About the Authors

  • The Authors and Contributors of "Patent Docs" are patent attorneys and agents, many of whom hold doctorates in a diverse array of disciplines.
2018 Juristant Badge - MBHB_165
Juristat #4 Overall Rank

E-mail Newsletter

  • Enter your e-mail address below to receive the "Patent Docs" e-mail newsletter.

Contact the Docs


  • "Patent Docs" does not contain any legal advice whatsoever. This weblog is for informational purposes only, and its publication does not create an attorney-client relationship. In addition, nothing on "Patent Docs" constitutes a solicitation for business. This weblog is intended primarily for other attorneys. Moreover, "Patent Docs" is the personal weblog of the Authors; it is not edited by the Authors' employers or clients and, as such, no part of this weblog may be so attributed. All posts on "Patent Docs" should be double-checked for their accuracy and current applicability.
Juristat #8 Overall Rank


« BIO Top Brass Meet the Press | Main | WIPO to Offer Seminars on PCT »

September 10, 2009


Kevin, I agree that it's sometimes difficult for district courts to distinguish between permissibly relying on the specification to construe claims, on the one hand, and impermissibly reading limitations from the specification into the claims on the other hand. But I don't think case illustrates that difficult - this one isn't even a close call. The claim recites an optically pure compound. The specification inherently defines optically pure as being free from the other enantiomer (100% enatiomeric excess), so there's no indefiniteness issue. And the specification teaches at least one method to make the optically pure material, viz. via HPLC, so there's no written description or enablement issue. That should have been the end of the discussion. There may or may not be other methods to make the claimed material, but that's irrelevant, because anyone who makes or possess the claimed material infringes. Note also that the spec says a separate application was filed directed to the purification method per se (although I didn't investigate the fate of that application), which would seem to auger for a process-free claim construction in the patent at issue.


"Quixotic jurisprudence on claim construction" eh? We are waxing "lyrical" today aren't we?

I do agree with you that the way the Federal Circuit handles claim construction can be very frustrating for district court judges. And Judge Newman has written some fairly "volcanic" dissents about how unrealistic it is to treat claim construction so much as a question of law, which is what Cybor commands. That, like this case, the Federal Circuit so often parses through the specification and file history suggests otherwise.

Dear Dan:

I don't dispute the CAFC's determination, just that claim construction is a place where reasonable people can differ. One way of looking at the spec and prosecution history is the way you and the court did. Another is to interpret the evidence as meaning that HPLC was the only way (or asserted to be the only way) to achieve this degree of optical purity. I don't know the answer, and I don't know if the district court considered any evidence on this point.

But my point is that these types of questions are best considered at trial, where testimony can be elicited on this topic. Clearly, if HPLC was but one tool in the chemist's toolbox for making optically pure oxaliplatin, then the district court made an error in putting so much weight on the fact that the patentee chose to use HPLC to go from 90% pure (the prior art) to 100% pure. But we don't know if that is the case; here, the issue was framed as one of pure claim construction.

Having claim construction at the district court be but a way-station to Federal Circuit review is a bad idea on many fronts. It is excessively costly, it delays final resolutions, it promotes premature summary judgment decisions, and it puts in the appellate court's hands questions that have sufficient factual predicates that some assistance from the district court might prove helpful in many cases. But why should a district court put in the time and effort to "get it right," if it knows any decision it makes will be second-guessed and likely overturned?

Finally, the current system makes patent claim meaning a crap shoot of which panel of the court you happen to get. This wasn't the increase in certainty that Congress wanted when it created the court. (Ironically, the Federal Circuit's animus against the doctrine of equivalents stems from the court's reluctance to have patentees introduce a similar level of uncertainty to patent claim "construction.")

I have no opinion about whether the district court or the Federal Circuit "got it right" - there are arguments for both positions. But I think the discussion about the process is important, so thanks for contributing.

Dear EG:

What do you have against lyricism once in a while?

Dear Kevin,

None, I was being very "tongue in cheek" on that! And I too waxed "lyrical" on the Cardiac Pacemarkers case. See IPWatchdog.

Just a thought but is the problem not that the Federal Circuit regards itself as somehow better qualified than the District Court judges to "do" claim construction properly, and that despite the factual background to claim construction, established by witness evidence at trial.

To me, from England, it seems very simple. Claim construction is a matter of law, so you need an appeal instance that is competent to lay down clear law, District Court judges who are competent to follow that law, and (to complete the virtual circle) CAFC judges who defer to competence, and reverse only when they feel they have no choice.

Otherwise, with rules that contradict each other, it's nothing but a downward spiral.

Dear Max:

No doubt, but the practice of having district court judges "sitting by designation" may help.

I think rather than a downward spiral it's more of a sine wave (or a roller coaster), but in any event the present practice doesn't promote certainty.

Thanks for the comment.

The comments to this entry are closed.

May 2024

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31