There is something psychologically fascinating about the way people behave when somebody else’s reputation begins to collapse in public. A person says something offensive, behaves badly or finds themselves accused of wrongdoing. The story begins circulating online, criticism gathers momentum and, within hours, thousands of people who have never met the individual are discussing their character, their relationships, their employment and whether they deserve to have a future in their profession. Some are genuinely concerned about what has happened. Others are angry on behalf of those affected. But there is also a sizeable audience that appears to be enjoying the spectacle.
Working in crisis PR, I spend a considerable amount of time thinking about the difference between accountability and punishment. They are often treated as interchangeable, particularly in conversations about cancel culture, but psychologically they serve different purposes. Accountability is concerned with responsibility, consequences and, ideally, some form of correction or repair. Punishment can serve those purposes too, but it can also satisfy other human impulses, including anger, social belonging, moral superiority and the desire to see somebody brought down.
What interests me is not whether people should be criticised for behaving badly. Of course they should. Public scrutiny has exposed serious wrongdoing, challenged powerful institutions and given people opportunities to speak about experiences that might otherwise have remained hidden. The more interesting question is what happens psychologically when a group of people becomes convinced that somebody deserves punishment, and whether that conviction changes the standards of behaviour the group applies to itself.
Because one of the peculiar features of online outrage is that people who consider themselves compassionate, fair-minded and socially conscious can sometimes participate in behaviour that is extraordinarily cruel. They may ridicule someone’s appearance, contact their employer, circulate details about their family or celebrate the deterioration of their mental health, all while maintaining that they are acting in the interests of justice.
How does somebody reconcile those two positions? And at what point does holding another person accountable become a socially acceptable opportunity to humiliate them?
The psychological comfort of being on the right side
Human beings have a strong interest in understanding themselves as fundamentally decent people. Our moral identities influence how we interpret our own behaviour, and we are remarkably skilled at finding explanations that preserve a favourable view of ourselves.
Psychologists have spent decades studying moral disengagement, a concept developed by Albert Bandura to describe the processes through which people justify conduct that would ordinarily conflict with their moral standards. These mechanisms include minimising the consequences of harmful behaviour, displacing responsibility, blaming the person being harmed and presenting questionable actions as serving a worthwhile purpose.
The relevance to online shaming is difficult to ignore. When somebody has been publicly identified as having done something wrong, their perceived wrongdoing can become the justification for almost anything that follows. Insults are reframed as consequences. Humiliation becomes accountability. Attempts to damage their livelihood are presented as protecting other people.
The underlying psychological mechanism is particularly interesting because it does not necessarily require somebody to abandon their moral principles. Instead, they reinterpret the situation so that their behaviour appears consistent with those principles.
A person who would ordinarily condemn bullying might feel entirely comfortable joining a campaign of public ridicule if they believe the target is sufficiently objectionable. They have not necessarily decided that cruelty is acceptable in general. They have decided that this particular person falls outside the boundaries of who deserves their sympathy.
This is where moral conviction can become psychologically dangerous. The stronger somebody’s belief that they are acting for a good cause, the less inclined they may be to examine the methods they are using to pursue it.
None of this means that strong moral convictions inevitably produce cruelty. Moral conviction can motivate extraordinary courage, compassion and resistance to injustice. But believing that your objective is morally justified does not automatically make every action taken in pursuit of it justified too.
And the internet is not particularly good at encouraging that distinction.
Why the crowd changes our behaviour
One of the most important differences between an argument involving two people and a public controversy involving thousands is the distribution of responsibility.
In an ordinary confrontation, we are relatively conscious of our own participation. We know what we have said, whom we have said it to and how the other person has responded. Online, those relationships become much less immediate.
Psychologists have studied this through concepts including deindividuation, social identity and the diffusion of responsibility. Although these theories describe different processes, they help explain why people may behave differently when acting as part of a group.
Deindividuation is often described as a reduction in individual self-awareness or accountability in group settings. More recent social identity research complicates the idea that people simply lose themselves in crowds. In many circumstances, people become more responsive to the norms of the group with which they identify.
That distinction matters. An online crowd is not necessarily a collection of people who have temporarily abandoned their values. It may be a collection of people whose behaviour is increasingly guided by what their particular community considers acceptable.
If the prevailing expectation is to express outrage, people may feel pressure to express outrage. If ridicule receives approval, ridicule can become normalised. If expressing sympathy for the person at the centre of the controversy risks attracting criticism, people may become reluctant to do so publicly.
The result can be an environment in which individuals behave in ways they might privately question, not because they have stopped thinking altogether, but because belonging to the group has become psychologically important.
There is also a striking imbalance between the experience of the participants and that of the person receiving their attention.
For somebody joining an online pile-on, writing a hostile comment may occupy thirty seconds of their afternoon. They may forget about it almost immediately. For the person at the centre, that comment arrives alongside hundreds or thousands of others, creating an experience that bears little resemblance to the individual act of criticism.
Each participant experiences their own contribution as relatively insignificant. The recipient experiences the accumulation.
This is one reason online punishment can become disproportionate without any single participant necessarily intending the overall result.
What the Stanford Prison Experiment can, and cannot, tell us
There is an interesting historical comparison here with the Stanford Prison Experiment, conducted by psychologist Philip Zimbardo in 1971. Volunteers were assigned roles as guards and prisoners in a simulated prison environment. Some of those playing guards subjected prisoners to degrading treatment, and the experiment was terminated after six days rather than the planned two weeks.
The study became famous for its apparent demonstration of how quickly ordinary people could adopt cruel behaviour when placed in positions of authority. However, subsequent investigations have raised substantial questions about its methodology, including the extent to which researchers encouraged particular behaviour. Its original conclusions should not be treated as established scientific fact.
What remains interesting is the broader question it raises about social roles and the rationalisation of harmful conduct. Under what circumstances do people begin to regard behaviour they would ordinarily condemn as acceptable, or even necessary, because of the role they believe they are performing?
The comparison with online shaming is not that social media users are equivalent to prison guards, nor that the experiment explains cancel culture. Rather, it invites us to consider how perceived authority and group expectations influence what people believe they are entitled to do to somebody else.
In an online controversy, authority may not come from an institution or an official position. It can emerge from the apparent consensus of thousands of people who agree that an individual deserves condemnation. Once that consensus becomes sufficiently visible, it can begin to feel like moral permission.
The uncomfortable question is whether people are evaluating the proportionality of their own behaviour or simply taking reassurance from the fact that so many others are doing the same thing.
The social rewards of outrage
Public outrage is not simply an emotional response. On social media, it is also a form of communication that can produce measurable social rewards.
Research published in Proceedings of the National Academy of Sciences in 2017 examined the relationship between moral-emotional language and the spread of messages on social media. The researchers found that moral-emotional language was associated with greater diffusion within the political networks they studied.
A separate study published in Science Advances in 2021 examined how social feedback can reinforce expressions of moral outrage online. It found evidence that positive feedback influenced subsequent expressions of outrage, with the effects depending on the surrounding social environment.
These findings do not establish that everyone who expresses anger online is seeking approval, nor do they mean outrage is inherently performative. They do, however, suggest that the architecture of social media can influence how moral emotions are expressed and repeated.
Consider what receives attention during a major public controversy. A carefully qualified argument acknowledging uncertainty is unlikely to produce the same immediate emotional response as an emphatic declaration that somebody is terrible and deserves to lose everything.
Nuance requires the audience to think. Outrage tells the audience how to feel.
Platforms have not invented our appetite for moral judgement, but they have created systems in which expressions of judgement can be rapidly distributed, measured and rewarded. Likes, shares, comments and follower growth make social approval unusually visible.
That visibility matters psychologically. We are social creatures, sensitive to signals about what other people approve of and what behaviour will earn us acceptance.
An individual might begin by expressing a sincere objection to something they have witnessed. If that expression receives substantial approval, they may become more inclined to express similar sentiments in future. Over time, the boundary between communicating a genuine belief and performing an identity that an audience rewards can become difficult to distinguish.
This is particularly apparent when a controversy develops its own community of spectators. People begin producing reaction videos, compiling timelines, interpreting apologies and searching for additional evidence of wrongdoing. The original incident becomes the foundation for an expanding collection of content.
The scandal acquires a life beyond the event that caused it.
And at some point, the public may become more interested in the individual’s downfall than in the original issue.
Why we enjoy watching people fall
There is a German word, Schadenfreude, which describes pleasure derived from another person’s misfortune. It is an uncomfortable emotion to acknowledge because it conflicts with how most of us would prefer to understand ourselves.
Yet research into Schadenfreude suggests that it can be influenced by social comparison, perceived deservingness and the status of the person experiencing misfortune.
The concept of deservingness is particularly relevant to public scandals. People may experience another person’s suffering differently depending on whether they believe the individual has brought it upon themselves.
A successful celebrity experiencing an embarrassing professional failure might provoke amusement where an ordinary person experiencing the same event would attract sympathy. The celebrity’s wealth, status or perceived arrogance may make their misfortune seem less troubling, or even satisfying.
There is also an element of social comparison. Watching somebody who appears more successful, attractive, influential or privileged lose their status can temporarily alter how we perceive our own position.
This does not mean that every criticism of a successful person is motivated by envy. It means that our emotional responses to another person’s downfall may be influenced by considerations that have little to do with the original wrongdoing.
I find this particularly interesting in cases where the public response seems to become more enthusiastic as the consequences become more severe.
An apology is issued, but the audience is dissatisfied. A professional relationship ends, but the criticism continues. A person loses their job, yet people continue searching for additional evidence that might justify further consequences.
If the objective was accountability, we might reasonably ask what outcome would satisfy it.
But if part of the attraction is the spectacle of somebody losing status, there may be no particularly satisfying endpoint. Each new consequence becomes another development in a story people have become emotionally invested in following.
The punishment becomes entertainment.
The problem with treating people as characters
Another feature of online controversy is the speed with which complicated human beings are reduced to relatively simple moral identities.
A person becomes the villain of a story, and once that identity has been established, new information is often interpreted through it.
This connects with confirmation bias, our tendency to seek out, notice or interpret information in ways that support existing beliefs. It also relates to research on impression formation and the difficulty of revising strongly established judgements.
Once somebody has acquired a negative public identity, ordinary behaviour can begin to look suspicious. Silence becomes evidence of guilt. Speaking becomes evidence of manipulation. An apology is described as insincere. A refusal to apologise is presented as arrogance.
The individual is placed in a position where almost every available response can be interpreted as confirmation of the original judgement.
This is one of the more difficult dynamics to manage in crisis communications, because the problem is no longer simply what the person has done. It is what the audience has come to believe about who that person is.
And those are very different reputational problems.
It is possible to correct a factual misunderstanding. It is considerably harder to change an interpretation of someone’s character, particularly when an audience has invested time and emotion in that interpretation.
The distinction also helps explain why apologies sometimes fail even when they contain the elements people originally demanded.
An apology can acknowledge responsibility, express remorse and identify steps towards change. But if the audience has already concluded that the individual is fundamentally dishonest or malicious, those same words may be interpreted as another example of their dishonesty.
The response becomes almost impossible to evaluate independently of the existing reputation.
There is a further complication in the way social media encourages us to experience real events through the conventions of entertainment. We discuss public figures as protagonists and antagonists, speculate about their motivations and anticipate the next dramatic development.
Yet unlike fictional characters, the people involved cannot simply disappear when the audience loses interest.
Their families, colleagues and professional relationships may all be affected by a story over which they have limited control.
When does the punishment end?
Perhaps the most difficult issue surrounding cancel culture is that there is rarely an agreed understanding of what proportionate accountability should involve.
In formal systems of justice, at least in principle, there are procedures for examining evidence, assessing responsibility and determining consequences. Those systems are imperfect, and public criticism is not a criminal trial, but the comparison highlights something important.
Online crowds have no shared process for deciding what constitutes sufficient evidence, what consequences are appropriate or when somebody has done enough to make amends.
There is also no single authority capable of declaring that a punishment has ended.
An individual might acknowledge wrongdoing, apologise, compensate somebody who has been harmed, change their behaviour and experience substantial professional consequences. Yet a new audience can discover the original controversy months or years later and begin the process again.
This is partly a consequence of the internet’s unusual relationship with time.
A comment made fifteen years ago can be circulated today without its original context, creating the impression of an immediate controversy. A person may have changed considerably since the event, but the material remains accessible in a form that preserves the original moment.
Digital permanence can be valuable when it allows important evidence to remain available. It can also create difficulties when the public record preserves somebody’s worst behaviour far more effectively than it preserves their subsequent development.
We tend to accept, at least in theory, that people are capable of learning, changing and becoming better versions of themselves. Yet public reputations can be remarkably resistant to that possibility.
There is a tension here between two beliefs that many people hold simultaneously: that individuals should take responsibility for wrongdoing, and that individuals should have opportunities to change.
If we believe in both, then we have to consider what meaningful rehabilitation looks like in a culture where reputational punishment can be indefinitely renewed.
Otherwise, accountability risks becoming a permanent identity rather than a process.
What this means for crisis PR
Working in crisis PR has made me particularly interested in the difference between the public’s perception of a scandal and the experience of the people directly involved.
From the outside, a controversy can appear relatively straightforward. Somebody has done something wrong, people are angry and a response is expected.
From the inside, there are often much more complicated questions. What is established fact? What remains an allegation? Who has been harmed? What responsibility should be accepted? What can be repaired? Which consequences are reasonable, and which are being demanded primarily because public anger has gathered momentum?
These are not questions that can always be answered by producing a carefully worded statement.
In fact, one of the mistakes organisations and individuals make during a reputational crisis is assuming that communication alone can resolve a problem that fundamentally concerns conduct.
If somebody has behaved badly, the answer is not necessarily better messaging. It may be acknowledging what happened, changing behaviour, addressing the harm and accepting that public trust will take time to rebuild.
But there is another side to this work that receives less attention.
A crisis adviser also has to recognise when the public response has become disconnected from the original incident, when allegations are being treated as established facts or when demands for accountability have developed into a campaign of humiliation.
That does not require defending the indefensible. It requires maintaining a distinction between what somebody has done and what other people believe they are entitled to do in response.
I also think the instinct to respond immediately can be counterproductive. When thousands of people are participating in an emotionally charged conversation, there is a temptation to believe that every accusation requires an answer and every hostile comment needs to be corrected.
But an online crowd is not a single audience with a shared objective. Some people want information. Some want an apology. Some want consequences. Others have already reached a conclusion and have little interest in anything that might complicate it.
Treating all those people as though they are asking the same question is rarely useful.
A more thoughtful approach requires understanding the psychological dynamics of the controversy, separating legitimate concerns from speculation and identifying the people whose trust genuinely needs to be restored.
It also means recognising that public approval and ethical behaviour are not always the same thing.
An organisation can make an unpopular decision for good reasons, just as it can receive public support for conduct that deserves criticism. The number of people applauding or condemning an action does not, by itself, establish whether that action is right.
That principle should apply equally to the person at the centre of a scandal and to the people participating in their condemnation.
The uncomfortable question we should be asking
I don’t think cancel culture can be neatly divided into people seeking justice and people behaving cruelly. Human motivation is rarely that straightforward.
A person can sincerely care about an issue and still enjoy the social rewards of expressing outrage. They can believe somebody deserves consequences while also experiencing satisfaction at watching their status decline. They can participate in behaviour that causes harm without fully appreciating the cumulative effect of their contribution.
And somebody facing public criticism can be genuinely responsible for wrongdoing while also becoming the target of treatment that is disproportionate or unnecessarily degrading.
These possibilities are not mutually exclusive, which is precisely why the subject is more psychologically interesting than the familiar argument about whether cancel culture is good or bad.
What concerns me is the ease with which moral certainty can discourage self-examination. Once we have decided that somebody deserves condemnation, we may become considerably less interested in scrutinising our own behaviour towards them.
The internet gives us extraordinary opportunities to expose wrongdoing, challenge authority and demand accountability. But it also gives us the ability to participate in another person’s humiliation at a distance, surrounded by people who reassure us that our participation is justified.
And perhaps that is the distinction worth paying attention to.
Accountability asks what somebody has done, what harm has resulted and what should happen next. Public punishment can become preoccupied with how much suffering we believe somebody deserves.
The first requires judgement. The second can be sustained by emotion alone.
I suspect most of us would like to believe that, faced with somebody else’s public disgrace, we would respond thoughtfully, proportionately and with some consideration for the person behind the headlines.
But the psychology of group behaviour suggests that our conduct is influenced by social expectations, emotional rewards and the behaviour of those around us, often more than we would care to admit.
So perhaps the most revealing question about cancel culture is not why people sometimes behave terribly online.
It is why so many of us find it easier to recognise cruelty when we disagree with the people committing it.
And why, when we believe somebody deserves punishment, we can become remarkably reluctant to ask whether we are behaving any better ourselves.