Is this case still good law?
You found the case. It says what you need it to say. Before it goes in the brief, one question has to be answered.
The methods, the free options and what they miss, and the treatment no citator will show you.
Last reviewed 2026-08-24
Two questions, not one
"Good law" collapses two separate questions. Keeping them apart is the whole discipline.
Direct history is what happened to this case, in this litigation. Affirmed, reversed, vacated, remanded, modified. Direct history is a fact about the docket.
Citing references are what other courts did with it afterward, in other litigation. Overruled it, declined to follow it, distinguished it on facts close to yours, criticized the reasoning, limited it to its facts. Citing references are a fact about the case's authority.
A case can have clean direct history and still be worthless to you. It was never appealed, so nothing happened directly. Six later courts in your circuit read it narrowly. Nothing about the docket tells you that.
| Term | What happened | Can you still cite it |
|---|---|---|
| Overruled | A court with authority over the deciding court rejected the holding. | No. Not for the overruled proposition. |
| Reversed | The appellate court reversed the judgment in this same case. | No. The reasoning did not survive. |
| Vacated | The judgment was wiped, often on procedural grounds or by settlement. | No. A vacated opinion has no precedential force. |
| Abrogated | A later decision undercut the holding without naming the case. | Rarely. This is the one that gets missed. |
| Superseded by statute | The legislature changed the rule the case interpreted. | Only for the pre-amendment period. Check the effective date. |
| Limited | A later court confined the holding to narrower facts. | Yes, inside the narrowed boundary. Read the limiting case. |
| Criticized | A later court disagreed with the reasoning but could not overrule. | Yes. Expect opposing counsel to quote the criticism. |
| Distinguished | A later court found the facts materially different. | Yes. If the distinguishing facts match yours, that is a warning. |
The distinction that matters most. Overruled, reversed, and vacated are binary. The case is out. Abrogated, limited, criticized, and distinguished are matters of degree, and no red flag appears for most of them. A brief that survives is one where somebody read the citing cases, not just the flag.
The free methods, and what each one misses
You can verify a case for zero dollars. What you cannot do for zero dollars is verify it quickly and completely. Here is the honest accounting.
Google Scholar
Free. Full text of most federal and state appellate opinions, plus a "How cited" panel listing later cases. Coverage is broad and the citing list is real.
Misses: no editorial treatment analysis. Google tells you a later case cited this one. It does not tell you whether the citation was approving, distinguishing, or overruling. You read every citing case yourself, or you learn nothing.
CourtListener
Free, run by the nonprofit Free Law Project. Millions of opinions, full dockets through RECAP, and a citation network showing both cited and citing authorities. This is the same public record Marcella indexes.
Misses: the citator layer is signal without a verdict. You get the network. You supply the legal judgment.
Your bar association benefit
More than 40 state bar associations provide a free Fastcase or vLex tier with membership. If you pay bar dues, check before you pay anyone else. This is the most underused research asset in the profession.
Misses: free bar tiers generally exclude the advanced AI features and, depending on the state, the full citator. Verify what your package includes rather than assuming.
The official reporter and the docket
Pull the opinion from the court's own site or from PACER. For direct history the docket is the primary source and nothing overrides it.
Misses: nothing about this case. Everything about every other case. The docket cannot tell you the Ninth Circuit abrogated the holding last year.
The free stack, in order
- Pull the opinion from the official reporter or the court's site. Confirm the citation is real and the quote is accurate.
- Check the docket for direct history. Appealed, reversed, vacated, or settled after judgment.
- Run the case through Google Scholar or CourtListener and open the citing list.
- Read every citing case from a court that binds yours. Not the headnote. The passage discussing your case.
- Find the statute or rule the case interpreted and check whether it has been amended since.
That sequence works. It costs twenty minutes to two hours per case depending on how heavily the case has been cited. Multiply by the number of authorities in the brief.
The paid citators
KeyCite on Westlaw, Shepard's on Lexis, and BCite on Bloomberg Law do the same core job. An editorial layer reads the citing cases, assigns a treatment verdict, and surfaces it as a flag.
That editorial layer is what you are paying for. It is genuinely valuable and it is why the incumbents have held this market for four decades. It is also why a research subscription costs what it costs. Published rates are on the Westlaw pricing page.
Each citator also reads only its own publisher's silo. That is a separate problem and it has its own page: the silo problem.
A flag is a starting point, not a clearance. A green symbol means the editorial layer found no negative treatment it classified as significant. It does not mean no court quietly limited the holding, and it does not mean the case fits your facts.
The failure nothing flags
There is a category of negative treatment below the threshold of every citator: the later opinion that reaches an incompatible result without ever naming the earlier case.
It happens constantly. A panel decides a question. Another panel decides the same question the other way three years later without citing the first opinion. The conflict sits unresolved until somebody briefs it. No citator flags a case that was never cited. The treatment is real and the signal is absent.
The same gap opens when a statute is amended and the interpreting case is never formally superseded, when a case is abrogated by implication, and when an unpublished decision from your own district cuts against the published authority you found.
Closing that gap requires reading the later law in the area, not checking a symbol. It is the most expensive part of verification and the part most likely to get skipped under deadline.
How Marcella handles it
Marcella retrieves from an indexed corpus of more than 10 million judicial opinions. If a case is not in the index it does not appear in the output. That constraint is architectural, which is why a fabricated citation cannot be produced. That is the Mata problem, and the architecture answers it rather than reducing its odds.
The citation treatment engine maps how courts have cited and applied authority across jurisdictions from the public record, which is why it crosses publisher silos instead of reading one. Every output carries a six-item Daubert attestation: real source, good law, overruled, distinguished, questioned, treatment history.
Marcella does not replace the reading. It removes the part where you find out at the hearing.
Questions
Can I Shepardize a case for free?
Not literally. Shepard's is a LexisNexis product and requires a subscription. You can perform the equivalent verification for free using CourtListener or Google Scholar plus the docket, at the cost of reading the citing cases yourself rather than reading an editor's verdict. More than 40 state bars also provide a free Fastcase or vLex tier that includes some citator function.
Does a green flag mean the case is safe to cite?
It means the citator's editorial layer found no negative treatment it classified as significant. It does not account for cases that reached an incompatible result without citing yours, for unpublished decisions in your district, or for whether the holding fits your facts. Treat a flag as the beginning of verification, not the end.
What is the difference between overruled and abrogated?
Overruled means a court with authority over the deciding court expressly rejected the holding. Abrogated means a later decision undercut the holding without naming the case. Abrogation is harder to catch because the later opinion contains no reference a citation network can follow.
How long does it take to verify one case properly?
Twenty minutes to two hours using free tools, depending on how many times the case has been cited. A paid citator compresses the first pass to seconds and leaves the reading of significant citing cases in place. Neither approach removes the reading.
What happens if I cite a case that was overruled?
Under Fed. R. Civ. P. 11 a filed document must have legal support. Courts have imposed monetary sanctions, bar referrals, and license suspensions for filings built on authority that did not hold. The running record of those decisions is maintained at heppner-problem.com.
Marcella runs inside your firm's existing Microsoft 365 tenant. Client data never leaves it. Waitlist members get 30 days free on any paid plan at launch.
Join the Waitlist