Abstract
Research on disinformation has expanded rapidly, producing substantial advances in understanding propaganda, influence operations, algorithmic amplification, belief formation and fact-checking. Yet one question remains only partially resolved: why do some communicative patterns persist long after their factual basis has been challenged or removed, while others disappear almost immediately?
Existing explanations usually emphasise the properties of messages, the psychology of audiences or the structure of communication networks. Each contributes important insight, but each tends to begin with discrete communicative events and explain their persistence after the fact.
This paper argues that persistence itself should become the primary object of analysis. Disinformation is not simply false information moving through a network. It is a recurrent communicative organisation sustained through the continual reproduction of relations between messages, participants, institutions and technologies across time. Its social force depends less upon the survival of particular claims than upon the persistence of the interpretive structures through which new claims become intelligible, credible and actionable.
The argument is not that relations matter in some general sense. It is that communicative continuity is a distinct organisational phenomenon that cannot be reduced to any individual message, participant or platform. Persistent disinformation is one visible instance of a broader process through which communicative systems emerge, reproduce themselves, transform, divide, merge or disappear.
The paper develops observable criteria for identifying such systems, examines examples of both persistence and collapse, and argues that digital platforms do more than amplify communication. They alter the relational and temporal conditions under which communicative organisations can survive.
1. Introduction
Every act of communication depends upon communication that has already occurred. A sentence is intelligible because a language already exists. A question presupposes conventions governing response. Institutions communicate through procedures established before any particular interaction. Even the identities of speakers and audiences are recognised through histories of previous communication.
No communicative event begins in isolation. Every communication inherits conditions that it did not create while contributing to the conditions under which future communication will occur.
This observation sits uneasily alongside much communication research. Classical transmission models explain how information moves between communicators (Shannon & Weaver, 1949). Later work expanded the analytical focus to include symbolic interaction (Mead, 1934; Blumer, 1969), language use (Wittgenstein, 1953), speech acts (Austin, 1962; Searle, 1969), cybernetic feedback (Wiener, 1948; Bateson, 1972), autopoietic organisation (Maturana & Varela, 1980) and the recursive reproduction of communication within social systems (Luhmann, 1995).
These traditions demonstrate that communication is considerably richer than the transmission of information. Meaning depends upon interaction, use, context, difference, feedback and institutional organisation. Yet messages, utterances, interactions and communicative acts generally remain the principal analytical objects.
Research on misinformation and disinformation has largely inherited this orientation. Considerable attention has been devoted to the production, diffusion and correction of false or misleading claims, the psychology of belief, algorithmic amplification, online influence operations and platform governance (Wardle & Derakhshan, 2017; Vosoughi, Roy & Aral, 2018; Lewandowsky et al., 2020; Bak-Coleman et al., 2021).
This work explains how misleading information is produced, transmitted and received. It explains less about why recognisable communicative organisations can survive the disappearance of the messages from which they appear to have been composed.
Consider a conspiracy narrative whose original prediction fails. The prediction may be abandoned, reinterpreted or forgotten, yet the larger communicative system continues. New events are inserted into the same pattern. Contrary evidence becomes evidence of concealment. Failed predictions become proof that unnamed forces intervened. Participants leave and others replace them. Platforms remove content, but the communicative organisation migrates elsewhere.
By contrast, countless rumours, slogans, memes and political campaigns generate brief attention and then vanish. They may spread widely for several days without developing the organisation required to sustain further communication.
The distinction is not adequately captured by asking which message was more persuasive or which network was larger. A communicative pattern may be highly visible yet structurally shallow. Another may initially appear marginal while possessing enough relational organisation to adapt, recruit and persist.
The central question is therefore:
Under what relational and temporal conditions do communicative patterns emerge, become self-reproducing, transform or disappear despite continual variation in messages, participants and circumstances?
This paper argues that communicative persistence should be treated as a primary explanatory problem. Persistent disinformation is not the survival of individual false claims. It is the reproduction of organised relations that shape how subsequent claims are interpreted, connected, repeated and acted upon.
The contribution is not simply to claim that communication is relational. It is to identify the persistent communicative organisation, rather than the message, as the relevant object of continuity. Messages are transient expressions through which that organisation is reproduced, modified or dissolved.
2. The Limits of Message-Centred Explanations
Communication theory has developed far beyond signal transmission. Across philosophy, sociology, linguistics, cybernetics and media studies, communication has been understood as interpretation, action, interaction, difference, feedback and social reproduction.
This development has been highly productive. Shannon and Weaver’s mathematical theory established a rigorous account of transmission while deliberately excluding questions of meaning (Shannon & Weaver, 1949). Speech act theory shifted attention from what language represents to what it accomplishes (Austin, 1962; Searle, 1969). Symbolic interactionism located meaning in social interaction rather than within symbols themselves (Mead, 1934; Blumer, 1969). Wittgenstein located meaning in use (Wittgenstein, 1953), while Bateson treated information as a difference situated within a relational system (Bateson, 1972). Maturana and Varela situated communication within the recursive organisation of living systems, and Luhmann described social systems as reproducing themselves through communication (Maturana & Varela, 1980; Luhmann, 1995).
These approaches move well beyond a simple sender-message-receiver model. The remaining limitation is subtler. The communicative event usually remains the point from which analysis begins. Explanation concerns how a message is produced, interpreted, transmitted, transformed or acted upon.
Contemporary disinformation research often follows the same pattern. It investigates why particular claims are believed, how they spread through networks, how platforms rank them and why corrections fail. These are necessary questions, but they begin from communicative events whose organisation across time is not itself fully explained.
The problem becomes visible when the constituent elements change while the communicative pattern persists.
Individual claims vary. Participants enter and leave. Platforms rise and decline. Narratives fragment, recombine and adapt to new circumstances. Yet observers continue to recognise a stable interpretive pattern linking successive communications.
Anti-vaccination communication provides one example. Particular claims have changed across vaccines, diseases, technologies and political contexts. One claim may concern toxicity, another fertility, another state surveillance and another pharmaceutical profit. The factual propositions differ, but the larger communicative organisation may persist through recurrent relations of distrust, concealed agency, bodily threat, institutional betrayal and suppressed knowledge.
The continuity does not reside in any single claim. It lies in the relational structure through which different claims become mutually reinforcing.
A contrasting case is a viral but short-lived rumour. It may be repeated thousands of times yet disappear once attention shifts because it has not established durable relations between identity, community, explanation, authority and future expectation. It achieved diffusion without achieving persistence.
Message-centred analysis can describe both events as successful transmission. It is less able to explain why one became part of a self-reproducing communicative system while the other did not.
The distinction is fundamental. A communicative event is transient. It occurs, is interpreted and passes. Persistence is expressed through many such events without being reducible to any one of them.
The relevant theoretical object is therefore not simply the message, but the organised continuity through which successive messages become connected.
3. Communicative Persistence
Individual communicative events are inherently transient. A spoken sentence disappears almost as soon as it is uttered. A social-media post is rapidly displaced by later content. News stories are superseded. Conversations end. Participants leave. Institutions replace their members.
If communicative organisation persists, its continuity cannot be identified exclusively with any particular message or participant.
This principle is familiar elsewhere. English persists although its speakers, vocabulary, pronunciation and usage continually change. A legal tradition persists despite new judges, statutes and precedents. A scientific discipline remains recognisable while its theories, methods and personnel are repeatedly revised. A religion may persist across centuries despite disagreement, institutional division and reinterpretation.
In each case, persistence does not require material or informational identity. It depends upon organised continuity across replacement.
Communication exhibits the same structure. Every communicative event inherits conditions established by previous communication while modifying the conditions under which future communication will occur. Questions create expectations of response. Agreements generate obligations. Public accusations alter later interpretations. Institutional decisions constrain what can subsequently be said or done. Narratives establish distinctions through which new events are classified.
Communication therefore possesses an historical organisation. It does not merely occur within a context. It contributes to the reproduction or alteration of that context.
Persistence emerges when successive communicative events reproduce enough of this organisation for later events to remain intelligible as part of the same process.
This continuity can be identified through several observable features:
1. Relational recurrence: similar relations between actors, authorities, causes, threats and responses recur despite changes in wording or content.
2. Interpretive transfer: new events are readily absorbed into an existing explanatory structure.
3. Participant replaceability: the communicative pattern survives the departure of particular speakers or audiences.
4. Cross-platform continuity: the organisation persists despite migration between technologies or institutional settings.
5. Generative capacity: the pattern produces new messages rather than merely repeating old ones.
6. Resistance to local correction: the removal or correction of individual claims does not dissolve the wider organisation.
7. Identity or role formation: participation creates recognisable positions such as believer, expert, traitor, victim, investigator or enemy.
8. Temporal extension: the pattern establishes expectations that organise interpretation beyond the immediate event.
These criteria distinguish persistence from visibility. A message may be popular without becoming self-reproducing. A communicative organisation may remain relatively small while possessing substantial continuity.
Persistence is therefore not repetition alone. It is the reproduction of an organised field of relations through variation.
This persistence may also be understood as a form of harmonic organisation. A communicative system does not need to repeat identical messages to retain continuity. It may preserve recurrent relations, temporal rhythms and phase dependencies across changing expressions. Different messages can occupy equivalent positions within the same relational structure, allowing the organisation to remain recognisable even as its immediate content changes.
The Wiener–Khinchin theorem establishes a formal relation between the autocorrelation of a stationary process and its power spectral density (Wiener, 1930; Khintchine, 1934). It does not itself constitute a theory of communication, but it provides a useful formal analogy. A persistent communicative system may be detectable not only through repeated content but through recurrent correlations distributed across a changing sequence of events. What persists is then less like a stored object than a reproducible mode: a harmonic structure regenerated through variation.
4. Emergence, Transformation and Disappearance
A theory of persistence must also explain emergence and collapse. Communicative systems do not exist eternally. They form, expand, stabilise, divide, merge and disappear.
4.1 Emergence
A communicative system emerges when previously separate events become recursively connected. A message refers to earlier messages, anticipates responses, assigns roles, establishes distinctions and generates reasons for further communication.
A single rumour may produce attention without producing organisation. It becomes more durable when it begins to organise relationships: who can be trusted, who is concealing information, which evidence matters, what future event should be expected and how disagreement should be interpreted.
The communicative pattern then begins to generate the conditions of its own continuation.
Political slogans illustrate the difference. Most slogans disappear with the campaigns that produced them. Some persist because they condense a wider organisation of identity, grievance, history and expectation. They become reusable interpretive devices rather than isolated statements.
4.2 Transformation
Persistent systems survive by changing. A communicative organisation that cannot incorporate new circumstances becomes brittle. It may collapse when its central claims fail or when its participants encounter incompatible realities.
More adaptive systems preserve their core relations while altering their messages. A failed prediction is reframed. A new enemy replaces an old one. A technological change becomes evidence for an existing suspicion. A new political event is placed within an established moral structure.
The system persists because transformation occurs without destroying recognisable continuity.
This is why persistent disinformation often appears inconsistent at the level of individual propositions. Its continuity does not depend upon propositional consistency. It depends upon the reproduction of deeper relational positions: hidden authority, persecuted truth, corrupt expertise, threatened identity or imminent revelation.
4.3 Bifurcation and merger
Communicative systems can divide when incompatible interpretations become sufficiently organised to reproduce themselves independently. Religious schisms, ideological factions and scientific disputes often develop in this way. The competing systems share a common history but reorganise its relations differently.
Systems may also merge. Distinct conspiracy narratives can become connected through shared enemies, sources or interpretive rules. Anti-government, anti-vaccination and financial conspiracy communities may converge when common relations of distrust and concealed control allow previously separate narratives to reinforce one another.
4.4 Disappearance
Communicative systems disappear when they fail to reproduce the relations necessary for further communication.
This may occur because:
– the pattern loses access to participants or institutions;
– its distinctions no longer organise experience;
– competing explanations absorb its concerns;
– its future expectations repeatedly fail without successful reinterpretation;
– technological environments cease to sustain its recurrence;
– participation no longer produces identity, status or practical consequence;
– the pattern becomes unable to generate new communication.
Most viral trends disappear in this way. They may achieve enormous exposure but lack sufficient organisational depth to survive the loss of novelty. Their messages are repeated, but the relations required for continued reproduction are not established.
A theory of persistence must therefore distinguish between diffusion and reproduction. Diffusion concerns how far a message travels. Reproduction concerns whether communication reorganises the conditions under which more communication of the same general kind becomes likely.
5. The Relational Organisation of Meaning
Messages are not self-contained units carrying fixed significance. Their meaning depends upon the relations they establish with previous and subsequent communication.
The same sentence may reassure, accuse, threaten or amuse without any change in its wording because its communicative position has changed. Meaning is produced through relations of timing, history, identity, expectation and response.
This is consistent with traditions that locate meaning in use, interaction or difference (Wittgenstein, 1953; Mead, 1934; Bateson, 1972). The argument developed here extends those traditions by locating continuity in the organisation of relations across time.
A banknote illustrates the point. Its material properties do not explain its value. Its significance depends upon a persistent system of legal authority, institutional recognition, expectation, accounting and exchange. Individual transactions are transient. The monetary system persists because each transaction reproduces relations that make later transactions possible.
Language operates similarly. No single sentence sustains English. Each utterance inherits and modifies a larger organisation of grammar, convention and expectation. Most utterances disappear immediately, yet the language persists through them.
Persistent disinformation follows the same general principle. A claim becomes consequential because it occupies a position within an organised structure of relations. It identifies trusted and untrusted authorities, assigns causes, establishes threats, interprets evidence and anticipates future events.
The significance of the message is therefore identical neither with its literal content nor with the private belief of an individual recipient. It lies in the relational dependencies that the message activates and reproduces.
Communication persists not by preserving identical messages but by maintaining the organisation through which changing messages continue to perform recognisable functions.
6. Persistent Disinformation as Relational Reproduction
Persistent disinformation is one expression of this more general communicative process. Its continuity does not depend upon the survival of a single false claim. It depends upon the continued reproduction of relations capable of generating and organising new claims.
Consider repeated narratives of electoral fraud. Particular allegations may be disproven, withdrawn or replaced. Courts may reject cases. Predictions may fail. Specific witnesses may lose credibility. Yet the broader communicative organisation can persist if subsequent events continue to be interpreted through relations of stolen legitimacy, corrupt administration, suppressed evidence and institutional betrayal.
The individual claim is replaceable. Its relational position is not.
The same structure appears in recurring health conspiracies. One substance, vaccine, institution or medical intervention may replace another, while the organisation connecting bodily threat, hidden profit, corrupt expertise and suppressed testimony remains stable.
This helps explain why factual correction often produces less change than expected. Correcting a proposition addresses the content of a communicative event. It does not necessarily alter the organisation through which that proposition acquired significance.
A correction may even reinforce the system when institutional denial has already been assigned the role of concealment. Evidence that would weaken an isolated proposition can strengthen the relation between persecuted truth and corrupt authority.
This does not mean correction is useless. It means its effects depend upon the communicative organisation into which it enters.
A correction is more likely to succeed when it alters relational conditions: when trusted participants carry it, when it offers an alternative explanation, when it preserves social identity, when it changes future expectations or when it weakens the roles through which contrary evidence is automatically absorbed.
Persistent disinformation is therefore not adequately described as false content surviving transmission. It is relational reproduction across changing communicative events.
7. Technology and Relational Topology
Digital platforms do more than accelerate or amplify communication. They alter the topology through which communicative systems form and persist.
Recommendation systems influence which messages become adjacent. Search systems determine which claims are encountered together. Ranking mechanisms establish temporal priority and visibility. Metrics create public evidence of attention. Group features stabilise identities and boundaries. Reposting functions allow messages to recur across contexts without preserving their original meaning.
These systems restructure relations among participants, messages and institutions. They reduce some distances, increase others, create new forms of recurrence and alter the temporal rhythm through which communication develops.
A communicative pattern that once required a local organisation, publication or recurring meeting can now maintain continuity across geographically dispersed participants. Archives preserve earlier messages for later reuse. Recommendation systems can repeatedly return participants to similar interpretive environments. Metrics can make visibility appear equivalent to credibility or consensus.
Technology also permits rapid migration. A communicative organisation removed from one platform may persist through other platforms, private groups, video channels, newsletters or interpersonal networks. Content moderation can reduce reach without dissolving the relations through which the system reproduces itself.
This produces a testable prediction: interventions directed only at visible messages will often displace persistent communicative systems rather than eliminate them. Elimination is more likely when intervention disrupts the relations connecting identity, authority, recurrence and recruitment across platforms.
Platforms also affect disappearance. Changes to recommendation systems, moderation policies or access can deprive a communicative organisation of recurrence. Yet decline will depend upon whether the pattern has developed alternative infrastructures and durable social relations beyond the platform.
Technology is therefore neither a neutral channel nor an autonomous cause. It participates in the organisation of communicative possibility.
8. Implications and Predictions
Treating communicative persistence as the primary object of analysis produces several implications.
First, message popularity should not be treated as a reliable measure of persistence. Some highly visible messages will disappear quickly, while less visible communicative organisations may endure.
Second, systems with strong participant identities, replaceable claims and adaptable interpretive rules should survive factual correction more successfully than systems organised around a single proposition.
Third, interventions targeting individual claims should have limited lasting effect when the broader organisation of trust, identity and interpretation remains intact.
Fourth, communicative systems should be more likely to persist when they can convert contradictory evidence into a recognised internal role, such as censorship, persecution or concealment.
Fifth, platform removal should produce migration rather than disappearance when the organisation already possesses cross-platform relations, stable identities and alternative means of recurrence.
Sixth, communicative collapse should become more likely when a system loses its capacity to generate new interpretations, recruit participants or connect present events to future expectation.
These propositions can be investigated empirically. Researchers could track whether relational roles persist across changing claims, examine the replacement of participants and platforms, measure the incorporation of new events, and compare systems that survive correction with those that disappear.
A comparative study might examine two rumours with similar initial reach. One may vanish after several days; the other may develop durable identities, trusted sources, explanatory rules and expectations. The relevant question would not merely be which message spread further, but which relations were reproduced strongly enough to organise future communication.
This reframes intervention. Lasting change requires more than replacing false messages with true ones. It requires altering the relations through which future messages acquire meaning.
9. Implications for Communication Theory
The argument extends beyond disinformation. Meaning, identity, institutions and collective memory can all be understood as effects of organised communicative continuity.
Individuals and organisations remain recognisable despite continual change because communication reproduces the relations through which they remain intelligible. Institutions persist because procedures, authority, expectations and obligations are repeatedly reconstructed through communication. Scientific disciplines survive the replacement of theories and researchers because their methods, problems, standards and histories retain organised continuity.
These phenomena need not be treated as separate theoretical problems. They are different expressions of the same underlying dynamic: recurrent communication reorganising the conditions of its own continuation.
This reverses the usual analytical priority. Rather than beginning with messages and asking how continuity emerges, analysis can begin with organised continuity and ask how individual messages participate in its reproduction, transformation or collapse.
The communicative system is the persistent object. Individual communications are the transient interactions through which it reproduces its organisation.
10. Conclusion
This paper began with a question that message-centred explanations do not fully resolve: why do some communicative patterns persist despite continual variation in content, participants and circumstances, while others disappear almost immediately?
The answer proposed here is that communicative persistence does not reside primarily in individual messages. Messages are transient events within longer processes of organised communication. What endures is the relational organisation through which successive events remain mutually intelligible, reinforce established expectations and modify the conditions under which further communication becomes possible.
Persistent disinformation makes this structure unusually visible. Particular claims may be corrected, abandoned or replaced while the organisation that made them credible and consequential remains substantially intact. New claims can occupy the same relational positions, reproduce the same distinctions and sustain the same patterns of interpretation.
Other communicative patterns disappear because they fail to establish such continuity. They may diffuse widely but remain unable to generate identities, expectations, interpretations or future communication. Visibility is not persistence. Repetition is not reproduction.
The theory therefore explains not only how communicative systems persist, but how they emerge, transform, divide, merge and collapse. It also clarifies why technologies matter: they reorganise the relational and temporal conditions under which communicative recurrence becomes possible.
This does not displace research on cognition, diffusion, political communication or platform governance. It situates those processes within a broader account of communication as organised continuity across time.
The principal task is no longer merely to explain how messages move between communicators. It is to explain how communicative systems reconstruct the relations required for their own continuation—and why, under different conditions, they fail.
References
Austin, J.L. (1962) How to Do Things with Words. Oxford: Clarendon Press.
Bak-Coleman, J.B., Alfano, M., Barfuss, W., Bergstrom, C.T., Centeno, M.A., Couzin, I.D., Donges, J.F., Galesic, M., Gersick, A.S., Jacquet, J., Kao, A.B., Moran, R.E., Romanczuk, P., Rubenstein, D.I., Tombak, K.J., Van Bavel, J.J. and Weber, E.U. (2021) ‘Stewardship of global collective behavior’, Proceedings of the National Academy of Sciences, 118(27), e2025764118. doi: 10.1073/pnas.2025764118.
Bateson, G. (1972) Steps to an Ecology of Mind. San Francisco: Chandler Publishing Company.
Blumer, H. (1969) Symbolic Interactionism: Perspective and Method. Englewood Cliffs, NJ: Prentice-Hall.
Khintchine, A. (1934) ‘Korrelationstheorie der stationären stochastischen Prozesse’, Mathematische Annalen, 109, pp. 604–615. doi: 10.1007/BF01449156.
Lewandowsky, S., Cook, J., Ecker, U.K.H., Albarracín, D., Amazeen, M.A., Kendeou, P., Lombardi, D., Newman, E.J., Pennycook, G., Porter, E., Rand, D.G., Rapp, D.N., Reifler, J., Roozenbeek, J., Schmid, P., Seifert, C.M., Sinatra, G.M., Swire-Thompson, B., van der Linden, S., Vraga, E.K., Wood, T.J. and Zaragoza, M.S. (2020) The Debunking Handbook 2020. George Mason University, Center for Climate Change Communication.
Luhmann, N. (1995) Social Systems. Translated by J. Bednarz Jr, with D. Baecker. Stanford, CA: Stanford University Press.
Maturana, H.R. and Varela, F.J. (1980) Autopoiesis and Cognition: The Realization of the Living. Dordrecht: D. Reidel. doi: 10.1007/978-94-009-8947-4.
Mead, G.H. (1934) Mind, Self, and Society: From the Standpoint of a Social Behaviorist. Edited by C.W. Morris. Chicago: University of Chicago Press.
Searle, J.R. (1969) Speech Acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press.
Shannon, C.E. and Weaver, W. (1949) The Mathematical Theory of Communication. Urbana, IL: University of Illinois Press.
Vosoughi, S., Roy, D. and Aral, S. (2018) ‘The spread of true and false news online’, Science, 359(6380), pp. 1146–1151. doi: 10.1126/science.aap9559.
Wardle, C. and Derakhshan, H. (2017) Information Disorder: Toward an Interdisciplinary Framework for Research and Policy Making. Strasbourg: Council of Europe.
Wiener, N. (1930) ‘Generalized harmonic analysis’, Acta Mathematica, 55, pp. 117–258. doi: 10.1007/BF02546511.
Wiener, N. (1948) Cybernetics: Or Control and Communication in the Animal and the Machine. Paris: Hermann & Cie; Cambridge, MA: MIT Press.
Wittgenstein, L. (1953) Philosophical Investigations. Translated by G.E.M. Anscombe. Oxford: Basil Blackwell.
Daedelus Kite 