Why We Push People Away

The Belonging Project, book 8 - Jamie Anson

WHY WE PUSH PEOPLE AWAY Fear, Rejection, and the Hidden Cost of Understanding Each Other

Jamie Anson Peer Review Manuscript · September 2026

AUTHOR'S NOTE Understanding is not agreement. Empathy is not allegiance. Closeness is not surrender.

This book begins with an ordinary human contradiction. We want to be understood. We are often terrified of what understanding someone else might cost us. That fear appears in intimate relationships when a partner says, "If I admit you have a point, I lose." It appears in families when one person's new identity seems to force everyone else to declare a side. It appears online when people publicly mock a stigmatized stranger less to persuade the stranger than to reassure an audience about where they stand. It appears in workplaces when defending the unpopular employee risks being classified with them. The book's central hypothesis is that alignment can carry social risk. If I move closer to your reality, other people may update their model of me. Understanding you may be mistaken for endorsing you. Defending your dignity may be mistaken for joining your camp. Admitting you are right about one thing may be treated as surrender on everything. Being seen with you may alter my status. That creates an incentive to push you away before anyone can confuse us. Some of this territory overlaps with established work on rejection sensitivity, stigma, social identity, fear of isolation, conformity, impression management, and the spiral of silence. This book is not claiming those literatures failed to notice that social rejection matters. The narrower contribution is an attempt to connect several phenomena under one question: What is the perceived cost of aligning, even partially, with this person? "Alignment" here does not mean obedience, trust, attraction, or access. It means moving some part of your model toward theirs: acknowledging a fact, recognizing an emotion, defending personhood, sharing social space, admitting similarity, or allowing their perspective to influence yours. Sometimes alignment is dangerous. People can manipulate empathy. Groups can recruit through apparent understanding. Abusers can use sympathy to regain access. Some ideas deserve rejection. Some people require distance. The goal is not universal closeness. It is to recover distinctions fear collapses. I can understand you without obeying you. I can agree with one claim without joining your tribe. I can defend your dignity without defending your behavior. I can love you without giving you access.

I can change my mind without losing myself. And perhaps most importantly: I can stand close enough to see you accurately without becoming you.

PROLOGUE: DON'T STAND TOO CLOSE Sometimes rejection is not aimed at the person in front of us. It is a message to everyone watching.

A stranger posts something online. The post is not especially persuasive. It is designed to provoke. A stigmatized person appears in the story. Within minutes, the comments organize themselves. Disgust. Jokes. Moral certainty. People compete to say how far they are from the target. I would never. Imagine being that. Keep those people away from me. What is the comment doing? Maybe it expresses a sincere belief. Maybe anger. Maybe prejudice. Maybe moral conviction. But sometimes the primary audience is not the person being mocked. It is the tribe. Look where I stand. Do not confuse me with them. I know the rule. I am safe. Humans are sensitive to association because association can alter how other people classify us. Who are your friends? Whose post did you like? Who did you defend? Who did you date? Who did you refuse to condemn? Who did you say was "actually kind of right"? The answer can move you socially. Research has long documented nearby phenomena. People may self-censor when they fear social isolation. Rejection-sensitive people may anxiously expect rejection, detect it in ambiguous behavior, and react in ways that can help produce the very rejection they fear. Stigma can spread by association under some conditions. But there is a useful everyday question hiding among those findings. How much social risk do I believe I incur by moving toward you? Call it alignment risk. If your status inside my group is low, the cost may be high. If the group is secure, the cost may be low. If my membership is precarious, the cost may be enormous.

If my identity is already questioned, I may distance from you aggressively just to make the boundary visible. This may help explain some behavior that otherwise looks bizarrely cruel. The person does not merely reject. They over-reject. They perform distance. They add contempt where disagreement would have been sufficient. Why? Because ambiguity is risky. The intimate version may be even more painful. I love you. I am afraid you will leave. So I become colder first. I fear you think I am weak. So I attack before you can see me asking for reassurance. I fear that understanding your complaint means admitting I am the bad person. So I refuse to understand. I fear that if I move one inch toward your reality, I will lose the argument, the relationship, my status, or myself. So I push. Then you push back. And both of us say: See? I knew you were unsafe. This book is about that move. The hand that creates distance while insisting it is protecting us from distance.

PART I — THE FEAR OF BEING MOVED Understanding can feel dangerous when identity, status, or membership seems to be at stake.

1. ALIGNMENT RISK What will it cost me if I move one inch toward your reality?

Imagine two people arguing. One says, "You hurt me." The other hears more than the sentence. If I acknowledge that, I am guilty. If I am guilty, I lose moral standing. If I lose moral standing, you control what happens next. So the reply is not: Tell me what hurt. It is: That is ridiculous. Now the argument appears to be about facts. Underneath, it may be about the perceived cost of alignment. Alignment risk is the anticipated social, relational, identity, or status cost of moving toward another person's position, experience, or humanity. The movement can be tiny. "I see why you felt that way." "You are right about that part." "I don't know enough yet." "They still deserve dignity." "I like that person." "I was wrong." Each sentence can carry more social meaning than its literal content. That is why people refuse sentences they privately know are true. A son knows his father has a point but refuses to concede because the relationship is already organized around dominance. A manager sees that the unpopular employee was treated unfairly but fears defending them will damage credibility with the team. A friend understands why someone left the group but stays silent because the group treats understanding as betrayal. The hidden question is: what will people infer about me if I align here? That is alignment risk. There is a useful way to make the idea more precise. Alignment risk is not simply fear of disagreement. It is the expected cost of being *seen moving*. That makes public and private alignment different. Privately, I may admit that someone I dislike made a good point. Publicly, saying so may create reputational consequences.

Privately, I may feel compassion for a family member everyone has condemned. At dinner, expressing that compassion may be interpreted as betrayal. Privately, a manager may think the unpopular employee was treated unfairly. In the meeting, defending them may alter how the manager's own judgment is evaluated. The audience changes the price. So does the security of the person paying it. A person with high status can often afford nuance because their membership is harder to revoke. A person whose status is already precarious may need to demonstrate loyalty more aggressively. That creates an uncomfortable implication. The people who perform the most contempt may not always be the people with the strongest underlying conviction. They may be the people who believe ambiguity is most dangerous for them. This is a hypothesis worth testing because it predicts something counterintuitive: secure insiders may sometimes show more tolerance for partial alignment than insecure insiders, even when their private beliefs are similar. If true, reducing stigma would sometimes require less moral instruction and more membership security.

2. UNDERSTANDING IS NOT AGREEMENT Fear becomes dangerous when distinctions collapse.

There are several separate acts that humans routinely fuse. Understanding. Agreement. Approval. Trust. Affiliation. Access. Loyalty. They are not the same. I can understand why you lied without approving the lie. I can agree that you were treated unfairly without agreeing with your worldview. I can defend your right to speak without wanting you in my home. I can love you without trusting you with money. I can trust your expertise without wanting your friendship. I can be your friend without believing you are right. Fear compresses these distinctions. If I understand, I approve. If I approve, I agree. If I agree, I join. If I join, I betray us. Once that chain becomes automatic, understanding itself becomes threatening. This creates a peculiar form of ignorance. People avoid accurate perception because accurate perception may produce morally or socially inconvenient information. Maybe the person we hate has a legitimate grievance. Maybe the family member we exiled was right about one thing. Maybe the employee everybody dislikes is still correct about the numbers. None of these facts requires surrender. But if our social system treats nuance as disloyalty, the facts become costly to notice.

3. THE SOCIAL COST OF SAYING YOU'RE RIGHT A concession can move more than the argument.

Most arguments contain a scoreboard nobody admits exists. Who is winning? Who is reasonable? Who owes whom an apology? Who gets to define what happened? This means a factual concession can become a status transfer. You are right about the time. You are right that I said it. You are right that I interrupted. Those should be small updates. But if the relationship is insecure, each can feel like handing over the whole case. People defend claims they do not care about because the claim has become a border. This is one reason repair fails over tiny facts. The disputed detail represents whether one person's reality can move the other's. A secure relationship lowers this cost. You can be right without becoming ruler. I can be wrong without becoming subordinate. Once that becomes believable, truth gets cheaper.

4. GUILT BY ASSOCIATION Sometimes the risk is not what you believe. It is who people think you are because of who you stand beside.

Stigma can travel through relationships. Researchers have described courtesy stigma or stigma by association: social disapproval directed toward people because of their association with someone or a group that is stigmatized. The effect is not universal. Context, prejudice, voluntary association, and the particular stigma all matter. But the underlying social fact is familiar. People infer identity from proximity. Who you defend. Who you date. Who you hire. Who you invite. Who you refuse to laugh at. This makes association informational. The group asks: Why are you standing there? Are you one of them? The person anticipating that inference may create distance before anyone asks. That distance can be physical. It can also be linguistic. "Those people." "I could never." "Not that I support them, but..." The disclaimer comes first because the speaker is protecting the social location from which the next sentence will be heard. Sometimes the disclaimer is prudent. Sometimes it is the sound of alignment risk becoming visible.

Association also explains why people sometimes abandon someone immediately after that person becomes stigmatized. The relationship did not necessarily change in one afternoon. The social meaning of the relationship did. Yesterday, standing beside this person cost nothing. Today, it communicates something. That can make loyalty suddenly expensive. A friend who was easy to love in private becomes difficult to defend in public. A colleague who seemed competent becomes reputationally dangerous. A family member becomes the subject nobody wants photographed beside. We often interpret the abandonment as discovery: Now everyone finally sees who the person really is. Sometimes. But another possibility is that everyone already knew much of the relevant information and the *cost of association* changed. That distinction matters because it shifts the question from: Why did everyone suddenly change their moral judgment? to: What happened to the social price of being seen as close? The answer can be ugly. But it can also be measurable.

PART II — THE REJECTION MACHINE Fear of rejection can produce behavior that makes rejection more likely.

5. I WILL LEAVE BEFORE YOU CAN Pre-rejection converts uncertainty into control.

Waiting to be rejected is excruciating. The partner is quieter. A message takes longer. A friend seems distant. A boss schedules a meeting. The nervous system starts forecasting. Maybe this is it. One way to end uncertainty is to create the ending yourself. Become cold. Pick a fight. Disappear. Insult. Withdraw affection. Announce that you never cared. Now at least the outcome belongs partly to you. This can look irrational because the person creates what they feared. From inside, it may feel like control. I did not get abandoned. I ended it. I did not get exposed as needy. I proved I did not need you. I did not wait for the tribe to downgrade me. I left first. The cost is obvious. The strategy protects against uncertainty by increasing the probability of loss. It replaces feared rejection with engineered rejection. That can become stable because the final outcome confirms the original model. People leave. Yes. But sometimes they leave after being repeatedly pushed toward the door. Pre-rejection can occur in groups too. A person senses that their standing is falling. They stop attending. Mock the group. Pretend they never cared. Attack the leader. Announce that everyone is fake. Maybe the criticism is valid. But the timing may also protect against the unbearable interpretation: I was becoming less wanted. The person converts social demotion into voluntary departure.

This can preserve dignity while destroying information. If every threatened membership gets rewritten as "I rejected them first," the person never gets to learn whether the relationship could have been repaired. That does not mean people should stay where they are unwanted. It means the motive for leaving deserves inspection. Am I leaving because the relationship is wrong? Or because uncertainty about my place has become intolerable? Those are different problems.

6. REJECTION SENSITIVITY Some people enter ambiguous moments already expecting the worst answer.

Psychologists Geraldine Downey and Scott Feldman used the term rejection sensitivity for a pattern in which people anxiously expect, readily perceive, and overreact to rejection. That framework matters here because ambiguous social information is unavoidable. A partner is tired. A friend forgets to reply. Someone laughs across the room. A manager gives brief feedback. None of these events arrives with a certified interpretation. Expectation helps fill the gap. If rejection is already the predicted outcome, ambiguous behavior is easier to classify as rejection. Then the response becomes defensive. Anger. Withdrawal. Hostility. Testing. Reassurance seeking. Coldness. Later work has described vicious cycles in which hostile or defensive reactions to perceived rejection can elicit distancing from the very people whose acceptance matters. This is extraordinarily close to the mechanism we have been exploring across the series. Fear creates behavior. Behavior changes the other person's forecast. Their response becomes new evidence. The fear says: See? I was right. The problem is not that rejection-sensitive people imagine everything. Sometimes rejection is real. The problem is that anticipation can change the probability distribution. A crucial point in the rejection-sensitivity literature is that defensive responses can become self-fulfilling without the original fear being imaginary. That subtlety matters. A person may have learned rejection sensitivity from real rejection. Their nervous system may be calibrated by history, not fantasy. The mistake is not necessarily expecting danger. It may be applying an old probability to a new relationship. Someone who grew up in a home where criticism predicted withdrawal may reasonably learn that disagreement is relationally dangerous. Then a later partner says, "I don't like what happened last night." The nervous system hears a much larger message. I am about to lose you.

The defensive response may be disproportionate to the current relationship and perfectly proportionate to the remembered one. Healing therefore requires more than telling the person not to overreact. They need enough new experiences for a different forecast to become rational. Disagreement. Then staying. Conflict. Then repair. No. Then love. Again and again. A secure relationship is partly a dataset.

7. THE TEST NOBODY CAN PASS If you loved me, you would survive the behavior I use to test whether you love me.

Relationship tests often begin with a legitimate need. I want evidence that you will stay. Then the test is designed in a way that damages staying. The person becomes cruel. Unavailable. Jealous. Demanding. Unpredictable. They create a scenario where the other person has to prove love under increasingly bad conditions. If you stay, maybe you pass. But passing teaches the tester that escalation works. If you leave, the fear is confirmed. Either result can preserve the test. The healthier target is not infinite proof. It is enough security that ordinary uncertainty can be tolerated without manufacturing emergencies. A relationship cannot become safe if safety requires repeated demonstrations that make the relationship less safe. Some tests are social rather than romantic. Say the correct thing. Condemn the correct person. Post the correct symbol. Laugh at the correct joke. Refuse contact with the correct outsider. The member learns that belonging is periodically renewed through demonstrations. This creates a strange insecurity even among conformists. If loyalty has to be proven repeatedly, nobody's status is fully settled. That can increase the demand for stronger signals. Yesterday's statement is not enough. Today's controversy requires another. The group becomes a loyalty treadmill. Members may eventually spend more energy proving they belong than deciding what they believe. The same structure can exist in families. Who called Mom? Who visited?

Who sided with whom? Who defended the outsider? Every interaction becomes a membership audit. People stop relating. They certify.

8. ANGER IS OFTEN FASTER THAN NEED It can be easier to say get away from me than please don't leave.

Need is vulnerable. Anger is armored. The person feels ignored. The primary experience may be hurt. But hurt exposes dependence. I cared enough for this to matter. Anger transforms the position. Now I am not the person who wanted you. I am the person rejecting you. A text goes unanswered. "Are you okay?" becomes "Forget it." A partner asks for space. "I'll miss you" becomes "Do whatever you want." A friend disappoints. "That hurt" becomes "You're fake." The angry sentence may contain real grievance. But it also performs distance. It says: do not look at how much power your absence has. This is one reason emotional literacy matters. If people cannot distinguish hurt from rage, the relationship receives the defensive emotion and never hears the original need. Then the other person defends against the anger. The need remains unmet. The anger increases. Another fear loop.

PART III — PERFORMING DISTANCE Sometimes cruelty is a public credential.

9. THE AUDIENCE IS THE REAL TARGET Public contempt often communicates sideways.

A humiliating comment aimed at one person may be written for a hundred others. That changes how we should read it. The literal target receives the insult. The social audience receives information about the speaker. I know the norm. I share the disgust. I am not confused. I am not one of them. This is why online contempt can become competitive. Once the group establishes the correct distance, members demonstrate even greater distance. A joke becomes a harsher joke. A criticism becomes disgust. A disagreement becomes dehumanization. The escalation may have little to do with persuading the target. It is a loyalty performance. The group is watching for ambiguity. The safest member is the one nobody could possibly mistake for sympathetic. That dynamic is especially likely when membership itself feels uncertain. People with secure status can sometimes afford nuance. People whose place is precarious may need stronger signals. The presence of an audience also changes what repair looks like. A person who insulted someone privately can apologize privately with relatively low status cost. A person who humiliated someone publicly may now face a second problem. The audience has seen the hierarchy. An apology changes it. That can make public repair much harder than public harm. The speaker may privately regret the cruelty and still resist correcting it because correction would reveal movement. This is one reason online conflicts become sticky. Thousands of observers become witnesses to identity. Changing your mind is no longer a cognitive update. It is a public repositioning. The cost rises with visibility.

A healthier public culture would make updating prestigious rather than humiliating. Until then, certainty will often be cheaper than correction.

10. MEMES AS MEMBERSHIP BADGES Provocation can compress an entire loyalty test into one image.

A meme is efficient because it eliminates argument. The format says: you either get it or you do not. You laugh or you become suspicious. The stigmatized target is simplified into one recognizable category. The joke rewards instant classification. That makes memes unusually good vehicles for belonging signals. Share this and your audience knows where you stand. Refuse to share and nothing happens. Object, and suddenly you may become the interesting person. Why are you defending them? That question exposes the mechanism. The content may be crude. The social test is sophisticated. Can you tolerate the possibility that defending the target's humanity changes how your group sees you? This is why counter-speech often becomes identity conflict. The original claim disappears. Now everyone is negotiating affiliation. The meme format also makes deniability cheap. It was just a joke. Relax. Why are you so sensitive? The speaker gets two benefits. The in-group receives the signal. The target is denied the right to treat the signal as serious. This asymmetry makes meme-style stigma especially efficient. The message can wound downward while remaining unserious upward. If challenged, the speaker retreats into comedy. If celebrated, the speaker accepts the social reward. The content therefore operates in two registers at once. Literal unseriousness. Social seriousness. That is one reason arguing with the meme at the literal level often feels useless.

The more interesting question is what membership work the joke is performing.

11. THE DISCLAIMER I'm not defending them, but... is sometimes cowardice and sometimes intelligent boundary-setting.

The disclaimer deserves more respect than it usually gets. It can be cowardice. The person knows the group will punish nuance, so they preemptively perform loyalty. But the disclaimer can also be useful. "I do not support what she did, and I think the punishment was disproportionate." That sentence separates behavior from personhood. "I disagree with his worldview, and he is right about this data." That separates global affiliation from local agreement. "I am not asking you to trust her again. I am asking you to stop humiliating her." That separates dignity from access. The problem is not the existence of disclaimers. It is when every act of understanding requires one. A healthy group should tolerate sentences whose meaning is local. You are right about X. Without immediately demanding: so whose side are you on?

12. THE STIGMA FLARE Groups sometimes intensify stigma exactly when boundaries feel blurry.

Why do some social moments produce sudden waves of ridicule? One possibility is boundary maintenance. A category that once seemed distant becomes visible. Someone famous joins it. A friend discloses membership. A norm changes. The old border feels less obvious. The group responds by making the stigma louder. The joke is not only about the target. It repaints the line. This can happen around class, addiction, sexuality, mental illness, criminal history, occupation, religion, body size, disability, or any trait a group uses to organize social distance. The specific moral content differs dramatically across cases. The mechanism can still rhyme. Ambiguity increases. Boundary signaling increases. People exaggerate distance. This is a hypothesis, not a universal explanation for stigma. But it suggests a useful question when ridicule suddenly becomes intense: What boundary is this group afraid is becoming less legible?

Stigma flares also explain why some people become more punitive when a category moves closer to home. A stranger is abstract. A neighbor is not. A family member is not. Once proximity increases, the old stereotype has to coexist with a person the group already values. That creates cognitive and social tension. One possible response is compassion. Another is stronger distancing. The group intensifies the negative category to protect the old rule. This is why the first person in a family or institution to violate a norm can receive a disproportionate response. The person is not merely breaking the rule. They are threatening the clarity of the category system. Later members may face less punishment because the boundary has already been redrawn. Again, this does not make every norm wrong. It explains why the first breach can feel existential.

PART IV — WHEN LOVE BECOMES A STATUS SYSTEM The people who matter most can create the largest alignment risks.

13. IF I TAKE YOUR SIDE, WHO AM I AGAINST? Families often convert understanding into coalition.

Family conflict loves triangles. Two people disagree. A third is asked to understand. Understanding is immediately converted into side-taking. You agree with Mom? So you are against me. You talked to him? After what he did? You can see her point? Whose family are you in? Now the third person faces alignment risk. They may privately see complexity and publicly simplify because complexity threatens a relationship with one side. This is how families lose reality. Each person gives the answer that preserves the most important bond. Eventually the family has many private models and one public fiction. The antidote is not neutrality. Some situations require clear judgment. The antidote is allowing local alignment. Mom is right about this. Dad is right about that. You were harmed. You also harmed someone. I love both of you. I am not available to carry messages. Those sentences refuse coalition without refusing relationship. Children become particularly sensitive to this in conflict between parents. A child may discover that empathy itself is dangerous. If I comfort Dad, Mom becomes hurt. If I defend Mom, Dad asks whose side I am on. The child learns to hide local truths. One parent was kind today. The other was unfair yesterday. Both still matter.

A child should not need diplomatic training to preserve access to both parents. The healthier adult system allows the child to align locally without assigning coalition. You can love your mother here. You can miss your father there. You do not have to solve our conflict in order to belong to both of us. That is a profound gift because it prevents the child's empathy from becoming a loyalty test.

14. THE JEALOUSY OF UNDERSTANDING Sometimes understanding a rival feels like giving them rank.

Jealousy is not only about sex. It is about priority. If my partner understands another person deeply, I may experience that understanding as intimacy. If my parent defends my sibling, I may hear preference. If my boss praises the colleague I distrust, I may hear rank movement. Now empathy itself becomes scarce. Why are you explaining them? Why are you taking their side? Why do you care what they think? The hidden fear is that understanding allocates relational priority. Sometimes it does. Attention is limited. Time is limited. Loyalty matters. But understanding does not automatically transfer rank. A secure relationship can tolerate the possibility that somebody else is intelligible. The insecure relationship needs the rival to remain wrong so the hierarchy remains clear. This dynamic is especially strong when priority is already uncertain. If I know I matter to you, your curiosity about someone else may feel interesting. If I am unsure, the same curiosity can feel like displacement. That suggests jealousy is partly about the interpretation of attention under uncertain rank. The act is small. You laughed at their joke. You defended them. You texted them back quickly. The inference is large. They are moving above me. This is why reassurance sometimes needs to be structural rather than verbal. Not only "you matter." Clear behavior about priority when priority actually matters. Who gets called in crisis? Whose plans are protected?

What promises exist? What boundaries organize the relationship? Predictable priority can make outside empathy less threatening because the relationship no longer has to infer rank from every small act.

15. THE ARGUMENT ABOUT THE ARGUMENT Couples often fight less about what happened than about whose interpretation will govern the relationship.

One person says: You ignored me. The other says: I was overwhelmed. Those explanations can coexist. But each person hears competition. If you were overwhelmed, maybe my hurt is invalid. If I was hurt, maybe your overwhelm is no excuse. The couple begins arguing over which truth is allowed to survive. This is where dialectical thinking becomes more than a therapy word. Two truths can alter the relationship simultaneously. You were overwhelmed. I was hurt. You did not intend rejection. I experienced rejection. My experience matters. Your intent matters. Neither sentence erases the other. Alignment risk falls when neither person's reality needs exclusive control.

16. APOLOGIES FEEL LIKE SUBMISSION WHEN POWER IS UNCLEAR I'm sorry becomes dangerous when it is interpreted as surrender.

Some people avoid apologies because they do not care. Others avoid them because apologies have been weaponized. The moment I admit fault, you get to punish me. The moment I say sorry, you become right about everything. The moment I acknowledge harm, I lose my ability to describe what you did. In such relationships, apology carries too much jurisdiction. A healthy apology should be local. I am sorry I said that. It does not mean everything you believe about me is correct. It does not mean your behavior no longer matters. It does not mean you now own the moral hierarchy forever. The ability to apologize depends partly on whether the relationship knows how to receive an apology without turning it into conquest. If admitting one inch causes total defeat, people will defend absurd ground.

The receiver has a role too. A good apology should reduce the need for punishment. If apology merely gives the injured person new ammunition, future honesty becomes less likely. The receiving sentence matters: Thank you for saying that. I am still hurt. We still need to talk about what happens next. I am not going to use your apology as proof that every part of my story is correct. That is relational containment. The harm is acknowledged without allowing one admission to absorb the whole person. When apology is safe, truth gets cheaper. When apology is conquest, defensiveness becomes predictable.

PART V — STRANGERS, GROUPS, AND THE COST OF EMPATHY The farther away someone is socially, the easier it can be to make them simple.

17. EMPATHY CAN FEEL LIKE CONTAMINATION The feared sentence is not I understand them. It is people may think I am like them.

Empathy is often presented as costless virtue. Just imagine how they feel. But empathy can carry identity consequences. If the target is stigmatized, understanding them can create suspicion. Why do you know so much? Why are you defending this? Have you done it too? Are you one of them? The person anticipating these questions may protect themselves by becoming less empathic in public than they are in private. This is one reason moral courage sometimes looks embarrassingly small. Not a speech. Just refusing to join the joke. Not correcting everyone. Just declining to make the target less human. The risk may still feel real. Social identity is partly made of other people's inferences. The question is whether we can build groups secure enough that curiosity does not automatically threaten membership. There is a second cost of public empathy: moral suspicion. People worry that explanation will become excuse. Why are you trying to understand someone who did something terrible? The fear is reasonable because people do sometimes use explanation to minimize responsibility. But refusing explanation creates another problem. We become unable to study mechanisms that produce harm. If every attempt to ask "why" is interpreted as sympathy, prevention becomes morally risky. This is especially dangerous in domains where the behavior is frightening. Violence. Abuse. Extremism. Addiction.

Betrayal. Understanding the path does not remove responsibility for the destination. It improves the chance of seeing the path earlier next time.

18. MISCLASSIFICATION RISK Sometimes we reject people less because of who they are than because of who others may decide we are if we understand them.

There is a particular fear underneath public distancing that deserves its own name. Misclassification risk. I understand you. Someone watching concludes I am like you. I defend one part of what you said. Someone concludes I endorse the whole thing. I remain your friend after a scandal. Someone concludes I share your values. I refuse to laugh at the joke. Someone wonders why I am sensitive. The feared event is not merely disagreement with the group. It is being assigned the wrong identity by the group. That matters because human beings do not control their social identities alone. Other people infer who we are from patterns of association, language, proximity, taste, loyalty, silence, and defense. Sometimes those inferences are reasonable. If I repeatedly defend a movement, fund it, organize for it, and adopt its symbols, observers are not irrational to update their model of me. The problem begins when tiny local acts are treated as global evidence. I acknowledge one valid grievance and become a supporter. I show compassion and become compromised. I question a punishment and become an apologist. I speak to the wrong person and become suspect. Now accurate perception carries reputational risk. That changes behavior before anyone explicitly threatens us. We begin editing what we notice. Not because reality changed. Because the social cost of noticing it changed. This helps explain why people sometimes perform more distance than their private judgment requires. A quiet disagreement may not be enough to protect identity. Contempt is clearer. Ridicule is clearer.

Public condemnation is clearer. The stronger the signal, the less likely observers are to misclassify the speaker. That gives cruelty a strange social advantage: it is unambiguous. This also suggests a prediction. Misclassification risk should rise when three things are true at once: the target is highly stigmatized, the response is public, and the observer's own membership or status is uncertain. Under those conditions, even mild empathy may feel expensive. The implication is not that everyone should ignore reputation. Reputation is real. Association matters. People can rationally care what others infer. The question is whether we can build social systems where a local sentence is allowed to remain local. I can say that person was treated unfairly without signing up for everything they believe. I can refuse dehumanization without accepting access. I can understand a motive without defending an act. I can remain accurate even when accuracy creates ambiguity about me. A society that cannot tolerate that ambiguity forces people to choose between truth and social legibility. Many will choose legibility.

19. THE WRONG PERSON CAN STILL BE RIGHT — AND STILL NOT GET ACCESS Local truth, human dignity, and relational access are separate judgments.

Humans do not evaluate claims in a vacuum. We evaluate speakers. If the speaker belongs to a distrusted group, the claim inherits cost. Agreeing may feel like giving them status. This creates a bizarre incentive to prefer a worse model over a socially inconvenient concession. The intellectually mature move is uncomfortable: That person is wrong about many things. They are right about this. This should be ordinary. Instead it can feel like betrayal. The cost is not merely polarization. It is information loss. Groups become less able to learn from adversaries because learning itself threatens identity. A secure mind can take accurate information from an enemy without becoming the enemy. But accuracy raises a second fear. If I admit the person is right about something, am I now required to trust them? Forgive them? Let them back in? No. This is where three judgments need to be separated. Is this claim true? Does this person retain human dignity? What access is safe? Those questions can produce three different answers. A person can be factually right about one issue and still be unsafe in a relationship. A person can have behaved terribly and still remain human. A person can deserve compassion and still never return to your home. This matters in cases involving violence, abuse, exploitation, repeated deception, or serious boundary violations.

Understanding is not exposure. Dignity is not trust. Compassion is not reconciliation. Accuracy is not reinstatement. Those distinctions make empathy safer because they remove the false bargain underneath so much defensive thinking. I do not have to make you wrong about everything in order to protect myself from you. And I do not have to let you close in order to admit that something you said is true. Once those decisions are separated, truth stops carrying jurisdiction it never deserved.

20. WHY PUBLIC SHAMING FEELS SO GOOD Condemnation can purchase both moral status and group security at once.

Public punishment has an obvious function. Groups enforce norms. Some behavior deserves condemnation. But public shaming can produce rewards beyond accountability. It tells the audience: I am one of the good people. The target's fall creates contrast. The more contaminated the target becomes, the cleaner the condemner can appear. This is why punishment can continue after its practical purpose is exhausted. The group is no longer merely responding to harm. It is producing identity. That does not mean all public criticism is performative. It means criticism can do two jobs at once. Address behavior. Signal membership. The more the second job dominates, the less proportional the response may become. Restoration becomes threatening because the target's permanent contamination is socially useful.

Shaming also creates a temporary coalition among the shamers. People who otherwise have little in common can unite around a target. That unity feels like belonging. The target becomes social glue. This is one reason groups can seem reluctant to let a controversy die. If the enemy disappears, so does the organizing principle. The group has to discover what else it is for. A healthy community can coordinate around shared purpose. An insecure community may coordinate around shared contempt. Both produce cohesion. Only one requires an enemy to remain available.

PART VI — SAFE ALIGNMENT The answer is not infinite empathy. It is learning to move toward reality without surrendering judgment.

21. PARTIAL AGREEMENT Civilization depends on the ability to agree locally.

A mature mind can partition. You are right about this sentence. Wrong about the next. Trustworthy in this domain. Untrustworthy in another. Kind to me. Cruel to someone else. A victim in one relationship. An aggressor in another. Human beings resist this because global categories are efficient. Good person. Bad person. Our side. Their side. Safe. Dangerous. Partitioning requires more cognitive work. It also reduces alignment risk. If agreement is local, I do not have to join you in order to learn from you. This may be one of the most important skills for pluralistic societies and complicated families alike. Local agreement allows information to cross borders without dragging identity behind it. Partial agreement is difficult because language likes wholes. I agree with her. That sentence sounds global. More accurate language is slower: I agree with her on this point. I think his description of the event is accurate even though I disagree with his conclusion. I believe that behavior was wrong and that the person was also treated unfairly afterward. Precision lowers the chance that local alignment will be misread as total allegiance. This may sound pedantic. It is actually social technology. The more polarized the environment, the more useful local language becomes. It lets reality move in pieces.

22. THE SENTENCE I CAN SEE THAT Recognition can lower threat without conceding the whole case.

Consider the phrase: I can see that. Not: you are right about everything. Not: I agree. Not: I will do what you want. Just: I can see how you got there. That sentence can transform conflict because it answers a belonging question hidden inside many arguments. Does my reality exist inside your model at all? People escalate when the answer appears to be no. They repeat. Speak louder. Add evidence. Recruit witnesses. Use more extreme language. The argument becomes partly an attempt to force representation. Accurate recognition can reduce that pressure. Again, this is not universal. Manipulative people can exploit recognition. Some conflicts need decisive boundaries. But in ordinary relational conflict, being represented accurately inside another person's mind can itself be regulating. Recognition can also be directed at fear itself. "I can see why that scares you." This sentence does not concede that the feared outcome is likely. It acknowledges the internal logic. That is often enough to reduce the person's need to make the fear louder. People escalate when they believe the other side is refusing to represent the threat. Once the threat is represented, probability can be discussed. I see why you are afraid I will leave. I am not planning to leave. Those are different sentences. Both can be true. The first creates contact. The second creates information.

23. AGREE WITHOUT DEFECTING You can move toward an outsider without moving away from everyone you already love.

Groups often treat alignment as zero-sum. If you understand them more, you understand us less. If you defend them, you betray us. If you change your mind, you abandoned the family. That assumption creates intellectual captivity. A healthier group can say: Bring back what you learned. Tell us where we are wrong. Keep loving us. Do not confuse loyalty with intellectual immobility. This is particularly important for children becoming adults, partners forming independent friendships, employees learning outside the company, and religious people encountering other traditions. The group becomes stronger when external information can enter without the messenger being treated as contaminated.

24. THE COURAGE TO BE MISCLASSIFIED Sometimes doing the humane thing means allowing other people to misunderstand where you stand.

You defend someone. People assume you endorse them. You refuse the joke. People become suspicious. You admit the rival has a point. Your group calls you disloyal. There may be no perfect explanation that preserves every audience. At that point alignment becomes a courage problem. Can you tolerate temporary misclassification? Not because reputation does not matter. It does. Not because every association is harmless. It is not. Because a life organized entirely around preventing incorrect social inference eventually becomes a performance for the most suspicious person in the room. Sometimes integrity requires saying: You may misunderstand me. I am still not going to lie about what I see.

The courage to be misclassified has limits. Reputation can affect employment, safety, custody, family relationships, and access to community. People with less power may rationally choose more caution. The book should not romanticize social risk from a comfortable distance. Sometimes the brave act is private. A quiet refusal to participate. A one-to-one correction. A relationship preserved across a taboo boundary. A decision not to forward the joke. Courage is not measured by audience size. The point is simply that perfect control over other people's inferences is impossible. At some point, a person has to decide whether avoiding misclassification is worth falsifying reality.

PART VII — STOP PUSHING The deepest change is not becoming fearless. It is noticing which defensive move is about to manufacture the rejection you fear.

25. THE PAUSE BEFORE DISTANCE There is a moment before the door closes.

The push-away move often happens quickly. Delete the message. Block. Insult. Withdraw. Post the joke. Say whatever. Refuse the apology. Leave the room. Announce the relationship is over. Sometimes those actions are necessary. Sometimes they are reflex. The useful question is: What am I trying to prevent right now? Being rejected? Being embarrassed? Being controlled? Being associated? Being wrong? Being seen needing someone? Losing rank? If the answer is visible, another response may become possible. Not always reconciliation. Sometimes a cleaner boundary. "I need an hour." "I think I am hearing this as rejection." "I disagree and I don't want to humiliate you." "I can see your point and I am not changing my boundary." "I am scared that if I admit that, I lose." Those sentences are not elegant. They are information. Information gives the other person a chance to respond to the actual threat rather than the defense. A useful pause has three questions.

What happened? What am I predicting? What am I about to do because of that prediction? These questions separate event, forecast, and defense. The message took six hours. I am predicting loss of interest. I am about to send something punishing. Now the mechanism is visible. The prediction may still be correct. But the person can choose whether the defensive behavior helps gather better evidence or merely forces the feared outcome faster. That tiny separation is where agency returns.

26. ASK BEFORE YOU ACCUSE A question leaves more futures available than a verdict.

"What did you mean?" is structurally different from: "So that's what you think of me." The first preserves ambiguity. The second resolves it against the relationship. Fear dislikes ambiguity because ambiguity is where rejection might be hiding. But premature certainty is expensive. The partner may have been tired. The friend may have been busy. The manager may have been brief. The stranger may not have noticed. Questions do not guarantee benign answers. They prevent fear from becoming the only witness. Questions work only when they are real questions. "Why do you always do this?" is an accusation wearing punctuation. "So you don't care?" is a verdict with a question mark. A real question contains answer space. Did you mean that as criticism? Are you asking for distance? Did I misunderstand what happened? What are you worried will happen if you agree with me? That last question can be especially revealing. It makes alignment risk explicit. Sometimes the argument changes immediately because the feared cost was never the literal subject.

27. SAY THE NEED BEFORE THE THREAT The relationship has a better chance when it hears what you want before what you will destroy.

"If you don't care, just leave." often means: Please show me that you care. "I don't need anybody." often means: I do not believe needing people is safe. "Do whatever you want." often means: I am afraid your choice will reveal my rank. The defensive sentence forces the other person to decode. Sometimes they do. Sometimes they obey the literal instruction. Then both people are devastated. Direct need feels riskier because it exposes the desired outcome. Please stay. Please reassure me. Please tell me where I stand. Please do not make me guess. Those sentences do not guarantee safety. They improve the information available to the relationship. Vulnerability is not always the right move, especially in coercive or unsafe relationships. But where mutual goodwill exists, concealed need often creates the very confusion it was trying to avoid. Direct need also reduces the amount of mind-reading required. Relationships deteriorate when each person has to infer the tender request hidden inside the defensive behavior. The partner is expected to know that "leave me alone" means "please come closer." The friend is expected to know that silence means "ask again." The parent is expected to know that rage means "tell me I still belong." Sometimes loving people decode correctly. Sometimes they respect the words.

Then the speaker experiences obedience as abandonment. A relationship cannot remain safe if the literal and intended messages routinely point in opposite directions. Say the need when it is safe enough to do so. Not because directness guarantees care. Because it gives care a target.

28. LET PEOPLE DISAGREE WITHOUT LEAVING Secure belonging makes ideological and emotional movement less catastrophic.

One of the strongest anti-polarization interventions may be surprisingly intimate: Prove that disagreement is survivable. A child disagrees and dinner still happens. A spouse says no and affection remains. An employee challenges the plan and is invited back. A friend changes their mind and is not humiliated. A believer asks a difficult question and is not treated as infected. Each event teaches the nervous system: Movement does not necessarily equal exile. That changes future alignment risk. People become more able to hear. Not because they were lectured into open-mindedness. Because curiosity became less socially expensive.

This is learned behavior, not merely a principle. Groups can practice it. Ask for dissent before decisions. Reward the person who identifies a real flaw. Let children hear adults change their minds without humiliation. Let friends hold different conclusions without turning every dinner into court. Let the person who apologizes remain respected. Let the person who says no remain invited where appropriate. Each event lowers the expected cost of future honesty. Eventually disagreement stops being a membership emergency and becomes information.

PART VIII — THE PEOPLE WE COULD BECOME A secure person does not need every border to be defended with contempt.

29. SECURE BELONGING PRODUCES BETTER EMPATHY When my own membership is less threatened, your humanity becomes less dangerous.

This may be the book's most optimistic hypothesis. Secure belonging could increase the capacity for accurate empathy. If I am confident that understanding you will not cost me my family, my status, or my identity, I can look longer. I can admit similarity. I can take in uncomfortable information. I can say you are right about something. I can change. This does not make secure people morally superior. It changes the price structure. Curiosity is cheaper. That suggests an unexpected route to better discourse. Do not only teach empathy. Build groups that do not punish their members for using it. Do not only teach critical thinking. Make updating less humiliating. Do not only demand courage. Reduce the first-mover cost of nuance. Good systems do not require every humane act to be heroic. This hypothesis also makes a prediction about power. The more secure member of a relationship or group can often afford to move first. The parent can tell the child disagreement is survivable. The boss can reward dissent publicly. The high-status member can defend the stigmatized outsider. The majority can make room for nuance. The person with more security absorbs more of the first-mover risk. This connects alignment risk to a principle from Love Is Infrastructure: power should carry more of the cost of making truth safe. Do not ask the most vulnerable person in the room to prove that empathy is safe. Let the secure person demonstrate it.

The mechanism matters because empathy is often treated as a trait. Some people are empathic. Others are not. But the same person can become more or less capable of seeing accurately depending on what accurate seeing appears to cost. Put someone in a secure relationship and they may be able to hear criticism. Put the same person in a relationship where every disagreement threatens abandonment and criticism becomes harder to process. Give an employee strong standing and they may defend the unpopular colleague. Make their own job precarious and silence becomes more rational. Give a teenager confidence that family membership survives difference and curiosity becomes cheaper. Make love conditional on ideological sameness and curiosity acquires a price. This suggests that empathy is partly infrastructural. We can ask people to become braver, more compassionate, and more open- minded. Or we can change the conditions under which those behaviors occur. A family can say explicitly that disagreement does not end relationship. A manager can defend the employee who brings inconvenient information. A group leader can model local agreement with an outsider without apologizing for the agreement. A friend can say, "You can understand them without losing me." Each act lowers the expected cost of future alignment. This is where secure belonging may create an epistemic advantage. If membership is less threatened, perception does not need to work as border patrol. The person can ask a cleaner question: What is actually true here? Instead of: What answer keeps me safe with my people? That difference is enormous. It means secure belonging may do more than make people feel better. It may improve the conditions under which people update. That claim is still a hypothesis. It should not be romanticized. Secure groups can be confidently wrong. High-status people can use security to ignore criticism. People can feel loved and remain prejudiced. Belonging does not mechanically produce wisdom.

The narrower prediction is comparative. When social punishment for partial alignment decreases, people should become more willing to represent an outsider's valid point accurately, all else equal. That can be tested. And if it holds, the implication is practical. Do not only teach empathy. Lower the cost of using it. Do not only tell people to speak honestly. Show them that honesty does not automatically terminate membership. Do not only praise independent thought. Build groups that can survive it. The safest tribe may not be the tribe that agrees most. It may be the tribe in which reality can move between people without every movement becoming a referendum on belonging.

30. THE PERSON YOU HATE IS NOT REQUIRED TO BE SIMPLE Moral clarity and psychological complexity can coexist.

There is a fear that understanding makes judgment weak. If I know why the person acted that way, perhaps I will excuse it. Sometimes explanation does soften judgment. Sometimes it makes judgment more precise. A person can have understandable motives and unacceptable behavior. A frightened person can be dangerous. A traumatized person can traumatize others. A rejected person can become cruel. A stigmatized person can also be wrong. Complexity does not abolish moral evaluation. It prevents moral evaluation from needing a cartoon. That matters because cartoons are easy to hate and impossible to learn from. If we want fewer harmful behaviors, we need mechanisms. Mechanisms require looking closer than contempt permits.

31. THE PEOPLE YOU LOVE ARE NOT REQUIRED TO AGREE Love becomes safer when disagreement stops functioning as an eviction notice.

Many relationships would become less frightened if one sentence became believable: You can disagree with me and still be mine. Not mine as possession. Mine as relationship. My child. My friend. My partner. My family. The relationship survives the difference. This is the architecture beneath intellectual freedom. A person who knows disagreement will not end belonging can examine ideas more honestly. A person who knows an apology will not create permanent subordination can apologize. A person who knows understanding a rival will not be treated as betrayal can listen. A person who knows they can say no can give a more meaningful yes. Secure belonging does not produce sameness. It produces room. The deeper promise is not "we will never change each other's minds." Good relationships should influence us. The promise is that influence is not ownership. You may persuade me. I may persuade you. We may both change. We may remain different. The relationship does not need to freeze either person in order to prove it exists. That is what makes alignment voluntary. I can move toward your view because the movement is not being extracted under threat. And I can remain where I am because disagreement is not automatically exile.

32. TEST ME WITHOUT EXILING ME Concern becomes safer when it can challenge a person without turning the challenge into a verdict.

There is a phrase I want to preserve because it captures much of the project: Test me without exiling me. Tell me what you see. Ask whether I might be wrong. Bring evidence. Name the pattern that worries you. Set a boundary if you need one. Do not pretend concern is impossible merely because judgment can hurt. But leave a path through which new evidence can matter. Do not convert the question into identity before the answer arrives. This principle applies almost everywhere. Parenting. Marriage. Friendship. Leadership. Medicine. Religion. Recovery. A good test can make a person more accurate. An exile threat makes accuracy compete with survival. That is a terrible design. If we want people to update, we need somewhere for the updated person to stand. The phrase also contains a standard for criticism. A test needs a proposition. What exactly are we testing? A behavior? A belief? A prediction? A pattern? Exile language skips the proposition and classifies the person. You are dangerous. You are crazy. You are selfish. You are one of them. Those labels may sometimes point toward real risk, but they are poor instruments for updating because they absorb too much. A better test is narrower. You said this would happen by June. It did not.

You told me this was a one-time event. It happened three more times. You say you want honesty, but you punish disagreement. Now the person can answer evidence with evidence. That is what makes correction possible without making belonging contingent on immediate submission.

33. WHAT IF ALIGNMENT RISK IS WRONG? The theory needs a way to lose.

Alignment risk is attractive because it can explain many different behaviors. Public contempt. Silence. Defensive anger. Pre-rejection. Stigma by association. Refusal to concede. That breadth is exactly why the concept needs discipline. How would we know if it adds anything? We could measure anticipated social cost of partial agreement. We could vary the status of the target, the security of the participant's group membership, the publicness of the response, and whether agreement is framed as local or global. We could test whether people become more willing to acknowledge a target's valid point when their own membership is explicitly secured. We could compare alignment risk with existing constructs such as fear of isolation, rejection sensitivity, identity threat, impression management, conformity, and stigma by association. If the new construct adds no predictive value, we should not keep it because the phrase sounds good. Maybe existing theories already explain the behavior better. That would be useful knowledge. The purpose is not to own a variable. It is to understand why humans sometimes make themselves crueler, lonelier, and less accurate in order to avoid being socially moved.

EPILOGUE: STAND CLOSE ENOUGH TO SEE You do not have to become someone in order to stop making them a silhouette.

We began with people pushing someone away. A joke. A cold reply. A refusal to concede. A public performance of disgust. A relationship test. A threat. A silence. The behaviors look different. The fear underneath can rhyme. If I move toward you, what happens to me? That is a legitimate question. Some people are dangerous. Some groups punish association. Some relationships use empathy as leverage. Some boundaries should not move. But fear can overprice proximity. It can tell us that understanding is surrender. That agreement on one point is allegiance. That compassion is contamination. That apology is defeat. That another person's humanity threatens our own membership. Those equations are often wrong. You can understand and still say no. You can see the wound and still protect yourself from the behavior. You can admit the enemy is right about one thing. You can let your partner's experience exist without surrendering your own. You can defend a stranger without joining their tribe. You can change your mind without disappearing. You can let somebody misunderstand your allegiance rather than performing cruelty for reassurance. And you can notice the moment you are about to push someone away because you are afraid they will leave. That moment contains possibility.

Maybe the relationship still ends. Maybe the group still rejects you. Maybe the person really is unsafe. This book does not promise closeness. It asks for accuracy before distance. Stand close enough to see. Then decide.

A NOTE ON EVIDENCE AND SCOPE Where established research ends and the alignment-risk hypothesis begins.

Several established research traditions overlap with the mechanisms in this book. Rejection sensitivity research, beginning with work by Geraldine Downey and Scott Feldman, describes a disposition to anxiously expect, readily perceive, and overreact to rejection. Subsequent research has linked perceived rejection in rejection-sensitive people with hostility and reactive aggression and has described vicious cycles in which defensive responses can elicit further rejection. Research on interpersonal rejection more broadly has found associations between rejection and anger or aggression, while proposed mechanisms include social pain, frustration, threatened self-esteem, attempts to restore control, and other pathways. These findings support the general claim that rejection can alter behavior; they do not establish the specific alignment-risk model proposed here. The spiral-of-silence literature examines how perceived opinion climates and fear of social isolation can affect willingness to express views. Studies have found that fear of isolation can predict self-censorship or reduced expression in some contexts, including online settings. Recent work also suggests that under some conditions people may publicly support a community's dominant opinion despite privately disagreeing. Stigma research describes identity threat and the social consequences of stigmatized identities. Work on courtesy stigma, also called stigma by association, shows that people can sometimes be socially evaluated because of their association with stigmatized others. Those effects are context-dependent and should not be treated as universal. The new proposal in this book is narrower: perceived alignment risk may be a useful way to describe the anticipated cost of moving partially toward another person's position or humanity when observers may interpret that movement as evidence about the actor's own identity, loyalty, rank, or membership. That construct is not established. It may overlap heavily with fear of isolation, impression management, social identity processes, conformity, rejection sensitivity, stigma by association, or relational uncertainty. The scientific question is whether it adds anything measurable. If not, the phrase belongs in essays rather than models. That is an acceptable outcome.

SELECTED SOURCES A short research trail for the concepts this book borrows from and tries to connect.

Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for intimate relationships. Journal of Personality and Social Psychology, 70(6), 1327–1343. doi:10.1037/0022-3514.70.6.1327. Romero-Canyas, R., Downey, G., Berenson, K., Ayduk, O., & Kang, N. J. (2010). Rejection sensitivity and the rejection-hostility link in romantic relationships. Journal of Personality, 78(1), 119–148. doi:10.1111/j.1467- 6494.2009.00611.x. Leary, M. R., Twenge, J. M., & Quinlivan, E. (2006). Interpersonal rejection as a determinant of anger and aggression. Personality and Social Psychology Review, 10(2), 111–132. doi:10.1207/s15327957pspr1002_2. Neuwirth, K., Frederick, E., & Mayo, C. (2007). The spiral of silence and fear of isolation. Journal of Communication, 57(3), 450–468. doi:10.1111/j.1460- 2466.2007.00352.x. Shin, D.-I., Lim, Y.-W., & Kwahk, K.-Y. (2022). SNS users' opinion expression: Focusing on suppression effect in spiral of silence. Telematics and Informatics, 72, 101859. doi:10.1016/j.tele.2022.101859. Oz, M., Shahin, S., & Greeves, S. B. (2024). Platform affordances and spiral of silence: How perceived differences between Facebook and Twitter influence opinion expression online. Technology in Society, 76, 102431. doi:10.1016/j.techsoc.2023.102431. Major, B., & O'Brien, L. T. (2005). The social psychology of stigma. Annual Review of Psychology, 56, 393–421. doi:10.1146/annurev.psych.56.091103.070137. Sigelman, C. K., Howell, J. L., Cornell, D. P., Cutright, J. D., & Dewey, J. C. (1991). Courtesy stigma: The social implications of associating with a gay person. The Journal of Social Psychology, 131(1), 45–56. doi:10.1080/00224545.1991.9713823. These sources support adjacent mechanisms. They do not establish alignment risk as a distinct scientific variable.