How to Find Federal Court Records for Historical Research
Federal Court Records: NARA, PACER, and the Paper Trail of American Justice
You've now learned how to read a source critically — to ask the hard questions about who created it, what they left out, whose interests it served. But critical reading only gets you so far if you can't actually find the source in the first place. The federal court record you analyzed earlier (Robert Williams's complaint) represents just one document in an archive system containing over 2.2 billion pages. This section is about the architecture of that system: where these records live, how they're actually organized, and how to navigate it without getting lost.
Here's what makes federal court records special. A single case file can contain sworn testimony, financial records, personal correspondence entered as exhibits, medical records, photographs, maps, expert reports, and the names of dozens of people who'd rather not be there. Courts compel disclosure in ways that almost no other institution does. When someone gets sued, when someone gets accused of a crime, when a business goes bankrupt — they have to produce documentation. That compelled transparency is now sitting in an archive or a database waiting for you. And unlike institutional record-keeping, which is voluntary, court records exist because the law demanded them. That's what makes them so reliable as evidence.
The Federal Court System: A Quick Map
If you're new to federal courts, here's the structure. It matters because knowing which court heard a case tells you where the records ended up.
District Courts are where most of the action happens — where cases are actually tried, where witnesses testify, where most of the documentary evidence gets filed. There are 94 federal judicial districts covering the United States and its territories. When people say "federal court records," they usually mean district court records. You'll find civil cases (contract disputes, civil rights claims, anything involving federal law), criminal prosecutions, and bankruptcy cases all happening here.
Courts of Appeals (Circuit Courts) sit above the district courts and hear appeals from them. There are 13 circuits. Their records don't contain the massive case files you'll find at trial courts — instead you get briefs, the lower court record, and the appellate decision itself. But those briefs are gold if you want a summary of what actually happened at trial. The appellate decisions are where lawyers lay out their best legal arguments in writing, which sometimes tells you things the trial record alone won't.
The Supreme Court reviews a tiny fraction of cases from the circuit courts (and occasionally from state supreme courts). It's handled differently from the rest of the system, which we'll get to.
graph TD
A[U.S. Supreme Court] --> B[13 Courts of Appeals / Circuit Courts]
B --> C[94 District Courts]
C --> D[Bankruptcy Courts]
C --> E[Civil Case Files]
C --> F[Criminal Case Files]
The Records Themselves: What You're Actually Looking At
When someone says "court record," they don't mean one document. They mean a collection of everything generated across the life of a case. Understanding what types exist helps you know what to ask for and what to expect.
Civil case files contain the complaint (which starts the whole thing), the answer, motions and responses, discovery materials like interrogatories and depositions, exhibits, trial transcripts if there was a trial, and the final judgment. For historical civil cases, this is often where you find personal letters, business records, and property descriptions that somebody had to enter into evidence. This stuff wouldn't have survived anywhere else.
Criminal case files include indictments, arrest records, bail documents, plea agreements, trial transcripts, jury instructions, verdicts, and sentencing orders. Pre-sentence reports show up here too — they contain biographical summaries of defendants that are invaluable if you're doing genealogical or biographical research. Fair warning: these reports are often sealed or restricted now, which is something to keep in mind when you're planning your search.
Bankruptcy records deserve their own mention. When someone files for bankruptcy, they have to disclose detailed financial information: assets, liabilities, income, debts, who they owe money to. For business research, a bankruptcy file can reveal organizational structures, subsidiary relationships, and financial histories that companies would never publish willingly. For genealogy or local history, bankruptcy records from the 19th and early 20th centuries sometimes contain personal statements, property inventories, and lists of business relationships that light up daily economic life in surprising ways. Important note: NARA holds all bankruptcy case files at the National Archives at Kansas City, which is a detail worth remembering when you're planning your trip.
Docket sheets are essentially indexes to everything filed in a case. Even if you can't access the underlying documents, the docket tells you the case number, parties involved, dates, and a one-line summary of every filing. Start here when you're orienting yourself in an unfamiliar case.
Tip: The docket sheet is your map. Before you order or download anything else, read through the whole thing to understand what actually happened in the case — when it was filed, what motions got contested, whether it went to trial or got settled. That context makes every individual document make sense.
Historical Records at NARA: The Pre-Electronic Era
For cases that closed more than roughly 15 years ago, your first stop is the National Archives and Records Administration. Court records don't stay with the courts forever. When a case closes, the paper file eventually gets transferred first to a Federal Records Center and then, if it's old enough or historically significant enough, to a NARA facility.
The geography matters more than you'd think. NARA's court records are scattered across regional facilities, and which facility holds a given case depends on where the court was located. New Hampshire federal court records are at NARA Boston (physically in Waltham, Massachusetts). Records from New York, New Jersey, Puerto Rico, and the U.S. Virgin Islands circuit courts are at NARA Philadelphia. Some states are split between two facilities depending on the time period. NARA staff can help you navigate those splits, which they do routinely.
To find a specific historical case, you need to know:
- Which court heard it (which district, which circuit)
- Roughly when it was filed or decided
- The case number, if you have it, or at least one party's name
If you have a case name but no number, start with the National Archives Catalog at catalog.archives.gov, which has finding aids for many court record collections. Some collections are described at the series level, meaning you know NARA holds records from that court for that period, but you'll need to contact the facility directly to find a specific case within the boxes.
Warning: Not all cases survive. Courts have records retention schedules — approved by both the Judicial Conference and NARA — that determine what gets kept and what gets destroyed. A routine civil case from the 1950s might be gone. A high-profile criminal case almost certainly survived. If you can't find something, it may have been legally destroyed, not just misfiled.
For very old records — anything from the 19th century or early 20th — the physical condition and how they're organized varies wildly. Some collections have been microfilmed or digitized. Others are original paper, sometimes in clerk's handwriting that takes practice to decipher. When in doubt, contact the regional facility before making a trip. NARA's staff are genuinely helpful; they can tell you whether a collection has a finding aid, whether it's actually accessible, and what format it's in.
CM/ECF and PACER: The Electronic Era
Starting in the late 1990s and accelerating through the 2000s, federal courts migrated to an electronic filing system called CM/ECF — Case Management/Electronic Case Files. Today, virtually all federal appellate, district, and bankruptcy court filings are electronic, and once something hits CM/ECF, it's immediately available to the public through PACER.
PACER (Public Access to Court Electronic Records) is the public-facing system for CM/ECF. You register for a free account at pacer.gov, and then you can search case indexes and download documents from any federal court with the system. It's clunky by modern standards — it looks like it was built in 2003 because it pretty much was — but it works.
Once you have an account, the most useful starting point for finding cases involving a specific person or organization is the Party/Case Locator at pcl.uscourts.gov. Reuters reporter Brad Heath, who has used PACER extensively for investigative journalism, recommends starting there: "That gets you to this search, which gives you access to the nationwide index of federal court cases and you can get from here into basically everything." Search by a name, get back a list of cases across all participating courts, drill down into individual dockets.
From there the workflow is straightforward:
- Find the case in the Party/Case Locator
- Pull up the docket report
- Read through the docket entries in order
- Download the specific documents you actually need
PACER's search limitations are real. You can search by party name, case number, attorney name, and date range — but you cannot do full-text search within documents. If you want to find all cases mentioning a specific drug, a specific address, a specific allegation, PACER won't do that. Third-party services like CourtListener (discussed below) offer better full-text search.
Cost Structure and Fee Waivers
PACER charges per page — currently $0.10 per page, with a $3.00 per-document cap regardless of length. That sounds cheap until you're downloading a 500-page civil case file and realize you've spent $30, or you're doing systematic research across hundreds of cases and the bills add up.
A few cost facts worth knowing:
- Fees are waived if your quarterly bill is under $30. Light or infrequent users often pay nothing.
- Written opinions are free on PACER. Courts publish decisions at no charge.
- You can view records for free at the courthouse. If you live near the courthouse, you can read everything on-site without paying download fees.
- Fee waivers exist for researchers who can show the information serves the public interest and won't be used commercially. Academic researchers and journalists sometimes qualify; you apply through the Administrative Office of the Courts.
Tip: PACER charges creep up faster than you'd expect when you're doing systematic research across multiple cases. Before downloading every document in a large docket, read the docket entries carefully and identify only the filings most likely to contain what you need. The complaint, key motions, and final judgment usually give you 80% of the story at a fraction of the cost.
CourtListener and RECAP: The Free Alternative
Here's the most important thing to know about PACER's cost structure: there's a substantial free alternative built on top of it. CourtListener, operated by the nonprofit Free Law Project, works through a browser extension called RECAP (PACER backward — yes, really). When RECAP users download documents from PACER, the extension automatically uploads a copy to CourtListener's public database. Over time, this has built up a massive freely-accessible collection of federal court documents — millions of records that would otherwise cost money to access.
The advantages of CourtListener over raw PACER:
- Free access to millions of documents already in the database
- Full-text search across documents (something PACER doesn't offer)
- Docket alerts — you can set up email notifications when new filings appear in a case, which PACER offers only through RSS in some courts
- Better search interface for finding cases by topic or content
- No per-page charges for documents already in the database
The catch is coverage: CourtListener only has documents that RECAP users have previously downloaded. For high-profile cases that lots of people follow, coverage is usually complete. For obscure cases in smaller districts, you might find nothing. Check CourtListener first; if the document you need isn't there, you'll need to pay PACER.
RECAP is available as a browser extension for Chrome and Firefox. Install it before your next PACER session and every document you download automatically becomes part of the public commons.
Sealed Documents, Redactions, and Privacy Restrictions
Not everything in a federal court case is public. Understanding what's sealed and why tells you something about the case itself.
Privacy redactions are mandatory. Federal rules require that anyone filing documents must redact specific personal information: Social Security numbers, taxpayer ID numbers, dates of birth, names of minor children, financial account numbers, and (in criminal cases) home addresses. This is why you'll see "XXX-XX-1234" where a Social Security number should be. The full information exists somewhere in the sealed court file, but the public version gets redacted.
Sealed documents are different — these are entire filings withheld from public access. Judges have discretion to seal documents in certain circumstances: protecting informants in criminal cases, shielding classified national security information, preventing disclosure of trade secrets in civil litigation, or protecting ongoing investigations. Some categories are sealed automatically: unexecuted warrants, bail and presentence reports, juvenile records, and documents that might reveal defense strategies.
When you see a sealed entry on a docket, you know something exists but can't access it. For a researcher, a sealed docket entry is itself information — it tells you something was sensitive enough to warrant protection, and the surrounding unsealed documents sometimes reveal what the sealed one likely contained.
For historical researchers: sealed documents don't remain sealed forever. Courts can unseal records as circumstances change, and researchers can petition for unsealing. It's uncommon, but journalists have successfully petitioned to unseal records in cases of significant public interest, sometimes decades after the original proceedings.
Supreme Court and Appellate Records
The Supreme Court operates outside PACER entirely. For researchers, this is actually an advantage: the Court's records are more consistently accessible and often better organized than district court holdings.
The Supreme Court's website (supremecourt.gov) provides free access to all opinions, argued cases, and briefs from recent terms. For historical records, NARA holds Supreme Court case files dating back to the early republic. The Oyez Project (oyez.org) offers searchable audio of oral arguments from 1955 forward — you can actually hear the arguments in landmark cases, which is an extraordinary resource.
For the Courts of Appeals, PACER and CM/ECF work the same as in the district courts. Appellate briefs are particularly useful because they typically summarize the entire factual record from the district court, giving you a narrative account of what happened at trial. If you're trying to understand a complex case and the trial record is enormous, the appellate briefs give you the Cliff's Notes version.
Historical circuit court records follow the same regional NARA geography as district court records, with one notable exception: circuit court records from New York, New Jersey, Puerto Rico, and the U.S. Virgin Islands are held at NARA Philadelphia.
Using Court Records Alongside Other Sources
Court records don't exist in isolation. The most productive research treats them as nodes in a larger network of documentation.
A few cross-referencing strategies that consistently pay off:
Newspapers + court records: When a case was newsworthy, contemporaneous newspaper coverage explains the context that the legal filings assume you already know. Historical newspaper databases (covered earlier) can tell you who the key players were, what the community thought was at stake, and sometimes quote witness testimony that didn't make it into the official record. Conversely, if you find a compelling newspaper story, search for the underlying court case to get the documentation behind the headlines.
Census + civil records: If you're researching a property dispute or family matter from the late 19th or early 20th century, the census records from surrounding years can identify the other parties, establish their ages and households, and sometimes reveal family relationships that explain why the dispute happened. A civil case about inheritance makes much more sense when you can see the family structure in the census.
Land records + court records: Property disputes often involve deeds, surveys, and chain-of-title documents entered as exhibits. If you're researching land history, court records can preserve copies of documents that the original deed books no longer contain.
Court records + FOIA: Sometimes a civil or criminal case references government agency documents — FBI reports, agency investigations, regulatory correspondence — that were submitted as evidence. The case file often gives you enough identifying information (dates, agency names, file numbers) to make a targeted FOIA request for the underlying agency records. Think of court filings as a finding aid for government documents.
graph LR
A[Court Case File] --> B[Newspaper Coverage]
A --> C[Census Records]
A --> D[Land/Deed Records]
A --> E[FOIA Requests]
B --> F[Richer Context]
C --> F
D --> F
E --> F
A Practical Research Workflow
Here's how an actual court records research session actually goes:
Start by figuring out whether your case is likely historical (pre-2000, probably at NARA) or recent (post-2000, probably in PACER/CM/ECF). For anything in the past 15 years, the case is almost certainly still at the court itself; contact the clerk's office if PACER doesn't have what you need.
For historical cases, search the National Archives Catalog and contact the relevant regional facility. Know the district, know the approximate year, and bring any other identifying information you can gather from secondary sources — newspaper accounts, published histories, family documents. The more you know going in, the faster the NARA staff can help you.
For recent cases, start with CourtListener to see if the documents you need are already freely available. If not, register for PACER (free to register; you pay only when you download), use the Party/Case Locator to find your case, read the docket carefully before downloading anything, then target the specific documents most likely to contain your information.
In either scenario, document everything you find, including what you searched for and didn't find. Negative results matter. If a company was involved in dozens of lawsuits and you can verify that none involved a specific allegation, that's meaningful data. The paper trail of American justice is genuinely vast — 2.2 billion pages at NARA and growing — but once you understand how the systems work, they're more rational than they first appear.
The researcher who understands how courts work, what records they generate, and where those records land when a case closes has access to a category of historical documentation that most people don't even know exists. That's the real advantage.
Only visible to you
Sign in to take notes.