Clearly Established · Chapter 3 · sample
The Sentence That Moved the Question
Why the modern test asks about the state of the case law rather than the state of a person’s mind — and what that change does to everyone who arrives afterwards.
Start with what the statute says, because almost every argument about this doctrine is conducted at a distance from the text. The provision now known as was enacted in 1871 and it does one thing: it makes a person who uses the authority of a state to deprive someone of a federal right liable to the party injured. There is no clause about good faith. There is no clause about reasonableness. There is no clause about immunity. Whatever else is true about , it is not in the sentence Congress wrote.
That absence is not a scandal by itself. Courts read statutes against a background of common-law understandings, and immunities were part of that background. What matters for this chapter is a narrower and stranger fact: the shape of the defence changed once, decisively, in a way that moved the question from a person to a library.
Before 1982, the defence had two halves. One was objective — had the official acted in a way a reasonable person would think lawful? The other was subjective — had the official acted with a malicious intention or with knowledge that the conduct was wrong? The second half is the interesting one, because a subjective question is a question of fact, and questions of fact belong to juries. A defendant asserting good faith could not usually escape a trial by asserting it. The claim had to be tested.
The Court held that government officials performing discretionary functions are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known. RECORD · Harlow v. Fitzgerald, 457 U.S. 800 (1982) · Supreme Court of the United States
Read the sentence again and notice its subject. It is not the official. It is the law. The question after Harlow is whether the right was , and that is a question a judge can answer by reading reported decisions, without hearing a witness, without a jury, and — crucially — before the expense of discovery and trial. The Court said as much: the point was to permit resolution of insubstantial claims before those costs are incurred.
Five years later the Court added the instruction that turned a standard into a procedure. In Anderson v. Creighton, 483 U.S. 635 (1987), it held that the right in question must be identified at an appropriate level of specificity rather than as a general proposition. It is not enough to say that the Fourth Amendment forbids unreasonable seizures; that is true and useless. The plaintiff must be able to point to law establishing that this conduct, in circumstances close enough to these circumstances, was unlawful.
Add one more piece and the machine is complete. Because an immunity from suit is worth nothing if you must first endure the suit, a denial of qualified immunity may be appealed immediately — an — rather than waiting for final judgment. Mitchell v. Forsyth, 472 U.S. 511 (1985). A plaintiff who survives the motion has not won; the plaintiff has earned an appeal.
Then there is the sequence. In Saucier v. Katz, 533 U.S. 194 (2001), the Court required courts to take the questions in order: first, do the alleged facts show a constitutional violation; second, was the right clearly established. That order produced something valuable almost as a by-product. Every time a court answered the first question, it wrote down whether particular conduct was unconstitutional — and that written-down answer became the precedent the next plaintiff would need. Eight years later, in Pearson v. Callahan, 555 U.S. 223 (2009), the Court held that the sequence was no longer mandatory. A court may now begin and end with the second question.
Here is where this book stops describing and starts arguing, and the change is marked so you can see it coming.
The specificity requirement and the optional sequence work together as a ratchet. The precedent a plaintiff must produce is precisely the precedent that courts are no longer obliged to write. Each unanswered constitutional question leaves the next plaintiff in the same position, holding the same empty hand. EDITORIAL CONCLUSION · Tracy Williams · dated, attributed, and revisable through the UPDATE door
That is a claim about consequences, and claims about consequences can be wrong. It is not a holding. No court has described the doctrine as a ratchet, and this project does not pretend otherwise; the paragraph above is coloured oxblood and marked as editorial for exactly that reason. What can be said without argument is that the two decisions exist, that they say what they say, and that the second one removed a requirement the first one had imposed.
There is also a counter-current, and honesty requires putting it in the same chapter rather than hiding it in an endnote. In Hope v. Pelzer, 536 U.S. 730 (2002), the Court held that officials had fair warning that the conduct alleged was unconstitutional even without a decision on indistinguishable facts. In Taylor v. Riojas, 592 U.S. 7 (2020), decided without argument, the Court held that no reasonable officer could have concluded that the conditions of confinement alleged were constitutionally permissible. Both cases show that a plaintiff can win when the violation is obvious enough. Whether that route is wide enough to matter at scale is the subject of Chapter 4 — and it is a question, not a conclusion.
One last note, for readers who came to this book through the federal side of the story. The equivalent action against federal officers, the , has travelled in the opposite direction: recognised in 1971, and narrowed repeatedly since, most recently in Egbert v. Boule, 596 U.S. 482 (2022). A reader who assumes the two tracks are symmetrical will misread both.
End of sample · Chapter 4 continues in the specified full edition