Compartmentalization and the Structure of Classified Power
Why secrecy systems are designed to prevent full internal visibility
Secrecy is usually defended as a way to keep sensitive information from enemies. That defense is real, but incomplete. Mature secrecy systems do more than conceal information from outsiders. They regulate who can see what inside the state itself. Once that happens, secrecy ceases to be only protective. It becomes a method for organizing authority by controlling visibility.
That distinction matters because confidentiality and compartmentalization are not the same thing. Confidentiality withholds information while preserving internal intelligibility. The relevant institution can still know what it is doing, who authorized it, and how it fits within lawful purpose. Compartmentalization operates differently. It fragments knowledge into restricted channels, limits integrated understanding, and prevents most participants from seeing purpose, scale, and consequence together. The issue is not simply that information is hidden. The issue is that internal visibility is broken apart by design.
One of the clearest examples is the intelligence budget. The public may be shown a large topline number while being denied meaningful detail about the programs, projects, and activities beneath it. Oversight in the strong sense requires more than awareness that money is being spent. It requires enough visibility to judge what is being funded, at what scale, under what authority, and toward what durable end. When that visibility is withheld, budget control remains in form while weakening in substance.
This is where the link to War and Budgetary Expansion becomes structurally important. That essay argues that modern conflict often fails in the language used to justify it while succeeding in the durable structures that surround it, especially appropriations, readiness claims, and institutional continuity. Classified budgeting helps explain how that continuity can survive even where public scrutiny intensifies. A war can fail publicly and still strengthen the institutions managing it. If the fiscal architecture behind those institutions is only partly visible, challenge becomes harder not because spending disappears, but because meaningful evaluation of its object becomes institutionally difficult.
The historical model for this architecture is the Manhattan Project. Its importance here is not merely that it was secret. It is that it proved the effectiveness of extreme compartmentalization. A project of vast scale was organized so that many participants understood only their immediate task while very few could see the whole. The modern secrecy state did not invent this logic later. It demonstrated it early: divide knowledge sharply enough, and large systems of power can operate with most participants unable to judge the full structure they are serving. Special access structures push this logic further by ensuring that clearance, rank, and formal office still do not guarantee visibility.
Once that model exists, rank no longer guarantees visibility. Formal office may remain intact, but substantive knowledge depends on controlled access. That is the point at which a second structure emerges inside the visible state: not a theatrical hidden government, but a narrower system in which who can know is determined less by constitutional position than by admission into the relevant compartment. That is what classified power means in practice. Authority still speaks through the visible hierarchy, yet integrated sight is reserved to a far smaller circle.
This is also why the new essay sits naturally beside The Military-Industrial Complex and the Persistence of Secrecy. That essay argues that defense structures preserve programs and authority after public scrutiny fails. The mechanism examined here is one of the ways that preservation occurs. Exposure may happen. Hearings may happen. Public controversy may happen. Yet if the structure itself prevents full internal visibility, correction becomes harder because the knowledge required for full correction is not broadly available even within the institutions nominally responsible for supervision. Secrecy then functions not merely as protection from the outside, but as a survivability mechanism within the system itself.
The companion essay Classification and the Limits of Public Accountability explains the outer boundary of this problem: states can hide actions they could not publicly defend. The present mechanism goes further inward. What happens when concealment is not only directed against the public, but organized inside the state? Public accountability weakens because the facts are classified. Internal accountability weakens because the facts are compartmentalized. The first restricts outside judgment. The second degrades inside judgment. Together they produce a structure in which authority becomes harder to trace, harder to evaluate, and harder to restrain.
This is why euphemism has to be avoided on hard examples. MKUltra was not mere drug testing, nor was it a vague controversy about research ethics. It was a covert CIA program involving mind-control and interrogation experimentation, including experiments on unwitting human subjects. That is the correct level of description. A system justified as protective carried out secret experimentation on people who had not consented. The importance of MKUltra here is not only moral. It is structural. It showed that once a program is hidden deeply enough, illegality and abuse can persist long after ordinary institutions should have stopped them.
Once knowledge is fragmented that severely, responsibility fragments with it. Each participant can understand only the immediate task. Each office can certify only a narrow segment. Each reviewer can be shown only part of the picture. Conscience is not removed, but the conditions under which conscience can form into informed institutional resistance are weakened. People may suspect that something is wrong while lacking the integrated visibility needed to name the problem clearly enough to stop it. That is not a side effect. It is one of the system functions of compartmentalization.
The usual defense is that secrecy is necessary because threats are real and exposure can cause harm. That can be true and still leave the central constitutional question untouched. If meaningful review depends on visibility, and visibility is withheld from all but a minute circle, how is restraint supposed to function? At that point oversight risks becoming ceremonial. Institutions can say rules were followed, budgets were approved, and channels were observed while lacking the integrated understanding required to judge whether the underlying activity should exist at all. Procedure remains visible. Substance recedes.
That is the deeper significance of compartmentalization. It does not simply keep a secret. It structures who can know, who can judge, and therefore who can restrain. Where that architecture becomes entrenched, the visible state continues to speak in the language of law, supervision, and democratic control. But beneath that language a narrower structure can persist in which access, rather than office, determines who can see; fragmentation, rather than hierarchy, determines who can judge; and continuity is preserved not because the system has been vindicated, but because the design of secrecy has made full evaluation institutionally difficult. That is the structure of classified power.