Reissue, Recapture, and Skirting Eligibility

by Dennis Crouch

Prior to 1952, the Reissue Statute text only allowed for (1) narrowing claim scope and (2) fixing a “defective or insufficient” specification.  The 1952 Patent Act edited the law to allow for reissue for “error[s]” where the patentee claimed either “more or less than he had a right to claim.” Thus, for the first time the statute expressly permitted broadening reissues.  At the time, Congress also added a  statutory caveat that broadening reissue must be applied for within two years of the patent grant.  On paper, these changes might appear significant, but they were not overly controversial since the Patent Office and Courts had already allowed broadening reissues along  with the two-year laches caveat.

The 1952 Act remained silent on another broadening caveat — the rule against recapture that the courts had developed in the 1880s in response to perceived abusive practices.  Following the ’52 Act, courts concluded that the rule against recapture remained effective, and in recent years have expanded its scope. The most recent expansion is found with In re McDonald, 2021-1697, — F.4th — (Fed. Cir. Aug. 10, 2022) (applying recapture to amendments to overcome eligibility rejections).

The recapture rule has ordinarily been used to prohibit later recapture of scope that had been eliminated in an attempt to avoid a prior art rejection.  Carl Moy’s treatise explains the rule against recapture as follows:

[The rule against recapture] seeks to hold the patentee to the positions she impliedly took during prosecution of the original patent regarding the scope and cont[ent] of the prior art.  Specifically, where the patentee relinquished claims of a given scope in the original application, in response to a rejection based on the prior art, this is taken as  tantamount to her admission that the relinquished claim is in fact unpatentable over the art. The patent owner is later held to that admission during reissue.

5 Moy’s Walker on Patents § 16:107 (4th ed.); See also, Robert A. Matthews, §4 Annotated Patent Digest § 25:76 (“Surrender arises from a patentee’s canceling or amending a claim to avoid a prior art rejection and relying on the amended matter as avoiding the prior art.”).

Likewise, the MPEP finds impermissible recapture only when the original surrender relied on by the patent applicant to overcome the prior art.

If an original patent claim limitation now being omitted or broadened in the present reissue application was originally relied upon by applicant in the original application to make the claims allowable over the art, the omitted limitation relates to subject matter previously surrendered by applicant.  The reliance by applicant to define the original patent claims over the art can be by presentation of new/amended claims to define over the art, or an argument/statement by applicant that a limitation of the claim(s) defines over the art.

MPEP 1412.02 (flowchart below as well).  See also, Mentor Corp. v. Coloplast, Inc., 998 F.2d 992, 996 (Fed. Cir. 1993) (“Error under the reissue statute does not include a deliberate decision to surrender specific subject matter in order to overcome prior art.”); Medtronic, Inc. v. Guidant Corp., 465 F.3d 1360, 1372–73 (Fed. Cir. 2006) (“[T]he deliberate surrender of a claim to certain subject matter during the original prosecution of the application for a patent made in an effort to overcome a prior art rejection is not such ‘error’ as will allow the patentee to recapture that subject matter in a reissue.”); Cubist Pharm., Inc. v. Hospira, Inc., 805 F.3d 1112, 1121 (Fed. Cir. 2015) (“the recapture rule applies only if the patentee surrendered subject matter in the original prosecution in order to overcome a prior art rejection.”); In re Clement, 131 F.3d 1464, 1469 (Fed. Cir. 1997) (“we look to … changes to the claims made in an effort to overcome a prior art rejection.”); QXMedical, LLC v. Vascular Sols., LLC, 408 F. Supp. 3d 996, 1005 (D. Minn. 2019) (applying the Cubist rule); Stephen G. Kunin & Kenneth M. Schor, The Reissue Recapture Doctrine: Its Place Among the Patent Laws, 22 Cardozo Arts & Ent. L.J. 451, 456 (2004) (the doctrine applies “if a patent applicant narrowed the scope of the claims during the prosecution of an application to overcome art of record (i.e., to distinguish the claims from the art)”).

In McDonald, the court rejected these treatises and statements found in a large number of prior cases limiting unavailable recapture to the “prior art” context. Instead, the court applied the doctrine to prevent recapture of scope relinquished to overcome an eligibility rejection.

McDonald filed his patent application back in 2008 claiming a computer-implemented method of displaying primary and secondary search results.  In 2010, the examiner rejected the claims as lacking eligibility and the applicant amended the claims to add a requirement that the search algorithms are “executed by a processor.”  That amendment was enough for the PTO to overcome the eligibility issue.  A patent eventually issued (in a continuation application), but the patentee later figured out that its “a processor” amendment was going to tie the claims to operating both queries on a single processor.  The patentee then filed its reissue application within the two-year timeline, but sought to recapture the scope that had been amended-away.   The PTO refused to allow the amendment as impermissible recapture. On appeal, the Federal Circuit has affirmed — holding that the court’s repeated prior statements limiting the rule to overcoming prior art were merely context specific dicta.

The public’s reliance interest on a patent’s public record applies not only to subject matter surrendered in light of § 102 or § 103 rejections but also to subject matter surrendered in light of § 101 rejections. Mr. McDonald focuses on the “prior art” language of certain decisions—citing language describing the initial amendments as made in an effort “to overcome a prior art rejection”—in an attempt to limit the recapture rule to § 102 or § 103 rejections. Although we previously addressed cases centered on prior art rejections, this does not mean that the recapture rule is limited to that context.

Slip Op.  The court has previously held that the recapture prohibition did not apply to amendments made to overcome a Section 112 indefiniteness rejection.  Cubist Pharm., Inc. v. Hospira, Inc., 805 F.3d 1112, 1121 (Fed. Cir. 2015).  In Cubist though the court gave two reasons for rejecting the recapture rules: (1) amendment were not for the purpose of avoiding prior art; and (2) amendment was not actually  broadening. Cubist.  In McDonald, the Federal Circuit recharacterized Cubist as a case only about the broadening question. Similarly, the court rejected the MPEP as having any authority.

In Cubist, the court wrote that the “recapture rule applies only if the patentee surrendered subject matter in the original prosecution in order to overcome a prior art rejection.”  On appeal here, however, the court concluded that none of its prior cases “expressly prohibit the recapture rule from applying to amendments made for other purposes.”  Slip Op.

Although the court has not previously applied the recapture doctrine in the eligibility context, there are a number of cases that don’t state the “prior art” requirement.  For example:

Where an applicant for a patent to cover a new combination is compelled by the rejection of his application by the patent-office to narrow his claim by the introduction of a new element, he cannot after the issue of the patent broaden his claim by dropping the element which he was compelled to include in order to secure his patent.

In re Byers, 230 F.2d 451 (C.C.P.A. 1956).

Because there is no express statutory rule here other than the word “error,” the court appears to be willing to consider policy arguments in making its conclusions.  In its decision, the appellate panel reflects that McDonald offered no persuasive justification for distinguishing amendments to avoid Section 101 rejections from amendments made to avoid prior art.  The largest consideration here is likely that that this issue appears already well settled — with the Federal Circuit previously treating the rule as “rigid.”

The doctrine here has its roots in equity and concern that a reissue will operate as a “trojan horse” to by an “unscrupulous attorney” to cover competing products after the original patent issued and expectations had been settled.  Of course, in the area of eligibility expectations were not settled at the time. Further, we should recognize that during this time Congress also amended the reissue statute to eliminate the patentee’s “deceptive intent” as a hurdle to obtaining reissue.  (Do be clear, there is no suggestion that McDonald’s attorneys acted improperly in any way).

I believe that there is a good chance that McDonald will petition for en banc rehearing.

23 thoughts on “Reissue, Recapture, and Skirting Eligibility

  1. 6

    “ On appeal, the Federal Circuit has affirmed — holding that the court’s repeated prior statements limiting the rule to overcoming prior art were merely context specific dicta.”

    Seems like the correct result for the right reasons.

    The bit about “context specific dicta” is important. There have been some extremely feeble arguments made here that the Supreme’s holding in Alice should be narrowed in a bizarre fashion. Those arguments ignore the fact that courts may refer to something (e.g., “an abstraction) in the singular simply because they are addressing the particular facts in the case or a hypothetical, and definitely not because they wish to limit the holding in some nonsensical fashion.

  2. 5

    In off topic patent news, the ol 1$ insulin patent comes up again in the news, as new insulin makers are making $$$, and MM will be glad to hear that biden officially rescinded Trumps thing to lower insulin prices.

    link to

  3. 4

    Yes, the “error” for re-issue purposes would be a 112 error of the form, “it makes it sound like the ‘processor’ can do everything all by itself without help.” The idea is that this would no longer be seen as a “broadening reissue.” Of course, it does have a potentially broadening side-effect because it creates the possibility of the “processing system” including second processor that the Applicant wanted.

  4. 3

    Re: “The 1952 Act remained silent on another broadening caveat — the rule against recapture.”
    That judicial rule applies to the PTO as well as patent suit defendants. But note that the second paragraph of 35 USC 252 provides another closely-related defense for patent suit defendants, and the reexamination and IPR statutes adopted it.

    1. 3.1

      BTW, I find some irony in the legitimate “concern that a reissue will operate as a “trojan horse” to by an ‘unscrupulous attorney’ to cover competing products years after the original patent issued and expectations had been settled.” [AKA “claim drafting from the catbird seat” long after defendants heavily invest in manufacturing and marketing, as stated re one case.] The irony is that this is frequently and easily done in the U.S. by keeping “continuations” pending for up to 20 years after a parent or grandparent patent issues and/or is in suit.

      1. 3.1.1

        Thanks Paul – I was going to mention exactly that: “The irony is that this is frequently and easily done in the U.S. by keeping “continuations” pending

        In view of:

        the courts had developed in the 1880s in response to perceived abusive practices.

        Some will — no doubt — equate continuation practice as being THE SAME abuse.

        But is it?

      2. 3.1.2

        The “irony” is diminished when one recognizes that the existence of pending continuations (which is a matter of public record) puts potential infringers on notice that broader or narrower claims may be forthcoming.

        A childless patent that was granted two years ago, on the other hand, isn’t going to produce any claim broader than what was issued (unless the PTO screws up).


          The irony is that, as you note, U.S. industry will know 2 years after a patent has issued what claims to avoid infringing, and how, and also have a 35 USC 252 defense, whereas by simply keeping continuations pending there may be no way for anyone to have any such certainty until 2 years after all the continuations finally issue many years later. There is no longer any “late claiming defense” in the U.S.


            Avoiding the question does not imbue irony.

            Some will — no doubt — equate continuation practice as being THE SAME abuse.

            But is it?

  5. 2

    Amending “processor” to “processing system” might have been safer since a “processing system” could reasonably cover two separate computer systems, each having its own processor chip, in which case both queries would be processed by the same processing “system” (albeit by different “processors” within that system). One could then couch the mistake as a 112(b) or 112(a) “mistake” by saying that “processor” would imply that the various other chips in a computer (e.g., memory chips, I/O hardware) were not involved in the procedure.

    1. 2.1

      Taking a term with well defined meaning in the art (“processor”) and renaming it to something with no meaning in the art (“Processing system”) and undefined in the spec is something prosecutors do a fair amoutn and is pretty dangerous. For apparatus claims especially, “processing system” is almost certainly going to invoke 112(f) and all that comes with it, while “processor” at this stage possible avoids it. The verbiage “at least one process or” or “one or more processors” or “a set of one or more processors” etc. etc. would be preferred.

      1. 2.1.1

        Excellent points.

        Though one wonders if a “processor” should avoid 112(f), as you’re just claiming everything functionally anyway.

        As long as we’re “fixing” 101 via statute revisions, we should fix the whole MPF issue too.


          as you’re just claiming everything functionally anyway.

          Reminds me of David Stein’s utter destruction of Malcolm vis a vis terms sounding in function LONG existing and being used outside of the computing arts…

          … with NO retreat to 112(f) nonsense.

          Also reminds me: “Vast Middle Ground.”

      2. 2.1.2

        … and undefined…?

        You mean, define with words (which themselves may well need to be defined…

        With words ([insert recursive “it’s turtles all the way down” loop here]…

  6. 1

    Can you go into more detail regarding why the patentee concluded that the “‘a processor’ amendment was going to tie the claims to operating both queries on a single processor”?

    1. 1.1

      Thanks Bryan – I don’t have any more information on their thoughts. The method involves two different searches, and the claims imply that it is the same processor that handles both queries. That may also be part of the issue.

      1. 1.1.1

        I recall briefly looking into this issue and coming up with case law going both ways–i.e., holdings that imply “a processor” means the same processor throughout and that imply “a processor” may encompass “one or more processors” working in conjunction. I’m not sure that it’s settled law. See, e.g., FotoMedia Technologies, LLC v. AOL, LLC, 2009 WL 2175845 (“The Court does not rule that there cannot be more than one server, each capable of performing all of the recited steps, in the claimed system. However, in light of the disclosure, the inventors did not claim a system that could distribute the steps of the claims at issue between various server computers.”), cf. Copy Protection LLC v. Netflix, Inc., 2015 WL 4639954 (“The Court concludes that there is no limitation in the claims that requires each device that forms a ‘server’ or ‘server computer’ to be capable of performing all of the recited server-side steps.”)

        If anyone is aware of more authoritative law on this issue I would be interested to know of it.

    2. 1.2

      It looks like just they used the common claim format:

      A method comprising: operation A, executed on *a* processor; and operaton B, executed on *the* processor.

      If so, this decision is going make a lot of SW patents pretty worthless. In some cases, the patentees ‘system’ boilerplate might save them, but…

      1. 1.2.1

        Most SW claims I’ve seen from good practitioners have explicitly accounted for this either through “first processor”, “second processor” language, “a set of one or more processors,” “at least one processor”, etc. … as wel as dependent claims “wherein the first processir is the same/different than the second processor.”

        Multiple processor computers, as well as distributed computing, have been well known for well over a decade now. It would take a pretty poor practitioner to not have accounted for this.


          YMMV. Google returns “About 33,789 results” for “(CL=(“executed on a processor”)) country:US”

      2. 1.2.2

        Dennis did not say that the singular interpretation of “processor” was part of this decision, rather that it was just the concern of the patentee that it might be if not changed by reissue.

Comments are closed.