Essay 02 · On Power over Visibility
Who Has the Right to Define Reality?
Why a dispute about power begins before a command or a vote—with the authority to name the problem, build the category, and decide whose bad news will be heard.
A report lies on the conference table: 92 percent of deadlines were met, complaints were almost nonexistent, and the project was judged a success. The numbers may be accurate. Yet a series of decisions already occurred before the numbers appeared: what counts as a deadline, which messages count as complaints, whom to ask, and which department lies outside the project's boundary. Perhaps the contractor closed tickets before repairs were actually complete, while dependent employees did not complain to the person who controlled their bonuses. “Is the report true?” cannot be separated from “Who designed the way of seeing?”
Power is often imagined as a reserve of force held by a superior: one gives an order and the rest comply. This picture notices coercion but misses earlier mechanisms. Someone defines the agenda before the vote, the language before the argument, and the price of refusal before consent. A category such as “successful,” “unreliable,” or “inefficient” does not merely describe a person; it changes their access to money, voice, and future opportunities. Once such decisions become habitual, a separate command may no longer be needed.
Yet it is also insufficient to see only oppression in every shared category. People cannot act together without a common description of the situation, a division of roles, and rules. Hannah Arendt calls power the human ability to begin and sustain action in common. Michel Foucault reveals power embedded in norms, knowledge, and everyday procedures. James C. Scott studies how a center simplifies a complex society in order to govern it. Pierre Bourdieu explains why voices carry unequal weight before an argument has even been made.
These views should not be reconciled too quickly. Each sees what the others may leave in shadow. Arendt protects the creative force of acting together from reduction to violence. Foucault prevents us from mistaking voluntariness for the absence of discipline. Scott shows the price of scale, and Bourdieu the hidden history of unequal positions. Together they lead to a more exact question—not “Is power present?” but “What kind of power is operating, how does it make the world visible, who can correct its description, and what sustains its right to decide?”
The agenda allocates reality before votes are cast
Visible power wins an argument. Less visible power decides whether the argument will happen at all. Every meeting has limited time, so an agenda is unavoidable: not everything can be discussed. But the criterion of selection is never empty technique. If a council considers only the cost of repairs, residents' safety appears as an “emotional addition.” If a school discusses average scores but not the students excluded from testing, part of the result disappears before analysis. What receives no line, no time, and no responsible owner has almost no influence on the decision.
A missing question is not necessarily suppressed by conspiracy. It may be new, poorly formulated, or genuinely less important. Suspicion alone is therefore insufficient. Traces of power appear where the same difficulty is systematically excluded, affected people lack access, and raising the question carries a special cost. Compare two cases: a proposal is postponed with a public reason and a date for its return; a proposal is called “irrelevant” for years, with no opportunity to change the criterion of relevance. Only the second procedure protects itself from discussion.
An agenda can be tested without turning a meeting into an endless forum. The rule for including questions can be public; a brief space can be reserved for what has been missed; rejected topics and reasons can be listed; a minority can have a right to return a question when new evidence appears. Albert Hirschman distinguished exit from voice: the ability to leave constrains a bad institution, while the ability to speak helps correct it. But voice exists not when someone may utter words, but when someone is obliged to hear, answer, and revise the decision when the grounds are sufficient.
CrystallizationPower over the agenda is the ability to turn a disputed interest into a topic that appears unworthy of dispute itself.
A category makes people governable—and changes their possibilities
Scott calls the production of legibility an important task of states and large organizations. Living differences are translated into standardized names, addresses, occupations, diagnoses, and measures. Without this, benefits cannot be assigned, water quality compared, or equal rights secured. A category is not an evil in itself; it is an instrument of scale. The error begins when the center's convenience is mistaken for the structure of life itself, and everything that does not fit is treated as noise, deviance, or the person's fault.
Consider a benefit rule for the “unemployed.” It can easily count a person without an employment contract, but may fail to see unpaid care for an ill relative, seasonal labor, or a health condition that has not been formally diagnosed. An official can apply the category with perfect accuracy and still make an unjust decision. The problem lies not only in the official's character, but in the translation: a complex situation has been reduced to the features available to the system. If a person cannot supplement the record with an explanation, administrative visibility becomes administrative voicelessness.
A category grows more powerful when resources are tied to it and when it begins producing the behavior it seems merely to measure. A school divides children into “able” and “weak,” then gives the first group difficult tasks and attention while assigning the second a simplified program. A year later the difference has grown and appears to confirm the original label. To avoid such a loop, ask: does the category describe a durable trait or a temporary position; what decisions does it trigger; can a person leave it; and where is the original experience preserved after compression?
CrystallizationA category acquires power when resources depend on it and the person affected cannot inspect, supplement, or challenge their description.
Bourdieu: arguments enter the room attached to unequal positions
Pierre Bourdieu invites us to see society as a set of fields: science, art, politics, education, business. Each field has its own stakes and forms of recognition. Money does not matter equally everywhere; credentials, connections, a familiar manner of speech, and reputation can be converted into influence, but the exchange rate varies. Participants therefore do not meet as pure minds. One is already recognized as an expert, another must first prove a right to speak, while a third carries experience for which the field has no prestigious language.
This helps explain symbolic power: the capacity to make a particular classification appear natural. One person's pronunciation is heard as education and another's as deficiency; a senior executive's confidence is read as competence, a junior employee's caution as weakness. The same argument therefore receives different weight depending on who carries it. Such power rarely operates as a conscious conspiracy. Participants absorb the rules, tastes, and expectations of the field, then sincerely recognize “quality” where they have learned to see it.
But a theory of position should not turn a person into a puppet of their origins. Not every expert defends status, not every newcomer brings a fresh truth, and fields change through coalitions, new standards, and unexpected achievements. The practical conclusion is more modest: assess not only the content of speech, but its route to listeners. Useful measures include blind review where possible, the right of affected people to present their experience, translation between professional and ordinary language, and examination of whose error is forgiven and whose becomes proof of unfitness.
CrystallizationA formally equal right to speak does not create an equal voice when status determines in advance whose words count as knowledge.
Bad news must be able to travel without demanding heroism
The higher a leader sits, the less reality they observe directly. Between event and decision stand forms, measures, managers, and presentations. Every intermediary has a reason to improve the message: not to look incompetent, to protect a team, to meet a target, to preserve a relationship. The report grows more attractive at every level even if no one orders anyone to lie. Power makes a leader dependent on translators, while also giving them the ability to punish an accurate translation.
“Do not be afraid to tell the truth” is nearly useless when the messenger's career depends on the recipient of the news. Amy Edmondson shows the importance of psychological safety: the ability to ask a question, admit an error, and express disagreement without unacceptable interpersonal risk. This is not comfort or the abandonment of standards. In a safe team, work can be criticized rigorously; reporting a problem cannot be turned into proof that the person does not belong. The point is to separate the bearer of the signal from the signal itself.
A reliable path for bad news is designed as infrastructure. It needs an early channel that does not forfeit a bonus, an independent destination for complaints about a direct supervisor, a duty to respond, protection from retaliation, and a record of what happened after the report. An anonymous box alone is not enough: it may collect pain without changing anything. An open conversation alone is not enough either: a dependent person still calculates the cost. The quality of power appears in whether an inconvenient signal becomes a reason to correct the system or a lesson that teaches everyone else to remain silent.
CrystallizationIf truth reaches power only through one person's courage, the system is already designed in favor of convenient ignorance.
Arendt and Foucault: common action is neither pure agreement nor merely a mask for coercion
For Arendt, power arises when different people can coordinate action. It does not lie in a ruler's vault and is not identical with violence. A superior can create movement through threat, but such submission lasts only while fear does. Common power rests on support and a promise to continue a shared undertaking. This matters because it frees politics from the image of an endless struggle for control: people can create what none of them could produce alone.
Foucault prevents this account from becoming too bright. People arrive at common action already shaped by norms: they know who speaks first, what kind of body counts as normal, and which success deserves respect. Power does not only prohibit; it produces categories, skills, and ways of understanding oneself. Sincere consent therefore does not prove the absence of power. A person may voluntarily pursue the measure by which they are ranked and police themselves more severely than any guard would.
There is no need to choose between Arendt and Foucault. One meeting may contain a genuine common purpose, unequal risks of dissent, and a language the participants did not choose. A test of legitimacy must therefore go beyond counting votes. We must ask whether those affected can shape the agenda, whether the grounds of decisions are available, whether dissent is protected, whether a category can be corrected, and whether the holder of authority can be replaced. Ostrom adds an important practical insight: durable common rules emerge when users can participate in changing them and have an accessible way to resolve conflict.
CrystallizationLegitimate common action requires not the absence of power, but power that permits objection, revision of rules, and replacement of its holders.
Conclusion
The authority to define reality must remain divided and correctable
No large system can do without agendas, categories, and representatives. A demand to see everything and hear everyone at every moment would destroy the capacity to decide. The task is not to eliminate selection, but to design selection that reveals its grounds and keeps open a path of correction. Power becomes dangerous not merely because it reduces complexity, but because it declares the reduction complete and makes the cost of objection higher than the cost of its own error.
Arendt reminds us that without trust and common action, people cannot create a common world. Foucault asks which norms have already shaped their agreement. Scott shows the loss produced when life is translated into a table. Bourdieu makes us notice that the right to speak is unequally distributed even before speech. Together they grant no simple license to the center, “local experience,” or the majority. Every source of knowledge can err and exclude; what differs is its power to impose that error on others.
A responsible order therefore preserves several independent routes to reality. It compares a metric with consequences, a manager's report with a worker's experience, an expert category with a person's right to explain. It separates the report of an error from punishment, publishes the grounds for exclusion, and knows how the rule itself can be changed. This does not guarantee justice. But such an order preserves the possibility of discovering injustice before it becomes the world's only official description.
Question after readingIf a person bearing the cost of your decision says that the category itself describes them falsely, what safe and effective path would let them change not only the record about them, but also the rule that created it?
Return to the map
Chapters that examine the mechanisms separately
Foundational texts
Sources for the argument
Hannah Arendt
On Violence
What it clarifies
Power is not a leader's possession: it exists for as long as people remain capable of acting in concert.
What it does not prove
Real regimes often mix consent, dependence, coercion, and violence; a clean distinction does not describe the entire mixture.
Hannah Arendt
The Human Condition
What it clarifies
Human action arises among people who are equal and distinct; no one fully controls the story begun by their own deed.
What it does not prove
The distinctions among labor, work, and action illuminate experience, but do not form a strict empirical classification of every life.
Michel Foucault
Discipline and Punish; lectures on governmentality
What it clarifies
Power works not only through commands from above, but through norms, categories, measurements, institutions, and habitual ways of governing conduct.
What it does not prove
When power is seen everywhere, it becomes difficult to distinguish its degrees and sources or to recognize the possibility of responsible common action.
James C. Scott
Seeing Like a State
What it clarifies
To govern many people, a center simplifies reality into categories and measures; together with noise, it may erase the knowledge on which life depends.
What it does not prove
Criticism of simplification does not imply that coordination at scale, measurement, and standards are always harmful.
Pierre Bourdieu
The Forms of Capital; the theory of fields and habitus
What it clarifies
A resource has force within a field: money, connections, education, manner of speech, and recognition operate by different rules and cannot be converted into one another without cost.
What it does not prove
The language of reproduction can underestimate unexpected action, creativity, and genuine changes in the rules.
Albert O. Hirschman
Exit, Voice, and Loyalty
What it clarifies
When a system deteriorates, a person can leave, try to change it through voice, or remain out of loyalty; the availability of each response changes the others.
What it does not prove
Real people may be trapped without safe exit or voice; loyalty can be either constructive or exploited.
Elinor Ostrom
Governing the Commons
What it clarifies
Between privatization and command from above lie many durable arrangements that users of a common resource create and monitor themselves.
What it does not prove
Self-governance does not work automatically: boundaries, trust, observability, the right to change rules, and proportionate sanctions all matter.
Amy C. Edmondson
Psychological Safety and Learning Behavior in Work Teams
What it clarifies
A team learns when reporting an error, asking a question, and disagreeing do not create unacceptable interpersonal risk.
What it does not prove
Psychological safety does not mean an absence of standards, conflict, responsibility, or uncomfortable feedback.