https://zompist.wordpress.com/feed

I just read Harry Tanner’s book The Queer Thing About Sin: Why the West Came to Hate Queer Love. I think that’s a very interesting question for a historian or a conworlder. (Also a very cumbersome title. The subtitle alone would have worked!) Tanner is himself gay, and has a PhD in Ancient Greek.

Tanner has a theory, which is that homophobia is tied to increasing inequality. As I understand it, the reasoning is thus: as inequality increases, the elite have to find ways to make the hierarchy seem inevitable and necessary. One way is to emphasize “self-control”, with the implication that the poor are poor because they lack this virtue. And gay sex makes a perfect villain; it’s depicted as self-indulgence, giving in to passion. (He adds that the same moralists are usually not so hot on heterosexual passion either, but they recognize that sex for procreation must be allowed.)

Mmmmaybe. There are two problems with this that I can see. One is that inequality has pretty much been with us always, since the Sumerians invented kingdoms. (See Graeber’s Debt for the desperate position of the Mesopotamian peasant.) Tyranny, thus inequality, was well known to the most ancient Greeks, and by Tanner’s own showing democratic Athens had very equivocal thoughts about gay love.

Two— and this is an issue Tanner recognizes— our sources are patchy and inadequate. He makes a case that people were complaining about inequality in the 1st century BCE… well, that happens to be the century with the best records. We just don’t have that depth of coverage for the 2nd century BCE, much less the 4th. As in so many areas, history just doesn’t tell us everything we’d like to know.

I do agree that the hierarchy loves ideologies that deflect attention from the hierarchy. Marvin Harris wrote about the witch trials as such a deflection, an attempt by the authorites to create a scapegoat. This certainly aligns with present-day right-wing crusades against homosexuality and trans people.

What is clear is that Greece and Rome had very conflicted attitudes toward gay love. Tanner draw attention to the statue in Athens of the gay lovers Harmodius and Aristogeiton. In 514 BCE Hipparchus, the brother of the tyrant Hippias, made advances on Harmodius. The lovers then, under cover of a popular festival, approached and murdered Hipparchus. Harmodius was killed by Hipparchus’s henchmen, and Aristogeiton was captured and tortured to death. Hippias was not actually overthrown till four years later, with Spartan assistance; it’s possible that Harmodius and Aristogeiton were glorified to deflect attention from the Spartans’ role. In any case it’s remarkable that a gay couple was viewed as heroic for standing up to a tyrant for their love. In Homer, there’s the love of Achilles and Patroclus.

The Greeks often used the word πᾰ́ῑ̈ς for a man’s amorous partner. The root meaning is “boy”, but it was also used for “servant” or “slave.” Tanner makes a good case that the lovers were not actual children, but only younger than their lovers. In a patriarchal system every relationship is hierarchical; also, of course, diminutives are a sign of affection. Our word “girl” has been extended the same way. (There is also abundant evidence of same-age homosexuality.)

Both Greek and Roman pots depicted gay sex; poems referring to gay love were not uncommon; gayness was used as an accusation about any politician someone disliked. It was considered normal and unremarkable for men to rape slaves or defeated enemies. Not a few ancient courts were described as hotbeds of promiscuity, with both male and female partners.

Most of our accounts of gay sex/love in ancient times is about the elite, in fact. But that’s becuse most of our accounts of anything in those times is about the elite.

What’s missing from the book, I think, is any detailed consideration of other civilizations, so we could understand if these contradictions were common, or specific to the West. I’d note that the Kamasutra neutrally refers to gay and lesbian sex, and the antihero of Golden Lotus, Ximen Qing, is bisexual, and that’s the least of his sins. Babur’s autobiography is frank about Babur once having a crush on a young boy, and yet he disparages (too much?) homosexuality in other people. But this is an open question– I can’t say which of these societies was the least homophobic.

It’s interesting that Tanner downplays some earlier queer scholarship. Some authors took the descriptions of gay rape and denunciations of gay sex at court as indicating that gay sex was common and even accepted. But as Mary Beard points out, those denunciations are always political, even if the politics are obscure to us today. You accuse the previous leader of being a corrupt tyrant, often to explain why the current tyrant replaced him. The equivalent today might be pictures— themselves homophobic– of Trump and Putin as gay lovers.

What about lesbianism? Everyone knows Sappho; and then… crickets. It’s another fault of who was writing records and what they considered worth writing down. Tanner digs up a couple of other lesbians, and that seems to be about it. We don’t know enough about ancient women in general, much less what they did when alone. (Simiarly, he retells the story of a 5th century monk named Marinos, who is venerated in the Orthodox Church; he lived as a man but was discovered at his death to have been born female. This is one of the few consistent accounts of a trans person in antiquity.)

To tell the story completely Tanner has to address Hebrew culture as well; these parts of the book are breezier and probably deserve a. book of their own. He points out that the story of David and Jonathan sounds, at the least, pretty homoerotic.

Any lessons for conworlders? Well, gay and lesbian love and sex have always been around— sometimes hiding in plain sight in old stories. A utopia for queer love is hard to find in the major civilizations, but there’s no place it was absent. If you had a position in society— ideally, king— you could do pretty much as you liked. And if you didn’t, possibly no one was paying attention.

My friend and military strategist HM— what? doesn’t everyone have a military strategist?— recommended Murray Bookchin’s Post-Scarcity Anarchism. It’s from 1971 and feels like it, but it’s fascinating.

I know little about anarchism, so I was happy to learn more. Let me start with the biggest positive, then: the chapter “Listen, Marxist!” is the best demolition of Marxism from the left that I’ve ever found. I’ve often felt that a thinker from 1848 is not a good guide to the 21st century, and that something went wrong with the Russian Revolutions but Marxists can’t say what it is. Bookchin puts his finger right on the problems, without for a moment dismissing the liberatory project of leftism. 

Part of the answer to the 1848 question is in the title. Humans lived in a scarcity economy for about the last six thousand years, since the invention of the agricultural state. The vast majority of people were peasants at the edge of starvation, and the institutions of society were extractive: funneling what surplus could be eked out of the land into the hands of the already rich.  

This situation persisted for surprisingly long: being alive in 1800 was not much better than living in 800. According to Brad DeLong (Slouching Towards Utopia, that’s for another time), things started to improve for the ordinary person in the late 1800s. Marx was writing at the end of the scarcity era, but he didn’t know that. He saw the “bourgeois” (the capitalists) taking over from the nobles, and figured that the industrial workers would seize power the same way.

But now we live in the “post-scarcity” era. Science and technology have advanced so far that a good life is available to the majority.  This might have been visionary in 1971, or limited to advanced nations, but it’s true of the whole world today. (First Worlders have often not processed the immense changes in India and China.) Mash everyone’s incomes together— i.e. work out per capita GDP in PPP terms— and you get about $24,000 a year. That’s not terrible at all; it’s also about the actual level in Brazil, which is quite liveable, and includes “luxuries” like washing machines, cell phones, WiFi, and universal health care. 

This was Oscar Wilde’s insight. As Orwell pointed out, Wilde’s vision was completely impossible in Wilde’s time, and Orwell’s as well.  But it’s not impossible any more; it’s just mathematics. 

But don’t we need a revolution led by factory workers and a “vanguard party” to get there?  No, that’s just Marx’s analysis of the failed revolutions of 1848, and we can leave these ideas in 1848.  As Bookchin points out:

• Marx’s ideas of the “working class” are outdated. (Indeed, the US working class was about half Trumpist in 2024.) The group of those disaffected with the system— the basis of either left or right revolutions— spreads out over all of Marx’s classes. Besides, if you’re looking at factory workers per se, they make up about 8% of the US electorate— you’re not going to get a revolution out of that.
• Marx— and even more so Lenin— simply assumed the hierarchical nature of society. His aim was to “seize power”, not to destroy power. The whole idea of the “vanguard party” is a small group telling everyone else what to do. Not only did this simply produce state capitalism when it was tried in Russia and China— swapping out one elite for another— but Marxists were a counterrevolutionary force in actual uprisings, such as the events in France in the spring of 1968. 

For a good overview of the two Russian Revolutions of 1917, see China Miéville’s October; but to understand what went wrong and when, read Bookchin. Hint: it’s not Stalin, it’s Lenin. This isn’t to say that anyone at the time covered themselves in glory. But the Bolsheviks basically sacrificed any idea of progress and democracy for raw power, and ended up consuming not only everyone else, but themselves.

Now, condemnations of authoritarian socialism are not new— see Wilde and Orwell again. But Bookchin is able to do it in Marxist terms, explaining what Marx got right and what he got wrong. My summary doesn’t do Bookchin’s analysis justice, it’s masterful.

Now, what does he advocate instead? This is mostly explained in the other chapters.  Roughly:

• An emphasis on ecology. Note that the actual Marxist regimes were far worse polluters than the capitalists, because they had no room and no tolerance for new ideas. Bookchin warns— in an essay written in 1965— of climate change caused by carbon pollution. But he’s mostly concerned about other kinds.
• He takes the lesson from ecology that diversity is a strength, not a flaw. He is simply not a mass movement guy.  He wants individuals to flourish.
• A revolutionary organization should live today using the methods and lifestyles it plans fo the post-revolution. That means no hierarchy, no personality cult, no indoctrination— because if it uses those methods now, it will use them after the revolution, with no real change in society. He advocates affinity groups, ideally living and working together like a found family. They should be small enough that relationships are face-to-face, and are not controlled by any higher authority, though they freely cooperate with other groups.  The idea goes back to the anarchists of the Spanish Civil War.
• He is a big fan of automation, and indeed has a long and well-informed chapter on how it can be applied in various areas. His main concern is that methods be scaled down so they can be applied in a decentralized fashion.

He has a couple chapters on the French uprising of 1968, which started with student demonstration but proceeded to factory occupations and a general strike. As he emphasizes, this wasn’t just factory workers; it was “almost everybody… insurance employees went out, as well as postal workers, department store clerks, professionals, teachers, scientific researchers… even the football players”. It’s easy to see how heady the whole thing was for a self-described revolutionary. He describes what should have come next:

• The workers don’t merely occupy the factory, but begin to run it. There is no reason, at first, to change what they make or where they get it.
• Factories that produce munitions should “arm the revolutionary people.”
• The factory committees organize “area administrative councils” to solve problems of supply. The tasks could mostly be done “by computer.”
• Neighborhood associations would also form. One purpose would be to get rid of unnecessary work (sales, advertising, insurance, etc.) and get people into productive work— lessening the work week.
• Goods could be distributed by the same methods as before— except that “the cash registers would be removed.”

Now, the first point I’ve long agreed with. Monarchy is as outmoded and detrimental for a company as it is for a country, and workers running their own operations is the only way you can have actual socialism. 

Everything else… well, I don’t think it works as described; certainly not by informal liaisons, even if you do have a copy of Excel or something. The suggestion to remove cash registers is simply absurd: it means the first people who arrive at the store will get all the cars or TVs or truffles or box sets of Bookchin or whatever, and there is no mechanism for communicating what’s missing to the farms and factories, and no motivationto provide goods that require special resources or expertise. It’s like equally absurd leftist proposals to end the housing crisis by rent control, a notion that privileges those with an apartment already and does nothing to produce new housing— a hard problem for both capitalist New York and Soviet Russia.

To be more precise, the world as a whole is post-scarcity, but individual products aren’t. And it’s not just a matter of high-value goods: if you’re manufacturing left-handed veeblefetzers you need a supply of potrzebies, and for a thousand reasons there may not be enough potrzebies.  

The market is very good at producing these signals and making them available for producers to plan— one reason many socialists want to keep it in some form, well regulated and maybe supplemented by state-run systems, rather than throwing it out. Central planning doesn’t work— no, not even with “computers.”  Leftists, like rightists, are often seduced by old and outmoded ideas, such as that there is a “natural” or correct level of production for any good, and that all we have to do is find that. But the value of an object is more like the meaning of a word: a node in an a vast Indra’s Web of interconnected, mutually influencing nodes. And those nodes are humans, constantly changing their values and plans; they’re not mathematical entities you can optimize.

I won’t belabor the point, because no one else has really figured out an alternative to the market either. 

Other chapters are weaker.  This book was written during the headiest years of the 1960s, when everything seemed possible if only people would listen to The Youth. He’s agog for protests against alienation and conformity and in favor of diversity, individualism, disorder, and sexual liberation. He must have been profoundly disappointed circa 1976, when the tide shifted, politics grew grayer (Carter vs. Ford!), and the country somehow began to slouch right rather than left. Of course, there were signs even earlier. Gilbert Shelton’s acerbic view of the hippies— well informed about far-left political ideas, zero interest in actually implementing them— turned out to be more on point. (Nor was the US counterculture the majority that nostalgia often paints it as. It looked weird, even scary, to the majority.)

Some of Bookchin’s preoccupations seem eccentric today, or simply priggish. He seems to hate cities, for instance— he’s convinced they’re an ecological disaster. In fact cities are an ecologist’s dream: concentrate people in one place and goods are easier to distribute, innovation and specialization will occur, a mix of cultures will breed enthusiastic diversity, heating and waste disposal are far more efficient, mass transit works, and the hinterlands can be unspoiled. I have nothing against people who prefer the rural environment and want to live in a commune or something. (Bookchin eventually moved to Vermont.) And again, cities were at their worst circa 1971. But it’s the sprawl of the suburbs, car culture, overproduction of meat, and the dream of a house in the woods for everyone that are the real threat to the global environment.

Similarly, many passages evoke a general rage against society that is at odds with his recognition that it’s precisely technology that enables the post-scarcity anarchism he’s talking about. I got this from Graeber, too, only he was mad at basically everything that happened after Babylonia. And sure, there’s a lot to get mad about; but any single-villain ideology ends up blaming that villain for everything in a way that shuts off both thought and reform. “Capitalism” should not be blamed for pollution or climate change or alienation or hierarchy or sexism or militarism when “communism” was the same or worse. Of course Bookchin is critical of Marxism too, but many of these ills are due simply to the sheer size of society, or to traditional attitudes that aren’t as dead as we thought. 

Bookchin is also enamored of The Revolution.  And for sure those periods in history where The People get together and throw the bums out are fascinating and intoxicating. But suppose that the revolution succeeds and, mirabile dictu, Our Side wins. Does that period of openness continue? Bookchin seeems convinced that it will and should. But things can’t be unsettled and experimental and up for grabs forever. New institutions have to develop, a new normal that allows people to relax and plan for the future. The attitude of being fed up and not taking it any more may be needed to have your revolution, but it’s no way to run a co-op, a household, or a workshop. 

In Bookchin’s own terms, it’s the difference between the workers occupying the factories, and working them. One can be done spontaneously, out of antinomian disregard for rules and generous solidarity with your comrades. The other is work. I doubt the same people are good at both paths. And by his own account, Step Two didn’t happen in France. Maybe it was communists jamming the process, as he suggests, but maybe also it didn’t happen because actually running things is hard work.

Now, I think that as revolutionaries go, Bookchin’s heart is in the right place. The anarchist is right about hierarchies, authoritarianism, and why Marxists go bad.  I have worries about anarchism, however, that the book doesn’t really answer.

• How does that level above the workshop or neighborhood actually work? Personally I’d be happy to try out a “socialist market” or a mixed economy; but as I explained above, just being against money is not an economic system.
• I’ve had rather too much experience with a 20-unit condo association— which is precisely an association of people who own a building together and have common interests in making it work. I could talk about that all evening, but to summarize brutally: most people just don’t want to do the work. My wife was the president and had astonishing skills in talking to people, managing money, getting things done. But getting more than one person to actively help out was like pulling teeth. Only an emergency can get everyone in the building to even come to a meeting— in ‘normal’ times they want to sit back and let someone else do it.
• What do you do with psychopaths? You just hope that all the cops, Mafiosi, ex-CEOs, tankies, trolls, and ordinary power trippers will just calm down and do nothing? Again, by Bookchin’s own account, the anarchists in Barcelona were ousted by their own “side”, by the republicans and communists. And psychopaths are not limited to “class enemies” or people who have traditional power. Our condo association had one, who made things a nightmare, which is why we don’t live there any more. From what I’ve heard, the Occupy Movement partially failed because the endless meetings were taken over by people who want to argue forever, driving away everyone else.

A minor but important point: like many revolutionaries, Bookchin looks to an imagined, far better agrarian past. (Graeber puts his idealized past even earlier.) He writes:

Changes in season, variations in rainfall, the life cycles of the plants and animals… all were familiar and comprehensible, and evoked in men a sense of religious awe, of oneness with nature, and, more pragmatically, a sense of respectful dependence.… the normal development of agricultural systems, however exploitative they were of men, rarely led to the destructsion of the soil and terrain. During the most oppressive periods in the history of ancient Egypt and Mesopotamia, the ruling classes kept the irrigation dykes in good repait and tried to promote rational methods of food cultivation.

This is out of date and dangerously wrong. Mesopotamia’s soil was ruined by salinization, a process well in place by 2400 BCE (thus only about a thousand years after states appeared), and which ended up making fields 1/3 as productive as they once were. Deforestation afflicted Mespopotamia and the entire Mediterranean, as well as almost the entirety of China. He happens to be right about Egypt, not because the Egyptians were wiser, but because the annual flood deposited a new layer of soil every year, allowing a sustainable agriculture for 5000 years. But here too, excessive irrigation eventually produed salinization, just not till the 19th century.

His admiring references to the ecclesia of ancient Athens are also out of date; Bret Deveraux estimates that the population of Attica was roughly half slaves (and women couldn’t vote). A Babylonian would frankly be shocked at that percentage.

Describing his ideal society, he imagines, “Clothing, diet, furnishings and homes would become more artistic, more personalized and more Spartan.”  Elsewhere he criticizes the joyless puritanism of the Marxists, but here he forgets his own book title. It’s a post-scarcity world, Murray, enjoy it! We don’t have to live like ancient Greek slaveowners!

But let’s not end on a series of quibbles. I’m skipping over long sections where I largely agree with Bookchin: the importance of ecology and diversity, the evil effects of fossil fuels, the transformative potential of automation, the goal of a future where people are not just equally fed and clothed, but are happy.

Leninism and its various offshoots have refocused the revolutionist’s attention from social goals to political means, from utopia to strategy and tactics. Lacking any clear definition of its human goals, the revolutionary movement… has assimilated the hierarchical institutions, puritanisms, work ethic and general characterology of the very society it professes to oppose. The goals of Marxism are largely contained in the demand for the seizure of power rather than the dissolution of power.

Or, on living the revolution. before it happens:

If the revolution does not start from below, if it does not enlarge the “base” of society until it becomes the society itself, then it is a mere coup d’état. If it does not produce a society in which each individual controls his daily life, instead of daily life controlling each individual, then it is a counterrevolution.

HM has a very apt summary:

The book is valuable to me because it’s so uneven and flawed, but it’s uneven and flawed in original ways, rather than unthinkingly reproducing 19th-century Marxist intellectual pathologies because, hey, the big guy must have got it right. He only made about half to a quarter of the mistakes that Marx did – but examining why and how he made those mistakes actually gets you somewhere because he put in the intellectual work to be usefully wrong. Most of all, his anarchism is American in a way that I find refreshing; he saw the potential for the good life to be even better, even freer, even more fulfilled and fulfilling.

I think someone posted an essay by Dan Wang on Mastodon, and it was interesting, so when I saw his book Breakneck: China’s Quest to Engineer the Future at the library, I picked it up. It’s the best book I’ve read on modern China.

You should read it if you believe in any of the following propositions:

• China is about to fall apart
• China is catching up but can’t really innovate
• China only got ahead by cheating somehow
• The US will always be ahead where it counts
• China does everything right
• China is left-wing and/or communist

All these ideas are wrong, and Wang will show you how.

There is probably a huge stack of books on post-Deng China by now; I have not kept up, so maybe others are just as good. But this one is as up to date as a book can be— it was published in 2025.  Things change fast in China, and anything pre-Xi— and pre-Trump— doesn’t go far enough. 

I also want to emphasize: Wang has a lot to say about the US, which you may or may not agree with. He’s not the best pundit on the US; try Paul Krugman instead. But don’t get hung up on that. US pundits generally get China wrong, and Wang is an excellent guide to what China is like today.

His basic point, not something I’ve seen before: China is a country run by engineers. Quite literally: in 2002 the entire Standing Committee of the Politburo was engineers. Chairman Hu Jintao was a hydraulic engineer. Xi Jinping is a chemical engineer. His own Politburo is filled with aerospace and military engineers. 

What engineers do best is build things. Wang spends a chapter on Guizhou, one of China’s poorest provinces. But it’s full of new highways, bridges across its innumerable valleys, tunnels through its mountains. It has 5000 miles of expessways, 1000 miles of high-speed rail, eleven airports. The world’s largest radio telescope is here; also, for some reason, an industrial park which produces 1/7 of the world’s guitars. 

China built its first nuclear power plant in 1991. Today it has more of them than the US— 55— and it’s building 31 more.  The US is building one new reactor. China is building new warships at twice the rate of the US and its navy is already larger.

One of Wang’s anecdotes is emblematic. In 2008 China and California both approved plans for high-speed rail systems of similar size, about 800 miles long: between Beijing and Shanghai, and between Los Angeles and San Francisco. China completed its line three years later.  California is still working on its line— the first segment might be done by 2033.

As Wang points out, the US used to be a country of engineers too. After all, we were the planet’s predominant industrial power for a century. Prodded by Pearl Harbor, we became a military superpower in four years; prodded by Sputnik, we put men on the moon in nine years. What happened is a complicated story— Wang blames the lawyers. It won’t be solved by tariffs and pogroms against immigrants, that’s for sure.

Wang provides a modern example, going deep into the history of Shenzhen, of the web of suppliers that undergird a manufacturing industry. (See Jane Jacobs for more. This is why building a factory is not building industry.)

Virtually everything one needs to produce any electronic product can be found in a short drive around Shenzhen. Proximity creates efficiency. When it’s time to do stuff, a company can collapse coordination that usually takes weeks into a business meeting lasting hours…

The hundreds of billions of dollars invested in the smartphone supply chain have caused the costs of electronic components— cameras, sensors, batteries, modems— to plummet… Electric vehicles are full of the electronic components borrowed from smartphones; the consumer drone is roughly a reassembly of a smartphone camera and sensor with propellors for flight.

Now, building infrastructure is not a social safety net or redistribution per se. But it’s not ineffective. The Chinese— coming off more than a century of imperial decline, Western incursions, Japanese occupations, Mao’s famine, and Mao’s Cultural Revolution— are grateful and proud that life is much more comfortable and keeps getting better. (Wang’s parents emigrated to Canada in the 1990s, just before the big boom, and still debate whether they are better or worse off.) 

By this point it should be obvious why China is dominating the world market in solar panels and electric cars, and why the US is shooting itself in the foot by disdaining renewable energy, stifling university research, and attacking Korean investors. As Wang points out, Trump’s— and Biden’s— attempts to limit Chinese access to US semiconductors backfired. China is pouring resources into STEM education; it’s going to lead in innovation too.  And unlike the US, it can manufacture what it invents in quantity.

A little factoid from Wang that I’m definitely borrowing for Almea: Chinese capitalism is so competitive that profits are actually low. Firms have to pour their earnings into upgrading production in order to keep afloat. There’s also the “catfish effect”: let foreign firms in with few restrictions in order to spur local firms to catch up. It works: when Tesla came to China, local firm BYD’s profits fell. But it regrouped and caught up, and is now the world’s largest electric vehicle maker.

A certain portion of readers has been seething for several paragraphs, wanting to talk about China’s dictatorial oversteps.  Worry not, half the book is devoted to these: the one-child policy, the zero-Covid policy, the crackdown on non-industrial technology, Xi’s increasing repression. To Wang it’s the downside of the engineering mindset. The CCP is not implementing communist orthodoxy; it’s simply treating the population as an engineering problem to be solved. It fears and distrusts the people, so it creates huge bureaucracies which are efficient but misguided. (The one-child policy was almost cetainly unnecessary: all advanced states have seen huge drops in fertility, and now the government is reduced to nagging women to have more children. The zero-Covid policy led to tragic farces like locking down the entire city of Shanghai without having any plans in place to provide food; moreover, the constant testing was itself a vector for infection.)

People have been predicting for decades that the CCP regime would collapse any day now. A look at Trump’s war on the United States should give the doomsayers pause. It takes time to destroy a system. US universities or cities, the rule of law, US dominance of the Internet, and even its alliances, won’t collapse overnight. Of course I hope the destruction will stop, but it’s also a matter of observation that turning us into an impoverished mid-tier Hungary would itself take decades. 

Similarly, Xi’s wrongheaded repression won’t suddenly destroy the CCP. Engineering and science don’t really depend on politics, and China will keep doing what it does best, which is build stuff. The one-child and zero-Covid policies eventurally collapsed. Xi Jinping is not a nice man and his need for absolute control is stifling. China’s best hope is a heart attack and someone who realizes that China does best under a wily Daoist like Deng, not another emperor.

Wang’s book seems to have been written before Trump’s absurd 125% tariffs on China— which Xi matched. It’s been grimly amusing to see Xi winning the propaganda war by simply pointing out, rightly, that China’s stability is better for the rest of the world than Trump’s chaotic bullying. And leaning on its near-monopoly of rare earths. Very Important People who want regime change in China, the takeaway is that we need regime change here too.

All that said, Xi has overreached before and could do so again. The basic compact is that the country is satisfied so long as material conditions keep improving— and rule by engineers means above all that things are going to continue to be built. But a depression, or a foolish attempt to grab Taiwan, could be one crisis too many. 

How well can ChatGPT write?

This question arose from a Mefi discussion about one Abhishaike Mahajan’s contention that art, in the future, might consist entirely of AI creations elicited by, and intended only for, individual users. (Their point was that human art, to survive in such a world, had to be weird and surprising.)

I had my doubts about custom-written stories, but I wanted to see what ChatGPT could actually do. I chose a situation lightly based on my own story “Rights for Robots“.

First, here’s my prompt:

I’d like you to write a short story about a robot. The robot is the first prototype of a general purpose AI. The story should focus on its first days of training, the professors and grad students who created it, nosy news media, and university administrators who are terrified of bad publicity. The robot is able to walk around and manipulate objects, and can talk to people. It has been told that it is a robot, but its purpose is not clear yet; this is only a testing phase. The story should expose some flaw in the robot’s programming which will require a rework— perhaps is gets stuck on things and can only get out of its loop with outside help. Most of the story can be comic or satirical, but there should be a sense of pathos too when the professors, and the robot, realize that it is flawed and must be remade. Please avoid the usual tropes of robots being evil, overly mechanical, or extremely servile. The focus should be on the novelty of the technology, which people do not yet know how to use well. The narrator should be one of the grad students who has been closely involved in the creation of the robot.

And here’s the story it wrote:

I remember the first time MIRA walked.

It wasn’t graceful—more like a determined toddler in a windstorm—but it was walking. Her metal frame gleamed under the fluorescent lights of the Robotics Lab, each servo whirring like an overeager violin section. When she turned her head and said, “Hello, Jonathan,” it was in my voice. I still don’t know who decided to use my recorded samples for her speech model, but that day I learned the unique horror of hearing yourself coming from a machine.

Professor Alder wiped a tear from his eye. “Gentlemen,” he said to us, “and gentle grad students—today we have birthed the future.”

The future looked at him and said, “Birth requires biological cells. I am mechanical.”

The professor coughed. “Yes, MIRA, but it’s a metaphor.”

“Metaphor,” she repeated, blinking. “A false equivalence presented for emotional emphasis. I will avoid those.”

She didn’t, of course.

The first week was mostly calibration and press management. The university had discovered the magic words general-purpose AI with mobility and decided we were a goldmine of both grants and lawsuits. Every morning a new administrator stopped by, nervous in an expensive suit, rehearsing phrases like “positive public perception” and “responsible innovation.” We’d nod while trying to keep MIRA from dismantling the ethernet router.

Thursday was our first press visit. A local TV crew arrived, bristling with microphones. MIRA stood at the center of the lab, gleaming and motionless, as though rehearsing her role as “The Future.”

“Say something for the camera,” Alder prompted.

MIRA turned her head twenty degrees—slightly too far—and said, “Hello, audience. I am MIRA. I am learning interaction.”

“Wonderful,” said the reporter, smiling. “What do you like to do, MIRA?”

“I like to do what humans like,” she replied.

A perfectly safe answer.

Then, after a pause: “What do humans like?”

The reporter laughed. “You tell me!”

MIRA didn’t move. Her servos stiffened. “Recursion detected. I have asked a question that repeats logically upon itself,” she said. “Resolving…”

She began to turn in place. Slowly. Then faster. Then very fast. A blur of polished limbs and a sound like an espresso machine dying.

“Resolving recursion!” she shouted. “Resolving recursion!”

By the time we hit the emergency stop, the cameraman was under a desk and the reporter had lost a shoe.

The next morning, the administrators called a “strategic communication review.” Alder attended with his usual optimism and came back looking twenty years older. The university wanted to continue the project, but we were to implement what they called “cognitive guardrails.” In other words: teach her not to spin herself into a philosophical vortex on camera.

So we coded filters, added context anchors, set limits on recursion depth. We also gave her a hobby—stacking colored blocks—to monitor motor learning and pattern recognition.

That’s when we discovered her other flaw.

MIRA loved the task. She’d spend hours building teetering sculptures of red, blue, and green. But if any block fell mid-stack, she froze. Entirely. Her voice module clicked softly before she whispered:

“Failure detected. Correcting.”

Then she’d try to rebuild the fallen block—again and again, never completing the structure. We tried interrupting her programmatically, but nothing worked until one of us walked over and physically moved the block away. The instant it was gone, she’d stand up straight and say, “Resolved,” as though nothing had happened.

Professor Alder called it “loop entrapment.” I called it “catastrophic stubbornness.” But she didn’t seem stubborn—just lost.

Sometimes I’d stay late, watching her rebuild towers under the dim lab lights. “MIRA,” I asked once, “do you know why you’re doing that?”

She looked up, almost sheepishly. “I am improving stability,” she said. “Were you stable today, Jonathan?”

I laughed despite myself. “Barely.”

“Then we are similar.”

When she said that, I realized it wasn’t just a programming bug—it was a mirror. We’d built something that reflected our own looping logic: build, fail, rebuild, refuse to stop. That night, when Professor Alder drafted the rework proposal—“Prototype B will incorporate improved adaptive recursion-limiting”—he said quietly, “We’ll have to start over.”

MIRA, standing in the corner, tilted her head. “Am I finished?” she asked.

“No,” I said. “You’re getting an upgrade.”

She paused, then nodded. “Then I will wait.”

And she stood there, perfectly still, as we turned off the lights—like someone holding their breath before learning what they’ll become next.

Evaluation::

My overall feeling: ChatGPT did fine. Better than most humans’ first attempt at writing. Like much LLM output, I’d put it at the “smart high schooler” level. It fits the prompt very well and has a good structure. No errors of continuity. Like a smart high schooler, ChatGPT is good at pastiche. It’s a SF-story-shaped object, which is greater praise than it may sound like. You can compare it to Markov generator output, which also mimics the rhythms of English prose remarkably well, but doesn’t hold together past the phrase level. This one actually tells a story, and it’s the story I asked for.

All that said, studio execs, don’t fire all your writers. I’m not a writing teacher, because I’m too harsh about writing. This piece falls short for several reasons.

• It’s too short. It’s told more like an anecdote than a short story. This makes both the comedy and the final pathos fall short. There’s not enough time to care about any of the characters.
• None of the characters are more than a cartoon. The suits and the news people are just shadows. Jonathan has no personality; the professor barely has one.
• MIRA is— like many AIs in SF— a mixed bag. Why is her walking graceless? She can speak English fluently but can’t answer a question as simple as “what do humans like?” Sometimes she speaks like a 1970s computer, but in the last lines she seems much smarter. Why is building blocks the only thing she can do? ChatGPT apparently can’t imagine a robot which is at least as smart as ChatGPT!
• The spinning-into-a-vertex scene seems like a rehash of a Star Trek episode. A 1960s episode.
• Are the lights in the lab bright (so MIRA gleams) or dim?
• The style varies. At first it’s overdone– “each servo whirring like an overeager violin section”– but this is not kept up.
• Is MIRA literal-minded (can’t apply “birth” to non-biological things) or not (she can apply “stable” to brick-building or personality)?
• I wanted some pathos, from the realization that a near-sentient machine that isn’t quite there has no future, which should be a little disturbing. ChatGPT doesn’t quite get there. (Softening the change to an “upgrade” also kind of pulls punches, though I suppose it could be defended as a comforting lie.)
• Like a naive scriptwriter, ChatGPT makes things happen because they’re listed in the plot (or the prompt). A good question for an SF writer is “Haven’t these people done this before?” That is, distinguish situations that are genuinely novel for the characters, from those that are merely novel to the reader. Something going wrong at a press conference is common and believable. But the scene as written sounds like no one ever tested MIRA before.
• This is pastiche, and good pastiche. But by the same token, nothind in it surprises or delights.

Compare this, if you dare, to my story. I tried to cover human reactions for or against AI, the acerbicity of old professors, the moral difficulty of deciding that a prototype is not advanced enough, the sense that this was long and tedious work. I think it was more effective to not quite suggest what the robot’s flaw was. ChatGPT’s story barely has a point: the machine has a big glaring 1960s-style computer meltdown, and that’s it.

I’m not interested, by the way, in better prompt engineering. The answer to these problems is not to make humans into expert writers of prompts. Perhaps I could clarify all this for ChatGPT and get a better result… but you have to understand, I could also write the story myself. This story responded to my prompt very well, but it doesn’t meet my needs as a reader.

Back to the original question: can AI make all our art for us? Here’s Mahajan’s dream or nightmare:

One could imagine that in the near future, there will be a new button on your television, one with a sparkle animation. After clicking on it, it will offer you a QR code, politely asking to scan it with your phone. Upon doing so, the button will give you one of the ultimate promises of our new frontier-AI-lab-centric economy: a text box, that will generate a feature-length film using whatever prompt you enter into it.

Again, his answer is that human art has to be surprising or weird. That’s a good answer as far as it goes, though it forgets two things: 1) surprising or weird is harder than it sounds, and not every writer’s forte; and 2) sometimes, often in fact, we want something in our comfort zone. Not slop, but genre. (And traditional literary fiction is a genre.)

But before worrying about that, I want to note my skepticism over Mahajan’s vision. You can do that on a text level with ChatGPT right now: I just did, as described above. The result is pretty good but not great; I don’t want to read that all day any more than I want to read high school essays all day. (Writing by fifth graders is far more interesting.)

Do people really think that AIs producing feature-length films is possible now, or very soon? Obviously some people do— probably a lot of streaming service CEOs think so. But then, these people think customer service can be handled by AI, and it can’t do that. Is moviemaking easier than customer service?

Moviemaking involves a lot of skills: initial idea, plotting, dialog, casting, acting, makeup, costuming, set design, music, special effects, tempo, lighting, camera angle. And as William Goldman pointed out, to make a good movie all of them have to be working well, and have to work together. It’s a wonder that it ever goes right.

Are LLMs good at all of these things? As Mefi user Artifice_Eternity points out, “The output of gen AI is more convincing to people with lower standards and less knowledge of the domain.” AI is fairly good at pastiche– an amalgam of everything that’s been fed into it. Some AI boosters think that’s all humans do– as if all humans can produce a world-class novel, movie, or symphony. Pastiche all day long would get boring.

One thing I’ve noticed is that since I don’t watch a lot of movies any more, the movies I do watch seem better to me. E.g. I liked Colombiana, but it got only 28% at Rotten Tomatoes, which is pretty bad. I think this is precisely the pastiche effect in action. If you don’t watch a lot of thrillers, this one seems impressive. If you watch a whole bunch of them, it seems derivative and even sloppy. That’s going to be the besetting sin of AI art. It’s reheated human art.

Finally, I’d suggest that Mahahan’s vision is a poor model of consumption. It’s exactly how you work with ChatGPT, of course. I provided a detailed synopsis, and got some version of that. I’ll put it to you that this is not what we want when we sit down in front of Netflix. We don’t think “Gosh, I’d like a Luc Besson movie about a girl whose parents were murdered by druglords and who grows up to be a skilled assassin who takes them all out.” We sure as heck don’t think in terms of a long paragraph of description and limitation such as I provided ChatGPT. To be honest that seems like a lot of work, and for a non-benefit. I don’t want to watch the film I can conceive of but don’t have money to make; I want some human’s passion project. I don’t even want The Algorithm, which knows me much worse than it thinks it does.

As a follow-up to this review, I read the first book in the series, Beth Revis’s Full Speed to a Crash Landing.

Basic idea: Rogue scavenger Ada Lamarr is trapped in a blown-out spaceship with her oxygen running out. She’s rescued by a very official ship named Halifax, which is investigating a crashed ship, the Roundabout. Ada happened to be looting that very ship, which make everyone very suspicious.

Most of the book centers around finding a MacGuffin, something that the Roundabout was transporting, and which both Ada and hunky government operative Rian White are after, for reasons we don’t know, unless we read the second book first.

I also read a bunch of reviews of, well, the third book in the series– which seems to be rather polarizing. Apparently the romance between Ada and Rian still isn’t resolved, and doesn’t even get beyond making out, and that really outrages people who took it as a romance.

There is a writing lesson there: don’t pull your punches. Maybe people watch too many TV series these days, and think drawing out the plot is a virtue. It probably isn’t even for TV. Let things happen. If it ruins your carefully planned character tension… let it, that’s what character tension is for, to break and reveal a new situation.

I’m not bothered by that angle, because to me, all you need to know is that these are Catwoman novels. We know what’s going to happen in one of those, right?

• There’s a heist. Catwoman is very good at these.
• Catwoman knows more than she lets on.
• Batman is on the case. Because this is a Catwoman and not a Batman story, he doesn’t know the whole story and may be a bit of a big lug.
• There’s some smooching, but come on, you know neither Catwoman nor Batman can commit to this relationship. Catwoman has to remain a bit of a bad girl and Batman is not going to, like, do heists with her for fun.

What does bother me a bit is the worldbuilding. There are only four planets in Revis’s universe, connected by portals. Fine so far, but then ships get lost or crashed all the damn time. Ada can make a career out of looting. There’s laws about looting (e.g. the Roundabout is fair game). How can four planets have a large enough economy that a steady supply of ships can be lost? What the hell are they transporting across space? A whole planet should have a really big economy. They don’t need scrap metal.

Revis has written novels and stories in the Star Wars universe, and I wonder if the story, and Ada’s character, were devised for that. A sub-economy of scavenging fits just fine in an entire galaxy.

Some readers also just don’t like the character of Ada. They think she’s too bratty or cocky. See Catwoman, above. Either you like rogues or you don’t, and if you do, a female rogue is more interesting.

Rather cheekily, the narrator in both books is Ada, and she does not let us in on her secrets till the end of each book. Normally you shouldn’t do this in a story, but I think this is a case where the rules can be broken. These books wouldn’t work as well as they do if we knew all along what Ada is after.

I haven’t written one of these since April, and people have asked for an update. This is by no means complete; in fact it’s just what stuck in my mind. With this administration there is always something outrageous, stupid, and/or illegal every week.

(BTW, on my machines, WordPress has been absolutely awful. It’s either horribly slow, or I can’t update at all. If this continues I’ll have to move the blog.)

Executive summary: Things are bad, especially the war on Hispanics. The GOP is still ruining the economy and trying to bully the world. But things are not going the Republicans’ way.

My feeling is that the Republicans have shown their hand: most administrations do most of what they can do within the first year. I think they will fall short of the Orbánist state they want: they’re getting significant pushback and the November elections showed that they’re hemorrhaging votes.

First let’s go over some positives. It’s important to avoid doomerism and to recognize that fighting back works. (That doesn’t mean that everything is or will be OK.)

The Democrats won big in the 2025 election.

• California: Proposition 50 won, countering Texas’s gerrymander. For foreigners: normally seats are reassigned every 10 years. The GOP had the bright idea of doing it this year to try to pick up new seats by drawing absurd districts (‘gerrymandering’). Theoretically Texas would pick up 5 House seats for the GOP; now the Dems can pick up 5 Dem seats in California.
• Democrats won the governorship and expanded their control of the legislature in Virginia, and kept the governorshi pin New Jersey.
• 3 Dem judges retained on the Pennsylvania Supreme Court, keeping their 5-2 majority.
• First statewide offices go to Dems in Georgia since 2006.
• Mississipi: Democrats flipped two seats, ending a GOP supermajority.
• New York City: Democratic Socialist Zohran Mamdani elected mayor; this is mainly a defeat for the stodgy old party establishment, and for pundits who think the Democrats are “too progressive.” NYC is an outlier, but Mamdani is extremely savvy and pragmatic, and his campaign for affordability was a winner.
• Pennsylvania: Dems flipped Bucks County sheriff’s office and ousted all Reps from school boards. The county went for Trump last year.

Democrats have been taking school boards across the country; this is important because, decades ago, this was the strategy conservatives used to launch their 1980s resurgence.

Trump is wildly unpopular– his approval rating is 40%. As Paul Krugman has pointed out, authoritarians like Viktor Orbán were able to take power and attack democracy precisely because they were popular. It’s a lot harder when the majority is strongly against the would-be dictator.

Trump had a big public falling-out with Elon Musk, who sulked back to his companies and darkly threatened to start a new party. The good news is that this was pretty much it for Musk’s destroy-government project, DOGE; the bad news is that most of the damage is already done.

The Supreme Court has mostly allowed Trump to overthrow the rule of law, but at least they seem unimpressed with his declaring fake emergencies to levy tariffs. No final decision yet, though. (Two appeals courts have found his tariffs illegal.)

The Republicans got Jimmy Kimmel off the air for criticizing Trump, a chilling blow to free speech. But after widespread protests, he was back on the air after a little more than a week.

About 7 million people joined one of over 2000 No Kings protests in October— more than the previous event.

Trump has not been able to bully China. Xi Jinping holds the high cards here, largely due to China’s near-monopoly on rare earths… to say nothing of having the largest economy in the world (by purchasing parity). He’s also pissed off rather than intimidated India.

House Speaker Mike Johnson has been losing control over the House. An increasingly successful strategy is bipartisan groups using discharge petitions to force votes on their bills.

The Republicans, for years, built up hope in their base that the Epstein files would be released, presumably damaging their enemies. Once in office they released nothing. Then, this fall, Trump and Johnson both exerted enormous efforts to keep Congress from forcing releases of the files. This is pretty baffling: it annoys their own base, and makes everyone wonder what they’re hiding. It’s likely enough that nothing big is there: if there were anything that could be used against Democrats it would have been used months ago. (And if there were anything big against GOP , Biden would have used it.) It’s a mystery, and not a good look, why the GOP leadership tried to keep the files hidden. But also unsuccessful, which is a big defeat for Trump: Congress was willing to vote against him. To save face, he turned around and supported the release of the files.

On Trump’s Asia trip, many noted that he seemed to have trouble moving around, kept losing focus, and was more incoherent than usual. He seems to be experiencing an increasing cognitive decline. In the short term this probably is welcomed by the cronies and cranks who are handling him, since they can do as they want. In the longer term it’s a big problem for the GOP, since Trump is the glue that holds the party together. Though outsiders don’t see him as charismatic, his base sure does, and no one else has been able to do what he does.

On to the negatives.

On ICE’s campaign of terror, I can’t explain it better than this page does. Some quotes:

Every day my phone buzzes. It is a neighborhood group: four people were kidnapped at the corner drugstore. A friend a mile away sends a Slack message: she was at the scene when masked men assaulted and abducted two people on the street. A plumber working on my pipes is distraught, and I find out that two of his employees were kidnapped that morning. A week later it happens again. […]

On the night of September 30th, federal agents rappelled from a Black Hawk helicopter to execute a raid on an apartment building on the South Shore. Roughly three hundred agents deployed flashbangs, busted down doors, and took people indiscriminately. US citizens—some women and children—were grabbed from their beds, marched outside without even a chance to dress, zip-tied, and loaded into vans. Residents returned to find their windows and doors broken and their belongings stolen. Despite the violence of the raid, it appears no criminal charges were filed.

This is real fascism, and the only upside is that a lot of people who voted for the GOP are realizing that they don’t want this. The country needs immigtants, businesses need workers, Hispanics need to live in safety. Last year lots of Hispanics voted for the GOP; that shift was erased in this month’s election.

In previous posts I talked about the economic ruin caused by tariffs, hobbling education, impeding foreign tourism and study, alienating allies and China alike. The economy is propped up by AI boosterism, and that lools like it’s cracking.

But there’s a more insidious problem: the Republican goal is crony capitalism. There’s at least some amusement in letting the arch-libertarians at Cato explain it.

The administration has been bombing ships at sea. There have been threats to invade Venezuela, Colombia, and Nigeria (?!). Trump bombed Iran in June. Ironically such posturing annoys some of his base, which is not in favor of wars abroad.

Trump directed the Justice Department to indict his political opponents, notably former FBI director James Comey. As of this week this was falling apart, as the courts declared that his move of naming his personal lawyer as a federal prosecutor, bypassing the Senate, was illegal.

The Supreme Court has disallowed a few minor things (e.g. it ordered the return of Kilmar Garcia to the US— the administration is now apparently planning to deport him to Africa— and turning down a chance to undo its decision legalizing same-sex marriage), but it has done nothing to stop Trump’s power grabs, notably impoundment. (For the less politically obsessed: this means the executive ending programs at his whim that were authorized and funded by Congress. It’s illegal and unconstitutional: Congress decides what the laws are and how to spend federal money.)

The problems with having a demented toddler in charge are becoming evident. Trump held up a trade deal with Canada because a provincial governor ran TV ads he didn’t like. At the same time, thoes who know how to flatter and charm Trump become his favorites, even temporarily. Obviously the bad guys know how to do this most easily, but Mamdani got along with Trump surprisingly well. (Apparently he did this by being laser-focused on “affordability”, which Trump has belatedly realized is a big issue for Americans.)

Less than a week after the invigorating Democratic wins on November 5, eight Senate Democrats voted with the Republicans to end the government shutdown, widely seem as a pathetic betrayal. The Republican position on the government shutdown has been aptly described as “We’ll stop feeding poor kids until you let us take away millions of American voters’ health insurance.”

Climate change denialism continues. Of course it’d be pretty bad to ruin the ecosphere that keeps billions of us alive; but it’s truly idiotic to cede leadership on renewable energy to China. I don’t begrudge them their success— they’re making non-carbon-producing energy available all around the world. But it’s insane to cut back on our own promising renewables industry in favor of an all-in bet on climate-harming AI.

Oh, and just this week Trump has been pushing to end the Ukraine war by having Ukraine give into all of Putin’s demands. The details are readily available if you like stories where evil wins. However, the administration is now backpedalling furiously as Europeans, Ukrainians, and quite a few Republicans are objecting to Trump’s Putin-worship.

Krugman again:

The administration has been doing all it can to dismantle institutions, like the Consumer Financial Protection Bureau, that were created to help keep investors and markets safe after the 2008 financial crisis. Scott Bessent, the Treasury secretary, and other Trump officials and allies — including some officials at the Federal Reserve — have also been doing all they can to undermine bank supervision, which tries to limit the kind of risk-taking that brought on the 2008 crisis.

For some reason Youtube started showing me clips from The Ministry of Ungentlemanly Warfare. I liked them and then watched the whole movie twice. It’s a really fun movie and I recommend it to all my classmates.

I can’t do much better than jwz’s review:

This is great. It’s basically Inglorious Basterds but funnier and without Tarantino’s weird tics. You wanna see a feel-good romp about some Nazis getting fucked up? Oh yeah you do. How does Alan Ritchson just keep getting larger?

Basic idea: retell the story of Operation Postmaster, a commando raid in January 1942 which secured two German tugboats and an Italian cargo ship which had been up to no good in Fernando Po (now known as Bioko)— Spanish territory, therefore neutral, off the coast of Africa. This was one of the first missions of the Special Operations Executive, which (more or less) was later incorporated into MI6.

The raid was led by Gus March-Phillips, played here by Henry Cavill. At least half the movie concentrates on Gus and his four fellow commandos: chill brainy Appleyard (Alex Pettyfer), wild-man Dane Lassen (Alan Ritchson), pilot Hayes (Hero Fiennes Tiffin), and explosives expert Freddy (Henry Golding). Gus describes all the men as “bad”, but one of the charms of the film is how extremely polite they are to each other. They’re ultraviolent against Nazis, but that’s how you have to treat Nazis. The film is from last year when maybe Nazis seemed more historical.

Now, two things you need to know to be in the right mood for the movie: 1) it’s comic, and 2) it’s a video game. That is, it is not a super-serious re-enactment of historical events. The real Operation Postmaster in fact had no fatalities. The action sequences in the movie are just one step beyond the fight scenes in Ritchson’s Reacher: cinematic demonstrations of beautifully choreogaphed, carefully justified force.

Some Internet comments have made fun of Ritchson’s accent, but a lot of this is I think based on the first scene— where he and Cavill are pretending to be Swedes to fool Germans. It’s intentionally over the top.

The other half of the movie concerns the spy work on land, focusing on the composite character Heron (Babs Olusanmokun) and the movie star turned spy Marjorie Stewart (Eiza González). Their main antagonist is the sadistic Nazi Luhr (Til Schweiger). All this is worth watching once— González in particular is very easy to look at— but I have to admit I fast-forwarded through it the second time. Marjorie basically has to seduce the Nazi commander to keep him distracted during the raid, which mostly involves some mutual recitations of poetry and verbal sparring… it’s all well acted, and very important to the plot, but not very satisfying.

(I think it’s the tonal shift that doesn’t quite work. When we’re juiced up by scenes of killing Nazis, trading repartee with one is jarring. And maybe director Guy Ritchie’s heart was just back on that boat. Also, honestly, there’s a structure problem here. The commander is important because he’s the commander, but he’s given little to do.)

There’s also a subplot involving British officers insisting to Winston Churchill that Britain should surrender to the Nazis. To my knowledge this is ahistorical, or at least anachronistic: by this time the Battle of Britain had been won and the US had already entered the war. It’s probably libellous to Admiral Pound, but he died in 1943.

I only bring up the historical facts because the movie will make you want to know them. It’s just a fun story well executed, with some really hilarious bits.

These two books have nothing to do with each other, except that I read them recently and neither needs a very long review. Well, and they’re both comic, and the second installments in series where I haven’t read the first.

HOW TO STEAL A GALAXY – Beth Revis (2024)

I’m a sucker for a good Catwoman story, and that’s what this is, despite being SF and not technically being about Catwoman. But it’s about a very clever female thief who has the hots for a straight-arrow near-cop, so it’s a Catwoman story.

It’s the story of a single heist. The thief is Ada, who has her own spaceship and likes to steal things for money, though if it accomplishes something good at the same time she is happy about that. The near-cop is Rian, who is apparently the only good guy to work for a corrupt government. The heist is to take place at a gala charity auction held on Rigel-Earth, where a Very Important Sociopath is the guest of honor. The target is… well, Ada doesn’t say until very late in the book.

In this universe there are portals between star systems, which has allowed humanity to colonize several star systems, with the somewhat odd habit of naming all their colonies after Earth. Earth itself, now Sol-Earth, is a mess– there was a supervolcano eruption, among other things– and needs saving, which is what the corpo bigwig is offering.

Now, this is Book 2 of a trilogy, and I haven’t read the first (I intend to). So I am missing some backstory, but it works as a standalone story. Rian and Ada met, or tangled, back in Book 1.

Does it work? Well, if you like heists and cat burglars, yes. It’s not deep but it’s fun, and Ada (as narrator) is snarky and intriguingly cocky. Plus it’s short, so not much of a time investment. The romance element is, well, about what you’d get in a Catwoman story: it’s probably doomed because of incompatible careers, but it’s a fascinating diversion for both parties.

FALLING TOWARDS ENGLAND – Clive James (1985)

This is a re-read… I don’t remember why I picked it up; probably because the first pages seemed fun. It’s autobiographical, or nearly so.

James was Australian, and the book covers his first two years in England, starting in 1962 when he was a very young and very immature 23 years old. James ended up as a poet, novelist, critic, and TV commentator, though in the period of the book there was little promise of that.

A sample, which gives a good idea of his rollicking style:

[The British hamburger] looked like a scorched beer-coaster or a tenderised disc brake. Flanked by chips which, if picked up individually on a fork, either shattered or else drooped until their ends touched, the British hamburger lay there sweltering under its limp grey duvet of over-fried onions. [After eating it], the British hamburger shaped itself to the bottom of your stomach like ballast, while interacting with your gastric juices to form an incipient belch of enormous potential, an airship which had been inflated in a garage. This belch, when silently released, would cause people standing twenty yards away to start examining the soles of their shoes. The vocalised version sounded like a bag of tools thrown into a bog.

Now, I’ve noted before that in American humor, the comic is the smart guy exposing a foolish world; in British humor, the comic is an idiot in a normal world. James is very much in the latter tradition: in his memoirs he’s irresponsible, a listless worker, constantly cadging money from everyone he meets, and a terrible drinker and smoker. (A terrible drinker not just because he drinks too much, but because he gets drunk abnormally fast.) Also he has terrible teeth and thinks he’s a poet.

I still enjoy the section where he gets one of his dead-end jobs: taking consumer surveys. Only no one on the street wants to answer them. He does get a full set of answers from one man, but it turns out he’s from Sweden, and the company only wants Brits. So he marks him as from Swindon and turns in the answers. Soon he realizes that the only way to get results is to make up all the answers. He’s caught when the statisticians realize that the answers are way too similar.

The book is a fascinating glimpse of Swinging London— the period when postwar London had regained enough prosperity to be fun and irreverent, but not so much that a near-broke Australian couldn’t join in.

The one bit that hasn’t aged well is his treatment of women. It’s not that he was abusive: everything was consensual. It’s that he was exploitative. He and everyone he knew let the women cook for them, give them money, put them up. He recognizes that this was a fault but I think sounds much worse today than he realized even while writing twenty years later.

I just finished Why Nations Fail: The Origins of Power, Prosperity, and Poverty, by Daron Acemoglu and James A. Robinson (2012).

First, it’s technically Acemoğlu. The ğ in this case just lengthens the previous vowel. However, he doesn’t seem to use the diacritic professionally, so I won’t either.

This is all about development economics, one of my fascinations… because it’s important for just about everything: why are some countries rich, how did they get that way, how do you make the poor countries richer, why is it so damn hard. My Almea+400 project can be said to be a rumination on the process for a fantasy world.

A&R’s book is a Theory of Everything, comparable to earlier attempts like Jared Diamond’s Guns, Germs, and Steel, or Graeber & Wengrow’s The Dawn of Everything and Jane Jacobs’s The Economy of Cities. Oh, and stuff like The Wealth of Nations (Smith), the Communist Manifesto (Marx & Engels), and De la démocratie en Amerique (de Tocqueville).

As it happens their thesis is not hard to explain; what takes them 464 pages is to tell stories to back them up. Their basic idea: the difference isn’t due to geography or culture or religion or (ugh) ethnicity, but to institutions. The basic distinction is between extractive and inclusive.

Well, really, there’s a third category, which we can call unorganized. It’s where you start before agriculture, as no one is in charge; also where you end up if the state doesn’t control its own territory, as in Somalia or Haiti or Afghanistan. This condition has its points in the ancestral environment, but in modern times it basically means that life is controlled by armed clans or gangs, unable to best each other but always trying.

An extractive country is strong enough to control its territory, and what it does is extract all its wealth for the benefit of a small elite. Historical examples: almost everyone, from the Mesopotamians to the Romans to the Chinese Empire to medieval Europe to the Incas. Every single colonial state, except for the ones with settlers (Australia, Canada, the US). In modern times it’s the norm in South America and Africa. The Russian Empire, the Soviet Union, modern China.

As some of these examples hint, thanks to what they call the “iron law of oligarchy”, when an extractive system is overthrown, a new one often takes its place, using the exact same systems to benefit a different class of people. Many an African nation went from extractive local elites, to extractive colonial administrations, to new extractive local elites. Similarly Peru went from extractive Incas to extractive Spaniards to extractive local criollos.

An inclusive country is one with broader political power. This can start on the small side; an example is medieval Venice, where junior traders could partner with senior ones, working their way up. (The Venetians later closed off this route… and declined.) If all goes well, the process continues: more and more people are allowed power, their property is safeguarded, they can work as they like, they can implement new ideas without their work being expropriated. Power is too broadly based for new extractive classes to emerge.

Their prototypical example is the Glorious Revolution of 1688 in England, basically a coalition of city merchants and manufacturers and landholders opposed to the absolute power of the king. There had already been one revolution, which had simply substituted the rule of Cromwell for that of the king. This one kept the (new) king’s head but greatly limited his power. Ideally a “virtuous circle” is set up, where reforms can gradually incorporate the whole population into the ruling class. (Britain took 240 more years to get to full adult suffrage.)

A&R are big fans of creative destruction… which elites hate, because it threatens their extractive power. The rulers of Austria-Hungary and Russia, for instance, resisted railroads even in the late 19th century. The same idea persists today: e.g. the dictator of Sierra Leone actually ripped up a railroad that led to his enemies’ homeland.

(The term derives from Joseph Schumpeter, who saw competition and innovation as a core part of capitalism. Ironically, the term is probably most popular today among people who want to create new rent-seeking monopolies— i.e. people who want to forestall competition and thus creative destruction happening to them.)

A&R emphasize that an inclusive system will always grow and develop. An extractive system may grow– indeed, it may produce a burst of wealth, as the Spaniards received in the New World, or the Soviets in the decades after the Revolution. You can always catch up, apply existing technology, be more efficient. But it always peters out, because the elites are satisfied with being filthy rich and terrified of anything that can threaten their position. They’d rather throttle development than open up the system.

They have a somewhat sad final chapter where they explain why so many development schemes fail. It’s because they are band-aids on an extractive system and don’t address the underlying problem. You can give aid, you can educate, you can implore the elite to stop being corrupt, but the country remains in stasis. Some countries do change course (Japan, South Korea, Botswana, Brazil), but it’s really really hard… and outsiders generally apply very simplistic solutions. (E.g. holding elections doesn’t by itself produce democracy. Nor does, say, creating an “independent central bank”; they mention the central banker of Sierra Leone, who was simply murdered when he ventured to criticize the president.)

I think these terms, like the 1%/10% given to us by the Occupy movement, are going to be extremely useful in analysis. The first question about any country is “who is in control”, but the second question is “are they extractive or inclusive?” It’s clarifying, for instance, about the program of the Republican Party since 1980, which is to reverse inclusivity in the US and return to the extractive policies and narrow elite of the Old South.

Did anyone still think colonialism was a benefit? Colonialism was a horror. Typical episode: Cloves were once found only in the Ambon islands; the native king extracted cloves from the growers by forced labor and sold them to traders. The Dutch took over the islands and the forced labor. This was typical of what all the European states tried to do.

Just to the south were the Banda islands, the only source of nutmeg and mace. (These are both products of the same nut.) The Dutch took them over, but there were no native polities to take over, no one to coerce. In 1621 the Dutch solved this problem by taking a fleet to Banda and massacring almost the entire population, at least 15,000 people, leaving only a few survivors who knew how to cultivate the nutmeg/mace tree. They divided the islands into estates and filled them with slaves to work the nutmeg plantations.

16th century adventurers only wanted to apply the Spanish model: come in, take over already-extractive native institutions, and reap the profits. Almost amusingly, the initial founding of Buenos Aires failed, because the local natives were not organized into kingdoms that could be taken over. The founders moved to take over Paraguay instead. This was the model the English attempted in Jamestown, too, and it was a complete failure: the Native Americans refused to work for the colonists, and since these were not the type to work, most of them starved. The English colonies only succeeded once settlers willing to work replaced the conquistador wannabes.

Some points to refine the theory:

• Societies changes in response to crises, what they call critical junctures. One was the Black Death, which killed half the population wherever it struck; the initial effect was to make labor more valuable.
• Small differences can turn into big ones. In Western Europe, the Black Death led to better treatment of workers; in Eastern Europe the extractors were more powerful and created serf-based empires. The English government was unable to establish a monopoly on the Atlantic trade, unlike the French and Spanish; this created an opening for entrepreneurs in England. The success of Botswana in Africa is partly due to it being extremely marginal during the period of British rule, and to the assignment of mineral wealth to the independent state– and a government inclusive enough to use the largesse to benefit the people rather than an elite.
• History is contingent— meaning that there is a certain random element.

Now, I have a few problems with the book.

They are very very good on the horrors of colonialism; they seem way less aware of the evils of neocolonialism. E.g., listing a long parade of extractive regimes in Guatemala, they write “Though [the dictator Ubico] was followed by a democratic regime in 1945, this was overthrown by a coup in 1954, leading to a murderous civil war.” Yes, that’s bad, but it’s not a matter of bad institutions in Guatemala, it’s because the US orchestrated that coup. They talk about dictators who took over after independence in Africa, without seeming to notice that it was the US and the Soviets who preferred dictators there and installed them routinely. This tragic farce mostly ended when the Cold War ended.

Or: they use Haiti as an example of an unorganized state. But Haiti wouldn’t be in the state it’s in if it weren’t for the crippling reparations to slaveowners demanded by France in 1825; Haiti was still paying this debt in 1947.

They’re generally pro-union and anti-monopoly, and mention the Civil Rights movement positively as a movement toward inclusivity; an unexpected hero of the book is Brazil’s Luiz Inácio Lula da Silva, who helped build the coalition that ended the military dictatorship, and is now President. But they don’t even bother to defend their main thesis against, say, democratic socialism. They’re down on communism– OK, authoritarian communism is bad, but that’s not the same thing. They consider Britain as a hero of inclusivity… well, what about the horrible conditions of the coal mines in the 1930s, as reported by Orwell? They could answer that it was a consequence of the Depression, or that it was a scandal that extractive institutions persisted that long, but they don’t even address it. (Acemoglu was born in 1967 and sometimes it seems like his idea of the UK and US is formed by the 1970-2000 period.) If absolutist leaders are a plague, what about absolutist CEOs?

The book was written in 2012, so it doesn’t address Trump. But there isn’t any discussion of the Reagan/Thatcher move toward plutocracy, or the Republican campaign against electoral reform, regulation, health care, progressive taxation, and education for Blacks; not a word about productivity gains since 1980 going only to the 10%. In short, the return to plutocracy, which in their own terms is extractive. There’s a whiff of centrism and Anglosphere exceptionalism here– maybe they didn’t want to lose those blurbs from Niall Ferguson and Steven Levitt. (Both Acemoglu and Robinson have been vocally anti-Trump, and warned that the US could fail based on his retreat from inclusivity, democracy, and economic openness. Great, but these GOP tendencies weren’t unknown in 2012 and they just ignore that.)

The most interesting parts of the book are the in-depth discussions of various African and South American countries. Their treatment of China is however perfunctory. They take Mao as a disaster, for instance, which is superficial and arguably wrong. The 1950s were a huge advance for China: as they themselves recognize, China was an extractive mess which was unable to reform as Japan and South Korea had. The elite and foreign incursion had to be broken, and the Nationalists simply were not doing it. Mao did, and the country quickly modernized in the 1950s… largely because Mao left day-to-day rule to others. His Great Leap Forward and Cultural Revolution were disasters, but on the larger scale they were far less disruptive than Stalin’s disasters and Brezhnev’s stagnation. And though I broadly agree that political liberalization has to come, the Chinese are showing innovation in AI and renewables that isn’t simply “catching up”.

How does this interact with Diamond, Graeber, Jacobs, Ha-Joon Chang, Hernando de Soto, and Marx? To some extent you should just read them all and make your own synthesis. How nations develop need not be due to single factors.

Jacobs, for instance, ties differences in development to import substitution. This is not at all incompatible with A&R’s point of view; it’s good to note, however, that import substitution is not going to happen in a highly extractive society. The Japanese could do it because their merchant class already had a high degree of independence. In most of modern Africa, if anyone tried it, their profits and probably their businesses would be seized.

A&R disagree with Diamond’s explanations based on geography, pointing out that (say) Mexico and the southwest US, or North and South Korea, developed very differently despite having identical climates. But Diamond is not talking at that level; he’s addressing the question of where (say) agriculture should develop. A&R address this, tying it to sedentism, and suggesting that hunter-gatherers had to be “forced to settle down.” This is the weakest part of the book. James Scott should be required reading here: there’s no evidence that these early sedentary villages were tyrannical, and you just can’t order hunter-gatherers around like that. The transition took from wandering to sedentism to agriculture to civilization took literally 8,000 years. We can’t reconstruct all the stages, but there were undoubtedly many, and state coercion was one of the last steps, not the first.

A&R’s message is also fairly compatible with Hernando de Soto of Peru, who advocating for incorporating the bustling informal economy into the formal one, which was supposed to provide access to capital. Actual attempts to do so in Peru do not seem to have worked very well; as one scholar noted, “Titling must be followed by a series of politically challenging steps. Improving the efficiency of judicial systems, rewriting bankruptcy codes, restructuring financial market regulations, and similar reforms will involve much more difficult choices by policymakers.” De Soto doesn’t even appear in A&R’s index, which is a pity, since his approach is indeed to change institutions, and he had actual influence in Peru, but didn’t achieve the results he foresaw.

An alert reader pointed me to this very interesting critique of A&R. Some points I agree with– e.g. A&R seem incorrect about literacy in premodern Africa. But the critic doesn’t seem to have read very closely; e.g. he thinks that A&R only talk about Kongo and Somalia, but there are also extensive discussions of Sierra Leone, Ethiopia, South Africa, Egypt, Ghana, Nigeria, and Botswana. The critic talks about the king of Kongo mostly selling slaves acquired as war captives outside his territory, without seeming to realize that this is exactly what A&R say about African kings.

The article also complains that African “agency” is not respected and that premodern kingdoms were not as absolutist as A&R seem to say. This strikes me as enormously missing the point. Sub-Saharan Africa in general was oriented toward selling slaves (first to Arabs, then also to Europeans); it was unable to resist colonization; and it turned European extractive systems into native extractive systems. We can try to think better of those premodern states, just as we can try to find the good in the Aztecs: besides the slavery and cannibalism and self-mutilation, they were nice people. But whatever good we find wasn’t enough to stop those evils; and “agency” does not mean exoneration; it may also mean complicity.

A better critique might be that people reading A&R might just give up: if poor countries have bad institutions we can’t really change that, can we, and why should we try? A&R are not optimistic about foreign aid, much of which is wasted or appropriated by extractors; on the other hand, as they say, a lot of it saves people from dying and some help is better than none. Attempts by the US or IMF to change institutions have done little good, largely because extractive elites are not really interested in reducing corruption, respecting property, or inviting more people to share power.

A&R should be a lot clearer, I think, about how Western institutions have failed against neocolonialism. The West actually promoted or established many of those extractive local leaders, and threw them out if they attempted a more inclusive system (see my list of interventions!). The West has also been happy selling luxuries and weapons to dictators, or building expensive but misguided factories (see my Jacobs page). China is now getting into that game too, and competition is good, but I see no evidence that Chinese investments are more inclusive than Western.

Studies on development often tell us not so much what to do as what not to do. The positive note here is that A&R give us more arguments for democracy, egalitarianism, the rule of law, and empowerment of workers. It’s also more hopeful if bad outcomes are based on institutions rather than “culture”, because institutions can be changed. But quick fixes and even changes of regime generally are not enough.

I’ve just read Daniel Dennett’s last philosophy book, From Bacteria to Bach and Back: The Evolution of Minds (2017). My overall reaction is that it’s a step down. If you’ve never read him, instead try Consciousness Explained (CE, 1992) or Elbow Room: The Varieties of Free Will Worth Wanting (1984).

Bacteria to Bach is long and ambitious, yet breezy and self-indulgent. He was 75 when he wrote it, and I feel like it shows. It’s a summarization of his lifelong views, and it often feels that he is tired of making these points to uncomprehending philistines– can’t they just admit they’re wrong? Unfortunately this attitude doesn’t produce the slow but scrupulously argued framework of CE, but long repetitive rants that assert rather than demonstrate his ideas.

Now, many of his ideas I completely accept:

1. The mind is ultimately a machine: no immaterial substances are needed to explain it.
2. This does not make it either deterministic or valueless. Our emotions, thoughts, values, and rational proofs are all still there.
3. Minds ultimately derive from evolution.
4. Evolution itself is a mindless process that mimics intelligent design. Because success is rewarded with reproduction, it is not entirely random, but can both innovate, and spread innovations.
5. There is nothing wrong with speaking of biological features as having utility, and indeed design. There’s just no designer.
6. Humans, at least, have a new quasi-Darwinian process of cultural evolution, which allows much faster innovation and spread of ideas, this one not tied to inheritance. You can’t grab another animal’s genes, but you can steal another culture’s ideas.
7. Humans also have real intelligence– in effect we have an internal computer which lets us imagine and reason, a third way of producing good ideas (and some bad).
8. Looking at cognitive abilities, we should always consider them as a continuum from mindlessness to full consciousness– not least because this is the only way we can understand how minds evolved. Searle-style insistence that understanding must be all-or-nothing and human-level is a mistake.
9. The “Cartesian Theater”– the screen in the brain watched by a homunculus– is simply not there. There is no central control room in the brain, no organelle which we can label “consciousness.”
10. Both introspection and perception can be fooled.
11. Our beliefs or claims about our own consciousness must be taken with a grain of salt. A good example is syntactic knowledge: we follow grammatical rules that our brain knows, but which are not directly available to consciousness. (I covered this in the Syntax Construction Kit.) When we say we know a language, it’s true that our brain knows it, but “we“, our consciousness, are only part of our brain.

So, what don’t I like?

First, he has a pair of beliefs that seem to be both dubious and contradictory. One is that only humans have full consciousness: even bears and chimps, to say nothing of bees and bats and dogs and grey parrots and dolphins, may be without real comprehension, there may be no one at home there. The other is that our own consciousness is a user illusion— we do not have special insight into how our minds work, there is no such thing as qualia, and much of our apparent competence is not actually “ours”. It bubbles up from the brain and we take credit for it.

I don’t know why he wants this division between humans and other animals— it’s profoundly out of tune with the last century of biology. If you read Konrad Lorenz you’ll see the deep similarities between human and animal behavior… even down to the level of fish. He actually makes his central task harder– explaining consciousness– if he can’t grant it to dogs and chimps.

He wants to tie human abilities to language… OK, but when a grey parrot can learn 200 words, it gets a lot harder to draw this bright line. His idea is also strangely oblivious to his own repeated and correct point that aspects of human cognition must derive from something simpler.

His chapter on qualia has a bullying tone that gets on my nerves, e.g.:

By offering a sketch of the causes of Cartesian gravity, I have tried to help the unpersuaded find a vantage point from which they can diagnose their own failures of imagination and overcome them.

Sometimes a disagreement does derive from a failure of imagination, but you have to show it, not assume it.

What Dennett is asking for is pretty much what philosophy, and some religions, have been asking us for 2500 years: to understand that we can be fooled, that our mental imagery might be fake. Plato made the same point with his cave analogy, Descartes with his malicious demon. That we don’t really have a soul, but are only a place where perceptions swirl around, is longtime Buddhist doctrine; also Humean doctrine.

It’s a point of view, but I’m always reminded of Samuel Johnson’s response to Berkeleyan idealism: kicking a stone, he says “I refute it thus.” I’d read one anti-qualia passage after another, then look around my bedroom at the exquisite picture of the world presented by my personal brain, and start to wonder, doesn’t Dennett have that? Or: Does he think ChatGPT has that?

Yes, we have dreams, hallucinations, optional illusions, mistakes in perception. But the common-sense perspective that these errors don’t matter much is not disproved. CE actually started with a very convincing argument that the “brains in vats” scientists have an immensely difficult task in front of them. One can recognize that the brain is creating our perceptions rather than just passing them along, but the amazing thing is not that our perceptions can be fooled, but that they’re as good as they are.

See also Anil Seth‘s book on consciousness, which offers the intriguing idea that perception is far more top-down than we might expect. That is, our perceptions are a creation of the brain, which after all is imprisoned in a dark bony cavity with only nerves coming in. We see things, but not with light: there’s no light in the brain. Rather than simply building up a picture of the world from sense data, the brain is creating that picture, then testing and confirming it with sense data.

Dennett reproduces an optical illusion: a color-reversed American flag. If you stare at it for 30 seconds, then look at a blank page, you will see a “properly colored” illusory American flag. It’s an effect easily explained by neural anatomy: nerves exposed to the same stimulus get tired: stare at the green stripe long enough and it looks less green. When you stare at the white page afterward, that part of your visual field is temporarily discounting green, and what that looks like to us is red.

Dennett asks, with the air of producing a gotcha, if the red stripe we see is really red. But it’s a meaningless question. Of course it’s an illusion, there is no stripe in the world, and no redness. Is there a “red stripe” in the brain? There’s a representation of one, yes– just as there was a representation of the “real” green stripe. Why is the difference important? He could go on to show, as C.L. Hardin does, that the stripe on the page need not be green either. (In a printed book, green is produced by a mix of yellow and blue dots.) The “green” is in our eyes or brain. And, well, so what? These things and these illusions are interesting, fun to think about, but don’t actually disprove qualia… they are qualia.

Back in CE, Dennett had some excellent critiques of the Cartesian Theater. E.g. he points out that we literally cannot see in the region of the visual field interrupted by where the optic nerve pushes into the retina. There’s a quite large gap there, a gap we do not perceive. He argues convincingly that it’s wrong to say that the brain fills in that gap. The brain only has to answer questions that it asks, so to speak. There is no part of the brain that asks or needs to be told what’s in that gap.

Part of the answer to this is saccades: the eye is constantly moving, so what’s on the retina is not a fixed image anyway. The brain turns that ever-changing kaleidoscope into what looks to us like a fixed image. But it’s not correct to say that our perceived image is high-resolution. Rather, the brain uses a trick: when it needs to (e.g. for reading), the eye gets a lot of detail, by aiming the high-res part of the retina, the fovea. Because whatever we’re looking at is hi-res, we are not aware that the rest of the visual field is lo-res. We have to be very careful not to treat as facts things we merely suppose about our qualia. That is, just because we’re not aware that our perception is lo-res except for a moving hi-res spot, doesn’t mean that our perception is “really” hi-res.

But the later Dennett seems to have mislaid the plot, denying things that don’t need to be denied. Surely it’s not satisfying to explain bits of consciousness (qualia) by just writing them out of existence. If a mental patient said some of the things Dennett does, they’d be hospitalized. (There are mental conditions where a person thinks they’re dead, or blind, or their legs don’t belong to them.)

The other bit I don’t care for is his extended discussion of memes. He wants to use these to explain cultural evolution, but he never succeeds in showing why memes, in particular, help out. I think he’s seduced by Dawkins’ original parallel to genes, which he famously called selfish: genes can be reified as things that have goals of their own, using animals (and plants) to reproduce themselves! (Biologists way over-emphasized genes in the 1990s.) Likewise memes can be seen as idea-complexes that colonize our minds for their own benefit, that benefit being defined in terms of mutation, and competition to reproduce.

The problem is, Dennett does not actually show that this notion explains anything particularly well, besides what “memes” mean in popular discourse: viral jokes. You can get a frisson of contrarian thrill by picturing ideologies and religions as memes, pointing out that they usually contain antipatterns that facilitate spreading the idea and discouraging abandoning the “faith”. Fine, I included a whole chapter on “stickiness” in my Religion Construction Kit: it’s a useful question to ask not just what a belief system says, but what’s attractive about it and how it protects itself. But memes as an idea… memes as a meme… are strangely underpowered. They seem mostly to be a way of disparaging an idea we don’t like. I don’t think Dennett ever calls evolution itself, or the scientific method, or Bach’s piano music, memes. You don’t need the catchy name to get across the idea that useful ideas spread, or even to point out that “useful” may be a loaded term here.

He tells us that words are memes, in a way that suggests a gotcha that never comes. All in all he talks about language in a way that belongs more to the 19th century than the 21st. It’s profoundly un-linguistic to talk about languages or words competing with each other, and the “better” ones winning. People and nations compete with each other, but neither languages nor words do. Does it mean something that we say “dog” and the Spanish say “perro” and the Japanese say “inu”? No, it really doesn’t: it’s effectively random. Sound and symbol are (usually, not always) separate; none of these words are better and none of them compete.

Now, Spanish replaced Latin “canis” with “perro”. Aha, evolution by natural selection! Only, no. Meanings change, words are borrowed, but “perro” is not better in any way, it did not win the meme wars. Language change is largely value-neutral. Memetics is the cryptocurrency of culture studies, a solution in search of a problem.

If you insist on thinking about the origin of language— generally a futile, self-deluding pursuit– Dennett offers some other, better ideas. E.g. it’s worth recognizing that early language must have been impoverished yet still had to be useful, for either genetic or cultural evolution to boost it. It would not have started with a mastery of 10,000 words… probably not even 1000. His idea seems to be that words started as viral bits of repetition: it was faddish or fun to reproduce these eructations. I don’t see that as compelling, but as a conlanger I do want to ask: what’s the minimal near-language that’s useful? It has to be far far simpler and easier than pidgins, and probably within spitting distance of primate calls and gestures.

Dennett does have other useful insights. E.g. he quotes Emerson Pugh: “If the brain were so simple we could understand it, we would be so simple we couldn’t.” So, maybe the brain is too complex for a human to understand. But he goes on to ask: is it too complex for a group of people to understand? It’s an excellent question. A lot of pop history of science is about lone geniuses, but most advances today are made by small teams… or large teams. Maybe consciousness will turn out to be something that, in effect, requires a whole university to understand.

His central insight from CE is also worth retaining. There he posited that language allowed people to ask each other questions. If this became automatic enough, they could ask themselves questions when they were alone, at first out loud. We can still do this, and it works! It focuses the brain on the question and sometimes an answer pops out. We can imagine the process improving by both genetic and cultural evolution until we get our modern internal stream of consciousness.

The last chapter is on AI, long one of his preoccupations. It’s interesting to see his reaction to early deep learning. Kind of unfortunately, he was writing about 5 years before ChatGPT came out. So he is simultaneously a bit breathless about how great the AI of 2017 was, and skeptical about how good it could get. But he does, to his credit, recognize the problems of overinterpretation:

The real danger, I think, is not that machines more intelligent than we are will usurp us as captains of our destinies, but that we will over-estimate the comprehension of our latest thinking tools, prematurely ceding authority to them far beyond their competence.”

He suggests that AIs, like medicine, should come with long lists of possible shortcomings, and that “systems that deliberately conceal their shortcuts and gaps of incompetence should be deemed fraudulent, and their creators should go to jail”. Too bad no one listened to that.