The Same War, Better Funded
11,000 years of organized violence, and what centralized power did with it.
Every war comes with a reason attached. National security. Liberation. Self-defense. Manifest Destiny. The gods require it. The reasons change with the century, but the shape of the thing does not. Somebody decides the killing is necessary, somebody else does the killing and the dying, and the two are rarely the same person. The ones who die are almost always the same category of person: young, strong, capable, and fertile. The ones who decide are rarely in the blast radius.
That arrangement is old. It predates writing, predates agriculture, predates the wheel. It is older than the idea of civilization itself, which is inconvenient, because civilization is supposed to be the thing that fixed it. Philosophy was supposed to fix it. Religion was supposed to fix it. Science was supposed to fix it. None of them did. What changed was not the impulse but the equipment, and the equipment is what makes the modern version so much worse than anything that came before.
This is an accounting exercise. Start at the bones in the sand and end at a spreadsheet in Stockholm, and see if the argument that we have not evolved, that we remain selfish, short-sighted, and organized mainly around finding new justifications for old appetites, survives contact with the record. It does. But the record also points to where the responsibility actually sits, and it is not evenly distributed. It concentrates wherever the power to organize violence has been allowed to concentrate. That is the part worth taking seriously.
It would be easy to read a list of massacres and death tolls as a case for despair, a shrug dressed up in footnotes. That is not the intent here. A pattern this consistent across this many centuries is not proof that the species is irredeemable. It is proof that a specific arrangement, small numbers of people acquiring the administrative capacity to marshal large numbers of people into organized killing, keeps producing the same result no matter which century, religion, or ideology surrounds it. That is a narrower and more useful finding than the claim that humanity is simply broken, and it points toward a narrower and more useful response.
The bones came first
For decades, the working assumption among archaeologists was that organized violence was a byproduct of settled life. Hunter-gatherers, the theory went, were too mobile, too sparse, too busy surviving to bother with war. Farming created surplus, surplus created property, property created reasons to kill your neighbor. Peace was the default state of nature. War was an invention of civilization.
The bones say otherwise. At Jebel Sahaba, in what is now northern Sudan, archaeologists in the 1960s uncovered a cemetery of 61 individuals dated to as early as 13,400 years ago. Roughly half showed evidence of violent trauma, and a reanalysis published in 2021 found more than 100 previously undocumented healed and unhealed lesions, a quarter of the skeletons carrying wounds from more than one violent episode. The 2021 study concluded the site was not one massacre but a long pattern of raids and ambushes, probably driven by competition over resources during a period of severe climate disruption at the end of the last Ice Age. People were fighting over the Nile Valley’s dwindling floodplains millennia before anyone had invented the state, the tax collector, or the standing army.
At Nataruk, in Kenya, a second site tells a similar story from roughly 10,000 years ago. Twenty-seven skeletons, including a pregnant woman and six children, were found unburied, left where they fell. Ten of twelve near-complete skeletons showed clear evidence of violent death: blunt force trauma to the skull, projectile wounds from arrows, an obsidian blade still lodged in one man’s head. The Cambridge researchers who led the excavation concluded the group was killed in a deliberate attack, and one interpretation offered since is telling: population density and resource abundance, not scarcity alone, may have been what triggered the raid. Conflict did not require civilization. It required people, proximity, and something worth fighting over. All three conditions have been present for the entire run of the species.
Researchers who study these sites are careful to note that violence in the prehistoric record has not been uniform. Some hunter-gatherer bands, particularly those with very low population density, appear to have gone long stretches without organized conflict of any kind, which is itself informative. It suggests the human default is not constant war, and never was. What the record instead shows is a species with the latent capacity for organized violence, a capacity that stays dormant under conditions of scarcity so severe that cooperation is the only viable strategy, and that activates as soon as population, proximity, and stakes rise enough to make raiding pay off. That capacity did not go away when farming, cities, and writing arrived. It found better tools and a bigger stage.
This matters for the argument this piece is built on because it forecloses the easiest excuse. You cannot blame the state for inventing war. The state did not invent war. What the state did, once it arrived, was scale it. It took a capacity that had always existed at the level of small bands and raiding parties and gave it institutional permanence: standing armies, tax systems to fund them, bureaucracies to conscript and supply them, and ideological machinery to explain to the people doing the dying why it was necessary. War went from something a village might do a few times in a generation to something an empire could do continuously, for centuries, against enemies its soldiers would never see and populations its soldiers did not quarrel with.
The state discovers scale
Human sacrifice is a useful hinge point, because it shows the transition in progress: violence still wrapped in religious language, but already organized by a centralized political authority with an interest in the outcome. The Aztec Empire practiced sacrifice on a scale that remains genuinely disputed among historians, precisely because the numbers were also a political tool. Spanish chroniclers after the conquest claimed the 1487 reconsecration of the Great Pyramid at Tenochtitlan involved 80,400 victims over four days, a rate of roughly 14 people per minute that later scholars, including Ross Hassig, have called a physical impossibility and likely propaganda designed to justify the conquest itself. More sober estimates from historians working off the Florentine Codex and archaeological recovery at the Templo Mayor put annual sacrifice across the empire somewhere between 1,000 and 20,000 people, with some demographic estimates running as high as 15,000 to 250,000 per year depending on population assumptions that are themselves contested.
The exact figure will likely never be settled. What is not contested is the function. Sacrifice reinforced the authority of the Aztec state, financed by tribute, staffed by a professional priesthood, fed by captives taken in wars whose entire purpose was to generate sacrificial victims. It was not spontaneous violence. It was a bureaucratic pipeline: campaign, capture, transport, ritual, repeat, all run by an institution with a monopoly on the definition of religious necessity. Archaeologists excavating the Templo Mayor in Mexico City uncovered the physical infrastructure of that pipeline directly: a rack, or tzompantli, built to display rows of skulls on wooden posts, flanked by two towers constructed largely out of human remains mortared together, sized by the excavation team at roughly the footprint of a basketball court and several meters tall. That is not the architecture of spontaneous cruelty. It is the architecture of an institution that had scheduled, budgeted, and permanently staffed the production of victims as an ordinary function of state religion, the same way a modern state permanently staffs a military rather than raising one only when attacked.
The passage that inspired this piece calls war’s justifications an echo of ancient ritual sacrifice, and the Aztec case makes the comparison almost too literal. The victims were, as the argument goes, the young, the strong, and the captured, offered up to sustain a cosmology that the people making the offering had themselves designed. It is also worth noting who was not offered up. The priests performing the sacrifice, the emperor who received tribute from conquered provinces, and the nobility who directed military campaigns to generate captives all occupied positions structurally exempt from the fate they administered for others, the same asymmetry that shows up in every subsequent example this piece will cover, right up to the present day.
Move the clock forward, and the religious language falls away, but the machinery gets more efficient. The Mongol conquests of the 13th century are estimated by historians to have killed tens of millions across Asia and the Middle East, with some estimates running past 40 million when famine and disease from the destruction of irrigation systems are included. That was achieved not by superior numbers but by superior organization: a decimal command structure, standardized logistics, and a political culture that treated conquest as the ordinary business of state. The lesson repeats across the Mughal campaigns in India, the Taiping Civil War in China, and the long list catalogued in the historical record of wars by death toll: whenever a centralized authority acquired the administrative capacity to mobilize large populations and sustain them in the field, the death tolls climbed by orders of magnitude over anything a raiding party or a tribal skirmish could produce. Centralization did not create the impulse to kill. It multiplied the impulse’s reach.
Rome offers the same lesson from a different angle: a state that made conquest into a career path. Roman legions were a standing, professionalized bureaucracy for organized violence, funded by taxation and tribute extracted from the very provinces the legions had previously conquered, a closed loop that ran for centuries and left an estimated body count across its campaigns that is still debated by historians precisely because record keeping was itself a state monopoly, controlled by the people with the least interest in an honest tally. The pattern shows up again in the colonial era, where European states industrialized extraction alongside violence. The Congo Free State, run as the personal property of Belgium’s King Leopold II rather than as a colony of the Belgian state, is estimated by historians to have caused the deaths of several million Congolese through forced labor, mutilation, and famine between 1885 and 1908, a toll generated not by an invading army in the field but by an administrative rubber quota system enforced with organized terror. Centralized authority does not need a battlefield to convert its subjects into casualties. It only needs the administrative reach to make extraction profitable and the ideological cover to make it look like progress.
Industrializing the sacrifice
The 20th century removed any doubt about where this trend line was headed once the state married conquest to industrial capacity. World War I mobilized more than 65 million soldiers and killed nearly 15 million people, roughly 8.8 million of them in uniform, using an administrative apparatus, conscription, rationing, and propaganda ministries that would have been unrecognizable to a Mongol khan but served the same function: convert a civilian population into a renewable supply of combatants.
World War II went further. Estimates of total mortality range from 60 to 75 million people, around 3 percent of the entire global population alive in 1940. Between 50 and 56 million of those deaths were directly caused by the fighting; the rest came from the famine and disease that followed in war’s wake, the predictable second-order effects of a state deciding to convert its whole economy into a killing apparatus. China alone lost more than 15.5 million people. Germany’s military dead and missing are estimated at 5.3 million, a substantial share of them men conscripted from outside Germany’s own borders. None of these people chose the war. All of them paid for it.
Add up the century, and historians land in a range that varies by methodology but not by magnitude. Political scientist Andreas Wimmer’s synthesis puts 20th-century war deaths at roughly 110 million. Historian Eric Hobsbawm’s broader count of people “killed or allowed to die by human decision,” which includes deaths from state-induced famine and internment alongside battlefield deaths, reaches 187 million, about 10 percent of the entire global population as it stood in 1900. A separate accounting effort by researcher Milton Leitenberg arrives at a comparable range of 136.5 to 148.5 million for war and conflict deaths alone, before adding the deaths attributable to state policy more broadly. However the number is drawn, it describes the deadliest century in the species’ history, achieved by the most technologically and administratively advanced civilizations the species had yet produced. Philosophy, science, and organized religion were all more developed in 1945 than they had been in 1300. The killing was worse anyway.
The demographic pattern inside those numbers is the part the original passage puts its finger on directly. War does not draw randomly from a population. It draws from the young, the healthy, and the reproductively capable, the exact people a society depends on to rebuild itself afterward. Conscription age has hovered around 18 to 25 for most of the industrial era for a reason: that is the age bracket a state can train quickly, deploy cheaply, and replace easily. It is also, not incidentally, the age bracket with the least institutional standing to refuse the order. A conscript at 19 has no seat on the general staff, no vote on the war cabinet, and in most conscription systems, no legal mechanism to decline. The men who did not come home from the Somme or Stalingrad or the Pacific were disproportionately in their late teens and twenties, at the exact point in life when a person’s labor, fertility, and future contribution to their community are at their highest value. Whatever the state loses by sending them, it treats as an acceptable cost, because the decision makers are, structurally, rarely the ones on the casualty list. Generals and heads of state die in combat at a rate too small to register statistically against the millions of conscripts who do not survive to see whether the war’s stated objective was ever achieved.
National security as an all-purpose license
If the first half of the 20th century was defined by total war between industrial states, the second half was defined by something more useful to the argument here: wars justified almost entirely by an abstraction, fought against countries that posed no plausible threat to the nation doing the invading, because failing to fight them constituted a security risk. Vietnam, Iraq, and Afghanistan are the obvious American examples, but the pattern is not uniquely American. It is the default posture of any state large enough to project force and confident enough in its own justifications not to examine them closely.
The Iraq War is instructive because the justification, weapons of mass destruction that did not exist, was demonstrably false by the time the occupation had run its course, and the war proceeded anyway, restructured around new rationales, regional stability, democracy promotion, and counterterrorism, as the original one collapsed under scrutiny. Afghanistan ran for two decades against an insurgency that had no capacity to threaten the continental United States by the time the occupation ended, consumed 2.3 trillion dollars by the Costs of War project’s accounting, and concluded in August 2021 with the same movement the 2001 invasion had been launched to remove retaking control of the country within weeks of the last American aircraft departing. Over 20 years, trillions of dollars, and tens of thousands of lives produced by the government’s own stated objective of preventing that movement’s return to power: no durable result at all. Both wars were sold as security necessities. Both drew their soldiers overwhelmingly from the same demographic bracket that has supplied every war since Jebel Sahaba: young, capable, and disproportionately drawn from communities with the least capacity to say no, often enlisting for reasons that had more to do with economic opportunity at home than any considered judgment about the strategic merits of the campaign they were about to join.
None of this required the population to foot the bill, in blood or in taxes, to consent to the specific reasoning in any granular way. It required only the invocation of the category. National security functions, in practice, as a bureaucratic override, a phrase that closes debate rather than opening it. That is not an accident of language. It is the load-bearing function of the phrase. A raiding party has to explain its reasons to the people it is asking to fight and die, because the people it is asking are its neighbors, and the group is small enough that consequences and decisions stay attached to the same faces. A modern state does not have that constraint. It can fund the war through debt, staff it through a volunteer force drawn from economically limited options, and manage public opinion through a media and messaging apparatus scaled to match the war itself. The distance between the decision and the consequence, which in a small community would be measured in feet, is measured in a modern state in institutional layers, and every layer is a place where accountability can be diffused until it disappears.
The accounting on this is now well documented, and it is worth sitting with the actual figures rather than the abstraction. Researchers at Brown University’s Costs of War project have tracked the post 9/11 wars in Iraq, Afghanistan, Pakistan, Syria, Yemen, Somalia, and elsewhere since 2010, and their most recent comprehensive estimate puts the total budgetary cost to the United States at roughly 8 trillion dollars, a figure that includes direct war fighting appropriations, homeland security spending, interest on the debt used to finance the wars, and an estimated 2.2 to 2.5 trillion dollars in future veterans’ care that has not yet been paid. That is more than the entire annual economic output of Japan and Germany combined. The human toll runs alongside it: an estimated 897,000 to 940,000 people killed directly by the violence, with the researchers’ own indirect mortality estimate, built on the standard ratio of four indirect deaths for every direct one, pushing the total toward 4.5 to 4.7 million people once famine, disease, and the collapse of medical infrastructure are counted. 38 million people were displaced across those war zones. None of it was paid for out of current revenue. Nearly all of it was financed through borrowing, which means the generation asked to shoulder the bill was not the generation that voted to start the war, a second and less visible version of the same demographic transfer that shows up on the battlefield itself.
The economics of continuing
A war, once started, generates its own constituency for continuation, and that constituency has nothing to do with the original justification. Between 2020 and 2024, private defense contractors received an estimated 2.4 trillion dollars in Pentagon contracts, about 54 percent of the department’s entire discretionary spending over that period. Researchers who study the economic effects of military spending have found that a million dollars of Pentagon spending generates an average of five jobs, compared with roughly thirteen jobs for the same money spent on education, nine in healthcare, and seven to eight in infrastructure or clean energy. Military spending is, by the government’s own numbers, one of the least efficient uses of public money available for producing employment, and it remains among the most politically protected categories of spending in most large states regardless of which party holds power.
The reason is not mysterious. An industry that manufactures the equipment for organized violence has every incentive to ensure that the demand for organized violence never disappears, and it has the lobbying budgets, campaign contributions, and revolving door hiring practices to make sure its interests are represented at the table where war and peace get decided. When the SIPRI figures released in 2026 showed European defense spending climbing 14 percent in a single year, the fastest pace since 1953, they also showed defense sector stocks rallying across the same period: shares in one major South Korean arms maker rose 193 percent in 2025 alone, a German armor and munitions manufacturer climbed 154 percent, and a Japanese heavy industries firm gained more than 100 percent. None of that is a coincidence, and none of it requires a conspiracy to explain. It requires only an institution structured so that a subset of its members profit directly from the continuation of the thing everyone else is asked to sacrifice for. That is precisely the incentive structure a community organized around voluntary cooperation and direct accountability is built to resist, because in a small enough group, the people who profit from a decision and the people who pay for it are the same people, or close enough to see each other clearly. Scale removes that visibility, and removing that visibility is not an unfortunate side effect of scale. It is close to the entire function.
Financing the war on debt rather than current taxation compounds the same problem across time instead of across space. A state that had to raise the full cost of a war from its living taxpayers in the year the war was fought would face a much shorter list of wars worth fighting, because the bill would arrive immediately and land on the same people being asked to approve the war in the first place. A state that borrows against its own future removes that check entirely. The soldiers conscripted or recruited to fight the war are paying with their bodies. The taxpayers of a later decade, many of them not yet born when the war started, are left paying the interest, a bill that by some estimates will keep compounding on the post 9/11 wars alone for decades after the last American unit has come home. Neither group had much say in the decision. Both are paying for it anyway. That is not a side effect of how modern states finance war. It is close to the whole reason debt financing became the default method in the first place, because it lets the decision makers spend money nobody in the room has to feel losing.
The present tense
None of this is history in the sense of being over. In 2025, global military expenditure reached $2.887 trillion, according to the Stockholm International Peace Research Institute, the highest figure ever recorded and the 11th consecutive year of growth. That figure represents 2.5 percent of global GDP, the largest share since 2009, and works out to $352 for every person on the planet, whether or not that person has any say in how the money is spent. The five biggest spenders, the United States, China, Russia, Germany, and India, accounted for 58 percent of the total. European NATO members increased their spending by 14 percent in a single year, the fastest rate since 1953. Germany’s military budget crossed 2 percent of GDP for the first time since before reunification. Ukraine, fighting for its survival, now spends an estimated 40 percent of its entire GDP on the war, the highest military burden of any state on record.
The human toll behind those budget lines is not abstract. The Armed Conflict Location and Event Data Project recorded at least 240,000 conflict deaths worldwide in 2025 alone. Sudan’s civil war, which began in April 2023 as a power struggle between two factions of the same military apparatus, has killed more than 20,000 people by ACLED’s count in a single recent 12-month period and displaced roughly 11 million more inside the country, with millions further pushed into Chad, Egypt, and South Sudan. Gaza’s death toll from the war that followed the October 2023 Hamas attack has been estimated at 47,000 or higher by local health authorities, with some epidemiological reviews suggesting the true figure, once famine, disease, and the missing are accounted for, runs considerably higher; over 70 percent of the strip’s structures have been damaged or destroyed. Ukraine’s war has produced Russian casualty estimates running into the hundreds of thousands. The United Nations recorded a record 117.3 million people forcibly displaced by conflict and violence worldwide as of the middle of this decade, a figure that keeps climbing.
Beyond the three conflicts drawing the most coverage, the pattern repeats in places that get a fraction of the attention. Eastern Democratic Republic of Congo remains contested by rebel proxies backed by neighboring Rwanda, a conflict that has displaced millions and killed tens of thousands with barely a fraction of the press coverage devoted to Ukraine or Gaza, a discrepancy that is itself a form of institutional priority setting: some populations’ deaths are treated as newsworthy and some are not, and the difference tracks geopolitical relevance rather than the scale of the suffering. Myanmar’s civil war has fragmented into more than 1,200 distinct armed factions by one count, a conflict so decentralized it barely resembles a traditional war and yet has still produced a death toll and displacement crisis on par with far more centralized fights. Pakistan’s simultaneous insurgencies and border clashes, a brief but intense war between India and Pakistan, and a Thailand and Cambodia border clash have all added their own tallies to a global total of at least 240,000 conflict deaths in 2025 alone, according to ACLED, a level of violence the organization describes as remaining high after years of steady increase rather than showing any sign of retreat.
None of these wars are being fought by hunter-gatherer bands disputing a floodplain. They are being fought by states, or by factions attempting to become states, using weapons, logistics, and financing that would have been unimaginable to anyone at Jebel Sahaba or Tenochtitlan, in pursuit of justifications, territorial security, sectarian survival, and resource control that would have been entirely legible to them. The rationale has not changed. The delivery system has, and the delivery system is precisely what a centralized, well-financed, administratively sophisticated state makes possible that a hunter-gatherer band, a Bronze Age city state, or a voluntary civic association never could.
The paradox at home
The original passage asks why, if war is conducted so blithely abroad, there is murder taking place daily and hourly within our own borders, as if society had turned its violence inward because it could no longer point it outward with a clean conscience. It is worth checking that claim against the data rather than simply repeating it, because the honest answer is more complicated, and the complication matters.
The United States homicide rate peaked at 10.7 per 100,000 people in 1991, fell to a low of 4.7 by 2014, spiked again during the pandemic years to 7.75 in 2021, and has since fallen further than the pre-pandemic baseline. Preliminary data for 2025 puts the national rate near 4.29 per 100,000, and some analysts tracking large cities describe it as the lowest homicide rate ever recorded in data going back to 1900, with 2023, 2024, and 2025 registering the three largest annual declines on record. Whatever is driving American homicide right now, it is trending in the opposite direction from the picture the original passage assumes. Interpersonal violence at the individual level has been getting rarer, not more common, for most of the last three decades, pandemic aside.
That complication does not undercut the larger argument. It sharpens it. What has not declined, what in fact keeps climbing, is violence organized at scale by institutions with the administrative capacity to sustain it: state militaries, paramilitary factions, and criminal organizations operating with a level of coordination that mimics state structure. Gang violence in Mexico, Ecuador, and Haiti has grown severe enough that ACLED now ranks all three among the ten most violence-affected places on earth, driven not by individual rage but by organized groups fighting for control of territory and trade routes using tactics, discipline, and logistics that look far more like small armies than like street crime. Haiti alone recorded more than 4,500 deaths from political violence in a single recent year, and Ecuador’s toll from organized criminal violence rose by more than 1,000 deaths year over year, in both cases the product of organizations that function as parallel governments with their own taxation, territorial control, and armed enforcement, not as an aggregate of individual disputes.
It is worth putting a finer point on the historical baseline, too. A demographic study of American homicide from the early 1980s calculated the lifetime risk of dying by homicide at roughly 1 in 133 for the population as a whole, a risk that has fallen substantially in the decades since, even accounting for the pandemic-era spike. That decline happened without any comparable decline in the capacity of organized institutions, whether state militaries or criminal syndicates, to kill at scale when it serves their purposes. The pattern holds across the whole record examined here: violence between individuals, arising from personal dispute, jealousy, or opportunity, is the kind of violence that social trust, economic stability, and community cohesion can and do reduce over time. Violence organized by an institution with a taxing authority, a supply chain, and an ideology to justify itself is a different animal, and it has shown no comparable sensitivity to any of the civilizing forces that have worked on individual behavior. The problem was never human aggression as a raw material. It is what gets built with it once an institution decides aggression is a resource worth harvesting at scale.
Progress that never touched the root
Every civilization that has practiced large scale violence has also produced serious people who argued against it. Legions of philosophers, in the phrase this piece takes as its starting point, and it is not an exaggeration. Stoic and Confucian ethics both counseled restraint and proportion. Christian, Buddhist, and Islamic traditions all developed extensive theological frameworks condemning unjust killing, and just war theory has been refined by serious thinkers for over a thousand years specifically to constrain when and how states may fight. The scientific revolution that followed gave humanity the tools to understand disease, agriculture, and material scarcity well enough that resource competition, the oldest driver of violence identified at Jebel Sahaba and Nataruk, should in principle be a solvable problem for most of the developed world.
None of it changed the trajectory of state violence in any way that shows up in the casualty figures. The 20th century, sitting on top of more accumulated ethical philosophy, more developed religious institutions, and more scientific capability than any prior era, was also the deadliest in the species’ history. That is not a coincidence to be waved away with an appeal to population growth, because the death tolls as a share of global population, Hobsbawm’s 10 percent figure among them, remain staggering even adjusted for scale. The century that produced the League of Nations, the United Nations Charter, and the Geneva Conventions is the same century that produced the atomic bomb, the firebombing of entire cities, and industrialized genocide carried out with the same bureaucratic tools used to run a railway or a census. The moral vocabulary got more sophisticated. The restraint it produced on actual state behavior, measured in bodies, did not keep pace.
Moral philosophy restrained individuals. It did essentially nothing to restrain institutions, because institutions do not have a conscience to appeal to. They have incentive structures, and the incentive structure of a centralized state with a monopoly on organized force has never included a meaningful check on how that force gets used, so long as the people making the decision are not the ones paying the price. A soldier can be taught just war theory in a classroom and still be ordered into a war that fails every test the theory sets out, because the theory constrains the individual conscience of the person holding the rifle, not the institutional decision of the government that issued the order. That mismatch, moral development at the level of the person, structural indifference at the level of the institution, is the actual mechanism behind the pattern this piece has been tracing, and it is a mechanism that centuries of religious and philosophical effort were never designed to fix, because they were aimed at the wrong target.
This is the piece’s answer to the question the original passage poses almost as an aside: is any of this intentional, is it somehow built into the species at the level of its biology. It is not necessary to reach for DNA to explain the pattern. The simpler explanation is structural. Wherever the capacity to organize mass violence has concentrated in a small number of decision makers insulated from its consequences, that capacity has been used repeatedly, across every civilization for which records exist, regardless of that civilization’s religious convictions, philosophical sophistication, or technological era. The species has not failed to evolve past violence. It has consistently built institutions that reward the concentration of violence and insulate the people who authorize it, and no amount of individual moral development has ever been sufficient to counteract an institutional incentive that strong.
Where the accountability actually sits
This is the point at which the diagnosis has to turn into something more useful than despair, because an essay that ends at “we are barbarians and always will be” is just a more articulate version of the shrug the original passage is pushing back against. The pattern across 13,000 years is not that human beings are incapable of organizing cooperative life without violence. Most human interaction, across most of history, has been exactly that: trade, kinship, mutual aid, and negotiated resource sharing that leaves no bones in the sand at all. The pattern is narrower and more specific. It is that whenever a coercive authority accumulates enough centralized power to wage war at scale, tax a population to pay for it, and conscript that population’s young people to fight it, that authority tends to use the power, and it tends to use it more often and more destructively as its administrative reach grows.
That is an argument about institutional design, not about human nature in the abstract, and it points toward an answer. A government whose legitimate function is limited to its irreducible core, courts, physical security, contract enforcement, and the narrow set of protections a community genuinely cannot provide for itself, does not have the administrative reach to fund a $2.887 trillion global arms buildup or conscript a generation into a war whose justification will be debated by historians for the next century. The capacity for total war is not a natural feature of political life. It is a capacity that specific institutions built, over centuries, by accumulating taxing power, conscription power, and the ideological machinery to make both look normal. Remove the accumulation, and the ceiling on how much damage any single decision can do comes down with it.
This also explains why moral appeals to citizens, be more peaceful, question your leaders, write to your representative, have never been a reliable check on state violence, while structural limits on what the state is permitted to fund and conscript have, historically, worked far better when they were actually enforced. The problem was never that ordinary people failed to feel strongly enough about war. Public opposition to Vietnam, to the Iraq invasion, and to any number of 20th-century conflicts was often broad, sustained, and well organized, and it rarely stopped the war it was aimed at, because the decision to continue did not sit with the public. It sat with an institution insulated by design from exactly the kind of accountability a protest march is built to apply. That is the structural reality the phrase national security is built to protect, and it is why the fix has to be structural too. You do not outargue an institution’s incentive to keep an $8 trillion war running. You remove its capacity to fund one in the first place.
This is where accountability belongs, and it does not belong diffused downward onto ordinary people carrying an inherited species guilt for the Somme or Tenochtitlan, nor upward onto some abstract failure of humanity that nobody in particular is responsible for fixing. It belongs with whoever holds the power to decide each specific case: the government official who chooses to invade, the legislature that appropriates the funding, the institution that insulates itself from the consequences it imposes on other people’s sons and daughters. Communities organized around voluntary cooperation, mutual aid networks, credit unions, worker-owned cooperatives, and civic associations built to solve a specific local problem have never in the historical record produced anything resembling the death tolls generated by centralized states, not because the people in them are more virtuous but because they lack the administrative machinery to convert local disputes into total war. That is not a coincidence. It is the entire argument for keeping that machinery as small as it can possibly be.
The obvious objection is that some wars are genuinely defensive, that Ukraine did not choose Russia’s invasion any more than Poland chose Germany’s in 1939, and that a framework built entirely around distrusting centralized military capacity risks leaving a free population defenseless against an aggressor with no such scruples. The objection is fair, and it does not actually conflict with the argument here. Physical security against external aggression is precisely the kind of function that sits inside the irreducible core, the thing a community cannot provide for itself through voluntary cooperation alone, because deterrence requires a credible, coordinated capacity to respond that a patchwork of local militias cannot reliably supply. The distinction this piece is drawing is not between a state that can defend its borders and one that cannot. It is between a defensive capacity sized to the actual threat and accountable to the population bearing its cost, and a capacity that has grown, in dollar terms and administrative reach, far past what any plausible defensive requirement explains, then gets deployed in wars of choice against countries that posed no threat at all, using the same justification, and no serious institutional check, that a genuine defensive war would require. Ukraine spending 40 percent of its GDP to repel an actual invasion and the United States spending 2.3 trillion dollars over twenty years in Afghanistan are not the same category of event, even though both get filed under national security, and treating them as if they were is exactly the rhetorical trick that lets the second kind of war borrow the moral legitimacy of the first.
What the alternative looks like in practice
None of this requires abolishing government, and it does not require pretending that courts, physical security, and contract enforcement are optional. Those are the functions a community genuinely cannot replicate through voluntary cooperation alone, because they require a monopoly on coercive authority to work at all. The argument is narrower: every function beyond that core is a place where centralized power has historically found a way to convert itself into capacity for larger and more distant wars, and every one of those functions is also a place where voluntary, mutual, and locally accountable institutions have a track record of solving the same problem without acquiring that capacity.
Mondragon, the worker-owned cooperative federation in the Basque region of Spain, has run for close to 70 years as an alternative to both centralized state control and unaccountable corporate ownership, without ever needing an army to enforce its internal decisions. Cooperation Jackson in Mississippi and the broader community land trust movement offer another version of the same principle: local ownership, local accountability, and a structural inability to convert the organization’s resources into anything resembling a war machine because the organization was never built with that capacity in mind. Credit unions, mutual aid networks formed around disaster response, and the volunteer fire and rescue structure this country still substantially depends on all share the same architecture. They solve a real problem; they are accountable to the people they serve because those people can see and reach the decision makers, and none of them has ever needed to conscript a 19-year-old to do it.
The point is not that these institutions are morally superior to the people who staff a national defense ministry. The point is that they are structurally incapable of the scale of harm a centralized state can produce, because nobody built them with a taxing power large enough to fund a trillion-dollar arms buildup or a legal authority broad enough to draft a population against its will. Keep the state’s coercive function limited to the core it cannot outsource, and hand everything else to institutions built on voluntary cooperation and direct accountability, and the ceiling on how much damage a single bad decision can inflict falls by orders of magnitude. That is not a sentimental argument about human goodness. It is an argument about where you put the largest lever and how far you let anyone reach for it.
The same principle applies closer to home, in the institutions this country already relies on without fully crediting them for it. Volunteer fire departments and the Coast Guard Auxiliary run substantial portions of the nation’s emergency response and maritime safety infrastructure through unpaid or lightly compensated civic commitment rather than centralized conscription, and they do it with a chain of accountability that runs directly to the communities they serve rather than through the layers of a federal bureaucracy. Scouting America and similar civic institutions have spent more than a century building young people’s competence and character through voluntary association rather than state mandate. None of these organizations has ever needed the power to tax a nation or draft its sons to function, and none of them has ever produced anything resembling a battlefield, because the entire architecture of voluntary civic life is built around solving a shared problem, not projecting force. That is the model worth defending, expanding, and citing whenever someone insists that only a centralized authority can be trusted to handle a serious problem. The historical record says otherwise, and it says so specifically at the point where centralized authority stops handling problems and starts creating casualties.
The honest ending
The original passage ends on a genuinely difficult question: whether any of this is intentional, whether the pattern is somehow baked into the species at a level nobody can reach or reform. It is a fair question to ask after looking at 13,000 years of evidence that points in the same direction every time. But the honest answer is not resignation. The evidence does not show a species incapable of anything but violence. It shows a species that has spent most of its existence cooperating, trading, raising children, building irrigation systems, burial customs, and cooperative federations, interrupted periodically by the specific and identifiable failure of letting too much coercive power concentrate in too few hands. That failure has a shape. It has been documented at Jebel Sahaba, at Tenochtitlan, at the Somme, and it is being documented right now in Sudan, Ukraine, and Gaza, in real time, by organizations counting the dead.
A specific pattern is not a verdict on human nature. It is a design flaw in how power gets organized, and design flaws can be corrected by people who decide to stop building the same structure and expect a different result. The species does not need to wait for evolution to solve this. It needs institutions small enough, local enough, and accountable enough that the people who decide to fight and the people who do the dying are never more than a few steps removed from each other again. That was the arrangement for most of human history before the state learned to put distance between the two. Closing that distance back down is not a utopian project. It is a maintenance job, and it is overdue.
It is also a job that does not require waiting on anyone else to start it. Every one of the counter examples raised in this piece, the worker cooperatives, the credit unions, the volunteer fire companies, the Auxiliary crews who show up for search and rescue with their own boats and their own time, the community land trusts holding property in trust for people who could never compete for it in an open market, exists today, right now, run by people who decided that a given problem did not require waiting for a centralized authority to solve it on their behalf. None of them required a declaration of war to get built. None of them will ever generate a casualty list. They are proof, sitting in plain view in nearly every American community, that the voluntary and the mutual are not a theory about what people might do under better conditions. They are a functioning alternative to centralized coercion that has been quietly doing the work all along, usually without credit, while the institutions built for conquest continue to absorb the headlines, the budgets, and the dead.
The question the passage that opened this piece leaves hanging, whether any of this is intentional, whether it is written into the species at some level beyond correction, deserves a direct answer rather than a shrug. It is not written in. It is built by specific people, making specific decisions, inside specific institutions, generation after generation, and every one of those decisions had an alternative available to it at the time it was made. The Aztec state could have stopped expanding its wars of capture. The Kaiser’s government could have accepted a negotiated peace before the trench lines hardened. The architects of the war on terror could have weighed twenty years of consequences against the actual size of the threat before the first trillion was spent. None of them did, and the reason none of them did traces back to the same structural fact each time: the people weighing the decision were never the people who would pay for it. Fix that one asymmetry, keep the power to decide and the obligation to bear the cost bound to the same hands, and the rest of the pattern this piece has documented, the death tolls, the budgets, the generations of young men and women spent on wars their grandchildren will still be paying interest on, stops looking like fate and starts looking like what it actually is: a choice, repeated, that a species capable of building cathedrals, cooperatives, and rescue squads is also fully capable of choosing not to repeat.