There is a particular kind of confidence that can look magnetic from a distance. A person may seem strong, protective, decisive, even intimidating in a way that feels attractive. But strength and aggression are not the same thing.

A confident partner does not need to frighten someone to feel powerful. A protective partner does not need to control another person's choices. And a person who loves deeply does not prove that love through threats, physical force, or relentless pursuit after a breakup.

This distinction matters because intimate partner violence rarely becomes easier to understand when we focus only on how someone looks. It becomes clearer when we look at patterns of behavior.

You Cannot Recognize an Abusive Partner by Appearance

There is no reliable "look" of a person who may become abusive.

Clothing style, tattoos, athletic build, short hair, posture, facial expression, tidiness, or a preference for personal space cannot tell us whether someone will hurt a partner. A person may look tough and be emotionally safe. Another may appear gentle, charming, and attentive while becoming increasingly controlling behind closed doors.

What matters is behavior.

According to the CDC, intimate partner violence can include physical violence, sexual violence, stalking, and psychological aggression intended to emotionally harm or control a partner.

That means the warning signs worth noticing are not broad shoulders or a serious expression. They are behaviors such as intimidation, threats, humiliation, coercive control, physical aggression, monitoring, and repeated violations of boundaries.

Sometimes these behaviors appear gradually. A person may initially seem unusually attentive or protective. Over time, however, "Where are you?" becomes constant monitoring. Concern becomes jealousy. Jealousy becomes accusations. Arguments become intimidation. And intimidation may eventually become physical violence.

The important question is not whether someone appears "brutal." It is whether they respect another person's autonomy when they are angry, jealous, disappointed, or afraid of losing the relationship.

Why Do Some People Return to Partners Who Hurt Them?

One of the most painful situations is watching someone return to a relationship in which they have already been harmed.

From the outside, the answer can seem obvious: Why go back?

From inside the relationship, it may be far more complicated.

A person can miss someone who hurt them. Love, fear, attachment, hope, shame, financial dependence, isolation, promises of change, and memories of good periods in the relationship can exist at the same time.

So it is misleading to say that people consciously choose abusive partners because they want to suffer.

Research does show that some people experience violence across more than one relationship, and childhood abuse and several other factors have been associated with a higher risk of later revictimization. But researchers describe revictimization as a complex phenomenon—not evidence that survivors secretly want abuse or deliberately search for violent partners.

This distinction is essential.

Studying why someone may repeatedly enter unsafe relationships can help with prevention and recovery. Turning that research into blame—"you chose this because you wanted it"—does the opposite.

The responsibility for violence remains with the person who uses violence.

When "I Can't Let You Go" Becomes a Warning Sign

Breakups can reveal behavior that was easier to overlook during the relationship.

A former partner may call repeatedly, send message after message, appear outside a home or workplace, monitor social media, leave unwanted gifts, demand another conversation, threaten, plead, or alternate between affection and anger.

Persistence is sometimes romanticized. But there is a point at which persistence stops being romance.

The CDC defines stalking as a pattern of unwanted harassing or threatening behaviors that causes fear or safety concerns. Examples can include:

• Repeatedly appearing at someone's home or workplace
• Following them
• Monitoring them through technology
• Tracking their location
• Repeatedly contacting them by phone, text, email, or social media

A person saying, "I love you too much to leave you alone," does not make repeated unwanted contact healthy.

Love requires the ability to hear "no."

When someone cannot accept the end of a relationship and begins monitoring, threatening, following, or repeatedly appearing without permission, the behavior deserves to be taken seriously.

The Strongest Predictor Is Often the Pattern We Have Already Seen

People understandably want certainty.

After an assault, someone may ask:

Will it happen again?

No clinician can predict an individual's future behavior with perfect accuracy. One violent incident also does not tell us everything about what will happen next.

But previous violence should never be dismissed as meaningless.

Intimate partner violence can occur as a single episode, but it can also become chronic and severe over time. Research on revictimization has found that recent physical violence and greater severity of violence are among the factors associated with further victimization.

Promises matter less than sustained behavior.

An apology after violence may be sincere in the moment. Tears may be sincere. Regret may be sincere. But regret and behavioral change are not the same thing.

What matters is whether the person accepts responsibility without blaming the partner, stops threatening or controlling behavior, respects boundaries, and consistently changes over time.

Is Abuse Caused by a Personality Disorder?

This is where psychological discussions can become dangerously oversimplified.

It would be incorrect to look at an aggressive partner and automatically assign a personality disorder.

Current U.S. psychiatric practice uses standardized diagnostic criteria, and personality disorders require careful professional assessment. They cannot be identified from clothing, facial expressions, jealousy, a short temper, or one relationship conflict. The American Psychiatric Association's DSM-5-TR is the primary diagnostic classification used in U.S. mental health care.

There is research showing associations between personality pathology and intimate partner violence. A large meta-analysis found that personality disorders as a group were associated with IPV perpetration, with particularly robust associations involving antisocial and borderline personality disorders. But an association is not destiny, and having a personality disorder does not mean that someone will abuse a partner.

The reverse assumption is equally problematic: an abusive person does not need to have a mental disorder for the abuse to be real.

Sometimes violence is better understood by looking directly at what the person does rather than searching for a psychiatric label.

Anger Matters, but Anger Is Not the Whole Explanation

Some abusive partners describe their behavior as losing control:

"I just exploded."

Anger and hostility are indeed associated with intimate partner violence. A meta-analysis involving dozens of studies found a moderate relationship between anger-related characteristics and IPV perpetration.

But this does not mean that domestic violence is simply an anger-management problem.

Someone who becomes angry is still responsible for what they do with that anger. Millions of people experience intense anger without hitting, threatening, stalking, or terrorizing their partners.

Exercise, sports, emotional regulation skills, psychotherapy, substance-use treatment when appropriate, and other interventions may help some people manage aggressive impulses. But sending someone to the gym is not a treatment for abusive behavior.

Meaningful change requires more than finding an outlet for anger. It requires recognizing the behavior, accepting responsibility for it, changing beliefs and habits that support abuse, and maintaining that change.

Research on interventions for people who perpetrate IPV suggests that treatment can help in some cases, although results vary considerably and there is no simple treatment that guarantees a person will stop being violent.

Do Not Confuse Fear With Love

Perhaps the most useful distinction is also the simplest.

A relationship can include conflict without becoming abusive. People can argue, become irritated, need space, make mistakes, and still remain respectful.

But fear changes the picture.

If one partner must constantly calculate what to say, what to wear, whom to speak with, where to go, or how to behave so that the other person does not explode, the issue is no longer ordinary relationship conflict.

If someone has already used physical force, threats, intimidation, surveillance, or stalking, the most important question is not:

"What personality type is this?"

It is:

"What is this person actually doing, and am I safe around this behavior?"

A powerful appearance can be attractive. Confidence can be attractive. Even a certain toughness can be attractive.

But genuine strength is compatible with restraint.

It respects boundaries.

And it does not require another person to be afraid.

References

• Centers for Disease Control and Prevention. (2026). About Intimate Partner Violence. National Center for Injury Prevention and Control.
  Provides current U.S. definitions of intimate partner violence, including physical violence, sexual violence, stalking, and psychological aggression. It also explains that IPV can range from a single incident to chronic and severe patterns over time.
• Centers for Disease Control and Prevention. (2025). Stalking. National Center for Injury Prevention and Control.
  Describes stalking as repeated unwanted or threatening behavior and lists examples such as following someone, appearing at their home or workplace, repeated unwanted communication, and technology-based monitoring.
• Bellot, A., Muñoz-Rivas, M. J., Botella, J., & Montorio, I. (2024). Factors Associated with Revictimization in Intimate Partner Violence: A Systematic Review and Meta-Analysis. Behavioral Sciences, 14(2), 103. doi: 10.3390/bs14020103.
  A systematic review and meta-analysis of 22 studies examining factors associated with repeated intimate partner violence against women. The authors emphasize that revictimization is complex and identify childhood abuse, recent violence, severity of violence, substance use, and PTSD symptoms among relevant factors. This journal uses article numbers rather than conventional page ranges.
• Collison, K. L., & Lynam, D. R. (2021). Personality disorders as predictors of intimate partner violence: A meta-analysis. Clinical Psychology Review, 88, 102047. doi: 10.1016/j.cpr.2021.102047.
  Analyzed 163 studies and examined associations between personality disorders and physical, psychological, and sexual IPV. The findings support an association between personality pathology and IPV while also showing why a diagnosis should not be treated as a simple predictor of individual behavior. The journal uses article number 102047 rather than a page range.
• Birkley, E. L., & Eckhardt, C. I. (2015). Anger, hostility, internalizing negative emotions, and intimate partner violence perpetration: A meta-analytic review. Clinical Psychology Review, 37, 40–56. doi: 10.1016/j.cpr.2015.01.002.
  Reviews the relationship between anger, hostility, negative emotional states, and IPV perpetration. The findings support an association with aggression while cautioning against reducing IPV to an anger-control problem alone.
• American Psychiatric Association. (2022). Diagnostic and Statistical Manual of Mental Disorders (5th ed., text rev.; DSM-5-TR). American Psychiatric Association Publishing. doi: 10.1176/appi.books.9780890425787.
  The principal contemporary diagnostic classification used by U.S. mental health professionals. It provides standardized criteria for personality disorders and helps distinguish formal diagnoses from informal personality "types" or labels.
• Karakurt, G., Koç, E., Çetinsaya, E. E., Ayluçtarhan, Z., & Bolen, S. (2019). Meta-analysis and systematic review for the treatment of perpetrators of intimate partner violence. Neuroscience & Biobehavioral Reviews, 105, 220–230. doi: 10.1016/j.neubiorev.2019.08.006.
  Reviews treatment approaches for IPV perpetrators and shows that intervention outcomes vary, supporting a more cautious position than assuming either that violent behavior is impossible to change or that a simple intervention will reliably stop it.