What Does Not Exist
Can Move the World

The disaster that did not happen

Imagine that the city receives a report about a possible flood. The document does not announce a certainty. It describes a scenario dependent on rainfall, maintenance, and vulnerabilities in the terrain. A decision is made to reinforce a defense, change a route, and conduct an evacuation drill. Months later, no catastrophe occurs.

Someone points to the budget and asks why so much was spent on something that did not happen. Another person replies that it did not happen precisely because of that spending. Both explanations are possible, but neither is proved by the absence of the disaster alone. Perhaps the intervention prevented harm. Perhaps the dangerous conditions never arose. Perhaps the model exaggerated a risk. More information will be needed to distinguish between the hypotheses.

This is the defensible meaning of the title. It does not claim that a nonexistent event exerts a physical force. It claims that systems capable of representing absent situations can organize their behavior around them. The bearer of the representation exists now; its content may refer to something that is not happening, is not yet happening, or will never happen.

Imagining, predicting, wanting, and promising are different acts

The city can relate to the same scenario in different ways. A designer imagines it to explore alternatives. An analyst calculates its probability. An authority proposes bringing it about. A group commits to trying. A campaign presents it as if it were guaranteed. The image may look identical while its function changes radically.

An imaginary company that declares “this will be our future” may be describing an aspiration, not a forecast. Failure to materialize does not make the aspiration false in the same way as a failed categorical prediction, but it can be irresponsible if it was presented as a guarantee to obtain resources or commit others. Its evaluation also depends on the relationship it established.

Likewise, a promise is not a simple prediction of the promisor’s behavior. It can generate an obligation even when future behavior remains uncertain. Treating it as a probabilistic estimate removes the dimension of commitment; treating every expectation as a promise attributes duties that were never undertaken.

A shared future can enable coordination without being guaranteed

An individual representation changes its practical nature when other people adjust their actions to it. Imagine a workshop that agrees to produce an object that does not exist. Each participant performs one part because they expect the other parts to happen as well. The final object does not materially organize the work from the future. Present plans, agreements, expectations, and resources do so, coordinated around its representation.

Beckert analyzes fictional expectations in the economy as orientations that enable action under uncertainty. “Fictional” does not automatically mean false or deceptive; it points to the role of imaginings about future states that cannot be treated as fully known outcomes. This prior work is central: the idea that represented futures organize economic decisions is not an invention of this trilogy (Beckert, 2013).

The requirement we propose adding is to trace the relationship between representation and decision. In the flood report, we would need to separate observations from assumptions, identify which actions risk reduction depends on and recognize conditions that might change even without intervention. It also matters who needs to accept or execute part of the plan. That reconstruction allows effects to be attributed without treating the scenario as a guarantee and makes it possible to evaluate what happens when its circulation changes many participants’ behavior.

The last question changes the problem. A forecast can influence what it aims to describe. Merton examined the self-fulfilling prophecy through the way an initially false definition of a situation can trigger behavior that brings that situation about. His banking example is a parable, not a historical episode we should present as documented (Merton, 1948).

The lesson is not that believing anything makes it true. Mechanisms and conditions are required: the capacity to coordinate actions, the situation’s sensitivity to those actions, and the absence of barriers that would prevent the outcome. A crowd can cause a specific shortage by behaving in a certain way; it cannot create any physical resource merely by increasing its conviction.

The opposite can also happen. A warning can trigger an intervention that prevents the anticipated event. In that case, an outcome that is not observed may be compatible with effective prevention. Nonfulfillment does not in itself prove initial falsity, just as fulfillment does not prove that a prediction was independent of its effects.

The room where it is decided which future will be visible

Now consider a media planning room. This is not a scene attributed to a real campaign. Three proposals appear on the screen: one shows people gaining time, another promises recognition, and a third presents security against a threat. The products may be similar; the personal futures each proposal makes visible are different.

Whoever designs the communication does not literally move people from a time that does not yet exist. They select present representations that could enter into people’s evaluations. The result depends on interpretation, context, credibility, material conditions, and many other relationships. We must not assume that exposure equals persuasion, or that persuasion equals action.

Responsible communication should distinguish a verified benefit from a conditional possibility, an aspiration from a guarantee, and an illustrative representation from a typical result. The aesthetic force of a future does not replace its justification. The more a narrative participates in costly or hard-to-reverse decisions, the more that separation matters.

The efficacy of a fiction does not demonstrate its truth

A false rumor can change behavior. An invented story can modify a commitment. A manipulated image can prompt a decision. In each case, efficacy belongs to a present causal relationship; it does not automatically validate the content.

Nor should we fall into the opposite extreme: “if every represented future is a construction, all promises are worth the same.” Uncertainty admits degrees, constraints, and tests. One scenario can be better grounded than another, even when both describe something unobserved. The difference lies in their assumptions, mechanisms, compatibility with evidence, sensitivity to change, and willingness to be corrected.

The criterion does not require eliminating all imagination from public communication. It requires that imagination not be granted the credentials of a demonstration because of its capacity to move us. Nor does it require every decision to await absolute certainty. It requires proportionality between what we know, what we claim, and what we ask others to risk.

What absence leaves behind

Let us return to the city without a flood. The defenses are there. The money has already been spent. The residents have changed a routine. The anticipated scenario did not materialize, but the history of its representation is part of what happened.

The causal relationship passes through present organizations acting according to differences among possible scenarios. Within them, represented content may help explain why one measure rather than another was adopted. That explanation must show how the content operated alongside sources, incentives, capacities and material conditions. An absent flood does not approve a budget; people and institutions do so guided by a situation they represent. Understanding that orientation allows us to examine both what the representation opened up and the alternatives it left out.

And when those systems coordinate others, futures acquire a public life before becoming events. They can be discussed, contested, manipulated, financed, or rejected. Our responsibility begins at that stage, not only when the project has already been built and its effects seem inevitable.

What has not yet happened can guide what we do. Precisely for that reason, we must demand more from the futures we are offered than the capacity to convince us.