Part I
Corrosive Fragmentation
Societies do not begin as unified communities that propaganda suddenly breaks apart. They already contain competing interests, political disagreements, inequalities, historical grievances, cultural differences and conflicting ideas about what the future should look like. Governments fail. Institutions make mistakes. Political parties lie. Communities sometimes have good reasons to distrust one another. People can be angry because something unjust really happened to them. None of this means propaganda has succeeded, and none of it is something a healthy society should expect to eliminate. Democracy depends upon disagreement. Protest can expose abuse, distrust can force institutions to become more accountable, and political competition can produce change. If we want to understand what makes propaganda corrosive, simply pointing to division is not enough.
What matters here is what happens to divisions that already exist. A grievance can remain one disagreement among many, something people argue about, investigate, negotiate or eventually resolve. It can also begin to carry much more weight than the evidence justifies. Political actors can keep returning to it, connect it to identity, simplify its causes and use it to explain more and more unrelated problems.¹ One crime can become evidence about an entire population. One corrupt official can become proof that every institution is corrupt. A genuine historical injustice can become a permanent explanation for the motives of people living generations later. None of this requires the original event to be invented. Manipulation can work perfectly well with facts and with grievances that deserve attention. The change happens in what those facts and grievances are made to represent.
Looking only for false information can miss part of the problem. A photograph may be genuine. A quotation may be accurate. A crime may really have happened. An institution may genuinely have failed. Yet truthful pieces of information can still be arranged into a deeply misleading picture. Ten real examples drawn from a population of millions can be repeated until they begin to feel typical. A genuine case of corruption can be presented again and again until the audience is encouraged to conclude that correction is impossible. A real event can be stripped of context and turned into proof of something much larger than the evidence allows.² The information is not necessarily false, but the conclusion built around it can still be corrosive.
The theory proposed here is that corrosive propaganda works cumulatively. Repeated attempts to exploit existing social fractures do not automatically produce fragmentation, but they can make it more likely by linking grievances more tightly to group identity, narrowing what counts as credible evidence and weakening the standards through which people previously judged institutions and political opponents. When these patterns become sufficiently entrenched, disagreement can begin to divide the shared rules through which disagreement was previously managed. That is what I mean by corrosive fragmentation.
There is nothing inevitable about this sequence. Propaganda is not the cause of every fracture, fragmentation can develop without propaganda, and a fragmented society does not necessarily descend into violence or cruelty. Repeated manipulative exploitation can, however, make certain social conditions more likely. It can help turn disagreement into enclosure, sustained enclosure into corrosive fragmentation, and severe fragmentation into an environment in which political and moral reciprocity becomes harder to sustain.
The deeper problem lies in what repeated manipulation can begin doing to the way people form, challenge and revise beliefs together.
From disagreement to enclosure
Imagine two political camps that strongly dislike each other. They disagree about immigration, taxation, national identity, war, religion or constitutional reform. Their newspapers describe events differently. Elections are bitter. Demonstrations fill the streets. People say terrible things about the other side. Yet they may still accept enough common rules for the conflict to continue without swallowing everything around it. A court can make a bad ruling without every court being presumed fraudulent. An election can be lost without the loss automatically proving that elections are meaningless. Evidence from outside one's own political camp can occasionally be accepted. Opponents can be considered misguided, selfish or even dangerous on particular questions while still retaining the right to organize, speak, vote and sometimes win.
Such a society may be highly polarized, but something important still holds it together. The disagreement is taking place inside at least some shared expectations about how disagreement works. The deeper problem begins when people stop disagreeing only about politics and begin dividing over the standards used to decide what is true, which institutions can ever be legitimate and whether opponents deserve the same political standing they expect for themselves. At that point the dispute is no longer confined to policy. The mechanisms for managing disagreement are becoming part of the disagreement itself.
Evidence then becomes our evidence and their evidence. A newspaper associated with the other side is dismissed before the article is read. A court is respected when it rules in our favour and corrupt when it does not. An election is fair when our side wins and suspicious when the result goes the other way. Political opponents are no longer simply people pursuing the wrong policies. Their participation itself begins to look illegitimate. The conflict is increasingly about whether the people on the other side are entitled to take part in deciding at all.
Fragmentation should not become another word for political conflict. A loud, argumentative society can still be healthier than a quiet society in which disagreement has been suppressed. The concern is not that people differ. The concern is that political difference begins closing off the paths through which people can still test claims, correct mistakes and recognize one another as participants in a shared society.
I use enclosure for this narrowing process. Enclosure does not mean people never encounter opposing views. They may see them constantly. The problem is that the conclusion about the source comes before the information itself. Someone from the opposing camp can no longer surprise us because we already know what kind of person they must be. A crime committed by one member of a disliked group becomes evidence about the group, while contradictory examples are dismissed as exceptions. An institutional failure no longer raises the question of what went wrong and whether it can be fixed. It confirms a larger belief that the entire system is controlled, fraudulent or beyond repair.
Political identity can then begin absorbing other parts of social life.³ People who might otherwise meet as neighbours, colleagues, parents, musicians, football supporters, artists or members of the same profession increasingly encounter one another first as political categories. The more complete that category becomes, the easier it is to interpret everything the other person does as further evidence of what we already believe about them.
A person stops being someone who holds a political view and becomes an example of what people like that are supposedly like.
The cumulative effect
A single inflammatory claim rarely transforms a society. What can become powerful is the accumulation of the same interpretive move across hundreds or thousands of messages. The individual examples change, but the underlying lesson remains familiar. Their crimes reveal who they are. Their institutions cannot be trusted. Their victories are illegitimate. Their suffering matters less. Their explanations are lies. Repeated often enough and reinforced by enough different voices, these claims can stop feeling like arguments and begin functioning as background assumptions through which new events are interpreted.
This does not require all of the people repeating them to be part of a coordinated operation. Some messages may be deliberately introduced. Others may arise domestically. Some people repeat them for political advantage, others because outrage brings attention or revenue, and many because they sincerely believe what they are saying. Repetition can produce a cumulative environment even when the individual acts of repetition have very different causes.
Manipulative communication is particularly suited to this because it does not need to invent a complete worldview. It can simply make one interpretation more emotionally satisfying, more socially rewarding and easier to repeat. Again and again, in different forms, the message becomes: we are suffering because they are doing this to us. The identity of they can change. Migrants, elites, minorities, bureaucrats, journalists, political opponents, foreigners, activists, corporations or intellectuals can all be placed in that role. Sometimes criticism of those groups is entirely justified. The corrosive move happens when evidence about particular actions becomes a judgment about the category itself, so that membership increasingly determines guilt, motive and legitimacy.
Repetition also changes the burden of proof. A claim heard once may invite skepticism. A claim heard dozens of times from apparently different people may begin to feel established even when every repetition ultimately traces back to the same weak evidence or the same selective framing. Familiarity can take the place of verification.⁴ An interpretation becomes something “everybody knows,” and once it reaches that stage new events are no longer evaluated independently. They are fitted into a story that already exists. That is one reason the cumulative aspect of corrosive propaganda deserves more attention than the individual message.
The social effect may come less from persuading people of one falsehood than from repeatedly teaching them how to interpret the next event.
Propaganda after the propagandist
The familiar picture of propaganda has a propagandist on one side and an audience on the other. One produces the message and the other receives it. Modern information environments are much messier. A narrative introduced deliberately by one actor can be picked up by domestic politicians because it helps them defeat an opponent. A media outlet can repeat it because it fits an existing interpretation. An influencer can amplify it because anger brings attention. Ordinary people can repeat it because they believe they are warning friends about a genuine danger. Communities then add their own examples, jokes, videos, anecdotes and explanations until the original source may hardly matter anymore.⁵
A message can begin strategically and survive for reasons completely unrelated to the intentions of whoever introduced it.⁶ It changes as it travels. New examples are added, old examples disappear, local grievances become attached to it and the surrounding society gradually takes ownership of the narrative. Eventually domestic participants may be producing far more material than the original propagandist ever could.
A hostile operation does not need to control a narrative forever if it can help establish an interpretation that others find useful enough to reproduce. This is what I mean by propaganda after the propagandist. The strategic intervention can disappear while the social habit remains.
Attribution is still important. Journalists, intelligence services, researchers and law-enforcement bodies should investigate who funds covert campaigns, operates coordinated networks or hides sponsorship. But attribution cannot provide a complete civic defence. Ordinary people cannot wait for an intelligence assessment before deciding what to make of every claim they encounter, and identifying a hostile source does not automatically prove that a claim is false. An adversary can say something true. A trusted domestic actor can manipulate. A political movement we support can use methods we would immediately condemn if the same methods were used against us.
The method matters independently of the cause. Is one example being turned into a claim about an entire group? Is uncertainty being presented as certainty? Can contradictory evidence change the argument, or does every contradiction somehow become further proof? Does criticism of one institution leave open the possibility of correction, or does every failure prove that everything is illegitimate? Are political rights being applied consistently, or does belonging to the wrong camp increasingly mean that ordinary protections no longer count?
One question cuts through much of this: Would we accept the same rule if it were being used by people we deeply distrusted? If electoral rules are legitimate only when our side wins, then the principle is not really electoral legitimacy. If censorship is intolerable when directed at us but necessary when directed at them, then the principle is not free expression. If collective punishment is barbaric when our group suffers it but deserved when another group does, then the standard has changed.
What once looked like a political principle has become a tribal one.
From fragmentation to reciprocity
A fragmented society does not inevitably become violent. Most political conflicts do not end in persecution, and it would be irresponsible to describe disagreement as if it naturally marches toward atrocity. The danger is more gradual. Severe fragmentation can weaken some of the restraints that normally limit what people are willing to justify against those they dislike.
One of those restraints is reciprocity. Reciprocity does not mean liking our opponents or pretending their ideas are harmless. It means that some rights, protections and standards continue to apply even when we strongly oppose them. We may want their party defeated, their policies reversed and their arguments rejected while still accepting that rules used against them should be rules we could live under if political power changed hands.
As enclosure deepens, this becomes harder. The out-group is increasingly treated as a category rather than as a collection of individuals.⁷ One member commits an offence and the offence becomes evidence about everyone associated with the group. Another behaves responsibly and is dismissed as an exception, a fraud or someone hiding their real intentions. The category becomes difficult to disprove because almost any evidence can be made to fit it.
Under those conditions, scapegoating becomes especially attractive. Most serious social problems have many causes. Economic insecurity can involve housing, technology, trade, education, taxation, demographics and government policy. Institutional failure can result from corruption, incompetence, bad incentives, underfunding or conflicting responsibilities. Crime, migration, war and cultural change are similarly complicated. Complexity rarely provides the emotional satisfaction of a simple explanation. A scapegoat does. It says that many different problems can be understood through one story: we are suffering because they are doing this to us.
Once that story becomes useful, the scapegoat can perform several functions at once. It turns uncertainty into certainty, gives unrelated grievances a common cause and provides a target for anger. It can strengthen belonging within the group while protecting leaders or institutions from scrutiny by moving responsibility elsewhere. A complicated failure no longer requires an investigation into policy, incentives or incompetence. Someone has already been identified as the cause.
The danger is not limited to explicit hatred or physical violence. Cruelty can become normal through smaller changes in moral judgment. People laugh at humiliation they would find appalling if it happened to someone they identified with. They tolerate collective punishment because the victims are assumed to share responsibility. They become less interested in suffering once the people suffering belong to the wrong category. Rights begin to look conditional. Exclusion is described as protection. Harassment becomes accountability. Treatment that would be considered unacceptable in ordinary circumstances is redescribed as necessary self-defence.
Repeated exposure to this reasoning can matter because each instance helps make the next one easier to accept. A punishment once regarded as excessive can become normal. A joke that once crossed a line can become ordinary. A political opponent once regarded as wrong can become dangerous, then illegitimate, and eventually undeserving of protections demanded for one's own side. This progression is not automatic, and societies can interrupt it. Severe fragmentation can still lower social barriers against treatment that would previously have been regarded as unacceptable.
People participating in this process may not think of themselves as cruel. They may believe they are protecting their community, defending justice or responding to a threat. The important change has taken place in who is still considered entitled to ordinary moral restraint. Corrosive fragmentation can therefore have a human consequence more serious than political gridlock. Under severe conditions it can make cruelty toward designated out-groups easier to excuse, celebrate or simply ignore.
Societies still have legitimate reasons to restrict violence, intimidation, corruption and attempts to strip others of their rights. Harsh measures are not automatically cruel, and exclusion is not always unjustified. The distinction should depend upon conduct, evidence and proportionality rather than category. The question is whether people are being judged for what they have done or whether membership itself has begun supplying the verdict.
The most dangerous consequence of fragmentation may be the gradual withdrawal of political and moral reciprocity.
What resistance would have to preserve
If the problem were simply false information, the response could be imagined mainly as better information. Fact-checkers correct claims. Journalists investigate. Schools teach media literacy. Platforms identify coordinated behaviour. Governments expose foreign influence. All of this remains necessary, but correcting false claims is not enough if people have also stopped listening outside their own group, assume that every institution is irredeemably corrupt or believe political opponents deserve fewer rights.
A person can become very good at spotting fake photographs while remaining completely enclosed inside a political community. Someone can learn every common manipulation technique and become excellent at identifying it when opponents use it while remaining blind to the same behaviour from allies. A citizen can know which sources are generally reliable and still believe that people on the other side should not have the same political standing. Information skills matter, but they operate inside a larger social environment.
Resistance requires some ordinary but demanding capacities. People need to be able to judge claims without responding to uncertainty by believing nothing. Better evidence should be capable of increasing confidence, contradictory evidence should be capable of reducing it, and changing one's mind should not have to feel like betraying one's identity. People also need some usable paths outside their own social and informational circles, not because everyone should be forced into political dialogue but because political identity should not become so complete that every relationship reinforces the same division.⁸
Political opposition has to leave room for equal standing as well. Democracy does not require anyone to respect every opinion, but it does require people to live with the fact that others they strongly disagree with are sometimes entitled to speak, organize, vote and win.⁹ Institutions require a similar balance. Blind trust is not resilience, and criticism can be entirely justified, but accountability depends upon the possibility that one institutional failure is not every institutional failure and that correction can restore some credibility.
People also need to believe legitimate action can still achieve something. If voting, organizing, protesting, volunteering, association and local action all feel meaningless, more closed forms of political participation can become attractive. None of this means citizens should simply accept existing institutions. Sometimes those institutions need to be challenged or changed. The important thing is that correction remains imaginable and participation does not feel entirely pointless.
These capacities cannot solve every problem that makes a society vulnerable. Economic insecurity, discrimination, corruption and poor government are not communication problems that disappear if citizens become better at evaluating information. Sometimes the proper response to distrust is institutional reform. Sometimes anger is justified. Sometimes the grievance itself needs to be addressed.
When people can still judge evidence without distrusting everything, encounter others outside their own political world, criticize institutions without declaring all institutions worthless, participate meaningfully and oppose one another without withdrawing basic political standing, repeated manipulation has fewer opportunities to turn ordinary fractures into closed tribal worlds.
The next question is whether the civilian side is organized in a way capable of preserving those conditions over time.
Part II
The Civilian Asymmetry
Corrosive influence has one structural advantage that is easy to underestimate: persistence. A serious influence campaign does not need to persuade everybody immediately. It can repeat, adjust, wait and return. State-backed operations can sometimes draw on long-term funding, personnel, media infrastructure, technology and institutional support.¹⁰ Domestic political actors may reinforce similar messages because they mobilize supporters. Commercial media and influencers may repeat them because outrage produces attention. Ordinary users then contribute further repetition because they believe what they are sharing.
The resulting pressure can be continuous without being centrally controlled. A narrative introduced from outside may combine with domestic grievances, partisan incentives and commercial algorithms until it is difficult to separate one source from another. The important feature is not perfect coordination. It is that the same corrosive interpretation can continue arriving from many directions over a long period.
Civilian responses often operate differently. A fact-check appears after a claim has travelled. An investigative article exposes an operation. A research institute publishes a report. A media-literacy course reaches a classroom or a group of volunteers. An NGO receives a grant for a limited project. Civil society organizations respond when they have staff, money and opportunity. Individual citizens try to correct misinformation in their spare time. Many of these efforts are useful, but useful does not necessarily mean continuous.
This creates an asymmetry that is larger than funding. The corrosive environment may be experienced every day, while the response arrives as an intervention. One side can keep applying pressure to social fractures, while the other often produces excellent but disconnected acts of repair.
There is a further problem. Much of the civilian response is specialized. Researchers speak to researchers, journalists speak through journalism, fact-checkers correct individual claims, security institutions concentrate on attribution, and media-literacy programmes teach specific information skills.¹¹ These functions are necessary and should not be replaced.¹² The difficulty is that corrosive fragmentation is not restricted to any one of those domains. It takes place across information, identity, relationships, institutions and political behaviour at the same time.
A fact-check can establish that a photograph is misleading without changing the social meaning attached to the group depicted in it. Attribution can demonstrate that a foreign operation promoted a narrative without preventing domestic actors from continuing to use the narrative once it becomes politically useful. A classroom can teach source evaluation¹³ while the wider social environment rewards certainty, outrage and tribal conformity. Individual interventions may succeed on their own terms while the cumulative process continues around them.
If Part I is approximately right, better corrections alone will not solve the civilian problem. The response may also need greater continuity.
Why counter-propaganda is not the answer
There is an obvious temptation here. If corrosive propaganda gains power through repetition, perhaps the answer is to build an equally powerful apparatus repeating better messages. That would be a mistake. A democratic response cannot simply be propaganda with preferred content.
The purpose should not be to teach citizens which political side deserves trust or to construct an authorized narrative that competes more effectively for attention. Doing so would reproduce the same basic structure: somebody decides which interpretation the public should hold and then develops techniques to make that interpretation socially dominant. Even if the intentions were benign, such a system would be vulnerable to political capture and would have little defence against becoming exactly what it claimed to oppose.
A civilian response should be different in kind. Rather than repeatedly supplying conclusions, it can repeatedly exercise capacities. It can create conditions in which checking sources is normal, revising a conclusion is socially possible, interaction outside one's political environment remains available and standards applied to opponents remain recognizable when power changes hands.
A society capable of resisting manipulation will still reach conclusions that many people dislike. It will contain criticism of governments, established institutions, immigration, military alliances, corporations, social movements and almost everything else. Resistance to corrosive fragmentation cannot require agreement on those questions without turning political pluralism into a security problem.
The goal is to maintain enough common ground around the process of disagreement that disagreement does not become complete enclosure.
The limits of the current educational model
Formal education has an important place in this response, but it has a reach problem. People who already believe media manipulation is a serious concern are more likely to attend a media-literacy event. People who associate the language of disinformation, democratic resilience or information integrity with political opponents may reject the invitation before encountering the material.¹⁴
A programme intended to reduce enclosure may end up serving mainly people who were already relatively open to its premises. The programme can still be useful, but it does little to solve the problem of reaching people outside the audience that already recognizes itself in the language of civic education.
Can some of the capacities identified in Part I be practised without requiring people to seek out an explicitly educational or political environment? Can they become part of ordinary social and cultural activity rather than something citizens encounter only when they agree to be taught?
If so, the civilian side may have resources it has not yet organized particularly well.
Part III
The Social Vehicle
Imagine two posters displayed next to one another. The first advertises Building Resilience Against Disinformation Through Media Literacy. The second advertises How Memes Travel. The first might contain excellent material about source verification, misleading statistics, emotional manipulation and image provenance, but it has already told potential participants what kind of problem they are expected to recognize and what sort of educational experience they are entering.
The second event requires much less initial agreement. Someone may attend because memes are funny, because they spend too much time online, because they enjoy visual culture or simply because a friend is going. During the evening they might trace where familiar images first appeared, compare captions attached to the same photograph, discover how cropping changes interpretation and find examples that have been recycled years later in completely different contexts. They could watch a joke move between communities and change meaning as it travels. In the process they would practise source tracing, provenance checking, contextualization and framing analysis without first having to identify themselves as people seeking instruction in disinformation.
The difference goes beyond marketing. It suggests reversing the usual educational model. Instead of beginning with a lesson and then searching for an audience willing to receive it, begin with an activity people genuinely want to participate in and design the activity so that useful habits are practised within it.
A photography event could give several groups the same collection of images and ask each group to tell the story of what happened. The groups would almost certainly construct different narratives, and the effect of selection and framing would become visible through experience. A local-history project could compare newspaper accounts of the same strike, demonstration or controversial event and ask participants to reconstruct what probably happened. A game could reward players for changing conclusions when new evidence appears rather than for defending the position adopted first. A collaborative art project might bring together people who would never voluntarily attend an explicitly political dialogue exercise.
These activities would serve different purposes. Some would mainly exercise judgment. Others would expose people to ambiguity. Some might create contact across social boundaries, while others might give people the experience of contributing something instead of simply receiving information. Many would also simply be enjoyable, which is not a defect in the model.
The activity has to be capable of standing on its own rather than existing merely as packaging for a civic lesson.
The basic idea
I would call the environment carrying these activities a social vehicle. The term is broad because the proposal is not tied to one kind of workshop, institution or teaching method. An exhibition, neighbourhood archive, cooking project, photography walk, game, performance, meme evening, local investigation or collaborative publication could all belong to the same vehicle without resembling one another very much.
There is no evidence yet that such a vehicle will work. The theory gives us a reason to test the idea. The hypothesis is that if fragmentation grows partly through repeated enclosure, civil society may be able to create environments in which some of the habits that enclosure weakens are repeatedly used instead. Participants might check where something came from because the activity makes provenance interesting. They might encounter competing interpretations without being forced to choose one immediately. They might discover that revising an opinion does not need to be humiliating. They might work successfully with someone whom they had previously understood mainly through a political category.
The public would not be asked to join a movement in order to participate. Someone should be able to attend an event, exhibit work, make something, volunteer for a project or simply enjoy an evening without agreeing with the organization's analysis of propaganda or its views on current political controversies. Participation should not require support for the Theory of Corrosive Fragmentation, a particular geopolitical position or an approved cultural identity. If ideological agreement becomes an informal price of admission, the vehicle has begun reproducing the enclosure it was intended to make more difficult.
The informational habits encouraged by the project should remain as source-neutral as possible. Participants can become accustomed to asking where something came from, what may have been omitted, what conclusion the available evidence permits, what information might change an interpretation and whether they are applying the same standard to people they agree with. Those questions remain useful whether the source is a hostile foreign actor, a government, an activist group, a newspaper, an influencer, a corporation, a political party or a friend.
The vehicle could practise media literacy without requiring media literacy to become its public identity.
Learning through the environment
The process could loosely be described as learning by osmosis, but the phrase should not imply concealed instruction. Participants should know what they are taking part in, especially in an organization concerned with manipulation.
The activities can be completely honest about what they are doing. A photography event can say that it explores how images tell stories, while a meme project can openly examine how online culture changes as it travels. A local-history group can describe itself as investigating different accounts of the city's past. The difference is that education does not need to be the principal reason someone attends.
Some learning then emerges through practice. Someone who spends an evening tracing images becomes more accustomed to asking where an image originated. Someone who watches several different stories constructed from the same pool of photographs has personally experienced how selection changes meaning. Someone who works successfully with a person they expected to dislike acquires a small piece of experience that does not fit perfectly into a categorical picture of that group.¹⁵
None of those experiences needs to produce political conversion. Conversion would be a poor measure of success because the purpose is not to bring participants toward the organization's preferred conclusions. It is to preserve their capacity to reach their own conclusions without becoming entirely enclosed within one interpretive community.
Art and cultural activity may provide useful environments for this, but they should not be romanticized. Art is not inherently democratic or tolerant, and cultural organizations can become intensely homogeneous. Its possible usefulness lies in its ability to tolerate ambiguity, present incompatible perspectives without immediately resolving them, create experience before demanding a verdict and give people reasons to enter a shared space that have little to do with politics. Propaganda often reduces complexity to certainty. Cultural experience can sometimes make complexity interesting rather than threatening.
That is enough to make art worth experimenting with, while keeping expectations modest.¹⁶
The social side
The social dimension may prove as important as the informational one. If fragmentation works partly by allowing political identity to absorb other ways in which people understand one another, then ordinary relationships that cross those categories can provide friction against enclosure even when nobody discusses politics.¹⁷
Twenty people might spend an evening building, cooking, photographing, researching or performing something together. One participant discovers that somebody they had mentally placed in a hostile political camp is funny. Another turns out to be irritating but extremely competent. Someone they expected to have nothing in common with knows a great deal about something they care about. A disagreement emerges and the project continues anyway. Nothing dramatic needs to happen. Nobody has to announce that a stereotype has been defeated, and nobody has to change political allegiance. The category of them simply becomes somewhat harder to maintain in its pure form because an actual person has introduced inconvenient information into it.
The possible effect here is modest. Social contact can go badly. A hostile encounter can reinforce prejudice, and forced interaction can leave people more enclosed than they were before.¹⁸ The vehicle would have to learn which kinds of shared activity encourage curiosity and cooperation, which simply reproduce existing social boundaries and which make things worse.
Social contact cannot be assumed to heal division, but ordinary relationships can complicate political categories that would otherwise remain socially uncontested.
From local persistence to a distributed response
A single event will not counter years of repeated manipulative communication, and one local project cannot compete with the financial resources, media infrastructure or political reach available to a major state-backed influence operation. If the social vehicle has any wider value, it will come from persistence at the local level and connection between local spaces.
Different activities do not need to repeat the same lesson or attract the same audience. What they can share is an environment in which curiosity, source awareness, cooperation and openness to revision remain available over time. One month a local group might host a photography project, another month a history project or an exhibition, while another activity may have almost no explicit informational component at all. No individual event has to carry the theory on its shoulders. The relevant question is whether a continuing social environment can make certain habits more ordinary than a one-off intervention can.
If the model develops beyond one place, federation makes more sense than ordinary organizational expansion. Social fractures differ by location. Historical memory, language, class, institutions and political culture make an activity that works in Maastricht potentially useless in Bucharest, Warsaw, Naples or Helsinki. A central organization cannot know local relationships as well as people who live inside them, so local groups would need genuine autonomy over their activities, partnerships and cultural language.
A wider network could provide shared capacity while leaving programming local. Research, practical tools, evaluation methods, technical support and institutional memory can travel between nodes. Something noticed in one city can become a question asked elsewhere. A pattern that appears independently across several places can be investigated collectively. Previous cases can be remembered rather than repeatedly rediscovered, and a useful response can travel without becoming a compulsory script. Different communities might respond to the same corrosive pressure through very different activities while working from the same verified understanding of what is happening.
Federation would therefore serve a practical function in the response. The local group provides sensitivity to its community, while the network provides a wider field of vision and the ability to learn across places.
When circumstances genuinely require a wider response, coherence matters more than uniformity.
A civic immune system
An immune system offers a useful analogy here, as long as we do not push the biology too far. An immune system is distributed throughout an organism. Local detection matters because problems first appear somewhere specific. Communication matters because what begins locally may become systemic. Memory matters because a system that must rediscover every threat from the beginning remains permanently disadvantaged. Coordination matters because some problems exceed the capacity of the place where they first appear.
A civilian network could operate on similar principles without pretending that political communication is biology. Local nodes possess knowledge of their communities and may notice shifts a distant institution misses. The wider network can help verify whether the same pattern is appearing elsewhere, connect people who have encountered similar techniques and preserve knowledge about earlier responses. Most activity can remain local, while some situations may justify many independent nodes responding during the same period with a common understanding of the problem.
The analogy also carries a warning. A healthy immune system must distinguish danger from what merely belongs in the organism. If it attacks everything unfamiliar, it becomes destructive. The equivalent civic danger would be treating political disagreement itself as a pathogen. Criticism of government, unpopular political opinions, radical ideas or hostility toward established institutions cannot become evidence of infection simply because they are uncomfortable. Democratic societies require disagreement, including angry and unsettling disagreement.
A civic response should focus on the corrosive process itself: deceptive amplification, categorical scapegoating, manipulation that makes evidence irrelevant, repeated attempts to turn institutional failure into universal illegitimacy, or narratives that gradually withdraw political and moral standing from whole groups.
A healthy civic response would need something comparable to immune tolerance, the ability to tolerate enormous political difference while remaining alert to processes that erode the conditions under which political difference can continue.
The vehicle must resist enclosure itself
A project created to make enclosure more difficult can become another enclosed community. It can develop insiders and outsiders, preferred political assumptions and people who quietly acquire ownership over the organization's culture. An artistic community can speak constantly about openness while nearly everyone in the room has the same education, politics and social background. A decentralized organization can claim nobody is in charge even though a few people control funding relationships, passwords, institutional knowledge and important decisions.
The theory has to apply to the vehicle itself. If enclosure can emerge through social sorting, selective standards and closed interpretive communities, a project concerned with resisting fragmentation has to remain capable of detecting those same tendencies in itself. Removing formal hierarchy does not remove power. It can simply make power less visible.¹⁹ Experience, trust, reputation, access to information and control over resources all create influence even when nobody holds a grand title.
A useful organizational principle might be maximum local freedom inside minimum common restraints. Public participation should remain easy, while responsibility is acquired more deliberately. People responsible for money, participant safety, data, shared resources or major decisions should know what powers they have and where those powers end. Authority should remain close to the work it affects, with wider structures becoming relevant only where decisions genuinely affect shared resources, legal responsibility, serious misconduct, safeguarding or the common principles of the network.
A small set of common principles could cover reciprocity, artistic and intellectual independence, transparency, financial accountability, safety, fair conflict procedures and the rule that ideological agreement is not required for participation. Discipline should concern conduct rather than belief. Threats, harassment, intimidation, misuse of funds or deliberate interference with other people's ability to participate concern behaviour. Holding an unpopular political view does not.
The vehicle should also periodically examine informal power instead of relying on its organizational chart. It can ask who actually shapes major decisions, who controls relationships with funders, who possesses information everyone else depends upon and whether the departure of one person would paralyse the project. The same scrutiny should apply to the composition of the community. If every event gradually attracts the same political, educational or cultural audience, the project should not simply celebrate having found its people.
It may have discovered that it is becoming another enclosure.
Shared infrastructure
A distributed network would still require some connective structure. A central organization powerful enough to dictate local political interpretation would undermine the model, but a network with no common infrastructure would struggle to learn, remember or act coherently. Any central structure should be limited to the functions local groups genuinely need to share.
Its role could include maintaining shared documentation, supporting research and evaluation, assisting with common legal or financial requirements, connecting local groups and preserving institutional memory. It could help verify emerging patterns and make previous experience searchable so that every local node does not have to begin from zero. Its role should be strongest where coordination genuinely adds value and weakest where local knowledge matters more.
It also helps with the problem of continuity. Local groups may appear and disappear, volunteers may burn out, funding may end and particular projects may fail. Persistence does not require every organization to last forever. It requires useful knowledge, relationships and methods to survive beyond individual projects.
That continuity is itself part of the intervention.
What would count as success?
The proposal should be tested without assuming in advance that it works. Research in fields such as media literacy, cross-group contact, participatory activity and cultural experience gives reasons to investigate parts of the idea, but evidence for individual components would not demonstrate that combining them into a social vehicle reduces corrosive fragmentation. Whether the combination works remains an open question.²⁰
Early experiments should ask specific questions and avoid grand claims that a community has become “resilient.” An information-related activity might examine whether people become better at distinguishing evidence from framing without simply becoming more suspicious of everything. A cultural project could test whether it attracts people beyond the social audience that normally attends civic events. A collaborative activity could examine whether participants actually form relationships outside the groups with which they arrived. Follow-up could ask whether any observed change persists after the immediate experience has faded.
The people who do not participate are as important as the people who do. If every event attracts participants who were already unusually open to other perspectives, positive results would tell us relatively little about deeper enclosure. If the same educated and politically similar audience appears repeatedly, that is evidence about the limits of the model. If contact increases hostility, that matters. If people become more confident in recognizing manipulation but also more indiscriminately distrustful, that matters. If artists begin changing their work merely to satisfy evaluation requirements, that matters as well.
Evaluation should preserve the same correctability the vehicle hopes to encourage in participants. An apparent effect should be allowed to disappear on follow-up. An appealing idea should be allowed to perform badly. A simpler intervention should be preferred when it works better. If an ordinary media-literacy workshop proves more effective for a particular purpose, use it. If a local sports club already creates exactly the cross-boundary relationships the project is trying to produce, supporting that club may make more sense than inventing a new institution. If an exhibition produces excellent art and no measurable civic effect, it should be allowed to remain an excellent exhibition.
The social vehicle deserves to continue only if reality gives us reasons to continue it.
The practical wager
The practical wager begins with a limitation. Cultural programming will not manufacture better citizens, and a loose network of local groups will never match the money, technology and political power available to major influence operators. A single local space will usually look insignificant beside those resources. Its value has to be judged on a different scale.
Civil society already occupies much of the social environment into which propaganda enters. It exists in neighbourhoods, associations, clubs, professions, schools, cultural spaces, friendships, volunteer networks and ordinary relationships. These structures are usually not organized as a response to manipulation, and there is no reason most of them should become one. Their significance lies in the fact that they are where people actually live social life.
If some of those spaces can become better at preserving curiosity, correctability, reciprocal standards and relationships that cross political categories, and if enough of them can share knowledge and occasionally act coherently when they recognize the same corrosive pressure, their collective capacity may become much greater than the visible power of any one project. This would not defeat propaganda message by message or eliminate political conflict. Its purpose would be to make societies somewhat less available to the processes corrosive propaganda exploits.
The immune-system analogy also helps explain what this could look like in practice without asking each local node to carry the whole burden. Most of the time a node simply has to remain locally rooted, healthy and connected. The wider network allows detection in one place to become knowledge elsewhere and past experience to become shared memory. At the same time, the system must preserve the equivalent of immune tolerance. It cannot begin treating disagreement itself as the disease, because its purpose is to protect the conditions in which disagreement remains possible.
If corrosive fragmentation can deepen gradually through repeated experiences of suspicion, enclosure and categorical blame, resistance may also accumulate through ordinary experiences moving in another direction. Someone checks before sharing. Someone changes an interpretation without being humiliated for it. Two people who expected to dislike one another successfully make something together. A local institution admits a mistake and repairs it. A community refuses to turn the actions of one person into a verdict on an entire group.
None of these moments defeats propaganda on its own. Neither does one local node, one workshop or one cultural project. But connected and preserved across ordinary civic life, they may form something more durable than the haphazard responses that currently appear after each new campaign has already begun. The social vehicle would have a more modest ambition. It would try to build a civilian capacity that is locally rooted, continuously present, able to learn and occasionally able to act together.
If that capacity can make enclosure harder to produce, fragmentation harder to deepen and cruelty harder to justify, it is a possibility worth testing.
And now, you know.
Endnotes
1. Jennifer McCoy, Tahmina Rahman, and Murat Somer, “Polarization and the Global Crisis of Democracy: Common Patterns, Dynamics, and Pernicious Consequences for Democratic Polities,” American Behavioral Scientist 62, no. 1 (2018): 16–42, doi:10.1177/0002764218759576; Jennifer McCoy and Murat Somer, “Toward a Theory of Pernicious Polarization and How It Harms Democracies: Comparative Evidence and Possible Remedies,” The ANNALS of the American Academy of Political and Social Science 681, no. 1 (2019): 234–271, doi:10.1177/0002716218818782.
https://doi.org/10.1177/0002764218759576
https://doi.org/10.1177/0002716218818782
2. Claire Wardle and Hossein Derakhshan, Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making, Council of Europe Report DGI(2017)09 (Strasbourg: Council of Europe, 2017).
https://edoc.coe.int/en/media/7495-information-disorder-toward-an-interdisciplinary-framework-for-research-and-policy-making.html
3. Lilliana Mason, Uncivil Agreement: How Politics Became Our Identity (Chicago: University of Chicago Press, 2018), doi:10.7208/chicago/9780226524689.001.0001.
https://doi.org/10.7208/chicago/9780226524689.001.0001
4. Steeven Ye, David Attali, Maria Ghazi, Arnaud Cachia, Mathieu Cassotti, and Grégoire Borst, “Systematic Review and Meta-Analysis of the Evidence for an Illusory Truth Effect and Its Determinants,” Nature Communications 17 (2026): 3270, doi:10.1038/s41467-026-70041-x.
https://doi.org/10.1038/s41467-026-70041-x
5. Alicia Wanless and Michael Berk, “The Audience Is the Amplifier: Participatory Propaganda,” in The SAGE Handbook of Propaganda, ed. Paul Baines, Nicholas O’Shaughnessy, and Nancy Snow (London: SAGE, 2019), 85–104, doi:10.4135/9781526477170.n7.
https://doi.org/10.4135/9781526477170.n7
6. Yochai Benkler, Robert Faris, and Hal Roberts, Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics (New York: Oxford University Press, 2018), doi:10.1093/oso/9780190923624.001.0001.
https://doi.org/10.1093/oso/9780190923624.001.0001
7. Eli J. Finkel et al., “Political Sectarianism in America,” Science 370, no. 6516 (2020): 533–536, doi:10.1126/science.abe1715. The literature supports the mechanisms of othering, aversion and moralization; it does not establish an inevitable progression from political fragmentation to cruelty.
https://doi.org/10.1126/science.abe1715
8. Diana C. Mutz, “Cross-cutting Social Networks: Testing Democratic Theory in Practice,” American Political Science Review 96, no. 1 (2002): 111–126, doi:10.1017/S0003055402004264.
https://doi.org/10.1017/S0003055402004264
9. John L. Sullivan, James Piereson, and George E. Marcus, Political Tolerance and American Democracy (Chicago: University of Chicago Press, 1982); Christopher J. Anderson, André Blais, Shaun Bowler, Todd Donovan, and Ola Listhaug, Losers’ Consent: Elections and Democratic Legitimacy (Oxford: Oxford University Press, 2005), doi:10.1093/0199276382.001.0001. These literatures provide established component evidence for political tolerance and acceptance of legitimate adverse outcomes; the broader concept of mutual political recognition used here is an integrative formulation.
https://press.uchicago.edu/ucp/books/book/chicago/P/bo3775122.html
https://doi.org/10.1093/0199276382.001.0001
10. European External Action Service, 4th EEAS Annual Report on Foreign Information Manipulation and Interference Threats (Brussels: European External Action Service, March 12, 2026); U.S. Department of Justice, “Justice Department Disrupts Covert Russian Government-Sponsored Foreign Malign Influence Operation Targeting Audiences in the United States and Elsewhere,” September 4, 2024. The EEAS documents the scale, technological adaptation and network infrastructure of contemporary FIMI, while the Justice Department’s Doppelganger case provides a documented example of a multi-year state-directed operation using public-relations companies, online infrastructure, paid advertising, influencers, fabricated personas and AI-generated content.
https://www.eeas.europa.eu/eeas/4th-eeas-annual-report-foreign-information-manipulation-and-interference-threats_en
https://www.justice.gov/archives/opa/pr/justice-department-disrupts-covert-russian-government-sponsored-foreign-malign-influence
11. Guanxiong Huang, Wufan Jia, and Wenting Yu, “Media Literacy Interventions Improve Resilience to Misinformation: A Meta-Analytic Investigation of Overall Effect and Moderating Factors,” Communication Research, published online October 4, 2024, doi:10.1177/00936502241288103; Almog Simchon, Tomer Zipori, Louis Teitelbaum, Stephan Lewandowsky, and Sander van der Linden, “A Signal Detection Theory Meta-Analysis of Psychological Inoculation Against Misinformation,” Current Opinion in Psychology 67 (2026): 102194, doi:10.1016/j.copsyc.2025.102194. Media-literacy and inoculation/prebunking interventions are related but distinct intervention families; both provide evidence that targeted information skills can improve discernment.
https://doi.org/10.1177/00936502241288103
https://doi.org/10.1016/j.copsyc.2025.102194
12. OECD, Facts not Fakes: Tackling Disinformation, Strengthening Information Integrity (Paris: OECD Publishing, 2024), doi:10.1787/d909ff7a-en. Existing information-integrity policy already includes media literacy, public communication, plural information sources, civil society and wider societal resilience; the argument here should therefore not be read as claiming that current responses ignore these dimensions.
https://doi.org/10.1787/d909ff7a-en
13. Andrew M. Guess, Michael Lerner, Benjamin Lyons, Jacob M. Montgomery, Brendan Nyhan, Jason Reifler, and Neelanjan Sircar, “A Digital Media Literacy Intervention Increases Discernment between Mainstream and False News in the United States and India,” Proceedings of the National Academy of Sciences 117, no. 27 (2020): 15536–15545, doi:10.1073/pnas.1920498117.
https://doi.org/10.1073/pnas.1920498117
14. Ryan C. Moore and Jeffrey T. Hancock, “A Digital Media Literacy Intervention for Older Adults Improves Resilience to Fake News,” Scientific Reports 12 (2022): 6008, doi:10.1038/s41598-022-08437-0. The study documents self-selection among participants and therefore supports the audience-selection concern discussed here, although it does not directly establish that prior concern about disinformation, or hostility toward the language of information integrity, causes willingness or unwillingness to attend.
https://doi.org/10.1038/s41598-022-08437-0
15. Joshua L. Kalla and David E. Broockman, “Reducing Exclusionary Attitudes through Interpersonal Conversation: Evidence from Three Field Experiments,” American Political Science Review 114, no. 2 (2020): 410–425, doi:10.1017/S0003055419000923. The findings support perspective-getting and narrative exchange as mechanisms capable of complicating categorical views of out-groups; they do not demonstrate that a cultural or social vehicle would reproduce the same effects.
https://doi.org/10.1017/S0003055419000923
16. Steve Rathje, Leor Hackel, and Jamil Zaki, “Attending Live Theatre Improves Empathy, Changes Attitudes, and Leads to Pro-Social Behavior,” Journal of Experimental Social Psychology 95 (2021): 104138, doi:10.1016/j.jesp.2021.104138; Reba Troxler, Thalia Goldstein, Steven Holochwost, Charles Beekman, Stephanie McKeel, and Muna Shami, “Deeper Engagement with Live Theater Increases Middle School Students’ Empathy and Social Perspective Taking,” Applied Developmental Science 27, no. 4 (2023): 352–372, doi:10.1080/10888691.2022.2096610. These findings support particular social and perspective-related effects of cultural experience, not a general claim that arts participation produces durable civic resilience.
https://doi.org/10.1016/j.jesp.2021.104138
https://doi.org/10.1080/10888691.2022.2096610
17. Jill Sonke, Virginia Pesata, Aaron Colverson, Jane Morgan-Daniel, Alexandra K. Rodriguez, Gray Davidson Carroll, Shanaé Burch, et al., “Relationships between Arts Participation, Social Cohesion, and Well-Being: An Integrative Review of Evidence,” Frontiers in Public Health 13 (2025): 1589693, doi:10.3389/fpubh.2025.1589693; Hanka Otte, “Bonding or Bridging? On Art Participation and Social Cohesion in a Rural Region of the Netherlands,” Poetics 76 (2019): 1–14, article 101355, doi:10.1016/j.poetic.2019.02.006. The evidence supports associations between arts participation, social connection and forms of cohesion, while causal claims about bridging and longer-term resilience remain limited.
https://doi.org/10.3389/fpubh.2025.1589693
https://doi.org/10.1016/j.poetic.2019.02.006
18. Thomas F. Pettigrew and Linda R. Tropp, “A Meta-Analytic Test of Intergroup Contact Theory,” Journal of Personality and Social Psychology 90, no. 5 (2006): 751–783, doi:10.1037/0022-3514.90.5.751; Fiona Kate Barlow, Stefania Paolini, Anne Pedersen, Matthew J. Hornsey, Helena R. M. Radke, Jake Harwood, Mark Rubin, and Chris G. Sibley, “The Contact Caveat: Negative Contact Predicts Increased Prejudice More Than Positive Contact Predicts Reduced Prejudice,” Personality and Social Psychology Bulletin 38, no. 12 (2012): 1629–1643, doi:10.1177/0146167212457953. The literature establishes that contact effects are conditional and that negative contact can increase prejudice; the specific warning about forced interaction is a cautious inference rather than a directly tested general finding.
https://doi.org/10.1037/0022-3514.90.5.751
https://doi.org/10.1177/0146167212457953
19. Jo Freeman, “The Tyranny of Structurelessness,” Berkeley Journal of Sociology 17 (1972–73): 151–164.
https://www.jstor.org/stable/41035187
20. Evidence for the proposed social vehicle remains componential rather than direct. Research supports particular mechanisms involving cultural experience, perspective-getting, intergroup contact, social connection and participatory activity, but it has not demonstrated that combining these mechanisms into an integrated social vehicle reduces corrosive fragmentation. See notes 15–18. The integrated proposition therefore remains a testable hypothesis rather than an established intervention effect.
Bibliography
Anderson, Christopher J., André Blais, Shaun Bowler, Todd Donovan, and Ola Listhaug. Losers’ Consent: Elections and Democratic Legitimacy. Oxford: Oxford University Press, 2005. doi:10.1093/0199276382.001.0001.
https://doi.org/10.1093/0199276382.001.0001
Barlow, Fiona Kate, Stefania Paolini, Anne Pedersen, Matthew J. Hornsey, Helena R. M. Radke, Jake Harwood, Mark Rubin, and Chris G. Sibley. “The Contact Caveat: Negative Contact Predicts Increased Prejudice More Than Positive Contact Predicts Reduced Prejudice.” Personality and Social Psychology Bulletin 38, no. 12 (2012): 1629–1643. doi:10.1177/0146167212457953.
https://doi.org/10.1177/0146167212457953
Benkler, Yochai, Robert Faris, and Hal Roberts. Network Propaganda: Manipulation, Disinformation, and Radicalization in American Politics. New York: Oxford University Press, 2018. doi:10.1093/oso/9780190923624.001.0001.
https://doi.org/10.1093/oso/9780190923624.001.0001
European External Action Service. 4th EEAS Annual Report on Foreign Information Manipulation and Interference Threats. Brussels: European External Action Service, March 12, 2026.
https://www.eeas.europa.eu/eeas/4th-eeas-annual-report-foreign-information-manipulation-and-interference-threats_en
Finkel, Eli J., Christopher A. Bail, Mina Cikara, Peter H. Ditto, Shanto Iyengar, Samara Klar, Lilliana Mason, et al. “Political Sectarianism in America.” Science 370, no. 6516 (2020): 533–536. doi:10.1126/science.abe1715.
https://doi.org/10.1126/science.abe1715
Freeman, Jo. “The Tyranny of Structurelessness.” Berkeley Journal of Sociology 17 (1972–73): 151–164.
https://www.jstor.org/stable/41035187
Guess, Andrew M., Michael Lerner, Benjamin Lyons, Jacob M. Montgomery, Brendan Nyhan, Jason Reifler, and Neelanjan Sircar. “A Digital Media Literacy Intervention Increases Discernment between Mainstream and False News in the United States and India.” Proceedings of the National Academy of Sciences 117, no. 27 (2020): 15536–15545. doi:10.1073/pnas.1920498117.
https://doi.org/10.1073/pnas.1920498117
Huang, Guanxiong, Wufan Jia, and Wenting Yu. “Media Literacy Interventions Improve Resilience to Misinformation: A Meta-Analytic Investigation of Overall Effect and Moderating Factors.” Communication Research. Published online October 4, 2024. doi:10.1177/00936502241288103.
https://doi.org/10.1177/00936502241288103
Kalla, Joshua L., and David E. Broockman. “Reducing Exclusionary Attitudes through Interpersonal Conversation: Evidence from Three Field Experiments.” American Political Science Review 114, no. 2 (2020): 410–425. doi:10.1017/S0003055419000923.
https://doi.org/10.1017/S0003055419000923
Mason, Lilliana. Uncivil Agreement: How Politics Became Our Identity. Chicago: University of Chicago Press, 2018. doi:10.7208/chicago/9780226524689.001.0001.
https://doi.org/10.7208/chicago/9780226524689.001.0001
McCoy, Jennifer, Tahmina Rahman, and Murat Somer. “Polarization and the Global Crisis of Democracy: Common Patterns, Dynamics, and Pernicious Consequences for Democratic Polities.” American Behavioral Scientist 62, no. 1 (2018): 16–42. doi:10.1177/0002764218759576.
https://doi.org/10.1177/0002764218759576
McCoy, Jennifer, and Murat Somer. “Toward a Theory of Pernicious Polarization and How It Harms Democracies: Comparative Evidence and Possible Remedies.” The ANNALS of the American Academy of Political and Social Science 681, no. 1 (2019): 234–271. doi:10.1177/0002716218818782.
https://doi.org/10.1177/0002716218818782
Moore, Ryan C., and Jeffrey T. Hancock. “A Digital Media Literacy Intervention for Older Adults Improves Resilience to Fake News.” Scientific Reports 12 (2022): 6008. doi:10.1038/s41598-022-08437-0.
https://doi.org/10.1038/s41598-022-08437-0
Mutz, Diana C. “Cross-cutting Social Networks: Testing Democratic Theory in Practice.” American Political Science Review 96, no. 1 (2002): 111–126. doi:10.1017/S0003055402004264.
https://doi.org/10.1017/S0003055402004264
OECD. Facts not Fakes: Tackling Disinformation, Strengthening Information Integrity. Paris: OECD Publishing, 2024. doi:10.1787/d909ff7a-en.
https://doi.org/10.1787/d909ff7a-en
Otte, Hanka. “Bonding or Bridging? On Art Participation and Social Cohesion in a Rural Region of the Netherlands.” Poetics 76 (2019): 1–14, article 101355. doi:10.1016/j.poetic.2019.02.006.
https://doi.org/10.1016/j.poetic.2019.02.006
Pettigrew, Thomas F., and Linda R. Tropp. “A Meta-Analytic Test of Intergroup Contact Theory.” Journal of Personality and Social Psychology 90, no. 5 (2006): 751–783. doi:10.1037/0022-3514.90.5.751.
https://doi.org/10.1037/0022-3514.90.5.751
Rathje, Steve, Leor Hackel, and Jamil Zaki. “Attending Live Theatre Improves Empathy, Changes Attitudes, and Leads to Pro-Social Behavior.” Journal of Experimental Social Psychology 95 (2021): 104138. doi:10.1016/j.jesp.2021.104138.
https://doi.org/10.1016/j.jesp.2021.104138
Simchon, Almog, Tomer Zipori, Louis Teitelbaum, Stephan Lewandowsky, and Sander van der Linden. “A Signal Detection Theory Meta-Analysis of Psychological Inoculation Against Misinformation.” Current Opinion in Psychology 67 (2026): 102194. doi:10.1016/j.copsyc.2025.102194.
https://doi.org/10.1016/j.copsyc.2025.102194
Sonke, Jill, Virginia Pesata, Aaron Colverson, Jane Morgan-Daniel, Alexandra K. Rodriguez, Gray Davidson Carroll, Shanaé Burch, et al. “Relationships between Arts Participation, Social Cohesion, and Well-Being: An Integrative Review of Evidence.” Frontiers in Public Health 13 (2025): 1589693. doi:10.3389/fpubh.2025.1589693.
https://doi.org/10.3389/fpubh.2025.1589693
Sullivan, John L., James Piereson, and George E. Marcus. Political Tolerance and American Democracy. Chicago: University of Chicago Press, 1982.
https://press.uchicago.edu/ucp/books/book/chicago/P/bo3775122.html
Troxler, Reba, Thalia Goldstein, Steven Holochwost, Charles Beekman, Stephanie McKeel, and Muna Shami. “Deeper Engagement with Live Theater Increases Middle School Students’ Empathy and Social Perspective Taking.” Applied Developmental Science 27, no. 4 (2023): 352–372. doi:10.1080/10888691.2022.2096610.
https://doi.org/10.1080/10888691.2022.2096610
U.S. Department of Justice. “Justice Department Disrupts Covert Russian Government-Sponsored Foreign Malign Influence Operation Targeting Audiences in the United States and Elsewhere.” September 4, 2024.
https://www.justice.gov/archives/opa/pr/justice-department-disrupts-covert-russian-government-sponsored-foreign-malign-influence
Wanless, Alicia, and Michael Berk. “The Audience Is the Amplifier: Participatory Propaganda.” In The SAGE Handbook of Propaganda, edited by Paul Baines, Nicholas O’Shaughnessy, and Nancy Snow, 85–104. London: SAGE, 2019. doi:10.4135/9781526477170.n7.
https://doi.org/10.4135/9781526477170.n7
Wardle, Claire, and Hossein Derakhshan. Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making. Council of Europe Report DGI(2017)09. Strasbourg: Council of Europe, 2017.
https://edoc.coe.int/en/media/7495-information-disorder-toward-an-interdisciplinary-framework-for-research-and-policy-making.html
Ye, Steeven, David Attali, Maria Ghazi, Arnaud Cachia, Mathieu Cassotti, and Grégoire Borst. “Systematic Review and Meta-Analysis of the Evidence for an Illusory Truth Effect and Its Determinants.” Nature Communications 17 (2026): 3270. doi:10.1038/s41467-026-70041-x.
https://doi.org/10.1038/s41467-026-70041-x
