Today I'm chatting with professor and military historian Si Sheppard. I want to talk to you about the conquest of the Aztec and the Inca Empires because these are some of the most shocking events in human history, unparalleled among all the military and diplomatic achievements ever made.
A few hundred conquistadors show up to a new continent they know nothing about. Within a matter of years, they subjugate empires which have millions of inhabitants in them. In the process of doing so, they are often winning battles where they are outnumbered 100 to 1, 1,000 to 1, and in many cases not even taking a single casualty.
They're doing this while upwards of 99% of their fighting forces are made up of the very natives that they're going to subjugate. I wanted to have you on to talk about something this episode. We're not going to focus on it, we're mostly talking about history here, but people think of AI takeover as this crazy sci-fi scenario but this would not be the first time in history that an alien force, with motivations that the natives don't understand, is able to use diplomatic cunning and technological advantages to outmaneuver and then take over a much larger and well-established order.
I wanted to have you on to understand what happened here. First, I would love to hear the stories. What did Cortés do? What did Pizarro do? How did these conquests actually go down?
Thank you. It's a great pleasure to be here. Yes, we're talking about the birth of our modern world, which played out between 600 to 500 years ago and created the modern globalized community that we all participate in today.
It began in the 15th century when Europe, backed up against the wall of the Atlantic, under immense pressure from external forces and cut off from the core economic power bases of the East — the sources of the spice and silk that had driven the global economy historically throughout Eurasia — was finally forced to innovate. Key names like Prince Henry the Navigator of Portugal work on turning the Atlantic Ocean from a barrier into a superhighway.
Generational knowledge accrued from captains reporting on tides and wind currents and whatever land they've managed to reconnoiter, building out the map as they go along, enables them to find a way around the southern tip of Africa directly to India and therefore cut out the middleman.
The Spanish, meanwhile, are heading west. We all know the story of Christopher Columbus: "In 1492, he sailed the ocean blue." That was entrepreneurship. This was an expedition to arrive at the Oriental markets by traveling west.
Of course, his figures were completely wrong. He was saved by the fact that, unknown to him, there was an entire continent in the place where he thought Japan, China, and the Spice Islands would be. Initially, that was perceived, ironically, as a disappointment. But over time, it became increasingly apparent this was a source of potential in land, in wealth, and in people that could be exploited.
A key turning point happened in 1511 when an ambitious Spanish nobleman, de Velázquez, seizes control of Cuba. He sends out from this power base expeditions exploring the coast of what we would call Mesoamerica, Central America.
Later in 1518, he assigns that task to a man with whom he's had a somewhat tempestuous relationship, a man named Hernán Cortés. He'd arrived in the New World in Cuba in 1511. He had some legal training, so he served as a notary and worked his way up to being alcalde, which is chief magistrate, on Cuba. But he had much greater ambition.
He takes this warrant from the governor and runs with it. He begins recruiting men to a much wider and much more ambitious goal: not just to explore the coast, but to penetrate into the interior, to find out what's there, and to seize control of it.
When the governor realizes what Cortés is up to, he tries to stop him. There's this great moment on 18th November 1518. He arrives at the waterfront just as Cortés gets into a boat to go out to his ships offshore. The governor calls out to him, "What are you up to?" Cortés assures him, "Don't worry. I have this all under control." But of course, the governor had lost control at that point. Now it was Cortés's expedition.
This is the background. Again, as you say, he is literally going off the side of the map. He's entering terra incognita, unknown to Europeans previously. His political antenna, honed by years in the islands, enables him to appreciate at a very early stage that there is a significant imperial power which has hegemony here in the mainland, and that its subject peoples chafe under the authority of whoever this lord or imperial power is.
He arrives in the lands of the Totonacs, who had recently been incorporated into the imperial orbit of a people who call themselves the Mexica — hence Mexico as we know it — but who have come down to history based on their mythical origins in a land called Aztlán as the Aztecs. For simplicity's sake, that's what we'll refer to them as.
The Totonacs resented this authority, as all subjugated and subordinated peoples do, and they saw the arrival of these outsiders as a potential opportunity to throw off the authority that had been imposed on them. Having established this relationship with the Totonacs, Cortés begins to march inland, and he enters the territory of the Tlaxcalans, who, like an autoimmune system, immediately rallied to defeat him, to expel him from their territory.
But he fights them to a standstill for such a period of time that the Tlaxcalans and he arrive at a new kind of arrangement. The Tlaxcalans again appreciate his strength, his power, his potential. They swing over to his side and will henceforward serve as guides and reinforcements for his expedition deeper into the interior.
Can we pause here and talk about the numbers? How many men had Cortés brought with him? Even the Tlaxcalans had much larger numbers than Cortés did. So even the standstill is an impressive military achievement, especially because he's fighting without native support. What are the numbers even at this point?
A few hundred on Cortés' side and multiple thousands on the Tlaxcalans' side. It's at this point that the critical technological differentials from the Old and New World really begin to come into effect. The classic text which got me invested in this whole scenario in the first place is Jared Diamond's book, Guns, Germs, and Steel. He enumerates in the title those critical advantages that the Spaniards had.
Guns in reality weren't so significant. Once the Native peoples got over the shock value, they appreciated, given the very slow reloading times involved in firearms at that time, that they weren't so significant as assets on the battlefield.
But steel was critical. Local weapons were carved from wood and from stone and from obsidian, that volcanic glass which is razor sharp when honed to a point and quite dangerous as a slashing weapon. But against the steel armor, and especially the steel helmets that the conquistadors were wearing, it was much less effective.
The conquistadors were drawing on military traditions of the Old World that went back generations, incorporating long Toledo steel sword blades. They used those as thrusting weapons to take out indigenous warriors who were coming in with cut-and-slash weapons, which of course open up the side of the body to be taken out very early. The combination of the technology with generations of inherited military training and skill meant that the conquistadors had that critical edge on the battlefield.
Now, they weren't invincible. Cortés was again playing those mind games. He was taking casualties, and he was careful to bury his dead underneath the houses that they were sheltering in so the Tlaxcalans wouldn't be aware that they were even having some successes in hand-to-hand combat. It was as much a mind game as it was a physical game.
The ultimate weapon, of course, that the conquistadors brought with them was their horses. I cannot overemphasize how vital the horse was to the conquistadors once they entered into the interior of the American continents. They had the external asset. They had sail power, which enabled them to land anywhere they wanted on the coast. But in the interior, the horse was decisive. There were no native equivalents.
The horse originated in North America, but it had subsequently died out. The local peoples had not just no horses, but no beast of burden generally. That was not integrated into their economies, into their ways of life, or into their military.
The Inca, whom we'll discuss in a moment, did have one, which was the llama. When they campaigned, trailing for miles behind their armies would be these endless baggage trains of pack animals, llamas. The Aztecs had no equivalent. Their only domestic animals were the turkey and the Chihuahua.
Speaking of dogs, the conquistadors also brought these big war mastiffs, these fighting dogs which wore armor and were trained to take down men. Again, imagine the terrifying impact that would have for the indigenous people. But the horse was the critical asset. On attack, you could blast through any native formation which didn't have not just the weapons, but the training or discipline to be able to stand up to massed shock cavalry.
If things were going bad for the conquistadors, they could retreat much faster than, of course, the enemies could chase them. The Tlaxcalans appreciate all these assets coming together, and they see this as their opportunity. They were the holdouts, the last remaining major entity in Mesoamerica which was still able to go toe-to-toe with the Aztecs.
They'd been completely surrounded. The Aztecs were upping the ante every year with these ceremonial flower wars, which were becoming more and more obviously wars of conquest. The Tlaxcalans had been cut off from the coast and from sources of salt. So they were being pushed to the brink. This is a Hail Mary for them. "If we ally with these outsiders, these strangers, who are just few in number after all, we may have some potential way to break out of this strategic trap that has been imposed upon us."
They lean over to the side of Cortés, and they travel with him to Tenochtitlan, the capital, the glittering jewel in the midst of Lake Texcoco, the capital of the Aztec Empire. Again, this is a dramatic moment when Cortés passes through the causeways out into the city and encounters Moctezuma II, the huey tlatoani as he was called, the Great Speaker, the emperor of the Aztec Empire. These two worlds come together at that moment.
The interface between them was Cortés's interpreter/mistress, Doña Marina, Malinche as she became known, who is translating from one side to the other. The dialogue that occurs then is one of those great tipping points in history.
What was the perspective of Moctezuma II?
He was aware Cortés was coming. This is his empire. He has envoys and emissaries and scouts and spies everywhere. They're reporting back to him, "Something has washed up on the shores of our empire, and it is determined to meet with us." He initially bargains with Cortés, urging him to leave. He provides him with precious gifts, which of course just inflames Cortés even more. Once he's got hold of Aztec gold, he's just determined to go all the way to the center.
Moctezuma continues playing this double game. His advisors, his council, urge him not to allow Cortés to proceed to the capital. His own brother, Cuitláhuac, tells him, "If you allow this man inside your house, he will take your house away from you." But Moctezuma proceeds, and for that he's been condemned by generations of historians. This was a mistake to bring the conquistadors into the heart of the empire itself.
I actually think that was his best option, whether he was aware of it or not, because it was better for him to have Cortés as his guest within Tenochtitlan than it was to have Cortés, this random element, at loose within his empire, talking to the subject peoples, inflaming them possibly against central authority. He felt perhaps Cortés was what he said he was, an envoy from a great emperor on the far side of the vast ocean.
Cortés, again with the mind games, had no warrant to say he was a diplomat or ambassador for anybody. He kept telling anyone who'd listen, "I speak for the mighty Charles V, the Holy Roman Emperor." Of course, Charles V had no idea who Hernán Cortés was. Cortés, again, is just winging it, just incredible improvisation here.
And for a moment… stasis. The great turning point comes when word arrives from the coast that one of the Aztec regional military officers, on his own initiative, has taken action against the conquistadors left behind on the coast, the ones with the Totonacs, and some of them have been killed.
Cortés seizes this as his pretext. He marches into the palace of Moctezuma II, the Aztec huey tlatoani, and tells him that for his own safety, he is now being taken under the protective custody of the Spanish, and he should come with him, or else there will be violence. This is the moment when Moctezuma's entire worldview just collapses.
Nobody had spoken to him like this before. His psychology just disintegrates at that point, and he meekly allows himself to be led away. He tells his people, "Don't resist. I don't want bloodshed. I will pass with these strangers."
Then once Cortés has rounded up the officers from the coast responsible for turning on his men there, he has them, in front of the emperor and the entire court, burned alive, which is a form of punishment totally unknown to the Aztecs. They watch this in grim silence.
Once Cortés has seized the emperor himself, once he has Moctezuma, he already has the empire. He governs through what is increasingly becoming his puppet. Moctezuma seems to have fallen almost under the spell of Cortés, the classic Stockholm syndrome. He gives him a grand tour of the city. Cortés, in again a subtle flex of his power, has brigantines, small sailing boats, built. He takes Moctezuma with him on a sailing cruise on Lake Texcoco, and the emperor's astonished at this new technology of using the wind to sail on the water.
Having the emperor was the key. The emperor's mistake wasn't inviting Cortés into the city. It was making himself vulnerable by staying there also in an adjacent palace. If he had gone to some further location, made some excuse, a pilgrimage or an expedition of some kind, and had ruled through subordinates, Cortés would've been baffled. He couldn't go chasing after the emperor. That would be obviously too provocative.
But without the emperor there, Cortés would have had his options extremely limited. I don't know what he would've done in that circumstance. But because Moctezuma stayed in his capital, he made himself the target. Once he was taken, this was the key to controlling the entire empire at that point.
Obviously the story continues from there, with La Noche Triste. What people might not appreciate about what happened here, in both the Pizarro story of the Inca Empire which we'll talk about later, and in the Aztec story, is that in both cases the Spanish were themselves divided and in fact fighting each other.
So maybe you should tell the story of how Velázquez sends another expedition to capture and arrest Cortés for going beyond the prerogative of his mission.
We left the governor on Cuba, of course, fuming. He rallies another expeditionary force because Cortés is stealing his thunder. He sends this under Pánfilo de Narváez, a larger force than the one Cortés brought with him originally. This lands on the coast. Cortés appreciates he has to move, so he leaves one of his subordinates, the hot-blooded Alvarado, behind in charge in Tenochtitlan, takes as many men as he can to the coast, confronts de Narváez, and defeats him in a minor battle.
Again, diplomacy wins over the entire expeditionary force that the governor had dispatched. He tells them, "My friends, why are we fighting? We have the empire at my disposal. Join with me and we can share in the riches." So he wins over the vast majority of this new force. It effectively becomes reinforcements, and he marches back to the capital.
He finds the mood has turned against him. The problem with being in charge is you can't be everywhere. His subordinate, assuming that the Aztecs were mobilizing against him, decided on a preemptive strike. There was a festival. The aristocracy, particularly the warrior caste, were celebrating, and Alvarado just descended on them and massacred everybody.
The survivors rally the population. When Cortés returns, he finds himself effectively under siege. The people are so appalled now with the failure of their huey tlatoani that when he tries to pacify them, they throw stones at him. Mortally wounded, he either dies or the Spanish finish him off, because he's no longer of any use to them.
Now Cortés is trapped in Tenochtitlan. He determines to break out, and that is, as you say, La Noche Triste, the Tragic Night, because so many of his men are left behind. They're either killed as they try to escape the city across the causeways or forced to retreat back to the palaces, where they're taken alive and sacrificed subsequently. Cortés gets away with a minority of the troops still at his disposal and makes it back to Tlaxcalan territory.
That was the critical aspect, the two features of the entire campaign. One, he has enough strike power with enough horses to defeat the Aztecs when they try to finish him off, at the Battle of Otumba. They feel he's on the run. They have the numbers. The conquistadors are dragging their tails between their legs as they flee. Cortés himself is grievously wounded. "We can finish them off." They had the numbers. They were winning.
But again, Cortés identifies where the Aztec general is. He rallies enough horsemen to blast through the Aztec ranks and take out their command and control. The Aztec army disintegrates, and the Spanish are able to flee. The battlefield advantages, their horses, their steel, were critical, but just as critical was the diplomacy. If the Tlaxcalans had turned on Cortés at that point, he would've been finished. Not one Spaniard would've survived. None of them would've made it back to the coast.
But the Tlaxcalans, for their own reasons, not out of any generosity or humanitarian impulse to welcome back the conquistadors, still see, "We are committed to these people and to our alliance." So they give Cortés shelter. He recovers physically, begins to rebuild his strength, and is determined to go back to Tenochtitlan for round two: "This time we're not going to be indulging diplomatic niceties."
The Tlaxcalans have their own perspective on this. If you go to Tlaxcala, the territory today, the governor's mansion there has murals on the wall, and these depict Cortés and the conquistadors from the Tlaxcalan point of view. It's fascinating to look at how the relationship is not between a superior and an inferior. It's between equals. You see them depicted in the same scale, engaged in diplomacy between partners, not between subordinates and dominant figures.
The Tlaxcalans drove a hard bargain. "We want territory carved out of the Aztecs. We want our own fortress in Tenochtitlan after this is over." They saw this as them using the conquistadors, not the other way around, which is very important to bear in mind.
The Aztec Empire from a distance, if you look at a map, seemed impressive. But this was an imperial construct, and not just the remaining independent entities like the Tlaxcalans, like the Tarascans, who we can discuss later, were hostile to the Aztecs. So were so many of their subordinated peoples. The Totonacs on the coast, the Chalco states on the shores of Lake Texcoco, they chafed under Aztec hegemony and authority.
They too saw the conquistadors as an opportunity to finally shrug off the authority and the demands of the Aztecs, and get revenge in a lot of ways. Cortés builds up for the final campaign against Tenochtitlan, against the new huey tlatoani, Cuitláhuac, and then, after he died, his brother, the last of the huey tlatoanis, Cuauhtémoc. He has indigenous people flocking to his banner from all over Mesoamerica. He emphasizes that he's still in the fight. He still has those advantages.
More reinforcements are arriving at the coast, and they're being integrated into the conquistador army. But of the force that marches on Tenochtitlan, the conquistadors were maybe 0.5% in raw numbers. The overwhelming majority were other Mesoamerican peoples who see this as their final reckoning with the Aztecs. This is the integration again of the technology with the diplomacy.
Cortés knows that to take the city, he has to control the lake. Again, the innovations… They need sulfur for their gunpowder. Their guns are useless without gunpowder. So an expedition goes into the mountains to the volcano, Popocatépetl, and they dangle with ropes inside the caldera of the volcano, scooping out the sulfur they need as the critical ingredient for their gunpowder.
Another one builds prefabricated brigantines, boats, in their different sections, and then has untold numbers of porters, human load-bearers, bring these prefabricated pieces over the mountains to the lake. They're assembled and let loose in the lake, and now they have control over the lake. Bringing together the technology and the innovations with the local manpower, this is the force directed against Tenochtitlan in the final battle.
I actually didn't know those stories. That's crazy. One of the ways in which, to the extent people are even aware of the story, they might not understand what actually happened, is that they have this model of the Aztecs or the Incas as this cohesive single nation. A better model might be something like, imagine Russia and Ukraine have been doing trench warfare for 100 years. Imagine the animosity, and then also the Russians are taking tens of thousands of Ukrainians and doing human sacrifices with them. Then some foreign force comes in and says to the Ukrainians, "We can help you march into Moscow." I think they'd have pretty decent odds of getting some agreement there.
The old phrase is, "The enemy of my enemy is my friend," and in diplomacy and war, that's always very true.
I don't think I actually appreciated the ambition of Jane Street's intern projects until last week, when I got to see some of them for myself. For example, one intern spent part of his summer researching LLM memorization, figuring out how to quantify and reduce it without hurting out-of-sample performance. This matters because whenever Jane Street evaluates a new model on historical data, they need to know that its performance reflects predictive power, not just memorization of past outcomes.
Another intern built a custom activation checkpointing scheme. Activation checkpointing lets you trade compute for memory: you recalculate some activations during backprop instead of storing them all after the forward pass. This intern found a way to identify the activations that are cheapest to recompute, but that still free up a lot of memory. And this allowed his version to outperform PyTorch's native one.
These are just two examples, but Jane Street has written about a bunch more. These write-ups are coming out in batches. The first few are live now, including these two, and the rest will be published in the coming weeks. If you want to read them, or if you want to apply for Jane Street's 2027 summer internships, go to janestreet.com/dwarkesh.
There are so many interesting questions to ask, everything from technology to diplomacy to disease to tactics. Before we get to that, just to round out the story, in many ways what's striking about the conquest of the Incas in Peru under Pizarro 10 years later is how similar it was to Cortés's story, how much he just followed the playbook. I want people to have a sense of roughly what happened here, starting with the Battle of Cajamarca, which might be the most astonishingly lopsided military encounter that has maybe ever happened in human history. If you can recapitulate at least that battle a bit, and then the overall story, then we can do some analysis from there.
Absolutely. When Cortés consummates his conquest of Tenochtitlan in 1521, the last of the huey tlatoanis surrenders to him, and he occupies the territory and, most importantly, the top of the social hierarchy. The existing structure, as much as possible, is allowed to persist. The conquistadors just decapitate the local nobility and insert themselves at the top.
Now, this is the playbook that will continue to be applied, less successfully, throughout the rest of Mesoamerica because the existing political entities there aren't so sophisticated. Nonetheless, there are rumors — there are always rumors — of lost cities of gold. More conquistadors are now flooding into the New World to seek advantage of this. One of these is a distant cousin of Cortés named Francisco Pizarro. He met with Cortés and got the playbook, as you say, about how this is to be done.
Then, in one of these conquistador business ventures, he finds the funding for men and ships and begins pushing south down the west coast of South America, crossing the equator and finally arriving in what is now modern-day Peru. He encounters what he's looking for: a sophisticated culture on the scale… in fact, in terms of territory, on a much broader scale than Cortés had encountered in Mesoamerica. This is the land of the Inca.
When he arrived, there was civil war. The previous Sapa Inca, Huayna Capac, had died in 1525, and his two sons, Atahualpa and Huáscar, turned on each other. Atahualpa had just won the civil war, and he was proceeding to Cuzco, the Inca capital, to be formally enthroned as the new Sapa Inca. That's when word arrives to him that strangers have entered his realm. Again, we're talking 168 men who have pushed into the Andes Mountains and are seeking him out for reasons unknown.
He encounters them outside a place called, as you say, Cajamarca in 1532. He sees not threat, but opportunity. Just like Moctezuma II had been advised not to treat with these people, so does Atahualpa's envoy, who reports to him, "These are not men of good faith. We can take them out. They tend to sleep together. At night, we can set fire to wherever they're staying and just take them all out in one fell swoop." But Atahualpa is interested in what they have to offer him.
He's seen the horses. In fact, Pizarro made a specific case of impressing the Sapa Inca with his horses. He sends one of his key lieutenants to put on a display of horsemanship. Famously, he charges right at the Sapa Inca and all his court. The horse stops just inches from Atahualpa's face, snorting and breathing. Atahualpa is absolutely stony-faced and professes to be unimpressed and says, "I will meet with your master at a time of my choosing."
After the Spaniards leave, he has all of his court, including his generals who flinched or showed fear, executed just to make the point, "We're not to be intimidated by these people." He sends a blocking force, tens of thousands of men, to cut off any retreat, and then he proceeds to meet with Pizarro at Cajamarca with the vast expanse of his army.
There's a saying in Peru that Atahualpa was planning to have Pizarro for lunch, but Pizarro ended up having him for breakfast. Once he enters into the confines of a narrow, constricted courtyard space, that's when Pizarro lets off the ambush. He lets fly with all his artillery and all his musketry for the shock value, and unleashes his horses. Atahualpa himself is overwhelmed and taken prisoner at that point.
This again is the playbook, the copybook, as you say. Go to the top. Seize whoever is at the very pinnacle of the social hierarchy. You control the entire structure.
Just so people know the numbers here, the Incan Empire overall has some 10 million inhabitants. During this first battle where Pizarro encounters the emperor, he has some 160-190 conquistadors. I don't think the numbers are known for exactly how many Atahualpa had, but he's facing off against the army that has just clinched the civil war in the Inca Empire, and it's tens of thousands big. I've seen some numbers that are like 100,000 or 200,000, but certainly in the high tens of thousands. So 100-something against almost 100,000.
That's the first battle. What is astonishing is that these 100 conquistadors kill… I saw some numbers that are like 10,000 of the enemy, or at least thousands of Atahualpa's troops. They suffer zero casualties. So they kill close to 10,000 of the enemy and suffer zero casualties. This is the first encounter that the Inca Empire has ever had with the conquistadors. That must have been shocking from both sides' perspective.
If you can do some analysis here, how is it possible? How can you have a military outcome where you have literally a KD ratio of infinity: 10,000 to zero?
The conquistadors themselves admitted that the night before that encounter, there were so many campfires from the Inca army scattered across the hills surrounding Cajamarca that it looked like the twinkling of stars at night. The conquistadors themselves admitted that when they saw the number of Inca troops filing into this courtyard, they were losing control of their own bladders out of fear. They knew what the risks were here. But the shock value of that initial impetus was enough to carry them through.
The Inca had never heard gunpowder weapons fired in any format before. The ancestral founder of their empire was Pachacuti, which means the thunder bearer. But to see a real-world exemplar of that must have been beyond shocking. The actual physical damage that the cannons and arquebuses inflicted was relatively minor, but the explosion and the fact that these strangers seemed able to control the thunder must have been psychologically devastating.
Then it came down again to that steel and those horses. When Pizarro unleashed the horses, this is something that the Inca… As you say, these are veteran warriors, the winners of a civil war, strategists at a tactical level within the framework of pre-Columbian Andean warfare. They were the top dogs, but this is something outside of their entire understanding of the universe. They'd never seen a horse before or knew what it could do if you integrate a rider into the saddle with stirrups and the shock value of that charge. They had no doctrine to push back against that.
The psychological impact was that they tried to flee. They were piling up against the walls in their desperation to scramble over them, to the point where the walls finally gave way under the press of all those bodies. Some of them escaped into the countryside, and the Spaniards rode after them, just running them down as they fled. The only thing that stopped the massacre at the end was that night was falling, and the Spaniards in the darkness couldn't continue the pursuit any longer.
It's what happens when something totally outside the universe of even veteran warriors plays out in a way that they have no mental appreciation of, or are unable to draw on any kind of library of how to respond to the situation. The fight or flight instinct completely takes over, and flight was the winner in this instance.
As you say, the only casualty was Pizarro himself. He's reaching for Atahualpa to take him prisoner, and one of his more enthusiastic comrades is slashing and manages to cut his hand as he does that. That was the only injury any of the conquistadors suffered at this critical moment.
Give me a sense of how it is that, even if they're on horses, they're just so impenetrable. Intuitively, you would think that if you have tens of thousands of troops, can't you just swarm a cavalry? Many of them will die, but just pull that guy in his steel armor off the stirrups, take him down, and then kill him.
Europeans knew how to do that. The Swiss, for example, had already proved that a mounted knight could be knocked out of the saddle and defeated if you have men shoulder to shoulder, locked in place with long pikes to keep those horses at a distance and break up the impetus of their charge. But that requires, one, the technology and, two, the discipline, training, and institutional knowledge to do that.
Mesoamerican warfare was very loose in its formation, warriors with a lot of space between them. It's the worst thing that you can do in regards to cavalry, because the horses can take advantage of that looseness to just blast through any formation that's coming at them.
Now, one depiction of the pre-Columbian societies is that they were somehow childlike, innocent, and couldn't have responded. They did try to. It's important to bear that in mind. At every level, both the Aztec and the Inca sought to put in place countermeasures to negate the advantages that the conquistadors had. One of the critical exemplars is confined space.
The Aztecs, when Cortés showed up in 1521 for round two, drew them into their city, which was a network of islands connected by bridges and causeways. That minimized the advantages of the horses, because that's not good horse country. They learned to fight from rooftops and pelt the enemy with stones or whatever they could from heights. They built man-sized holes in buildings so they could slip from one to the other where the horses couldn't chase them. So they learned to use those kinds of advantages.
They learned the stopping power of water. Interestingly, both did this completely unconnected. They had no relationship with each other. Against Cortés, when he was taking one of the smaller cities on Lake Texcoco, Iztapalapa, the Aztecs broke the dikes in the lake and flooded the city, which took out a number of the conquistadors and their allies. In that environment, of course, the metal armor works to the disadvantage of the conquistador. It drags them to the bottom. If they'd waited until nightfall when the Spaniards were sleeping, they might have bagged all of them.
The Inca too, later, when the conquistadors moved on Ollantaytambo, flooded the river. They allowed it to burst its banks and again bogged down the conquistador cavalry. So they were adapting. They took captured enemy weapons. They learned to fix conquistador swords to the end of long wooden poles, improvising pike tactics. Again, without being instructed to, they just intuited that keeping horses at a distance with long spears would be a way to negate the advantages they had.
The Inca employed the bolas, which they'd used for hunting the great birds of South America, to tangle the legs of horses. In one famous incident, an Aztec warrior allowed himself to be speared by a conquistador and gripped hold of the lance as it was in his body, fixing the conquistador in place so his colleagues could swarm and take him out of the saddle. So they were trying to improvise.
The Aztecs used their captured Spanish prisoners to learn how to use captured crossbows. The Inca, when they captured Spaniards, would use them to give instruction on how to use these firearms. And Manco Inca, the leader of the Inca resistance, made a point of learning to ride a horse himself to demonstrate to his people, "We can do this." They were trying, but the learning curve was so steep, and the timeframe to internalize foreign technology and foreign tactics was too short for them to really be able to deploy it to their advantage.
This is worth emphasizing. Cortés takes over the Aztec Empire in what is now Mexico in two and a half years. A couple hundred men, an empire of 6 million people, two and a half years. Pizarro does the same thing for the Inca Empire 10 years later in about five years. It's insane how fast these conquests happened, and just how little time they had to adapt, even though in principle they could have figured out ways to combat cavalry and things. All the while, their whole social structure is disintegrating, there's disease, et cetera. There's all this psychological shock value.
Now, the obvious question to ask at this point, before we go to the other ones, is if you reran the simulation again and again, and you just changed a few things here and there — changed the leader for both the Aztecs or for the Spanish in these conquests — in what fraction of the simulations do the Spanish end up in control of these empires anyway within a matter of a few years? Because when you first learn about them, it sounds like an outrageous and impossible outcome that a few hundred men would take over empires of millions of people. When you learn more about it, you begin to think, "Well, I guess it couldn't have been any other way." But I don't know if that is your opinion.
In the grand scheme of things, there are multiple wheels that define history. The grand wheels that grind slowly are geography and climate, and then the small wheels in the foreground are the individuals, the key players themselves. This is an integration of that process, the big wheels with the small ones.
Yes, if Cortés had been taken out early, which could have happened on multiple occasions, the conquistadors, leaderless and rudderless, may well have just disintegrated, and the conquest may have been aborted at that level. With Pizarro again, Atahualpa's critical mistake was inserting himself into Pizarro's trap. If he had sent envoys to keep the negotiations at a distance, he may well have been able to frustrate Pizarro's ambitions. As the head of what remained an integrated Inca Empire, he may have been able to use those numbers to win the campaign in the short term.
When the Inca did finally turn on the conquistadors, Francisco Pizarro had seized Cuzco and stripped the temples of their gold and silver — "the sweat of the sun and the tears of the moon," as it was called — and shipped it back to Europe. He goes back to the coast to set up what would become the city of Lima, the modern capital of Peru.
The danger with Cuzco is the fact that it's elevated and away from the coast, which is where the Spaniards, with their absolute monopoly over naval power, have the advantage. His problem was he left his half-brothers in charge back in Cuzco. They provoke Manco Inca, who'd been set up as Pizarro's puppet ruler, into outright rebellion. He rallies his people. He lays siege to Cuzco. He takes Sacsayhuamán, the great fortress that dominates Cuzco, and he has one of his generals, Quizu Yupanqui, cut off reinforcements arriving from the coast.
The Inca could win set-piece confrontations if they used their terrain and their knowledge of it to their advantage. Pizarro from Lima on the coast is taking reinforcements and pushing them into the interior to relieve Cuzco. Quizu is waiting till they enter narrow defiles and passes through the mountains, blocking both ends, and then using boulders and other material to just obliterate them from the heights. More and more of these Spanish expeditions are being annihilated.
At that level, when they use the terrain and their institutional knowledge of it to their advantage, the Inca are able to win set-piece battles, but they can't seal the deal in Cuzco. Manco makes the mistake of ordering Quizu to attack Lima. He feels, "We have the advantage now. I've bottled them up in Cuzco. We've annihilated their relief columns. Take out their city on the coast, and the problem is over." But attacking Lima meant passing across open ground, where the Spanish then unleashed their advantage. The general is killed, and the whole assault collapses.
Back in Cuzco, yes, there's a handful of Spaniards holed up in just a couple of fortified buildings, desperately holding on. Day and night, they're under assault: the psychological aspect of it, the beating of drums, the chanting, and the constant rain of missiles to set their buildings on fire. They would've been wiped out except for two reasons. One, the open courtyards and plazas of Cuzco are good terrain for cavalry, and they still have their horses.
And secondly, critically, the asset through diplomacy of the local peoples — the subordinated mitimaes, they were called, or yanakuna, serfs, the defeated and subordinated people the Inca had brought forcibly within their imperial structure, chafing under those authorities — who see these outsiders as their chance to finally break free of Inca authority.
So strangely enough as it may seem, more local peoples kept joining the besieged Spaniards during the whole Battle of Cuzco, which played out over many, many months. The Spaniards end up with more allies than they started with during this battle, because there were so many people within the Inca Empire — from the Cañari, who were from Ecuador, to the Chachapoya from the Amazon Basin, to even the Aymara, who used to be rivals with the Inca for control of Cuzco itself — who increasingly side with the Spaniards.
That local manpower was the critical asset. Without that, yes, they would've been wiped out at Cuzco. Again, it's the Spaniards inserting themselves into an imperial orbit, which from a distance looks impressive with its geographic expanse and its total population, but which had been ruthlessly dictated to by a tiny elite of a hierarchy who had imposed themselves over the entire system.
Remember, only a small number of people spoke Nahuatl, the imperial language of the Aztecs. Only a small number spoke Quechua, the imperial language of the Inca themselves. It was the Inca Empire, but it was constructed from multiple different ethnicities. When they saw their chance, they took it to side with the Spaniards.
It's worth remembering that the Aztecs and the Incas were themselves the equivalent of the conquistadors in their own world. There's 100,000 or so actual Incas that are controlling an empire of 10 million people. So they are themselves an imperial force.
Going back to the question of how contingent these victories were, it sounds like you're saying they are contingent, because you're pointing out all these cases in which, if things had gone slightly differently, the siege of Cuzco would've worked, or Cortés would've been killed in battle.
The ultimate factor comes down to those background forces, the great wheels that grind through history. Geography and climate structure one other factor. The fact that Eurasia was so stacked with domesticable animals — sheep, cows, goats, mules, pigs, and above all, horses — and human interaction in biomes with those domesticated animals, created the context for the evolution of contagious diseases, endemic or pandemic, that ripped back and forth through Eurasia. We're familiar with all their names.
That never occurred in the New World. The people there had no acquired immunity to diseases such as the big one, smallpox. But beyond that, there was a whole galaxy of communicable diseases that we're now familiar with as being universal. Even if the Aztecs and Inca had won the short-term proximate victories, the longer-term demographic crisis would have undermined the functional capacity of their societies to remain integrated and to resist much longer.
The pressure within them from their subject peoples, as they felt the authority of their overlords beginning to weaken, would've led to their fragmentation, which is the overarching cycle of pre-Columbian societies generally. They're extremely impressive, in fact, more impressive when you consider the disadvantages they had to struggle with. Their pyramids were built not through the assistance of beasts of burden or even the wheel, but simply through pure manpower and the application of that to construct them.
The subject peoples would have broken for freedom, and we see the recurring cycle of Mesoamerican history, for example. When Cortés arrived, Tenochtitlan was one of the greatest metropolises perhaps on Earth: beautiful, rich, and heavily populated. 100 years later, if he'd arrived in 1619 as opposed to 1519, he would've found a swamp with moss-covered mounds and hills rising out of it.
The parallel to that, of course, is Teotihuacan, the previous great metropolis which dominated Mesoamerica. It rose. It expanded. It attracted a huge population. It overstressed its environment, and it collapsed. I think the Aztecs naturally would've gone the same way.
One way or another, the Spanish, or even subsequent waves of European colonists, would've entrenched themselves at the coast and would have perhaps more slowly pushed inland. It's possible that you would see a more balanced relationship between the colonists and the indigenous peoples, but I think in the longer term they all would've been folded into the grand European imperial framework.
So we've gotten to the point now where, if you have a computer-based task that a personal assistant could knock out in an hour, AI can probably just do it for you. AI has gotten more and more general over time, so these new AI assistants need much less hand-holding than they used to.
For example, xAI gave Grok Bot access to their contracts, vendor spend, and usage data and asked it to identify potential savings. It ended up finding over $100,000 worth, mainly made up of unused SaaS seats and unnecessary add-ons. I tried this on my own spend, and Grok Bot found more than I expected.
It went through all my SaaS tools and figured out which seats I should stop paying for by looking at the last active date for each user. It looked at my billing settings and told me which tools I should pay for annually instead of monthly. And it also messaged me about a couple of subscriptions it doubted that I was actually using, like LinkedIn Premium, and asked me if I wanted to cancel them, which I did.
All in, Grok Bot found over $5,000 of savings, even though we're a pretty small team. It seems quite likely that within six months I'll get to completely eliminate some of the most annoying and least productive-feeling hours of my week. You can already give real work to AI assistants like Grok Bot, which is making a bunch of small things in my life easier. Go to x.ai/bot to try it out.
We do know examples of other native forces who resisted European powers for a very long time. I think there's some platoon or army that has just finished fighting during the American Civil War that has to go down and then fight some native tribes. Even late into the 19th century, native forces were able to resist European incursions. Of course, we know about the Comanche and so forth.
It does seem to me that maybe just because they were so centralized, you get the emperor and the whole thing falls, whereas other tribes were more dispersed.
You're absolutely right. The key term there, as you used it, was dispersal. The vulnerability of the Aztec and the Inca was that hierarchical nature of their society and the absolute loyalty that the system had invested into whoever was atop the social hierarchical tree: the huey tlatoani for the Aztecs, the Sapa Inca for the Inca themselves.
Taking that individual destabilizes the whole system, and potentially even means that it can be functionally governed through the use of intermediaries or even puppets. The dispersed peoples, as you say, the Comanche, even the Maya in Mesoamerica, whose city-states jealously asserted and defended their autonomy against each other, meant that the Spanish couldn't take out the Maya in one fell swoop. They couldn't do to the Maya what they'd done to the Aztecs and the Inca. They had to reduce their cities one by one.
That was a long, laborious process that took almost to the end of the 17th century to finally consummate. Even then there was still smoldering dissent. In many ways, you can go to the highlands of Central America and they still speak the ancestral languages as the first one before they speak Spanish. Yes, colonizing powers can find aggressive bite-back.
Just look at Custer's Last Stand as late as 1876 here in America, the entire 7th Cavalry wiped out at the Little Bighorn because they lost the initiative. At that point, they were fighting warriors who had come together temporarily, riding horses, firing guns, and then dispersing afterwards. It was very hard to pin them down. That was often a frustrating aspect.
It took a long process of what amounts to really total war — setting up barbed wire everywhere and of course slaughtering the buffalo to take away the basis for their culture and their food chain — to finally bring the last holdouts to heel. So it can be a long, protracted process. The Inca and Aztecs themselves found that out. They never consummated even incorporating their entire zones within their empires.
The Aztecs had surrounded and were slowly reducing the Tlaxcalans. But to the west, in what's Michoacán, the Purépecha people, or as they're called today, the Tarascans, had an empire of their own, and they were able to go toe-to-toe with the Aztecs. The first time the Aztecs crossed into Tarascan territory, it was a disaster. A key reason for that was actually technology. The Aztecs had no connection with South America. The Tarascans somehow did. We know from linguistic evidence and costume choice there was some exchange of flora and fauna, and above all in technology. The Tarascans could smelt metal. Like the Inca, they could smelt copper, and
They used that to form the points of their arrows. Their massed archers with their copper-tipped arrows were decimating the invading Aztec forces. The exchange, if it had been on a broader scale between north and south, would've been potentially an equalizing force. If the Aztec had had llamas from South America, it would've been dramatically to their advantage.
There is tin in the New World. So no bronze? Why do we not get bronze with copper and tin?
They'd never made it, certainly not in Mesoamerica, where everything was stone. They had the worst of both worlds. They could smelt silver and gold to attract the greed of the Spaniards, but they couldn't smelt iron to hold them off.
But before that, you would've thought that they would've done copper, tin… get bronze.
They did have bronze in the Inca world. If the Aztecs had had access to that, again, it would have given them a much more aggressive fighting stance. Bronze can't stand up against iron, but it's better than stone. Or obsidian. This is an important part of what happens here, which is that the Europeans are learning from each other, but the natives are not learning from each other. Cortés learned from the conquest of Hispaniola. Of course, he has a ton of improvisation in the way he approaches the conquest of the Aztecs. Then Pizarro knows the whole details of what Cortés did a decade earlier and basically follows the same playbook. It's not clear to me that the Incas even knew the Aztecs existed. So you just have a much greater accumulation of knowledge than the other one is able to accumulate. They're closed worlds. They were very insular in their societies.
To go further back in time, the only divergence that would've given pre-Columbian peoples a real long-term chance to hold off the Europeans is if we reverse the clock another 500 years back to Leif Erikson and the Norse in Greenland. They knew of the existence of North America — Helluland, Markland, and Vinland, as they called them, as they explored them. They tried to colonize the New World. Their numbers were too few, the local resistance was too fierce, and they were forced out. If the Vikings had been able to claim a long-term presence in the New World, they may have been able to drip-feed European technology, European beasts of burden, and above all, European diseases slowly into the Americas, so that 500 years later, when the conquistadors did show up, they'd have met a people who had some resistance to European diseases, knew how to work iron, and were riding horses. If that was the case, all bets are off.
That would be such an interesting history. What does the New World look like if the Vikings had actually set up there? That's very interesting. What was Pizarro's greatest innovation on Cortés's strategy?
Essentially, the key to it was following the blueprint as closely as possible. Go right to the top, establish a coastal settlement — which Cortés also was very invested in — so you have a secure base to fall back to if what's happening in the interior is not playing out. A base on the coastline was key. Originally with the conquistadors, but more broadly the Portuguese, British, and French subsequently, they would build those key strongholds on the coast, which, because of their monopoly over nautical technology, meant they were effectively unable to be extricated by any local indigenous population. That's one thing that people tend to underestimate with the success of European expansion. At the end of the day, they had the fortresses, and again, the long institutional knowledge of what a perfect fortress should look like: those jagged-edged star parapets and revetments and battlements which maximized firepower. Even Eurasian cultures, even the Qing Chinese, struggled to be able to deal with that. The peoples of the pre-Columbian Americas had no real chance of storming defenses on that kind of a scale. The Inca and Aztecs didn't even have siege warfare. They didn't have siege weapons, battering rams, catapults, nothing like that. Once the Europeans dug into fortified positions, it was very hard to move them out of that.
Pizarro's own environment was quite different from that of Cortés. The highlands of Mexico are quite elevated, but the Andes Mountains that the conquistadors were moving into… I've been at elevation in the Himalayas and I was struggling to breathe with the altitude sickness. How they were able to move to those heights and then be able to fight battles is really astonishing.
With the many dozens of pounds of armor and equipment.
