Categories
cybernetics

Before the Message: A Relational Account of Persistent Disinformation

Disinformation does not persist because individual false claims survive. It persists because the communicative relations that make those claims meaningful continue to reproduce themselves.

Abstract

A claim can disappear while the relations that made it credible, contentious or worth repeating continue to produce new claims. Persistent disinformation is a case of this wider problem of communicative continuity. Relations precede the identities they distinguish, and communicative activity reconstructs the conditions through which those identities persist. This account follows recurrence, delay, meaning and mutual influence through changes in claims, participants and platforms. It connects those dynamics to observable sequences of interaction and to the provisional effects of intervention.

1. Relations Before the Message

A prediction fails. Some participants abandon it. Others take the failure as evidence that somebody intervened. The next claim can concern an entirely different event and still renew the same relations of trust, suspicion and concealed agency. The failed prediction has changed the conditions in which its replacement will be heard.

Before concerns logical priority. A statement acquires a role through relations among what has already been said, who is expected to answer, which sources carry authority and what remains unresolved. An absent reply can be consequential. So can the interval between a claim and its correction. These differences and dependencies are the conditions through which a message, a speaker and an audience become distinguishable in the first place.

In Applied Field Logic, identities are provisional continuities of relation. A participant becomes recognisable as a witness, critic or authority through recurring interactions that give those positions practical consequences. Each intervention in the exchange inherits some of those relations and changes what can follow. The same problem of identity through change is developed in How Things Persist.

2. The Problem of Persistence

Shannon and Weaver (1949) isolate the engineering problem of transmission. Wittgenstein (1953) examines meaning through use, and Bateson (1972) locates information in consequential difference. Luhmann (1995) treats communication as the operation through which social systems reproduce themselves. The question here is which relations carry that reproduction through a change of words, participants or medium, and how we can distinguish their continuity in observation.

A rumour can travel widely and disappear when attention moves elsewhere. A recurrent account of concealed agency can survive the loss of particular allegations because its participants continue to recognise sources, anticipate objections and interpret new events through related distinctions. Diffusion concerns the passage of a communication. Reproduction concerns the changes it makes to the conditions of further communication. Reach measures one part of that history.

Begin by identifying a relation that could be followed across changing expressions. Does a new claim call on the same sources of authority? Does criticism repeatedly return as evidence of concealment? Do those responses generate further claims? The relevant evidence is the sequence of dependencies through which one exchange makes another possible. A repeated phrase supplies one trace of that sequence.

3. Recurrence, Delay and Harmonic Structure

Recurrence concerns what returns. Recursion concerns how the consequences of activity enter into the conditions of its next occurrence. An accusation can provoke a defence, which changes the terms of a subsequent accusation. The intervals between those responses affect who encounters them, which evidence is available and which interpretations acquire force. Timing belongs to the relationship itself.

Take the sequence in the opening example. Identify recorded exchanges in which a challenge to a claim is followed by a response treating the challenge as concealment. Record which earlier communication each response addresses, when it occurred and when the response followed. Use the same criteria when the subject, wording or participants change. Counting those responses in successive equal intervals gives a time series, Xn, where n indexes the intervals. The recorded links and delays remain available alongside the count.

Autocovariance compares deviations from the mean at two times separated by a lag. When the mean μ is constant and the covariance depends on the lag alone, C(k) gives the autocovariance at lag k.

C(k) = E[(Xn − μ)(Xn+k − μ)]

Here E denotes the expected value. For a stationary series in this second-order sense, with absolutely summable autocovariance, the Fourier transform of C(k) gives the power spectral density of fluctuations about the mean (Wiener, 1930, and Khintchine, 1934). The spectrum expresses the distribution of variation across frequencies. The lag description expresses how activity varies together across intervals. A changing series calls for comparisons across periods and explicit attention to when the stationary model is adequate. See recurrence and frequency.

A rhythm in the count could arise from the ordinary rhythm of participation, a common news event or the repeated interaction being studied. Compare the count with overall activity and examine the recorded sequences before attributing a cause. A claim about mutual influence requires evidence of how each process changes in response to the other.

Where a recurring cycle can be identified, phase specifies position within it. A sustained phase difference allows activities to remain distinguishable while keeping a characteristic timing relationship. Coupled oscillator models establish that such relationships can emerge through mutual influence under specified conditions (Kuramoto, 1984). In communication the corresponding questions are which cycles can be measured, which delays carry effects between them and how those relations change across timescales. Harmonic structure concerns this pattern of recurrence and timing across scales.

4. Transformation and Disappearance

A communicative pattern becomes self-propagating when its activity reconstructs the relations that make further activity of that kind more likely. A failed prediction can redirect attention to a different source, change expectations about future evidence or give new significance to an old grievance. Each of these changes affects the possibilities available to the next exchange.

The logical orbit names the constitutive incompleteness of a local configuration. A particular claim draws its force from dependencies that exceed what it expresses, including histories of trust, available channels and unresolved expectations. Recursive displacement is the dynamical expression of this dependence. Each recurrence changes the conditions through which another recurrence becomes possible.

In a probabilistic description, Sn denotes the selected relational configuration at step n. The transition rule Pn assigns probabilities to possible successors given that configuration. The current rule governs the transition, and the transition updates the rule.

Sn+1 ∼ Pn(· | Sn)
Pn+1 = Φ(Pn, Sn, Sn+1)

The first line says that the next configuration is drawn according to the current transition probabilities. In the second, Φ gives the rule by which the realised transition changes those probabilities. A quantitative application specifies the alternatives, the history retained in the description, the update rule and the evidence for it. Judging whether an interaction favours continuity requires a criterion of persistence and a comparison under comparable conditions. See recursive probability.

Dependencies can be renewed, divided, joined with other patterns or lost. A faction can develop its own sources and expectations from a shared history. Previously distinct accounts can reinforce one another through a common relation of suspicion. A pattern disappears when the relations that regenerate it cease to be sustained. Similar configurations at two times leave open whether those relations continued through the intervening period or were subsequently reconstructed.

5. Meaning and Relational Invariance

The same sentence can reassure, accuse or amuse as its context changes. Different sentences can perform a recognisably similar function in an exchange. Repeated use makes some interpretations familiar. Those interpretations then affect what is said next, who is expected to respond and which responses are taken seriously. Meaning develops through these relations and participates in their further recurrence.

Meaning is invariance under transformation. A paraphrase preserves meaning where the relations that give the expression its significance are preserved. An exact mathematical invariant has the same value under each specified transformation. Assessing continuity in observed communication requires a criterion for recognising a pattern through variation. A role can pass to a new participant, divide into several roles or acquire a different significance. The question is which dependencies remain consequential through that change. See relational invariance.

Trust, authority and opposition can persist in this way. Participants can change their beliefs and positions while the exchange continues to assign familiar functions to claims and responses. The recurring distinction between trusted witness and suspected authority is itself produced and revised through these encounters. Its continuity is an achievement of their relations.

6. Persistent Disinformation

Judging accuracy and attributing deceptive intent require evidence about a communication and its circumstances (Wardle and Derakhshan, 2017). Explaining persistence requires evidence about relations across successive acts. The same exchange can contain deliberate deception, mistaken repetition, accurate quotations and criticism. Each of these acts participates in the timing and feedback through which the exchange develops.

In the failed-prediction example, a correction supplies evidence about what happened. Its consequences for the wider exchange depend on who receives it, how it is interpreted and whether it changes trust or subsequent expectations. A participant can revise a belief. Another can take the same correction as confirmation of concealment. Opposition in meaning can feed the same sequence of replies and renewed allegations. These effects are questions for observation. See frequency domain intervention.

The distinction between a claim and the relations sustaining its recurrence gives factual correction a more precise place. A correction that changes who counts as credible, which evidence is shared or which response follows can also change the conditions that generate replacements. Those conditions explain why effects differ across audiences and occasions.

7. Technology and Relational Topology

Ranking, search and recommendation change which communications encounter one another and in what order. An archive makes a previous allegation available for reuse. A recommendation can bring a response to people who missed the exchange that prompted it. These functions alter the topology of communication, the pattern of routes through which participants, claims and responses become connected. Changing access to a platform can weaken a pattern, redistribute it or leave other routes of recurrence available.

A reply network, a record of timing and an account of meaning each distinguish relations within the same activity. Their categories arise through the practices, interfaces and observations that make those relations visible. Comparing them requires showing how recorded links, temporal sequences and interpreted roles correspond under the transformations being studied. A change in one description prompts an inquiry into what changed in the others. This is part of the many spaces problem.

8. Implications and Predictions

Compare two rumours with similar initial reach. The account predicts longer persistence where later exchanges regenerate trusted sources, recognised roles and expectations that draw in new claims. Specify these relations before examining the later history. Record whether they recur across changes of wording and participation, and compare with cases where attention falls away. This gives a test of persistence beyond a count of repetitions.

Follow corrections through their replies and subsequent use. The account predicts different effects where a correction changes the relations of trust and interpretation than where the same relations are reproduced in responding to it. Track the intervening exchanges. A change in reach, a shared news event and a change in the measured relation provide competing explanations to examine.

When a platform removes content, trace whether sources, roles and response sequences remain connected through other channels. Continuity requires evidence of the dependencies that carry activity across the move. A later appearance of similar content also permits an explanation in terms of renewed activity after a break. The distinction matters to any assessment of what the intervention achieved.

9. Intervention and Provisional Control

Control is always provisional. Changes in pacing, visibility or access modify which exchanges occur and when their consequences become available. The participants, platforms and institutions involved adapt to those changes. An intervention enters the same field of feedback that it seeks to influence.

An institution’s classifications and procedures also become conditions of subsequent communication. A designation changes what receives attention and how people respond. Its consequences must enter the next assessment. Trace the time spent verifying claims, the demands placed on participants and the work shifted to other channels. Persistence and disruption both distribute costs across the relations involved.

An information entropy measure requires a specified distribution of possible outcomes. Physical entropy production requires an account of physical processes and exchanges. In this inquiry, the immediate work is to identify which possibilities are constrained, whose work sustains the constraint and how these conditions change after intervention.

10. From Theory to the Handbook

The handbook introduction sets out the sequence from this account of communicative persistence to formal description, observation and intervention. The same question runs through each stage. Which relations are being reproduced, and how does their reproduction change what becomes possible next?

Key takeaways

  • Follow the relations through which changing claims become consequential.
  • Measure recurrence with a defined signal, recorded delays and a comparison against ordinary activity.
  • Assess accuracy, intention and persistence with evidence suited to each question.
  • Evaluate interventions through the dependencies and costs they change.

References

  • Bateson, G. (1972) Steps to an Ecology of Mind. San Francisco: Chandler Publishing Company.
  • Khintchine, A. (1934) ‘Korrelationstheorie der stationären stochastischen Prozesse’, Mathematische Annalen, 109, pp. 604–615.
  • Kuramoto, Y. (1984) Chemical Oscillations, Waves, and Turbulence. Berlin: Springer.
  • Luhmann, N. (1995) Social Systems. Translated by J. Bednarz Jr, with D. Baecker. Stanford, CA: Stanford University Press.
  • Shannon, C.E. and Weaver, W. (1949) The Mathematical Theory of Communication. Urbana, IL: University of Illinois Press.
  • 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.
  • Wittgenstein, L. (1953) Philosophical Investigations. Translated by G.E.M. Anscombe. Oxford: Basil Blackwell.

Comments Closed

Your email address will not be published. Required fields are marked *