Thomas Hobbes painted a picture that has haunted Western thought for four centuries. Imagine humans without government, without law, without civilization. What would they do? They would kill each other. Constantly. Without mercy. In this "state of nature," life is "solitary, poor, nasty, brutish, and short." Every person's hand is against every other. Trust is impossible. Violence is incessant. Morality is meaningless. The only solution is a powerful sovereign—a Leviathan—who imposes order through threat of violence greater than any individual could muster. Better to live under a tyrant, Hobbes argued, than to suffer the anarchy of natural human conflict.
This image has become foundational to Western political thought. It justifies monarchy, justifies authoritarian rule, justifies the concentration of power. It says: humans are naturally violent, naturally selfish, naturally prone to war. Civilization tames these impulses through coercive force. Without that force, we revert to savagery.
Steven Pinker resurrected and modernized this argument in The Better Angels of Our Nature. He presented statistical evidence that violence has declined over history. Fewer people die in war as a percentage of population. Fewer people are murdered. Torture and slavery, once common, have become rare. His conclusion: civilization, law, and reason have made humans less violent. We are progressively taming our savage nature through the institutions of the modern state.
Both arguments share a common assumption: humans have an innate propensity for violence that civilization either channels or tames. The question is whether civilization makes us more or less violent overall. But beneath both arguments lies a deeper assumption that goes unexamined: that humans are naturally violent, that "the state of nature" is one of endemic warfare, that without external constraint we would kill each other constantly.
This assumption is wrong. And the consequences of believing it are profound.
The problem with Hobbes' account is that it is not based on observation. It is based on political philosophy. Hobbes lived in seventeenth-century England during a period of civil war. He extrapolated from his historical moment to a claim about human nature. He never encountered a "state of nature." No human has. All human societies have laws, norms, hierarchies, and conflict-resolution mechanisms.
When anthropologists actually studied contemporary hunter-gatherer societies—the closest we have to understanding how humans organized before agriculture and states—the picture was far more complex than Hobbes imagined. Some hunter-gatherer societies were relatively peaceful. Others were violent. Some had virtually no warfare. Others engaged in chronic raiding. The variation was enormous. There was no universal "state of nature." There were diverse states of nature.1
Some societies had cultural norms strongly against violence. The Mbuti foragers of Central Africa had mechanisms for defusing conflict before it escalated to killing. The San peoples of the Kalahari engaged in minimal warfare compared to pastoral societies in the same region. Other societies—some pastoral groups, some agricultural societies—had higher rates of male-male violence and raiding than others.
The crucial point: the variation in violence across human societies shows that violence is not a fixed human nature that emerges without constraint. It is conditional on social organization, resource abundance, and cultural norms. Some societies with minimal government still had low violence. Some societies with strong government were internally peaceful but engaged in conquest warfare. The simple dichotomy—violent state of nature vs. peaceful civilization—does not match the evidence.
Pinker correctly observes that violence as a percentage of population has declined in most developed nations over recent centuries. Fewer people are murdered per capita in modern democracies than in many tribal societies. Slavery and torture, once institutionalized, are now rare in wealthy nations. Wars between developed democracies have become almost unthinkable. These observations are accurate.
But Pinker conflates two distinct claims. The first claim—violence has declined in developed nations—is factually defensible. The second claim—this decline proves that reason and civilization have tamed human savage nature—is less defensible. The decline in violence in developed nations correlates with state development, wealth, literacy, and interconnection, but it does not prove causation. And it obscures something crucial: state development also created new capacities for violence that did not exist before.1
The twentieth century saw the largest-scale organized violence in human history. Hundreds of millions of people were killed by state apparatus—the Holocaust, the Soviet purges, the Cambodian killing fields, the atomic bombs, industrial war. As a percentage of global population, the twentieth century was probably the most violent in human history. Pinker acknowledges this but argues that state violence was declining by the end of the century and that even at their peak, these atrocities represented a smaller percentage of population than tribal warfare in some societies.
This argument misses something essential. The question is not what percentage of population dies in violence. The question is what enabled that violence. Tribal societies did not have the organizational capacity to commit genocide. They could not maintain slavery systems at industrial scale. They could not develop weapons capable of destroying cities. States created those capacities. And yes, states also created legal systems and norms that constrain violence in many contexts. But the net effect is ambiguous. States simultaneously reduced certain types of violence (murder) while enabling others (organized atrocity).
The deepest problem with the "savage state of nature" narrative is that it becomes a self-fulfilling prophecy. If you believe humans are naturally violent without external constraint, you organize society around coercive control. You build police forces. You create surveillance systems. You establish harsh punishments. You assume people will cheat and lie without incentive to cooperate. You make systems of mutual suspicion.
And what happens? The systems of coercion and suspicion activate threat response in people. They breed resentment and resistance. They make cooperation harder, not easier. The very act of organizing society around the assumption that humans are naturally violent and selfish creates conditions that produce violence and selfishness. The prophecy fulfills itself.
Contrast this with societies organized around different assumptions about human nature. Some societies assume that humans are capable of cooperation and trustworthiness. They build systems with minimal surveillance, minimal coercive punishment, and high levels of local autonomy. These societies often have lower violence than societies organized around coercive control. Why? Because the assumption shapes the reality. If you assume people will cooperate, you create conditions that make cooperation easier. If you assume people will cheat and betray, you create conditions that make betrayal more likely.
The narrative matters. The belief about human nature shapes the society that emerges from that belief.1
Neither Hobbes nor Pinker asks a crucial question: why did Western thinkers develop such a dark view of human nature in the first place? Why is the image of "the savage state of nature" so compelling to western philosophy?
The answer lies in history. Hobbes lived during the English Civil War. European societies were experiencing the devastation of warfare at unprecedented scale—the Thirty Years' War had killed millions. The Wars of Religion had made Europe a charnel house. Violence was endemic. Hobbes looked at his historical moment and extrapolated it to human nature itself.
But this extrapolation was distorted. The violence Hobbes witnessed was not tribal raiding. It was state-organized warfare. It was the product of organized religion and dynastic conflict. It was warfare enabled by new technologies and organizational structures. Hobbes took a historical accident—the European wars of religion—and treated it as proof of eternal human nature.
The same distortion appears in Pinker. He observes that modern developed nations have lower violence than historical societies and contemporary tribal societies. But he does not ask: which societies are we comparing? Modern developed democracies are wealthy, have strong legal institutions, and have experienced centuries of state consolidation. They are not typical of human history. Pinker selectively compares modern democracies to historical warrior societies and tribal cultures known for high violence. He ignores the many historical and contemporary societies that had lower violence than modern warfare but higher wealth and state organization than tribal societies.
The comparison is distorted. It proves not that civilization reduces violence but that wealth and stable institutions reduce certain types of violence while enabling others.
Nature vs. Nurture in Violence: Is human violence a fixed nature that civilization constrains, or is it conditional on social organization? The tension reveals that the dichotomy is false. Humans have the capacity for both violence and cooperation. Which emerges depends on context, norms, and social structure—not on an immutable nature.
Tribal vs. State Violence: Tribal societies had higher rates of male-male killing but lower capacity for organized atrocity. States have lower murder rates in many contexts but created the capacity for genocide. Which is "more violent"? The tension reveals that "violence" is not a single dimension. We cannot reduce complex historical questions to a single metric like "percentage of population dying in violence."
Belief as Self-Fulfilling Prophecy: If societies organized around the belief that humans are naturally violent create conditions that produce violence, are we observing human nature or creating it through our beliefs? The tension reveals that the distinction between discovering human nature and constructing it through belief may be illusory.
Behaviorally, the minimal group paradigm and dehumanization research show how humans can be manipulated into in-group favoritism and out-group hostility. But these mechanisms operate best when they are justified—when people believe that the in-group bias and violence toward the out-group serve a legitimate purpose.
Historically, the "savage state of nature" narrative provides that justification. If you believe humans are naturally violent and selfish, then coercive control seems necessary and justified. If you believe that "civilization tames savagery," then state power seems like a gift rather than oppression. Political philosophy becomes a technology for making populations accept control and violence directed at them.
The tension between domains reveals something critical: political theories about human nature are not neutral descriptions. They are behavioral blueprints. A society that believes Hobbes' account of human nature will organize itself differently—with more surveillance, more punishment, more coercive control—than a society that believes humans are capable of cooperation and trustworthiness. The theory shapes the institutions. The institutions shape behavior. The behavior then seems to confirm the theory.
This is not a discovery of human nature. It is the creation of human nature through belief and institutional design.
The "savage state of nature" narrative provided justification for colonialism. Indigenous peoples were living in the "state of nature." They were violent and uncivilized. Therefore, conquering them and imposing European order was not oppression—it was civilizing. The narrative justified atrocity by describing it as the taming of savagery.
The same narrative justified slavery. Africans were supposedly closer to the "state of nature," more prone to violence and irrationality, less capable of civilization. Therefore, enslaving them was not wrong—it was bringing them toward civilization. The narrative made atrocity morally acceptable.
In the twentieth century, the narrative shifted but persisted. European civilization was supposedly rational and progressive. Other peoples were supposedly tribal, violent, prone to ethnic conflict. The Nazis used narratives about Jewish tribal nature and racial savagery to justify genocide. Rwandan perpetrators used narratives about Tutsi nature and Hutu tribal identity to justify killing. The narrative changes but the function remains: justify the violence by describing the target as savage, as naturally violent, as deserving of coercion or elimination.
The cross-domain insight is that historical atrocity relies on narratives about human nature. If you can convince people that an out-group is naturally violent, naturally selfish, naturally prone to aggression, then violence toward that group becomes defensible. You are not committing atrocity. You are protecting civilization from savagery. You are taming nature.
The question this raises: how do we escape the narrative? If we reject the "savage state of nature" account, what replaces it? If we acknowledge that humans are capable of both cooperation and violence, both selfishness and altruism, then our theories of human nature become conditional. Violence emerges under specific conditions. Cooperation emerges under other conditions. This makes for a more complex and harder to mobilize narrative. It cannot justify atrocity as the inevitable taming of human nature. But it also cannot guarantee peace.