Whither the Reverse Doctrine of Equivalents

by Dennis Crouch

The Reverse Doctrine of Equivalents can shrink the scope of a patent to exclude acts literally covered by the claims.  The Supreme Court in Graver Tank explained that the doctrine comes up to prevent abuse of the patent system:

[W]here a device is so far changed in principle from a patented article that it performs the same or a similar function in a substantially different way, but nevertheless falls within the literal words of the claim, the doctrine of equivalents may be used to restrict the claim and defeat the patentee’s action for infringement.

Graver Tank & Mfg. Co., Inc. v. Linde Air Prods. Co., Inc., 339 U.S. 605 (1950).  The Graver Tank decision traced its provenance back to Westinghouse v. Boyden Power Brake Co., 170 U.S. 537 (1898) (“correspond[ence] with the letter of the [claim] does not settle conclusively the question of infringement.”).  Both Westinghouse and Graver Tank describe this scope-shrinking as an aspect of the doctrine of equivalents that can either increase or decrease literal scope.   The term “reverse doctrine of equivalents” was seemingly first used by Judge Markey in a 1977 decision rejecting application of any reverse doctrine of equivalents in the validity context. Ex parte Hogan, 559 F.2d 595, 607 (Cust. & Pat. App. 1977).   Hogan places the term within quotation marks–“reverse doctrine of equivalents”–but did not indicate whether those quotes were for emphasis or citing some prior usage. The usage was later solidified by the Federal Circuit in SRI International v. Matsushita Electrical Corp. of America, 775 F.2d 1107 (Fed. Cir. 1985).  In Roche Palo Alto LLC v. Apotex, Inc., 531 F.3d 1372 (Fed. Cir. 2008), the court endeavored to cabin-in the doctrine’s use by explaining that (1) “The reverse doctrine of equivalents is rarely applied”; and (2) the Federal Circuit “has never affirmed a finding of non-infringement under the reverse doctrine of equivalents.” Id.

The Reverse DOE has not seen much action of late.  In Steuben Foods, Inc. v. Shibuya Hoppmann Corp., CV 19-2181-CFC-CJB, 2021 WL 4775996, at *1 (D. Del. Oct. 13, 2021), the court denied summary judgment.  The defendant apparently admitted that its product was covered by the literal scope of the patent, but argued no infringement based upon the reverse doctrine of equivalents.  In particular, the defense argued that its sterile region was substantially different than that disclosed by the invention.  The jury was asked whether the accused machine “is no more than insubstantially different from the second sterile region of the invention.”  The jury marked the box labelled YES-INFRINGED.

Unlikely that the defendant will choose this issue to appeal given the history of the Federal Circuit and reverse DOE, but we’ll see.  In another recent Delaware decision, the court noted that Teva had initially raised the reverse doctrine of equivalents, but later dropped that defense.  Vifor Fresenius Med. Care Renal Pharma Ltd. v. Teva Pharm. USA, Inc., CV 18-390 (MN), 2022 WL 3562555, at *6 (D. Del. Aug. 18, 2022).  In, Kewazinga Corp. v. Microsoft Corp., 558 F. Supp. 3d 90, 112 (S.D.N.Y. 2021), the patentee sought summary judgment rejection of Microsoft’s reverse DOE defense. The district court refused, holding that any decision regarding reverse DOE must follow a finding regarding literal infringement. Id.

In many ways, the Reverse Doctrine of Equivalents has the potential of doing some of the same work as our doctrines of full-scope enablement & written description. A key difference though is that the outcome is a narrowed patent claim rather than an invalid patent claim.

 

15 thoughts on “Whither the Reverse Doctrine of Equivalents

  1. 3

    The thing about the reverse doctrine of equivalent and claims scope is that you don’t know if a claim is “too broad” or if the claim scope is fully covered until a new non-obvious way to do the same thing that is claimed within the claim scope is invented.

    That is the key. Scope of enablement.

    1. 3.1

      The rDOE is really the only objective way to police claim scope.

      The other ways that are mentioned particularly “enabled the full scope of the claim” is subjective based on how a judge feels about your claim. The fact is that you have enabled the claim and there is a WD for the claim and any “full scope” nonsense is pure speculation. The judge doesn’t know what could or will be invented.

      The rDOE on the other hand relies on determining if the accused device literally infringes but performs the functions in a different non-obvious way. Those are things that are traditionally can be determined. It is merely an infringement analysis, a reverse DOE analysis, and a 103 analysis.

      The “full scope” nonsense is like “directed to an abstract idea.” You may as well just write a statutory that says, “The judge may read the claims and invalidate the claims based on how he or she feels about the claims.”

      1. 3.1.1

        The rDOE is really the only objective way to police claim scope. The other ways that are mentioned particularly “enabled the full scope of the claim” is subjective based on how a judge feels about your claim… The rDOE on the other hand relies on determining if the accused device literally infringes but performs the functions in a different non-obvious way.

        Two points:

        (1) Asking an English-major district court judge to determine whether “the accused device literally infringes but performs the functions in a different non-obvious way” is every bit as subjective as asking the judge to determine whether the full claim scope is enabled.

        (2) The application of RDoE occurs entirely in the context of Markman, which is a step in the process that belongs entirely to the judge. By contrast, the determination of whether the full claim scope is enabled is one of those mixed questions that is typically entrusted to the jury.

        1. 3.1.1.1

          (2) I think the way the RDoE is done now is flawed.

          (1) I don’t agree. I think that “full claim scope” is wildly subjective. Why? Because the question is speculating on whether other solutions to the problem exist.

          Basically, I think that scope of enablement with a modified RDoE is the way to go. And, you know, there is no reason that the claims couldn’t be invalidated under the RDoE, which I think may be a fair outcome.

        2. 3.1.1.2

          There are other RDoE cases that analyze this a bit differently I think.

          Been a few years since I read them. I might poke around and see what is out there.

  2. 2

    “the doctrine of equivalents may be used to restrict the claim”

    NEVER. Ulness, the aim is to make all Tech people to “go Galt”, hahahha

  3. 1

    In many ways, the Reverse Doctrine of Equivalents has the potential of doing some of the same work as our doctrines of full-scope enablement & written description. A key difference though is that the outcome is a narrowed patent claim rather than an invalid patent claim.

    Exactly, which is why the doctrine is little used. RDoE turns into a sort of heads-I-win,-tails-you-lose for the patentee. Right now, if a patentee writes a claim that exceeds the scope of that which the patentee has described and enabled, then the end result is that the patentee’s claim is held invalid and the patentee ends up with nothing. Under the RDoE, however, the worst that happens to the patentee is that s/he suffers a challenging Markman, at the end of which s/he ends up with a claim scope that does not quite reach as far as s/he needed in order to win.

    Under such a rule, it is much more worthwhile for the patentee to press its luck when drafting claims. That is not an incentive that society should want to create. It is better that the rule be that the claim fails as invalid when it exceeds the scope of that which is described and enabled in the supporting specification. That helps to keep everyone honest.

    1. 1.1

      Agreed. It would be a very unfortunate way to address functional claiming problems. On the other hand, invoking the principle behind the doctrine and citing the cases where the RDOE was applied might help a defendant who infringes a functional claim but who achieves the function in a substantially different way than the manner disclosed in the specification (or whatever manner the patentee happened to be in actual possession at the time of filing).

    2. 1.2

      +1

      This is exactly what I was thinking as I read this article, but I could not have said it better. In most, of not all, of the rest of the world such an over-broad claim would be invalid. Over time, the US courts seem to have been generous towards claims that have been drafted narrowly (DOE) or broadly (RDOE). To be fair, though, the US law does uniquely provide that a patent should be presumed valid, and one way of giving effect to this is to prefer a construction that preserves validity over one that does not.

      1. 1.2.1

        US law… uniquely provide[s] that a patent should be presumed valid…

        Is it true that the U.S. is unique in this respect? All that “presumed valid” means in this context is that it is the challenger who bears the evidentiary burden to prove the claim invalid, rather than the patentee bearing the evidentiary burden to prove the claim valid.

        Are there jurisdictions in the world where—once the defendant pleads invalidity as a defense—the evidentiary burden then shifts to the patentee to prove that the challenged claims are valid? Can anyone point me to an actual decision in which such an allocation of the evidentiary burden is applied?

        1. 1.2.1.1

          I am unaware of another jurisdiction with the equivalent of 35 USC 282(a), or where the standard of proof is ‘clear and convincing’ evidence. You are correct about the onus, but the usual ‘balance of probabilities’ standard typically applies, i.e. there is no presumption that examination has any positive effect on the likelihood of validity.

        1. 1.2.2.1

          I’m not convinced that a finger on the scales of justice is justified. Why should the court “prefer” a claim construction that preserves validity of the claim over one that does not? What social policy is thereby satisfied?

          Would it not be preferable for the court to divine, disinterested, what the claim means, and then adjudicate the consequences, deal with the cards as they fall? At least then those who are paid, years (or even decades) earlier, to render opinions on infringement and validity have a clear run, to call it as they see it.

          1. 1.2.2.1.1

            A few thoughts (both in response to your question and to Mark’s remarks above):

            (1) Although the U.S. may be unique in codifying a presumption of validity, I think that pretty much every jurisdiction around the world applies a presumption of validity. That is to say, everyone requires the patent challenger to bear the burden of persuasion, and a tie goes to the patentee everywhere.

            (2) The essential logic of the “clear and convincing” evidence requirement in the U.S. to prove invalidity rests on the intuition that it should be easier for an accused infringer to escape liability by demonstrating non-infringement (which only get the immediate defendant off the hook) than by demonstrating invalidity (which gets the immediate defendant off the hook, but also all potential future defendants). So far as it goes, this intuition seems correct to me.

            (3) The logic of the clear-&-convincing requirement, therefore, really offers no support for the idea that the claim construction should be bent or twisted to avoid a finding of invalidity. The claim construction should be the same whether the defendant pleads invalidity or mere non-infringement. The quantum of evidence necessary from the defendant changes according to those two, but not the claim construction.

            (4) There is, however, a rule that—where the claim is susceptible of two plausible constructions and one leads to invalidity but the other does not—then “res magis valeat quam pereat.” Corning v. Burden, 56 U.S. 252, 269 (1854). This rule does not seem crazy to me, although the contrary rule would be equally sane. In actual practice, however, it is very rarely applied. My experience is that the U.S. courts will not bend over backwards to save validity where a claim deserves to fail. You really have to be talking about a Markman dispute in absolute equipoise before the Corning rule will kick in.

            1. 1.2.2.1.1.1

              I hope my slow reaction to your valuable post, Greg, will not have the consequence that you miss this reply.

              I understand the logic of the Latin in Corning, but wonder whether the rarity of its application in a real case has some echo in the law of clarity/indefiniteness. I’m thinking of the “irresolvably ambiguous” idea. The text we write can never be so clear that a zealous lawyer finds absolutely no clarity attack available. But the chances of two rival claim constructions being exactly equally plausible, and that dilemma for the court being determinative of the case at bar, must be vanishingly small, I should have thought.

Comments are closed.