Public-domain · open source
OpenJurist

The Empirical Record

Not summaries of the law — measurements of it.

The idea

Most legal research tools tell you what an editor thinks a case means. OpenJurist measures what courts actually did with it: which later decisions cited it, which sentences they quoted, whether they followed it or pulled away from it, and how all of that changed over time. That body of evidence — built from the citations, quotations, and treatments running through millions of published opinions — is what we call the empirical record of the law.

The record is held to an evidentiary standard. A quotation appears only when we find the exact words in the cited opinion. Machine-derived analysis is labeled as machine analysis and links to the methodology that produced it. When a claim can't be verified against source text, it isn't shown.

The corpus it's built from

Millions of published federal and state opinions; the United States Code, versioned across its release points; the state codes we host; the founding documents; law dictionaries spanning two centuries; and the biographies of the federal bench. All of it public domain, all of it cross-linked — tens of millions of citation links connecting cases, statutes, and judges into one navigable graph.

What is built so far

Each capability has its own page explaining what it measures, with live examples to try.

How to verify what we show

Every empirical surface links to its methodology page, every methodology names its limits, and our editorial and sourcing policy explains how content is compiled and reviewed. The primary sources — the opinions and statutes themselves — are on the site, one click from every claim built on them.