

Judicial claim correction has a small but important role in patent law. As Gideon Korrell explains, in a recent Federal Circuit decision, Canatex Completion Solutions, Inc. v. Wellmatics, LLC, the court clarified when mistakes in patent claims can be fixed instead of being used to invalidate a patent. This case shows how courts balance strict rules with fairness to patent owners.
Understanding Judicial Claim Correction
Judicial claim correction allows a court to fix small errors in patent claims during a case. However, this power is used very carefully to protect how patents inform the public.
The rule comes from the Supreme Courtâs decision in I.T.S. Rubber Co. v. Essex Rubber Co.. Over time, courts have turned this into a strict three-part test:
- The mistake must be clear from reading the patent
- The correction must not be open to reasonable disagreement
- The fix must be a small clerical or typing error
Courts stress that this rule is narrow. It is not meant to rewrite claims but to show what the patent always intended to say.
The Canatex Case and the Drafting Error
In the Canatex case, the patent described a mechanical device used in oil and gas work. The invention had two parts: a âfirst partâ placed deeper underground and a âsecond partâ closer to the surface.
The problem came from the claim language saying a mechanism would ârelease the connection profile of the second part.â This created two issues:
- There was no clear support for a âconnection profileâ on the second part
- The wording did not match how the device actually worked
Throughout the patent, only the first part was described as having a connection profile. The second part only had parts that hold and release that profile. Because of this mismatch, the district court first said the claims were unclear.
Why the Federal Circuit Allowed Correction
The Federal Circuit disagreed and corrected the claim. It found the mistake was clear and that only one fix made sense: changing âsecond partâ to âfirst part.â
The court relied on key parts of the patent, including:
- The full claim language
- The written description
- The drawings
All of these showed that the connection profile belonged to the first part. The court said a skilled reader would quickly see the mistake.
The court also rejected the idea that repeating the wrong word made it intentional. Just because an error appears more than once does not mean it was done on purpose.
PTO Denial vs. Judicial Authority
During the case, the patent owner asked the Patent Office to correct the mistake, but the request was denied. The Office said the change might affect the scope of the claim.
The Federal Circuit said this denial did not decide the issue. Court corrections and Patent Office corrections follow different rules. Courts look at what the claim always meant, while the Patent Office may look at changes differently.
This shows that courts have the final say on claim meaning in legal cases.
Distinguishing Earlier Case Law
The court also discussed Chef America, Inc. v. Lamb-Weston, Inc., a case often used to refuse corrections. In that case, the court did not fix the claim because more than one meaning was possible.
In contrast, the Canatex case had only one clear meaning. There was no sign that the mistake was intentional or that other meanings could work.
Practical Implications for Patent Survival
This decision gives some useful lessons for patent owners and lawyers:
- Clear mistakes may not destroy a patent if the meaning is obvious
- The patentâs own content is very important when asking for correction
- Courts will not rewrite claims, but they will fix simple mistakes
- Patent Office decisions do not control court rulings
At the same time, the case shows why careful writing is important. Even small mistakes can lead to expensive legal fights.
Conclusion
Judicial claim correction helps protect patents from being lost due to small drafting mistakes. The Canatex decision shows that courts will protect patents when the meaning is clear and not in doubt.
Gideon Korrellâs view matches the courtâs approach: correction should happen only when it is clearly needed. This keeps patents reliable for the public while avoiding unfair loss of rights because of simple errors.





