The most dangerous moment in a political society is not when people become angry.
It is when they become convinced that there is nothing meaningful they can do with their anger.
A person can vote.
Speak.
Write.
Donate.
March.
Organize.
Petition.
Sue.
Call representatives.
Argue with friends.
Document what is happening.
Appeal to journalists, courts, regulators, universities, professional institutions, religious organizations, political parties and every other mechanism a society presents as an alternative to violence.
None of those mechanisms guarantees that person will get what they want. Nor should they. Democracy cannot mean that sufficiently intense conviction automatically becomes policy.
But something much more serious happens when people stop believing those mechanisms are capable of addressing grave wrongdoing at all.
That is where moral injury becomes politically dangerous.
Not because moral injury automatically produces violence. It does not.
Not because killing becomes justified when institutions fail. It does not.
And not because the United States is presently populated by millions of people preparing to commit political violence. The newest national evidence actually points in the opposite direction.
The danger is subtler.
Repeated moral injury can produce anger, betrayal, disillusionment and loss of trust. Research on political violence separately finds that grievance becomes more dangerous when people believe conventional political participation is ineffective. Threat-assessment research finds unresolved grievances repeatedly appearing along pathways toward targeted violence.
Put those findings together carefully and a disturbing picture emerges:
A society that repeatedly creates profound moral grievance while convincing people that peaceful redress is futile is degrading one of the psychological barriers that keeps political conflict nonviolent.
The overwhelming majority of people may never cross that barrier.
A functioning country cannot afford to assume nobody will.
Moral Injury Is More Than Political Anger
The phrase moral injury is often used loosely now, but it describes something much deeper than being offended, losing an election or encountering a policy one strongly dislikes.
The U.S. Department of Veterans Affairs describes moral injury as the psychological, behavioral, social and sometimes spiritual aftermath of experiences that violate deeply held moral beliefs. Those experiences can involve doing something oneself, failing to prevent something, witnessing it, or experiencing betrayal by people or institutions in positions of trust. Common reactions include guilt, shame, disgust, anger and betrayal.
A major 2025 review in the Annual Review of Clinical Psychology describes moral injury as involving impairing moral emotions, beliefs and behaviors, along with damage to beliefs about humanity, meaning and purpose. Particularly relevant here is what the authors describe as trust-violation-related moral injury, which can involve grievance, outrage, cynicism, disillusionment, diminished faith in authorities and institutions, and persistent justice-seeking. The review is worth reading directly.
That distinction matters.
Some moral injury turns inward.
A person thinks:
I did something unforgivable.
But betrayal-based moral injury can turn outward:
Something unforgivable was done, the people responsible know it, the institutions that should stop it will not, and nobody is going to make this right.
That is a very different psychological problem.
Institutions Can Inflict Moral Injury Too
Moral injury entered the modern clinical conversation largely through military experience, but the concept is no longer confined to soldiers.
The VA recognizes morally injurious experiences among health-care workers, including circumstances involving impossible decisions, witnessing suffering, institutional failure and perceived betrayal by leaders. Research has since examined moral injury in many noncombat populations.
That expansion makes sense.
Human beings build institutions partly so that enormous moral questions do not have to be resolved personally.
We create courts because individual vengeance is dangerous.
We create elections because political succession through violence is worse.
We create legislatures because millions of people cannot personally negotiate every disagreement.
We create inspectors general, watchdogs, journalists, regulators, professional bodies, ethics systems and protest rights because societies need mechanisms for saying:
Something is wrong, and there is somewhere legitimate to take that wrong.
Those systems do something more important than process paperwork.
They absorb grievance.
They convert anger into procedure.
They give outrage somewhere to go.
When they function, imperfectly but credibly, they tell the injured person:
You do not need to personally restore the moral order. There is a process for this.
That is an extraordinary civilizational achievement.
And we should be far more frightened than we seem to be when people stop believing it.
Peaceful Redress Is Not Just a Democratic Ideal. It Is a Safety Mechanism.
There is research remarkably close to this exact question.
Karin Dyrstad and Solveig Hillesund examined survey data from Guatemala, Nepal and Northern Ireland to study the interaction between grievance and people’s belief that they could meaningfully influence politics.
Their findings, published in the Journal of Peace Research, were not that grievance inevitably creates violence.
They found that support for political violence increased with grievance and decreased with perceived political efficacy. In several analyses, grievance was most strongly associated with support for violence among respondents with low confidence that ordinary political channels cared about their views. The full study is available here.
The authors put the underlying problem plainly: aggrieved people may become more willing to view violence as legitimate when they also conclude that peaceful methods of influence are ineffective.
That finding should receive far more attention outside academic conflict studies.
Because it means that the repeated instruction to “use the system” carries an obligation on the other side.
The system has to remain worth using.
That does not mean every complaint must prevail.
It does not mean every protest movement is correct.
It does not mean losing a court case constitutes oppression.
It means that people must continue to have credible evidence that facts can matter, evidence can be heard, wrongdoing can be investigated, rules can constrain powerful actors, participation can alter outcomes and peaceful pressure can occasionally produce meaningful change.
Otherwise, “use the system” slowly stops sounding like a solution.
It starts sounding like a taunt.
The Most Dangerous Word May Be “Futile”
A more recent American study makes this point difficult to ignore.
Researchers followed participants in a large nationally representative U.S. cohort and examined whether major life events were associated with changes in support for political violence between 2022 and 2023.
Most events were not.
But one of the few statistically significant findings was striking:
Respondents who reported giving up on politics showed an increase in support for political violence.
The effect was modest. It certainly does not establish that political withdrawal causes violence. But the researchers themselves noted the importance of political disengagement in understanding changes in attitudes toward violence. The study was published in Injury Epidemiology.
Read that again carefully.
Not becoming more politically obsessed.
Giving up.
That should bother us.
We tend to imagine dangerous political psychology as a person becoming increasingly engaged: angrier speeches, more demonstrations, more arguments, more activism.
Sometimes the more important warning is resignation.
The person who still writes letters believes letters might matter.
The person who still protests believes being seen might matter.
The person who still files complaints believes someone may eventually read them.
The person who still votes believes political power can be changed.
The person who has abandoned all of those things may have simply become apathetic.
Usually, that is exactly what happens.
But for a small minority, another possibility exists.
The question can change from:
How do I persuade the system?
to:
Why am I continuing to restrain myself for a system I no longer believe will restrain anyone else?
That is an extraordinarily dangerous question for a society to allow to become psychologically plausible.
Moral Injury Is Not the Same Thing as Radicalization
This distinction has to remain explicit.
There is no good evidence that moral injury automatically causes radicalization, much less political violence.
A systematic review examining the relationship between moral injury and radicalization found significant conceptual overlap between the two bodies of research.
People experiencing moral injury and people moving toward radicalization can experience some similar conditions: victimization, perceived betrayal, anger, shame, disrupted identity, loss of significance and difficulty making sense of events.
But the researchers also found an important limitation: the studies they reviewed did not directly demonstrate that moral injury causes people to radicalize. The authors concluded instead that moral injury may provide a useful framework for understanding part of the radicalization process and potentially preventing it. The systematic review was published in Studies in Conflict & Terrorism.
That is exactly the level of certainty this subject deserves.
Moral injury is not a violence switch.
It can, however, help create some of the psychological terrain in which a violent worldview becomes easier to construct.
And betrayal appears especially important.
A 2024 study of Army veterans found that exposure to potentially morally injurious experiences involving betrayal was associated with overall aggression, anger, hostility and verbal aggression. Betrayal itself did not independently explain aggression once other variables were considered, which is an important limitation. Arousal symptoms were more predictive. But the association reinforces the larger picture: betrayal-based moral injury is not simply sadness with a political vocabulary. The study was published in Clinical Psychology & Psychotherapy.
The Line Is Crossed Before the Violence Happens
The FBI’s behavioral threat-assessment literature approaches the problem from another direction entirely.
Its job is not to study moral injury.
It studies people who actually move toward targeted violence.
And grievance appears again.
In a recent FBI Behavioral Analysis Unit study examining 59 active shooters, all 59 had evidence of a grievance, 90% had evidence of violent ideation, and 73% had evidence indicating an intention to cause harm. The FBI is emphatic that these pathway indicators are not a checklist for predicting who will become violent, and many internal steps may never be visible beforehand. The Bureau’s Pathway to Intended Violence research summarizes the findings.
The logical direction matters enormously.
Millions of people have grievances.
Almost none become active shooters.
So grievance does not predict mass violence in any simple way.
But when researchers begin at the other end — with people who actually committed severe targeted violence — grievance appears again and again.
The FBI’s longer threat-assessment guide, Making Prevention a Reality, describes the familiar pathway as beginning when a person develops a deeply held grievance or humiliation involving a real or perceived injustice. Possible motivations can include revenge, the offender’s conception of justice, recognition, or a perceived need to solve an unbearable problem.
And buried in that same guidance is an observation that deserves far more public attention.
The FBI identifies continued pursuit of legal and socially sanctioned conflict resolution — complaints, communications, litigation and other attempts to obtain redress — as a potential mitigating factor in threat assessment.
Why?
Because continued investment in legitimate remedies can indicate that the person does not regard violence as the only remaining option.
Think about what that means.
A complaint system is not merely bureaucracy.
A functioning court is not merely procedure.
Political participation is not merely civic theater.
Nonviolent redress can be part of the architecture that keeps grievance from becoming something worse.
We Have the Causality Backwards
After political violence occurs, institutions almost always rediscover morality.
Suddenly violence is unacceptable.
Suddenly democratic norms matter.
Suddenly rhetoric has consequences.
Suddenly every grievance must be handled through proper channels.
All of that is true.
But it is a remarkably cheap morality if it begins only after someone dies.
The harder moral work happens beforehand.
It happens when an institution receives the hundredth complaint about something it would rather ignore.
It happens when officials know that a legal process exists on paper but not meaningfully in practice.
It happens when evidence contradicts a politically convenient story.
It happens when accountability would be embarrassing.
It happens when peaceful protesters become inconvenient.
It happens when rules are technically available but visibly applied differently depending on who needs protection from them.
It happens when institutions repeatedly tell people that the correct response to profound moral outrage is patience while providing no evidence that patience will accomplish anything.
A peaceful society cannot endlessly demand restraint while treating the mechanisms that make restraint rational as optional.
That arrangement contains an obvious contradiction.
“Do not take matters into your own hands” only works as a social contract when people continue believing that someone else will take the matter seriously.
This Does Not Transfer Responsibility for Violence
This point needs to be stated without hesitation.
The person who deliberately commits an act of political violence is responsible for choosing that violence.
A failed institution does not pull the trigger.
Moral injury does not erase agency.
Grievance does not create moral immunity.
Political futility does not transform murder into legitimate democratic participation.
Understanding why a probability changes is not the same thing as morally absolving the person who eventually occupies the terrible end of that probability distribution.
But individual responsibility and institutional responsibility are not mutually exclusive concepts.
A drunk driver can be responsible for killing someone while engineers can separately investigate whether a dangerous road design increased the probability of crashes.
A company can be responsible for violating safety rules while regulators can separately be responsible for ignoring warning signs.
And a violent offender can be entirely responsible for an attack while society still asks whether preventable conditions helped move that individual from grievance toward violent ideation.
Pretending otherwise is not moral clarity.
It is refusing to study causation because causation makes us uncomfortable.
The Latest U.S. Evidence Is Reassuring. It Is Not Permission to Be Stupid.
There is another fact that belongs in this article precisely because it complicates the story.
The latest large national survey does not show Americans collectively racing toward political violence.
Researchers with the UC Davis Centers for Violence Prevention surveyed 7,752 members of the Ipsos KnowledgePanel between May 15 and June 15, 2026. Across 53 measures involving expectations, justification and willingness to engage in political violence, 51 moved downward compared with 2025, although many changes were small. Three-quarters of participants said violence in general to advance a political objective they supported was never justified.
The researchers concluded that support for political violence broadly decreased from mid-2025 to mid-2026. The findings are currently reported as a preprint and should be understood accordingly. UC Davis published the study and methodology in September 2026.
Good.
That should be reassuring.
It should also keep us from writing stupid, apocalyptic narratives about an American population supposedly waiting for permission to begin killing one another.
Most people reject political violence.
Most morally injured people do not become violent.
Most aggrieved people do not become violent.
Most people who lose faith in politicians do not become violent.
That is not the end of the analysis.
It is the beginning of the correct one.
Because political violence is a tail-risk problem.
A country of hundreds of millions of people does not require 20%, 10% or even 1% of its population to begin committing political violence before the consequences become horrific.
A tiny fraction crossing that threshold can produce assassinations, bombings, shootings, retaliatory attacks and escalating cycles of fear.
More important still, there is a difference between the number of people willing to commit violence and the number beginning to view someone else’s violence as understandable, morally ambiguous, forgivable or inevitable.
The latter population can become a social environment in which the former operates.
That is the threshold worth watching.
Not:
How many people want to kill someone?
But:
How many people are slowly losing the conviction that political violence is categorically beyond the pale?
Those are not the same question.
The Real Breaking Point Is Moral
There is a tendency to imagine that political violence begins when someone becomes sufficiently furious.
I suspect the more important transition is moral.
Violence changes categories.
At first it is unthinkable.
Then horrifying.
Then wrong, but understandable.
Then perhaps still wrong, but increasingly difficult to condemn.
Then exceptional circumstances begin appearing.
Then someone decides the exception has arrived.
The physical act comes last.
The dangerous transformation happened earlier.
This is why moral injury matters.
A human being can tolerate losing.
People lose elections constantly.
They lose lawsuits.
They lose arguments.
They watch legislation fail.
They spend years supporting causes that never prevail.
Democratic life requires enormous tolerance for disappointment.
But moral injury is not simply disappointment.
It arises when the person comes to believe that something fundamentally wrong is occurring and that the people or institutions expected to uphold the moral order have instead participated in the violation, protected it, excused it or abandoned their obligations.
Then the failure of redress carries a different meaning.
The person is no longer merely losing.
They believe the system is asking them to participate in the wrong by peacefully accepting it.
That perception may be mistaken.
It may be exaggerated.
It may be politically manipulated.
Or it may be substantially grounded in reality.
From a prevention standpoint, we would be fools to believe the distinction eliminates the danger.
People act on the world they believe they inhabit.
Institutions Do Not Get Unlimited Withdrawals From the Public’s Restraint
There is a strange arrogance in assuming that peaceful political behavior simply exists.
As though citizens wake up every morning constitutionally incapable of anything else.
As though centuries of institutional development had nothing to do with converting personal vengeance into law, factional warfare into elections and political retaliation into adjudication.
Peaceful conflict is constructed.
It is maintained.
It requires legitimacy.
It requires enough trust.
It requires people to believe that restraint is not the same thing as surrender.
Every functioning democratic institution therefore has two jobs.
The obvious job is whatever appears in its charter.
The less obvious one is preserving the public’s belief that disputes can still be resolved without bloodshed.
Courts do it.
Elections do it.
Investigations do it.
Independent journalism does it.
Public records do it.
Legitimate protest does it.
Accountability does it.
Even the ability to stand in public and say, loudly and offensively, “This is wrong” without disappearing into a cell does it.
These mechanisms do not eliminate conflict.
They civilize it.
Destroy enough of their credibility and the underlying human conflict does not disappear.
It becomes homeless.
The Warning Is Not That Violence Is Inevitable
Nothing in the evidence supports that conclusion.
Political violence is not inevitable.
The latest American survey data are evidence against fatalism.
Nor does the existing research establish a neat causal chain in which moral injury produces radicalization and radicalization produces violence.
Human beings are far more complicated than that.
The warning is narrower.
And more serious.
Grievance matters. Betrayal matters. Moral injury matters. Political efficacy matters. And maintaining credible nonviolent avenues of redress appears to matter.
Research from entirely different disciplines keeps finding pieces of the same structure.
Clinical psychology describes what happens when fundamental moral expectations and trust are shattered.
Radicalization research finds overlapping themes of betrayal, humiliation, meaning and identity.
Political-science research finds grievance becoming more dangerous where peaceful political efficacy is weak.
Longitudinal American research finds increased support for political violence among people reporting that they had given up on politics.
Threat-assessment research repeatedly encounters grievance along pathways toward targeted violence and treats continued investment in lawful remedies as potentially protective.
None of those findings alone proves the whole argument.
Together, they make complacency indefensible.
Do Not Wait for the Bloodshed to Discover the Warning Signs
There is a terrible pattern in public life.
Before violence, the aggrieved person is unreasonable.
Their anger is excessive.
Their complaints are repetitive.
Their protests are annoying.
Their language is uncomfortable.
Their demands are unrealistic.
Their insistence on accountability is divisive.
Their inability to “move on” is evidence of instability.
Then somebody crosses the line.
Suddenly everyone wants to know what happened.
Why was this person so angry?
Why did they lose faith?
When did their thinking become violent?
Were there warning signs?
Could anything have interrupted the pathway?
Those questions should be asked.
They should simply be asked much earlier.
A person speaking is still speaking.
A person marching is still marching.
A person suing is still suing.
A person arguing is still arguing.
A person furiously demanding that the system work is, at some level, still demonstrating faith that it might.
We should be much more worried about the day they stop.
Not because silence means violence.
Usually it means exhaustion.
But because one of the strongest protections against political violence is precisely the belief that another route remains open.
A society interested in preventing violence should therefore stop treating every peaceful expression of moral injury as a nuisance to be managed until it goes away.
Sometimes it does go away.
Sometimes the person accepts the loss.
Sometimes circumstances improve.
Sometimes people heal.
And sometimes unresolved injury calcifies into grievance, grievance into moral certainty, moral certainty into permission, and permission into something nobody can take back.
The responsibility for crossing that final line belongs to the person who crosses it.
But institutions have a responsibility too:
Do not spend years destroying people’s faith in every alternative and then act mystified that alternatives were what kept the peace.
A functioning society does not survive because human anger has no breaking point.
It survives because people continue believing there is something meaningful they can do before reaching it.
We should behave accordingly.
References and Further Reading
Moral Injury: Definition and Clinical Evidence
Moral Injury: An Overview of Conceptual, Definitional, Assessment, and Treatment Issues — Annual Review of Clinical Psychology (2025) Brett T. Litz and Hannah E. Walker. A major contemporary review of moral injury, including trust-violation-related injury, grievance, anger, disillusionment, institutional distrust and loss of meaning.
Moral Injury — U.S. Department of Veterans Affairs, National Center for PTSD Clinical overview explaining potentially morally injurious experiences, including perpetration, witnessing, omission and betrayal, and common responses such as guilt, shame, anger and loss of trust.
Moral Injury — VA MIRECC / Center of Excellence Explains the historical development of the concept, including Jonathan Shay’s emphasis on betrayal by leadership and later clinical models of moral injury.
Moral Injury, Aggression and Radicalization
The Relationship between Moral Injury and Radicalisation: A Systematic Review — Studies in Conflict & Terrorism Victoria Williamson and colleagues. Reviews overlaps between moral injury and radicalization, including betrayal, victimization, anger, shame, loss of significance and disrupted identity, while emphasizing that direct causal evidence remains limited.
Posttraumatic Stress Disorder Symptom Clusters, Exposure to Potentially Morally Injurious Events, and Aggression Among Army Veterans — Clinical Psychology & Psychotherapy Reports associations between betrayal-related morally injurious experiences and aggression, anger, hostility and verbal aggression while identifying other factors as stronger independent predictors.
Grievance, Political Efficacy and Violence
Explaining Support for Political Violence: Grievance and Perceived Opportunity — Journal of Peace Research Karin Dyrstad and Solveig Hillesund. Comparative survey research from Guatemala, Nepal and Northern Ireland examining how grievance and perceived ability to influence politics relate to support for political violence.
Life Events and Change in Support for Political Violence in the United States — Injury Epidemiology Longitudinal U.S. cohort study finding that relatively few individual life events predicted changes in political-violence attitudes, but reporting an increase among respondents who said they had “given up on politics.”
Targeted-Violence Prevention
The Pathway to Intended Violence: Active Shooter Pathway Indicators — FBI Behavioral Analysis Unit Recent analysis of 59 active shooters finding documented grievance in every case and violent ideation in 90%, while cautioning against treating pathway indicators as a mechanical prediction tool.
Making Prevention a Reality: Identifying, Assessing, and Managing the Threat of Targeted Attacks — FBI Threat-assessment guidance describing grievance as one possible starting point on pathways toward targeted violence and identifying continued pursuit of lawful, socially sanctioned conflict resolution as a potentially mitigating factor.
Current U.S. Attitudes Toward Political Violence
Decreased Support for Political Violence in the United States in Mid-2026 — UC Davis Centers for Violence Prevention September 2026 preprint reporting results from 7,752 U.S. adults surveyed through the Ipsos KnowledgePanel. Most measures of support for political violence declined from 2025 to 2026, reinforcing that political violence remains rejected by the large majority of Americans.
Editorial currency note: Research on moral injury outside military settings, radicalization, political violence and behavioral threat assessment continues to develop. The 2026 UC Davis findings discussed above are currently reported as a preprint. Associations described in this article should not be interpreted as deterministic predictors of individual violent behavior.
