The One-Year Clock and Its Four Starting Points

Federal collateral review runs on a one-year period, and the hard part is not the length but the start. Two nearly parallel provisions supply four possible starting events each, and the statute takes the latest one that applies. Most filings are governed by the first of them, but the other three exist because some claims cannot be built until something changes, and the calculation begins with identifying which event the claim actually depends on.

The four triggers for a federal movant

Section 2255(f) measures the year from the latest of four dates. The first is the date the judgment of conviction becomes final. The second is the date on which an impediment to making a motion, created by governmental action in violation of the Constitution or laws of the United States, is removed, and it applies only where the movant was in fact prevented from filing by that impediment. The two halves of that requirement are both litigated.

The third is the date on which the right asserted was initially recognized by the Supreme Court, where the right is newly recognized and has been made retroactively applicable to cases on collateral review. The fourth is the date on which the facts supporting the claim could have been discovered through the exercise of due diligence. That last trigger measures discoverability, not discovery: a movant who could have learned the facts earlier does not get a later start by not looking.

The parallel triggers for a state petitioner

Section 2244(d)(1) mirrors the structure for a prisoner in state custody. Subparagraph (A) runs from the date the judgment became final by the conclusion of direct review or the expiration of the time for seeking such review. Subparagraph (B) runs from the removal of a State-created impediment in violation of the Constitution or laws of the United States. Subparagraph (C) runs from the date a constitutional right was initially recognized by the Supreme Court and made retroactively applicable to cases on collateral review, and subparagraph (D) from the date the factual predicate could have been discovered through due diligence.

The parallel is close enough that the two bodies of case law are read together, and the differences in wording are small but not empty. The state provision defines finality on its face; the federal one does not. The state provision speaks of a constitutional right; the federal one speaks of a right without that qualifier. Neither difference has produced a settled divergence in result, but both are argued.

Computing finality under the first trigger

Finality is the trigger that governs the great majority of cases, and it is computed from the end of direct review rather than from sentencing. Where a certiorari petition is filed, the judgment becomes final when the Supreme Court rules on it. Where the appeal is decided and no petition follows, finality arrives when the ninety-day certiorari period expires. Where no appeal is taken at all, the judgment becomes final when the fourteen-day period for a notice of appeal lapses.

That sequence has to be reconstructed before anything else is calculated, because an error at this step is not recoverable later. A resentencing, an amended judgment or a remand can reset the date, and the effect of a new judgment also bears on whether a later filing counts as successive, a question taken up in the bar on second or successive filings. The distinction between what closed on appeal and what remains open is the subject of the comparison of appellate and collateral review.

The questions the circuits answer differently

Two disagreements recur. The first is whether the discovery trigger operates claim by claim or fixes a single date for the application as a whole. The majority position is claim by claim, which limits the reach of a late-surfacing fact to the claim it supports. The minority position treats the provision as setting one start date for everything filed, on the reading that the statute speaks of a period for the application rather than for individual grounds.

The second concerns who may make a new right retroactive. Some courts read the newly recognized right trigger to require that the Supreme Court itself have declared the rule retroactive. Others hold that any court may make the determination, pointing to a textual contrast: the gatekeeping provisions for successive filings say expressly that the rule must be made retroactive by the Supreme Court, while the limitation triggers omit that phrase. The contrast is real, and the courts have not resolved what to draw from it.

A limitation period, not a jurisdictional bar

The one-year period is a statute of limitations. It can be waived by the government, forfeited by a failure to raise it, and set aside on equitable grounds in the narrow circumstances the courts recognize. That character is what makes the doctrines built around it possible at all, because a jurisdictional deadline would admit no exception and no excuse.

Two mechanisms operate on the running of the period once it has begun. For a state petitioner, the statute itself suspends the count while a state collateral application is pending, described in statutory tolling during state review. Beyond that, both federal movants and state petitioners may seek relief on the equitable grounds described in the diligence and extraordinary circumstance test. Neither doctrine changes the starting point; both operate on the days that follow it.