---

---

Chapter 8

The Appeal Stamp

The stamp has a modest vocabulary. Reconsidered. Reversed. Remanded. Denied. Granted. Vacated. Returned for further proceedings.

It does not look like humility. It looks like office equipment. Ink touches paper, or its digital equivalent marks a docket. A first decision is no longer the last event in the file. Someone has looked again, or has been ordered to look again, or has decided that looking again does not change the result.

In 1971, one of those words reached a highway planned through a park in Memphis. The case was Citizens to Preserve Overton Park v. Volpe. The Secretary of Transportation had approved the route of an interstate highway through Overton Park. Federal law did not simply ask whether the road would be useful. It required that parkland not be used for the highway unless there was no feasible and prudent alternative, and unless the project included all possible planning to minimize harm to the park.¹

That statutory language created pressure inside the decision. A highway promises mobility, commerce, and years of administrative coordination. Finality matters because a project cannot be redesigned forever. But a park is not empty space on the map. The statute forced the agency to ask whether the apparent route was truly necessary in the legal sense. The question was not whether the Secretary preferred the route. The question was whether the decision could be defended under the standard Congress had written.

The Supreme Court did not decide that the highway could never pass through Overton Park. It did something more institutional. It rejected the idea that judicial review could rest on litigation affidavits produced after the decision. Review under the Administrative Procedure Act required a substantial inquiry into whether the Secretary had acted within authority, considered the relevant factors, and made a decision that was not arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law. That inquiry had to be based on the full administrative record before the Secretary at the time of the decision, not on post hoc rationalizations supplied after citizens challenged the action.

Then came the stamp: remanded.

Remanded did not mean the citizens had won the final highway dispute. It meant the approval could not simply be treated as self-justifying. The Supreme Court remanded the case to the district court for plenary review under the proper standard. It did not itself become the Secretary of Transportation, and it did not directly remand the highway decision to the agency as a substitute for that judicial review. The district court had to examine the Secretary's decision on the administrative record. The agency might need to disclose what was before it. If the record did not explain enough, the court could require more explanation from the officials who made the decision. The word on the docket forced the institution back toward the reasons it had at the moment it exercised power.

That is why remand can be consequential without being theatrical. No one has to declare the agency malicious for review to matter. The agency may have expertise. The highway may have real public reasons behind it. The first decision may even survive after proper review. But after Overton Park, the route could not be defended merely by saying, in effect, trust the Secretary now. The institution had to show its work from the time the work was done.

The stamp matters because institutions make mistakes in ways individuals cannot easily correct on their own. A form is misread. A deadline is miscalculated. A fact is omitted. A rule is applied to the wrong category. A hearing officer credits the wrong evidence. An agency explains too little. No serious institution can promise that every first decision will be right. The question is whether it treats first decisions as correctable.

Appeal is often described as a right of the losing party, and that is true in many settings. But for this book the deeper point is institutional. Review is designed reconsideration. It gives the system a lawful way to say: the first decision may have been incomplete, unsupported, unlawful, inadequately explained, or wrong. That admission need not destroy legitimacy. It can strengthen it. A system that can correct itself is more trustworthy than a system that must pretend it never errs.

The strongest reason for finality deserves respect. Decisions cannot remain open forever. People arrange their lives around outcomes. Agencies and courts have limited capacity. Opposing parties need repose. Public projects cannot function if every determination can be relitigated indefinitely with no threshold, record, or deadline. Finality protects people from endless process, strategic delay, and the exhaustion of having to win the same dispute repeatedly.

The stamp therefore should not be romanticized. Appeal can be slow, expensive, technical, and unevenly available. It can reward persistence as much as rightness. It can bury people in deadlines and standards of review. A review system can appear open while remaining practically inaccessible. Still, a consequential first decision with no meaningful path of review asks too much trust from the people affected. It asks them to accept that the first official saw everything that mattered, understood the law, weighed the evidence properly, explained the reason adequately, and acted within authority. Sometimes that trust is warranted. Sometimes it is not. Legitimacy requires a path for telling the difference.

The modern administrative state made this problem unavoidable. Agencies do not merely advise. They license, inspect, adjudicate, distribute benefits, enforce regulations, impose penalties, manage public lands, administer immigration systems, oversee workplaces, approve drugs, regulate markets, and decide questions that shape ordinary life. The more decisions institutions make, the more important it becomes to design pathways for correction.

The Administrative Procedure Act of 1946 is one of the central legal structures for this problem in federal law. The Act does many things, including setting default procedures for rulemaking and adjudication. For this chapter, its importance lies in the architecture of record, reason, hearing, and review.² It does not assume agencies are illegitimate. It assumes agency power must be made governable.

That distinction matters. Review is not anti-administration. Agencies have expertise. They handle volume. They apply statutes in complex settings. The problem is not that expertise exists. The problem is expertise without a path for challenge.

The APA's judicial review provisions, now reflected in 5 U.S.C. §§ 701-706, establish a default framework for courts to review agency action where review is not precluded and where action is not committed to agency discretion by law.³ Courts may decide questions of law, interpret statutes, and determine whether agency action is arbitrary, capricious, an abuse of discretion, contrary to law, unsupported where required by the record, or procedurally defective. Different standards apply in different contexts. But the structural idea is constant: power should leave reasons behind.

Reasons are the hinge between authority and review. A decision without reasons may be efficient in the moment, but it is hard to contest and hard to learn from. A later reviewer cannot tell whether the right standard was applied. The institution itself cannot see patterns of error. Reasons slow the decision just enough to make accountability possible.

Overton Park also belongs with the Chenery principle: agency action is judged on the grounds the agency invoked when it acted, not on a lawyer's later reconstruction of why the action might have been lawful.⁴ If an agency can act first and invent adequate reasons during litigation, review becomes theater. The reviewing court would be judging a better decision than the agency actually made. The affected public would be told that power is lawful because someone later found a defensible explanation for it.

Record review resists that drift. It asks what was before the agency when it acted. Which alternatives were considered? Which factors were treated as relevant? Which statutory commands were understood as binding? The record does not make every answer obvious. But without it, review has no object.

Hearings serve a related function. They are not always formal trials, and they should not always be. An agency that required courtroom procedure for every minor matter would collapse under its own fairness. But where important interests are at stake, the chance to present evidence, respond to adverse material, and be heard by an appropriate decision-maker can be the difference between administration and command.

Internal appeals occupy a middle space. They may be faster and more specialized than judicial review. They can correct factual mistakes before a court is needed. They can allow an agency to maintain consistency across front-line decisions. But they also raise questions. Is the reviewer independent enough from the first decision? Can new evidence be introduced? Does the reviewing body have authority to change the outcome, or only to check whether the first official followed a script? Does the person appealing understand what must be shown? A stamp that says "reviewed" but only confirms that someone else glanced at the file is not reconsideration. A stamp that says "denied" without explaining why may be final, but not necessarily legitimate.

Remand is institutional discipline without institutional replacement. In Overton Park, the Court required review of whether the Secretary had followed the statutory command and whether the decision could be sustained on the record. The broader design lesson is that error should be sent back toward its source so the institution must do its own work in a form that can be tested. That is a different kind of correction than substitution. It tells the agency: your authority remains yours, but authority exercised under law must be traceable to law.

This is why revision should not be treated as humiliation. An institution that reverses or explains itself has not necessarily become weak. It may have become more accurate. The embarrassment lies not in correction, but in the inability to correct. A system that must defend every first decision as if admitting error would destroy its authority eventually teaches people that its authority depends on denial.

The appeal stamp also changes the first decision. If a front-line official knows that reasons may be reviewed, the first decision becomes more disciplined. The official may read the file more carefully, cite the correct rule, and avoid unsupported shortcuts. Review is not only a later safeguard. It is a prior influence on behavior. But review can also distort. If decision-makers become afraid of reversal, they may write defensively rather than clearly. If the appeal process is too complex, only people with lawyers or institutional knowledge can use it. The existence of a review path is not enough. Its design determines whose mistakes can be corrected.

The book's core problem appears here in administrative form. Agencies need rules, standards, deadlines, eligibility criteria, and procedures because otherwise public power becomes personality. A transportation secretary cannot treat a statutory protection for parks as optional because a route is administratively convenient. If an agency applies the wrong standard, treats the wrong facts as decisive, ignores alternatives Congress made relevant, or offers reasons only after being sued, then the institution must have some lawful capacity to reconsider. Appeal is one way to make that capacity visible. It places the first decision inside a sequence: application, decision, notice, reasons, record, appeal, review, remand, revision, finality.

Notice protects because people cannot contest a reason they never receive. A record protects because memory and impression are too fragile for serious review. A deadline protects because institutions need closure. An opportunity to submit evidence protects because first files are often incomplete. A standard of review protects because not every disagreement justifies replacement of the first decision.

The difficult work is arranging these protections so they do not cancel each other. Too much finality can harden error. Too much openness can make every decision provisional. Too much deference can shelter unlawful action. Too little deference can make expert administration impossible. Too much procedure can exclude the people it was meant to protect. Too little procedure can leave them at the mercy of a file they cannot see.

Accommodation shows that equality sometimes requires individualized adjustment. Appeal shows that individualized correction must be institutionalized. The exclusion order shows what happens when emergency and category overwhelm the possibility of either. The stamp is the quieter middle technology: the system's way of pausing before its own authority becomes irreversible.

Modern institutions often speak of correction as failure. High reversal rates may reveal poor initial decision-making. But zero correction is not necessarily a sign of excellence. It may mean no one can challenge the system, or that the review process is too weak to detect error. Competence is not the absence of revision. Competence is the ability to make revision unnecessary where possible and available where necessary.

Return to the stamp. Its ink is small, but its implication is large. In Overton Park, remanded meant that a highway approval had not been sealed against thought. It meant citizens could ask whether the first decision fit the statute. It meant the institution had built a lawful place for second sight.

The stamp does not guarantee justice. A reversed decision may still come too late. A remand may return to the same habits. But without some version of the stamp, the people before the institution are left with only protest from outside the system. Review gives protest a door.

Administrative procedure and review protect by requiring public power to follow forms, give reasons, and stay within law. Appeal protects by letting the institution see again when its first classification may have missed what the statute required it to notice. Public authority must be defended by the reasons and record belonging to the decision when it was made, not by a better explanation invented afterward.

Reconsideration is not the enemy of authority. It is one of the disciplines that keeps authority from becoming mere insistence.

Footnotes

1. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402 (1971), Cornell Legal Information Institute, https://www.law.cornell.edu/supremecourt/text/401/402. ↩

2. Administrative Conference of the United States, "Administrative Procedure Act," https://sourcebook.acus.gov/wiki/Administrative_Procedure_Act. ↩

3. 5 U.S.C. chapter 7, "Judicial Review," U.S. House Office of the Law Revision Counsel, https://uscode.house.gov/view.xhtml?edition=prelim&path=/prelim@title5/part1/chapter7. ↩

4. SEC v. Chenery Corp., 318 U.S. 80 (1943); see also 332 U.S. 194 (1947), Cornell Legal Information Institute, https://www.law.cornell.edu/supremecourt/text/318/80. Overton Park relies on the principle that agency action must be judged on the grounds invoked by the agency when it acted. ↩