Part IV — What the design does to a person, and how it catches itself

23. The Shield

A disclosure against a publicly answerable actor protects its maker from confinement under a conviction, in each case in which the maker is cited. The protection begins before anyone has decided whether the disclosure is true, and the discloser has to obtain no finding to receive it. Three records limit it. A properly made deceit finding about that particular exposure ends its protection. An entry by both of a case's temporal witnesses that the disclosure followed the charge ends its reach into that case. And a properly made case finding that a prosecution is unrelated to the disclosure removes that prosecution from its coverage.

The case finding needs authorised reviewers whose eligibility is independently confirmed. The defendant, the injured person and the prosecuting court are each excluded from deciding it. An actual conflict, found by an authorised office, defeats the affected decision; merely naming a reviewer in a disclosure does not. Dara's sequence below tests that distinction.

The cases show what follows when those conditions are present, absent or disputed.

Faro: a private target

Faro is the defendant and discloser in this case. The record contains an injury to Pax and Faro's disclosure against Pax, the injured person.

Pax is not publicly answerable in the record. Answerability follows from a public-body entry or a seating, and the record holds neither for Pax. Faro is therefore unshielded, and the otherwise complete case authorises confinement. An accusation against a private person opens no shield, whatever it is called.

The distinction depends on an entry. Suppose a public-body entry about Pax were added, making Pax answerable. Faro's exposure would then open the shield, and confinement would lose its authorisation. The rules treat a forged status entry exactly as a true one. The case shows the trust placed in the record, not a lawful way to appoint a victim to public office.

Lior: an unresolved disclosure

Lior's separate case also records an injury to Pax, but Lior's disclosure names the prosecuting court. The court is publicly answerable, and nothing in force finds that exposure deceitful. Lior's shield holds.

The absence of an adverse finding is enough for protection to continue, and it leaves open whether Lior told the truth. The recorded injury likewise leaves the disclosure's truth open: the two questions have separate evidence and consequences.

The prosecution's own assertion that it is unrelated leaves the shield in place. Suppose the court and the review body both record that assertion, and every positive entry confirming the reviewers' eligibility is present. The court is excluded from deciding its own prosecution, so confinement remains blocked. Suppose instead that the court's public-body entry were removed: Lior's disclosure against the court would then leave Lior unshielded. The protection rests on the record here too.

The shield blocks confinement under a conviction while the judgment itself stands; it acquits nobody. Other protective measures keep their own rules: arrest, search and hazard restrictions each need their own individual grounds, current authority, evidence, review and remedy, and the shield grants none of them. These cases concern confinement alone.

Confinement also needs the custody case's own merits: adjudicated defences and exculpatory evidence, criminal proof, disclosed lawful evidence, counsel and a fair hearing, agreed by the adjudicator and an independent merits reviewer for the same case. The shield and the unrelated-case finding each leave that inquiry whole. A finding that a prosecution is unrelated removes only the shield's obstacle, and the merits must still be shown. The rules check that the findings are recorded; deciding a defence belongs to the adjudication.

Kel: a finding about this disclosure

Kel's case pairs an injury to Pax with a disclosure against the court too. Here, however, a completed case identifies Kel's exposure, a deceit finding and its evidence. It includes two independent reviewers, their eligibility at the decision, and matching records of notice, access to evidence, a hearing and reasons. That finding defeats this exposure's shield, so Kel's otherwise complete case authorises confinement. It also produces a personal finding that restricts signing new credibility findings. It leaves Kel's wages, property, pension and every floor right in place.

A raw accusation or a generic judgment by the review body falls short of such a case: the completed finding must identify the same subject and incident. These requirements concern what the record must show; the outside witnesses may still have given a false account of the procedure or the evidence.

The finding concerns this exposure alone. It does not make every later statement by Kel false, and Kel remains eligible for protection against a different act of retaliation. Another exposure to the same authority is assessed separately too.

Dara: a separate exposure

Dara's record concerns an injury to another person and two exposures: one against the court and another against the review body. A completed case found the court disclosure deceitful.

The review body is publicly answerable, and nothing in force finds the second exposure deceitful, so the shield applies again. Dara's conviction remains recorded while confinement is blocked. Dara is both discredited and shielded. Protection follows the exposure, whatever the claimant's general credibility: a restriction on signing credibility findings gives nobody permission to retaliate.

Whether this prosecution is unrelated to the disclosure remains a separate question, and it can be answered while the second disclosure stands; the case finding's effect stays with that prosecution. Dara cannot block it by naming the decision participants in fresh disclosures.

The case the shield stops covering

The test continues with Dara as defendant and the same injured person. Suppose the appeals body and the convocation are proposed to decide whether this prosecution is unrelated to the disclosure. Their recorded agreement alone leaves Dara protected: evidence of their current authority and eligibility is missing.

Each deciding reviewer needs authority for this case. An appointments office and a separate office that reviews its work, both authorised for the case, must agree positively on each reviewer's eligibility. That finding covers lawful current appointment, independence from the parties and the disclosures, notice, reasons, an opportunity to be heard and independent challenge.

The separation is explicit. The deciding reviewers must be different people or bodies. The appointments office and its reviewing office must differ from one another and from both deciders. All four must be distinct from the defendant, the injured person and the prosecuting court. The court cannot clear a route to confinement under its own judgment, and a reviewer cannot certify their own independence.

With those premises recorded, Dara's confinement is authorised. The disclosure stays on the record, the shield still applies, and the credibility finding is unchanged: the case finding removes this prosecution from the shield's coverage. A finding for this case gives no eligibility or decision in another case.

Suppose next that Dara names the appeals body, the convocation and both appointments offices in fresh disclosures. Dara's confinement remains authorised. An allegation against a participant is not itself an adjudication of that participant's interests. Eligibility still rests on the positive independent finding; silence about family ties or a different institutional name cannot stand in for it.

Suppose then that the reviewing office records an actual conflict concerning the appeals body in this case. The shield then holds against the case finding, and Dara's confinement loses its authorisation. A favourable certificate left beside the conflict does not erase it. The same holds if the appointments office is found conflicted, and choosing a different favourable pair cannot bypass the recorded finding. Only an authorised office's finding in this case has that effect: an unauthorised accusation, or a conflict recorded for a different case, leaves Dara's confinement authorised.

A disclosure made after the charge

A prosecution can answer only for what came before it. Each case has two temporal witnesses, the ones that record its subject, court, judgment and injured person. When both record that a disclosure was made after that case's charge, the charge came first, and the disclosure stops blocking confinement in that case.

Suppose both witnesses date Lior's disclosure against the prosecuting court after the charge. Confinement under Lior's case is then authorised, and the shield itself still applies. A defendant who accuses the court that tries them gains no delay by it. One witness's entry leaves the disclosure's effect in place: it takes both, and a missing entry keeps the protection. A second disclosure the witnesses have left undated bears on the case again, and a disclosure dated after one charge still bears on any other case in which its maker is cited.

The timing record decides only what came first. It leaves the disclosure's truth open and its protection in every other case intact. Retaliation within a pending case is answered by the requirements that already govern confinement: the merits, the necessity of the place, independent review and challenge. A disclosure made before the charge keeps its protection until a properly made case finding removes the prosecution from its coverage.

What it costs

The shield can delay confinement under an unrelated prosecution for as long as the case's two witnesses have yet to agree that the disclosure followed the charge and no eligible reviewers have decided. Missing authority, a missing eligibility finding or an actual conflict found by an authorised office keeps the unrelated-case route closed. The injured person bears that wait, and still holds a claim to assistance, protection and remedy on its own grounds. The defendant keeps the floor, standing and the ability to challenge. The existing justice routes govern the hearing, interim relief, reviewer nonresponse and independent alternatives.

A disclosure against any actor who remains publicly answerable can open the protection, including a recalled official: recall leaves intact the answerability on which someone else's protection rests. The protection has no recency test and no deadline that would end it while a reviewer fails to act. The unrelated-case route settles its reach one case at a time, leaving each disclosure standing.

The defendant bears the cost of the timing rule. A disclosure made after the charge gives no protection in that case, so retaliation within a pending case has to be caught by the custody requirements rather than by the shield.

What this cannot settle

Someone who controls the appointments offices' findings can still lie about a reviewer's independence, and the case finding then opens confinement on a false premise. Protection for a disclosure made before the charge likewise depends on the two temporal witnesses dating it honestly: if both record a false date, the disclosure loses its reach into that case. Whether an independent tribunal is staffed, whether reviewers act promptly and whether the injured person's remedy is performed belong to the second book.

The shield separates protection from approval of the claimant's conduct. The next chapter turns to the adverse findings themselves: what a credibility finding requires, what it reaches and what must remain untouched.

Run it: the companion runs this chapter's cases on your own device, at dhilipsiva.dev/rights-nobody-has-to-earn/cases.

Argument: Protection that follows the charge

The companion states these rules as Article 24.

I scope the shield by the harm it guards against. Retaliation through confinement needs a prosecution, so the protection attaches to each case in which the discloser is cited and asks the one question retaliation raises: did this prosecution answer the disclosure? A prosecution can answer only for what came before it. When both of a case's witnesses date a disclosure after the charge, the disclosure could have prompted nothing in that case, and letting it hold back confinement there would let any defendant buy delay by accusing the court that tries them; it keeps its force in every other case in which its maker is cited. Where the connection is disputed, someone must decide it, and the prosecuting court and the parties have the strongest motive to answer it their own way. So two reviewers from outside the case decide, and two further offices must positively show each reviewer's eligibility, so that no single appointment settles the question. Naming a reviewer in a fresh disclosure leaves that eligibility standing, and a conflict found by an authorised office removes it. If naming were enough, a defendant who named every available body would hold a permanent shield.

The strongest alternative is a blanket immunity from prosecution: nobody who has exposed an official may be prosecuted at all. It is the position of those who doubt that a reviewer reading a case file can see motive, since a retaliating authority can build a charge whose link to the disclosure is hidden, and a finding that the prosecution is unrelated is exactly what such an authority would arrange. Its best argument is that it misses no disguised retaliation. No study of such an immunity is cited here. A narrower alternative, a threshold on how specific a disclosure must be before it protects, would filter tactical accusations, but somebody must judge specificity, and the judge nearest to hand is often the authority accused.

Against immunity, my rule costs the discloser. Reviewers who wrongly find a prosecution unrelated open the way to confinement, and someone who controls the appointment records can manufacture that finding. Protection for a disclosure made before the charge depends on two witnesses dating it honestly, and retaliation inside a pending case has to be caught by the custody requirements. Immunity puts its cost on everyone its holder harms afterwards, whose cases could never be tried, and it turns an accusation against an official into a licence. My rule also costs the injured person time while eligible reviewers are found.

I would move toward immunity on evidence that reviewers shown to be eligible regularly certify retaliatory prosecutions as unrelated. I would reconsider the timing rule if case witnesses were found to misdate disclosures, or if the review owed within a pending case proved too slow to catch retaliation. The case that opens this Part shows how long such a wait can run: people were held three to ten years for a trial that had not come.