Media highlights 2024 (July to Dec)

There’s this great blog post, from Luke Muelhauser, called “Three wild speculations from amateur quantitative macrohistory”. Here is the nub:

I help myself to the common (but certainly debatable) assumption that “the industrial revolution” is the primary cause of the dramatic trajectory change in human welfare around 1800-1870, then my one-sentence summary of recorded human history is this: Everything was awful for a very long time, and then the industrial revolution happened.

This is Clark's starting point. The task of the book is to explain perhaps the three most important features of long-run economic history:

(1) The ‘Malthusian’ era, in which subsistence incomes persisted throughout the preindustrial world (from settled agriculture to around 1800), despite better medicine and advancing scientific knowledge, owing to a coupling between fertility and income, plus a fixed trade-off between population and material income per person (2) The Industrial Revolution — sustained and rapid economic growth, fueled by increasing production efficiency through advances in technology; and the demographic transition, a decline in fertility relative to income, translating the efficiency advances of the Industrial Revolution into a stratospheric rise in incomes per person (in much of the world); (3) The ‘Great Divergence’, in which the efficiency advances and sustained economic growth triggered by the Industrial Revolution ended up very unevenly distributed across the world, such that income per person in the richest countries is around 100 times greater than those in the world's poorest countries, who are today not significantly better off than before 1800 and in some cases worse-off.

The Malthusian era is the most counterintuitive but the best understood. It is really a kind of predictable default: (i) birth rates increase and death rates decrease as material living standards increase, and (ii) material living standards decrease as population increases (because there is less to go around, and because of dimishing returns to more labor holding other inputs fixed). The result (modulo short-term shocks) is subsistence — birth rates equalling death rates.

This gets counterintuitive: anything that makes life at a given income more deadly — poor sanitary practices, war, disease, etc. — increase the income required to maintain subsistence, so increase material living standards in the long run, and vice-versa. A technological advance — like a way to more efficiently grow crops — increase population size feasible at an income and income feasible at a population size. But assuming the birth and death rates as functions of income do not change, then the result is simply more people, not raised material standards of living.

Hence real income in Malthusian societies was fixed by birth and death rates alone, and then population size depended just on how many people a society could support at that income level given the available fixed inputs like land and production technology:

Jane Austen may have written about refined conversations over tea served in china cups. But for the majority of the English as late as 1813 conditions were no better than for [hunter-gatherers].

Yet there are no deep mysteries to this (from the modern perspective) upside-down world. Per Clark, “[i]n economics the known world thus stretches from the original foragers of the African savannah until 1800.”

The real question is how much of the world broke permanently free of the Malthusian regime, and got extremely rich in historical terms. Clark is good at knocking down popular single-factor explanations here, like the scientific institution-centric view of Joel Mokyr, the Protestantism/individualism-centric view of Henrich, or the ‘inclusive institution’-centric view of Acemoglu and Robinson. Contra Acemoglu and Robinson, Clark makes a pretty compelling case that all the institutional foundation-setting you could possible want was already in place centuries pre-1800. That is, to the extent it was in place at all around 1800, since the innovators in English textile manufacturing — surely those people most responsible for igniting the growth to come — were rewarded spectacularly poorly. And contra Henrich:

Protestantism may explain rising levels of literacy in northern Europe after 1500. But why after more than a thousand years of entrenched Catholic dogma was an obscure German preacher able to effect such a profound change in the way ordinary people conceived religious belief? The Scientific Revolution may explain the subsequent Industrial Revolution. But why after at least five millennia of opportunity did systematic empirical investigation of the natural world finally emerge only in the seventeenth century?

As for the divergence in fortunes between post-industrial countries:

At a proximate level capital per person explains perhaps a quarter of income differences across countries in the modern world. But with capital free to flow across countries, and earning a rental that differs little across income levels, efficiency differences explain most of the variation in capital stocks. So at a deeper level efficiency differences are the core of the variation in income per capita across economies since the Industrial Revolution.

This claim is well evidenced but not well explained. Indeed the book seems to end a couple chapters early, and the bow is never really tied on the overall argument.

Still, this is a phenomenally insight-dense and non-patronising book (especially the dozens of tables of historical economic data that presumably took a bunch of grad students hundreds of hours to source).

God clearly created the laws of the economic world in order to have a little fun at economists’ expense.

Bacteria haven't become significantly more complex in billions of years. After all, they would need a larger genome to carry the extra information required to code for more complex functions, but extra genetic material is just costly baggage before it reaches a critical ‘usefulness’ threshold. So there is in fact downward selective pressure on bacteria, toward simplicity. Yet at some point in the history of life eukaryotes emerged from the prokaryotic ooze, and they got a foothold on the evolutionary ladder. Cells became differentiated and organised together into big complicated multicellular organisms, their genomes increased in size; and this is the reason anything interesting happens.

But how? There isn't really any cosmic force toward the complex; no ‘great chain of being’. We might imagine (as Lane does) that most ‘simple’ life in the universe simply gets stuck as bacteria-style sludge without ever hitting on multicellularity, sex, predation, and interestingness. The truth is we don't exactly know, but we have some fascinating guesses.

Also, I love the chutzpah to write a fairly technical nonfiction book about a tiny organelle and choose that title.

An epic and basically successful 2-part attempt to write a sweeping political history of the world from the first states to the 21st century.

The key organising device was to think about societies in terms of state capacity, the rule of law, and accountability. ‘State capacity’ meaning something like ‘a bureaucracy to accumulate and administer the power of the state’. It can be efficient and well-organised; or corrupt, nepotistic, clientelistic. ‘The rule of law’ meaning some kind of judiciary, with some authority even over political leadership, and largely insulated from their whims. And ‘accountability’, meaning the decisions of the state (and the choice of leadership) are subject to correction by more or less of its subjects — basically, free and fair elections. For Fukuyama, ideal modern states (liberal democracies) score highly on all three counts.

But, when you think about it, it's really not obvious how some societies got a relatively strong state, rule of law, and accountability. Beginning with an autocratic state, for instance, why would it ever ‘agree’ to subject itself to democracy and courts?

I liked these books because Fukuyama is not cherry-picking from history to illustrate his ‘one big idea’. He is trying to take political history in all its massive complexity, figuring out how to organise it in englightening ways, drawing out themes but noticing exceptions, showing connections and nicely insightful explanations for them, and mostly steering clear of sweeping, clever-sounding, abstract theses (a la Sapiens).

The analysis of the US as ‘vetocracy’ was especially good I thought. And despite his reputation as the ‘end of history’ guy, we get the right answer to “will more and more societies join the ranks of liberal democracy”, which is, “it’s complicated”.

On most counts, the deadliest conflict since WWII (including civilian deaths) is the 1998 Second Congo War. The figure is three million lives through 2004, and estimates run over five million accounting for all conflict from 1996 to the present day. That surpasses Iraq, Vietnam, Korea, Afghanistan, Russia-Ukraine.

Like (presumably) most people, I knew close to nothing about the DRC and its history, and it gets very little airtime on western news. Perhaps for similar reasons: “NATO sent 50,000 troops from some of the best armies to Kosovo in 1999, a country one-fifth the size of South Kivu [one of 26 provinces in DRC]. In the Congo, the UN peacekeeping mission plateaued at 20,000 [ill-equipped] troops”

Unlike other conflicts, where there is often a clearer ‘opressor’ and ‘opressed’, or a single injustice to right; “[t]he Congo war had no one cause, no clear conceptual essence that can be easily distilled in a couple of paragraphs […] it is a mess of different narrative strands—some heroic, some venal”.

And when conflict in the DRC does show up in journalism or popular consciousness (in the west), Stearns complains about the attitude of morbid fascination, only “focusing on the utter horror of the violence” and portraying “an inscrutable and unimprovable mess”; ultimately too incurious or impatient to understand the politics.

There was an official ceasefire of sorts in 2002 but conflict is basically unresolved and ongoing. “[Survivors] didn’t have anything to help them address their loss […] The Congo is something of an outlier in this sense: Sierra Leone, Kosovo, East Timor, Rwanda, and the former Yugoslavia have all had tribunals to deal with the past.” It's a history with frayed loose ends. (Indeed Stearns has a new book, [The War That Doesn't Say Its Name).

Not easy to get through (because of the nature of the events described, but also because the story itself is quite complicated). But fascinating; if that's the right word.

If a faculty for abstract reasoning is a ‘superpower’ for mapping out and navigating the world, then: (i) why are humans apparently the only animal capable of reasoning; and (ii) why are humans so often terrible at reasoning?

The proposed answer is that the (evolutionary) function of reasoning is to provide socially defensible justifications for beliefs and behaviors. W're more interested in finding creative reasons which support our side than we are in finding reasons to switch side. On its face, this is a pessimistic theory: if reasoning is selfish, then why would anybody change their mind? Hence intractable disagreements on politics; unfalsifiable conspiracy theories; nobody gets ever smarter.

But there would be no point to giving justifications if anything goes: the flipside of giving reasons is being good at discerning good arguments from bad (since it's selfishly useful to avoid being duped). Similarly, even if it pays to be a lying bastard as long as others fall for your lies, it also pays to discern others' lies from honesty, and to change your tune if others catch you lying. Similarly, it pays to change your mind if your reasons for your beliefs are no longer socially justifiable.

So the takeaway isn't that humans are terrible at reasoning; it's more like “reasoning is more of a social than an individual phenomenon than you think”. Here's a way to make sense of the ‘my-side bias’ we exhibit in arguing about some unsettled question: just how an economy is more productive if different people specialise in different trades, so a conversation is more productive if its participants specialise in defending a particular point of view they're attached to and their identity is tied up with. Each individual participant might be irrationally stubborn, but (since accurate beliefs are never literally incoherent) the truth has an edge. Group deliberation is surprisingly effective!

At birth, the minds of human babies are only subtly different from the minds of newborn chimpanzees. We are friendlier, our attention is drawn to different things, and we have a capacity to learn and remember that outstrips the abilities of newborn chimpanzees. Yet when these subtle differences are exposed to culture-soaked human environments, they have enormous effects. They enable us to upload distinctively human ways of thinking from the social world around us.

This book is reacting to a bunch of nativist views in ‘high church’ evolutionary psychology, which claim that distinctively human patterns of thought are passed down through modern humans' genetic firmware. Our firmware supposedly includes: instincts for social learning, like the ability and disposition to imitate others during development (maybe through ‘mirror neurons’?); a basic instinct for language, as in Chomsky's universal grammar and ‘language acquisition device’; and a distinctive ‘mind reading’ module — some kind of innate ability to guess what other people are thinking and feeling based on how they act.

Then there's the school studying ‘cultural evolution’. The big idea here is that (modern) humans set themselves apart by their ability to accumulate culture over generations, through being able to communicate complex new ideas (language instinct) and wanting to learn them (adaptations for social learning). We have all these “copy what works” instincts (like through selectively imitating elder or high-prestige group members), so the learning rate can win out against the forgetting rate across generations, so successive generations can build up complex cultures and technologies, so humans learn to hunt and make fire and farm and eventually smart fridges. See: The Secret of Our Success and The WEIRDest People in the World. This book is reacting to those views, or rather extending them.

Heyes agrees that modern humans are set apart, at least significantly, through the ability to accumulate culture via social learning. But culture runs deeper, she claims. The suggestion is that the enabling ‘instincts’ of cultural evolution aren't part of the human genetic ‘firmware’, but are themselves learned and passed on through culture(!)

So we do not have hard-coded modules for theory of mind, or specific mechanisms of social/imitation learning, or a language faculty. Instead, we use more basic and domain-geneal faculties (like associative and operant learning) to boostrap distinctively human software.

For example, we are not (Heyes argues) born with a preinstalled faculty for knowing how to imitate the physical movements of other people. But adults like to imitate infants (e.g. doing an exaggerated smile if a baby smiles). So the infant, initially not knowing which motor commands correspond to smiling, learns an association between those commands and smiling, by watching adults. In turn infants are rewarded for imitating others, as in learning to walk, so they learn a disposition to imitate their own children, and the “cognitive gadget” of imitation is passed down as software, not firmware.

We know at least one very automatic and natural faculty which can't have a dedicated module encoded via our genes: reading. Written language happened more recently than 10,000 years ago, which isn't nearly enough time to install some kind of genetic “print reading” module. Instead, learning to read seems to borrow from relatively unspecialised brain real estate (the so-called ‘visual word form area’, which wasn't used for reading in the ancestral environment).

So it goes for language, forms of social learning, and even theory of mind, on Heyes' account. These are “gadgets” passed down by culture in much the same way “how to build a canoe” or “how to prepare berries” are passed down culturally.

Heyes writes in academese, so it's often easy to miss how big these arguments are. Since cultural evolution is really only a phenomenon of modern humanity, the implication is that distinctively human “instincts” are really only innovations of the last 10,000 years or less. If this is true then (for example) theory of mind — being able to infer and empathise with what someone is thinking in any kind of sophisticated way — is <10,000 year-old invention, which our hominid ancestors significantly lacked. That is wild!

In other words: modern humans are less like specialised computers (entirely firmware, not easily programmable); and more like general purpose computers with a minimal operating system out of the box (you need to install the programs yourself). Programmable computers have this really key property that you can store and manipulate programs like data (see: ‘homoiconicity’). And for Heyes, programs, data, and programmable computers relate in the same way that “cognitive gadgets”, culture, and human minds relate. Humans pass on programs (“gadgets”) just like anything else, and then use those programs to learn and use and eventually pass on more stuff.

Is this good or bad news? The case for bad news is that cognitive gadgets are more fragile against catastrophe than instincts. If distinctively human gadgets are innate, then the next generation would rebound. If they are maintained by culture alone, they can vanish.

The case for good (?) news is that human minds are more pliable to big social/technological changes if more of our minds are software and less firmware than we thought. Heyes:

[T]he cognitive instinct view, famously implies that "our modern skulls house a Stone Age mind" (Cosmides and Tooby, 1997); in other words, the cognitive processes with which we tackle contemporary life were shaped by genetic evolution to meet the needs of small, nomadic bands of people, who devoted most of their energy to digging up plants and hunting animals. In contrast, cultural evolutionary psychology, the cognitive gadgets theory, suggests that distinctively human cognitive mechanisms are light on their feet, constantly changing to meet the demands of new social and physical environments. If that is correct, we need not fear that our minds will be stretched too far by living conditions that depart ever farther from those of hunter-gatherer societies. On the cognitive gadgets view, rather than taxing an outdated mind, new technologies—social media, robotics, virtual reality—merely provide the stimulus for further cultural evolution of the human mind.

Bill Gates: “One country alone has lifted 500 million people out of abject poverty […] To have done this essentially in one generation is an unbelievable accomplishment and is unique in the history of the world.”

Fleet Foxes are very unselfconscious and unironic; their sound is not innovative; and their lyrics are not pointed, political, or current. They are “2010s hipster starter kit” music. Their former drummer went solo as the irony-drenched Father John Misty, making them the Disney Channel to Sabrina Carpenter's musical career. Fleet Foxes are uncool by consensus.

But what sad reasons not to love this band!