The Shape of the Cage

Every system that destroys a person leaves a shape behind.

Not a smoking gun. Not a signed order. A shape — a geometry of pressure that, once you learn to recognize it, appears across documented cases, decades apart, on different continents. The names change. The architecture tends to recur.

This piece describes that architecture. It is not about any single case. It is a structural model — assembled from federal investigations, congressional inquiries, declassified intelligence directives, and international human rights findings — that describes how a person can be neutralized without anyone deciding to do it.

The common objection to any account of sustained institutional harm is: who ordered it? The answer, in many documented cases, is: nobody had to. The system can produce the outcome the way a river produces a canyon — not by intention, but by the sustained application of force along the path of least resistance.

What follows is a catalog of the forces and the paths.

A Note on Scope and Method

This article is a structural analysis, not an allegation. It compares the architecture of documented systems — the shapes they leave behind — without claiming that any two cases are equivalent in severity, intent, or moral weight. Three distinctions matter:

Systemic emergence vs. coordinated conspiracy. The model described here does not require a mastermind or a plan. It describes how institutional incentives, information asymmetries, and procedural design can produce outcomes that look coordinated without anyone coordinating them. Some of the historical precedents cited below (Stasi Zersetzung, COINTELPRO) involved deliberate coordination. The structural observation is that comparable outcomes can also emerge without it — and when they do, they are harder to identify and harder to reform.

Documented mechanisms vs. subjective experience. Every mechanism described in this article is drawn from a documented case: a government investigation, a court record, a declassified directive, or a peer-reviewed finding. Where the article describes a general pattern, the claim is that the pattern has been observed in those documented cases — not that it is universal or that any specific reader’s experience necessarily fits the model.

Evidence vs. inference. The article distinguishes between (a) mechanisms that are directly documented in primary sources, cited in the Notes; (b) structural analogies between documented cases, marked as “shape matches”; and (c) the inference that these analogies reflect recurring architectural features rather than coincidence. A reader can accept (a) and (b) while remaining skeptical of (c). The “Observable Outputs” checklist following Section I describes what a reviewer could look for to evaluate whether the architecture is present in any specific case.

I. The Neutralization Stack

The pattern, abstracted from documented cases across multiple jurisdictions and eras, has seven layers. They do not require coordination. They require only that each actor, at each layer, choose the lowest-risk option available to them.

Layer 1 — Identification and Visibility. The target becomes visible to institutional systems earlier or more intensely than peers. This can happen through talent programs, family background, proximity to sensitive environments, or simply being the wrong person in the wrong room. The visibility itself is not harmful. It becomes harmful when later layers use it as a predicate.

Layer 2 — Information Asymmetry and Demonstrated Access. In documented cases, targets become aware that information about their private circumstances is available to actors who should not have it. This awareness typically arises through observable institutional channels rather than direct confrontation: a detail from a sealed proceeding appears in an unrelated third-party filing; a confidential report’s contents are reflected in an administrative decision by a body that was not a party to the original matter; a piece of information shared only in a restricted setting is referenced in subsequent institutional correspondence.

The Stasi formalized this as Verunsicherung — the deliberate creation of uncertainty about what is known and by whom.¹ But the mechanism does not require a formal program. Information asymmetry is an inherent feature of systems that generate sealed records, confidential proceedings, and restricted-access databases. When that asymmetry becomes visible to the person it concerns — when they can observe that restricted information has migrated to an unexpected context — the effect is a persistent awareness of exposure, regardless of whether the migration was deliberate or incidental.

What distinguishes this from ordinary information sharing is auditability: in cases where the mechanism has been documented, the information trail is traceable. A specific detail moved from a specific restricted source to a specific downstream action. The question for any reviewer is whether such a trail exists in the case at hand.

Layer 3 — Deniable Coercion. Threats, warnings, and pressure are delivered in forms that preserve plausible deniability. “It was a joke.” “That’s not what I meant.” “You’re reading into it.” The target experiences coercion. A third-party observer sees nothing actionable. This asymmetry is a structural feature of deniable communication — whether or not it is intentional, it functions as one.

Layer 4 — Legal and Administrative Leverage. The target acquires a formal institutional vulnerability — an indictment, a proceeding, a filing, a sealed record — that can be activated or referenced by downstream actors. The vulnerability does not need to result in conviction or even adjudication. Its existence is sufficient. It changes the risk calculus for anyone considering whether to help the target.

Layer 5 — Reputational Poisoning. A stigmatizing allegation is placed into a channel the target cannot access, cannot rebut, and often cannot confirm exists. A sealed court record. A confidential personnel file. A whisper network among institutional gatekeepers. The allegation does not need to be believed. It needs only to be present — to create a hesitation, a doubt, a reason for the next reviewer to close the file.

Layer 6 — Resource Depletion. Housing, employment, savings, relationships, and health are degraded through the cumulative weight of the preceding layers. No single actor needs to intend this outcome. The target, engaged in sustained defensive action across multiple fronts, simply runs out.

Layer 7 — Oversight Exhaustion. The target files complaints. The complaints enter systems that route them into confidential processes, jurisdictional limitations, time-barred windows, and self-referential review bodies. Each complaint is handled in procedural isolation. Rarely are they evaluated in the context of the others. The system’s own accountability mechanisms become the final containment layer.

The stack is not deterministic. Not every case exhibits all seven layers, and the layers do not necessarily appear in this order. The model describes a structural tendency, not a fixed sequence.

Social, informational, legal, and economic pressure can reinforce one another. The critical insight is that no layer requires the actors in other layers to know what they are doing. Each layer’s output becomes the next layer’s input. The stack can assemble itself.

Observable Outputs

If this architecture is present in a specific case, it should produce observable, auditable indicators. A reviewer — journalist, attorney, oversight body, or researcher — can look for:

• Complaint routing patterns. How many distinct bodies received complaints about the same set of facts? Were any cross-referenced? Did any reviewing body obtain the primary record (audio, documentary evidence) or rely solely on summaries and representations?
• Sealed-record prevalence. Are there sealed, confidential, or access-restricted records in the case history? Has any downstream actor’s behavior changed in ways consistent with awareness of those records?
• Temporal correlation. Do disruptions to employment, housing, or professional relationships cluster around complaint-filing dates, public statements, or other identifiable advocacy actions?
• Jurisdictional handoff patterns. Was the matter referred between bodies? Did any referring body’s stated reason for non-jurisdiction rely on a characterization that the complainant could not access or contest?
• Disposition documentation. When a complaint was closed, did the closing body state in writing what primary evidence it reviewed? If not, did it state why?
• Information migration. Can specific restricted or confidential details be traced from their original source to an unexpected downstream context — a filing, a decision, an institutional communication — through retrievable documentation?

The absence of these indicators does not prove the architecture is absent, but their presence — particularly in combination — is consistent with the model. Their absence should prompt caution against applying the framework.

II. The Central Mechanism: Stigmatize and Seal

Of the seven layers, Layer 5 — reputational poisoning inside a protected channel — is the most structurally potent and the least understood.

The mechanism operates through three distinct channels that can converge:

Channel A — Sealed and confidential records. An allegation is made in a proceeding or record that is subsequently sealed, classified, or placed under confidentiality restrictions. The target cannot see it, cannot rebut it, and cannot confirm to third parties what it says. But actors with formal access — judges, investigators, oversight staff, employers conducting background checks — can review it directly. The access pathway is institutional: the record exists in a system with defined access controls.

Channel B — Informal reputational networks. Information from restricted channels migrates — through professional gossip, collegial conversations, or off-the-record briefings — to actors who do not have formal access. A journalist evaluating whether to pursue a story may not read the sealed record, but may hear its characterization from a source who has. A potential employer may not run a background check, but may receive a phone call. The access pathway here is social and largely unauditable.

Channel C — Algorithmic amplification. When stigmatizing information enters digital platforms — through social media, public records databases, or search engine indexing — recommendation algorithms can amplify it. Platforms that optimize for engagement tend to surface high-valence content (conflict, scandal, accusation) over neutral content. The access pathway is automated and indiscriminate: anyone who searches can encounter the amplified signal.

The convergence of these three channels — not any single one — is what makes the mechanism durable. A sealed record primes institutional gatekeepers (Channel A). Social migration extends the stigma beyond formal access (Channel B). Algorithmic amplification makes it discoverable to anyone with a search engine (Channel C). The target faces a credibility deficit that operates before any evidence is evaluated:

• If the target addresses the allegation publicly, they risk appearing guilty, unstable, or obsessed.
• If they do not address it, it remains a silent frame through which subsequent interactions may be filtered.
• If they file complaints, the complaint itself may be assessed by reviewers who have already encountered the stigmatizing characterization through one or more of these channels.

This is not speculation. It is the documented operational logic of systems ranging from Cold War–era political psychiatry² to modern watchlisting.³ The methods vary. The architecture is structurally comparable: place a stigmatizing label inside a channel the target cannot reach, and institutional risk-aversion tends to do the rest.

A neutral gatekeeper does not need to believe the allegation. Two dynamics are sufficient:

Risk aversion: “If there is even a chance this person is what the record suggests, I should not engage.”

Ambiguity bias: When evidence is inaccessible — sealed, classified, confidential — the mind fills the gap with priors and institutional heuristics. The default heuristic, in the documented cases reviewed here, tends toward avoidance.

The seal converts an allegation into a credential that resists falsification. It can travel across institutions with little degradation. It does not automatically expire. And it costs little to maintain.

III. The Geography of Proximity

The neutralization stack tends to operate most efficiently in small geographic areas.

This is not intuitive. Coercion and institutional pressure are usually imagined as large-scale operations requiring significant resources. But the documented cases — from East German Zersetzung¹ to the UK undercover policing scandal⁴ — suggest the opposite. The most effective operations documented in these cases were hyperlocal.

The following are structural features of bounded communities — small towns, tight neighborhoods, island jurisdictions, or constrained professional networks — that the documented cases illustrate. They are descriptions of proximity dynamics, not allegations about any specific locale:

• Access demonstrations cost less. In a bounded area, repeated encounters occur naturally. A single reference to restricted information in a shared social setting can establish awareness of exposure without requiring sustained operational effort.
• Social signaling propagates faster. Dense, overlapping acquaintance networks carry information without formal channels. A characterization introduced in one social cluster can reach adjacent clusters within days.
• Institutional proximity increases. The local courthouse, the local police station, the local media — the mechanisms that convert social pressure into legal or administrative outcomes — share the same social ecosystem. Professional and personal relationships overlap. The structural effect is that fewer intermediaries separate social reputation from institutional action.

The analytical term for this is a control surface: a bounded area within which a target can be monitored, isolated, and pressured using existing infrastructure. The documented cases suggest it does not require a dedicated budget. It requires only density and proximity.

IV. Gossip Networks as Intelligence Architecture

A structural observation about networked coercion is that civilian gossip networks and professional intelligence platforms share comparable topology.

An intelligence platform works by:

1. Collecting information across multiple channels.
2. Routing it through bridge nodes that connect otherwise separate networks.
3. Surfacing patterns that no single channel could reveal.

A gossip network operates similarly. Private Facebook groups, group chats, workplace cliques, nightlife scenes, and community organizations are the channels. People who belong to multiple groups — the person who is in the local parents’ group and the nightlife group and the professional network — are the bridge nodes. Information hops across circles by human carriers, then can be amplified by algorithmic recommendation systems that optimize for engagement, which in practice often means optimizing for drama.

The critical feature is overlapping membership. Any single group is a closed channel. But when a person belongs to three or four groups simultaneously, they carry information across all of them, often without deliberate intent. The topology becomes functionally comparable to a collection platform — except it runs on consumer messaging apps instead of certified intelligence infrastructure.

This means “surveillance” does not necessarily require wiretaps or spyware. It can emerge from social routing plus platform optimization, with occasional hard-surveillance moments used primarily for access demonstrations. The target may be visible to the network before any institutional actor takes interest. And the network’s output — who is connected to whom, who is vulnerable, who is isolated — is available to anyone with access to the right bridge nodes.

This is not a novel observation about technology. It is an observation about architecture. The professional intelligence platform that maps relationships, surfaces patterns, and identifies vulnerabilities is the institutional version of what the gossip network already does. The difference is certification, legal constraint, and oversight. The information-flow topology is structurally comparable — which is precisely what makes the informal version harder to regulate.

V. Closed Loops: When Oversight Becomes Containment

The final structural element is the most important for anyone attempting to use legitimate channels: the self-referential oversight loop.

A closed loop exists when the body responsible for investigating misconduct shares personnel, funding networks, confidentiality obligations, or institutional incentives with the actors being investigated.

Three variants appear repeatedly in documented cases:

The Judicial Loop. Judges investigating judges. The most thoroughly documented example is Chicago’s Operation Greylord, in which federal investigators discovered that the Cook County court system had been captured by a corruption network so thoroughly that internal oversight was structurally compromised — it took an FBI sting operation, run for years inside the courthouse itself, to break the cycle.⁵ The Pennsylvania “Kids for Cash” scandal demonstrated a comparable architecture: judges using authority in systematically abusive ways that internal review mechanisms failed to detect or correct until federal intervention.⁶

The Executive Loop. When the attorney general’s office investigates executive-branch corruption while reporting to the executive branch. The structural conflict is self-evident and has been litigated extensively. The relevant legal principle, stated by the Hawaii Supreme Court in Amemiya v. Sapienza (1981): “doubts should be resolved in favor of disqualification.”⁷

The Law Enforcement Loop. Police officers whose misconduct is investigated by their own department, with termination decisions subject to labor arbitration under collective bargaining agreements that can reinstate officers after formal findings. Arbitration can reinstate officers even after sustained findings, which can structurally weaken the accountability mechanism.

Each loop tends to be insulated by confidentiality. Judicial conduct proceedings are typically confidential. Internal affairs investigations are typically confidential. Sealed court records are confidential by definition. The effect: public accountability mechanisms often cannot be publicly verified to work or fail.

The combined operation of these three loops can produce an accountability vacuum in which procedurally valid steps yield substantively null outcomes. A complaint is filed. It is acknowledged. It is routed. It enters a confidential process. The process closes. The complainant is told: insufficient evidence, no jurisdiction, matter closed.

Every step was correct. The outcome was nothing.

VI. The Documented Precedents

The architecture described above is not theoretical. Its components have been documented in federal investigations, congressional findings, declassified intelligence records, and international human rights proceedings.

The catalog that follows is a structural comparison — not a claim that any specific case replicates any other. The logic of comparison is this: different systems, operating at different scales, with different levels of intent and different degrees of severity, can share structural features. A Stasi directive is not the same as a bureaucratic incentive cascade; a COINTELPRO operation is not the same as a gossip network amplified by social media. The claim is narrower: that certain mechanisms — stigma placed in unreachable channels, oversight routed through captured bodies, information asymmetry used to create uncertainty — recur across these cases in ways that are structurally comparable. Where this article uses the phrase “shape match,” it means: this documented case exhibits a mechanism that is architecturally similar to a component of the model — not that the cases are equivalent in intent, severity, or moral weight.

Psychological Decomposition

Stasi Zersetzung (East Germany, 1950s–1989). The Ministry for State Security’s Directive 1/76 codified a program of “decomposition” — targeted psychological disruption designed to “switch off” dissidents without arrest. Tactics included reputation sabotage, career interference, social isolation, engineered relationship breakdowns, and the creation of a pervasive sense of being watched. The directive is preserved in the Stasi Records Archive and has been extensively analyzed by historians of the GDR.¹

The shape match: deniable interference with life circumstances; demonstrated-access operations; “make them look unstable” as a strategy that converts the target’s rational response to harassment into evidence of instability.

Disrupt and Discredit

FBI COINTELPRO (United States, 1956–1971). The Church Committee documented a program of covert action aimed at disrupting and discrediting domestic political organizations and individuals. Tactics included infiltration, rumor dissemination, anonymous mailings, and manipulation of media.⁸ The most notorious episode — the FBI’s anonymous letter to Martin Luther King Jr. urging him to commit suicide — demonstrates the extreme end of the reputational-poisoning mechanism.⁹

The shape match: neutralization through credibility destruction rather than prosecution; use of institutional access to manipulate the target’s social environment; the target’s inability to identify the source of pressure.

Stigma Under Secrecy

U.S. No Fly List and Watchlisting (2001–present). The Terrorist Screening Center’s consolidated watchlist creates high-impact stigma with constrained contestability. Individuals are placed on lists through processes they cannot observe, for reasons that may not be disclosed, based on standards that have been litigated repeatedly. The Ninth Circuit’s decision in Kashem v. Barr recognized constitutional concerns with the redress procedures.³ The ACLU’s extensive litigation record documents the practical impact: a label that travels across agencies and borders, affecting employment, travel, and institutional trust, without a meaningful opportunity to challenge it.

The shape match: institutional stain that operates across agencies; the target experiences concrete harm while observers see only “administrative process.”

Political Abuse of Psychiatry

Soviet Punitive Psychiatry (1960s–1980s). The systematic use of psychiatric diagnosis — particularly “sluggish schizophrenia” — to discredit and institutionalize political dissidents is documented in peer-reviewed literature and international human rights findings.² The mechanism: reframe dissent as pathology, convert complaint-making into a “symptom,” and strip the target of credibility by assigning a diagnostic label that resists disproof within the institutional framework that created it.

The shape match: a credible-sounding institutional label that converts the target’s attempts to seek accountability into evidence of the condition being attributed to them.

Infiltration and Intimate Exploitation

UK Undercover Policing (1968–2010s). The Undercover Policing Inquiry has documented officers who maintained long-term deceptive intimate relationships with targets, fathered children under false identities, and reported to supervisors who were aware of the deception. The inquiry’s interim findings describe institutional awareness and management failures spanning decades.⁴

The shape match: social embedding as a control mechanism; the use of personal relationships as intelligence infrastructure.

Procedural Pressure as Suppression

Strategic Lawsuits Against Public Participation (SLAPPs). Anti-SLAPP doctrine exists because courts recognized that litigation itself can be weaponized — that the cost, stress, and reputational damage of defending a lawsuit can silence speech regardless of the merits. The Reporters Committee for Freedom of the Press catalogs the legislative response across jurisdictions.¹⁰

The billionaire-funded Hogan v. Gawker litigation demonstrated that third-party financing can amplify this mechanism: a well-resourced actor can impose existential legal pressure on a target without appearing as a party.¹¹

The shape match: create legal leverage, impose cost, force exit from the arena — without formal censorship.

Employment Denial Systems

Hollywood Blacklist (1947–1960s). The HUAC-era blacklist demonstrated that stigma plus institutional coordination can destroy livelihoods without formal legal process. Studios maintained lists. Agents refused calls. The mechanism was social, not statutory — and it was devastating.¹²

UK Construction Blacklist (Consulting Association, exposed 2009). A secret database used to deny employment to construction workers based on union activity and political views. Exposed by an Information Commissioner’s Office raid. A rare case where a literal deny-list was documented and proven.¹³

The shape match: livelihood disruption as a control lever; reputational metadata traveling across employers and industries.

Captured Judicial Systems

Operation Greylord (Chicago, 1978–1986). An FBI undercover investigation that exposed systemic corruption in the Cook County court system — fixed cases, bribed judges, complicit attorneys. The operation demonstrated that judicial ecosystems can be captured so thoroughly that internal oversight mechanisms become part of the corruption. Federal intervention was the only mechanism that broke the loop.⁵

“Kids for Cash” (Pennsylvania, 2003–2008). Two juvenile court judges received millions in payments from private detention facilities in exchange for sentencing children to those facilities. The scandal demonstrated that judicial authority can be exercised in systematically abusive ways for extended periods without detection by oversight mechanisms.⁶

The shape match: procedure as weapon; gatekeepers as chokepoints; the difficulty of correction from within a captured system.

Surveillance Commoditization

NSO Group / Pegasus Spyware (2010s–present). Citizen Lab and Amnesty International have documented the use of commercial spyware to target journalists, human rights defenders, and political dissidents across multiple countries.¹⁴ The significance is not the technology itself but its availability: access-demonstration capabilities that once required state intelligence agencies are now commercially purchasable.

The shape match: “you are observable” demonstrations delivered through commodity tools; private communications becoming contestable evidence; intimidation-by-capability rather than intimidation-by-confrontation.

Confidentiality as Silencing

Non-Disclosure Agreements in Misconduct Cases. The post-Weinstein policy movement around NDAs used to suppress misconduct reporting demonstrates the mechanism: a contractual obligation that prevents the target from mobilizing public support, sharing evidence, or even confirming the existence of a dispute.¹⁵ The shape is structurally comparable to the sealed record: can’t rebut publicly, can’t discuss terms, can’t mobilize support.

VII. The System Without a Mastermind

The most important conclusion from this catalog is negative: these mechanisms do not necessarily require central coordination.

The Stasi had a directive. COINTELPRO had a program. But the architecture documented above can assemble itself from local actors making local decisions for local reasons.

A school administrator protects the school’s reputation. A police officer avoids paperwork. An attorney preserves a client relationship. A judge manages a docket. A journalist declines a story that cannot be independently verified because the records are sealed. An oversight body applies its jurisdictional rules as written.

Each actor’s behavior is individually rational. None necessarily requires malice. The information topology — gossip networks, recommendation algorithms, institutional databases — can carry the effects across actors without anyone needing to coordinate. The procedural topology — sealing, confidentiality, jurisdictional time limits — can prevent any single reviewer from seeing the full picture.

The system can produce the outcome the way a river produces a canyon. Whether it does so in any specific case depends on whether the observable outputs are present.

This is what makes the pattern so difficult to fight, and so important to name. A conspiracy can be exposed. An incentive structure can only be redesigned.

VIII. What Breaks the Loop

If the architecture is structural rather than conspiratorial, the response must also be structural. Exposing bad actors is necessary but insufficient. The following are the pressure points where the architecture is weakest:

Primary records break the narrative. The most durable containment mechanism is the sealed record that no one thinks to retrieve. The most effective counter is forcing retrieval. If a reviewer must listen to an audio recording rather than rely on a file summary, the incentive structure shifts. Binary questions — does the recording contain X, yes or no — are harder to route into ambiguity than narrative complaints.

Written dispositions force accountability. When an oversight body can close a matter with a form letter, the closure is costless. When the body must state, in writing, whether it obtained and reviewed the primary record before reaching its disposition — and state the specific basis if it did not — the cost of non-review increases.

Temporal documentation defeats post-hoc fabrication claims. If a complainant can demonstrate that they identified an allegation and a corroboration target before a key denial occurred, the “he made it up after the fact” defense collapses. Dated law-enforcement intake records, emails, and phone logs become the decisive evidence — not because they prove the underlying allegation, but because they prove the allegation was contemporaneous.

Publication creates a cost for silence. Institutions optimize for quiet. A public record — not a social media post, but a structured, citeable, verifiable public record — changes the risk calculus. Silence is no longer costless if the silence itself has been documented and published.

Cross-jurisdictional filing defeats single-loop containment. A complaint filed with only one body can be absorbed by that body’s internal closure mechanisms. The same complaint filed simultaneously with multiple bodies — state bar, federal prosecutors, journalism outlets — forces each body to account for the others’ existence. No single loop can contain what multiple loops are being asked to review.

None of these are guarantees. They are pressure points. They work because they target the architecture’s actual load-bearing element: the ability to dispose of a matter without creating a retrievable record of the disposition.

Every system that destroys a person leaves a shape behind. The shapes recur because the architecture recurs. And the architecture recurs because it works — until someone maps it, names it, and makes the map available to the next person sitting in the same cage, wondering why no one comes.

IX. How to Use This Map

This article presents a structural model. Models are tools, not truths. The following principles should guide anyone applying this framework to a specific situation.

Do not over-attribute. Not every institutional failure is this architecture. Bureaucratic incompetence, individual bias, resource constraints, and bad luck produce outcomes that can resemble the neutralization stack without involving it. Before applying the model, ask: is there retrievable evidence of the specific mechanisms described here, or am I pattern-matching against a narrative?

Prioritize primary evidence. The architecture’s most durable feature is that it operates through channels that resist verification. The most effective counter is insistence on primary records: audio recordings, original filings, dated correspondence, timestamped communications. Summaries, characterizations, and institutional representations are not substitutes. If a claim cannot be grounded in a retrievable document, it remains an inference, and should be labeled as such.

Document before you interpret. If you believe you are observing these mechanisms, build a chronological record of observable events — dates, documents, institutional communications — before fitting them to the model. The record is durable. The interpretation can be revised. Reversing that order invites narrative capture.

Beware narrative capture. Any sufficiently general model can appear to explain everything. If this framework seems to account for every setback, every institutional interaction, every piece of bad news — that is a signal to step back, not to lean in. A useful model should be wrong about some things. If it is never wrong, it has become a lens rather than a tool.

Focus on what is retrievable. The architecture described in this article is built on confidentiality, sealing, and jurisdictional fragmentation. The counter-architecture is built on retrieval: forcing the record into the open, one document at a time. The question is not does this model explain what happened? The question is: what documents exist, who has them, and what do they say?

— Ekewaka Lono, 24 February 2026

Notes