Why Certainty Arrives Before Facts — and What It Costs Us
Why we prefer monsters to explanations — and what changes when we stop
The first response
When people hear about a horrific act of violence — especially inside a family — the first response is not a thought. It is something in the body. A tightening behind the ribs. A flash of image no one asked for. The mind places loved ones inside the story before it can be stopped. For a moment the distance between one’s own life and the headline disappears entirely.
What follows, if it is not immediately converted into something else, is pain. The recognition that lives have been destroyed. That a child, a partner, a family has been broken beyond repair. That somewhere people are standing in the kind of silence that has no words. For parents especially, the pain is almost unbearable, because the mind has already projected their own family into the scene.
This pain is the honest response. It is what happens when one human being registers the destruction of other human lives. It does not need to be converted into anything else to be valid. It is not weak. It is not passive. It is the part of a person that understands what was lost.
But pain is difficult to sustain. It offers no resolution, no action, no sense of control. It leaves a person exposed to the reality that terrible things happen to people who did not deserve them, and that the distance between their life and the headline may be smaller than they need it to be.
Anger arrives to manage this problem — not as a deeper truth but as a regulatory shift. Research across psychology and neuroscience consistently shows that anger functions as a transformation of aversive states rather than an independent reaction to a separate cause. Leonard Berkowitz demonstrated that pain and other aversive experiences generate anger as a downstream response — the anger is produced by the distress itself as a coping mechanism (Berkowitz, 1990). Carver and Harmon-Jones showed that anger is specifically an approach-related affect: it shifts the nervous system from withdrawal states associated with pain, helplessness, and fear into approach motivation, producing a sense of directed agency where passivity existed moments before (Carver & Harmon-Jones, 2009).
Richard Lazarus’s appraisal theory explains why this conversion feels like moral clarity rather than psychological defense. Grief and pain involve an appraisal of irrevocable loss — something terrible has happened and cannot be undone. Anger restructures the appraisal: it introduces a responsible agent, a target, and the sense that something can be opposed. The situation shifts from overwhelming to actionable. This is not a conscious strategy. It is an automatic reappraisal that occurs faster than deliberation (Lazarus, 1991).
The conversion happens so quickly and so reliably that most people experience anger as the first reaction. But it is not the first reaction. It is the first defense — a transformation of helpless pain into active force. Underneath every surge of public outrage after a tragedy, there is a layer of grief that was too uncomfortable to sustain.
Helen Block Lewis documented this pattern clinically: vulnerable emotional states — grief, shame, helplessness, exposure — rapidly convert into anger because anger restores the sense of agency and control that vulnerability destroys (Lewis, 1971). At the collective level, Thomas Scheff showed the same dynamic operating in groups: shared vulnerability and unacknowledged pain transform into shared aggression directed at a target, producing cohesion through rage rather than through the harder, slower work of mourning together (Scheff, 1994).
None of this means the anger is false or that the outrage is performed. It means the anger is doing something — converting a state the mind cannot easily tolerate into one it can act on. The problem begins when this conversion becomes permanent. When the defense becomes the explanation. When the pain that could have opened into understanding is sealed over by a certainty that closes all questions.
Why the mind closes the story before the story is known
After a severe act of violence, especially one that ruptures ordinary expectations — a parent harming a child, a partner destroying a family — public narratives stabilize with remarkable speed. Long before motives are investigated, before mental states are assessed, before anyone with direct knowledge has spoken, the person at the center of the event becomes a type. Monster. Psychopath. Pure evil. A moral verdict arrives while the factual record is still almost entirely empty.
This is not a failure of character in the people making these judgments. It is a predictable property of human cognition under threat.
Daniel Kahneman’s research on intuitive judgment shows that the mind generates conclusions before it engages in reasoning (Thinking, Fast and Slow, Part II). When faced with an emotionally charged event, the brain does not wait for evidence, weigh it carefully, and then form a view. It produces a rapid story — one that feels obvious and complete — and then searches selectively for information that confirms it. The conclusion comes first. The justification follows.
In the case of extreme violence, the fastest available story is one about character. Lee Ross called this the fundamental attribution error: the deeply automatic tendency to explain other people’s behavior through who they are rather than what they were experiencing (Ross, 1977; Heider, 1958, ch. 5). When an act is terrible enough, the explanation collapses into essence. The person did this because they are this.
The reason this explanation feels so right is not that it is accurate. It is that it is stabilizing.
An unexplained act of destruction leaves the mind in a state of genuine threat. If the cause is unknown, it is impossible to predict whether it will happen again, whether it could happen nearby, whether the world still operates by the rules previously assumed. That uncertainty is not merely uncomfortable. Research on explanatory cognition shows that the nervous system processes unresolved causal gaps as active danger (Lombrozo, 2012). The brain treats not-knowing as a form of ongoing exposure to harm.
A monster solves this problem immediately. Monsters have a nature. Their behavior follows from what they are. The cause of the event is located inside a single individual, which means it can be contained by containing that individual. The world becomes predictable again — damaged, but navigable.
The mind is not choosing the most accurate explanation. It is choosing the one that most quickly restores the feeling of living in a world that makes sense.
What certainty accomplishes — and where it stops
The rush toward moral certainty performs real psychological and social work.
René Girard described a pattern visible across cultures and centuries: when communities experience internal disruption — violence, disorder, the cracking of shared norms — they restore cohesion by concentrating the disorder onto a single figure (Violence and the Sacred, ch. 1–3). Blame becomes a focusing mechanism. Collective condemnation reorganizes the group around a shared boundary: we are not like that. Modern moral psychology finds the same dynamic operating at the level of individual cognition: moral judgments function as coordination signals, aligning group perception and reinforcing shared values (Greene, Moral Tribes, ch. 6–7).
In the aftermath of a violent tragedy, collective certainty accomplishes three things simultaneously:
It ends unbearable uncertainty. The event has a cause. The cause is a person. The case is morally closed.
It restores the boundary between acceptable and unacceptable. By declaring the act monstrous, the community reaffirms that its norms still hold — that this was a violation of the order, not evidence that the order does not exist.
It synchronizes the group. Shared condemnation creates a temporary experience of solidarity. People who may disagree about everything else find common ground. This is experienced as moral community, and it is genuinely stabilizing after events that fracture the sense of shared reality.
But notice what has happened in this process. The original response — the pain, the grief for destroyed lives — has been entirely overwritten. First by anger, then by certainty. The community has moved from witnessing devastation to performing judgment without ever passing through understanding. The grief that Scheff describes as requiring collective acknowledgment has instead been bypassed through collective rage (Scheff, 1994).
And this is where the cost begins. When certainty becomes the permanent explanation — when “this happened because the person is evil” is treated as a complete account — then inquiry stops. The event is emotionally resolved but causally unexamined. The pain that might have motivated genuine learning has been consumed by a process that produces only closure.
The tragedy is honored in words and abandoned in practice. Because the one thing that could actually reduce future tragedies — understanding how this one became possible — has been foreclosed by the very reaction meant to show that the tragedy mattered.
Why understanding feels like a threat
After violent events — especially those involving families or children — a particular statement surfaces reliably in public conversation: mental health is not an excuse.
The intensity of this reaction reveals something important about what is actually being defended. It is not merely a legal argument about culpability. It is a defense of something psychologically essential: the belief that the world is organized by choice and that safety is earned through being a certain kind of person.
Melvin Lerner spent decades studying what he called the belief in a just world — the deeply held conviction that people’s outcomes correspond to their decisions and their character (The Belief in a Just World, 1980). This belief is not a naive error. It is a survival mechanism. It allows people to plan, to invest effort, to raise children, to walk through daily life without being overwhelmed by the randomness of suffering. If terrible outcomes happen to people who make terrible choices, then safety depends on behavior. The world is controllable.
This explains a pattern that might otherwise seem contradictory: the same public that rejects “mental health” as an explanation will often accept “insanity” or “psychopath” without resistance. The reason is that these words are doing entirely different psychological work. When someone calls a perpetrator insane or psychopathic in ordinary conversation, they are not offering a clinical description. They are saying: this person is fundamentally other. Their mind operates in a way mine never could. They belong to a different category of being. This language sounds medical, but it functions identically to “monster” — it preserves the wall between their life and mine.
When someone says “this person was experiencing a mental health crisis,” the implication is the opposite. It means: this arose from recognizable human suffering — isolation, depression, shame, accumulated stress — the kind of suffering that exists on a spectrum, the kind that in different degrees touches ordinary lives. This does not preserve the wall. It dissolves it. And that dissolution is experienced not as information but as threat.
The implicit logic:
If the causes are alien — an incomprehensible nature, a category of person fundamentally unlike anyone the listener knows — then safety is assured. The danger is contained within a type.
If the causes are ordinary — accumulated stress, untreated suffering, social failure, collapsed support — then the line between safe and unsafe lives is not a wall. It is a matter of degree. And degree can shift.
This is why causal explanations provoke such strong resistance. They are experienced as an attack on the structure that keeps the listener’s own world feeling secure. Understanding the perpetrator feels like weakening the protection around everyone else.
Lewis’s research illuminates the mechanism: recognizing shared vulnerability is itself a shame state — it exposes the fragility of one’s own safety — and shame states convert rapidly into anger and rejection (Lewis, 1971). The hostile reaction to causal explanations is not merely disagreement. It is the mind defending itself against the distress of recognizing that catastrophe does not require a special category of person.
But the protection is illusory. Declaring someone a monster does not make anyone safer. It only makes safety feel less urgent — because the danger has been assigned to a type rather than a set of conditions. And conditions, unlike types, can be changed.
By insisting that violence arises only from a special kind of person, the very information that would allow the next tragedy to be seen forming is removed. The comfort of a closed case is chosen over the discomfort of an open investigation. And the next family pays the price.
What was actually happening
When researchers examine people who have committed acts of severe violence — not from a distance, not through headlines, but through years of clinical and empirical study — a consistent picture emerges. It is not the picture the public narrative prepares anyone for.
What looks from the outside like a single monstrous decision usually turns out, on examination, to be the final moment in a long unraveling.
Robert Sapolsky’s work across neuroscience, endocrinology, and behavioral biology shows that any human action — especially an extreme one — is the product of interacting processes that span every timescale of a life (Behave, Parts II–IV). The neurochemistry of the seconds before the act was shaped by the hormonal environment of the preceding hours. That hormonal state was shaped by weeks or months of stress. The stress response itself was calibrated by years of developmental experience. And the developmental environment was structured by culture, economics, and social conditions that no individual chose.
These layers do not merely add up. They reshape each other. Chronic stress literally alters brain architecture — the prefrontal cortex, which supports impulse regulation and long-term reasoning, is measurably degraded by sustained exposure to threat and helplessness. The person who commits a violent act under extreme stress is not making a calm decision with an intact mind. They are operating with a nervous system that has been progressively damaged by the conditions leading to that moment.
In Determined (2023), Sapolsky extends this further: once every contributing factor is genuinely accounted for — genetics, epigenetics, prenatal environment, childhood experience, hormonal state, neurological condition, immediate context — the idea that behavior emerges from an autonomous inner essence, a free-floating will that simply chose evil, becomes scientifically unsupported. This does not mean society should stop holding people accountable. It means that explanations built around moral essence — “they did it because of who they are” — describe less than the full story, and often miss the parts of the story that are actionable.
James Gilligan spent decades working with men who had committed the most severe acts of violence — homicides, assaults that nearly killed, acts that seem to place a person beyond the boundary of comprehensible humanity. What he found, consistently, was not what the public imagines. These were not men overflowing with confidence and power. They were men who had been destroyed by shame — shame so total that it had annihilated their sense of being a person at all (Violence: Reflections on a National Epidemic, ch. 2–4).
The acts occurred not when hostility gradually increased past a threshold. They occurred when the last remnant of psychological structure collapsed — when shame became unsurvivable and no alternative to violence was cognitively available. The person did not choose violence over peace. They reached a state where the internal architecture that makes choice possible had already failed. Lewis’s clinical observations of shame-rage conversion operate here at their most extreme: the unbearable vulnerability of total shame produces explosive anger as the only remaining regulatory mechanism (Lewis, 1971).
Margaret Archer’s sociological framework offers a complementary view: human action always emerges from an interaction between internal concerns and external structures (Structure, Agency and the Internal Conversation, ch. 1–2). People are not isolated decision-makers operating in a vacuum. They are embedded in relationships, institutions, economies, and cultural systems that shape what options they perceive, what pressures they experience, and what responses feel available. An act that looks from the outside like a single free choice often turns out, on closer examination, to be the final point in a long process of narrowing — where options closed one by one until what remained was catastrophic.
None of this produces the clean satisfaction of declaring someone evil. But it does something that moral certainty cannot: it tells us where the process could have been interrupted.
An act is not a permanent property of a person. It is a long unraveling reaching its worst moment.
Where intervention actually lives
If violence is caused by evil, there is exactly one lever available: identify and punish evil people. If violence is caused by a process — a convergence of failures across time — then multiple levers appear, each one representing a moment where the outcome could have been different.
Imagine someone slowly losing hold over a period of months. Not a monster assembling a plan — a person running out of exits. Each week a little more isolated. A little more ashamed of what their life has become. A little more convinced that no one can help and no option remains. The distress is visible to people around them, but no one knows what to do with it, or the systems that should respond are inaccessible, or the person has learned — through years of experience — that asking for help leads to punishment rather than support.
Now ask: where could someone have reached them?
Consider a simplified chain:
- An academic or professional collapse that triggers identity crisis
- Withdrawal from relationships out of shame
- Growing internal pressure with no outlet and no language for what is happening
- Fear of disclosure — because vulnerability has been punished before
- Escalating desperation experienced as entrapment
- An acute confrontation that overwhelms remaining regulatory capacity
- Access to means of harm at the worst possible moment
Each point in this chain is a barrier that failed. This is structurally similar to what safety science calls the Swiss cheese model: catastrophic outcomes occur not because of a single cause but because multiple layers of protection each had a gap, and the gaps aligned. No single failure would have been sufficient. The convergence was necessary.
Evil cannot be repaired. But barriers can. Crisis intervention can be funded and staffed to actually reach people in shame spirals. Mental health systems can be built to be accessible before collapse rather than only after. People in communities and institutions can be trained to recognize withdrawal and escalation — not as signs of a person becoming dangerous but as signs of a person losing the capacity to cope. The conditions that narrow a person’s options over months and years — economic precarity, social isolation, cultures of silence around suffering — can be addressed. Access to lethal means during acute crisis periods can be restricted, a measure consistently shown to reduce fatalities even when it does not eliminate the underlying distress.
Each of these interventions is modest on its own. None of them requires minimizing harm. All of them require seeing violence as a process rather than a character revelation.
Understanding is not softer than certainty. It is more actionable.
Accountability in a causal world
There is a fear — a legitimate one — that causal explanation weakens consequences. Courts will give lighter sentences. Institutions will say “it’s complex” and avoid responsibility. People who harm others will learn to use the language of causation to escape accountability.
This fear has historical basis. It deserves a direct answer.
Causal understanding does not remove accountability. It changes what accountability is for.
In a framework built on moral essence, accountability means: you did something terrible because you are terrible, and punishment balances the moral ledger. The logic is retributive — harm answers harm.
In a causal framework, accountability serves a different function: you caused severe harm. Safety may require confinement. But safety does not come from caging one person alone — it comes from understanding and repairing the system of failures that allowed this convergence to occur. The individual is held responsible, and the conditions are examined so that the next person in a similar trajectory encounters a functioning barrier instead of an open path to catastrophe.
Consider the analogy of a bridge collapse. Engineers investigate every contributing factor: material fatigue, design errors, inspection failures, environmental stress. This investigation does not exonerate the builder. The builder may face legal consequences. But no one argues that investigating causes is disrespectful to the people who died on the bridge. And no one suggests that the appropriate response is to condemn the builder and leave the building codes unchanged.
When the causes of human violence are refused examination — when the event is treated as fully explained by the character of the individual — the building codes are effectively left unchanged, and the wait for the next collapse begins.
Causal morality is more demanding than condemnation. Condemnation requires a single act of judgment. Causal morality requires sustained systemic attention — to mental health infrastructure, to community support, to institutional design, to the social conditions that produce entrapment. It asks more of society, not less.
Causal language can also be weaponized. People who harm others can and sometimes do use psychological explanations to evade responsibility, to manipulate sympathy, to reframe their choices as things that happened to them. This is real. The answer is not to abandon explanation but to insist that explanation always serves prevention — never evasion. When someone uses complexity as camouflage for avoiding consequence, they are not practicing causal understanding. They are exploiting it. The distinction matters and must be maintained.
The hardest cases
Everything written so far might seem to apply most naturally to cases that already evoke some compassion — a person overwhelmed by crisis, a moment of collapse after long suffering. But what about calculated, repeated, predatory violence? What about acts that involve planning, apparent enjoyment, sustained cruelty over time? These are the cases where the word “monster” feels not like a cognitive shortcut but like a precise description.
If the framework fails for the hardest cases, it is not a serious framework.
When researchers examine people who have committed the most disturbing forms of violence — serial, predatory, apparently purposeful — they do not find an absence of causal history. They find causal histories of extraordinary severity. Sapolsky’s neurobiological work documents how extreme early environments — neglect, abuse, chaotic attachment — can produce lasting alterations in brain systems governing empathy, impulse regulation, and threat perception. Gilligan’s clinical work with the most violent offenders consistently found not people who were born without conscience but people whose developmental environments systematically destroyed the psychological structures that make conscience operational.
This does not make the horror smaller. But it does something critically important: it reveals patterns. And patterns, unlike essences, can be detected before they reach their worst expression.
If someone is a monster — if that is a complete and sufficient explanation — then there is nothing to learn and nothing to detect in advance. Monsters appear without warning because their nature is hidden until it erupts. But if someone’s trajectory toward severe violence followed a sequence of identifiable conditions — early abuse, institutional failure, absent intervention, escalating behavior that was visible but unaddressed — then there were moments where a different response could have changed the outcome.
The hardest cases do not break the causal framework. They demonstrate why it is most needed exactly where certainty is most intense.
How digital environments close the window
Most people encounter tragedies of violence not in person but through screens. The headline appears in a feed between other content. The reaction is immediate. The platforms carrying this reaction are not neutral — they are architectures that shape it in specific directions.
The online disinhibition effect reduces empathic processing when the target of discussion is a symbolic figure rather than a physically present person (Suler, 2004). It becomes easier to speak about someone in language that would never be used face to face — not out of malice but because the environmental cues that normally activate social caution are absent.
Emotional and morally charged language spreads faster and further than neutral or complex description (Brady et al., 2017, PNAS). Content that produces a strong moral reaction — outrage, disgust, righteous condemnation — is more likely to be shared, commented on, and promoted by algorithmic systems designed to maximize engagement.
Outrage specifically generates more interaction than other emotional registers (Crockett, 2017, Nature Human Behaviour), which means platforms inadvertently select for the interpretation of events that produces the fastest emotional resolution. Complexity is not suppressed by policy. It is suppressed by architecture. Nuanced responses receive less engagement, reach fewer people, and disappear from view more quickly.
What appears in a feed after a violent tragedy is not a representative sample of human thought. It is the output of a system that has filtered for speed, intensity, and emotional closure. The responses closest to genuine pain — the grief, the bewildered sorrow, the honest admission of not understanding — are structurally disadvantaged. The most confident, most categorical declarations rise to the top.
Every comment, share, and reaction added to this environment is a small act of closing the case or keeping it open. Each quick declaration of a person’s essential nature makes it slightly harder for everyone else in the conversation to stay with the pain long enough to ask a useful question.
The medium is not neutral. It shapes what kind of response feels normal and what kind feels excessive.
A practice for the space between pain and response
Between the first response — the pain of witnessing destroyed lives — and whatever is said or done publicly, there is a space. What happens in that space determines whether what follows contributes to closure or to learning.
Here is something small and concrete:
Next time a case produces a strong reaction, before commenting or sharing or forming a settled view, write down three questions whose answers are genuinely unknown. Not rhetorical questions. Real ones. For example:
- What was happening in this person’s life in the months before?
- Who around them noticed distress, and what options did those people have?
- Where could a realistic intervention have changed the trajectory?
These questions will feel unsatisfying. They do not produce the resolution that a moral verdict provides. That is the practice working. It is training the capacity to stay with pain and uncertainty longer than is comfortable — because the comfortable response and the useful response come from different places.
In conversation, when someone closes the case with a declaration — “that person is a monster, end of story” — argument in that moment usually produces defensiveness rather than reflection. A different approach:
“It is devastating. What do you think was going wrong in the time before anyone noticed?”
This does not minimize what happened. It redirects attention from the endpoint to the preceding sequence — which is where every opportunity for prevention exists.
The practice will feel incomplete. It will not produce certainty. But the goal is not to feel resolved. The goal is to build the habit of looking at the trajectory rather than only the terminus — because the trajectory is where the next tragedy can still be prevented.
The cost of certainty
Every year, tragedies occur that are structurally similar to previous ones. The details differ. The locations change. The names are new. But the underlying patterns — isolation, shame, collapsed support, institutional blindness, access to means — repeat with a regularity that should disturb more than it does.
It does not disturb because each time, the monster model erases the pattern. Each event is experienced as a unique eruption of individual evil rather than a recurrence of recognizable conditions. Shock repeats because each previous time, the explanation chosen was the one that prevents learning.
This is what certainty arriving before facts actually costs. Not that certainty feels wrong — it feels deeply right, because it converts unbearable pain into manageable anger and manageable anger into settled judgment. But it purchases emotional resolution at the price of structural ignorance. And structural ignorance is paid for by people who have not yet been harmed but will be.
The next time a headline stops someone — and it will — the pull toward the immediate story will be there. The person is a monster. The case is closed. The world makes sense again. That pull is real. It is managing something real.
But underneath it, if anyone stays long enough, there is the original pain. The pain of knowing that lives have been shattered. That someone’s child is gone. That an entire family’s future has been destroyed. And underneath the pain there is a quieter question: what was breaking down before anyone noticed? Where were the months of visible suffering that no one knew how to address? What systems failed? What barriers had holes?
That question will not feel satisfying. It will not resolve quickly. It will not produce the solidarity of shared condemnation.
But it is the question that, answered honestly and often enough, actually changes what happens next.
What is refused understanding cannot be prevented. Studying causes is not sympathy with harm. It is care for people who have not yet been harmed — and still can be protected.
A society that only condemns repeats its tragedies. A society that stays with the pain long enough to learn from them reduces their frequency.
The deepest compassion is not a reaction after suffering. It is fewer people who suffer in the first place.
References
Archer, M. (2003). Structure, Agency and the Internal Conversation. Cambridge University Press, ch. 1–2.
Berkowitz, L. (1990). On the formation and regulation of anger and aggression: A cognitive-neoassociationistic analysis. American Psychologist, 45(4), 494–503.
Brady, W. J., et al. (2017). Emotion shapes the diffusion of moralized content in social networks. PNAS, 114(28), 7313–7318.
Carver, C. S., & Harmon-Jones, E. (2009). Anger is an approach-related affect: Evidence and implications. Psychological Bulletin, 135(2), 183–204.
Crockett, M. J. (2017). Moral outrage in the digital age. Nature Human Behaviour, 1, 769–771.
Gilligan, J. (1996). Violence: Reflections on a National Epidemic. Vintage Books, ch. 2–4.
Girard, R. (1972). Violence and the Sacred. Johns Hopkins University Press, ch. 1–3.
Greene, J. (2013). Moral Tribes. Penguin Press, ch. 6–7.
Heider, F. (1958). The Psychology of Interpersonal Relations. Wiley, ch. 5.
Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux, Part II.
Lazarus, R. S. (1991). Emotion and Adaptation. Oxford University Press.
Lerner, M. (1980). The Belief in a Just World. Springer.
Lewis, H. B. (1971). Shame and Guilt in Neurosis. International Universities Press.
Lombrozo, T. (2012). Explanation and abductive inference. Cognition.
Ross, L. (1977). The intuitive psychologist and his shortcomings. In Advances in Experimental Social Psychology.
Sapolsky, R. (2017). Behave: The Biology of Humans at Our Best and Worst. Penguin Press, Parts II–IV.
Sapolsky, R. (2023). Determined: A Science of Life Without Free Will. Penguin Press.
Scheff, T. J. (1994). Bloody Revenge: Emotions, Nationalism, and War. Westview Press.
Suler, J. (2004). The online disinhibition effect. CyberPsychology & Behavior.
