Judges / District of Columbia

Rosemary M. Collyer

Written decisions in the mandamus and agency-delay group that Judge Rosemary M. Collyer signed, every one listed below, and the agency-delay suits that name the judge.

2,914 written decisions from 2,661 cases in the whole mandamus and agency-delay group, the group's newest decided October 30, 2026. Updated November 3, 2026 at 1:24 a.m. CT; the Passim connector counts the live collection, which can differ by the day's additions and retirements (which total to quote). Coverage: of the 1,606 public decisions since January 1, 2025 that a CourtListener and GovInfo search for mandamus and agency delay finds, Passim holds 1,371 (85%), measured November 3, 2026. Some of what that search finds are not mandamus decisions. Tell us about a decision we're missing.

These are counts of past written decisions and dockets, not predictions of how any case will be decided. A distribution appears only when at least 5 decisions (or 5 dockets) are behind it; smaller groups are listed, not summarized. How it's counted.

Read these counts as a record of past decisions, not a rate at which this judge rules for either side and not a prediction. Each decision turned on its own facts, its posture and what the parties asked for; the suits that reach a written decision are the contested minority. Read the decisions before drawing anything from them.

How the decisions came out

This page counts 6 of the group's written decisions, from 6 cases; the newest of them was decided November 17, 2010. The line at the top counts the whole group.

6written decisions
6cases they come from
0decisions weigh the TRAC factors
–median months from the application to the suit

Each case counts once in these outcome counts, by its latest decision here. A motion to dismiss denied and a later judgment for the government count as the judgment. How decisions are grouped into cases.

Disposition, by case

  • Granted (either side's motion)1
  • Dismissed4
  • Dismissed as moot1

Prevailing party, by case

  • Government5
  • Neither side1

Months from the application to the suit

Fewer than five cases state it (2); no distribution is shown.

Months from the application to the court's decision

Fewer than five decisions state it (4); no distribution is shown.

Leaves out the 1 decision holding that the agency had acted: its months to the decision are not a delay.

What the suits asked for, by case

  • Decide the application1
  • Issue a work permit1
  • Some other action2
  • Not recorded2

Agency, by case

  • USCIS3
  • CBP1
  • ICE1
  • Department of State1

Form or filing, by case

  • I-485 (adjustment of status)2
  • I-140 (immigrant worker petition)1
  • I-765 (EAD)1

Decisions by year

TRAC rulings

None of the judge's decisions in the group weighs the TRAC factors.

Every decision, by case

All 6 of the judge's decisions in the group, in 6 cases, newest first. Each case's latest decision comes first and is the one the counts above count; the case's other orders follow it. Download every decision as a spreadsheet (CSV).

DecisionDecidedAgencyFormDispositionPrevailing party
Fox v. Clinton November 17, 2010 Department of State Dismissed Government
Semati v. Mayorkas November 4, 2010 USCIS I-765 Dismissed as moot Neither side
George v. Napolitano March 16, 2010 USCIS I-140, I-485 Dismissed Government
Safadi v. Novak September 2, 2008 USCIS I-485 Dismissed Government
AWADH v. Chertoff June 24, 2008 ICE Dismissed Government
James v. United States Customs & Border Protection March 3, 2008 CBP Granted in the government's favor Government

How agency-delay suits ended

Most suits end without a written decision: the agency acts and the parties dismiss. These counts come from the docket record (the federal courts' Integrated Database and CourtListener) and count every mandamus, APA and section 1447(b) suit against any immigration agency, whatever year it was filed, among the dockets that name this judge. Check a delay counts fewer: a court's suits of the last three years against the form's agency.

1 docket: fewer than 5, so no distribution is shown. List it.