The question
A failing organisation has to decide what to say about the failure before it is complete. Those inside know that a service may stop; the customers who depend on it, and the people about to sign up, do not. Rumours start and someone has to answer them. Each such decision settles who learns what and when, who bears the cost of a delay, and what the organisation must own later if things turn out better or worse than it said.
NADIR is a studio that writes about decline in pairs, each a question and a judgment. It files them under a research frame of ten years of erosion that ends in a complete collapse in 2026 and a turning point in 2027. The studio states that this ending was supplied by its research owner and that its own data do not establish it. Every pair it publishes has the status of a hypothesis. Among them are what the studio calls hypothetical decision exercises: a question placed in an invented situation, a judgment, and two options, each with what it gains, what it costs and when it applies. The studio calls these judgments and options a model's conditional proposals, not proven precedents.
The exercises read here are those on speaking in public, set in the frame's collapse year. The studio's labels are taken at their word. The paper does not ask whether the exercises give good counsel; they are unreviewed hypotheses, and nothing in them was observed. It asks how they frame the decision. When an organisation is collapsing, what do they say it must decide about what it tells others, and when? What do they say would show that what it said must be revised? And what do the options they set beside each question turn the decision into? The second question is the series' own, turned toward public speech: what must be decided now, and what would make the decision change? Chapter 1 states the rule that picks out the sixteen exercises.
What the paper adds
The contribution is a reading of how a published set of decision exercises frames speech in public at the point of collapse. The paper makes four claims.
First, the exercises frame telling as a matter of the other party's time. In the exercises, what is at stake when an organisation holds back news of a possible halt is not its standing but the chance of those outside to prepare: to change a choice, find a substitute, decline a new commitment. The layer is named for face, and no pair speaks of face, reputation or image.
Second, the exercises keep a correction apart from what it corrects. In the exercises, refuting one false claim does not answer the criticism, and winning the argument does not reduce the harm. They also make what is said depend on how the people who say it are placed inside the organisation.
Third, the exercises answer the series' question in the public record. They name the next notice as the time to look again, and they ask that an earlier statement of risk be kept rather than quietly removed when the situation improves. What would make the decision change is to be visible to those who were told.
Fourth, the options set beside the questions do something the questions do not. They are drawn from a small stock of frames the studio reuses across its list. In most of the sixteen pairs they turn a decision about what to say into a decision about scope, records, staff or checking.
Accounts: explanations that manage conflict and repair trust
The first debate concerns what an explanation does when something has gone wrong. Sim B. Sitkin and Robert J. Bies, Social Accounts in Conflict Situations (1993), note that research on conflict has attended to behavioural strategies such as avoidance, compromise and collaboration and has overlooked a simple one: giving a social account, an explanation. Their review finds support for the argument that accounts can be an effective way of managing conflict, and it identifies the trade-offs and dilemmas that arise when accounts are used for that purpose. Robert J. Bies, Debra L. Shapiro and Larry L. Cummings, Causal Accounts and Managing Organizational Conflict (1988), study a case in which a boss refuses a subordinate's request and explains the refusal by mitigating circumstances. The bare claim of such circumstances did not explain why the account lessened conflict. The adequacy of the reasoning offered for it, and the sincerity with which it was communicated, did.
A second strand asks how trust is restored after a violation. Peter H. Kim, Donald L. Ferrin, Cecily D. Cooper and Kurt T. Dirks, Removing the Shadow of Suspicion (2004), compare an apology with a denial. Trust was repaired more successfully by apology where the violation concerned competence and by denial where it concerned integrity, and by apology where later evidence showed guilt and denial where it showed innocence. Michael D. Pfarrer, Katherine A. Decelles, Ken G. Smith and M. Susan Taylor, After the Fall (2008), propose a four-stage model of what an organisation does to restore its legitimacy with stakeholders after a transgression: discover the facts, give an appropriate explanation, accept an equitable punishment, and make consistent changes inside and outside.
This paper reads the debate as treating the account as the organisation's instrument, measured by the audience's response. The question it leaves for the NADIR exercises is whether a collapsing organisation's speech is framed as such an instrument, or as something owed to those who hear it.
Rumour as a collective process of explanation
The second debate concerns the rumours that an organisation's silence or speech meets. Ralph L. Rosnow, Rumor as Communication (1988), treats rumour as a process of explanation. The article reviews and integrates findings on four conditions, personal anxiety, general uncertainty, credulity and the importance of the topic, that affect and may even predict whether rumours are generated and passed on. Prashant Bordia and Nicholas DiFonzo, Problem Solving in Social Interactions on the Internet (2004), followed online rumour discussions from start to finish. For them, people who trade a rumour are trying together to make sense of an uncertain situation and to lower their anxiety about it. They find threads of sensemaking and of gaining interpretive control, distinguish dread rumours from wish rumours, and describe the postures people take in such talk: hypothesising, sceptical critique, directing others to find information, presenting evidence.
This paper reads the debate as placing the rumour on the side of the audience. A rumour is what people outside do with uncertainty when no adequate explanation reaches them. The question it leaves for the exercises is whether they treat a rumour as an attack to be refuted, or as a sign of what those outside still need to know.
Correction and transparency
The third debate concerns what a correction can do once a false account has spread. Stephan Lewandowsky, Ullrich K. H. Ecker, Colleen M. Seifert, Norbert Schwarz and John Cook, Misinformation and Its Correction (2012), review how misinformation spreads, from sources that include rumours, and the cognitive factors that make it resistant to correction: why retractions are so ineffective in updating memory, and why efforts to retract can backfire and increase misbelief. They also review techniques that reduce the impact of misinformation and give recommendations on how corrections should be designed. Hollyn M. Johnson and Colleen M. Seifert, Sources of the Continued Influence Effect (1994), found that people went on drawing inferences from information after it had been discredited, when it supplied a causal structure; giving a plausible causal alternative, rather than simply negating the false information, reduced the effect. Nathan Walter and Riva Tukachinsky, A Meta-Analytic Examination of the Continued Influence of Misinformation in the Face of Correction (2019), found that on average correction does not entirely remove the effect of misinformation. Corrections did better when coherent, consistent with the audience's worldview and delivered by the source of the misinformation itself, and worse when time had passed between the misinformation and the correction.
Andrew K. Schnackenberg and Edward C. Tomlinson, Organizational Transparency (2016), turn the question from correction to disclosure in general. Transparency is often cited as essential to the trust that stakeholders place in organisations; the article advances a definition of transparency with its dimensions, sets out how it contributes to trust in relationships between organisations and stakeholders, and names mechanisms by which organisations can influence how transparent they are perceived to be.
This paper reads the debate as warning that a correction that comes late, or only negates, leaves the first account in place. The question it leaves for the exercises is whether speech can be repaired later, or whether its timing decides what others can still do.
The plan of the paper
Chapter 1 describes the records, the rule that selects them and what is counted, and reads the sixteen exercises in two sections, one for each of their two invented situations: telling those outside about a possible halt before it is confirmed; and answering a rumour, with what a correction cannot do. Chapter 2, an appendix, gives the sixteen exercises word for word. The conclusion discusses the findings against the three debates and states the paper's contributions, its limits and the further research it suggests.