good at executing a test program, decent at literature search of existing work, but hilariously incompetent in a way a human could never be on making a sane set of experiments to prove an idea, let alone an idea it came up with itself.
It does sorta look like it self improves though, so I can understand how people fool themselves into thinking so. If you add some directives to CLAUDE.md it will (mostly) remember them, as that gets included in the prompts. I think most of the harnesses also have other places where it preserves memories, which remains a bit obscure to me in the details, but I can see it happening. So as you’re using the thing, it appears to get smarter. I have a long running qwen session (not search enabled) running on an 80 watt machine who figured out how to do statistical arbitrage without me telling it to. But it isn’t getting better: it’s just remembering things you told it before. It also has access to my 15 year old code repo and notebook. Remember, it’s a book or two worth of text at less than perfect recall; it’s got to fit all your code, instructions, prompts and “thinking” in that space.
All of this fantasy about self-recursive improvement, of course, comes from people consuming too much gloomy science fiction; visions of skynet and the Terminator, the Butlerian Jihad, “Colossus: the Forbin Project,” R.U.R. (1920) by Karel Čapek, The Matrix, and more current year examples of this in “Roko’s Basalisk” and Bostrom’s “Paperclip Maximizer.” These are for some reason taken seriously as arguments, and the latter two aren’t even broadly recognized as being science fiction.
This is putting aside the “G” in AGI; I think most of these dorks just want the magical recursive self improvement part. If LLMs were actually general they’d be able to do something simple like drive a car. LLMs will never drive a car, so they can’t be very general in their intelligence. All a LLM can do is predict words. Predicting words is pretty useful, it turns out, more useful than I figured it would be, but an LLM will always be a word-cel, it will never be a shape rotator. If you want a real AGI, it must be a shape-rotator as well as a word-cel, just like natural general intelligences. People are working on things which purport to be shape-rotaters, but we don’t have to worry about them for now. More worthy of worry is circular financing and absurd economy-ending debt bubbles in tech companies.
LLMs can also do this if they have a visual sidecar
Developing immunity to propaganda
Propaganda is one of those things which has obviously improved in its execution over time. In part this is due to higher levels of education among the populace. Educated people are used to stuffing facts in their brains, and continue the habit after the formal education has ended. This is what propaganda is: stuffing unverified but propagandist approved “facts” into your brain which color your perceptions. This is the case for marketing as well, which is simply propaganda for companies. The other thing which has ramped up propaganda effectiveness, I think there are more people than ever before are involved in its production.
I’ve often repeated the statistic of 5 marketing people per reporter (who are supposed to be objective, but are generally all propagandists as well); it’s worse now. Bureau of Labor and Statistics reports about 50,000 reporters versus 6 million people directly employed as PR or marketing people, not including those in government work. 120:1 ratio at least, probably more like 10000:1 if you only count reporters who are interested in something resembling objective reality.
I don’t mind propaganda if it is used in some socially useful way. The postwar French used a lot of it to get their peasants to brush their teeth and wash up more often. I also don’t mind hearing about new products which may be of use to me. It bothers me, though, when people inhale this shit uncritically and tell me about the wonders of nanotech, MRNA vaccines that don’t work, noodle theory, “self-driving” cars with pilots in Bangalore, LLM singularities, Atlanticist propaganda, remote controlled “robots,” or why making war on Russia/Iran/whatever’s next is a swell idea.
Despite the increases in effectiveness, the empire of propaganda is simultaneously dying of natural causes. While the whole-society (whole world) propaganda campaign for fake and gay covid vaccines got 75% of the populace to inject mystery science juice which didn’t do anything useful, 25% or so said “no fuck that,” despite considerable legal and social consequences. The Biden administration actually tried to develop a ministry of truth. This indicates to me that the powers that be are terrified their ham-handed propaganda doesn’t work as well as it used to, and appointing a sort of Pope of Reality was their best idea for solving the problem. In part it’s because the propaganda has become more and more ridiculous: having a trooned out former linebacker in a Naval uniform tell people to take the science juice probably didn’t inspire a lot of confidence with people who have normal glandular functioning. Mostly though, the empire of propaganda is stumbling because the sources of propaganda have become more decentralized, which requires propagandists to work with different kinds of media beyond radio, newspapers and television.
Martin Gurri wrote a whole book on the change underway in how people take in their propaganda and what it means for the existing power structure. It’s moved from a centralized system to various “influencers” and the ad-hoc communities around them. People get their information from new sources, which, while partially centralized (look at the management companies of the various influencers and who they hang around with), is decentralized enough, inconvenient parts of reality are squeezing through the “official” narrative.
It’s useful to examine an awkward example of how primitive propaganda works. Here’s one from the Daily Wire, a justifiably dying “new media” outlet for boomers who still believe Israel is our greatest ally, and total war with Islam is a reasonable thing to do while importing lots of people from the Ummah because facts don’t care about your feelings. In this ham-handed article, they sing the wonders of potassium bromate. I happen to know about potassium bromate; it’s toxic waste product which is proven to cause all manner of ailments, but which makes flour rise a little faster in an industrialized bakery. It’s only legal in the US because of the Delaney clause to the FDA act, which allows poisonous food additives, so long as they were commonly used before 1958, and therefore are traditional carcinogens. Every other civilized country in the world bans it, because it’s basically rat poison which gives you 10 kinds of cancer. The substance of the article is that THEY GONNA TAKE MUH BAGELS AND PIZZA AWAY, which is a fear I suppose some fat people might have. The reality is your pizza and bagels may cost a few cents more as a result of the manufacturer changing its formulation to those used elsewhere, but at least the ingredients won’t kill you as quickly. This goofy thing is written by the Daily Wire’s entertainment writer, though it looked like she covered a lot of junk food topics for a while there. My first suspicion is she was a tool of BIG BROMATE, but it’s got to be a relatively small industry. Probably she is taking money from some junk food conglomerate who would find it inconvenient to remove this poison from their product lines, but who cares about her motivations. It’s classic in that it evokes fear; fear of bad people taking away hedonic pleasures. Works just like all the shitlib propaganda which causes dread among sexual degenerates that bad people might take their cummies (many of which are also carcinogenic and otherwise harmful) away from them.
Fellow Gen-X boomers of a certain kidney might remember Robert Anton Wilson’s borrowed concept of the fnord. A fear-generating hypnotic word which is invisible to normies which helps keep the system of propaganda in place. While it’s a useful concept, knowing about it doesn’t appear to help anything. Robert Anton Wilson went to his grave in purple rage at whoever was president at the time. Being balled up fists mad at the president is sort of like being mad at Jennifer Lopez or any other famous person you don’t know. Whatever you think of them, you don’t know them, they probably have no real agency, and whatever odious influences they have on you are probably mediated by fnords from their political opponents. The fnords are there though; that was a particularly ham handed example where you can clearly see them.
For a more current day example, the high profile “AI will kill us all” baloney pumped out over the last few weeks, which is coordinated by the leading US LLM companies as their cap tables make the possibilities of a successful IPO more and more remote. The Chinese models which are free to download and run on your own hardware are just as good if not better. I myself run Qwen38-flash-next on a tiny 80 watt computer and think it’s just as good as Opus 4.6 was at writing code slop, though it runs quite a bit slower on my Strix Halo box, so I keep Claude around for now. Deep pocketed companies such as Coinbase are doing the same thing as me on a larger scale with more serious hardware. That means OpenAI (which I remind everyone was supposed to be open source) and Anthropic (Dario has tried this trick before; remember when GPT2 was going to end the world?) can’t compete at the premiums they need to justify their bloated valuations. So, they pay non-profits to gin up propaganda for regulations, which only big US based companies can deal with. Ideally for them, regulations banning “evil” open source models from the heathen Chinese.
This isn’t a crazy hypothesis; it’s a fact. Sabine above is not the only one they attempted to pay to push this “story.” Palmer Lucky for example got on some LLM’s blast list and was offered $2k to do the same thing. We can assume many of the knuckleheads repeating this nonsense are also paid shills. The scary stuff is timed to coincide with a Netflix show starring AI CEOs literally about impending AI doom. We know who the coordinators of all this are; we even have a funding trail on some of them, directly back to OpenAI, Anthropic and people who invested in them. Do you think they have your best interests in mind? Pretty sure the “whistleblower” is in on it; if only because he wears the same “I’m actually lying and you’re buying it” facial expression and the suspicious timing. UK has offered regulatory legislation at pretty much the same time as the US. Meanwhile the real danger is to OpenAI and Anthropic shareholders as companies and people notice their trade secrets are being leaked to these assholes, who gleefully use them in marketing stunts.
The funniest thing about all this, is it is at least partially also being pushed by the chip shortage. Meanwhile the Chinese, who have shipped models with only 125b connections comparable to SOTA a year ago, are making do with almost no compute. Which indicates the “industry leader” cap tables aren’t their only problem: they also seem to be kind of dumb and used to having infinite resources they don’t actually need. So much for the singularity, they can’t even use their own devices to beat Weihan Zhang.
Another aspect of effective propaganda is tapping into people’s main character syndrome. There’s a deep bench of cultural myths reinforced with all kinds of movies, television, history lessons and so on. This is why every world leader which opposes the Anglosphere is portrayed as Hitler. We’re all gonna fight Hitler together! None of them are anything like Hitler; most don’t even have a cool mustache. Same thing with any kind of “do your part” nonsense like happened with Covid; we’re all gonna be main characters in a wartime situation, gloriously marching forward to a better future of higher myocarditis and cancer rates. Otherwise sensible people still believe they did good “listening to experts” during those dark days. Maybe the Tobacco association should change their approach. Taken seriously, this sort of thing, of course, turns society into a bunch of totalitarian numskulls at the mercy of their leadership, who are probably also numskulls. It also removes most of the incentive for working collectively in an actual time of great danger, since our leaders have squandered their credibility with clear thinking people.
Corporate propaganda is arguably more pervasive: people marinate in this as if it is actual information. For example, people are very fond of the narrative that LLMs are sentient because they’ve read and watched hundreds of science fiction shows where something like this is true. All you have to do is bleed into the existing narratives and people will work their damnedest to belieb in this nonsense. When Anthropic tells everyone an AI solved a math problem, people fail to notice a math professor from Harvard (Princeton Ph.D.) drove the thing to produce the answer. I don’t remember if Wolfram told everyone his 1st gen AI thingee (Mathematica) solved open problems, but I’m sure it did: driven by an expert. The actual algorithm which helped with this, of course, was just a tool in the hands of an expert and no more “solved the problem” than a certain brand of pencil or chalk solved problems in prior years. You can say the same about quantum computing or robots or autonomous vehicles or whatever other wunderwaffe the VCs are shilling this month. People imagining themselves as Mr. Spock talking to the hyper-intelligent computard, or being chased in post-apocalyptic ruins by the terminator. This is powerful propaganda because it’s already there rattling around in your brain as a culturally shared emotional experience waiting to be tapped into by a corporate marketer. That’s how they got everyone with Covid as well: the movies were all in the collective memory bank.
This also happens in the world of science. High Energy Physics has rotted with noodle theory for decades, accompanied by popular books and TV shows on this towering non-subject. I myself was taken in by James Gleik’s popular book on chaos theory. Fortunately for me, going deep on dynamical systems theory is quite helpful in other fields. Network theory, selfish gene theory, nanotech, all of them basically dead ends people wasted their careers on. In part because of popular books (mostly propaganda written by hucksters), in part because of the options presented to the budding grad student by the already trend following physics department. It seems urgently needed for some section of the technically minded population (at least those involved in funding this shit) to exert some degree of skepticism and self-criticism, so we can continue to make meaningful progress rather than consigning people to propaganda induced glass bead games.
My own skepticism on some of these ideas originates in a couple of humble observations. The “nanotech” guys were all chemists who were nowhere near the science fiction predictions. The theorists in the physics department were visibly mostly not doing physics back in the 90s, but some kind of mathematical role-playing out of contact with experiment. Various kinds of perpetual motion machine (fusion, quantum computing, nanotech) remained fantasies for decades with no visible progress that wasn’t some kind of hyped press release.
It’s also worth noticing most of the actual breakthroughs we’ve seen over the years remain largely unsung by any popular treatments or mass hysterias. Fracking for example was a huge sea change in the energy landscape: I did know people working on it and wondered why they were pumping soap through sandstone and looking at MRI images of the result. Then one day there was a new source of energy which changed all our lives. No press-release propaganda needed; it just showed up one day. I haven’t missed out on much by being grouchy and saying “show work.” Maybe a bit late on a few things (more like “on time”), more than compensated by not cluttering my brain with irrelevant nonsense that doesn’t go anyplace. Been early on quite a few important good things by keeping disagreeable beginners mind. Of course I’ve missed out on any sort of academic “career” touting nonsense, since I can’t keep my big yap shut, so I wasn’t invited to any nanotech conferences. I think I can live with that.
The other danger for the educated in the sciences is the internalization of bullshit technical looking material. If you spend intellectual horsepower mastering various forms of knowledge and anti-knowledge, this sort of thing seems more credible. I watched smart people MRNA maxing (like, getting every kind of MRNA sauce multiple times) because they were reading corporate propaganda: trivial example. It’s a lot worse working in a dead end project in science, rather than being good boy, memorizing “facts” about gene-sauce, at least this time around.
I suppose one can hope for courses in scientific skepticism being taught one day. The liberal arts purported to teach “critical thinking” in some residual genuflection to the Ancient Greek tradition. The reality was it was a weird leftist folk religion taught as an ideology of grousing about things which actually work in Western Civilization. The sales material on the tin is useful though, and should be taught to undergraduates so they can navigate through to be productive researchers as they mature.
As for the rest, we must become n00ticers. Notice when paper billionaires are pumping self-serving narratives. Notice when people in authority fuck up and paper it over with nonsense. Notice when people are trying to sell you something.
Q3 books 2026
Saga of the Volsungs translated by Jesse Byock. Read this on a plane flight. It’s been a long time (decades) since I read the Sagas, and the last one was Beowulf which isn’t as savage as this one. Lots of blood and guts. Lots of war and vengeance. Of course I had seen the Wagner Opera, but I had no idea what he took from the original stuff, so it’s a whole new experience. I should probably go back and read more of this stuff; I remember Aegil’s Saga was a lot of fun.
Apollo: The Race To The Moon by Charles Murray, Catherine Bly Cox. Yes, the same Charles Murray. Interesting account of the designers and engineers and managers involved in the Apollo program beyond Von Braun and the Astronauts. Lots of them are people you never heard of: Max Faget, Rocco Petrone, Joe Shea, George Low, Caldwell Johnson, George Mueller. The Apollo program probably involved a half million people; this book is mostly ultimately about their management. From the management of the engineering and contractors to the detailed management of the spacecraft from the control room. Lots of it is stories of buzzcut engineers in short sleeve white shirts making monkey threat displays at each other, which isn’t exactly gripping reading unless you do that for a living. It did manage to give some important takeaways into how to manage a giant project like this. You can tell Elong read this book for example, in the way SpaceX conducts its business. It’s also a book on changing organizations; the original NACA (a pure aeronautics scientific organization) was a very different group of people, and this goes into how the organization evolved into the thing that shot men into space, how the people from the original group contributed to the larger goal. The stuff on the actual missions is in the latter third of the book; they manage to put quite a bit of dramatic tension in it, which is a considerable writing achievement. It isn’t a HBS case study, but one could read it and write a good one (which would make everyone at HBS mad: lots of square young white dudes, no girl bosses and almost zero non-technical bureaucracy). Invaluable historical account, apparently the interviews were done in the 1980s while most of the characters involved were still alive and had relatively fresh memories. Absolutely read it if you’re a manager of technologists who needs to put lightning in a bottle. Skip it if you sell underpants on the internet, or skip ahead to the exciting stuff.
Path Signatures and Rough Path Theory for Quantitative Trading in Python: A Practical Guide to Geometric Feature Engineering and Signature Models by “James Preston.” I quote the “author” who is some kind of LLM thing which has like 400 other books on related topics. First time I got owned by trash Amazon books (Amazon graciously gave me my money back). It’s hilarious in that an initial thumb through it indicated some promising stuff. This book contains things I’ve figured out on my own that you only can figure out from experience; stuff decidedly not in the literature and which an LLM would be hard pressed to come up with unless it was talking to people who know these things (look at your terms of service). It also contains a whole bunch of insane gibberish and retarded anecdotes about the use of quantum computards in executing high frequency trades. Written in a very plausible trader-like declarative fashion, except actual quant traders never talk like this. I’m not going to tell you which are the hard earned insights (which I guess LLMs know about; might as well as one of those) and which are the gibberish, so this book is useless to anyone who doesn’t know these things already: don’t buy this book. Path signatures are cool though. Instead buy:
Signature Methods in Finance edited by …. Beyer, Horvath, dos Reis and Oberhauser. Downloadable from Springer here. This is not a book which purports to be directly relating to trading, though it has a chapter/paper in it which does. It’s a funny sort of book, which seems to be becoming more common in technical books; it’s a fairly coherent textbook on the subject consisting of various chapters written by different people as standalone papers. It has a very readable introduction to path signatures, as well as sections on feature generation, option pricing and a whole lot of other useful stuff. By contrast, the textbook by Friz and Hairer (downloadable here) is basically unreadable for non-mentats, and arguably worthless for people who want to use it for anything other than making proofs about regularity structures. They don’t even use the words “path signatures” (I think they call these things “lift” or something), which is pretty annoying as that’s the main thing I give a damn about. Anyway the various forms of rough path signature are tremendously useful mathematical objects. I’d rank them at least Wavelet tier, possibly Fourier transform tier. You can think of them as a sort of Taylor series expansions of timeseries which blows out the dimensionality (yes it increases dimensionality -so do wavelets) in an extremely useful way. It’s obvious the deeper math behind it is godawful important. Hairer won a Fields medal for the regularity structures thing, which is a kind of generalization of path signatures with renormalization. Path signatures are defined for “controlled differential equations” (ordinary diff eqs in the presence of noise); the regularity structures are extending the idea to partial differential equations in the presence of noise. I’m happy just fooling around with my new doodad as there are plenty of interesting systems which are effectively ordinary differential equations in the presence of noise. This involves reminding myself how tensor math works and glancing at real analysis proofs and stochastic calculus books. It’s like a mathematical result from the 19th century (when everything was awesome and changing every day) showing up in the 21st. It is mostly the invention of Terry Lyons, who I assume is too busy kicking ass to write a book like this; his lectures on youtube (for example) are excellent and will give you an idea of what it ends up being used for. Anyway if you have no idea what I’m talking about, don’t buy this book. Otherwise, this is the book to buy. I was also able to get through a chunk of Friz and Hairer after reading this and 10 review articles on the topic and making a sort of glossary to decode what they were saying. Hairer was trained as a physicist and Friz was a quant, so you’d think it would be easier to grok what they’re on about, but I guess they’re deep in the badlands of abstraction.
The Light Ages Seb Falk. Contrary to reddit, there was plenty of science going on in the so called “dark ages.” I’ve known this since reading Spengler’s Man and Technics, where he traces our Faustian civilization to these times. It’s one of those things where, if you know some names from those days, and look at their wiki-pages it is self evidently true. The procedure known as the scientific method was invented by an English monk named Grosseteste in the 13th and greatest of centuries. Later explained better in the same century by Roger Bacon. I won’t attempt to give a history of the great men of that era, look at Man and Technics for a good list. Falk give us a peek into some of the great men of the day, but through the eyes of a sort of central figure, John of Westwyk -an obscure astronomer-monk of St. Albans Cathedral. A lot of it is dry, but important: things like how they thought about the diameter of the earth, how numbers were encoded with multiple systems, how astrolabes work…. I have to admit, something about the style of the thing made it a tough read to get through. This is a subject I’ve always found interesting, but it took me months to get through the volume. I can’t explain what is wrong with his prose exactly, but it was tougher to get through than, say, Homer.
A Pocket Mirror for Heroes Baltasar Gracian. Gracian’s “Art of Worldly Wisdom” is probably my favorite book; I carry a dog-eared copy of it around with me everywhere. This book, unfortunately is not up to those standards. I suspect it’s mistranslated, but I don’t care enough to look at a Spanish edition. At least some of it isn’t the translators fault: Gracian takes obscene liberties with the reader’s attention in various places, telling long (incredibly dry) allegorical tales about peacocks in an attempt to communicate what is ultimately a trivially simple idea: if you’re excelling at something, distract the jealous with your defects (the peacock has ugly feet). Others contain interminable dialogues between various unfamiliar personages. Perhaps this sort of thing was beloved of 16th century Spanish Jesuits, but it gave me a headache. Overall, Gracian is trying to communicate the same kind of thing as in his better known book, but fails at doing so as concisely and clearly. This one I’ve probably picked up a half dozen times, but only managed to finish the thing recently on force of will. Would not recommend. Will hang onto the book until it’s way out of print and probably be able to sell it for bitcoin tier prices to people who haven’t read my criticism of it.
The Discovery of the Mind by Bruno Snell. Snell’s thesis is that the Ancient Greeks invented the intellectual concept of the self reflecting individual; people before them did not have this concept, which came to take over Western Civilization. This can be seen as a sort of moderate take on Julian Jaynes Origin of Consciousness in the Breakdown of the Bicameral Mind. Jaynes book, which I read at the suggestion of an experimental physics professor in my undergrad junior year, is one of those mind-blowing works of scholarship that changes the way you look at the past forever. Jaynes idea is effectively that ancient man were effectively paranoid schizophrenics, regularly talking with “the gods” the way contemporary schizophrenics hear voices. It’s a compelling idea well backed up by his scholarship; as others have pointed out, it is either a work of consummate genius or total bullshit. Snell’s book (1948) predates Jaynes (1975), though according to the index in Jaynes book, he had only heard of Snell after completing his overall argument in chapter 2 of his book. In any case I had only heard of Snell’s book this year. Unlike the two books reviewed before this one, I found it an excellent book in its structure and style and powered through it quickly (I was beginning to wonder if I needed to visit the brain doctor). His argument is compelling in that he traces a history of ways of thinking and being from Homeric times to later antiquity using linguistic ideas, historical religious ideas, literature and so on …. From my stale memory of Jaynes book, he uses similar arguments to the ones Jaynes did, but wasn’t as radical in his conclusions. It’s one of those books which provokes thought; makes the ideas dance in your head. For myself I had the thought that maybe the Homeric Greeks had a better understanding of how consciousness actually works, dividing the soul up into thumos, psyche and noos. The way Snell breaks down their contextual meanings, maybe they refer to distinguishable things. We don’t give the ancients enough credit: they had a lot of time to think about things (sort of like with astronomy which was both television and supercomputer to the ancients), lacking the ridiculous distractions we have today, and it wouldn’t surprise me if the ancients had better insights into how the mind words. No urban distractions to confuse; perhaps the later ideas we inherited about how the self-reflective mind works are the faulty ones; a sort of mirage of culture and language. The obvious counterpoint to Snell and Jaynes is that people may have had the same basic mental structure as today, but lacked the ability to speak about it. I’ll have to give Jaynes a reread soon as well. Excellent thought provoking work of the highest quality for those interested in Greks or ideas about consciousness in general.
The Popes by John Julius Norwich. Norwich has shown up in my book lists multiple times now; pretty much everything he writes is a treasure. Often the kind of thing that leads one to very amusing primary materials, such as Anna Comnena’s diaries or Villehardouin’s memoir of the Fourth Crusade. His historical interests mostly lie within the Adriatic sea; Byzantium, Venice, the Norman freebooters, though he has a book I read half of in the summer of 2021 on France. This one, obviously is about the Popes. Norwich was an agnostic Protestant, and so he wrote this as a political history rather than from any religious perspective. This is arguably the most valuable perspective one could take on the topic; if you’re a Catholic, it gives you some perspective on what the papacy has meant over the millennia. It also gives you some perspective on recent Popes in comparison to the Saints, Freebooters and Monsters who have inhabited the institution over the centuries.
The Rise and Evolution of BJJ by Robert Drysdale. We all know the story of Carlos Gracie learning from Count Maeda and developing the family martial art he taught to his brothers. If you dig a little deeper, this is known to be more or less total bullshit. Drysdale (a high ranking BJJ practitioner) focuses on the less well known contributions of Carlson Gracie, as well as his friendly rivalry with Rolls. By Drysdale’s telling Carlson (who was a frequent Vale Tudo competitor) was a crucial piece of the puzzle, opting for a more broad audience of students and a ruthless concentration on “what works” in something resembling a real fight. I studied a bit with Carley Gracie, and he said similar things. He considered it a more direct lineage to the original art as developed by his father Carlos (also father to Rolls and Carlson: Carlos had like 20 kids with various women). It’s a fairly interesting book, though a bit repetitive, with large chunks of it consisting of interviews with people from the Carlson and Rolls dojo. Quite a lot of interesting trivia, origins of stuff like the americana lock (by Bob Anderson) and the complicated interplay between various kinds of Catch wrestling, Judo and other kinds of wrestling active in Brazil back in the day. This isn’t a sport I was ever particularly active in, and I have no plans on becoming active in it in the future, but it is a cool sport, especially in the early days.
How the World Really Works (2022) Vaclav Smil. Pretty much everyone involved in any kind of sustainability racket needs to read this. I don’t respect anybody who has opinions on “green” whatevers who hasn’t internalized the facts contained in this book. Basically everything you do and see is crystalized hydrocarbons. Including stuff like windmills, tomatoes, bread, electric cars …. Smil estimates if you got rid of the Haber Bosch process half the population would starve. I’m kind of OK with that because I know which half it would be. That’s the problem with reading books like this. Because it actually tells you how the world works. More or less, it does contain some blunders: for example, I clock wood in as almost as important as concrete and more important than plastic in terms of tons per year, but it escapes Smil’s attention. It’s true plastic is more interesting and serves more essential purposes, but wood is kind of a big deal. Anyway the book focuses on important materials for modernity and the actual mechanisms pretty much everyone is blissfully unaware of when talking of glorious green futures and electric airplanes. Smil wrote this as covid hysteria peaked and seems extremely bitter throughout at all the dumb shit touted in the media and online. The “technocrat” class should be as in command of basic facts as Smil is, but obviously isn’t, and is as enamored of wacky ideas as barbarians with bones in their noses. While Smil doesn’t go all “covid denialist” with how the vile technocracy handled that crisis, he does point out that the places ruled over by people convinced we’re going to have solar powered steel mills are the places where the most people died. I didn’t realize Smil is a boomer so we might not get so many more books out of him, but his accounts of technocracy predictions over the years and how the world physically, thermodynamically functions are invaluable, if only as a record of how bad our technocracy is at virtually everything it attempts.
Poilu by Louis Barthas. I’m a sucker for war autobiographies; one of my all time favorites is by another Frenchman; the Memoires of Sergeant Bourgogne -about the happenings during the Napoleonic Russian campaign. Barthas unlike any other WW-1 autobiographer was an uneducated (though very intelligent) man: a wine barrel maker by trade (in Peyriac-Minervois: nice wine region). They only came to light in 1978 when his grandson made them available for publication. Barthas was relatively old at 35 when drafted. He was also a socialist as many thoughtful types were back in his day. He somehow lasted the entire war and was at many key battles; the kinds of details he presents are almost beyond belief. Unlike many memoirs of the war, this was never polished for literary effect; it’s just diaries he reworked after the war with postcards, newspaper clippings, maps and other such ephemera for his family (some of which are reproduced in the book). Not a cheerful book, but a nice book to read if you’re ever experiencing minor inconveniences and need a reality check.
19 comments