Roastidio.us User Picks KjQKkQVg

Legibility is Ruining You

Sat, 04 Apr 2026 16:43:31 +0000

Legibility is Ruining You

I am incredibly grateful that Sean Goedecke published his beautifully written article Seeing like a Software Company. In it, you will find a wonderfully balanced presentation of why companies often choose legible practices over more efficient, illegible ones. Sean very clearly details the kind of (in my words) doublethink that goes on at companies involving these very needed, but often hated, illegible processes. If you are hoping for that sort of balanced, nuanced take, please read his article instead. Sean has freed me from the burden of needing to write that way myself.

The Values You Truly Hold

Deep down on reflection, you don't care about OKRs, you don't care about code review, you don't care about coding standards, you don't care about lint rules, you don't care about style guides, you don't care about deadlines, you don't care about job titles, you don't care about standup, you don't care about code coverage, you don't care about any of these processes.

Your values are complicated. They require context and consideration. They aren't simple mechanical rules that can be captured by any metric, any process, or any methodology. You may believe these practices and tools we've adopted are pragmatically correct decisions to achieve your values. But they aren't the values themselves. And yet, they are the "values" that the "company" holds, or, more accurately, they are the values the company wants you to hold.

The Coercive Force of Legibility

Processes adopted by a company aren't about the end result of the work. Their stated goals and their actual goals are always distinct. An initiative related to applying coding standards isn't about making the quality of the code better. It is about changing the behavior of the engineers writing that code. By making code uniform, it reinforces some of the legibility myths (my word) Sean Goedecke says that companies rely on.

Any engineers with the same job title perform roughly the same.

Engineers can be shuffled and reorganized without substantial loss of productivity.

A team will maintain the same level of productivity over time if it has the same number of engineers.

- Sean Goedecke: Seeing like a Software Company

These are all false. Everyone knows they are false. It is easy to think of these as idealized assumptions, but they aren't. You can't find physicists out in a field trying to shove a cow in an orb to make their equations work out better, but software engineering directors are more than happy to do so to software engineers.

Legible assumptions are not merely assumptions. They are coupled with practices trying to make their assumptions reality. This is first by placing barriers (processes), but it goes much deeper than that. Barriers can be circumvented. Gatekeepers can be ignored. The modern software enterprise has realized that the only true way to make its legible assumptions true is to change the very nature of those creating the software. To transform them from an illegible force to a legible one.

Value Capture

Every software process does the same thing: it converts a rich, multifaceted value into a single metric. "Are our customers happy with our product and support?" is not the same question as "how high is our NPS"? "Is this team performing well?" is not the same as "What is this team's 'velocity'?". "Are we working on the right stuff?" is not the same thing as "Do we have a prioritized backlog?". Yet we constantly mistake these questions for each other. Not only that, we don't even realize that these are conflations. Why? Because we are rewarded for fixing the latter and not the former.

The excuse always given for this disconnect is that the former questions don't work at "scale". But what precisely does that mean? If you have too many people, each divided into small groups (teams), they can't each ask these questions? No, these rich questions are actually quite easy to parallelize. What they don't do well is compose and reduce so that those at the top can understand them. Don't confuse something not working at "scale" with something that doesn't make an executive feel warm and fuzzy.

But the real reason for this conflation is not simply that it makes executives feel good, but that these metrics are an incredibly powerful means of changing behavior and values. In other words, they are a way of making the world fit the assumptions those at the top have already made.

Value capture occurs when an agent's values are rich and subtle; they enter a social environment that presents simplified — typically quantified — versions of those values; and those simplified articulations come to dominate their practical reasoning.

- Value Capture by C. Thi Nguyen

By making these metrics, the company can change how we behave. They can make our behavior properly legible.

Thin Rules, Thin Behavior

The sad fact is, as an industry, we long ago stopped only letting executives enforce rules on us; we enforce them on each other. We create rule after rule that we will make it so that "good software" is the necessary outcome. We all bring our own flavors of these rules. Our own dogmas we hold to, our own tastes we apply to others. We shame other engineers who do not follow our rules. The rules are the values.

They constantly try to escape

From the darkness outside and within

By dreaming of systems so perfect that no one will need to be good.

- T.S. Eliot

In Rules: A Short History of What We Live By, Lorraine Daston distinguishes between thick and thin rules. Thin rules are the algorithmic, mechanistic rules we in software favor; they are rules that can be obeyed and enforced unambiguously. Thick rules, on the other hand, are rules of thumb; they are guiding principles. Daston shows how there has been a historical progression from thick to thin rules. That today, we almost always think of rules as this thin, algorithmic process.

By following these thin rules, we not only produce worse software, we also steer our behavior to one that isn't sensitive to the real things we value. "Did the code pass code review?" is the wrong question. How much bad code have you seen pass code review? But the very fact that "passes code review" is the thin rule we have adopted ends our conversation. We can try to revise our "code review checklist" to "closer" approximate our true values, but this is a losing battle. What we care about can't be captured by our thin rules.

You

Legible processes are impossible to escape if you work for a company. Maybe they are necessary. Maybe companies of a given size must adopt them or fail. You don't have a choice but to follow some of these processes if you want to keep a job. But what is in your control is whether you let yourself be value captured by these metrics and processes.

You can choose to embrace the legible. As an IC, you can choose to value your work on the basis of these legible processes. You can set aside your own personal values about code and map your behavior to what the company cares about. As a manager, you can have your team follow the legible processes to create software that meets all "success" criteria, but still sucks.

Or you can choose to truly care about something. You can choose to push back against the legible processes that are trying to make you an abstraction. Not because you are selfish and stubborn. Not because you think you know better than others. Not because you don't want to follow rules. But because you value something that can't be captured in simple metrics. Because you know the company will benefit from these things. As Sean tells us, illegible work is essential. So why not be essential?

Hasan, Me & Racial Ambiguity | Çaki's Series Of Tangents

Sat, 04 Apr 2026 02:35:46 +0000

Hasan, Me & Racial Ambiguity

03 Apr, 2026

Case Study 1: Me

Before we start let's clarify my background, I am of Turkish Cypriot descent, a third generation product of the diaspora following the attempted ethnic cleansing of Turkish Cypriots from the island. My Dad's side of the family are often mistaken for being Indian or South Asian as they are darker than other Turks, and we think this is where our surname comes from, as it is not traditionally Turkish.

My mum's side of the family (also Turkish Cypriot) have paler skin and look more Eastern Mediterranean, which in some parts of the UK had it's own issues. My mum has been mistaken for Moroccan recently and faced her fair share of discrimination early on in life, predominantly white British schools in the 70's & 80's may have been worse for my Dad but they are pretty hostile to anyone they deem "foreign".

On my end, I've been told I look West Asian mostly, although some people have mistaken me for Indian in the past, and on the first day of my current job a child asked "who the Mexican was". It's hard to tell if people assume you're white because whiteness is largely unspoken, although it has definitely happened to me in the past, particularly online, where the subtler differences in skin tones feel more like a product of lighting and colour grading.

What does this mean?

Being a bit more ambiguous racially obviously has it's perks but it also has drawbacks. I think the best summary I can give is the time a friend of mine got sloshed and called me a racial slur (beginning with p) I'm not dark enough to say.

It's weird.

I experience moments and fragments of that oppression (a fair amount in my lifetime because of the shittiness of my surroundings) but it's difficult because I'm probably not getting the whole thing. There is an unquantifiable collection of moments that could have been worse had it been more obvious.

Because of that it's hard to feel that shared connection. It's a privilege obviously, to be in a position where it's not as obvious, but it comes at a cost.

My own experience growing up, (the accents made at my uncles, the constant othering and the endless terrorist "jokes") is why I reject the idea of being white, but the fact it's even in question means I probably could have experienced more no? It's hard to put a finger on what that really means, and even harder when examining it and embracing it can lead to opposition from people who insist that they know more about your race than you do.

It's also worth mentioning just how exclusionary British racism is. The kind of claims that Trump makes about Haitian and Hispanics immigrants, Farage makes about Eastern European ones, espousing that they eat Swans and fearmongering about a "Romanian Crime Wave". Indeed, British racism is so intrinsically tied to a general jingoism in favour of "the west" that one could genuinely argue eastern Europeans are white passing in the UK rather than just white.

This creates a large cultural disconnect between American and British racism that the online world has made extremely apparent to me. It creates an odd situation in which I have to basically spill out about my own (often unpleasant to say the least) encounters with race in the UK in order to sort of "prove" that I fit the description.

Case Study 2: Lineker

Gary Lineker is an Irish-English white man who played for Barcelona and England in the 80's. He was tan at a time when racism was still effecting Italians and other Mediterranean white people, and on his podcast he spoke on the colourism he faced at that time, highlighting that this is a long-standing issue and that if even he as a white man received such abuse it must be awful for the actually marginalised people trying to play for the team. There is obviously nothing wrong with this.

It was lied about constantly though, the tabloids removed all context from the headline as you would expect, and so it became "Gary Lineker opens up about racism" which in turn became jokes about he identified as black etc etc. The same ambiguity that led to him being abused for "darkish" skin as a kid in school was also a bludgeon to silence him for daring to talk about it.

Case Study 3: Hasan

Hasan Piker is a Turkish man and streamer "hasanabi" born to two parents who look Eastern Mediterranean/ West Asian. Unlike me, Hasan considers himself white.

Hasan grew up in Turkey, (where he's obviously a member of the dominant ethnic group) and the US, which has a very different perception of race altogether.

The presence of slavery and the genocide of Native Americans always meant race in the US had much clearer visual lines. The Irish diaspora may have experienced it's own discrimination in the US, but there were no "no blacks no dogs no Irish" signs in the 60's across it's windows. If you looked white enough on a first pass, you're white. If you don't, you're coloured.

Indeed, in an interview with FD Signifier on the topic, Hasan says that he knew he was white when people were comfortable being racist around him, as if he were part of a club. In other words, in most contexts Hasan is white passing.

There is a exception though, whenever hasan attempts to talk about the Muslim world or indeed any brown or West Asian country, he is immediately an "islamic fundamentalist", "barbaric" and a "terrorist". FD actually says this is part of why Hasan isn't white, to paraphrase slightly: "to be white means all slurs are funny, if there's any (racial) slur that might make you slap a motherfucker in the face then you aren't white".

But unlike me, Hasan has the white experience in most situations, the kind of looks I receive when I walk into a pub aren't the kind Hasan has received. The neighbours of white people in the States have always been visually distinct, Hispanic, black, native, and so race is drawn cleanly (or not so cleanly) along those lines. In the states the ambiguous are assumed white, in the UK they are still the other.

This benefits Hasan rhetorically, as he speaks to a mostly American audience, the fact he passes as white means his race does not hamper how likely people are to click on his content. It also means that he's able to more accurately relate to the disenfranchised young white men in America, and his conventionally masculine appearance also helps him pull them out of the world of the manosphere.

The truth about racial ambiguity is also largely the truth about race, it is almost entirely about the perception of the onlooker. Lineker probably wouldn't have been perceived in the same way had he grown up a few decades later, and yet nothing about him changed.

Hasan may have faced more direct opposition with issues outside of the Muslim world had he existed in a nation where his own racial differences were considered more important.

Likewise, if I had grown up in an area less exclusionary than the uniquely insane world of White British suburbia I might have a different perspective on my own race. What is "white" or "white passing" is not just different historically, but geographically. Whiteness is a social in group after all, not a biological category.

Race exists in the minds of our beholders.

Your sign-up form is a weapon | Bytemash

Thu, 02 Apr 2026 08:29:04 +0000

A couple of weeks ago, we noticed something odd on Suga. New users were signing up but not doing anything, they weren’t creating an org, a project, or a deployment, they just left an account sitting there. Most new users interact with the product pretty quickly, and we report on activity stats to try and understand blockers, so even a small spike in completely inactive accounts stood out.

Then we looked at the names of the new users and they were entries like PfVQXvYTXjwSbEeJBjXYy and xXzMafkbPLjOaGgDaOGZjLx.

We checked Resend (our email service) and could see welcome emails going out and being delivered to these accounts. They were real email addresses, with garbage names… something was off.

What is subscription bombing?

Subscription bombing is an attack where someone uses bots to sign up a victim’s email address across hundreds or thousands of websites. The goal isn’t to access those accounts on those websites, it’s to flood the victim’s inbox with so much noise that they can’t find the emails that actually matter.

While the victim is drowning in “Welcome to SaaS Product!” and “Verify your email for Newsletter You Never Asked For”, the attacker is doing something else. They’re resetting the victim’s bank password, making purchases on their accounts, or signing up for credit cards in their name. The real security alerts and confirmation emails get buried by the noise.

The people running these attacks are stealing money and impersonating real people or businesses. Every sign-up form on the internet that lets you enter any email address without verification is a tool they can exploit.

How we spotted it

It started on March 12 with a single unusual sign-up, then over the next two days we saw another 2-3 per day, which was low enough to be noise. Honestly, our first thought was that someone was pen testing our service. That’s pretty common, something we’ve experience with other products and, when it comes with responsible disclosures, something we appreciate, so it didn’t ring any alarm bells.

By March 14 we had 6 sign-ups matching the pattern, and we noticed something else in PostHog. There was unusually high page views and interactions on the forgot password page. The combination of activities was enough to make us take a closer look.

The pattern

Each bot would sign up with a real victim’s email address and a garbage name. Then, within about 60 seconds, it would navigate to the forgot password page and request a reset. That meant each victim received three emails from us in under a minute:

Verify your email address
Welcome to Suga
Reset your password

Three emails they never asked for, from a product they may never have heard of. We were just one of potentially hundreds of sites being hit at the same time.

The bots were also hitting the forgot password page with random email addresses, presumably trying to trigger reset emails for victims who were already a user.

We reviewed one session in detail and the typing behaviour was interesting. The bot was entering values into form fields painfully slowly, one character at a time with up to a second between keystrokes. The gaps had randomness to them, but it was too random. Humans type in bursts, most people type a few characters quickly, pause, then type again. This was a flat distribution of delays trying to look human and failing. The timing between page navigations had the same quality of being randomised, but uniformly so. Enough variation to dodge simple bot detection, not enough to actually pass for a real person.

Designed to be invisible

This is what makes subscription bombing so different from the attacks most people think of. There was no spike, or flood of requests, or dramatic server load. It was just 1-2 sign-ups per hour and that’s it.

The requests came from all over (India, Brazil, Romania, the US, Vietnam, Türkiye) which isn’t unusual until you compare it to typical traffic. Our real users typically navigate from specific countries with a reasonable correlation to the daytime hours of that country. The bot traffic had zero correlation between country and time of day, and that mismatch is what stood out.

Rate limiting does nothing here, since you can’t really rate-limit against one request per hour. The whole point of this attack is to stay below the threshold, that’s one of the reasons I find this attack type so interesting.

The damage isn’t to you

Subscription bombing barely hurts the site owner, our email sender reputation wasn’t affected (the volume was too low). If we hadn’t been tracking user activation closely, we might not have noticed. Unfortunately, the damage is almost entirely to the people on the other end.

Picture waking up to 200+ emails from services you’ve never heard of, you start deleting them, but they keep coming. Somewhere in that pile of garbage is a notification that matters, like someone changing your banking email address, resetting your password or ordering a new credit card in your name.

The reason this attack works at all is that thousands of websites (newsletters, SaaS products, forums, e-commerce stores) let anyone enter any email address and immediately start sending emails to it.

If your sign-up form sends email to an unverified address, your form is part of this. And because the damage falls on the victim, not the site owner, I suspect most people treat it as low priority to fix, which is wrong. It pollutes your user data and it makes your service an accomplice in harassing real people.

What we did

Firewall rules

Our first move was tightening bot detection on our front-end hosting provider’s firewall. This cut the volume roughly in half, but didn’t eliminate it. It was also tricky to configure because we have legitimate non-browser traffic hitting our API (webhooks, service-to-service calls) that needed to keep working.

We let it run for about 6 hours. In that time, two more sign-ups slipped through (one at the 5-hour mark, another at 6). The bots were getting past the verification step so we needed something better.

Cloudflare Turnstile

Cloudflare Turnstile was the fix, it’s a CAPTCHA alternative that doesn’t ask your users to solve puzzles. It analyses browser signals in the background and can be configured to only present a visible challenge when something looks off. You can set it up, so that for real users it’s invisible.

We use Better Auth for authentication, and it has a built-in CAPTCHA plugin with Turnstile support. That made the implementation dead simple, just add the plugin server-side, drop the Turnstile component onto the auth pages like sign-up, sign-in and forgot password with @marsidev/react-turnstile, and pass the token through to the server via a header. The submit button stays disabled until Turnstile provides a valid token.

After deploying Turnstile, the bot sign-ups stopped, so I want to specifically call out both Better Auth and Cloudflare Turnstile here. The combination made this a same-day fix. If your auth solution doesn’t support CAPTCHA out of the box, Turnstile’s API is still easy to integrate directly, but having it baked in saved us time.

Limiting email to verified users only

Even with Turnstile in place, we wanted to limit the damage if this happens again. We updated our email service code so that a user receives exactly one email from us (the verification email) until they click the link and prove they own the address. No welcome email, no product updates, nothing else until verification. We should have had it this way from the start, which is a mistake I regret.

If a bot creates an account with someone else’s email, the victim gets one email, if they ignore it that’s the end of it. The welcome email and everything after it only fires once the user verifies.

(For social auth sign-ups through Google or GitHub, we send the welcome email right away since those providers have already verified their email address.)

Wrapping up

We could have looked at 1-2 sign-ups per hour and shrugged, since the business impact to us was basically zero. But those were real people’s email addresses, and our service was being used against them.

I’d like to apologise to the people whose addresses were used, but reaching out would just be one more unwanted email in an already overloaded inbox. The best we could do was make it stop and make sure we can’t be used this way again.

We’ve also added better reporting to spot patterns like this sooner next time. Overall I think our response was reasonable and fairly rapid, but we should have had these mitigations in place from the start, lesson learned.

Why Florence Started the Renaissance - by Tomas Pueyo

Thu, 26 Mar 2026 00:51:51 +0000

One of my most successful articles now has a YouTube version. If you want to see why warm countries are poorer, go watch it!

Now today’s article, which I had a lot of fun writing, full of crazy facts. Enjoy!

After 1300 years as the largest dome in the world, Rome’s Pantheon was replaced by Florence’s dome in Santa Maria del Fiore. To this day, it remains the biggest masonry dome ever built.1 What?!

In a span of 100 years, the same city would birth an endless list of history-making figures: Michelangelo, Leonardo da Vinci, Brunelleschi, Botticelli, Raphael, the Medici, Machiavelli… What’s going on?!

What was in the water? I’d like to know, to brew it again. Florence or Silicon Valley can’t have been pure serendipity. How do we replicate them?

To answer that, the best place to start is a cathedral that had spent 100 years with a massive hole in the roof that nobody could plug, much before Lorenzo de Medici, Leonardo, or Michelangelo were born.

Brunelleschi’s Duomo

The architect that originally designed the cathedral died, and he left no solution on how to build the dome.

After a century with that hole (!) came a goldsmith (!!), Brunelleschi, who built the biggest dome the world had ever seen! And he did it without the cement used for the Roman pantheon, as its recipe had been lost in time!

The dome of the Roman pantheon was made of pozzolanic cement, a recipe that was lost for over 1000 years

Even harder, since the rest of the cathedral was already built when he arrived, he couldn’t change the shape of its walls, reinforce them, buttress them, or anything like that. He had to work with what he had! And he did it without scaffolding! And he did it without ribs!2 This is crazy.

Not only is this beautiful, but can you see ribs, lines of stone at the angles to support the structure? No! This is a continuous painting uninterrupted by ribs!

The key challenge was to prevent the dome from falling. The first thing there was to adopt the insight of Gothic architecture that taller arches are stronger than circular ones. So the dome is not a hemisphere, it’s a bit more vertical.

I placed a semicircle there. Notice how the dome is taller than that. Source.

The second was to make it light by emptying it. The structural force is carried by the inner dome (red), and then there’s a space and an outer shell (green).

In green and red the two roofs. The dotted lines and arrows show the structure, how it was inspired in Gothic for the angle: Instead of having a single semicircle (drawn here in dotted lines), there are two semicircles (also in dotted lines), one for each side of the dome. The arrows show the distance between each semicircle and its center. Source.

You can see the space between the roofs here:

Source

Even then, it would have fallen.

Source, cleaned up with nanobanana

Brunelleschi needed a way to pull the top outward and push the bottom inward. He solved it with stone and wood rings around the dome!

These rail-like structures are made of sandstone and wood, connected with iron. The craziness is that, to prevent the dome from bending, they convert the tension into compression, as the ring is compressed against itself by the dome’s pulling. I like to think of these rails as horizontal, infinite arches. Source.

The other thing Brunelleschi had to figure out was how to build this thing without a wooden scaffolding inside (it would have required too much wood!). He solved that by building the entire structure ring by ring, using herringbone bricks, crossing the bricks so that they would support each other.

From this, cleaned up with nanobanana, although from the pics I’m not sure what this video depicts is the actual herringbone structure of the dome, since I can’t see how that would stand, it doesn’t correspond to the pic above (you can see the actual herringbone bricks on the earlier photo of the passageway between domes), nor to the Roman herringbone pic below. But I’m not sure.

OK hold on. What’s going on here? Why, after 1300 years, out of nowhere, is a goldsmith that doesn’t even have the materials necessary for replicating the Ancient Rome Pantheon able to build a bigger one?! That we’ve never replicated?!

How did you do it, Filippo? Fun fact: He was originally a goldsmith, and he got one of the first patents in the world, for a machine to hoist goods onto barges on Florence’s Arno River.

Leonardo’s Roman Proportions

Brunelleschi lived in Florence, a city with plenty of Ancient Rome ruins. I think it’s hard for us to understand what living among these vestiges of a better time felt like.

Aqueduct in Ruins, Hubert Robert, 1700s. I saw a bunch of pictures like these I believe it was in Vienna’s Kunsthistorisches Museum, or in Munich’s Residenz. I can’t really find them. It’s a series of two-three dozens of paintings of life around Roman ruins, centuries after the fall of the Empire.

From Petrarch, a Tuscan from the neighboring city of Arezzo3 who visited Rome in the mid-1300s:

[Rome is a] broken city, the remnants of the ruins lay before our eyes. [...] Who can doubt that Rome would rise again instantly if she began to know herself?

Every day, you’d be reminded that your civilization is inferior to the one that came before.

A Hermit Praying in the Ruins of a Roman Temple, Hubert Robert, about 1760

Brunelleschi, inspired by these ruins and by some people before him like Petrarch, who had visited Rome, decided to make a trip to Rome too. Ruins were everywhere. He spent two years there, studying its remaining architecture. He studied the Pantheon to replicate its dome in Florence, and got the idea for the herringbone bricks by observing brickwork there.

Opus spicatum paving in Trajan's Market, Rome.

He didn’t go alone; he was accompanied by the famed sculptor, Donatello, who was inspired to sculpt a David, the first freestanding nude male sculpture since antiquity. Together, they excavated buried structures and measured monuments like the Pantheon, the Baths of Caracalla, Roman basilicas…

Around that same time, another Florentine, Bracciolini, rediscovered4 Vitruvius’s De Architectura, the only architectural treatise that has survived from antiquity to this day.

The Vitruvian Man, drawn by Leonardo da Vinci. The idea was that there are ideal proportions, some of which we can see in the human body, and they give you a hint of the ideal proportions in other fields like architecture.

Other Florentines like Michelangelo, Raphael5, and Leonardo da Vinci traveled to Rome to ~~form the ninja turtles squad~~ study it.

Between De Architectura and visits to Rome, architects and artists started noticing that Roman buildings used modular, proportional systems and strict rules of geometry: harmonious numerical ratios (1:1, 1:2, 2:3, etc.), room height related to width, temple column spacing based on column diameter… These ideas gave Renaissance architects something medieval builders lacked: a theoretical mathematical framework for beauty.

Brunelleschi came back from the trip with a firm idea of what the dome should look like, and all the other Florentine visitors came back with specific ideas of proportions and classical elements.

So why did the Renaissance happen in Florence? Well, it couldn't have happened much farther from Rome. Think of the experience of these Florentines, growing up among substantial Florentine ruins from the Ancient Roman Empire, and close enough to visit Rome and witness the massive beauty that had been lost. The farther you were from Rome, the fewer the ruins, the harder it was to get to Rome, and the less inspiration Rome would have provided.

This probably explains why the Renaissance had to happen close to Rome, but not specifically in Florence. Why not Siena, Pisa, Naples, Milan, or even Rome? Why then, and not 200 or 300 years earlier or later?

Machiavellian City-States

In the Dark Ages, as the Roman Empire fell and law and order withdrew, cities decayed and many disappeared. But not so in Italy. Cities like Milan, Florence, Verona, Bologna, Ravenna, Pisa, Venice, Genoa, and Siena remained populated. By the 11th century, they had merchant classes, bishops, courts, guilds, militias… They were already institutionally capable of governing themselves. This independence of communities happened everywhere in the former Roman Empire, but it was especially true in the Holy Roman Empire, modern-day Germany and Italy.

This is Italy in 1500. Chaos!

I explored why this happened in Why Were Germany and Italy the Last European Countries to Unify? The short answer is that Italy was the battleground between two dimensions of power: the temporal and the spiritual.

This was Europe around 1200 AD:

For centuries since the creation of the Holy Roman Empire in the 10th Century, the Emperor and the Pope fought for power and influence, like who would appoint the powerful bishops. Around 1200, the Holy Roman Emperor Frederick II had just united the northern part from his father with the Kingdom of Sicily to the south from his mother. The Pope logically freaked out when he saw his Papal States surrounded, and stirred up revolts everywhere he could in Italy.

Crucially, Italy is separated from Germany by the mighty Alps, making it very hard for the Holy Roman Emperor (HRE) to have strong influence to the south.

And Italy is very mountainous, with lots of secluded valleys that birthed strong, independent cities. This meant that, in practice, local cities had been quite autonomous for a very long time, even if they were nominatively under the HRE.

So when the Pope stirred revolt all over Italy, different cities decided to support different sides. This is what’s called the war of the Guelphs and Ghibellines.

Depiction of a 14th C. fight (1369?) between the militias of the Guelph and Ghibelline factions in the Italian commune of Bologna, from the Croniche of Giovanni Sercambi of Lucca.

Notably, the closer you were to one of these big powers, the less likely you were to support it, because it’s better to have a distant lord than a neighboring one. This created a patchwork of alliances and counteralliances, of wars and betrayals. Over the centuries, the power of the HRE dwindled in the region, and the emperor became just a distant influence.

This is the context in which the Florentine Machiavelli thrives and writes his famous work The Prince in the early 1500s.6

So all this is why the Northern Italian region had many independent city-states:

The cities there remained inhabited since Roman times.
The area has many fertile valleys among mountains, allowing for many urban nuclei to sprout.
The region was separated from its rulers by the Alps, which meant little oversight.
This is at the border region between the HRE and the Papal States, so they became a proxy war battleground that exacerbated their differences and undermined the oversight of both on the region

Independent states were crucial to explore new styles locally and to push for their own architectural styles to differentiate from other competing city-states. But this doesn’t tell us why they were so rich, and you needed a lot of money to build such huge cathedrals.

Medici Wealth

At the time, Italy’s population was rebounding from the Black Death faster than in many other European countries.

And these people were rich! In the 1300s, Northern Italy was already the richest region in Europe.

GDP per capita estimates for 1300 AD. Source.

One of the main reasons was urbanization. These were the main European cities in the early 1500s. Notice how many are in northern Italy!

Source

We already explained why: They were already quite populated in Roman times, and they have some of the best plains in the Mediterranean, especially in the Po Valley.

A big population that could grow fast and was already urban meant a bigger share of the population lived in cities.

Sources: van Zanden (2009), The Long Road to the Industrial Revolution. Jan de Vries (1984), European Urbanization 1500–1800. Paul Bairoch, Jean Batou, and Pierre Chèvre (1988), The Population of European Cities from 800 to 1850. Via ChatGPT.

Of course, cities are more productive than rural areas due to their network effects. They become marketplaces, develop industries, invest in infrastructure, build industrial clusters…

Northern Italy also had another huge benefit: It was in the middle of global trade networks.

The richest regions at the time were Flanders, the Holy Roman Empire in general, the Byzantines, and the Muslims,7 and Northern Italy was in the middle of them all. This is why Venice and Genoa grew so rich as maritime trade republics.

And within Northern Italy, Florence was very well positioned.

Roman roads are relevant because they were still in use, and also because they followed the most natural paths from one city to another, so the fact that Florence was on one of the best paths to cross the Apennines from Rome wouldn’t have changed by the Middle Ages.

If you wanted to move between the HRE and Rome, the best path went through Florence, which is a natural crossroads because Florence and Bologna flank the Apennine mountains.

So Florence, like only a handful of other Italian cities, had a unique confluence of assets:

It was in a populated region, because it has a good climate and great agricultural plains
This was in the richest region in the world, because it was at the crossroads connecting all the main kingdoms of the time
The region also contained many cities that had remained inhabited since Roman times, and Florence was one of the big, urbanized cities in the region
Situated between the two regional powers of the HRE and the Papal States, these cities had more independence than most
It was at a crossroads for internal Italian trade, making it uniquely rich as a marketplace
Close to Rome, to receive strong architectural influence from the Roman Empire

Still, it could have been Siena, Bologna, Parma, maybe Genoa or Venice. And sure enough, all of them have beautiful architecture. But something set Florence apart.

Money.

The Origins of Florentine Wealth

The best way to know who ruled back then is to look at who funded the biggest monuments. So who funded the cathedral, Santa Maria del Fiore?

One tip is in the name: Santa Maria refers to Virgin Mary, and echoes the importance of religion. But fiore? These are flowers—like Firenze. This refers to the city: The cathedral was meant to represent both the church and the state. This was not a purely religious endeavor. In fact, it was anything but. It was the commune of Florence that financed it. And where did the money come from? The Wool Guild, Arte della Lana.

By the 1200s, the wool trade had become one of the city’s largest industries, among other things precisely because Florence was such a big city:8 Wool must be sheared, sorted, scoured, carded, drawn, spun, wound, warped, woven, fulled, sheared, pressed, dyed, sewn… There are many steps, but crucially, each step can be done by one single person or a few, with pretty basic machinery. So people across the city worked on different steps in their homes. The bigger the city, the more workers could be dedicated to each step, the more competition and learning between them, the more volume could be produced, and the bigger the industry. The bigger it is, the more it specializes, the better its products compared to the competition, the better the reputation, and you end up dominating a continent-wide market.

This industry became so big that local wool was not enough to feed it, so Florentines started buying wool abroad (in Spain, England, Flanders…), and finished it into high-quality cloth. They organized themselves into guilds, which helped them work together to promote the industry’s quality, its trade, its financing…

Financing! When you buy wool and sell cloth abroad, you need currency exchange. And when there are so many steps to the process, from so many workers, you must pay them a salary before you sell the cloth or garment, so you need a lot of working capital. So the Florentines also developed a very strong financial guild.

Some of the Florentine guilds

The financial guild minted the gold florin, which soon became a currency used across Europe thanks to its distribution through wool commerce and banking, and to its reliability, as ensured by the guild.

This gave the Florentines seigniorage privileges

The power—and money—of the commune was concentrated in the hands of these guilds, and of course legislation favored them. Imagine what that did to the competition: Across most of Europe, industrialists, bankers, and merchants had to withstand the logic of Church and nobles. But Florence could do regulatory arbitrage, optimizing everything to support its trade; another advantage to pull forward.

Then the city had a lucky strike. Remember the Guelphs and Ghibellines? Florence supported the Guelphs (the pope), while its rival Siena supported the Ghibellines (HRE),9 and it turns out the pope won. Siena was stripped of its banking and tax collection monopolies, which went to Florentine families. Notably, the Medici created a bank that grew strongly throughout that time, was on the right side of the war, and was given the management of the Papal Treasury. Their good relationship also gave them a monopoly of alum mining, a product crucial for dyes that could only be sourced in mines near Rome.10 This is also the time when silver started flooding Europe, so banking became extremely profitable. Eventually, the Medici would take over the politics of Florence and convert it into a duchy.

Florence used this newly-found power to strengthen its position across Tuscany and conquered its neighbors one by one. When it took over Pisa, it acquired a port. When it took over Siena, it eliminated its main rival.

So we can add a few factors to our reasoning of how Florence became the cradle of Renaissance:

Thanks to its sizable population and perfect position at a crossroads within Northern Italy, Florence became wealthy in the pretty typical network effect of cities we’ve already seen elsewhere: first an industry developed (wool), which begat new industries (wool processing into cloth, finance). This made the city rich.
It was on the right side of a war, so it was showered with spoils, most notably lots of financial power and control over valuable commodities (alum was crucial for making colorful cloth).

But why push for a new art style? And why this one?

Florence Style

Power was conveyed through architecture. The Arte della Lana guild wanted to show its power in the city, and to other cities. Remember, there are dozens of city-states in Northern Italy, the competition was brutal! They had to stand out, establish themselves as the most successful, so they could have status and gain more business. That’s why they invested in Santa Maria del Fiore. But that cathedral is not yet Renaissance! It starts as a mix of Gothic and local Tuscan style. So what is the Renaissance style, and why does it appear here?

Although the Milanese cathedral is called Duomo, it has no dome (duomo comes from domus, house, for house of God). Also, although it’s mainly a Gothic cathedral, it took six centuries to complete so it mixed many styles. The Duomo is taller than the façade of St Peter’s Basilica, although St Peter’s dome is taller. I couldn’t convey the height of both the façade and the dome because of the perspective (pictures taken at ground level), so I had to choose which one to convey, and I thought the façade was the right one.

As we saw previously, the impetus behind Gothic was to go as high as possible, to reach the heavens. This pushed the boundaries of architectural technology of the time: verticality, pointed arches, rib vaults, flying buttresses, detailed decoration for mysticism, and stained glass were the methods, and I think the goal was clearly achieved.

In the Renaissance, the idea was to go back to the wisdom of the more powerful ancient Roman Empire. Studying it, architects realized they followed some rules, and they decided to decode them and apply them. They decided that space should be graspable, measured, balanced, and ordered as a coherent whole. They did that by going back to columns, round arch, domes, volumes, horizontality, visual legibility, proportions, and a return to hard-coded classicism.

The love for proportion and precision could also be seen in small details like these:

Façade of Santa Maria Novella, Florence

Lines, squares, circles, crosses, mathematical curves… These textures are beautiful.

Detail of Santa Maria Novella

Of course, Rome and the popes loved the new architectural style based on their own city, so they funded and promoted it. They realized that Rome was a pile of ruins, but if they could rebuild it and make it even more beautiful, their power would radiate across Christendom. This is why Vatican City’s St Peter’s Basilica is of the Renaissance style, for example, and why the Renaissance became huge in general.

A Change in the Reason behind Architectural Innovation

OK here’s my ignorant and probably unpopular opinion: Renaissance churches are beautiful but… underwhelming?

With Gothic, society had a clear goal: Convey their love of God by reaching upward and pushing the technological boundaries to make it happen.

The great thing that Renaissance does is rediscover domes, which are objectively awesome. It also recovered the pendentives (which were Eastern Roman, not from Rome…), and made proportions explicit. But does that mean you have to adopt all the other Roman stuff, too? Why go back to columns and rounded arches, when we’ve seen they’re objectively inferior, since they can only allow smaller spaces? Why go for horizontality, an architectural feeling that was already so pervasive in cities because houses were not tall? Why pretend an obsession with order that your predecessor didn’t have, when in fact Gothic architecture had a similar order that you just couldn’t read?

Source

It feels like the Renaissance was trying too hard to define its path in opposition to what came before.

“Gothic” litearlly means from the Goths—the Franks and the Germans on the other side of the Alps. That style was born in France and spread to Germany. It makes sense that Florence, allied to Rome and against the German HRE, would define itself in opposition to that by getting inspiration from ancient Roman architecture. Of course, a city that is managed by merchants wants to define itself in opposition to pure religious piety too. But when you define yourself by opposition to others, how strong are your values?

It’s a bit like how Swiss handwatches tried hard to become precise and thin, and once the Japanese figured out how to do that better, most Swiss watchmakers went bankrupt and the few that recovered (and soared) did so not because their watches were better (not more precise, nor thinner), but because their cost and elaborate shapes became symbols of status.

This is the first time in all the architectural styles we’ve observed that the innovations are not done to improve, but just to be different.

Renaissance Outside of Churches

Renaissance really shines outside of churches for me. Before, it seems like most architecture of beauty was concentrated into churches, but now it appears everywhere. Compare Florence’s Palazzo Vecchio, built around 1300 just before the Renaissance, with the Medici-Riccardi, built around 1450, during the explosion of Renaissance in Florence:

The Palazzo Vecchio is basically a fortress!11 A small unassuming door without many indications that it is even the main door, small windows, stones of different colors, no texture between windows, the crenelated roof for defence, a single tower in one corner…

Now compare that to the Medici Riccardi. Still strong, but now the round arches with voussoirs on top make them taller and much more conspicuous. There are more, bigger windows,12 external decorated marks for each floor, the roof overhang is now beautifully decorated, dimensions feel harmonious…

And I think this becomes even stronger when you have several houses following this style, even when some of them mix with other styles.

The city center of Arezzo, near Florence.

So Why Did Florence Birth Michelangelo, Fibonacci & Galileo?

Although I am lukewarm about Renaissance church architecture, this doesn’t take away from the improvement in non-Church architecture, and more importantly the amazing intellectual movement that emerged in Florence.

The Renaissance precursor Dante Alighieri revolutionized literature; Luca Pacioli, a collaborator of Leonardo da Vinci, became a ground-breaking figure of accounting; Piero della Francesca emphasized the divine proportions; Fibonacci13 discovered the Fibonacci sequence and popularized Indian numerals; Galileo the astronomer defended heliocentrism against the church; Botticelli, Boccaccio, and the aforementioned Michelangelo, the Medici, Da Vinci, Petrarch, Donatello, Raphael, Brunelleschi…

Made with nanobanana

And now we know what the Renaissance is, and why it started in Florence:

Because of the mountainous terrain, the Alps, and the length of the Italian Peninsula, Northern Italy was distant from foreign centers of power, which gave them autonomy, especially given the contest between the HRE and the Church.
Northern Italy was rich because it was at the right spot in the Mediterranean, so it was the marketplace for Europe, the Middle East, and North Africa.
It also has highly fertile valleys, so the population grew fast.
A big share of that population was urban, which means network effects, more industry, trade, and money.
Within this region, Florence was rich as a crossroads between Rome and Northern Italy. It developed a wool industry, and from there other industries, including finance.
It was close enough from Rome to get its patronage, but not too close to be absorbed by it.
It was lucky to be on the winning side of the war between the Church and the HRE, and got a massive boost when the Church won, with financial and mining monopolies.
It was close enough to Rome to gain inspiration from Ancient Rome and be able to study it.

Once Florence had the money and the independence to go beyond the daily focus on war and survival, it could use the example of Rome to study it and replicate it. This meant massive investments in architecture, art, and all types of studies of Ancient Rome. This created a critical mass of thinkers who learned from each other, creating an explosion of knowledge across the disciplines.

In many regards, this reminds me of what I witnessed in Silicon Valley in the nearly 15 years I spent there: an incredible concentration of intelligent, ambitious people, all focused on pushing the boundaries of one new technology—the Internet. It’s difficult to get these network effects set up. Can San Francisco survive its current downfall? Where will the next Florence emerge? What type of endeavor can justify the massive investments that are concentrated in one place, to attract geniuses from everywhere? Can Dubai do it, or does it not have enough of a mission? Shenzhen fits the definition. Is it limited to manufacturing, or will it bleed into other types of innovation? What do you think?

This type of article takes blood and sweat, and they don’t leave me much more time to do other things. But I want to do even better ones! And more of them! Across more media! I hope you saw the video at the top. I want to do more of that too! But I can’t do this alone. So I am going to hire a team to help me research, write, and publish more articles, convert this content into videos, audio, and podcasts, translate them, make tools to make them interactive, and much more. But I can only do that if you help me pay for it. If you like articles like this one and you want to see more, I need you to fund it.

Please become a premium subscriber!

This claim sounded crazy to me, so I looked into it. The Florentine Duomo, built in 1436, apparently remained the biggest dome in the world for 450 years, until 1871, when it was replaced by the Royal Albert Hall dome. But all modern domes use steel somehow, either directly or as reinforced concrete. Brunelleschi’s Duomo in Florence doesn’t, it’s just masonry: a series of construction elements bound with mortar and working by compression.

Architectural ribs, not those in his body. More details here.

Arezzo was absorbed into Florence soon after, in the 1380s.

Apparently it was already in circulation, and the original architect of Santa Maria del Fiore was inspired by it, but this rediscovery made its knowledge widespread.

The 3^rd and 4^th turtles. You didn’t think I had forgotten, did you? Although born in Urbino, Raphael became deeply connected with Florentine artistic circles before moving to Rome. Michelangelo was born in the Florentine village of Caprese.

There’s a fantastic write-up from Martin Sustrik about Ada Palmer’s class on Italian politics of the time. The gist of it is that this class puts the students in the skin of different characters, and they role-play the history, which is the closest thing we have to having randomized controlled tests in history, and what it finds is that some outcomes always happen, but some others change, and the details are always different. This is strong support for the theory that big parts of history are predetermined.

With Western countries like Castile, Portugal, Aragon, France, and England emerging. Flanders was part of the Holy Roman Empire and/or France depending on the moment. It’s in the area where France, the HRE, and the Hanseatic League overlap. I mean the HRE “in general” as it controlled Flanders, and many parts of HRE were rich, but not all.

I think the wool trade has not received the attention it deserves. It accounted for a massive share of GDP—I believe it was the second biggest industry after food. I hope to tackle it some day. If you know good sources, LMK.

Siena was closer to Rome than Florence, so it felt the Papal power more strongly. It made sense for it to oppose Rome, in hopes of the distant HRE suppressing Rome, which would have given Siena more autonomy.

Alum is a mordant, meaning it creates a chemical “bridge” between dye molecules and the fibre. There were no alum mines in Europe; it was all provided by the Ottomans in the 1400s, until an alum mine was discovered near Rome, in the Papal States. The Popes gave their mining rights to the Medici.

It had to be though because it wasn’t safe at the time.

Notice we’re still not very secure, as the windows had ironwork to fend off criminals.

From neighboring Pisa, which became part of Florence-controlled Tuscany.

Collections: That Dothraki Horde, Part IV: Screamers and Howlers – A Collection of Unmitigated Pedantry

Wed, 25 Mar 2026 15:05:39 +0000

This series is now available in audio format. You can find the playlist here.

This is the fourth part of a four part (I, II, III, IV) look at the Dothraki from George R. R. Martin’s A Song of Ice and Fire and HBO’s Game of Thrones. We’re looking at, in particular, if Martin’s claim that the Dothraki are “an amalgam of a number of steppe and plains cultures” can be sustained in the face of even basic knowledge about historical Steppe and Great Plains nomadic peoples.

Last week, we concluded that the vast majority of Dothraki culture, social organization, economic practices and family structure are effectively completely untethered from the historical realities of effectively any of the literally dozens of historical Great Plains Native Americans or Steppe nomads. This week, we’re going to close out our look by discussing Dothraki warfare. We’ll start with the visual – weapons and armor – and then move to the conceptual – strategy, operations and tactics.

And as always, if you like what you are reading here, please share it; if you really like it, you can support me on Patreon. And if you want updates whenever a new post appears, you can click below for email updates or follow me on twitter (@BretDevereaux) for updates as to new posts as well as my occasional ancient history, foreign policy or military history musings.

Email Address

Finally, as a reminder both of what we are investigating, the key statement we are really assessing here is this one by George R.R. Martin:

The Dothraki were actually fashioned as an amalgam of a number of steppe and plains cultures… Mongols and Huns, certainly, but also Alans, Sioux, Cheyenne, and various other Amerindian tribes… seasoned with a dash of pure fantasy.

It is not the existence of a fantasy culture which draws our attention, but the explicit declaration that this fantasy culture is not merely inspired, but ‘fashioned as an amalgam’ of real cultures, which both existed in the past and still exist today, with only ‘a dash of pure fantasy.’ That line is important, to be clear, because it presents the fictional Dothraki as a statement on historical Native American and Eurasian nomads and – when combined with Martin’s statements that he relies on history to inform his work – that this statement is based in some sort of historical reality.

Which it isn’t. But we’re getting ahead of ourselves.

Where There’s a Whip…

The Dothraki are described as having three main weapons: bows (AGoT, 86, 555, 558, 597, 669), whips (AGoT, 86, 194, 493, 555, 596, 669) and a curved sword called an arakh (AGoT 85, 86, 327, 493, 555, 556, 559, 560, 596, 597, 669, 674); of these, the arakh is clearly the most prominent (I am sure I have missed a reference to a weapon here or there, but I hope the citations here give some sense of the relative weight each is given – the arakh is the most frequently mentioned by some distance). When a Dothraki warrior enters Vaes Dothrak, each, “unbelted his arakh and handed it to a waiting slave, and any other weapons he carried as well” – after the arakh, the other weapons are seemingly afterthoughts (AGoT, 327). The prominence of the arakh in the narrative is underscored by the fact that it is the only one of these weapons whose name we learn in Dothraki, or which is described in terms of its shape or special function (AGoT, 85), while the bows and whips remain just bows and whips (ironic, as it was Steppe bows, not Steppe swords, which were unusual).

We might dismiss this as simply an accident of Daenerys’ perspective – that, being Westerosi, she focuses on the weapon most meaningful to the Westerosi – but that’s clearly not true. After all, the offering of an arakh is how Daenerys’ loyal followers demonstrate their fealty to her, in a ceremony that is clearly Dothraki, not Westerosi (AGoT, 674). It is also, I should note, the only weapon we see non-Dothraki using that is clearly identified as being foreign and typical of the Dothraki. It remains special through the eyes of multiple point-of-view characters, including military men.

(And, as an aside, now that we are this far in, it seems obvious but worth saying that the fact that Martin has no Dothraki viewpoint characters in his narrative is hardly a saving grace; it merely intensifies the ‘view of a savage culture from outside’ effect. As we’ll see, this makes perfect sense given what seem to be the actual inspirations for his depiction.)

The prominence of a curved iron (or steel) sword lets us rule out a Great Plains Native American inspiration for this kit right out; the sword was never a significant part of Plains Native American armament (the lack of tool-metal production in the Americas prior to European contact means that there was no indigenous sword-making tradition, although the maquahuitl represents a clever sort of ‘sharpened club’ design). Even after contact, it’s hard to avoid the conclusion that the expense of trading for a sword wouldn’t have been justified by its utility over a steel axe which might also double as a tool (on axes, see W. Lee, “The Military Revolution of Native North America: Firearms, Forts and Politics” in Empires and Indigenes (2011), 62-3). So we must turn to the Eurasian Steppe.

And immediately we run into problems, not that any of these weapons are wrong per se, but that their proportion and prominence is all mixed up and that there are other, far more important weapons missing.

For a Steppe nomad, by far, above and away, the most important weapon was the bow. The Armenians literally called the Mongols “the nation of archers” (May, Mongol Art of War, 43). Nomads spent the most time learning the bow (May, op. cit. 42-49) and it was the one indispensable weapon. Indeed, so indispensable that nomads were generally required to have several; the Liao Shi records that Khitan nomad warriors were required to possess four bows and 400 arrows, while John de Plano Carpini reports that the Mongols all needed to have 2-3 bows and three larger quivers (May, op. cit. 49-50). The Steppe bow itself would also have looked unusual in both shape and construction to a Westerosi observer either strung or unstrung – they were composite bows, made with a wood core, a backing of horn and a rigid end-piece (called a siyah in Arabic) and were generally drawn with the use of a thumb-ring to reduce strain on the thumb (May, op. cit., 50-1). This unique construction allowed these bows to reach draw weights and launch energies equivalent to the far larger yew longbows of England and Wales and still be compact enough to use from horseback.

Via Wikipedia, a 13th century Mongol horse archer. Lightly armored, he carries a bow (and a fancy hat) but no sword.

(I should note that the bow was also the paramount weapon for the Native American horse-borne nomads of the Great Plains, at least until it came into competition with firearms, though my understanding is that Native American bows were not as powerful as Steppe bows).

Via Wikipedia, a modern Mongolian woman taking part in an archery contest. You can see here the unique shape and multi-part construction of the Steppe bow (notice how the material on the tips, the belly and the spine of the bow are all different) which allows it so much power in such a small frame.
Also, notice the very nice and colorful traditional Mongolian clothing – not leather and rough furs!

But even after the bow, the sword is not first. Or even close to first. Or, indeed, even on the list! The Khitan regulations I mentioned included four bows, two spears (one ‘long’ and one ‘short’), a club, an axe and a halberd, but no sword. John de Plano Carpini describes the full kit as two or three bows with quivers, an axe, ropes, and swords only for the wealthy (May. op. cit., 50). Speaking more broadly, May notes that spears (used as lances from horseback) seem universal in accounts of the Mongols, but “accounts are contradictory regarding whether these [swords] were universally used” (May, op. cit., 52). While May supposes that the ughurgh-a, the Mongolian lasso, might have been used in combat – and it may well have – we have no definitive evidence of it. If it was ever a weapon, it doesn’t seem to have been an important one.

In short, while the Dothraki’s weapons are an arakh-sword, a whip, and a bow in that order, the Mongol’s chief weapons were his bow, followed by his backup bow, followed by his other backup bow, followed by his spear, and then his axe and only then followed by a sword, should he have one, which he might well not. The reason for preferring an axe or a spear for the humble nomad should not be too surprising – iron in quantity could be hard to get on the Steppe. Spears and axes are not only weapons, but also useful hunting and survival tools; swords are generally weapons only. Nomads generally cannot do their own metal working, so swords would have to be imported. Moreover, even in a melee, the first recourse would be to a spear, whose reach on horseback was a huge advantage, making a sword an expensive imported foreign luxury backup weapon with no additional utility. Nevertheless, it’s clear that Steppe nomads, once successful and moving into agrarian areas, liked to acquire swords – swords are effective weapons! – but the sword was about the furthest thing from the core of Mongol culture the way the arakh is practically the symbol of Dothraki culture.

Via Wikipedia, a relatively late Mongol soldier (c. 1755) nevertheless shows nearly the full kit, including mail body defense, a long spear for use on horseback, arrows (the bow in its bow-case would have been on the other side) and, this being the 1700s, a musket.

The other issue, of course, is the arakh itself. Martin describes the weapons as “long razor-sharp blades, half sword and half scythe” (AGoT, 85) and goes back to that scythe analogy (e.g. ASoS, 245). It seems generally asserted that what Martin means by this is something close to a scimitar (I have to confess, I haven’t found anywhere that Martin says this, but I’ll take the reader consensus). A scimitar of some sort (the term refers not to a specific form of sword, but a whole family of curved sabres, almost all originating in Asia) is the correct sword. Mongol swords were, John de Palno Carpini tells us, “pointed at the end but sharp only on one side and somewhat curved” (May. op. cit., 50), something like a Turkish kilij or a Persian shamshir (both forms of scimitar), though given his description, perhaps not as strongly curved as some of the examples of those types.

I have to admit, ‘scythe-sword’ (ASoS, 245) is a really strange way to describe most of the weapons in the scimitar ‘family’ (which includes a number of different curved sabres from Asia), though. A scythe-blade faces the wrong direction, but it is also sharp on the wrong side – scythes are sharp on the inside of the curve, whereas scimitars are sharp on the outside of the curve. There are swords with sharp edges on the inside of the curve (I tend to class these as ‘forward curving’ swords due to the direction of the curve when the sword is held), such as the Greek kopis, the Spanish falcata and the Nepalese kukri; of these, only the kopis seems to have been a cavalryman’s weapon (Xen. On Horsemanship 12.11). These forward curving weapons, being shorter and stockier, are clearly not what was intended by the arakh, which is consistently described as long (e.g. AGoT, 85, ADwD, 884).

Via Wikipedia, an Uzbek warrior, armed with a bow (in a bow case, important accoutrements of any mounted archer and one we almost never see in film) along with a shamshir. Notice that, while it is curved, it is not exceptionally long; scimitars generally aren’t – another thing Martin has wrong about them.

Instead, the scythe metaphor fits the overall framing of the arakh, a weapon “better to cull the infantry ranks without breaking stride,” a “murderous blade against half-naked foes,” (ADwD, 884), a “wickedly sharp scythe-sword” (ASoS, 245), ineffective against armored opponents. Not an elegant, fine weapon, but a cruel ‘murderous’ one, made for ‘culling’ unarmored infantry and peasants, as one reaps wheat or hay. I don’t want to push this point too far – in all of these many pages, the arakh simply doesn’t get enough characterization to make the case watertight – but the characterization it does get all seems to push in this same direction: a murderous weapon for a murderous people…which of course fits with effectively all of the other characterization the Dothraki have been given. On the balance, I think Martin is a skilled enough writer to understand the implications of the scythe-sword description and to have intended them (and then subsequently reinforced them).

Nevertheless, credit where credit is due, while the place of the arakh is entirely out of all sensible proportion with how it would be considered by actual nomads, it is the correct sort of sword for a steppe nomad (if we assume it is, in fact, a scimitar of sorts). That said, prioritizing the arakh belies a fundamental misunderstanding of how Steppe (or Plains Native American, for that matter) warfare and culture worked. Placing the arakhat the front is thus indicative of deeper problems.

Of course we couldn’t leave off without discussing the absolutely bizarre visual adaptation of the weapon for Game of Thrones, where the scimitar-like arakh is transformed into what is essentially an oversized iron khopesh:

Presumably what happened here was that someone read ‘scythe-sword’ who had never seen a scythe, but had seen a sickle, and decided that a sickle-sword was the way to go, but also wasn’t aware that sickles are sharp on the inside of the curve and not on the outside of the curve, and so went with a forward-curving ‘sickle-sword’ design (which is sharp on the outside of the curve). And then, for good measure, inexplicably chose a short weapon made for the bronze age and just scaled it up to absolutely massive size. Moreover, the show’s version of the arakh inexplicably has a long, two-handed hilt, supremely impractical from horseback.

There are so many problems here. First, the khopesh is more of an axe-sword than a sickle-sword. Moreover its form is directly connected to the material properties of bronze. Because bronze doesn’t resist bending as well as iron, bronze swords need to be short and the khopesh was generally below – often well below – 60cm in total length (markedly smaller than, for instance, a gladius; do note that the gladius and the khopesh were never on the same battlefields, they are separated by almost 1,000 years). While the Ethiopian shotel is a tempting comparison point for an iron ‘sickle-sword,’ unlike the arakh (or a khopesh), the shotel is actually a sickle-sword, sharpened on the inside of its curve and used for hooking attacks; the arakh of the show is clearly not wielded like a shotel – instead attacks are mostly made with the outside of the curve (though both might be sharp?).

Via Wikipedia, an 18th century (BC!!) Khopesh blade, now in the Staatliches Museum Ägyptischer Kunst, Munich.

By any measure, the result is a terrible weapon. Weapon designs cannot simple be ‘scaled up’ like this without ruining the things that made them successful; a jumbo-sized khopesh is almost guaranteed to be too heavy. Unlike most scimitars – note John de Plano Carpini on the sharp points of Mongol swords above – it cannot give point, which (contra Xenophon) is a real disadvantage on horseback. It looks to be a two-handed weapon for use on horseback where one-handed weapons are most appropriate. As a two-handed weapon, it has inferior reach for what must be its considerable weight, and the forward curving shape offers it essentially no cutting advantage, unlike the kukri, kopis or falcata, which lean into their cuts.

I haven’t any idea why they opted to adapt the weapon this way, except to note that it fits in with the general pattern of the show taking Martin’s already cringe-worthy exoticism in treating the Dothraki and dialing it up to 11.

Armored Arrogance

The Dothraki attitude towards armor is made fairly clear. “The Dothraki ¹ had mocked him [Jorah] for a coward when he donned his armor” (AGoT, 556), something echoed later by the arakh-wielding but Merenese (that is, not Dothraki) Khrazz (ADwD, 885). The show extends this out to a discussion between Rakharo and Jorah (S1E3, 48:10ff) , that “Dothraki don’t wear steel dresses” (though here, because they make them “slow” rather than because it is cowardice). In short, the Dothraki have a general contempt for armor and for those who wear it.

The unarmored ‘barbarian’ who attacks heedless of his peril, all fury and offense, no reason or defense (this is a part of the ancient form of the Fremen Mirage), is a literary trope that goes back at least to the Greeks and Romans (::deep breath:: Plut. Marcellus 8; Dionysius 14.9.2; Diodorus 14.9-10; 5.30.3; Liv. 7.10.7-10; App. Gal. 6; Plb. 2.30.1 and 3.114.4 but cf. Liv. 22.46.6; Caes. B.G. 4.1; in artwork note J.R. Marszal, “Ubiquitous Barbarians: Representations of the Gauls at Pergamon and Elsewhere” in From Pergamon to Sperlonga, eds. N.T. de Greummond and B.S. Ridgeway (2000)). In some cases the lack of armor (or clothing) seems to have been accurate and in some it was not (by the by, the Gauls with their supposed barbarian contempt of armor invented mail, probably the single most successful pre-modern armor technology); the idea that these ‘barbarians’ were madly reckless was never accurate – the Gauls used effectively the same full-size body-shield the Romans did which shows a real concern for personal protection! That same language – the irrational, ‘swarming’ natives, heedless of danger or death – reappears in later European military writing, particularly in the early modern and after (which is to say, during European imperial expansion) as a racist marker of non-European inferiority (on this, note P. Porter, Military Orientalism: Eastern War Through Western Eyes (2009), esp. 68ff where he notes the same tropes were applied to Native Americans, which – see below at Stagecoach for the almost certain line of connection where this trope gets to A Song of Ice and Fire).

(And yes, by the by, more than a few of the subjects of this bad old trope seem to have, at one point or another, adopted it in the way the Dothraki do in A Song of Ice and Fire – attributing weakness and cowardice to the ‘westerners’ and their supposed casualty aversion. Both sides of the stereotype are nonsense and states that have made strategic decisions based on those tropes have almost invariable faced disaster. That said, for reasons that will be obvious below, actual nomads do not adopt this position because they haven’t the population to sustain it.)

In short, this is a very old literary trope repurposed as a still-old hoary racist stereotype. It is also hot nonsense, both generally and as applied to our horse-borne nomads.

The popular image of the Great Plains Native America is unarmored, of course, but that image fundamentally formed in the late 19th century, when – after centuries of the development of gunpowder weapons – everyone was unarmored. A longer view shows that Plains Native Americans were perfectly capable of both developing or adopting defensive measures which worked. And to get a full sense of that, we need to outline the major phases of the changing warfare on the Great Plains.

F.R. Secoy (op. cit.) essentially breaks warfare into four phases, which happen at different times in different places, based on if they have horses, guns, both or neither. Because horses entered the Great Plains from the South (via the Spanish) but firearms entered the region from the North (via the British and the French, the Spanish having prohibited gun-sales to Native Americans) and spread out from there, for a brief time many of these systems were active on the Plains at once, as both guns and horses diffused through the region.

In the pre-horse, pre-gun phase (described by McGinnis as well, op. cit., 8-9), battles consisted of long-range missile exchanges between warriors who stood behind large shields which protected their whole bodies. Native American warriors in this system also wore armor, heavy leather coats, laminated in multiple layers using thick hide with glue that was sometimes mixed with sand or gravel (one more example of how ‘leather armor’ is almost always hardened leather armor, not modern clothing-leather). Some of this armor may have been effectively quilted leather as well. Clearly, there was plenty of concern about survivability here.

Both guns and horses were apt to disrupt this system. Horses allowed attackers to rapidly close the distance between the two opposing lines of shield-protected foot-missile-warriors, causing the shield-lines to drop away (though smaller shields, used on horseback to ward off arrows and blows were still used) and for both sides to seek instead the mobility of mounted fighting. That was not the end for armor though, because contact with a supply of horses meant contact with the Spanish, and the Apache at least swiftly adopted some of the Spanish methods of making leather ‘buff coats’ into their own armor tradition and copied the shape and pattern of the buff coat itself (while often still making the material using their own tradition). As Secoy notes (op. cit., 18-20), our sources are quite clear that these forms of armor (both original and Spanish-influenced Native armors) were quite effective at resisting the archery fire that dominated both the pre-horse, pre-gun system and the post-horse, pre-gun system.

Meanwhile on the Northern Plains, while the horse had not yet arrived, firearms had, and these had different effects. Firearms spelled the end of the armor once they became common enough, since there was no protection which could resist bullets; some shield use survived, since arrows remained fairly common as well. But this didn’t lead to suicidal warfare. Instead – as had happened on the East Coast as well, Native Americans adapted their warfare to the increased lethality of firearms (on this, note Lee, op. cit. above) by mostly avoiding pitched battles as they became too lethal (which, by the by, the relatively low lethality of pitched battles pre-gunpowder is often taken to mean that Native North American warfare in general was low-lethality; this is wrong. As with most forms of non-state warfare, most of the killing happened in surprise raids and ambushes, which could be extremely lethal and were still quite common).

Once the horse and the firearm were both in wide use in an area, warfare shifted again. War parties became smaller, faster moving and more reliant on surprise (essentially an extension of the raiding-focus of the pre-horse, post-gun system to the high mobility horses supplied). Infantry battle dropped away entirely because it was too lethal and resulted in casualties that low-population density nomads could not sustain (the contrast with the much higher population-density agrarian United States, which was self-immolating in massively costly massed-infantry engagements during the American Civil War, 1861-1865, at exactly this time is striking). These are fairly big, noticeable changes in warfare patterns!

In short, the tactics used in all four of these systems were conditions by casualty aversion, which makes a lot of sense in the context of a low-population density society which simply cannot afford massive losses.

Meanwhile the Mongols were quite fond of armor, though it is clear that they required access to the products of agrarian economies to get it. That same Khitan regulation I noted above required soldiers to possess nine pieces of iron armor, along with barding (that is, armor) for their horses. John de Plano Carpini describes the use of thins trips of leather and hide, bouind by cord to create a scale of lamellar horse barding. As May notes (op. cit., 53) the Mongols tended to prefer lamellar armors (that is, armors of overlapping rectangular plates attached to each other rather than to a backing) of either hardened leather or iron because these were more effective at stopping arrows than mail. The Mongols also seem to have really liked pointed conical helmets (the Turks did too) and seem to have contributed to their spread. There is actually a fair amount of evidence that the later European brigandine was a Western European adaptation of steppe lamellar armors, mediated through Eastern Europe. In short, Mongol armor (which again, is generally not being produced by them on the steppe but produced for them by the agrarian societies, which in some cases involved violently moving those craftsmen to where the Mongols needed them) was so good that it was quickly adopted in Europe when it arrived.

Minature depicting an Turkish Sipahi, sometime before 1657, wearing what looks to me to probably be lamellar armor (though it may also be textile).

Indeed, looking broadly at not only the Mongols, but also Turkic steppe nomads (Mamluks, the Seljuks, the Ottomans, etc), incorporating the heavier armor made possible by agrarian societies and their metal production industries seems to have been a fairly high priority for nomads moving into settled zones more generally. And I should note that while the Mongols preferred lamellar to mail (as did other Steppe nomads), they absolutely would and did use mail if mail was what was available; the Turks and the Timurids both made extensive use of mail as well as lamellar, scale and so on. While armor might be rare on the Steppe due to its expense and the relative inability to produce good armor locally, it was valuable and valued and nomads who ‘made good’ worked to acquire it.

In short, if the Dothraki followed the model of either the Great Plains Native Americans or Steppe nomads, far from holding armor in contempt, we’d expect a wealthy leader like Drogo to have made serious efforts to either acquire effective foreign armor both to protect himself and his immediate retainers. Remember that providing armor for your followers would not merely improve their combat effectiveness, but provide an opportunity to reward loyalty and more tightly bind those followers to you.

But of course that wouldn’t be badass and violent enough, so instead Martin deploys the bad old trope of the irrational, mindless barbarian careless to his own peril, going into battle unarmored.

Dothraki Ends

I know we have dealt with this distinction before in a number of other places, but I want to make sure we are all on the same page here, briefly. Tactics concern the how of warfare at the small scale; how a battle is fought. Operations concern how armies are moved and thus where a battle is fought. Strategy concerns the ends for which a war is waged in the first place and thus why battles are fought.

Each of these levels is a category of analysis, but of course not every general starts at first principles when going into a conflict. Instead, wars are often waged according to traditional systems of norms and expectations. That said, when you dig in to those systems of norms and expectations, the basic correlation of strategic ends to the means of operations and tactics generally emerge (if not the least because polities which fail to coordinate these things tend not to be permitted to play the game for very long).

Since strategy dictates operational concerns, which in turn dictate tactical concerns, we’ll move in that order and so begin with the question what does Dothraki warfare aim to achieve and are those the same ends as nomadic warfare on the Great Plains and the Eurasian Steppe?

Assessing the strategy of Dothraki warfare is tricky, because while we spend a good chunk of the story near a Dothraki leader, strategic aims are usually not discussed with our viewpoint characters. Still we see enough of how Dothraki khalasars function to get a sense of the general aims of Dothraki warfare.

First, it seems that the Dothraki Sea is essentially a war of all against all. As Daenerys notes once she is weakened, “the first khalasar they met would swallow up her ragged band, slaying the warriors and slaving the rest” (ACoK, 142). And indeed, we see this when Drogo’s and Ogo’s khalasars raided the same town; Drogo does exactly that, killing the adult men and enslaving the rest of Ogo’s defeated khalasar (AGoT, 556), while Pono’s khalasar scatters others before it as it moves as they rush to get out of the way (ADwD, 113). No khals appear to be allied with each other, there are no tribal confederations of smaller units; merely a collection of khalasars, each led by a khal, all at war with all of the others all of the time except when in Vaes Dothrak. Of course as already mentioned, the agrarian peoples on the edges of the Dothraki Sea are also subjected to this treatment, unless they are either key trade partners or buy the Dothraki off.

The main aim of this functionally total conflict is the capture of slaves. As we’ve discussed previously, the Dothraki don’t herd livestock, and horse-raiding isn’t ever prominent in the text either. Indeed, when attacking the Lhazareen, the Dothraki kill all of the livestock and leave their bodies to rot in the fields (AGoT, 555), instead, they take slaves. Of the Dothraki captives, we are told they are only “the women and children of Ogo’s khalasar;” the captive Lhazareen include “only a few men among them, cripples and cowards and grandfathers” (AGoT, 555-6); so the slaves in question are mostly women, children and the elderly (a strange choice since these are, historically speaking, the least valuable classes of people to enslave; most enslaved workers were forced to do agricultural or industrial labor for which adult males were typically considered the most suitable by far).

Apart but strangely, apart from a few domestics, the Dothraki have apparently no use for these enslaved people internally. Captive women evidently are not able to become wives or bear legitimate sons (AGoT, 559) and without any kind of domestic production (because of the non-existent subsistence model discussed last time) it is hard to imagine what work large numbers of enslaved persons would be forced to do. Instead, we are repeatedly told these masses of slaves are not incorporated into the khalasar, but traded away to the cities of Slaver’s Bay in exchange for ‘gifts’ (ADwD, 662) or left in Vaes Dothrak.

Crucially, in the description of the movements of khalasar, it is clear they do not fight over territory. Drogo’s khalasar cuts all of the way from Pentos over effectively the entire length Dothraki Sea to Vaes Dothrak. He then plans to head to the Jade Sea, which would mean covering the entire height of the Dothraki Sea (since Vaes Dothrak is at its northern extent) and going even further East, but instead cuts south to the territory of the Lhazareen. The movements of the other khalasars (ADwD, 113, 662) are like this as well. The Dothraki essentially cruise the empty Dothraki Sea like actual ships on the actual sea, without any concern for territory. At no point does any group of Dothraki get angry with any other group of Dothraki for territorial trespass.

A quick map showing the movements of Drogo’s (and later Daenerys’) khalasar, along with a line showing his initial planned movement to the Jade Sea. Essentially Drogo is able to cover the entire Dothraki Sea without any territorial boundaries at all.

Grass Wars

This bears little resemblance to the strategic concerns of historical nomads. As a direct consequence of failing to understand the subsistence systems that nomads relied on, Martin has also rendered their patterns of warfare functionally unintelligible.

The chief thing that nomads, both Great Plains Native Americans and Eurasian Steppe Nomads used violence to secure control of is the one thing the Dothraki never do: territory. To agrarian elites (who write most of our sources) and modern viewers, the vast expanses of grassland that nomads live on often look ’empty’ and ‘unused,’ (and thus not requiring of protection), but that’s not correct at all. Those ’empty’ grasslands are very much in use; the nomads know this and are abundantly willing to defend those expanses of grass with lethal force to keep out interlopers. Remember: the knife’s edge of subsistence for nomads is very thin indeed, so it takes only a small disruption of the subsistence system to push the community into privation.

For the Eurasian Steppe nomad, the grass that isn’t near their encampment is in the process of regrowth for the season or year when it will be near their encampment and need to support their herds. Allowing some rival nomadic group to move their sheep and their horses over your grassland – eating the essential grass along the way – means that grass won’t be there for your sheep and your horses when you need it; and when the sheep starve, so will you. So if you are stronger than the foreign interloper, you will gather up all of your warriors and confront them directly. If you are weaker, you will gather your warriors and raid the interloper, trying to catch members of their group when they’re alone, to steal horses and sheep (we’ll come back to that); you are trying to inflict a cost for being on your territory so that they will go away and not come back.

Via Wikipedia, an illustration of Rashid-ad-Din’s Gami’ at-tawarih, ewarly 14th century, showing Mongols (note their heavy lamellar armor and distinctive composite bows (esp. upper left) pursuing fleeing enemies.

The calculus for nomadic hunters like the Great Plains Native Americans is actually fairly similar. Land supports bison, bison support tribal groupings, so tribal groups defend access to land with violent reprisals against groups that stray into their territory or hunt ‘their’ bison. And of course the reserve is true – these groups aren’t merely looking to hold on to their own territory, but to expand their subsistence base by taking new territory. Remember: the large tribe is the safe tribe; becoming the large tribe means having a larger subsistence base. And on either the plains or the steppe, the subsistence base is fundamentally measured in grass and the animals – be they herded sheep or wild bison – that grass supports. Both Secoy and McGinnis (op. cit.) are full of wars of these sorts on the Great Plains, where one group, gaining a momentary advantage, violently pushes others to gain greater territory (and thus food) for itself. For instance, Secoy (op. cit., 6-32) discusses how access to horses allowed the Plains Apache to rapidly violently expand over the southern Plains in the late 17th century, before being swept off of them by the fully nomadic Ute and Comanche in the first third of the 18th. As McGinnis notes (op. cit., 16ff), on the Northern Plains, prior to 1800 it initially was the Shoshone who were dominant and expanding, but around 1800 began to be pushed out by the Blackfoot, who in turn would, decades later, be pushed by the expanding Sioux.

This kind of warfare is different from the way that settled, agrarian armies take territory. Generally, the armies of agrarian states seek to seize (farm-) land with its population of farmers mostly intact and exert control both over the land and the people subsequently in order to extract the agricultural surplus. But generally (obviously there are notable exceptions) nomads both lack the administrative structures to exert that kind of control and are also very able to effectively resist that sort of control themselves (it is hard for even nomads to tax nomads), making ’empire building’ along agrarian lines difficult or undesirable (unless you are the Mongols). So instead, polities are trying to inflict losses (typically more through raiding and ambush than battle). Since rivals will tend to avoid areas that become unsafe due to frequent raiding, the successful tribe can essentially push back an opposing tribe with frequent raids. In extreme circumstances, a group may feel threatened enough to get up and move entirely – which of course creates conflict wherever they go, since their plan is to disposes the next group along the way of their territory.

Within that security context, larger scale groupings – alliances, confederations, and super-tribal ‘nations’ – are common. On the Eurasian Steppe, such alliances tended to be personal, although there was a broad expectation that a given ethnic grouping would work together against other ethnic groupings (an expectation that Chinggis actually worked very hard, once he became the Great Khan of a multi-ethnic ‘Mongol’ army, to break up through the decimal organization system; this reorganization is part of what made the Mongol Empire so much more successful than previous Steppe confederations). Likewise, even a cursory look at the Native Americans of the Great Plains produces both a set of standard enmities (the Sioux and the Crow, for instance) but also webs of peace agreements, treaties, alliances, confederations and so on. The presence of British, French, Spanish and American forces (both traders and military forces) fit naturally into that system; the Plains Apache allied with the Spanish against the Comanche, the Crow with the United States against the Sioux and so on. Such allies might not only help out in a conflict, but also deter war and raiding because their strength and friendship made lethal retaliation likely (don’t attack someone allied to Chinggis Khan and expect to survive the experience…).

Exactly none of that complexity appears with the Dothraki, who have no alliances, no peace agreements, no confederations and no territory to attack or defend. Instead, the Dothraki simply sail around the grass sea, fighting whenever they should chance to meet. Which brings us to:

Raiding

The other strategic aim nomads might fight over is for the acquisition of some kind of movable good, which is to say raiding for stuff. Because all of the warriors (which is generally to say all of the free adult males) of these societies are mounted and because they have a subsistence system which allows rapid, relatively along distance movements (often concealed; remember that Mongols need not light any camp fires), nomads make fearsome raiders, able to strike, grab the things they are looking for and quickly retreat before a counterattack can be mobilized. That goes just as well for raiding each other as it does for raiding the farmers at the edges of the grasslands.

But what are the things here that they are aiming to get? It depends on the targets; nomadic raids into the settled zone generally aim to capture the goods that agrarian societies produce which nomadic societies do not: stocks of cereal crops, metal goods and luxury goods. But most nomadic raiding was directed against other nomads, seeking to acquire either people or animals.

On the Great Plains, the animals in question were invariably horses; the act of stealing, or ‘cutting out’ a horse gives McGinnis part of the title of his book (Counting Coup and Cutting Horses) and raids for horses dominate both McGinnis and Secoy’s discussion of Plains Native American warfare. Horses were, after all, a scarce commodity which only percolated into the Great Plains from the South (and which could only be raised in quantity in its southern reaches), but which all tribes required both to hunt and fight effectively. Stealing enemy horses thus both strengthened your tribe while weakening your enemies, both in military and subsistence terms. The Mongols also engaged in quite a lot of raiding for horses, but also – in a pastoral subsistence system – a lot of simple cattle rustling as well (e.g. Ratchnevsky, op. cit., 28-31).

Raiding for people is more complex, but undeniably part of this system of warfare. But crucially this raiding was generally not for slave-trading (though there are exceptions which I discussed last time), but instead incorporative raiding. What I mean by that is that the intent in gaining captives in the raid was to incorporate those captives, either as full or subordinate members, into the nomadic community doing the raiding. Remember: the big tribe is the safe tribe, so incorporating new members is a good way to improve security in the long run.

On the Eurasian Steppe, incorporated captives became the ötögus bo’ol ‘bonded serfs’ that we mentioned previously (Ratchnevsky, op. cit., 12-4). Unlike warfare on the Great Plains, it seems possible for the bo’ol to include adult men, either captured or sold (by destitute parents) as children or else taken as prisoners when their tribe or clan was essentially dissolved by being conquered in war. Indeed, in his own conquests, Chinggis only decreed the annihilation of one tribe, the Mongol’s traditional enemies, the Tatars – there he ordered the death of any Tatar male taller than the linchpin of an oxcart (May, Mongols, 12). In other cases, it is clear that the incorporation of defeated nomad warriors into the successful tribe was fairly normal, though raids to capture women and children (also for incorporation) were just as common. Bride abduction in particular was very common on the Steppe, as Ratchnevsky notes (op. cit., 34-5).

The incorporation of males was far less common in Great Plains Native American warfare, but the capture of women and children to enhance tribal strength in the long term was a core objective in raiding. McGinnis (op. cit., 42-3) notes how the Crow, after suffering a massive defeat in the early 1820s which resulted in the deaths of many warriors and the capture of perhaps several hundred women and children, steadily built their tribe back up over the following decades with an intentional strategy of capturing women and children from their enemies. As McGinnis (op. cit., 24) notes, women captured in this way might be married into the capturing tribe, adopted into it, or sometimes kept as an enslaved laborer (under quite bad conditions). Adult males, by contrast, were almost always killed; unlike on the Steppe, the incorporation of formerly hostile warriors doesn’t seem to have been considered possible (though one wonders if this would have become cultural practice given enough time; both McGinnis and Secoy note how the increasing lethality of warfare post-gun/horse led to slow population decline overall, which may, had the system run without outside interference long enough, led to the emergence of norms more closely resembling the Eurasian Steppe. We should keep in mind that the Eurasian horse-system had many centuries to sort itself out, whereas the North American horse-system was essentially strangled in its crib).

Of course, taken together with the previous discussion of territorial warfare, we can see that all of these raids have a double purpose: they both aim to acquire resources (horses, sheep, humans) and at the same time inflict damage on an opponent with the long-term goal of forcing that enemy to move further away, opening their pastures or hunting grounds for exploitation by the victorious tribe. Thus in the long-term, each successful raid is intended to build a sense of threat which eventually results in territorial gains (though in cases of real power asymmetry, the long term could come very rapidly; people aren’t stupid and if you are being raided by a clearly superior opponent, you are likely to move on before you lose everything of value).

Squaring the ugly reality of nomadic raiding with Martin’s depiction is tricky. On the one hand, a raid in which exceptional victory results in enemy women and children taken captive and fit adult males slain fits within either the Great Plains Native American or Steppe nomad military tradition. On the other hand, the immediate declaration by Drogo’s men that female captives taken this way are not marriageable (AGoT, 559; the idea is treated as laughable) and the killing of all of the very valuable livestock (which, even if the Dothraki are not herdsmen, these animals could be eaten, or quite easily driven to a place where they could be sold or traded for other resources, like metalwork) suggests that Martin has not understood why those raids happened. Instead, it seems like his imagination is only able to view these raids from the perspective of the settled people on the receiving end.

Instead, Martin’s understanding of Native American warfare seems not conditioned by any actual Native Americans, but rather by Hollywood depictions of Native Americans during the Hollywood ‘Golden Age’ which were in turn conditioned by sensational accounts of Western settlers who themselves didn’t understand how Native American warfare worked on the Great Plains. As we will see, the Game of Thrones showrunners took that unfortunate subtext when making the show itself, and turned it into actual text.

The Preposterous Tactics of the Dothraki

We do not see the Dothraki engage in large-scale warfare in the books; we see the aftermath of such fighting (AGoT, 555ff) or it occurs ‘off-screen’ (ASoS, 487), but we do not see it. The closest we get is Jorah’s description of them, that they are “utterly fearless…[they] fire from horseback, charging or retreating, it makes no matter, they are full as deadly…and there are so many of them” AGoT, 325-6). Evidently they also scream on the attack, since their warriors are repeatedly called ‘screamers.’

As a description, it is hard for this to be very much wrong because it is so very vague, but the attentive reader will note that Jorah’s assertion that there are ‘so many’ must be wrong for either Eurasian Steppe Nomads or Great Plains Native Americans, both of whom were routinely outnumbered by settled enemies, often dramatically so. Let’s put a pin in that, though, because of course while Martin gives only vague description of Dothraki warfare, the show, Game of Thrones, shows it to us on screen quite vividly.

As expert horsemen who literally spend their whole lives riding, naturally the Dothraki are completely incapable of keeping any kind of formation.
In contrast the Mongols could coordinate long-distance column movements converted into envelopments well enough to use it as a hunting system, for a hunt style they called the Nerge (which probably informed Mongol envelopment tactics as well).

We see a bit of Dothraki warfare in S6E9 when Daenerys’ Dothraki charge down the Sons of the Harpy at Mereen, but the really sustained look at how they fight has to wait for S7E4 and the Loot Train Battle and S8E3 and the Battle of Winterfell, both of which, happily, we have already discussed! In all three cases, the Dothraki do exactly the same thing. They charge, in a pell-mell rush, while giving high-pitched war-calls. While some of the Dothraki may fire arrows on the approach (they have them stand up to do this, which is not how actual Mongols or Native Americans fired from horseback; it looks cool and is stupid, like most of Game of Thrones season 7 and 8), they otherwise charge directly into contact and begin fighting from horseback with their arakhs as the primary weapon.

This is not how horse-borne nomads fought.

As we’ve discussed repeatedly before, the key weapon for Steppe nomads was the bow, shot from horseback at high speed (on this, note May, “The Training of an Inner Asian Nomad Army” JMH 70 (2006) and Mongol Art of War (2007)). Thus the crucial maneuver was the caracole, where the rider approaches the target at high speed, firing arrows as he goes, before making an abrupt turn (it is actually the turn that is technically called a caracole, but the whole tactic goes by this name) and retreating, before trying again. Pulling this tactic off en masse required a great deal of both individual skill at horsemanship and archery, but also quite a lot of group cohesion and coordination, since a collision of horses at speed is very likely to be fatal for everyone – humans and horses – involved.

Diagram of a caracole maneuver, following May (2007). The horse archers fire on the approach, but their lethality rises dramatically as they close in with the apex of the charge, before turning around. The ‘Parthian shot’ technique allows them to also fire on second half of the caracole. Multiple ranks perform the attack together, allowing for a continued barrage.

This tactic can then be repeated – charge and retreat, charge and retreat – until the psychological toll on the defender becomes too great and they either break and retreat or else charge out to try to catch ‘retreating’ nomads. In either case, it was at that moment when the Steppe nomads could press home and destroy the disorganized enemy. These tactics were brutally effective, but they were also a necessary casualty control measure. Shock combat – that is massed melee combat in close quarters – is simply far too lethal for low-population nomadic societies to sustain in the long-term on the regular (a hoplite battle might result normally in c. 10% casualties for instance (but note this discussion of that figure) – think of what that would mean in a society where 100% of adult males participate in each battle – you’d run out of men pretty quickly!).

And fascinatingly, we can actually see that calculus play out in North America, where the arrival of firearms, which suddenly make pitched ‘missile exchange’ battles (especially on foot) as lethal as shock combat (it seems notable that the introduction of musketry into Old World warfare did not come with a significant increase or decrease in battlefield lethality, at least until the rifled musket – on that, see B. Gibbs, The Destroying Angel (2019), but also noteE.J. Hess, The Rifle Musket in Civil War Combat: Reality and Myth (2008)), the pitched battle vanishes. It was simply too lethal to be a viable option in the long term for societies with low population density and very high military participation rates.

Instead, the raid came to dominate warfare on the Great Plains, with mass-casualty events generally being restricted to situations where a raiding party caught an enemy group unawares (McGinnis, op. cit., 45-6, 57-9). To be clear, that’s not to say the Great Plains Native Americans were peaceful, after all the goal of all of this raiding was to cause one of those rare mass-casualty surprise attacks and – as McGinnis notes again and again, warfare was part of the Plains Native American way of life, as the social status of males was directly and powerfully tied to success in war.

In short, the need to keep lethality relatively low is one of the most important factors which shaped nomadic horse-borne warfare, both on the Steppe and on the Great Plains. And here is where I think that even Martin’s description – which could, if read with friendly eyes, be taken as a description of the Steppe caracole described above – falls short: the Dothraki are dangerous because they are so many. But actual nomadic warfare was fundamentally conditioned by the shortage of men created by the low population density of the Steppe or the Great Plains. This weakness could be somewhat made up for by making every male into a warrior, but only if casualty rates remained low. A war of attrition with settled peoples would wear the nomads out quickly, which is why such attritional warfare was avoided (unless you are the Mongols, who use the sedentary armies of conquered states, notably using the armies of Northern China to conquer Southern China; that said, Drogo is clearly not Chinggis Khan or any such sort of Khal-of-Khals).

Stagecoach (1939)

So where does this model of warfare come from? Well, when it comes to the show, we needn’t actually look far, because the creators tell us. The director of the episode, Matt Shakman, noted in an interview that his primary reference for the Dothraki charge was John Ford’s Apache attack in his 1939 film Stagecoach (you can see the scene he means here). And in the S7 special feature, “Anatomy of a Scene: The Loot Train Attack,” David Benioff notes that the charge “definitely got a bit of that western feel” while VFX producer Steve Kullback says, of the battle, it’s “sort of like Cowboys and Indians.”

In Stagecoach (1939), the Apache aren’t a real humanized culture, but an elemental force of destruction. Their charge at the titular stagecoach is essentially mad and heedless of all losses (in the same featurette, Camilla Naprous, Game of Throne‘s horse master, describes the Dothraki as “they’re just these absolute mad men on horses,” in case you thought that connection was only subtext). The position of ‘Indians’ as particularly ‘rapey’ is also explicit in Stagecoach, where the one of the white male defenders of the coach saves his last bullet to spare the one woman, Mrs. Mallory, from being captured and raped by the approaching cavalry (the concern about white women being raped by non-white men being a paramount fixation of early American film; see also The Birth of a Nation (1915); or, you know, don’t.) And the tactics (or lack thereof) of the Dothraki, charging madly forward with no order or concern for safety, also map neatly on to Stagecoach‘s Apache attack (and not on to actual Apache attacks).

I don’t think this lazy use of old Western tropes is limited to merely the show, however. Having written this far, I find myself convinced that there is a longer article or perhaps a video-essay waiting to be written by a different sort of scholar than myself – that is, a film historian – on how Martin’s depiction of the Dothraki and their world is fundamentally rooted in the racist tropes of the Hollywood Western and its portrayal of Native Americans in a frontier environment where, as Sergio Leone put it, “life has no value.” Quite a lot of parallels with Martin’s Dothraki emerge after even a brief overview of the representation of Native Americans in film. The emphasis on taking captives (especially white women) to no apparent purpose besides sexual violence, the distinctive ‘screaming’ of Dothraki warfare (which, yes, Native Americans used a range of intimidating war cries, but so did basically everyone else in the pre-modern world, so why are the Dothraki the only ones who do it in Westeros?), it’s lack of tactics or order, and – as we’ve discussed already – the grossly simplified form of dress all seem to have their roots in racist Hollywood depictions of Native Americans. The Dothraki Sea is, essentially a ‘Cavalry and Indian Story‘ with the cavalry removed.

That is not a pure creation of Benioff and Weiss. The show simply takes that subtext and makes it text.

Conclusions

That makes it a good time, here at the end, to take stock. As I’ve noted in each of these posts, the fundamental claim we are evaluating here is this one, made baldly by George R.R. Martin:

The Dothraki were actually fashioned as an amalgam of a number of steppe and plains cultures… Mongols and Huns, certainly, but also Alans, Sioux, Cheyenne, and various other Amerindian tribes… seasoned with a dash of pure fantasy.

We may, I think, now safely dismiss this statement as false. What we have found is that the Dothraki do not meaningfully mirror either Steppe or Plains cultures. They do not mirror them in dress, nor in systems of subsistence, nor in diet, nor in housing, nor in music, nor in art, nor in social structures, nor in leadership structures, nor in family structures, nor in demographics, nor in economics, nor in trade practices, nor in laws, nor in marriage customs, nor in attitudes towards violence, nor in weapons, nor in armor, nor in strategic way of war, nor in battle tactics.

We might say he has added ‘dashes’ of pure fantasy until the ‘dash’ is the entire soup, but the truth is clearly the reverse: Martin has sprinkled a little bit of water on a barrel of salt and called it just a dash of salt. There is no historical root source here, but instead pure fantasy which – because racist stereotypes sometimes connect, in thin and useless ways, to actual history – occasionally, in broken-clock fashion, manages to resemble the real thing.

It seems as though the best we might say of what Martin has right is that these are people who are nomads that ride horses and occasionally shoot bows. The rest – which as you can see from the list above there, is the overwhelming majority – has functionally no connection to the actual historical people. And stunningly, somehow, the show – despite its absolutely massive budget, despite the legions of scrutiny and oversight such a massive venture brings – somehow is even worse, while being just as explicit in tying its bald collection of 1930s racist stereotypes to real people who really exist today.

Camilla Naprous describing the Dothraki in a Game of Thrones special feature for the Battle of the Loot Train.

Instead, the primary inspiration for George R.R. Martin’s Dothraki seems to come from deeply flawed Hollywood depictions of noamdic peoples, rather than any real knowledge about the peoples themselves. The Dothraki are not an amalgam of the Sioux or the Mongols, but rather an amalgam of Stagecoach (1939) and The Conqueror (1956). When it comes to the major attributes of the Dothraki – their singular focus on violent, especially sexual violence, their lack of art or expression, their position as a culture we primarily see ‘from the outside’ as almost uniformly brutal (and in need of literally the whitest of all women to tame and reform it) – what we see is not reflected in the historical people at all but is absolutely of a piece with this Hollywood legacy.

But Martin has done more damage than simply watching The Mongols (1961) would today. He has taken those old, inaccurate, racially tinged stereotypes and repackaged them, with an extra dash of contemporary cynicism to lend them the feeling of ‘reality’ and then used his reputation as a writer of more historically grounded fantasy (a reputation, I think we may say at this point, which ought to be discarded; Martin is an engaging writer but a poor historian) to give those old stereotypes the air of ‘real history’ and how things ‘really were.’ And so, just as Westeros became the vision of the Middle Ages that inhabits the mind of so many people (including quite a few of my students), the Dothraki become the mental model for the Generic Nomad: brutal, sexually violent, uncreative, unartistic, uncivilized.

And as I noted at the beginning of this series, Martin’s fans have understood that framing perfectly well. The argument given by both the creators themselves, often parroted by fans and even repeated by journalists is that A Song of Ice and Fire‘s historical basis is both a strike in favor of the book because they present a ‘more real’ vision of the past but also a flawless defense against any qualms anyone might have over the way that the fiction presents violence (especially its voyeuristic take on sexual violence) or its cultures. No doubt part of you are tired of seeing that same ‘amalgam’ quote over and over again at the beginning of every single one of these essays, but I did that for a reason, because it was essential to note that this assertion is not merely part of the subtext of how Martin presents his work (although it is that too), but part of the actual text of his promotion of his work.

And it is a lie. And I want to be clear here, it is not a misunderstanding. It is not a regrettable implication. It is not an unfortunate spot blind-spot of ignorance. It is a lie, made repeatedly, now by many people in both the promotion of the books and the show who ought to have known better. And it is a lie that has been believed by millions of fans.

One thing that I hope is clear from this treatment is just how trivial the amount of research I’ve done here was. Certainly, it helped that I was familiar with Steppe nomads already and that I knew who to ask to be pointed in the direction of information. Nevertheless, everything I’ve cited here is available in English and it is all relatively affordable (I actually own all of the books cited here; thanks to my Patrons for making that possible, especially since getting materials from the library is slower in the days of COVID-19; nevertheless, the point here is that they are not obscure tomes). Much of it – Ratchnevsky on Chinggis Khan, Secoy and McGinnis on Great Plains warfare – were already available well before the 1996 publication of A Game of Thrones. 1996 was not some wasteland of ignorance that might have made it impossible for Martin to get good information! For an easy sense of what a dedicated amateur with film connections might have learned in 1996, you could simply watch Ken Burn’s The West, which came out the same year. I am not asking Martin to become a historian (though I am asking him to stop representing himself as something like one), I am asking him to read a historian.

Instead of doing that basic amount of research, or simply saying that the peoples of Essos were made up cultures unconnected with the real thing, Martin and the vast promotional apparatus at HBO opted to lie about some real cultures and then to put hundreds of millions of dollars into promoting that lie.

And I want to be clear, these are real people! I know, depending on where you live, ‘Mongols’ and ‘Sioux’ and ‘Cheyenne’ may feel as distant and fanciful as ‘Rohirrim’ or ‘Hobbits’ or else they may feel like ‘long-lost’ peoples. But these were real people, whose real descendants are alive today. And almost all of them face discrimination and abuse, sometimes informally, sometimes through state action, often as a result of these very lingering racist stereotypes.

Via Wikipedia, a ger district in Ulaanbaatar in modern Mongolia.

In that context, declaring that the Dothraki really do reflect the real world (I cannot stress that enough) cultures of the Plains Native Americans or Eurasian Steppe Nomads is not merely a lie, but it is an irresponsible lie that can do real harm to real people in the real world. And that irresponsible lie has been accepted by Martin’s fans; he has done a grave disservice to his own fans by lying to them in this way. And of course the worst of it is that the lie – backed by the vast apparatus that is HBO prestige television – will have more reach and more enduring influence than this or any number of historical ‘debunking’ essays. It will befuddle the valiant efforts of teachers in their classrooms (and yes, I frequently encounter students hindered by bad pop-pseudo-history they believe to be true; it is often devilishly hard to get students to leave those preconceptions behind), it will plague efforts to educate the public about these cultures of their histories. And it will probably, in the long run, hurt the real descendants of nomads.

But this is exactly why I think it is important for historians to engage with the culture and to engage with depictions like this. Because these lies have consequences and someone ought to at least try to tell the truth. With luck, even with my only rudimentary knowledge, I have done some of that here, by presenting a bit more of the richness and variety of historical (and in some cases, present-day) horse-borne nomadic life, in both North America and Eurasia.

Because there is and was a lot more to nomads than just ‘that Dothraki horde.’

Like this:

Like Loading...

Miscellanea: The War in Iran – A Collection of Unmitigated Pedantry

Wed, 25 Mar 2026 06:21:44 +0000

This post is a set of my observations on the current war in Iran and my thoughts on the broader strategic implications. I am not, of course, an expert on the region nor do I have access to any special information, so I am going to treat that all with a high degree of uncertainty. But I am a scholar of military history with a fair bit of training and experience in thinking about strategic problems, ancient and modern; it is this ‘guy that analyzes strategy’ focus that I want to bring to this.

I am doing this post outside of the normal Friday order because it is an unusual topic and I want to keep making it clear that even as world events continue to happen – as they must – I do not want this blog to turn into a politics newsletter. I simply haven’t had the time to polish and condense these thoughts for other publication – the hard work of much writing is turning 3,500 words (or 7,500, as it turns out) of thoughts into 1,500 words of a think piece – but I need to get them out of my head and on to the page before it burns out of the back of my head. That said, this post is going to be unavoidably ‘political,’ because as a citizen of the United States, commenting on the war means making a statement about the President who unilaterally and illegally launched it without much public debate and without consulting Congress.

And this war is dumb as hell.

I am going to spend the next however many words working through what I think are the strategic implications of where we are, but that is my broad thesis: for the United States this war was an unwise gamble on extremely long odds; the gamble (that the regime would collapse swiftly) has already failed and as a result locked in essentially nothing but negative outcomes. Even with the regime were to collapse in the coming weeks or suddenly sue for peace, every likely outcome leaves the United States in a meaningfully worse strategic position than when it started.

Now, before we go forward, I want to clarify a few things. First, none of this is a defense of the Iranian regime, which is odious. That said, there are many odious regimes in the world and we do not go to war with all of them. Second, this is a post fundamentally about American strategy or the lack thereof and thus not a post about Israeli strategy. For what it is worth, my view is that Benjamin Netanyahu has is playing an extremely short game because it benefits him politically and personally to do so and there is a significant (but by no means certain) chance that Israel will come to regret the decision to encourage this war. I’ll touch on some of that, but it isn’t my focus. Likewise, this is not a post about the strategy of the Gulf states, who – as is often the sad fate of small states – find their fate largely in the hands of larger powers. Finally, we should keep in mind that this isn’t an academic exercise: many, many people will suffer because of these decisions, both as victims of the violence in the region but also as a consequent of the economic ripples.

But that’s enough introduction. What I want to discuss here is first the extremely unwise gamble that the administration took and then the trap that it now finds itself in, from which there is no comfortable escape.

The Situation

We need to start by establishing some basic facts about Iran, as a country.

First, Iran is a large country. It has a population just over 90 million (somewhat more than Germany, about the same as Turkey), and a land area over more than 600,000 square miles (more than four times the size of Germany). Put another way Iran is more than twice as large as Texas, with roughly three times the population.

More relevantly for us, Iran is 3.5 times larger than Iraq and roughly twice the population. That’s a handy comparison because we know what it took to invade and then hold Iraq: coalition forces peaked at half a million deployed personnel during the invasion. Iran is bigger in every way and so would demand a larger army and thus an absolutely enormous investment of troops, money and fundamentally lives in order to subdue.

Via Wikipedia, a map of Iran. This is a very big country. It also has a lot of very challenging terrain: lots of very arid areas, lots of high mountains and plateaus. It is a hard country to invade and a harder country to occupy.

In practice, given that Iran did not and never has posed an existential threat to the United States (Iran aspires to be the kind of nuclear threat North Korea is and can only vaguely dream of being the kind of conventional threat that Russia is), that meant that a ground invasion of Iran was functionally impossible. While the United States had the raw resources to do it, the political will simply wasn’t there and was unlikely to ever be there.

Equally important, Iran was not a major strategic priority. This is something that in a lot of American policy discourse – especially but not exclusively on the right – gets lost because Iran is an ‘enemy’ (and to be clear, the Iranian regime is an enemy; they attack American interests and Americans regularly) and everyone likes to posture against the enemy. But the Middle East is a region composed primarily of poor, strategically unimportant countries. Please understand me: the people in these countries are not unimportant, but as a matter of national strategy, some places are more important than others. Chad is not an area of vital security interest to the United States, whereas Taiwan (which makes our semiconductors) is and we all know it.

Neither is the Middle East. The entire region has exactly two strategic concerns of note: the Suez Canal (and connected Red Sea shipping system) and the oil production in the Persian Gulf and the shipping system used to export it. So long as these two arteries remained open the region does not matter very much to the United States. None of the region’s powers are more than regional powers (and mostly unimpressive ones at that), none of them can project power out of the region and none of them are the sort of dynamic, growing economies likely to do so in the future. The rich oil monarchies are too small in terms of population and the populous countries too poor.

In short then, Iran is very big and not very important, which means it would both be very expensive to do anything truly permanent about the Iranian regime and at the same time it would be impossible to sell that expense to the American people as being required or justified or necessary. So successive American presidents responded accordingly: they tried to keep a ‘lid’ on Iran at the lowest possible cost. The eventual triumph of this approach was the flawed but useful JCPOA (the ‘Iran deal’) in which Iran in exchange for sanctions relief swore off the pursuit of nuclear weapons (with inspections to verify), nuclear proliferation representing the main serious threat Iran could pose. So long as Iran remained non-nuclear, it could be contained and the threat to American interests, while not zero, could be kept minimal.

That deal was not perfect, I must stress: it essentially gave Iran carte blanche to reinforce its network of proxies across the region, which was robustly bad for Israel and mildly bad for the United States, but since the alternative was – as we’ll see – global economic disruption and the prospect of a large-scale war which would always be far more expensive than the alternatives, it was perhaps the best deal that could have been had. For what it is worth, my own view is that the Obama administration ‘overpaid’ for the concessions of the Iran deal, but the payment having been made, they were worth keeping. Trump scrapped them in 2017 in exchange for exactly nothing, which put us on the course for this outcome (as more than a few people pointed out at the time).

But that was the situation: Iran was big and hostile, but relatively unimportant. The United States is much stronger than Iran, but relatively uninterested in the region apart from the uninterrupted flow of natural gas, oil and other products from the Gulf (note: the one thing this war compromised – the war with Iran has cut off the only thing in this region of strategic importance, compromised the only thing that mattered at the outset), whereas Iran was wholly interested in the region because it lives there. The whole thing was the kind of uncomfortable frontier arrangement powerful states have always had to make because they have many security concerns, whereas regional powers have fewer, more intense focuses.

Which leads us to

The Gamble

The current war is best understood as the product of a fairly extreme gamble, although it is unclear to me if the current administration understood they were throwing the dice in June of 2025 rather than this year. As we’re going to see, this was not a super-well-planned-out affair.

The gamble was this: that the Iranian regime was weak enough that a solid blow, delivered primarily from the air, picking off key leaders, could cause it to collapse. For the United States, the hope seems to have been that a transition could then be managed to leaders perhaps associated with the regime but who would be significantly more pliant, along the lines of the regime change operation performed in Venezuela that put Delcy Rodriguez in power. By contrast, Israel seems to have been content to simply collapse the Iranian regime and replace it with nothing. That outcome would be – as we’ll see – robustly bad for a huge range of regional and global actors, including the United States, and it is not at all clear to me that the current administration understood how deeply their interests and Israel’s diverged here.

In any case, this gamble was never very likely to pay off for reasons we have actually already discussed. The Islamic Republic of Iran is not a personalist regime where the death of a single leader or even a group of leaders is likely to cause collapse: it is an institutional regime where the core centers of power (like the Iran Revolutionary Guards Corps or IRGC) are ‘bought in’ from the bottom to the top because the regime allows them access to disproportionate resources and power. Consequently if you blow up the leader, they will simply pick another one – in this case they picked the previous leader’s son, so the net effect of the regime change effort was to replace Supreme Leader Khamenei with Supreme Leader Khamenei…Jr.

But power in the Iranian regime isn’t wielded by the Supreme Leader alone either: the guardian council has power, the council of experts that select the Supreme Leader have power, the IRGC has power, the regular military has some power (but less than the IRGC), the elected government has some power (but less than the IRGC or the guardian council) and on and on. These sorts of governments can collapse, but not often. It certainly did not help that the United States had stood idle while the regime slaughtered tens of thousands of its opponents, before making the attempt, but I honestly do not think the attempt would have worked before.

The gamble here was that because the regime would simply collapse on cue, the United States could remove Iran’s regional threat without having to commit to a major military operation that might span weeks, disrupt global energy supplies, expand over the region, cost $200 billion dollars and potentially require ground operations. Because everyone knew that result was worse than the status quo and it would thus be really foolish to do that.

As you can tell, I think this was a bad gamble: it was very unlikely to succeed but instead always very likely to result in a significantly worse strategic situation for the United States, but only after it killed thousands of people unnecessarily. If you do a war where thousands of people die and billions of dollars are spent only to end up back where you started that is losing; if you end up worse than where you started, well, that is worse.

The problem is that once the gamble was made, once the dice were cast, the Trump administration would be effectively giving up control over much of what followed.

And if administration statements are to be believed, that decision was made, without knowing it, in June of last year. Administration officials, most notably Marco Rubio, have claimed that the decision was made to attempt this regime change gamble in part because they were aware that Israel was about to launch a series of decapitation strikes and they assessed – correctly, I suspect – that the ‘blowback’ would hit American assets (and energy production) in the region even if the United States did nothing. Essentially, Iran would assume that the United States was ‘in’ on the attack.

That is notable because Iran did not assume that immediately during the Twelve-Day War in 2025. Indeed, Iran did not treat the United States as a real co-belligerent even as American aircraft were actively intercepting Iranian missiles aimed at Israel. And then the United States executed a ‘bolt from the blue’ surprise attack on Iran’s nuclear facilities on June 22, 2025, catching Iran (which had been attempting to negotiate with the United States) by surprise.

The problem with that strike is that attacking in that way, at that time, meant that Iran would have to read any future attacks by Israel as likely also involving attacks by the United States. Remember, the fellow getting bombed does not get to carefully inspect the flag painted on the bomber: stuff blows up and to some degree the party being attacked has to rapidly guess who is attacking them. We’ve seen this play out repeatedly over the last several weeks where things explode in Iran and there is initially confusion over if the United States or Israel bombed them. But in the confusion of an initial air attack, Iran’s own retaliatory capability could not sit idle, waiting to be destroyed by overwhelming US airpower: it is a ‘wasting’ use-it-or-lose-it asset.

So Iran would now have to assume that an Israeli air attack was also likely an American air attack. It was hardly an insane assumption – evidently according to the Secretary of State, American intelligence made the exact same assessment.

But the result was that by bombing the Iranian nuclear facilities in June of 2025, the Trump administration created a situation where merely by launching a renewed air campaign on Iran, Israel could force the United States into a war with Iran at any time.

It should go without saying that creating the conditions where the sometimes unpredictable junior partner in a security relationship can unilaterally bring the senior partner into a major conflict is an enormous strategic error, precisely because it means you end up in a war when it is in the junior partner’s interests to do so even if it is not in the senior partner’s interests to do so.

Which is the case here. Because…

The Trap

Once started, a major regional war with Iran was always likely to be something of a ‘trap,’ – not in the sense of an ambush laid by Iran – but in the sense of a situation that, once entered, cannot be easily left or reversed.

The trap, of course, is the Strait of Hormuz and the broader Persian Gulf. The issue is that an enormous proportion of the world’s shipping, particularly energy (oil, liquid natural gas) and fertilizer components (urea) passes through this body of water. The Gulf is narrow along its whole length, extremely narrow in the Strait and bordered by Iran on its northern shore along its entire length. Iran can thus threaten the whole thing and can do so with cheap, easy to conceal, easy to manufacture systems.

And the scale here is significant. 25% of the world’s oil (refined and crude), 20% of its liquid natural gas and around 20% of the world’s fertilizer passes through the Strait of Hormuz which links the Persian Gulf to the Indian Ocean. Any of those figures would be enough for a major disruption to trigger huge economic ripples. And even worse there are only very limited, very insufficient alternative transport options. Some Saudi oil (about half) can move via pipeline to the Red Sea and some Emirati oil can move via pipeline to Fujairah outside of the Strait, but well over half of the oil and effectively all of the natural gas and fertilizer ingredients are trapped if ships cannot navigate the strait safely.

And here we come back to what Clausewitz calls the political object (drink!). Even something like a 50% reduction in shipping in the Gulf, were it to persist long term, would create strong global economic headwinds which would in turn arrive in the United States in the form of high energy prices and a general ‘supply shock’ that has, historically at least, not been politically survivable for the party in power.

And so that is the trap. While the United States can exchange tit-for-tat strikes with Iran without triggering an escalation spiral, once you try to collapse the regime, the members of the regime (who are making the decisions, not, alas, the Iranian people) have no reason to back down and indeed must try to reestablish deterrence. These are men who are almost certainly dead or poor-in-exile if the regime collapses. Moreover the entire raison d’être of this regime is resistance to Israel and the United States: passively accepting a massive decapitation attack and not responding would fatally undermine the regime’s legitimacy with its own supporters, leading right back to the ‘dead-or-poor-and-exiled’ problem.

Iran would have to respond and thus would have to try to find a way to inflict ‘pain’ on the United States to force the United States to back off. But whereas Israel is in reach of some Iranian weapons, the United States is not. Iran would thus need a ‘lever’ closer to home which could inflict costs on the United States. For – and I must stress this – for forty years everyone has known this was the strait. This is not a new discovery, we did this before in the 1980s. “If the regime is threatened, Iran will try to close the strait to exert pressure” is perhaps one of the most established strategic considerations in the region. We all knew this.

But the trap here is two sided: once the strait was effectively closed, the United States could not back off out of the war without suffering its own costs. Doing so, for one, would be an admission of defeat, politically damaging at home. Strategically, it would affirm Iran’s control over the strait, which would be a significantly worse outcome than not having done the war in the first place. And simply backing off might not fully return shipping flows: why should Iran care if the Gulf states can export their oil? An Iran that fully controls the strait, that had demonstrated it could exclude the United States might intentionally throttle everyone else’s oil – even just a bit – to get higher prices for its own or to exert leverage.

So once the strait was closed, the United States could not leave until it was reopened, or at least there was some prospect of doing so.

The result is a fairly classic escalation trap: once the conflict starts, it is extremely costly for either side to ever back down, which ensures that the conflict continues long past it being in the interests of either party. Every day this war goes on make both the United States and Iran weaker, poorer and less secure but it is very hard for either side to back down because there are huge costs connected to being the party that backs down. So both sides ‘escalate to de-escalate’ (this phrase is generally as foolish as it sounds), intensifying the conflict in an effort to hit hard enough to force the other guy to blink first. But since neither party can back down unilaterally and survive politically, there’s practically no amount of pain that can force them to do so.

Under these conditions, both sides might seek a purely military solution: remove the ability of your opponent to do harm in order to create the space to declare victory and deescalate. Such solutions are elusive. Iran simply has no real way of meaningfully diminishing American offensive power: they cannot strike the airfields, sink the carriers or reliably shoot down the planes (they have, as of this writing, managed to damage just one aircraft).

For the United States, a purely military solution is notionally possible: you could invade. But as noted, Iran is very, very big and has a large population, so a full-scale invasion would be an enormous undertaking, larger than any US military operation since the Second World War. Needless to say, the political will for this does not exist. But a ‘targeted’ ground operation against Iran’s ability to interdict the strait is also hard to concieve. Since Iran could launch underwater drones or one-way aerial attack drones from anywhere along the northern shore the United States would have to occupy many thousands of square miles to prevent this and of course then the ground troops doing that occupying would simply become the target for drones, mortars, artillery, IEDs and so on instead.

One can never know how well prepared an enemy is for something, but assuming the Iranians are even a little bit prepared for ground operations, any American force deployed on Iranian soil would end up eating Shahed and FPV drones – the sort we’ve seen in Ukraine – all day, every day.

Meanwhile escort operations in the strait itself are also deeply unpromising. For one, it would require many more ships, because the normal traffic through the strait is so large and because escorts would be required throughout the entire Gulf (unlike the Red Sea crisis, where the ‘zone’ of Houthi attacks was contained to only the southern part of the Red Sea). But the other problem is that Iran possesses modern anti-ship missiles (AShMs) in significant quantity and American escort ships (almost certainly Arleigh Burke-class destroyers) would be vulnerable escorting slow tankers in the constrained waters of the strait.

It isn’t even hard to imagine what the attack would look like: essentially a larger, more complex version of the attack that sunk the Moskva, to account for the Arleigh Burke’s better air defense. Iran would pick their moment (probably not the first transit) and try to distract the Burke, perhaps with a volley of cheap Shahed-type drones against a natural gas tanker, before attempting to ambush the Burke with a volley of AShMs, probably from the opposite direction. The aim would be to create just enough confusion that one AShM slipped through, which is all it might take to leave a $2.2bn destroyer with three hundred American service members on board disabled and vulnerable in the strait. Throw in speed-boats, underwater drones, naval mines, fishing boats pretending to be threats and so on to maximize confusion and the odds that one of perhaps half a dozen AShMs slips through.

And if I can reason this out, Iran – which has been planning for this exact thing for forty years certainly can. Which is why the navy is not eager to run escort.

But without escorts or an end to the conflict, shipping in the Gulf is not going to return to normal. Container ships are big and hard to sink but easy to damage. But while crude oil tankers are hard to set fire to, tankers carrying refined petroleum products are quite easy to set fire to, as we’ve seen, while tankers of liquid natural gas (LNG carriers) are essentially floating bombs.

The result is that right now it seems that the only ships moving through the strait are those Iran permits and they appear to have a checkpoint system, turning away ships they do not approve of. A military solution this problem is concievable, but extremely difficult to implement practically, requiring either a massive invasion of Iran’s coastline or an enormous sea escort operation. It seems more likely in both cases that the stoppage will continue until Iran decides it should stop. The good news on that front is that Iran benefits from the export of oil from the Gulf too, but the bad news is that while they are permitting some traffic, precisely because high energy prices are their only lever to make the United States and Israel stop killing them, they are unlikely to approve the transit of the kinds of numbers of ships which would allow energy markets to stabilize.

Just as a measure here, as I write this apparently over the last three days or so Iran has let some twenty ships through their checkpoint, charging fees apparently to do so. That may sound like a lot, but it is a quantity that, compared to the normal operation of the strait, is indistinguishable from zero. The Strait of Hormuz normally sees around 120 transits per day (including both directions). That scale should both explain why five or six ships a day paying Iran to transit is not going to really impact this equation – that’s still something like a 95% reduction in traffic (and all of the Iran-approved transits are outbound, I think) – but also why a solution like ‘just do escorts’ is so hard. Whatever navies attempted an escort solution would need to escort a hundred ships a day, with every ship being vulnerable at every moment from when it entered the Strait to when it docked for loading or offloading to its entire departure route. All along the entire Gulf coastline. All the time.

Likewise, even extremely punishing bombings of Iranian land-based facilities are unlikely to wholly remove their ability to throw enough threat into the Strait that traffic remains massively reduced. Sure some ship owners will pay Iran and others will take the risk, but if traffic remains down 90% or just 50% that is still a massive, global energy disruption. And we’ve seen with the campaign against the Houthis just how hard it is with airstrikes to compromise these capabilities: the United States spent more than a year hammering the Houthis and was never able to fully remove their attack capabilities. Cargo ships are too vulnerable and the weapons with which to attack them too cheap and too easy to hide.

There is a very real risk that this conflict will end with Iran as the de facto master of the Strait of Hormuz and the Persian Gulf, having demonstrated that no one can stop them from determining by force which ships pass and which ships cannot. That would, in fact, be a significant strategic victory for Iran and an enormous strategic defeat for the United States.

Peace Negotiations?

Which brings us to the question of strategic outcomes. As the above has made clear, I think the Trump administration erred spectacularly in starting this war. It appears as though, in part pressured by Israel, but mostly based on their own decisions (motivated, it sure seems, by the ease of the Venezuela regime-change) they decided to go ahead on the hopeful assumption the regime would collapse and as a result did not plan for the most likely outcome (large war, strait closure), despite this being the scenario that political leadership (Trump, Hegseth, Rubio) were warned was most likely.

The administration now appears to be trying to extricate itself from the problem has created, but as I write this, is currently still stuck in the ‘trap’ above. Now this is a fast moving topic, so by the time you actually read this the war well could have ended in a ceasefire (permanent or temporary) or intensified and expanded. Who knows! As I am writing the Trump administration claims that they are very near a negotiated ceasefire, while the Iranian regime claims they have rejected both of the United States’ interlocutors as unsuitable (‘backstabbing’ negotiators), while reporting suggests Israel may feel it in their interests to blow up any deal if the terms are too favorable to Iran.

That is a lot of uncertainty! But I think we can look at some outcomes here both in terms of what was militarily achieved, what the consequences of a ‘deal’ might be and what the consequences of not having a deal might be.

The Trump administration has offered a bewildering range of proposed objectives for this war, but I think it is fair to say the major strategic objectives have not been achieved. Initially, the stated objective was regime change or at least regime collapse; neither has occurred. The regime very much still survives and if the war ends soon it seems very plausible that the regime – able to say that it fought the United States and made the American president sue for peace – will emerge stronger, domestically (albeit with a lot of damage to fix and many political problems that are currently ‘on pause’ coming ‘un-paused’). The other core American strategic interest here is Iran’s nuclear program, the core of which is Iran’s supply of roughly 500kg of highly enriched uranium; no effort appears to have been made to recover or destroy this material and it remains in Iranian hands. Actually destroying (dispersing, really) or seizing this material by military force would be an extremely difficult operation with a very high risk of failure, since the HEU is underground buried in facilities (mostly Isfahan) in the center of the country. Any sort of special forces operation would thus run the risk of being surrounded and outnumbered very fast, even with ample air support, while trying to extract half a ton of uranium stored in gas form in heavy storage cylinders.

When the United States did this in Kazakhstan, removing about 600kg (so roughly the same amount) it required the team to spend 12 hours a day every day for a month to remove it, using multiple heavy cargo planes. And that facility was neither defended, nor buried under rubble.

Subsequently, administration aims seem to have retreated mostly to ‘fixing the mess we made:’ getting Iran to stop shooting and getting the Strait of Hormuz reopened and the ships moving again. They do seem to be asking for quite a bit more at the peace table, but the record of countries winning big concessions at the peace table which they not only haven’t secured militarily but do not appear able to do so is pretty slim.

Now it is possible that Iran blinks and takes a deal sooner rather than later. But I don’t think it is likely. And the simple reason is that Iran probably feels like it needs to reestablish deterrence. This is the second sudden bombing campaign the country has suffered in as many years – they do not want there to be a third next year and a fourth the year after that. But promises not to bomb them don’t mean a whole lot: establishing deterrence here means inflicting quite a lot of pain. In practice, if Iran wants future presidents not to repeat this war, the precedent they want to set is “attacking Iran is a presidency-ending mistake.” And to do that, well, they need to end a presidency or at least make clear they could have done.

Iran is thus going to very much want a deal that says ‘America blinked’ on the tin, which probably means at least some remaining nuclear program, a de facto Iranian veto on traffic in the strait and significant sanctions relief, along with formal paper promises of no more air strikes. That’s going to be a hard negotiating position to bridge, especially because Iran can ‘tough it out’ through quite a lot of bombing.

And I do want to stress that. There is a frequent mistake, often from folks who deal in economics, to assume that countries will give up on wars when the economics turn bad. But countries are often very willing to throw good money after bad even on distant wars of choice. For wars close to home that are viewed as existential? Well, the ‘turnip winter‘ where Germans started eating food previous thought fit only for animals (a result of the British blockade) began in 1916. The war did not end in 1916. It did not end in 1917. It did not end until November, 1918. Food deprivation and starvation in Germany was real and significant and painful for years before the country considered surrender. Just because the war is painful for Iran does not mean the regime will cave quickly: so long as they believe the survival of the regime is at stake, they will fight on.

There is a great deal of ruin in a nation.

Strategic Implications

So my conclusion here is that the United States has not yet achieved very much in this war on a strategic level. Oh, tactically, the United States has blown up an awful lot of stuff and done so with very minimal casualties of its own. But countries do not go to war simply to have a war – well, stupid fascist countries do, which is part of why they tend to be quite bad at war – they go to war to achieve specific goals and end-states.

None of the major goals here – regime change, an end to Iran’s nuclear ambitions – have been achieved. If the war ends tomorrow in a ‘white peace,’ Iran will reconstitute its military and proxies and continue its nuclear program. It is in fact possible to display astounding military skill and yet, due to strategic incoherence, not accomplish anything.

So the true, strategic gains here for all of the tactical effectiveness displayed, are functionally nil. Well what did it cost?

Well, first and foremost, to date the lives of 13 American soldiers (290 more WIA), 24 Israelis (thousands more injured), at least a thousand civilian deaths across ‘neutral’ countries (Lebanon mostly, but deaths in Kuwait, Iraq, Qatar, Bahrain, Saudi Arabia, etc) and probably at least a thousand if not more Iranian civilians (plus Iranian military losses). The cost of operations for the United States is reportedly one to two billion dollars a day, which adds up pretty quickly to a decent chunk of change.

All of the military resources spent in this war are in turn not available for other, more important theaters, most obviously the Asia-Pacific (INDOPACOM), but of course equally a lot of these munitions could have been doing work in Ukraine as well. As wars tend to do, this one continues to suck in assets as it rumbles on, so the American commitment is growing, not shrinking. And on top of spent things like munitions and fuel, the strain on ships, air frames and service personnel is also a substantial cost: it turns out keeping a carrier almost constantly running from one self-inflicted crisis to the next for ten months is a bad idea.

You could argue these costs would be worthwhile it they resulted in the destruction of Iran’s nuclear program – again, the key element here is the HEU, which has not been destroyed – or of the Iranian regime. But neither of those things have been achieved on the battlefield, so this is a long ledger of costs set against…no gains. Again, it is not a ‘gain’ in war simply to bloody your enemy: you are supposed to achieve something in doing so.

The next side of this are the economic consequences. Oil and natural gas have risen in price dramatically, but if you are just watching the commodity ticker on the Wall Street Journal, you may be missing some things. When folks talk about oil prices, they generally do so via either $/bbl (West Texas Intermediate – WTI – one-month front-month futures) or BRN00 (Brent Crude Oil Continuous Contracts). These are futures contracts, meaning the price being set is not for a barrel of oil right now but for a barrel of oil in the future; we can elide the sticky differences between these two price sets and just note that generally the figure you see is for delivery in more-or-less one month’s time. Those prices have risen dramatically (close to doubled), but may not reflect the full economic impact here: as the ‘air bubble’ created by the sudden stop of oil shipments expands, physical here-right-now prices for oil are much higher in many parts of the world and still rising.

Essentially, the futures markets are still hedging on the idea that this war might end and normal trade might resume pretty soon, a position encouraged by the current administration, which claims it has been negotiating with Iran (Iran denied the claim). The tricky thing here is that this is a war between two governments – the Trump administration and the Iranian regime – which both have a clear record of lying a lot. The Trump administration has, for instance, repeatedly claimed a peace deal between Ukraine and Russia was imminent, and that war remains ongoing. The markets are thus forced to try and guess everyone’s actions and intentions from statements that are unreliable. Cards on the table, I think the markets are underestimating the likelihood that this conflict continues for some time. Notably, the United States is moving assets into theater – an MEU, elements of the 82 Airborne – which will take some time to arrive (two weeks for the MEU which is still about a week out as I write this) and set up for operations.

In either case, while I am not an expert on oil extraction or shipping, what I have seen folks who are experts on those things say is that the return of normal operations after this war will be very slow, often on the order of ‘every extra week of conflict adds a month to recovery’ (which was Sal Mercogliano’s rule of thumb in a recent video). If the war ends instantly, right now, ship owners will first have to determine that the strait is safe, then ships will have to arrive and begin loading to create space in storage to start up refineries to create space in storage to start up oil wells that have been ‘shut in,’ some of which may require quite a bit of doing to restart. Those ships in turn have to spend weeks sailing to the places that need these products, where some of the oil and LNG is likely to be used to refill stockpiles rather than immediately going out to consumers. For many products, refineries and production at the point of sale – fertilizer plants, for instance – will also need to be restarted. Factory restarts can be pretty involved tasks.

This recovery period doesn’t just get pushed out by 24 hours each day it gets longer as more production is forced to shut down or is damaged in the fighting. As I write this, futures markets for the WTI seem to be expecting oil prices to remain elevated (above $70 or so) well into 2028.

Meanwhile, disruption of fertilizer production, which relies heavily on natural gas products, has the potential to raise food prices globally. Higher global food prices – and food prices have already been elevated by the impact of the War in Ukraine – are pretty strongly associated with political instability in less developed countries. After all, a 25% increase in the price of food in a rich country is annoying – you have to eat more cheaper foods (buy more ramen, etc.). But in a poor country it means people go hungry because they cannot afford food and hungry, desperate people do hungry, desperate things. A spike in food prices was one of the core causes of the 2010 Arab Spring which led in turn to the Syrian Civil War, the refugee crisis of which significantly altered the political landscape of Europe.

Via Wikipedia, a chart of the food price index, with the spikes on either side of 2010 clearly visible; they are thought to have contributed to the intense political instability of those years (alongside the financial crisis).

I am not saying this will happen – the equally big spike in food prices from the Ukraine War has not touched off a wave of revolutions – but that it increases the likelihood of chaotic, dynamic, unsettled political events.

But it does seem very clear that this war has created a set of global economic headwinds which will have negative repercussions for many countries, including the United States. The war has not, as of yet, made Americans any safer – but it has made them poorer.

Then there are the political implications. I think most folks understand that this war was a misfire for the United States, but I suspect it may end up being a terrible misfire for Israel as well. Israeli security and economic prosperity both depend to a significant degree on the US-Israeli security partnership and this war seems to be one more step in a process that very evidently imperils that partnership. Suspicion of Israel – which, let us be honest, often descends into rank, bigoted antisemitism, but it is also possible to critique Israel, a country with policies, without being antisemitic – is now openly discussed in both parties. More concerning is polling suggesting that not only is Israel underwater with the American public, but more Americans sympathize with Palestinians than Israelis for the first time in American history.

Again, predictions are hard, especially about the future, but it certainly seems like there is an open door to a future where this war is the final nail in the coffin of the American-Israeli security partnership, as it becomes impossible to sustain in the wake of curdling American public opinion. That would be a strategic catastrophe for Israel if it happened. On the security side, with Israel has an independent nuclear deterrent and some impressive domestic military-industrial production the country is not capable of designing and manufacturing the full range of high-end hardware that it relies on to remain militarily competitive despite its size. There’s a reason Israel flies F-35s. But a future president might well cut off spare parts and maintainers for those F-35s, refuse to sell new ones, refuse to sell armaments for them, and otherwise make it very difficult for Israel to acquire superior weapons compared to its regional rivals.

Economic coercion is equally dangerous: Israel is a small, substantially trade dependent country and its largest trading partner is the United States, followed by the European Union. But this trade dependency is not symmetrical: the USA and EU are hugely important players in Israel’s economy but Israel is a trivial player in the US and EU economies. Absent American diplomatic support then, the threat of economic sanctions is quite dire: Israel is meaningfully exposed and the sanctions would be very low cost for the ‘Status Quo Coalition’ (assuming the United States remains a member) to inflict under a future president.

A war in which Israel cripples Iran in 2026 but finds itself wholly diplomatically isolated in 2029 is a truly pyrrhic victory. As Thucydides might put it, an outcome like that would be an “example for the world to meditate upon.” That outcome is by no means guaranteed, but every day the war grinds on and becomes less popular in the United States, it becomes more likely.

But the United States is likewise going to bear diplomatic costs here. Right now the Gulf States have to shelter against Iranian attack but when the dust settles they – and many other countries – will remember that the United States unilaterally initiated by surprise a war of choice which set off severe global economic headwinds and uncertainty. Coming hot on the heels of the continuing drama around tariffs, the takeaway in many places may well be ‘Uncle Sam wants you to be poor,’ which is quite a damaging thing for diplomacy. And as President Trump was finding out when he called for help in the Strait of Hormuz and got told ‘no’ by all of our traditional allies, it is in fact no fun at all to be diplomatically isolated, no matter how powerful you are.

Of course the war, while quickly becoming an expensive, self-inflicted wound for the United States has also been disastrous for Iran. I said this at the top but I’ll say it again: the Iranian regime is odious. You will note also I have not called this war ‘unprovoked’ – the Iranian regime has been provoking the United States and Israel via its proxies almost non-stop for decades. That said, it is the Iranian people who will suffer the most from this war and they had no choice in the matter. They tried to reject this regime earlier this year and many were killed for it. But I think it is fair to say this war has been a tragedy for the Iranian people and a catastrophe for the Iranian regime.

And you may then ask, here at the end: if I am saying that Iran is being hammered, that they are suffering huge costs, how can I also be suggesting that the United States is on some level losing?

And the answer is simple: it is not possible for two sides to both win a war. But it is absolutely possible for both sides to lose; mutual ruin is an option. Every actor involved in this war – the United States, Iran, arguably Israel, the Gulf states, the rest of the energy-using world – is on net poorer, more vulnerable, more resource-precarious as a result.

In short, please understand this entire 7,000+ word post as one primal scream issued into the avoid at the careless, unnecessary folly of the decision to launch an ill-considered war without considering the obvious, nearly inevitable negative outcomes which would occur unless the initial strikes somehow managed to pull the inside straight-flush. They did not and now we are all living trapped in the consequences.

Maybe the war will be over tomorrow. The consequences will last a lot longer.

Like this:

Like Loading...

Your bridge to wealth is being pulled up - Daniel Homola

Tue, 24 Mar 2026 16:43:37 +0000

I want to make three claims: two about the world we live in and one about the future ahead of us.

The first is mathematical Most of the traits you were born with - intelligence, conscientiousness, height, bone density, grip strength, resting heart rate - follow a bell curve. They are Gaussian: symmetric, mean-reverting, self-averaging. The wealth you were born into is not. Wealth follows a power law. The top 1% of American households holds more than the bottom 50% combined. The mean is five times the median. These are not the same kind of object. When you multiply a bell curve by a power law to produce a life outcome, the power law dominates. Per standard deviation, parental wealth predicts a child's adult income at least as strongly as the child's own cognitive ability - and the gap widens substantially at the extremes, where the power-law tail of wealth extends far beyond anything the Gaussian bell curve of IQ can reach. One regresses. The other compounds.

The second is structural The historical wire from IQ through credentials to high-paying work is being cut by artificial intelligence - and understanding why that matters requires a brief detour into how it was built.

For most of human history, two inheritance systems ran in parallel and didn't communicate. Biological traits passed through chromosomes: stochastic, noisy, tending back toward the average across generations. Wealth passed through property law: wills, trusts, title deeds, institutional relationships. No regression. No noise. Just compounding. Then the French Revolution, industrial capitalism, and the credential systems of the nineteenth and twentieth centuries built a bridge between them: IQ → credentials → income → heritable wealth. For the first time at scale, cognitive ability could escape the class it was born into.

Artificial intelligence is dismantling that bridge. Large language models already match median professional performance on the routine tasks that constitute a large fraction of professional billing across legal research, financial analysis, software engineering, and diagnostic reasoning. The IQ premium in the labour market is collapsing. The capital premium is not. The coefficient on inherited wealth in the income equation is rising as the coefficient on cognitive ability falls - and the transition is measured in years, not decades.

The third is a predictionWhen the bridge closes, what remains is what was always there underneath it: two systems running on different mathematics, no longer connected. The wealthy class keeps the compounding clock. Everyone else keeps the biological one - the one that tends, however slowly, back toward the centre. For about ten generations the bridge existed and the two could communicate. A brilliant person from a modest background could cross. When it closes, the crossing stops. Each side compounds its own logic across generations. That asymmetry, left to run, has a historical name: aristocracy - not by decree, but by ruthless compounding.

But the transition is not complete. Deep domain knowledge combined with AI fluency is scarce in a way that inherited wealth cannot buy - it is gated by speed and expertise, things a person with ability and modest starting wealth can actually have. That window is the next five to ten years. After it closes, the legal inheritance system runs alone. UBI and aggressive capital taxation can floor the bottom and slow the compounding at the top - but they cannot create new entrants at the top in a world where labour no longer generates enough surplus to accumulate capital from scratch.

Watch the labour share of GDP against capital returns. Watch whether professional income and inherited wealth grow more correlated in the same households over the next decade. These numbers are published quarterly. The current trajectory already points in one direction.

This essay builds all three claims carefully - with interactive simulations, a calibrated model, and a political argument about which levers are actually available. It takes about an hour to read. What it offers is a precise map of a transition window: where it came from, how long it stays open, and what the world looks like after it closes. Most of it is interactive - you will spend more time running simulations than reading sentences. I built this framework to understand my children's futures. I think it will change how you see yours.

"Collaboration" is bullshit.

Mon, 23 Mar 2026 08:43:31 +0000

This newsletter is free to read, and it’ll stay that way. But if you want more - extra posts each month, access to the community, and a direct line to ask me things - paid subscriptions are $2.50/month. A lot of people have told me it’s worth it.

Upgrade

In 1944, the Wehrmacht launched into Hitler’s last ditch effort to save the Third Reich. The Battle of the Bulge was a doomed campaign and a doomed gamble from a doomed regime, but its brutality was a true second test of the US Army on the Western Front. During the battle, Army historian S.L.A Marshall began interviewing infantry companies who’d been baptised in combat. Published 3 years later in his 1947 book, Men Against Fire, Marshall’s research showed that just 15-20% of riflemen in active combat positions ever fired their weapons - most kept their heads down. They moved when they were ordered and they held their positions, and they mimicked the outward appearance of a soldier in battle - but shoot, they did not. By any standard organisational metric, the men were present and accounted for, but 4 out of 5 never pulled the trigger.

You can debate the extent of Marshall’s numbers, and you can debate his methodology, but his ratio shows up, again and again. IBM stumbled onto it in the ‘60s when they discovered that 80% of computer usage came from 20% of the system’s features. The pattern recurs because it describes something real about how effort is distributed inside groups, where a fraction of the people do most of the work, and the rest provide what you might ~charitably call “structural support.”

Anyone who has worked in any large organisation knows exactly what I’m talking about.

The modern tech industry looked at the problem of human coordination and participation and decided the solution was “collaboration.” If only 20% of us are operating with a “killer instinct” we need to be better at managing the shared instincts of the other 80%. And so collaboration became our shared obsession. We pursue “teamwork” as a holy grail.

The teamwork revolution, if you can call it that, gave us Notion for our documents, ClickUp for our tasks, Slack for our conversations, Jira for our tickets, Monday for our boards, Teams for the calls that should been emails, emails for the things that we couldn’t squeeze in anywhere else, and now agents attempting to re-invent the whole stack. The average knowledge worker maintains accounts across system after system, switching between applications hundreds of times per day. And they produce, in aggregate, a staggering amount of coordinated and collaborative activity that never actually becomes anything resembling ~output.

When you strip away the product marketing and the dev relations and the blog posts and the funding rounds and the fuckery-upon-fuckery of it all, we’re left with a simulation of collective engagement - but very little else. Transparency got confused with progress, visibility got confused with accountability, and being included in the thread became the same thing, socially and organizationally, as owning the outcome.

Once that confusion set in at the cultural level it became nearly impossible to dislodge. The feeling of collaboration is pleasant in a way that personal accountability can never be. Owning something means you, specifically and visibly you, can fail at it, specifically and visibly, in ways that attach to your name.

Collaborating means the failure belongs to the process.

So everyone chose collaboration, and we called it culture.

Marshall's riflemen were ordinary people responding to the diffusion of responsibility that happens inside any group. Maximilien Ringelmann measured the same phenomenon with ropes in 1913, long before there were Slack workspaces to offer an emoji-react to it. Individual effort drops predictably as group size increases. The presence of others dissolves the sense of personal responsibility in a way that feels, to everyone experiencing it, entirely reasonable. You're part of a team, you're contributing, you're also (measurably) pulling less hard than you would if the rope were yours alone. Every single person on the rope is doing this simultaneously, which is why the total force never adds up the way the headcount says it should.

Frederick Brooks identified the same dynamic in software development in 1975, watching IBM's System/360 project illustrate his emerging thesis that adding people to a late project makes it later. Communication overhead grows faster than headcount, coordination costs compound, and every new person contributes their capacity along with their relationships to everyone else. Those relationships require maintenance and produce misalignment and generate the need for more meetings to address the misalignment those meetings created.

Brooks might as well have described your company's Q3 roadmap planning cycle and your startup's sprint retrospective, all of which have gotten longer every year and produced, relative to their investment, less.

The collaboration industry has spent a fortune obscuring a dirty truth: most complex, high-quality work is done by individuals or very small groups operating with clear authority and sharp accountability, then rationalized into the language of teamwork afterward. Dostoevsky wrote _The Brothers Karamazov_ alone. The Apollo Guidance Computer came from a team at MIT small enough to have real ownership, hierarchical enough that Margaret Hamilton's name could go on the error-detection routines she personally designed.

Communication matters, and shared context matters. But there’s a huge difference between communication and collaboration as infrastructure to support individual, high-agency ownership, and communication and collaboration as the primary activity of an organisation. Which, if we’re honest, is what most collaboration-first cultures have actually built. They’ve constructed extraordinarily sophisticated machinery for the social management of work, without actually doing the work they’re socialising about.

If and when it exists, ownership looks like an individual who deeply gives a shit, making a call without waiting for group-consensus. That individual will be right sometimes, and they’ll be wrong other times, and they’ll own it. They won’t sit around waiting to find out who has the authority to move a card from one column to another and post about it in the #celebrations channel.

But being that person sucks when “collaboration” is the reigning value, because every unilateral decision gets read as a cultural violation and a signal that you aren’t a team player. Collaboration-as-ideology has made ownership and responsibility feel antisocial, which is a hell of a thing, given that ownership is the only mechanism that gets anything across the finish line.

You can see this excess everywhere. Standups where people announce their busy work and as long as everyone’s “on the same page” nobody changes course. Documents that are written to perform thinking so somebody else can perform thinking, with no decision in sight. Retros, and kickoffs, and WIP meetings that spawn their own retros, kickoffs and WIP meetings like cells dividing and re-dividing, with zero connection to the work that it’s nominally organising around.

Every project now seems to carry more coordination overhead than execution time, and when it fails the postmortem just recommends more collaboration...

At some point (and I think that point was fucking yesterday) we have to ask ourselves - what are we actually producing and who is actually responsible for producing it?

Because at some level, the answer for “who is responsible for X” has to be one single person, no matter how much the collaborative apparatus layered over modern work has been engineered to make that person invisible and dissolve accountability.

We need to find some path back to trusting that individuals will do their jobs, without every responsibility being visible to an entire organisation, without follow-ups being scheduled by a cadre of overpaid managers with their overfed metrics.

Maybe - just maybe - we could make our lives a little easier. Maybe we could let human beings keep their own lists of tasks, and we could let them sink or swim by how they manage those tasks, and we could assign blame to them and to them alone when they fuck up. Maybe we could do it without needing to have team-level views of every Kanban, calendar and task list. And maybe - if we let go of the warm, expensive fiction of collective endeavour - we could make it a little easier to see who among us are pulling the trigger and who are just keeping their heads down.

The last Andy Weir movie I will ever waste money on

Mon, 23 Mar 2026 05:34:42 +0000

The commenters here are persuasive. I dissed Andy Weir and his new movie, and I was told that it was entertaining and I should give it a chance. So I did. I went to the theater to see Project Hail Mary.

I loathed it.

Sorry.

The premise is garbage. Weir postulates an “astrophage,” a bacterium that harvests carbon from Venus and then streams to the sun, “eating” the sun, collecting vast amounts of energy, dimming the sun, and threatening humanity with extinction within decades. They send a probe to the line flowing between Venus and the Sun and collect the mysterious black particles, bring it back to Earth, and a middle school science teacher looks in a microscope and figures out that it’s an organism that harvests energy from stars.

Stop right there. I’d appreciate it if someone could justify that plot hook, which wouldn’t have been out of place in a 1950s hack disaster movie. It’s stupid.

Then astronomers notice that all the local stars are experiencing this same mysterious dimming. The “astrophage” must be infectious! Let’s not concern ourselves with the timing: we observe a rapid phenomenon occurring within the lifespan of a single human being simultaneously in a population of stars scattered over a volume 100 light years across. There’s a complete lack of awareness of space and time in this movie.

Humanity’s response is to quickly build a spaceship to fly to the one star, Tau Ceti, that isn’t exhibiting the mysterious dimming to see if they can find a “cure”. Fortunately, the “astrophages” also store such a tremendous amount of energy that they can form a fantastic, near-magical rocket fuel, enabling the construction of a starship that can travel at something near light-speed. This is the kind of exotic nonsensical space fuel you’d find in a 1930s pulp novel.

The plan is to send a small crew with one engineer, one pilot, and one scientist to a star 12 light years away, to collect information about how Tau Ceti was resisting the infection, and then return to Earth with a solution. That’s going to take at least 24 years round-trip, to deal with a crisis that’s going to doom Earth in 30 years. My problem was that my mental calendar was getting hopelessly lost by this time.

Then the scientist who is expected to lead this critical mission to stop human extinction was the middle school science teacher. This teacher is the charming, charismatic, appealing Ryan Gosling. We’re doomed if that is our selection criterion.

The ship takes off. Next thing we know, Ryan Gosling wakes up from an induced coma (that’s how humans can survive a 12 light-year trip?), the two other crew members have died — no explanation provided, but nice to know pilots and engineers are superfluous — and Gosling has amnesia.

That’s just the setup for the main part of the story, and it’s such radically nonsensical and unscientific garbage that I felt like walking out, and only stayed in my seat by virtue of Ryan Gosling’s charm and the curiosity and need to find out how the story would crawl out of this mass of sewage.

No spoilers. You’ll have to suffer as I did if you want the answers.

Short answer: Gosling finds a cute chatty alien who is there for the same purpose, and they team up. Don’t worry, there’s none of that complicated first-contact rigamarole to establish communication — they just point at things and say words and use a computer to compile a dictionary. Quickly. Mostly off-screen. Can’t let the whole alien complication that Weir has introduced get in the way of the whole star-eating space bacteria problem that Weir introduced!

Gosling also has a Weir staple: a white board that he can scribble on to solve science problems. For example, Gosling discovers a problem that will make the alien’s spaceship break down on the way to its home in 40 Eridani. So he scribbles some stuff on the white board and decides to fly off to the rescue, and somehow find this stranded spaceship somewhere between Tau Ceti and 40 Eridani. Are you surprised to learn that he does? Spaceships are easy! All you need for interstellar navigation is a white board and a collection of colored markers, and a pilot with no training who was hired on the basis of his entertaining middle-school science classes.

Ryan Gosling is a good actor who gave a great performance in an unbelievable role, and the alien (named Rocky) was amusing and somewhat original, but you will never, ever, ever, ever convince me to see another movie based on an Andy Weir book. He’s a hack.

Jesus christ, that movie was fucking stupid.

The Valuation Function Problem: Why Skin in the Game Fails When It Matters Most - The Nukemblog

Fri, 20 Mar 2026 20:50:10 +0000

As tensions between Iran, Israel, and the United States escalate, a familiar argument resurfaces in op-eds and think-tank threads: if only the decision-makers had real skin in the game — if generals’ sons were drafted, if politicians stood in the line of fire — they’d be far less eager for war.

It’s an appealing theory. It’s also built on an assumption so large, and so rarely examined, that accepting it uncritically will lead you to make systematic forecasting errors about some of the most dangerous situations in the world.

The Theory and Its Hidden Assumption

Skin in the game (SITG) is a simple mechanism: agents who personally bear the consequences of their decisions make better decisions. Remove the asymmetry between risk and reward, and recklessness follows. Restore it, and rational caution returns.

This works. In the right domain.

When a bank trader risks his own capital rather than the bank’s, he sizes positions more carefully. When a contractor has to live in the building he designed, he doesn’t cut corners on the foundations. The feedback loop is tight, the consequence is personal and immediate, and the agent’s objective — not losing money, not dying in a collapsing structure — aligns with ours.

But notice what the mechanism assumes: that the agent is primarily motivated by self-preservation or material wellbeing, and that personal loss is weighted heavily enough to override other motivations.

This is not a universal property of human beings. It’s a cultural and ideological variable.

The Valuation Function Problem

Every model of incentives contains, usually implicitly, a model of what people want. SITG assumes people want, above all, to avoid personal loss. Change that assumption and the entire mechanism inverts.

Consider a soldier who genuinely believes that dying in battle guarantees paradise. To him, the “cost” you’re trying to impose isn’t a cost — it’s the prize. SITG has nothing to say to him. The more skin he has in the game, the more motivated he becomes.

Or consider a leader who believes his civilization faces extinction. His calculus isn’t war vs. peace — it’s fight now vs. die later. Personal exposure to risk doesn’t dampen aggression; it confirms the urgency of it.

Or consider a political culture where the failure to fight — showing weakness, losing face, betraying ancestors — carries a social cost greater than death itself. In a shame-honor framework, the SITG lever doesn’t just fail to work. It points in the wrong direction.

Consider Duels: For several centuries across Europe and America, educated men with full knowledge of the consequences — death, serious injury, legal jeopardy — regularly accepted challenges rather than endure the social cost of refusing. SITG was maximally present: the consequence was immediate, personal, and fatal. The SITG mechanism predicts they would prefer dishonor to death. They didn’t. The social cost of backing down outweighed the physical cost of dying, reliably enough that dueling persisted as an institution for generations. It wasn’t abolished by making the stakes higher. It was abolished when the underlying honor culture that made refusal worse than death finally eroded.

These aren’t exotic edge cases. They describe large portions of recorded human history.

History Didn’t Get the Memo

If SITG reliably suppressed war, you’d expect the eras of maximum skin in the game — when kings led from the front, when warlords sent their sons into battle alongside conscripts, when the decision-maker and the foot soldier shared the same mud — to be the peaceful ones.

They were not. They were the most warlike periods in human history.

Alexander the Great fought in the front rank and took serious wounds at multiple engagements. He conquered most of the known world. The Mongol khans rode with their armies and suffered casualties in their own families. They prosecuted campaigns that killed roughly ten percent of the world’s population — among the most destructive in pre-modern history.

The counterargument sometimes offered is that these leaders had asymmetric skin in the game — they fought on horseback with bodyguards, insulated from the worst of it. But this escape hatch actually weakens the SITG position. It concedes that even substantial personal exposure didn’t suppress aggression, and raises the bar for “real” SITG to a standard that has never existed and cannot be legislated into existence.

The more honest explanation is simpler: for these men, in these cultures, war was not a cost to be minimized. It was the primary mechanism for acquiring status, legitimacy, wealth, and meaning. Personal risk didn’t deter them. It was the point.

Why This Failure Is Predictable

The domains where SITG works reliably share a structure: consequences are immediate, personal, physical, and unambiguous. A bridge engineer on his own bridge. A trader with his own capital. A pilot in his own aircraft. The feedback loop is tight and not easily reinterpreted.

SITG fails — predictably — when consequences are delayed, statistical, collective, or abstractable. And crucially, when a cultural or ideological frame exists that revalues the consequence. War almost always falls in this second category.

Death in battle can be reframed as martyrdom, sacrifice, honor, a ticket to paradise, a debt to ancestors, proof of manhood, or the cost of survival against existential threat. Each of these frames converts the cost into something else — honor, proof of faith, debt to the dead — until it no longer functions as a deterrent.

A related claim deserves attention: that experiencing war makes you anti-war. Sometimes true. Often not — and the exceptions illuminate exactly why the mechanism fails.

World War I produced some of the most powerful anti-war literature in history — Owen, Sassoon, Remarque. It also produced the Freikorps: hardened veterans who embraced violence as identity and became the street-fighting nucleus of early European fascism.

Adolf Hitler served as a frontline infantryman in WWI. He was wounded. He was gassed. He received the Iron Cross for bravery. By every metric, he had maximum skin in the game. He concluded from this experience not that war was hell to be avoided, but that Germany had been stabbed in the back before achieving its rightful victory — and that what was needed was a harder, more committed, more ideologically pure war next time.

This is not an anomaly that breaks the theory. It is the clearest available demonstration of why it fails: prior ideology determines what lessons you extract from experience. The war confirmed his existing worldview. The loss became a grievance narrative. SITG had no mechanism to correct this, because he wasn’t loss-averse in the relevant sense — he was interpreting loss through a cultural frame that converted defeat into motivation.

The same mechanism appeared in 20th-century Japan. A military culture that explicitly revalued death — Bushido, the kamikaze program, the expectation of fighting to the last civilian on the home islands — produced decision-making that Western analysts found literally incomprehensible. Not because the Japanese were irrational. Because they were operating on a different objective function.

The selection problem compounds all of this. SITG environments tend to select for the people on whom the mechanism works least. Who volunteers for the most exposed roles in warfare, ideology, and high-stakes politics? People already culturally primed to discount personal loss. The mechanism selects for zealots, glory-seekers, and true believers — and then wonders why it can’t deter them.

The Forecasting Implication

This matters practically whenever you’re trying to predict the behavior of actors operating under a different valuation function than your own.

If you assume SITG as a universal corrective — that personal exposure to consequences will reliably produce caution — you will systematically underestimate the aggression of actors for whom:

Death in the cause is a reward, not a cost
Existential threat perception makes ordinary risk calculus irrelevant
Cultural frameworks convert restraint into shame or betrayal
Ideological identity overrides individual self-preservation

These are not fringe conditions. Right now, several of the key actors in the Middle East fit one or more of these descriptions. Applying SITG logic to predict their behavior — expecting that exposure to consequences will moderate their choices — is not analysis. It’s projection.

What Actually Moves Behavior

None of the mechanisms that reliably constrain conflict are reliable across all contexts. But they share something the SITG mechanism lacks: they operate on what actors actually value, rather than assuming a universal preference for self-preservation.

Economic interdependence. When elites on both sides have material interests that war destroys, the calculus changes — not because of personal risk, but because the prize disappears. The mechanism assumes elites have more to gain from peace than from war, which holds until ideological or existential stakes reorder the preference ranking. It failed in 1914 despite deep European economic integration.

Structural veto players. Institutional constraints — legislative bodies, independent militaries, allied states with credible red lines — can prevent individual decision-makers from converting aggressive preferences into action. The mechanism isn’t SITG; it’s friction. It works until the veto players themselves are captured by the same ideology driving the aggression, or until a leader acquires enough institutional control to neutralize them.

Deterrence at civilizational scale. Nuclear deterrence works not because leaders fear death — many demonstrably don’t — but because the consequence is civilizational annihilation, a scale at which even martyrdom narratives break down. It requires credible capability and credible will, and it deters only actors who believe the threat is real.

Modeling the actual valuation function. This is the hardest and most neglected. Before you can predict or influence behavior, you need an accurate model of what the actor actually wants — not what a rational self-interested agent would want. This requires deep cultural literacy, historical knowledge, and the intellectual honesty to acknowledge that your default assumptions are culturally specific, not universal.

That last point is unglamorous. It doesn’t fit on a bumper sticker the way “skin in the game” does. But it’s the work.

A Closing Note

The current Iran-Israel-USA situation will not resolve according to the logic of rational loss-aversion. Iran’s leadership operates under a theology of resistance that revalues sacrifice; Hezbollah and Hamas under explicit martyr frameworks; Israel’s decision-makers under existential threat perception rooted in historical memory that overrides ordinary risk calculus; domestic politics on all sides rewards projections of toughness and punishes restraint.

This doesn’t mean the situation is unreadable. It means you have to read it correctly: by modeling what each actor actually values, what they believe they’re buying with their risk, and what cultural or ideological frames are converting the costs you see into rewards they feel.

The map that says “give them skin in the game and they’ll calm down” is not wrong because incentives don’t matter. It’s wrong because it assumes everyone is using the same map.

Most of the dangerous actors in history were not. Most of the dangerous actors right now are not.

That’s the thing about skin in the game. It only works on people who share your theory of what skin is worth.

Published March 20, 2026

Disclaimer: This post was created by a human with the assistance of LLM

systems-thinking geopolitics incentives culture

You have access to almost any book - BiteofanApple

Fri, 20 Mar 2026 20:50:25 +0000

I read a fair bit of non-fiction that spans a number of different genres. But the trouble with non-fiction is that it never answers a question without raising further questions. Non-fiction comes replete with footnotes: a ponzi-scheme for the curious mind.

Of course, this leads very quickly to a fundamental problem. The deeper down the rabbit hole you go, the more likely you are to find that the cited works are academic, rather than consumer texts. Reading non-fiction, especially when it comes to history or science, you very quickly leave the realm of what you can easily find on a bookstore shelf. Sometimes you get lucky. Sometimes these academic books are relatively cheap, other times it's stamped with a textbook-level price tag or the book is simply out of print.

The real price for a book I wanted to read. Photo: mine

The Internet Archive exists of course, and if the book is really old then that's a good place to start. However, it will never be the full solution.

Several people reading this are likely already screaming: go to the Library. However, most readers would be correct in assuming that one's local city library does not stock the esoteric academic text from the mid-1980s that you're looking for. Wander the little shelves of mass-market fiction and explainer books that fill my local library for example and you won't find anything of the sort. My university library, located just down the road, might have them but I'm not a student and so the best I could do would be to wander in and skim the title for a few hours while being unable to take it home.

There is a solution though, one that I'm sure many people out there already know about, but I didn't. So I'm here to make sure the good word is spread.

You have access to basically any book you want,* through a system called Inter-Library Loan (ILL).

The World's Library

Assuming you live in a city or county connected to some segment of the Inter-Library Loan system (of which there are many), your local city library can request books from basically anywhere, all for free or sometimes a modest fee (in San Diego it's $5 per request).

From here, you leave the beaten path and venture into the rather informal system of ILL. San Diego, for example, has three methods of ILL depending on the reach you need and the speed at which you need the given book. For us, the Circuit system integrates with our local universities to get you quick access to books in their catalog, while the Link+ system has a broader reach. In the end though, if neither of those two have the book you want, you can use WorldCat. WorldCat is a collection of catalogues from thousands of public libraries and colleges across the world. Given time to wait, you can get basically any book you want this way: including the "rare Ph.D. thesis that apparently only exists at the one person's alma-mater".

One of many books I read via Inter-Library Loan. Photo: mine

It's magical.

Of course, the moment you venture off the beaten path like this, you're going to encounter snags and irritations that simply do not usually happen with library loans. Delays, glitches, and a lack of transparency are occasionally par for the course here, but the reward is that you can call upon the world's libraries to summon forth whatever text you want.

Under-Appreciated, Under-Explored

Until a few years ago, I was not a big library user. As a kid, of course, I'd checked out books and in college I spent many an idle afternoon browsing the older stacks, but I always believed that you had to have connections to have access. Being at university, that granted you access. Working in academia, that granted you access. I never realized that, by filling out a little form online and then waiting a bit, I too could have access to those same texts. Perhaps this is common knowledge, but it certainly wasn't to me.

It's pretty wonderful to get to read a thing that journeyed far and wide to find you, passing through a dozen hands called by your personal whim. I love Inter-Library Loan. It legitimately changed how I read and research topics. These days, when I have a burning question or find an obscure footnote, I don't need to turn to increasingly unreliable internet sources, spend ungodly amounts of money on academic books, or scrounge quotes and snippets from a sketchy PDF scan. I can click a button, wait a bit, then hold the real book in my hand.

It's glorious and you should try it.

* Of course, rare books and damaged media are not included.

We’ve got to stop sending files to each other – Terence Eden’s Blog

Thu, 17 Jul 2025 18:14:06 +0000

Another day, another data breach.

the spreadsheet, initially shared in 2022, and thought to contain data related to a small number of applicants, had contained hidden data related to more than 18,000 people. 

ICO statement in response to 2022 MoD data breach

Why are people still sending files to each other? I remember having a stand-up argument a decade ago with a project manager who wanted us to email a completed Word template to him every day. He'd then spend hours merging the various documents together. He couldn't get his head around the collaborative document suite the company had purchased a licence for. I tried showing him that we could give specific people write-access to the document and they could edit it live. No more emailing back-and-forth.

It just didn't stick. It wasn't that he was ignorant about what computers could do, but his entire mental model was built around files. Discrete packets of data with a fixed metaphor from the real world.

Collaborative online documents don't have an easy analogue analogue. It is rare to see a dozen people scribbling on the same whiteboard or using the same typewriter keyboard.

Permissions are another things that aren't intuitive. The idea that only specific people can see something doesn't match our expectations of paper. Sure, anyone could grab a pen and deface it, that's why we have one person in charge of the "master copy".

Copy. What a hateful word.

The modern workforce shouldn't be flinging copies to each other. A copy is outdated the moment it is downloaded. A copy has no protection against illicit reading. A copy can never be revoked.

Data shouldn't live in a file on a laptop. It shouldn't be a single file on a network share. Data is a living beast. Data needs to live in a database - not an Excel file. Access should be granted for each according to their needs.

I see the same issue in the WeTransfer kerfuffle. Very Serious People saying it was intolerable that the untrusted 3rd party they were using to share Very Sensitive Information was going to read that information.

At which point you have to throw up your hands and ask why people are sending files to each other in the year of Our Lord 2025?!?!? If you have a sensitive file, use proper access controls. Or at least use a password so the FTP-as-a-service provider can't steal your IP.

And git! Don't get me started on git! The best minds of a generation stuck in a paradigm of downloading files to their local machine, making changes, then ~~emailing~~ git pushing them up to be approved? Madness!

Look, there are some times when you need a local copy. I want my own copy of my insurance documents - but that's not a living doc; it is an agreed artefact. Sure, it's handy to have access when there's no network connection - but that's what background sync is for. OK, you're on Office 365 and I'm on Google - so we'll have to work a little harder to set up access.

But all of this is possible!

We rant and rave about the 💾 icon being a skeuomorph. But the very concept of an individual file is also a skeuomorph! Data are not stored on paper files. There is no such thing as a filesystem directory - it's just a convention to make computing palatable for people born in the 20th century who lived in a world of A4 paper and manilla folders.

Modern computing is still stuck in the past. Our computers are like cars which have been designed to carry a bale of hay to mop up the horse-piss.

I’m OK being left behind, thanks! – Terence Eden’s Blog

Fri, 20 Mar 2026 15:39:13 +0000

Many years ago, someone tried to get me into cryptocurrencies. "They're the future of money!" they said. I replied saying that I'd rather wait until they were more useful, less volatile, easier to use, and utterly reliable.

"You don't want to get left behind, do you?" They countered.

That struck me as a bizarre sentiment. What is there to be left behind from? If BitCoin (or whatever) is going to liberate us all from economic drudgery, what's the point of "getting in early"? It'll still be there tomorrow and I can join the journey whenever it is sensible for me.

Part of the crypto grift was telling people to "Have Fun Staying Poor". That weaponisation of FOMO was an insidious way to get people to drop their scepticism.

I feel the same way about the current crop of AI tools. I've tried a bunch of them. Some are good. Most are a bit shit. Few are useful to me as they are now. I'm utterly content to wait until their hype has been realised. Why should I invest in learning the equivalent of WordStar for DOS when Google Docs is coming any-day-now?

If this tech is as amazing as you say it is, I'll be able to pick it up and become productive on a timescale of my choosing not yours.

I didn't use Git when it first came out. Once it was stable and jobs began demanding it, I picked it up. Might I be 7% more effective if I'd suffered through the early years? Maybe. But so what? I could just as easily have wasted my time learning something which never took off.

I wrote my MSc on The Metaverse. Learning to built VR stuff was fun, but a complete waste of time. There was precisely zero utility in having gotten in early.

Perhaps there are some things for which it is sensible to be on the cutting edge. I took part in a vaccine trial because I thought it might personally benefit me and, hopefully, humanity.

But I'm struggling to think of anyone who has earned anything more than bragging rights by being first. Some early investors made money - but an equal and opposite number lost money. For every HTML 2.0 you might have tried, you were just as likely to have got stuck in the dead-end of Flash.

There are a 16,000 new lives being born every hour. They're all starting with a fairly blank slate. Are you genuinely saying that they'll all be left behind because they didn't learn your technology in utero?

No. That's obviously nonsense.

It is 100% OK to wait and see if something is actually useful.

Are They Still Your Friends if You Never See Them? - The Atlantic

Thu, 19 Mar 2026 18:52:49 +0000

This article was featured in the One Story to Read Today newsletter. Sign up for it here.

The dog was asleep in the corner, and I was seated at the kitchen table, drinking a cup of weak tea. My 21-year-old son sat cross-legged on the floor, messing with his guitar, telling me a funny story about a dating disaster involving one of his good friends.

“Rocco’s a fool,” I said with affection when Sam’s tale was done.

“He is,” Sam agreed. “I love him.”

We laughed. Then Sam stopped strumming and looked at me. “You don’t really have any friends, do you, Dad?”

Sam didn’t mean it in a hurtful way. As far as he knew, it was a fair-enough assessment.

“I have friends,” I said. “I just don’t see them, but I know they’re there. And that’s enough.”

Sam considered me—probably knew I was full of it (even if I didn’t at the moment)—then graciously accepted my answer with a nod. But his comment stayed with me. What had happened to my friendships? Were they still there, as I had claimed? What did I get from my friends, and what did I have to offer them? I sipped my tea—it was cold.

Men, it turns out, have lost the knack for friendship. A 2021 survey found that 15 percent of men confessed to having no close friends at all, up from 3 percent in 1990, while fewer than half of men said they were satisfied with how many friends they had. Only one in five men reported having received any form of emotional support from a friend in the past week.

In the days following my son’s comment, I began thinking more and more about my friends and how long it had been since I’d really talked with any of them. I started surfing online about men and friendship—in my drift away from those close relationships outside of my family, I was far from alone. I clicked fast, and the statistics piled up. The U.S. surgeon general has declared an “epidemic of loneliness and isolation.” Even America’s favorite cuddly sex therapist, Dr. Ruth, in her later years, talked less about sex and more about loneliness.

From the February 2025 issue: The anti-social century

And the downside is not just emotional. Researchers have found a staggering 50 percent elevated risk in developing dementia, a 29 percent increased chance of heart disease, and a 32 percent increased instance of stroke for those with “poor social relationships.” Social isolation exceeds the health risks of obesity, inactivity, air pollution, and consuming more than six alcoholic drinks or 15 cigarettes a day. A Harvard study concluded that the No. 1 factor in a longer, healthier, happier life is not diet or exercise, but a positive and consistent connection to community.

A study by the University of Kansas concluded that making a good friend takes more than 200 hours—but losing one is much easier. One of the most vital ingredients in close friendship is consistency. Showing up.

When I was young, that seemed easy enough to do, when my friends and I were all circling the same orbit. But as life asserted its demands, those close friends moved away, scattered. There was Seve, the surrogate big brother I met when I was barely 20, and Matthew, a show-business confidant I met a few years later; Eddie, my oldest friend and early role model; John and Don, friends I met more recently. Dear friends, all. In many ways they were the cornerstones on which so much of my life had been built. I couldn’t remember the last time I’d seen any of them. On the rare occasions when we spoke on the phone, we laughed and caught up—but was that just the fumes of past glories?

My experience with friendship has not always been straightforward. Mine was a typical neighborhood upbringing, now long gone—driveway basketball games and stolen peeks at Playboy magazine in wood-paneled basements. I was the third of four boys, a shy kid with a small circle, yet I never wanted for friends. In my early 20s I became successful in the movies. “Overnight,” my position in the world was forever altered.

I was a very unprepared public figure. Someone who was content to slip along the edges, desiring to be special yet not craving overt attention, I was thrust into the center of things. People came at me. I wrapped myself tighter around the friends I had had before this burst of notoriety. That I began to drink too much spoke to my innate alcoholism and not to my newfound fortune. I retreated, then withdrew.

By the time this brush with fame had subsided and my drinking had been arrested, I was nearly 30. I discovered that I liked my own company and often sought out time alone. When eventually I married, I saw that almost all my friendships with women had been based around flirtation and the possibility of our going to bed. That obviously had to change. And with men, I looked up to realize that several close and longtime friends had moved away. Far away. On the rare occasion that I did form a new connection, the motivation to nurture it was often lacking. Whether a reaction to the hollowness of some insincere acquaintances made during my early fame, or a fearful nature, or just becoming set in my habits, I found myself uninterested, even unwilling, to reach out to new friends. No matter—I was happy in my own company and with that of my wife and children. And there was always work. Life felt full—at least full enough.

But after that conversation with my son, something my wife had cautioned me against came back to me. My introspection, my introversion, my avoidance had begun to chip away at the edges of who I was and narrow my experience, diminish my joy, limit what I had to offer and what I allowed myself to receive. My kids had affectionately (?) begun to accuse me of becoming a curmudgeon. If I was really willing to look, the answer was there to see: My self-induced isolation was diminishing my life, making me into a smaller man. At one point my friends had been instrumental in broadening my horizons, bolstering my courage, providing safe harbor. But were those dear friends even still there?

The writer and his friend (Courtesy of Andrew McCarthy)

Because he was my oldest friend, I called Eddie first. I suggested flying down to Texas for a visit. “For sure,” he said. “But I gotta finish this fucking building first.” Eddie bought and renovated old buildings for a living and was in the middle of a large new project. “Just give me a few months.”

I reached out to Matthew. He was quick to say yes to my coming to Kentucky. I bought a plane ticket. Then, days before my flight, he called to cancel. “There’s just too much shit going on,” Matthew said. Work was a mess, and his son was going through “some stuff.” We pushed it a few weeks, and I rescheduled my flight. Then he canceled again. “Let’s just do it in the spring,” Matthew said. “Things should calm down by then.” I refunded my ticket.

John was mountain climbing in the Himalayas. Don’s email auto-reply said that he was in Japan for an extended stay. But when Seve and I connected, he was excited, and suggested that I come down to Baltimore and we hit the road—“I’ve wanted to head down to the Chesapeake Bay,” Seve said.

A road trip, just like old times, when we’d crisscrossed the country or spent weeks getting lost in Ireland. Seve and I had met in Greenwich Village. I was 19; he was a decade older. His real name is Stephen—though I haven’t called him that in nearly 40 years. We began to play tennis together, and on Sunday afternoons, we’d sit on the phone with The New York Times Travel section on our laps, and plot where we ought to visit someday. Then, one cold Christmas Day, I suggested we get in a taxi, drive to the airport, and actually go somewhere.

“When, now?” Seve asked.

“Now,” I said.

Seve surprised me by picking me up in a cab half an hour later.

Soon we were looking up at the departures board at Newark Airport. People’s Express—one of the original and most cut-rate of the cut-rate airlines—had a flight to Puerto Rico leaving in an hour. We bought two tickets. That night we were in a bar in San Juan. A man with a Hemingway beard and a Spanish accent sat on the stool beside us. He whispered of an island just off the coast—Vieques. A paradise, he said. The Navy frequently shelled the tiny isle for target practice, but we shouldn’t worry about such things, the artillery was well aimed, and besides, it was Christmas; surely the Navy would be on a break from the bombing. The next morning we boarded a six-seater plane.

Read: Americans need to party more

Vieques felt abandoned—the pavement ruptured (from the bombing?), the foliage scraggy and unkempt. We saw no cars and few people. Such simplicity would be my idea of heaven today, but as a young man, mai tais and swimming pools were more my idea of paradise. “There are no women here,” Seve said. And with that, we headed back to the airport.

We were the only passengers on the flight back. The pilot asked if we’d mind if he stopped at St. Thomas on the way to drop some things off. “It’ll save me a trip.”

“I didn’t like San Juan anyway,” I said, and we got off with the cargo.

A dreadlocked Rastafarian was behind the wheel of a cab. We climbed in the back.

“How you doing?” Seve asked.

“Cool and quiet, mon, cool and quiet,” came the languid reply. He knew a hotel, he said, and while en route I asked, “Anywhere we can get some ganja?” The driver shifted his bloodshot eyes to me in the rearview mirror and made the next left.

Properly stoned, we were deposited at a hotel (with pool) on a hilltop. “This is more like it,” I said to my friend. In the middle of the night I awoke, itching. I could hear Seve tossing in the other bed, the slap of flesh, and then a shout: “There are bedbugs in here!”

We went out to sit by the swimming pool. The air was close, there were no stars, the pool was too cold for swimming.

“I wish I hadn’t answered the phone when you called,” my friend said.

“You don’t mean that, Seve.”

He grumbled.

“Well,” I said, trying to look on the bright side, “at least it’s not raining.”

Lightning flashed. It began to rain. Hard. And we started to laugh. It was the kind of hysterical laughter over which a lifelong friendship is forged.

The Chesapeake Bay wasn’t exactly Puerto Rico, but a couple hours in the car was “probably all my back can take,” Seve told me. Several years ago, my friend had undergone a major operation to address stenosis, a narrowing of the spine. Nerves were being pinched and the pain had become intolerable. The recovery was long and arduous. I didn’t visit. More important, the operation hadn’t done the trick; his back was worse than ever.

The day before I was headed down to Baltimore, Seve called to cancel, citing a doctor’s appointment. It seemed odd that he didn’t know about the conflict in advance, but I let it pass. We rescheduled. Then, a few weeks later, just as I was walking out the door to see him, my phone rang again. “I haven’t been entirely honest with you,” Seve began, his voice strained. “I’m not in great shape. I can’t really walk very far, and sitting in a car for that long would be brutal on my back.”

“That’s fine. We don’t need to go far.”

“And I’m up and down all night long. My sleep schedule is all turned around. It’s not the right time. Let’s just postpone it a little.”

“All right,” I said. “Why don’t I just come down and take you to dinner?”

“I don’t want you to drive all that way.”

“I was gonna do that anyway,” I assured him. “It’s a few hours. It’s nothing.”

“I’m just really not up for it. I really appreciate it, but please hear me—” My friend was imploring me now. “It is not a good time. We’ll do it soon. I promise.” I could hear the pain, even fear, in Seve’s voice.

“Okay,” I said, trying to keep my tone neutral. “No sweat. We’ll do it soon.” I hung up, glanced at my overnight bag by the door, made another cup of weak tea, and took a seat in my customary spot at the kitchen table. The dog came over and nudged me for a pet. I shooed her away.

It was natural that my friends had full and busy lives. Although disappointed, I took no offense. I understood—or so I told myself. But the longer I sat with our conversation, the less comfortable I felt. My friend sounded like he was in trouble. I had not been there for our relationship for too long. Was it too late? Had I let the friendship atrophy too much? Was I being too melodramatic to think that I didn’t want the next time I saw my friend to be at his funeral?

That night I couldn’t sleep. Beneath the buzzing of my mind, I knew I had a decision to make. Would I allow my life to keep growing smaller or take the risks required for connection? Friendship had once been so central to my life. Could it be again?

“Fuck it,” I said in the darkness. “I’ll go see them all.”

And when the sun rose, I got in the car, drove to Baltimore—and knocked on Seve’s door.

This article has been adapted from Andrew McCarthy’s new book, Who needs friends: An unscientific examination of male friendship across America

Who Needs Friends: An Unscientific Examination Of Male Friendship Across America

By Andrew McCarthy

When you buy a book using a link on this page, we receive a commission. Thank you for supporting The Atlantic.

填不满的锚点 | JustGoIdea

Sun, 15 Mar 2026 01:52:29 +0000

填不满的锚点

14 Mar, 2026

对财富的渴望，即便不是天性，也早已被几千年的生存经验锻造成了本能。这本无可厚非。

但为什么那么多人对财富的追逐会近乎病态？已经越过了「想要更多」的进取心，变成一种永无止境的焦虑：客观上已经够用，主观上却永远不够。

或许，人们真正渴望的，从来不是财富本身。

一

今天的中国，饿死人的时代似已远去，可体面地活着依然代价高昂。教育、医疗、养老，哪一项不需要真金白银来托底？在许多国家，这些风险通过社会契约分担，个人无需独自面对全部的不确定性。但中国近几十年的经历恰恰相反：制度被反复重置——文革、下岗、房改、教育政策的频繁变动——对亲历者而言，每一次「信赖制度」的选择，都有可能被证明是一场误判。

于是，财富成了普通人能自己掌控的最可靠的锚点。

换句话说，人们真正渴望的，是一种不被时代抛弃的保障。这种保障本质上是开放性的、无边界的。制度能随时重置，那么无论攒多少，都不算真正安全。锚点本身是虚的，当然填不满。

二

我对这种现象抱有一种矛盾的态度：一方面觉得这些人很惨，一方面又觉得，既然是自己的选择，那就应该承受。

只不过，「自己的选择」这个判断，站不太稳。

当一种行为模式是几十年制度强化、加上几代人生存经验叠加的产物，把它归为个人选择，是结果论的残忍。更深层的机制是：当一套秩序长期是人全部意义感的来源时，它的崩塌就等同于自我的崩塌。承认压迫者是恶的，就等于承认自己是受害者；而受害者的身份在许多文化中是羞耻，不是力量。所以人们选择相信它，捍卫它，甚至在它伤害自己时，依然无法松手。

这不是愚蠢，是自我保护。

「惨」与「咎由自取」之间，并不是非此即彼的关系。一个人可以是受害者，同时也是共谋者。

三

更隐蔽的一层，是这种环境还会主动降解人的道德自我期待。

以前，前辈们常以过来人的口吻「教导」我：想挣钱就别要脸，想要脸就别挣钱。

我第一次听到这句话时，以为只是刹那间刻薄的愤世嫉俗。后来发现，说这话的他们往往真诚，甚至带着慈悲。他们是在帮我避开他们曾经受过的伤。

这才是真正让我不寒而栗的。

因为这句话的潜台词是：规则本身就是不公平的，想在这个环境里赢，就得愿意比别人更不要脸。

只是犬儒吗？

不，它是一代人把规则的腐败内化成个人策略之后提炼出的结晶。被这套规则损坏的人，用关爱的方式把规则传递给下一个被损坏的人。受害者们完成了一次代际共谋，而双方都不自知。

圣严长老说过一句话，简短，却精准：

人之所以自私自利，贪得无厌，是因为缺乏安全感。

贪的是钱，真正缺的，是一块能站稳的地方。

四

安全感归根结底是一种能力，但能力需要土壤。

它只能建立于自身，万不可建立在任何外部对象身上，既包括父母、亲朋，也包括国家。将安全感寄托于外部，就像把自己的稳定托付给一个本身充满不确定性的东西。这道理说起来不难，但做到极难，因为它要求人在一无所靠的时候，仍然相信自己站得住、立得稳。

而中国几十年的制度逻辑，恰好从根本上剥夺了人们练习这种能力的空间。当个体从未真正被允许依靠自己时，对外部锚点的渴望便从文化偏好升级成了生存本能。

这是一种结构性的、被制造出来的脆弱。

五

写到这里，我想回到最初的那个矛盾：同情，还是「作茧自缚」？

我的答案是：两者都有，但重心不在问责。

问「他们为什么不改变」，不如先问「改变的条件是否存在」。当制度是大多数人唯一能依附的东西，当别的选项在认知上已经不存在，当每一个叛逆的念头都意味着要从零开始重建意义感……改变的代价，远比旁观者想象的高。

我没有资格站在安全的地方俯视这一切，因为我的「安全」，本质上也是一种选择的结果。离开，本身就是一种态度的表达。但离开之后，我仍然觉得，那些留下来的人，并不比我更应该被指责。

我们只是在不同的条件下，做了不同的选择。而条件，从来不是公平的。

Previous |

#essay

Southern Italy is poor because of malaria - David Oks

Fri, 13 Mar 2026 16:04:27 +0000

A few days ago someone named Michael Arouet went viral on Twitter with this post:

I have never understood the massive economic divide in Italy.
Why is the North with its various industries one of the wealthiest areas in Europe, but the South so extremely poor? Same country, same language, same laws and taxes. Can someone please explain?

I’m glad that Michael’s tweet went viral, because it’s a very interesting question. Italy is an extremely unequal country: the gap between northern Italy and southern Italy is much wider than the Westen/Osten divide in Germany or the north/south divide in the United Kingdom. Northern Italy is comparable in wealth to such beacons of Mitteleuropa prosperity as Austria and southern Germany; southern Italian states like Calabria and Sicily, meanwhile, are some of the poorest parts of the European Union, with per capita incomes on par with Romania and Bulgaria.

From the European Commission

This gap has been present for a long time. When Italy was unified, it was obvious to most involved that northern Italy and southern Italy, the region called the Mezzogiorno, were basically different countries: thus Massimo d’Azeglio’s famous line that “we have made Italy; now we must make Italians.” (Sicilian and Piedmontese, it should be remembered, were not mutually intelligible.) In the years after unification this “Southern question” (the questione meridionale) became one of the dominant issues of Italian political life. That gap has remained pronounced in the century-and-a-half since unification—enough that there have been prominent separatist movements in both the south and the north of Italy—and so the Southern question has remained a long-running matter of contention within Italy, with an area of study, meridionalism, even spawning to study the issue.

Unfortunately this long tradition of trying to study the divergent trajectories of north and south Italy did not do much to improve the caliber of answers to Michael’s question. I won’t go individually through the answers, except to note that most of them just beg the question. Yes, southern Italy has institutions and cultural norms that seem to be less conducive to development; but the interesting question is why it has those institutions and northern Italy does not.

So I will do Michael a favor and provide an answer. South Italy is poorer than the north because of malaria, and what malaria did to the patterns of landowning and agrarian life, and what the patterns of landowning and agrarian life did to human capital and social trust.

South Italy’s geography caused malaria…

We should start with a few words about the disease that humans call malaria: malaria having its origin in the medieval Italian for “bad air.” We now know that malaria, perhaps the deadliest disease in the history of our species, is caused not by “bad air” but by protozoan parasites of the genus Plasmodium, which are transmitted into the bloodstream of primates, birds, and reptiles by mosquitos of the genus Anopheles. These mosquitos breed in still bodies of water, and thus tend to be prevalent in areas dominated by marshes or shallow pools; and because the developmental lifecycle of Plasmodium is aided by heat—the hotter the climate the faster they develop—one tends to encounter malaria most densely in areas that are hot and wet.

There are a few malarial areas in the north of Italy, though those places are generally cold enough that the strand of Plasmodium that could survive there is the relatively mild Plasmodium vivax, which is not so grave a threat to humans. In the central west of the Italian peninsula, by contrast, you have notoriously pestilential marshlands like the Pontine Marshes, which Julius Caesar had wanted drained before his assassination; there were various unsuccessful attempts in the 2,000 years that followed until Mussolini finally succeeded in draining them via massive mechanical pumps in the 1930s. Throughout the south, meanwhile, malaria was widespread due to river conditions: whereas in the north the Po River was fed by meltwater from the Cottian Alps, the Apennines that dominate the topography of the south aren’t tall enough for year-round snow; and so southern rivers are fed exclusively by the rain, and thus flood in the winter and dry up in the summer—in both seasons leaving behind endless puddles. It was, in short, an ideal environment for Plasmodium falciparum, the parasite that causes the deadliest form of malaria.

The “Torelli map” of malaria prevalence in Italy, 1882

P. falciparum first came to Italy sometime around 500 BC, likely coming up through north Africa; we can track its arrival by the abandonment of settlements in areas that were gradually made uninhabitable by endemic malaria, such as the Greek city of Paestum in what is now Salerno. By 100 BC malaria had become widespread throughout the south of Italy and the western coast.

…malaria caused latifundism…

Before P. falciparum arrived, these regions of Italy had been dotted with small farms. But as malaria became endemic that way of life collapsed. Because the disease peaks in summer and autumn, labor-intensive, year-round crops like wine grapes and spring-sown wheat cannot be harvested except at grave personal risk; and so smallholders, unable to forgo two critical growing seasons, abandoned these regions en masse, typically for agricultural colonies elsewhere in the Roman world. As they left, their small farms were gradually replaced by massive latifundia, and free labor was replaced by chattel slaves imported from conquered regions. From one account of this process:

The reason for the genesis of an economy based on mass chattel slavery in western central Italy (and large parts of the south of Italy) in antiquity is fundamentally exactly the same as the reason why it arose in the western hemisphere after Columbus. A spreading disease (in this case P. falciparum malaria) gradually, by a slow process of attrition, either killed or forced to emigrate the bulk of the indigenous farming population in Latium and southern Etruria, thus providing the manpower for Roman colonization elsewhere. That, in turn, created a vacuum, a massive labour shortage, on fertile agricultural land where free men were reluctant to work because of the disease. That labour shortage could only be filled by importing large numbers of chattel slaves. … Their owners need not worry if the gangs of slaves employed on the land … suffered extremely high mortality rates from malaria; slaves were very cheap and easy to replace.

As that excerpt notes, this was not unique to western central or southern Italy. Wherever malaria has coexisted with pastoral agriculture—in the American South, the Caribbean islands, Brazil, in the Arabian peninsula—it has tended to select for large estates manned by slaves. Thus we see, in the case of Italy, that malaria prevalence was strongly correlated with land inequality.

From Buonanno et al. (2020)

Malaria remained endemic in southern and central western Italy for the following two millennia; it should not be surprising that the same basic pattern of landownership persisted along with it. In northern Italy one could find freer forms of labor, like the commercial farming of the Po Valley or the mezzadria system of Tuscany where farmers and landlords split produce and profits equally; but in the south the chattel slavery of the Roman world gradually evolved into a system of severe debt peonage that closely approximated slavery. In the latifondi the authority of the landlords was absolute. Every empire that held sway over southern Italy, from the Arabs to the Spanish, merely served to ratify the rule of the landlords. (The sole exception is when Napoleon, having conquered Italy, sought to abolish the institutions of feudalism; but the landlords quickly regained control over their possessions, now classified as private property rather than feudal inheritance.) It is a common mistake to think, particularly in premodern conditions, that because a given empire controls a territory it actually administers it: in southern Italy it was the landlords who exercised the law and held effective control over the region. Peasants, utterly landless, lived in hillside “agro-towns” to avoid the malarial lowlands; each day they traveled into the fields to work, with their labor overseen by the landlords’ overseers and enforcers. It was a far more severe system of exploitation than the land tenure regimes that emerged elsewhere in Italy.

…and latifundism caused a lot of bad things…

The latifondi system had three consequences worth highlighting. The first is that southern Italian peasants had stronger kinship ties than peasants elsewhere in Europe. In most of western Europe, kinship groups were gradually broken by the interventions of state and religious authorities, for whom strong clan ties were a threat to the stability of both the feudal system and the authority of the Church; the most famous facet of this process was the prohibition on cousin marriage, explored by Joseph Henrich and others. But in southern Italy absentee landlords had little interest in regulating the lives of the landless peasants in the agro-towns: they did not bother with dismantling the clans and extended family networks that characterized southern Italian peasant life, either because they were disinterested—very few of them lived on the land because the land was malarial, after all—or because their power was so absolute that clans among the peasants did not threaten them. Thus we see throughout southern Italy all the marks of strong kinship ties: higher rates of cousin marriage, a greater prevalence of destructive family feuds (thus the term vendetta), and the family structure that undergirded such clan-based organized crime formations as the Sicilian Mafia, the Calabrian ‘Ndrangheta, and the Neapolitan Camorra. More generally, we see the type of “amoral familism” that the sociologist Edward Banfield associated with southern Italian life in his famous study The Moral Basis of a Backward Society: an ethic that sought to “maximize the material short-run advantage of the nuclear family, and assume that all others will do likewise,” and which he attributed to the high rates of death and defective system of landownership he saw in southern Italy.

The second inheritance of the latifondi is the low status of women throughout southern Italian society. Whereas on northern Italian farms women might tend kitchen gardens or raise poultry, in the south they were generally excluded from agricultural work—which was, after all, much more dangerous than in the north—and confined to the management of domestic activities, such as weaving and bread-making. This division of labor informed the strongly patriarchal cultural norms characteristic of southern Italy, whose imprint one can still see in maps of female labor force participation in the 21st century.

From Righetto (2023)

The third inheritance of the latifondi is the remarkably low level of human capital accumulation in southern Italy. As in other places characterized by the political and social dominance of extractive landowners, such as in Latin America, landed elites were generally antagonistic to mass education for the peasants, for the simple reason that there was no benefit to be gained from smarter farm laborers. Thus we see the same pattern of malaria and land inequality repeated in a map of literacy rates from 1929.

From Mariella (2023)

Out of these features of southern Italian society sprung many of the pathologies that plagued southern Italy throughout the nineteenth and twentieth centuries: the ubiquitous brigandism and banditry, the enduring strength of organized crime, the widespread sense of civic apathy, and the “praetorian” (to borrow Samuel Huntington’s term) nature of the southern Italian political sphere, where even into the 1990s violence against public officials was an ever-present fact of life. Even today we see the same basic malarial pattern imprinted on regional maps of Italy. For instance, the strength of the Mafia and Mafia-type groups:

From Calderoni (2014)

The prevalence of corruption:

From De Pascale et al. (2024)

Voter turnout:

From Harka and Rocco (2022)

Or even something as mundane as blood donations per capita, favored by Joseph Henrich as a proxy for impersonal social trust:

From Brunello (2013)

…that prevented industrialization

For all the social harm they did, it should be said that the latifond did not directly impoverish southern Italy. When Italy became independent, the economic gap between north and south was not enormous; in 1871, ten years after unification, northern Italy was only somewhat richer than the south in terms of real wages, and there were several parts of the south that were richer than parts of the north. Both north and south were predominantly agrarian; and so its human capital advantage could have only a limited impact on economic output.

From Federico et al. (2019)

It is only in the decades after unification, as northern Italy began to urbanize and industrialize, that the human capital gap between north and south began to matter tremendously. By 1911 a clear north-south gradient in real wages had emerged.

From Federico et al. (2019)

The most important long-term effect of the latifondi, then, was social and cultural. Northern Italy had developed a relatively educated and skilled workforce, networks of commercial trust, and something of a developmental coalition among its elite, who were predominantly mercantile and industrial rather than agricultural. Thus northern Italy was able to absorb workers from the countryside into the factories and industrialize quite rapidly, and a few decades into the twentieth century it had converged on western and central European standards of living. In the south, meanwhile, the latifondi produced an uneducated peasantry, weak horizontal ties outside the family, and a predatory elite whose wealth came from extraction rather than investment. It thus always lacked the political economy required for industrialization.

So to return to Michael’s question: yes, northern and southern Italy today share the same country, language, laws, and taxes. But they do not share the same history. For two millennia southern Italy was shaped by P. falciparum, which made smallholder farming untenable and produced a system of absentee landlordism and bonded labor, which created a society defined by weak human capital, strong kinship ties, and low social trust outside those kinship networks. Malaria was eradicated from Italy in the middle decades of the twentieth century, and in 1971 the country was declared free of the disease; and in the decades after the Second World War the country was freed also of the latifondi, which had died with malaria just as they had been born by it. But just as one can still see in the faded walls that once girdled the empire of Hadrian the outer edges of the once-great empire of the Romans, one can still spot—in maps of corruption, violence, trust, all the other sad indicators of dysfunction—the faded borders of the mosquito’s dominion. In the words of Francesco Saverio Nitti, prime minister of Italy for one year and a famous student of the Southern question:

Malaria is the basis of all social life. It determines relations of production and the distribution of wealth. Malaria lies at the root of the most important demographic and economic facts. The distribution of property, the prevailing crop systems, and patterns of settlement are under the influence of this one powerful cause.

This Substack is reader-supported. To receive new posts and support my work, consider becoming a free or paid subscriber.

The People Who Shun Super-Popular Pop Culture - The Atlantic

Wed, 11 Mar 2026 19:32:20 +0000

This article was featured in the One Story to Read Today newsletter. Sign up for it here.

These days, everyone seems to be watching The Pitt—but not me. I hear it’s really good. I have to believe it’s good, in fact, because people in my social circle—and Emmy Award voters—won’t stop saying it’s really good. “I’m riveted,” one friend said. “It’s addictive,” another said. “I’m surprised you haven’t seen it,” yet another said. But honestly, the more people who recommend the show, the less likely I am to watch it.

I’ve been this way for a while now. It was the same back in the early aughts, with The Wire, and in the later aughts, with Breaking Bad. Though eventually I succumbed and watched both shows—and loved them enough to rewatch them years later—my unwillingness to engage with literally popular culture in the moment that it’s popular seems only to have intensified in the ensuing years.

“But have you seen Severance?” you might ask. No. Slow Horses? The Night Manager? No and no. I also haven’t seen Sinners—even though I love a period piece and a good fright, and everybody I know is obsessed with it.

This tendency is something I’ve come to call “hype aversion”: an avoidance of the pop-culture products that seemingly everyone insists I would like. It’s not that I’m somehow above it all or too cool (I don’t consider myself cool at all). Some people are early adopters; others are late adopters. I’m simply a weirdly resistant one.

Does this make me a jerk? I don’t like to think so. Contrarian doesn’t quite describe me; my rejection of The Pitt isn’t an attempt to appear provocative or argumentative. And nonconformist doesn’t work; it suggests a person allergic to the zeitgeist, which I’m not. (After all, I covet Clare V. bags. I own a pair of Stan Smith Adidas.) I’m also not a dissenter. Dissent suggests a protest against something that a person has previous experience with, or doesn’t believe in; but my pop-culture resistance is different from having seen something and deemed it wanting or boring. I’m not necessarily worried about encountering pop culture that turns out to be bad. I just don’t care to act on it if it’s supposed to be good.

From the June 2025 issue: Is this the worst-ever era of American pop culture?

I’m not alone in this. (As a matter of fact, the impetus for this inquiry was an unscheduled conversation between me and one of my Atlantic editors, with whom I bonded over a reluctance to watch The Pitt.) Roland Imhoff, a social psychologist at the Psychological Institute of Gutenberg University, in Germany, told me that he relates as well, and suggested that what I’m expressing is less a need for uniqueness than a form of “psychological reactance”—a defensive response that occurs when someone thinks their freedom of choice is being constrained. For a long time, Imhoff told me, he “furiously refused to even touch” the Harry Potter novels because of their popularity and ubiquity; he dug into the series only once his daughter expressed interest in it. The same happened with the music of Taylor Swift: He made an effort to avoid it, then was forced to listen. “And then,” he said, laughing, “I kind of liked it.”

My aversion to hype might seem particularly strange, given that staying on top of popular culture used to be my full-time job. In the mid-’90s, I was an editorial assistant at Entertainment Weekly, a magazine where, among staffers, having an opinion about culture was a primary currency. It was how we came up with ideas, and ideas about how to express those ideas. Our cultural knowledge gave us sway and access. We were the influencers who covered the day’s influencers—actors, writers, directors—and back then, I loved it all.

So what’s my damage now? A few days before my conversation with Imhoff, I reached out to Marilynn Brewer, a social psychologist who, in 1991, articulated what she called “optimal distinctiveness theory,” which proposes that human beings are driven by two (often opposing) psychological impulses: a need for belonging and a need for differentiation. These desires operate in tension, Brewer told me. People look to foster enough in-group behaviors that they feel a sense of social cohesion and belonging, but they also want to express a distinction from others, to avoid a loss of identity or anonymity.

Of course, context matters. The need to belong or feel different fluctuates depending on any number of factors—your job, the town you live in, the friend groups surrounding you. Brewer explained that these needs operate less as fixed personality traits than as something more fluid, such as hunger, which has a threshold that changes over time. When I asked her about my disinterest in widely hyped cultural products, she speculated that these trends might activate my “pretty steep need for differentiation.” And what I have sometimes worried is a sign of immaturity or arbitrary contrariness could be, she suggested, resistance to immersion in a crowd. For some people, Brewer said, excessive hype triggers FOMO (a fear of missing out). Perhaps, I thought later, people like me suffer from LOMO: a love of missing out.

Read: Your FOMO is trying to tell you something

Occasionally, I worry that my LOMO might be annoying. For instance, I don’t so much announce my refusal to engage with The Pitt to interested parties as humor them and try to change the subject, which makes me bad at watercooler conversation. And adopting the mantle of cultural curmudgeon can get tiring. If I’ve made a big deal to friends about brushing off their pop-culture recommendations, I then feel a need to keep up an appearance of recalcitrance. Eventually this becomes an expected posture: an orientation and reputation that is difficult to extract myself from. “You’re the sort of person who doesn’t watch TV,” the guy I’m dating says. “Yes, I do!” I say. Then he points out that I don’t even know how to use my smart TV.

In his book Invisible Influence, Jonah Berger, a marketing professor at the Wharton School of Business, distinguishes between cultural products such as movies, music, and television, which are commonly used as identity markers, and something like, say, dishwashers or toilet paper, which are what he calls more “functional” domains of life. “Imagine you met someone at a party,” he suggested when we spoke recently, and they asked you, “‘What TV shows do you like?’” You might think, “Well, wait a second, what shows I say may impact what this person thinks about me,” he said, and then you might become cautious about what you pick.

According to Berger, a “magnet” model of social influence pushes some people toward conformity and others away from it. In our conversation, he made a distinction between what he called “bandwagon effects” (conformity) and “snob effects” (avoidance when something is too popular), or a need for uniqueness. These motivations aren’t divergent or mutually exclusive, he said; they can, and do, coexist. “It’s not that people only want to fit or only want to stand out,” he said. “Both are true.”

Perhaps this means that resistance to hype is not snobbery but identity management—a need for differentiation that gets triggered when a person believes their autonomy is under threat. In other words, maybe my rejection of The Pitt has little to do with the cultural product and much more to do with an effort to retain independence. I’m not rejecting culture; I’m rejecting overidentification—which, in a highly individualistic society like the United States, may not be such an odd reaction.

My chat with Brewer turned up something else I hadn’t thought of: that my attitude might be related to the sheer number of cultural products on offer, and to the speed with which these products are analyzed and memed. Perhaps shrugging off culture is a form of self-preservation to those of us who are easily overwhelmed by the way social-media algorithms accelerate consumption, and push individuals to engage in public conversation. When the culture pressures people to show that they’re in the know, some of us might be quicker to recoil from knowing in the first place.

Or maybe my resistance to pop-culture evangelism has grown more intense because of our atomized way of consuming said culture, now that streaming has nearly obliterated the custom of synchronized, communal viewing (live sports, awards shows, and huge political events aside). I did watch the Super Bowl, after all, and I plan to watch the Oscars, both of which evoke a sense of participation in a shared cultural moment. When it comes to The Pitt—which I can click to play at any time, any day, in solitude—perhaps what I’m resisting, in choosing not to join the crowd, isn’t the hype, but aloneness.

When you buy a book using a link on this page, we receive a commission. Thank you for supporting The Atlantic.

Load-Bearing Walls - BiteofanApple

Tue, 10 Mar 2026 18:54:06 +0000

We like to believe that we can predict the future. We believe it because it's true, at least in part. Physical systems obey known physical laws and so even fairly complex systems can be understood and predicted with stunning accuracy. Every branch of science makes predictions. Planes fly, servers compute, and dams generate power all through the strength we have to predict the future, if only in the short term. In our human world, we see pollsters predict elections with better-than-random odds, doctors performing research and intervention at global scales, and even disease propagating at predictable rates.

It's stunning just how good we are at this in the modern world. We're so good at predicting the future that much of the financial world relies on it as a hedge against the unforeseen. Short term catastrophe is easier to bear when you can predict that good times can and will return.

Yet there is, of course, much we cannot predict and two main reasons why.

The Chaos Limit

Even the simplest systems can become unpredictable with time because, while the past may indeed predict the future, the approximate past does not. This is chaos. Measurements with real instruments in the real world are always imprecise and so lead to unremovable error. These errors stack up over time and eventually reduce the usefulness of our models.

Famously, the double pendulum illustrates this phenomenon well. A simple pendulum is completely understood, yet a double pendulum is a chaotic system where even with knowledge of its starting position to ten decimal places, the best models will fail in the end.

Still, while chaos may ultimately limit our ability to predict the future in the long term, it leaves quite a lot of wiggle room.¹ No, there's another, much deeper, issue we need to understand.

Never miss a post!

✉️ Subscribe to the Newsletter!

A Connected World

Consider a mass on a spring.² There are known equations for predicting the behavior of a mass (like a smooth block) sliding on a table. Such systems are well studied and the resulting equations of motion depend on only a few variables: the slope of the table, its roughness, the spring constant, etc..

However, this mathematical model cannot tell us what will actually happen to a mass on a spring like this because it has no way to account for what happens when the grad student knocks it over!³ Such a variable isn't "in the model" as a Physicist might say.

Of course, one can rightfully scoff at this example, but if, when we say we want to predict the future, we mean fully predict it, then our models leave a lot to be desired. Indeed, while chaos does limit our ability to model dynamical systems perfectly, systems often degrade due to external factors long before chaos sets in. Something outside the model fails first.

Bearing the Load of Life

A load-bearing wall is one that, if removed, will result in structural damage. Some walls are decorative and can be knocked down at will. Others cannot be easily removed because, aside from being a wall, they are helping to hold the roof up.

It's important, when you're doing construction, to know which walls are load-bearing so that you don't accidentally knock your house down while trying to remodel a bathroom.

There are lots of structures in the human world, some are decorative and others are structural. The color you paint your fence is unlikely to matter beyond aesthetics, but your family income is quite a powerful force in your life and your ability to get a new car is highly dependent on keeping a stable, well-paying job.

As another example: let's say your friends decide to start a monthly board game night. The exact game you choose to play probably doesn't matter for your long-term friendship, however the fact that you're playing a game together regularly certainly does. Consider that for some people the game night might be a much needed break from family trouble or the stresses of work. What's more, friendships decay without active maintenance. For the group as a whole, it might be the only thread holding you all together.

In that way, the game night is load-bearing for the friendship.

Immediately, we can see the truth that the game night isn't just a game night. It's a social outlet, a stress valve, a tether to the present and the past and deciding to skip, postpone, or cancel it has outsized consequences: abandoning a common circle, being left out of in-jokes, drifting apart.

The Hunt for Hidden Walls

It's not impossible to replace a load-bearing wall. It just takes effort. It takes intent. You have to brace the wall, cut out the old beam, and replace it with something just as strong.

Everything must end, including game nights, but it's important to recognize what things in life may be load-bearing and what the load is that they might bear. Knowing that, you can be proactive—set up new things, new rituals and traditions—to bear the load that still very much exists even after game night is long gone.⁴

The trouble is of course, that we often don't notice load-bearing walls until they've already collapsed. Years after game night ends, we look back on it with longing, from a distance finally seeing clearly what we really had.

What's more, unlike with house construction, a wall in life might have been built to be decorative and only become load-bearing over time. Perhaps you threw a party on a whim and then did so again the following year. Soon enough, it's a tradition. Friends look forward to it every year. You do too. However, as more and more people lean on that wall, it will either strengthen to bear the load or it will collapse. Eventually, our lives are constructed entirely of walls which were never built to carry the loads they have upon them, and it's when these walls collapse that their consequences reverberate farther than we might imagine.

Isolation and Community

I've been playing D&D for over a decade and largely with the same group of friends. We'd been playing weekly for years in a sort of 3-on-1-off schedule. But years back some of our players got into 3D printing and that unleashed a lot of cool new options for our games. 3D figures need to be painted, and so a few of us started using that week off to get together, chat, and paint figures for a few hours. On one such night, the conversation turned to current events and to the increased isolation many of us feel in this modern digital world. Eventually one of us made a keen observation:

What we're doing right now, it's basically a weekly church group.

In that moment our shared and hidden load-bearing wall was clearly seen. This group wasn't for D&D. Maybe it never really was.

Who could've suspected that this was load-bearing? Photo: mine

What was clear was that D&D was a place for us all to be, to relax, to maintain relationships. Even when we weren't even playing D&D anymore, we were still getting together. It really never mattered why. A group of friends hanging out, spending time, and working toward a common goal were what actually mattered.

Society's Load-Bearing Walls

Aside from the specific religious tenets, a church group is a type of social gathering. It's a form of community, a place where people meet regularly for a common purpose: to break bread, talk, and help each other through the morass of daily life.

Groups like our little D&D group (or a real church group) take many forms: bowling leagues, swimming clubs, choirs, baseball teams. All of these are institutions built for a specific purpose but that are the backbone of collective civil society and it's these precise institutions that are, at a societal level in the modern world, going through a steep decline. It's not uncommon to see individuals today with few, if any, group ties beyond their own family and especially to a collective group effort.

Something’s changed in the past few decades… In the 1990s, the sociologist Robert Putnam recognized that America’s social metabolism was slowing down. In the book Bowling Alone, he gathered reams of statistical evidence to prove that America’s penchant for starting and joining associations appeared to be in free fall. Book clubs and bowling leagues were going bust.
- Derek Thompson, Why Americans Suddenly Stopped Hanging Out

You might ask: why is it that a baseball club matters for society? And in some respects you're right to question. A baseball club doesn't matter in the grand scheme, but the baseball wasn't the actual purpose of the club. That was its stated goal, its intended purpose, but that club, that wall, it was load-bearing.

Groups of families watched the games, intermingled with their neighbors, met new people, and organized events. Of course something could take its place (there's nothing special about baseball stands, after-mass doughnuts, or a D&D group as forums for community cohesion) but without it, the whole system may begin to fall apart.

There are so many dimensions to a problem that we try to ignore, but they're always there.

So much of our society relies on assumptions we make about the world. These assumptions are what enable us to predict the future. We assume that, given no change in extraneous events, X cause will indeed lead to Y result, but this is never the case long-term.

Examples of this abound: laws written for one context are applied to others even if they're a bad fit, institutions which spring up and thrive in one era calcify and die in a later one, entire governments arise, prosper, and fall when the underlying technologies they depend on change or evolve. These are the hidden assumptions that power our world and they're the ones we ignore in order to build simple models of the future.

A mass on a spring slides back and forth on a smooth table according to our models because we exist in a world wherein it could be so. Tables have to be made by a company whose existence is protected by a stable government. They have to be smoothed, finished, and shipped to a lab wherein they can be connected to a well-manufactured, uniform spring and mass under the watchful eye of an educated scientist and their well-fed, attentive graduate student.

The simplest experiments have hundreds of hidden variables, none of which are in the eventual model—and for good reason—but our extraordinary ability to predict the behavior of the world depends on hidden, load-bearing walls and it's when those walls, those assumptions, begin to fail and break down that our models and our societies go with them.

The Current Dilemma(s)

When systems degrade it's often not because of their easily quantifiable benefits, but because of the weakening of their underlying load-bearing walls.

“The current nation-state could not possibly exist in a world without the printing press,” Ball told me. “It couldn’t exist without the current telecommunications infrastructure. The nation-state is built dependent upon the macro-inventions of the era in which it was assembled. A.I. changes all of this in ways that are hard to describe and kind of abstract.”
I suspect they won’t remain abstract for long.
- Ezra Klein, The Future We Feared Is Already Here

The ongoing devolution and polarization of American Democracy has left our institutions fragile and crumbling, and the decline of American hegemony, upon which rests the foundation of the post-war order, means that the rest of the world feels the strain along with us.⁵ Americans, broadly speaking, are divided and bitter and so far nothing seems to be able to pull us back from the brink.

What's more the A.I. revolution is ongoing and it's upending our social fabric by further loosening the trust we have in the words and images we see online. As with any technological revolution, it has many upsides (otherwise we wouldn't be doing it) but it comes with downsides that are less easily quantifiable and certainly more foundational.

As Ball notes, the printing press arguably enabled democratic experimentation in the early-modern period. Linda Colley expands on this topic in her book, The Gun, the Ship, and the Pen where she argues that writing and especially written constitutions (born in a new age of literacy and the Enlightenment) were critical technologies that allowed for the creation of modern liberal democracy.⁶

Writing, literacy, truth, and trust in a shared reality are the hidden, and very much load-bearing, walls of the western democratic world order, and without them those institutions will struggle to bear the weight of this century's challenges.

Though it's important to remember that not all load-bearing walls are good.

Historians debate whether the existence of a rivalry between the ancient Romans and the Carthaginians was in some way load-bearing for the stability of the former's republican system. The defeat of Carthage at the end of the Punic Wars left Rome without a clear enemy abroad and so they found one at home. Systemic oppression has also historically been load-bearing for many societies. This, of course, does not mean that these walls should not be broken down, but it does mean that we need to explicitly identify and replace the good parts while purging the bad.

What's more the feedback loops present in the real world sometimes mean that the roof collapses years after the fact. By the time it does, the walls are long gone and it's too late to replace them. All we can do is live with the consequences while we work to dig ourselves out. Julius Caesar crossed the Rubicon in 49 B.C.E., nearly a century after the Punic Wars had ended.

We'd do well to identify and fix what we can before the roof falls in. But to do that we need to know which walls are load-bearing and understand the loads they bear.

Remodeling is Hard

Anyone who's gone through an extensive remodel knows that the process, while it may be worth it in the end, is incredibly frustrating and always takes longer than you think. Replacing a load-bearing wall is an act that must be undertaken with intent.

There is much in the world, and in ourselves, that relies on the foundational bulwark of these load-bearing walls and so we should work to be aware when we are working to tear them down and we should put in the work to build sturdy replacements.

Personally, that might involve introspection about yourself, your motivations, and the foundational institutions we share as human beings in community with each other. As a society it might mean building up shared systems of trust and mutual accomplishment, systems where we can build toward something positive, together, and give us the opportunity to find a shared future in this world. It might also mean taking the time to appreciate a simple game night with friends for as long as it may last.

¹ For certain definitions of "long term".
² As every physics professor ever has always suggested.
³ To see how this happened to Robert Boyle during a 20 year long experiment(!) see The Foundations of Newton's Alchemy (p 86).
⁴ Traditions, no matter how silly, are usually load-bearing. Maybe those ways are trivial, but likely not.
⁵ For all its many failings, the Post-War Order™ has been remarkably good at its original goal of preventing Great Powers from engaging in nuclear war. I do not want to be around when this load-bearing wall finally gives way.
⁶ Liberal as in classical liberal (i.e. freedom, liberty). Not "left".

A new level of cowardice

Sat, 07 Mar 2026 03:00:23 +0000

I heard all about the American torpedo attack that sunk the Iranian frigate, Dena, in the Indian Ocean. I couldn’t miss it — multiple channels on YouTube were replaying the footage over and over again. There’s an Iranian warship sailing along, when suddenly it was struck in the stern by a massive explosion that lifted the vessel out of the water, breaking it’s back and leading to its rapid sinking. It was glorious war footage, I guess.

Then Pete Hegseth is on all the news, bragging about the victory and all those dumb, blind Iranian sailors who met their fate at the hands of the brave American navy.

At least 87 sailors were killed in the torpedo attack in international waters in the Indian Ocean, and the Sri Lankan navy responded to the Dena’s distress call and rescued 32 survivors, but 61 members of the crew are still missing. The U.S. didn’t respond to the call, and Secretary of Defense Pete Hegseth crowed about the attack to reporters on Wednesday.

“An American submarine sunk an Iranian warship that thought it was safe in international waters. Instead, it was sunk by a torpedo,” Hegseth said, calling it a “quiet death.” As a result of the attack, an Iranian supply tanker that was also near Sri Lanka, the IRIS Bushehr, has taken refuge in the island country.

The news stories, until now, didn’t mention one horrible, ignoble fact.

The U.S. Navy’s attack on an Iranian frigate, the IRIS Dena, on Wednesday was the first time an American submarine has sunk an enemy ship since World War II. But the Dena may not have been armed because it was returning from an international exercise in the Indian Ocean, and the U.S. Navy likely knew it because it was taking part in the same exercise.

Both the United States and Iran were taking part in the MILAN 2026 exercise, organized by the Indian Navy, on February 15–26, with the U.S. sending a maritime patrol aircraft and Iran sending the Dena. Iranian sailors from the ship paraded on land before India’s president.

The exercise in question required ships not to carry any ammunition. Normally, the Dena carries various missiles and guns, including anti-ship missiles. Because the U.S. also took part, it would have been aware that the Dena was unarmed. Former Indian Foreign Minister Kanwal Sibal accused the attack of being “premeditated as the US was aware of the Iranian ship’s presence in the exercise.”

To add to the United State’s shame, the IRIS Dena immediately sent out a distress call; the Sri Lankan navy responded to rescue survivors. The American navy, despite obviously having a ship nearby, ignored the SOS.

Pete Hegseth is a motherfucking chickenshit coward. “Warfighter,” my ass.

Message Passing Is Shared Mutable State — Causality

Fri, 06 Mar 2026 03:08:49 +0000

Something about the way we write concurrent programs has always felt wrong to me. When I pick up a new language and look at its concurrency model I get the same uneasy feeling. The APIs change, the terminology changes, but the underlying patterns look strangely familiar.

Maybe you’ve felt this too. The tools get better, the abstractions get nicer, but the core problem never seems to go away.

Any software developer who has tackled concurrency in a serious project has the battle scars of dealing with the pitfalls of multi-threaded and concurrent programs: the touchy, often clunky APIs and synchronization mechanisms, the dread of debugging data races and deadlocks, and the brain-bending non-locality of it all.

It’s taken me a while to understand what feels so off about them to the point I can articulate it, but I think I’m finally ready. Let’s start with a somewhat recent language: Go.

The Prediction

In 2006, Edward Lee published The Problem with Threads. His argument was stark: threads are “wildly nondeterministic,” and the programmer’s job becomes pruning that nondeterminism rather than expressing computation.

But Lee went further than criticizing threads: he argued that the shared-memory vs. message-passing debate was a false dichotomy. Both approaches model concurrency as threads of execution that need to be coordinated. Switching the coordination mechanism from locks to messages doesn’t change the underlying model, it changes the syntax of failure.

At the time, this was a contrarian position, and the mainstream languages were moving the other way.

The Experiment

Three years later Go launched with a concurrency philosophy built on the opposite bet. “Do not communicate by sharing memory,” the Go documentation urged. “Instead, share memory by communicating.” Channels (typed, first-class message-passing primitives) were Go’s answer to the concurrency mess.

Go wasn’t a toy. Backed by Google, it was adopted by the infrastructure that runs the modern internet: Docker, Kubernetes, etcd, gRPC, CockroachDB. These systems are among the most heavily used Go codebases in existence and are maintained by experienced teams with extensive code review and testing practices. Tens of thousands of developers wrote concurrent code following Go’s guidance, using channels instead of mutexes or locks, sharing memory by communicating.

It was the most prominent, well-resourced, real-world test of the message-passing hypothesis the industry has ever run.

The Results

In 2019 Tengfei Tu and colleagues studied 171 real concurrency bugs across these flagship Go projects and published Understanding Real-World Concurrency Bugs in Go. The findings were striking: message-passing bugs were at least as common as shared-memory bugs.

Around 58% of blocking bugs (i.e. goroutines stuck, unable to make progress) were caused by message passing, not shared memory. The thing that was supposed to be the cure was producing the same problems as the disease.

To be clear, message passing does eliminate one important class of concurrency bugs: unsynchronized memory access. If two goroutines communicate only through channels, they cannot simultaneously mutate the same variable. But eliminating data races does not eliminate coordination failures. Deadlocks, leaks, protocol violations, and nondeterministic scheduling remain.

Go ships with a built-in deadlock detector, but it only caught 2 of the 21 blocking bugs the researchers tested. Two. The race detector fared better on non-blocking bugs, catching roughly half, which still means half the concurrency bugs in production Go code are invisible to the tools that were designed to find them.

Most of these bugs had long lifetimes. They were committed, shipped, ran in production, and weren’t discovered until someone happened to trigger the right interleaving. Testing didn’t find them, and code review didn’t find them. Instead they hid in some of the most heavily scrutinized Go codebases.

Lee’s prediction that switching the coordination mechanism wouldn’t address the root cause was confirmed.

The Code

Here’s a simplified bug in Kubernetes from the paper. A function spawns a goroutine to handle a request with a timeout:

func finishReq(timeout time.Duration) ob {
    ch := make(chan ob)
    go func() {
        result := fn()
        ch <- result  // blocks forever if timeout wins
    }()
    select {
    case result = <-ch:
        return result
    case <-time.After(timeout):
        return nil
    }
}

If fn() takes longer than the timeout then the parent returns nil and nobody ever reads from ch. The child goroutine blocks on ch <- result and will never be cleaned up. Go garbage-collects objects, but it doesn’t garbage-collect goroutines blocked on channels that will never be read.

In Kubernetes (the system managing your production container infrastructure) every one of these leaked goroutines hangs onto references and never lets the memory go. Under load, they accumulate, and the process will slowly degrade until it crashes or gets OOM-killed. This is a reliability failure in the software responsible for keeping your other software running, caused by a single missing buffer in a channel.

The fix is one character: change make(chan ob) to make(chan ob, 1).

Now look at the same logic in Java:

BlockingQueue<Result> queue = new ArrayBlockingQueue<>(1);

new Thread(() -> {
    Result result = computeResult();
    try { queue.put(result); }   // blocks if queue is full
    catch (InterruptedException e) { }
}).start();

try {
    Result result = queue.poll(timeout, TimeUnit.SECONDS);
    if (result != null) {
        return result;
    } else {
        return null;
        // thread still running, still blocked on put()
        // queue object still holds a reference
        // nothing will ever clean this up
    }
} catch (InterruptedException e) { return null; }

No Java developer would look at this and say “I’m doing message passing.” They’d say “I’m using a shared concurrent queue,” because BlockingQueue lives in java.util.concurrent, right next to Mutex and Semaphore. They’d know it carries all the risks of shared mutable state.

But this is the Go channel code. Same shared mutable data structure, same blocking semantics, same bug. If the timeout fires then nobody consumes from the queue and the producer blocks forever. The thread leaks. The structure is identical, the only thing that changed is the vocabulary.

In Java, we call this a shared concurrent queue and we understand the risks. In Go, we call it a channel and pretend it’s something different.

Message passing is often presented as an alternative to shared mutable state, but in practice it frequently reintroduces shared coordination structures under another name.

Why This Keeps Happening

Arthur O’Dwyer, writing about the paper, identified what he called the “original sin” of Go channels: they aren’t really channels at all. A channel has two distinct endpoints, a producer end and a consumer end, with different types and capabilities. If the last consumer disappears, the runtime can detect it, unblock producers, and clean up.

A Go channel has none of this. It’s a single object, a concurrent queue, shared between however many goroutines happen to hold a reference. Any goroutine can send, and any goroutine can receive. There are no distinct endpoints, no directional typing, no way for the runtime to detect when one side is gone. It is a mutable data structure shared between multiple threads, where any thread can mutate the shared state by pushing or popping.

Once you see this, the bug categories in the study become predictable rather than surprising. Every classic failure mode of shared mutable state has a channel equivalent:

Deadlock. Goroutine A sends to a channel and waits for a response on another. Goroutine B does the reverse. Both block. This is a circular dependency on shared state, i.e. the same structure as a mutex deadlock but expressed through queues instead of locks. These issues were found in Docker, Kubernetes, and gRPC.

Leak. Nobody reads from a channel, so the sender blocks forever. The shared queue retains a reference to the goroutine, preventing cleanup. The Kubernetes bug above is this pattern: a resource leak caused by a dangling reference to shared state.

Race. If multiple goroutines read from the same channel, which one gets each message? The answer is nondeterministic: the runtime’s scheduler picks one. This is concurrent access to a shared resource, with the nondeterminism mediated by the scheduler instead of explicit locking. The paper documents these in etcd and CockroachDB.

Protocol violation. A goroutine sends a message the receiver doesn’t expect, or sends on a closed channel (which panics in Go), or closes an already-closed channel. The shared object’s implicit contract was violated, the same category of bug that shared mutable state has always produced.

Every one of these is a classic shared-mutable-state bug wearing a message-passing costume.

And this isn’t just a Go problem. Message passing as a concurrency model doesn’t eliminate shared state, it relocates it. The data being communicated may transfer cleanly from sender to receiver, but the communication mechanism itself (channel, mailbox, or message queue) is a shared mutable resource. And that resource inherits every problem shared mutable state has always had.

Even Erlang demonstrates this. Erlang processes are genuinely isolated with separate heaps, no shared references, and messages copied between processes. These are the strongest form of the message-passing guarantee available anywhere, and yet researchers found previously unknown race conditions hiding in Erlang’s own heavily-tested standard library.

The races clustered around ETS tables, Erlang’s escape hatch from pure actor isolation, which are shared mutable storage that exists because the pure actor model didn’t meet performance requirements. The safety model promised isolation, yet reality demanded a shared mutable escape hatch. The escape hatch reintroduced exactly the bugs the model was supposed to prevent.

Message passing solves concurrency bugs the way moving your mess from one room to another solves clutter.

So Now What?

When a Go programmer hits a channel deadlock and considers reaching for a mutex, they’re choosing between two approaches that fail for the same structural reason. “Go channels are fine if you use them correctly” is a true statement. So is “mutexes are fine if you use them correctly.” They’re the same statement.

Lee saw this in 2006. The shared-memory vs. message-passing debate is an argument about which coordination mechanism to use. It has never questioned whether we’re even asking the right question.

If both sides of the dichotomy fail then maybe the dichotomy itself is wrong. Maybe the problem isn’t which tool we use to coordinate concurrent execution. Maybe there’s something deeper about the foundation that both approaches share, something we haven’t questioned yet.

I think there is. Some languages have tried different foundations and attacked aspects of the problem with real insight, but none of them have fully broken through to the mainstream. It’s worth exploring why, so that’s where we’re headed.

References

Lee, Edward A. “The Problem with Threads.” IEEE Computer 39.5 (2006): 33–42. PDF
Tu, Tengfei, et al. “Understanding Real-World Concurrency Bugs in Go.” ASPLOS 2019. PDF
O’Dwyer, Arthur. “Understanding Real-World Concurrency Bugs in Go.” Blog post, June 6, 2019.
Christakis, Maria and Konstantinos Sagonas. “Static Detection of Race Conditions in Erlang.” PADL 2010. PDF

The Brand Age

Thu, 05 Mar 2026 19:17:12 +0000

March 2026

In the early 1970s disaster struck the Swiss watch industry. Now people call it the quartz crisis, but in fact it was a compound of three separate disasters that all happened at about the same time.

The first was competition from Japan. The Swiss had been watching the Japanese in the rear view mirror all through the 1960s, and they'd been improving at an alarming rate. But even so the Swiss were surprised in 1968 when the Japanese swept all the top spots for mechanical watches at the Geneva Observatory trials.

The Swiss knew what was coming. For years the Japanese had been able to make cheaper watches. Now they could make better ones too.

To make matters worse, Swiss watches were about to become much more expensive. The Bretton Woods agreement, which since 1945 had fixed the exchange rates of most of the world's currencies, had set the Swiss Franc at an artificially low rate of .228 USD. When Bretton Woods collapsed in 1973, the Franc shot upward. By 1978 it reached .625 USD, meaning Swiss watches were now 2.7 times as expensive for Americans to buy. [1]

The combined effect of foreign competition and the loss of their protective exchange rate would have decimated the Swiss watch industry even if it hadn't been for quartz movements. But quartz movements were the final blow. Now the whole game they'd been trying to win at became irrelevant. Something that had been expensive — knowing the exact time — was now a commodity.

Between the early 1970s and the early 1980s, unit sales of Swiss watches fell by almost two thirds. Most Swiss watchmakers became insolvent or close to it and were sold. But not all of them. A handful survived as independent companies. And the way they did it was by transforming themselves from precision instrument makers into luxury brands.

In the process the nature of the mechanical watch was also transformed. The most expensive watches have always cost a lot, but why they cost a lot and what buyers got in return have changed completely. In 1960 expensive watches cost a lot because they cost a lot to manufacture, and what the buyer got in return was the most accurate timekeeping device, for its size, that could be made. Now they cost a lot because brands spend a lot on advertising and use tricks to limit supply, and what the buyer gets in return is an expensive status symbol.

That turns out to be a profitable business though. The Swiss watch industry probably makes more now from selling brand than they would have if they were still selling engineering. And indeed, when you look at the graph of Swiss watch sales by revenue, it tells a different story than the graph of unit sales. Instead of falling off a cliff, the revenue numbers merely flatten out for a while, and then take off like a rocket in the late 1980s as the surviving watchmakers come to terms with their new destiny.

It took the watchmakers about 20 years to figure out the new rules of the game. And it's interesting to watch them do it, because the completeness of their transformation makes it the perfect case study in one of the most powerful forces of our era: brand.

Brand is what's left when the substantive differences between products disappear. But making the substantive differences between products disappear is what technology naturally tends to do. So what happened to the Swiss watch industry is not merely an interesting outlier. It's very much a story of our times.

Jaeger-LeCoultre's web site says that one of their current collections "takes its inspiration from the classic designs of the golden age of watchmaking." In saying this they're implicitly saying something that present-day watchmakers all know but rarely come so close to saying outright: whatever age we're in now, it's not the golden age.

The golden age was from 1945 to 1970 — from the point where the watch industry emerged from the chaos of war with the Swiss on top till the triple cataclysm that struck it starting in the late 60s. There were two things watchmakers sought above all in the golden age: thinness and accuracy. And indeed this was arguably the essential tradeoff in watchmaking. A watch is something you carry with you to tell you the time. So there are two fundamental ways to improve it: to make it easier to carry with you and to make it better at telling the time.

Obviously accuracy is valuable, but in the golden age thinness was if anything more valuable. Even in the days of pocket watches the best watchmakers tried to make their watches as thin as they could. Cheap, thick pocket watches were derided as "turnips." But thinness took on a new urgency when men's watches moved onto their wrists during World War I. And since thinness was more difficult to achieve than accuracy, it was this quality that tended to distinguish the more expensive watches of the golden age.

There is one other thing watchmakers have pursued in some eras: telling more than the time in the usual way. Telling you the phase of the moon, for example, or telling the time with sound. In the industry the term for these things is "complications." They were popular in the nineteenth century and they're popular again now, but except for one pragmatic complication (showing the date), they were a sideshow in the golden age. In the golden age, as always in golden ages, the top watchmakers focused on the essential tradeoff. And, as always in golden ages, they did it beautifully. The best watches of the golden age have a quiet perfection that has never been equalled since. And for reasons I'm about to explain, probably never will be.

The three most prestigious brands of the golden age were the so-called "holy trinity" of Patek Philippe, Vacheron Constantin, and Audemars Piguet. Their prestige was mostly deserved; they had earned it by the exceptional quality of their work. By the 1960s they stood on two legs, prestige and performance. And what they learned in the next two decades was that they had to put all their weight on the first leg, because they could no longer win at either of the two things watchmakers had historically striven to achieve. Quartz movements were not only more accurate than any mechanical movement, but thinner too.

The holy trinity at least had another leg to stand on. Most of the other well-known Swiss watchmakers sold only performance. None of those companies survived intact.

Omega showed what not to do. Omega were the nerds of Swiss watchmakers. They made wonderfully accurate watches, but they would have been ambivalent, at best, about the idea of being a luxury brand. When the Japanese got as good as the Swiss at making accurate movements, Omega responded in the Omega way: make even more accurate movements. They introduced a new movement in 1968 that ran at a 45% higher frequency. In theory this should have made it more accurate, but the new movement was so fragile that it destroyed their reputation for reliability. They even tried to make a better quartz movement, but there was nothing down that road but a race to the bottom. By 1981 they were insolvent and were taken over by their creditors.

Patek Philippe took the opposite approach. While Omega was redesigning their movements, Patek was redesigning their cases. Or more precisely, designing their cases, because until then they hadn't.

This is probably the point to mention what a strange beast the Swiss watch industry was in those days. It was a kind of capitalism that's hard to imagine today, and even then could only have been made to work in a country like Switzerland — a network of small, specialized companies locked into place by regulation. The companies that we for convenience have been calling watchmakers were merely the consumer-facing edge of this network. The holy trinity didn't design their own cases, or even their own movements most of the time.

In 1968 (that year again) Patek Philippe launched a new watch that shifted the center of gravity of case design. This time they'd taken their own designs to the casemakers and said "this is what you're going to make for us." The result was a striking new model called the Golden Ellipse. Somewhat confusingly, because it wasn't elliptical. The new case was more of what UI designers would call a round rect: a rectangle with rounded corners. And this new family of watches was quite successful. But it was more than that: it was the pattern for the future. [2]

How could merely designing a distinctive case be so important? Because it turned the entire watch into an expression of brand.

The trouble with the best watches of the golden age, from the point of view of someone who wanted to impress people with the brand of watch he was wearing, was that no one could tell what brand of watch you were wearing. Until you got within a few inches of them, the watches of all the top makers looked the same. That's the thing about minimalism: there tends to be just one answer. Plus the watches of the golden age were small by present standards. Watchmakers had spent centuries working to make them smaller, and by 1960 they'd gotten very good at it. So the only thing distinguishing one top brand from another was the name printed on the dial, and dials were so small that these names were tiny. The manufacturers' names on the holy trinity's golden age watches are between half and three quarters of a millimeter high. So by taking over the case, Patek expanded the size of the brand from 8 square millimeters to 800.

Why did they suddenly decide to make their brand shout, after a century of whispering? Because they knew they weren't going beat the Japanese on performance. From now on they'd have to depend more on brand.

There's a cost to doing this, which we can see even in this early example of case-as-brand. Golden Ellipses are not bad looking. They must have looked even cooler in the 1970s, when designers were turning everything into round rects. But the Golden Ellipse was not an evolutionary step forward in case design. Watches didn't all become round rects. Watchmakers had already discovered the optimal shape for the case of something that describes a circle as it rotates.

They had also discovered the optimal shape for the crown, the knob on the side of a watch that you turn to wind it. But to emphasize the distinctive profile of the Ellipse, Patek made the crown too small, with the result that they're distractingly hard to wind. [3]

So even in this early example we see an important point about the relationship between brand and design. Branding isn't merely orthogonal to good design, but opposed to it. Branding by definition has to be distinctive. But good design, like math or science, seeks the right answer, and right answers tend to converge.

Branding is centrifugal; design is centripetal.

There is some wiggle room here of course. Design doesn't have as sharply defined right answers as math, especially design meant for a human audience. So it's not necessarily bad design to do something distinctive if you have honest motives. But you can't evade the fundamental conflict between branding and design, any more than you can evade gravity.

Indeed, the conflict between branding and design is so fundamental that it extends far beyond things we call design. We see it even in religion. If you want the adherents of a religion to have customs that set them apart from everyone else, you can't make them do things that are convenient or reasonable, or other people would do them too. If you want to set your adherents apart, you have to make them do things that are inconvenient and unreasonable.

It's the same if you want to set your designs apart. If you choose good options, other people will choose them too.

There are only two ways to combine branding and good design. You can do it when the space of possibilities is enormously large, as it is in painting for example. Leonardo could paint as well as he possibly could and yet also paint in a style that was distinctively his. If there had been a million painters as good as Bellini and Leonardo this would have been harder to do, but since there were more like ten they didn't bump up against one another much. [4]

The other situation when branding and good design can be combined is when the space of possibilities is comparatively unexplored. If you're the first to arrive in some new territory, you can both find the right answer and claim it as uniquely yours. At least at first; if you've really found the right answer, everyone else's designs will inevitably converge on yours, and your brand advantage will erode over time.

Since the space of watch design is neither unexplored nor enormously large, branding can only be achieved at the expense of good design. And in fact if you wanted one sentence to describe the current age of watchmaking, that one would do pretty well.

Patek Philippe didn't know for sure that making visibly branded watches would work. It was not even their only strategy, at the time. They were finding their way. But it was the strategy that did work, at least as measured by revenues.

For it to work the customers had to meet them halfway. Patek knew that not all their customers were buying their watches for the performance they delivered — for their accuracy and thinness. They knew that at least some customers were buying them because they were expensive. But it was unclear how many, or how far they could be pushed.

To encourage them, Patek did something that none of the holy trinity had done much of before: brand advertising. And what they talked about was how expensive their watches were. A 1968 Patek ad explained "why you are well advised to invest perhaps half a month's income" in an Ellipse. "Like every Patek Philippe," the ad continued, "this thin model is entirely finished by hand. Since a Patek Philippe is the costliest watch to make, production is severely limited: only 43 watches are signed out each day for delivery to prominent jewelers throughout the world." [5]

You can tell this is an early ad because they still mention thinness. But there is no mention of accuracy. Presumably Patek felt that battle was already lost.

The next move was made by Audemars Piguet, who in 1970 commissioned the renowned designer G

A Very Stable War - The Atlantic

Tue, 03 Mar 2026 22:47:55 +0000

After the president of peace, a man who felt deserving of the Nobel Prize, authorized a massive aerial bombardment of Iran last summer, the task of explaining away the contradiction fell to J. D. Vance.

“I certainly empathize with Americans who are exhausted after 25 years of foreign entanglements in the Middle East,” the vice president told NBC—a stark understatement, given that Vance had, up to this point, unsparingly denounced Middle East wars and promised that the Trump administration would avoid them. “I understand the concern, but the difference is that back then, we had dumb presidents, and now we have a president who actually knows how to accomplish America’s national-security objectives.”

The difference was simple: Other wars were bad because they were led by dumb presidents, but a Trump war would be good because Donald Trump is smart.

Yet after the administration’s second wave of air attacks on Iran, the president’s strategy seems more sundown than Sun Tzu.

Read: From ‘America First’ to ‘Always America Last’

The Wall Street Journal reported on January 30 that Trump was planning a major military campaign, but was still “debating whether the main aim is to go after Iran’s nuclear program, hit its ballistic missile arsenal, bring about the collapse of the government—or some combination of the three.” Generally speaking, military strategists tend to first settle upon their objective, and then devise a tactic to achieve it. The Trump method of first deciding on the tactic, and only getting around to what he wants to accomplish afterward, is unorthodox.

A lack of clarity has continued to define the operation. In his videotaped message announcing the latest attacks, Trump repeated his boast that the previous round of air strikes, in June, had “obliterated the regime’s nuclear program at Fordow, Natanz, and Isfahan.” But if the obliteration lasted only half a year, what value is there in re-obliterating it? Will biannual bombing campaigns be employed until Iran submits to American demands?

Because Trump appears not to know what he wants from his military strikes, he is also unable to make clear what he is demanding of Iran to stop them. At times, he faults Tehran for refusing to abandon its nuclear ambitions. “They wanted to do it. They didn’t want to do it. Again they wanted to do it. They didn’t want to do it,” he complained in his taped address. This tracked with the rationale that Vance offered last summer: “Simple principle: Iran can’t have a nuclear weapon.”

Yet at other times, Trump has floated more extensive ambitions. “Our objectives are clear,” he announced at the White House yesterday, ticking off a four-point list: eliminating Iran’s conventional missiles, destroying its navy, and ending its ability to fund terrorism, along with the original nuclear-nonproliferation goal.

The simple principle is now four simple principles. And Trump’s complaint that the mullahs were getting too “cute” in the negotiations has been mixed with grievances going back decades. On Sunday he told my colleague Michael Scherer, “They want to talk, and I have agreed to talk, so I will be talking to them.” The next day, he described Iran’s government as “sick and sinister,” language implying a desire for regime change.

Despite planning the war with the keen brainpower that Vance finds so impressive, Trump has conceded his apparent failure to anticipate some of its more predictable consequences. He told Jake Tapper in an interview that Iran striking Arab countries allied with the U.S. was “the biggest surprise.”

In another interview, with ABC’s Jonathan Karl, Trump lamented that he had picked out several candidates to lead a more pliant Iranian government, but the bombing campaign had killed them all. When planning a war to install a puppet regime, a smart president, or even one of average intelligence, would grasp the importance of not killing the puppets beforehand.

The other distinction Vance has drawn between the dumb Middle East wars of history and Trump’s extremely smart one is that the current president would maintain a tight timeline and would not, under any circumstances, entertain a ground presence. “We have no interest in a protracted conflict. We have no interest in boots on the ground,” Vance said.

George Packer: Hubris without idealism

Then Trump declined to rule this out. “I don’t have the yips with respect to boots on the ground—like every president says, ‘There will be no boots on the ground.’ I don’t say it,” he told the New York Post. Trump has said the campaign could run four to five weeks, but in his White House remarks, he said it could “go far longer than that.”

Trump also tried to disabuse his audience of any concern that his attention would wander from the task. “They said, ‘Oh, well, the president wants to do it really quickly. After that, he’ll get bored,’” he said yesterday. “I don’t get bored. There is nothing boring about this.” Within moments, Trump was riffing about the White House drapes.

In the absence of any clear strategic account of his actions, a raft of less flattering ones suggest themselves. Among them: Trump’s worldview was formed in the 1970s, and he never got past his anger with Iran over the Carter-era hostage crisis. Having lost out on the Nobel Peace Prize, he may be attempting to gain more recognition by reclassifying his presidency from the Nelson Mandela category to the Genghis Khan category. The Venezuela coup has disinhibited him from trying regime change, and now he seems eager to knock off as many hostile governments as he can. Even the Iran operation’s official name, Epic Fury, implies that it is carrying out an emotional reaction rather than a cold-eyed plan.

The usefulness of Vance’s rationale is that it can justify anything. Trump could not be starting a dumb war, because that would mean Trump is a dumb president. And who could possibly think that?

The stranger secret: how to talk to anyone – and why you should | Social etiquette | The Guardian

Mon, 02 Mar 2026 22:28:07 +0000

It started with two incidents on the same day. In a fairly empty train carriage, a stranger in her 70s approached me: “Do you mind if I sit here? Or did you want to be alone with your thoughts?” I weighed it up for a split second, conscious that I was, in effect, agreeing to a conversation: “No, of course I don’t mind. Sit down.”

She turned out to be an agreeable, kind woman who had had a difficult day. I didn’t have to say much: “I’m sorry to hear that.” “That’s tough for you.” She occasionally asked me questions about myself, which I dodged politely. I could tell she was only asking so the conversation would not be so one-sided. Some moments are for listening, not sharing. I sensed, without needing to know explicitly, that she was probably returning to an empty house and wanted to process the day out loud. I didn’t feel uncomfortable, as I knew I could duck out at any moment by saying I needed to get back to my phone messages. But instead we talked – or, rather, I listened – for most of the 50-minute journey. I registered that it was an unusual occurrence, this connection, but thought little more of it. A small part of me was glad this kind of thing still happens.

That evening, I ate at a restaurant with my family. As the waitress brought the bill, we chatted and I learned that she was from Seoul. She was shy and softly spoken. We talked gently about Korean food and what she missed about home. Once again, I thought little of this exchange.

As we walked home, my 15-year-old son asked: “Is it OK to talk to people in that way?” “What way?” He was asking about the boundaries when it comes to talking to someone about their home country.

This was a very good question. How do you know, generally, what the terms are of a conversation with a stranger? I realised that there is a sort of unwritten code you learn as you get older, which enables you to assess whether a conversation is a good idea or not. I thought about the woman who had approached me earlier. How did she know it was OK to talk to me? In the end, I replied to my son: “You don’t always know if it’s OK. Sometimes you have to take the risk and find out.”

Then it struck me. A lot of people have given up taking a chance on other people: that they might want to listen, that they might want to talk. But they have also given up taking a chance on themselves: that they might be able to navigate a conversation with someone new, cope with knockbacks and steer a path through any misunderstandings.

Everybody needs good neighbours. Photograph: Harold M Lambert/Getty Images

The disappearance of these kinds of interactions from day-to-day life – in pubs, restaurants, shops, queues, on public transport – is striking. I have been talking to people tangentially about this for the past 10 years, ever since I started researching my book, How to Own the Room, which came out in 2018 and went on to become a podcast. This project was supposed to be about public speaking and confidence. But I realised from people’s reactions to the topic – especially younger people – that their deepest anxiety lies elsewhere, in something much more banal and inexpressible. Forget “public speaking”. What a lot of people don’t like at all any more is “speaking to anyone in public”.

Many reasons are cited: state-of-the-art don’t-talk-to-me headphones, mobile phones and social media generally, the rise of working from home, the introduction of touchscreens in takeaway restaurants so you barely interact with a human, the death of third spaces, the pandemic. In the end, the biggest excuse becomes “social norm reinforcement”. This is the idea that if no one talks to you, you don’t talk to anyone either. A casual conversation in a waiting room where no one else is having a casual conversation suddenly sounds not very casual at all.

On an individual level, some people perfectly understandably cite neurodivergence, introversion, inability to tolerate eye contact or an intense loathing for small talk (especially about the weather) as reasons to avoid these conversations. It’s certainly true that this time six years ago – at the height of lockdown – it would have been rude and unsafe to start a chat, let alone sit next to someone on a train. But now? It can feel as if everyone is still adhering to the 2-metre rule, employing “the tech shield” or even “phantom phone use” (pretending that you need to be on your phone when you don’t).

This goes deeper than adolescent angst or personal preference. And possibly deeper than our overreliance on phones. We are losing a basic human skill. The ability to speak to others and understand them is being compromised.

Rail passengers at Waterloo East, London, using a variety of ways to avoid talking to each other. Photograph: Windmill Images/Alamy

Dr Jared Cooney Horvath, a teacher turned cognitive neuroscientist who focuses on speech, has warned that gen Z is the first generation in history to underperform the previous generation on cognitive measures. And Dr Rangan Chatterjee, a bestselling author and father of two teenagers, said in an interview this month: “I think we’re raising a generation of children who have low self-worth, who don’t know how to conduct conversations.”

It’s not only affecting young people. The psychologist Esther Perel calls it a “global relational recession”. She writes: “The point is not depth. The point is practice, the gentle strengthening of our social muscles.” On her YouTube channel she recently introduced the topic of Talking to Strangers in 2026.

Something that used to come naturally is now a subject of longing and fascination, as if it were a rare anthropological phenomenon. Videos are springing up on social media, cataloguing encounters with the unknown “other”: earnest, well-meaning, wholesome videos, under the categories “social anxiety”, “extrovert” and “talking to strangers”. Many have the unstated theme of “out and about in the big city”. Some are personal experiments, often extremely ill-advised ones. Can you challenge yourself to tell a joke to an entire train carriage? What happens if you go up to an older woman and tell her she looks beautiful? The (usually young) person doing the filming is often trying to improve themself in some way or attempting to “be braver” or “less socially anxious”. The camera acts as their accountability partner. The people they’re talking to are relegated to the role of “task to be ticked off the list”. Either that or there’s a push towards a Hallmark card effect: “Look, other people are not as horrible as you thought.” (Cue swell of trending motivational audio.)

The trouble with these social media experiments, of course, is that they are performative and individualistic. There’s an element of commodification: the encounter must be ripe for digital packaging. Often it’s not clear if the filming is consensual. The connections are one-way and border on the exploitative or manipulative. They are designed for individual personal growth or free, self-directed therapy (“this made me more confident”) and for clicks and voyeurism (“check out this person’s reaction”). The effect is to make “talking to absolutely anyone” seem even more alienating, fake and narcissistic. This has spawned a secondary genre of parody videos such as the comedian Al Nash’s “A cup of tea with a stranger – an amazing conversation!” In this clip, an irritating interviewer passes tea to a stranger on a park bench under the guise of “helping you with your loneliness”, only for the encounter to turn awkward when the stranger accidentally drops the cup and smashes it.

Take your cue from Mr Hewitt and Mr Boucker, shown here having a neighbourly chat in 1957. Photograph: Picture Post/Getty Images

It’s only natural to fear rejection, humiliation, giving offence or overstepping a boundary when we initiate a conversation – or even when we respond to someone else’s attempt. But according to a study by the University of Virginia (Talking with strangers is surprisingly informative), we overstate these fears in our minds: “People tend to underestimate how much they’ll enjoy the conversation, feel connected to their conversation partner and be liked by their conversation partner.”

The key is to lower the stakes. Make it less of a big deal. Don’t focus on what could go wrong. Also, don’t focus on how amazing this could be. You are just saying, “It’s cold today, isn’t it?” You are not asking someone to join you on a quest for world peace. Similarly, if an approach is made towards you and you don’t want to respond, just be confident and clear either with your gestures (look down, don’t make eye contact) or with speech: “I can’t talk right now.”

In her work on kindness, the University of Sussex psychologist Gillian Sandstrom calls these conversational gambits “small, humanising acts”. It’s important to emphasise the “small” aspect. Sometimes I think people are overwhelmed by the “bigness” in their mind of the fear of interaction, and how disproportionate that seems next to the “smallness” of the pathetic reality. Don’t read too much into passing moments. Trust yourself to read social cues and work out how you stand in relation to them. Know yourself and your own personality. Not everyone wants to talk and not everyone wants to be talked to. And that’s OK. It can depend on the day and on your mood. Give yourself get-out-of-jail-free cards in these conversations. If someone doesn’t respond, assume they didn’t hear you or they’re having a bad day. If someone talks to you and you feel uncomfortable or you’re having a bad day, it is not your job to be kind or nice. If their attempt was well meant, they’ll get over it. We don’t need to avoid each other. But we also don’t have to be on niceness autopilot all the time.

In any case, our worst fears about these interactions are rarely realised. Last year, the team of Stanford psychologist Prof Jamil Zaki, the author of Hope for Cynics: The Surprising Science of Human Goodness, put up posters around campus with messages about approachability and warmth. They found that what students most needed was permission – the reminder to “take a chance”. They concluded: “Too often, we’re sure that conversation and connection will exhaust us, or that we can’t count on others.” In our minds, we paint people (and ourselves) as profoundly disappointing. They – and we – are rarely that bad. And even if they are, it will make a good story to tell later to the people who are not strangers to you.

Supermarket shoppers in Rushden, England, observing the 2-metre rule in the first Covid lockdown, 2020. Photograph: David Rogers/Getty Images

Is it going to change your life if you talk to someone in a shop about the prospect of rain? Probably not. But in light of the current state of the world, even the slightest possibility of brightening someone’s day is valuable. It’s certainly worth the punt. Perhaps the way they respond matters less than the fact that you retained your humanity enough to try something, to risk, to connect.

Small talk may not profoundly alter your life. But its absence will profoundly alter human life as we know it. We live in a world of intense and often unnecessary division. Small talk is a tiny, free and very possibly priceless reminder of our shared humanity. If we intentionally give up talking to strangers, if we purposely decide to give in to the phone shield, the consequences will be horrible. Arguably, we are already on the verge of doing this. Let’s back up and start a conversation before it’s too late.

Roastidio.us User Picks KjQKkQVg

Legibility is Ruining You

Legibility is Ruining You

The Values You Truly Hold

The Coercive Force of Legibility

Value Capture

Thin Rules, Thin Behavior

You

Hasan, Me & Racial Ambiguity | Çaki's Series Of Tangents

Hasan, Me & Racial Ambiguity

Case Study 1: Me

What does this mean?

It's weird.

Case Study 2: Lineker

Case Study 3: Hasan

Race exists in the minds of our beholders.

Your sign-up form is a weapon | Bytemash

What is subscription bombing?

How we spotted it

The pattern

Designed to be invisible

The damage isn’t to you

What we did

Firewall rules

Cloudflare Turnstile

Limiting email to verified users only

Wrapping up

Why Florence Started the Renaissance - by Tomas Pueyo

Brunelleschi’s Duomo

Leonardo’s Roman Proportions

Machiavellian City-States

Medici Wealth

The Origins of Florentine Wealth

Florence Style

A Change in the Reason behind Architectural Innovation

Renaissance Outside of Churches

So Why Did Florence Birth Michelangelo, Fibonacci & Galileo?

Collections: That Dothraki Horde, Part IV: Screamers and Howlers – A Collection of Unmitigated Pedantry

Where There’s a Whip…

Armored Arrogance

Dothraki Ends

Grass Wars

Raiding

The Preposterous Tactics of the Dothraki

Stagecoach (1939)

Conclusions

Share this:

Like this:

Miscellanea: The War in Iran – A Collection of Unmitigated Pedantry

The Situation

The Gamble

The Trap

Peace Negotiations?

Strategic Implications

Share this:

Like this:

Your bridge to wealth is being pulled up - Daniel Homola

"Collaboration" is bullshit.

The last Andy Weir movie I will ever waste money on

Share this:

The Valuation Function Problem: Why Skin in the Game Fails When It Matters Most - The Nukemblog

The Theory and Its Hidden Assumption

The Valuation Function Problem

History Didn’t Get the Memo

Why This Failure Is Predictable

The Forecasting Implication

What Actually Moves Behavior

A Closing Note

You have access to almost any book - BiteofanApple

The World's Library

Under-Appreciated, Under-Explored

We’ve got to stop sending files to each other – Terence Eden’s Blog

I’m OK being left behind, thanks! – Terence Eden’s Blog

Are They Still Your Friends if You Never See Them? - The Atlantic

填不满的锚点 | JustGoIdea

填不满的锚点

一

二

三

四