by Dennis Crouch
The new petition in SawStop v. USPTO (Supreme Court 2022) focuses the question of whether COURTS have power to create non-statutory patentability doctrines.
Does the judiciary have the authority to require a patent applicant to meet a condition for patentability not required by the Patent Act?
SawStop Petition for Certiorari. You might imagine this argument being raised as tied to any number of patent doctrines: atextual eligibility limitations; assignor estoppel; doctrine of equivalents, etc. This particular case focuses on obviousness-type double patenting (OTDP). Question 2 of the petition focuses the question further: “Is the judicially created doctrine of nonstatutory double patenting ultra vires?”
Section 103 of the Patent Act generally defines the doctrine of obviousness and instructs tribunals to look to the prior art and consider whether the claimed invention is sufficiently beyond the prior art. But, even if a claims are found patentable under Section 103, the USPTO still considers obviousness-type double patenting. Even if there is no prior art, the USPTO still bars a patentee from obtaining a separate patent claiming an obvious variation of already-issued claims held by the patentee/inventor. Thus, although 102(b)(2) explains that certain prior-filed patents are not prior art when attributed to the applicant/inventor, OTDP steps in on the back-side to effectively assert those excused documents as prior art. Pause here to note that one big difference between obviousness and OTDP is that applicants can skirt OTDP rejections by filing a terminal disclaimer that effectively links the two similar patents together along two axes: same term and same owners.
My middle school shop teacher was missing fingers courtesy of a table saw. Yours? Steve Gass has a PhD in physics and also a patent attorney. He was doing some amateurs woodworking when he conceived of his SawStop technology. The basic idea: if the spinning blade ever touching human skin, the electrical connection triggers an emergency brake that stops/retracts the blade within 5 milliseconds. “Fast enough to turn a life changing event into a minor cut.” Gass has over 100 patents and runs the innovative product company SafeSaw.
Gass’s U.S. Pat No. 9,927,796 was filed back in 2002 — claiming priority to a 2001 provisional application. That patent finally issued in 2018–only after SawSafe filed a civil action and received a court-judgment in its favor. A 1974 patent had disclosed use of a safety circuit and a braking-means. However, the district court concluded that the prior art was not enabling — i.e., a person of skilled in the art would not be able to construct (or even design) the claimed invention without undue experimentation. [2016 Decision].
Another 16 issued patents also claim priority back to this 2002 application. One application is still pending, Application 15/935,432, that claims an obvious variant of what has already been patenting. The USPTO rejected the ‘432 application based upon the judicially created non-statutory doctrine of obviousness-type double patenting. SawStop did not want to file a terminal disclaimer on this valuable patent and instead appealed. On appeal, the Federal Circuit affirmed without opinion in a R.36 Judgment. Now, SawStop has petitioned the Supreme Court asking that the doctrine be eliminated.
= = =
Although the petition does not cite the Supreme Court’s recent abortion decision, it certainly picks-up upon some of the same themes:
- A right to abortion “is nowhere mentioned in the Constitution.” Similarly, obviousness type double patenting has no grounding in the Patent Act.
- The fact that a precedent is old does not convert that precedent to a sacred text.
The Supreme Court in Bilski addressed this issue to some degree in the context of the non-statutory categorical bars of abstract ideas; laws of nature and natural phenomenon.
While these exceptions are not required by the statutory text, they are consistent with the notion that a patentable process must be “new and useful.” And, in any case, these exceptions have defined the reach of the statute as a matter of statutory stare decisis going back 150 years.
Bilski v. Kappos, 561 U.S. 593 (2010). The Bilski court found a small textual hook, but also identified the doctrine as built upon longstanding precedent.
The longstanding nature of precedent is important in the patent context. But unlike in the Constitutional abortion context, we have always had direct statutes guiding patent issuance and enforcement, beginning with the First Congress in 1790. In the area of obviousness, the doctrine was developed by courts and then implemented by statute in 1952. In Graham though, the Supreme Court indicated that, with minor exceptions, the 1952 Act maintained all the old precedent.
Double Patenting in the Statute. In its 2014 Abbvie decision, the Federal Circuit wrote that Obviousness Type Double Patenting is “grounded in the text of the Patent Act.” In particular, the court honed-in on Section 101’s use of the singular article “a” in describing what is granted to an inventor: “a patent.”
While often described as a court-created doctrine, obviousness-type double patenting is grounded in the text of the Patent Act. . . . Section 101 reads: “Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, … may obtain a patent therefor.” 35 U.S.C. § 101 (emphasis added). Thus, § 101 forbids an individual from obtaining more than one patent on the same invention, i.e., double patenting.
Abbvie Inc. v. Mathilda and Terence Kennedy Inst. of Rheumatology Tr., 764 F.3d 1366, 1372 (Fed. Cir. 2014). But Judge Dyk’s conclusions in Abbvie are not well grounded themselves as there are several ways this provision could be interpreted. It also proves too much, since there are ways for a patentee to obtain multiple patents covering patentably indistinct inventions: (1) via terminal disclaimer; (2) via restriction. If the patentee files a terminal disclaimer then the patent office will allow multiple patents to issue. If the patentee files a divisional application (following a restriction) then the patent office will allow multiple patents to issue even without a terminal disclaimer.
If restriction practice and examination practice were reasonable, especially in some arts, a single, broader patent SHOULD issue instead of multiple narrower variants on the same theme. But, because examiners are rewarded on being adverse to the public and quantity over quality, patentees are forced to unfairly narrow claims and then file CONS or DIVS to cover other useful embodiments of the actual invention. I don’t like the games played by either examiners or applicants, but I’m rooting for Saw Stop on this one. Even if it’s bull pucky, maybe it’ll spur some re-evaluation of examination practices here. What am I saying. Of course it won’t. But I’m still rooting for Saw Stop because OTDP is abused by examiners almost as badly as restriction practice.
In view of all the OTDP discussion here, this is to note a concise helpful article by a patent attorney who has written well on other patent law subjects: “A Look At The U.S. Doctrine of Obviousness-Type Double Patenting” by Courtenay C. Brinckerhoff (Foley&Lardner LLP) June 13, 2012
Among the complexities of OTDP practice is whether to apply a one way or a two way obviousness test. As held at p.15 of In re Hubble (Fed. Cir. 2013): “ ..In re Berg, 140 F.3d 1428, 1432 (Fed. Cir. 1998). Under that test, “the examiner asks whether the application claims are obvious over the patent claims.” Id. In certain circumstances, however, a two-way test may apply, wherein the examiner “also asks whether the patent claims are obvious over the application claims. If not, the application claims later may be allowed.” Id. The two-way test, which is “a narrow exception to the general rule of the one-way test,” arose out of the concern to: prevent rejections for obviousness-type double patenting when the applicants filed first for a basic invention and later for an improvement, but, through no fault of the applicants, the PTO decided the applications in the reverse order of filing, rejecting the basic application although it would have been allowed if the applications had been decided in the order of their filing. Id. We have explained that the two-way test is appropriate only in the “unusual circumstance” where “the PTO is solely responsible for the delay in causing the second-filed application to issue prior to the first.”
P.S. I forgot to thank the commentator below who pointed out, as to the Fed. Cir. rationale for OTDP [a public policy against granting more than one patent to the same inventor claiming the “same” [obvious claims] invention], that 35 USC 121 [as well as 101] arguably supports a statutory suggestion for that “public policy.”
“[as well as 101]”
Clearly not.
As you well know, Paul, 35 USC 101 has but two statutory requirements:
1) the invention fits into at least one statutory category; and
2) the invention has utility (under the patent sense of the word).
That
Is
It
The problem is that most applicants cave to OTDP rejections rather than fight them.
Most of my clients are into building portfolios, so this is often not a battle worth fighting.
This is one of the better inventions of all time, I hope the patentee wins against all unreasonable attacks. What a totally GREAT SOLUTION to a LONG STANDING PROBLEM IN THE ART !! Preventing ppl from losing fingers by a saw. Give that Man a Blue Ribbon.
blah blah, stare decises, go re-read Erie vs. “the incredible Mr. Tompkins” for the Court’s view on that doctrine. Reality vs. appearances, sort of stuff.
Rule 56 and the Supreme Court and CAFC case law on inequitable conduct related to failure to disclose information is not in the statutes. (I have seen some filings alleging inequitable conduct as being a failure to comply with 35 USC 112. Are you kidding me?). Is the non-statutory nature of the law now a defense against an allegation of inequitable conduct?
George, this is an excellent point and the logical fallacy of the argument the petitioner here is making. The certiorari petition cleverly tries to characterize OTDP as a judicially created “condition of patentability” that exceeds the conditions in the Patent Act, but you could argue that it’s just another of many recognized judicially-created doctrines barring issuance or enforcement of patents. The broader argument here seems to be that a patentee can’t have its patent rights lost or reduced because of something that’s not expressly spelled out in the Patent Act. If that argument received any traction, then a lot of other judicially-created doctrines might also go bye-bye, like res judicata, collateral estoppel, inequitable conduct, prosecution laches, patent misuse, shop rights, sanctions for spoliation of evidence, and pretty much every single procedural vehicle not spelled out in the statute (but well-recognized in litigation and prosecution) that could result in forfeiture of patent rights.
Are you conflating various other legal paths outside of “conditions of patentability?”
If OTDP were eliminated, isn’t there a risk the Office would start using prosecution laches more aggressively to reject continuation applications based on unreasonable delay?
Perhaps the brief should have cited, Henry Schein, Inc. v. Archer & White, Inc., 139 S.Ct. 534 (2019) (unless it is in fact cited). That case seems to over-turn a judicially created doctrine albeit of just some circuits, and states, “We may not rewrite the statute simply to accommodate that policy concern” and “it is not this Court’s proper role to redesign the Act” which at least makes for good “sound bites.” 🙂
LOL – yes, that case is one of the two shears of the Kavanaugh scissors.
They sure look like double to me. But that was before I left the Enville Post Office. Of course GROUP 3600 confirmed it.
Here see they really were taken by Fax. Berkenstock knew what I did not.
DC: “ The fact that a precedent is old does not convert that precedent to a sacred text”
This is beyond “quaint” and into “I’m a hapless stooge carrying water for the worst people in the world” territory.
The 6 Ayatollah’s don’t care about the age of a “precedent”, nor do they care about the strength of the reasoning behind that precedent. All they care about is increasing and maintaining the power of their Dominionist friends to control women, punish gays and non-whites, and shovel more money into the deserving hands of their ultra-capitalist benefactors. They are quite open about this.
The “law” is a dead letter. Heckuva job, Dennis and Jason.
Your feelings are noted, Malcolm
OT, but this is interesting:
link to uspto.gov
Thanks. This is interesting. Two bits that surprised me:
(1) “[T]he highest earnings [are] in the copyright-intensive industries, followed by earnings in the utility patent-intensive industries… .” I would have expected utility-patents to generate more value than copyrights.
(2) “[R]elative to workers in non-IP-intensive industries, workers in IP-intensive industries are more likely to… be veterans.”
..(1)… really? You never hear of the Mickey Mouse parade (very old tune), and the supplanted Big Cinema and Big Music (now across a plethora of channels)…?
Yes, Chisum basically summarizes my position that I have disclosed over the course of the last decade.
That not a flook [sic].
This seems unlikely to go anywhere.
On the parallels to Dobbs, the Supreme Court has said a million times that stare decisis is stronger when courts aren’t the last word. If the Supreme Court gets a constitutional question wrong, Congress can’t overrule it. If the Supreme Court construes a statute incorrectly, Congress can amend the statute. If the Supreme Court wrongly invents obviousness-type double patenting, Congress can abolish it. My understanding of OTDP is that it comes from the Supreme Court, it’s been around for more than 150 years, courts and interested parties have assumed its existence and planned their affairs around it, and Congress has made no move to abolish it. There’s some intuitive appeal to the argument, but I doubt that even the most rigid textualist on the Supreme Court would find this interesting.
Interesting – would it be incorrect then to say that the NEW stare decisis OF Dobbs is stronger than the stare decisis of Roe (given that absence of Constitutional right is eminently fixable by normal legislation, as opposed to such normal legislation not being able to address what would have been deemed a Constitutional right)…?
I’ve never thought about that way. I have mixed feelings about the answer, and think it’s in some tension with the purpose of having constitutional rights.
Logically, it works. If the Supreme Court reads constitutional rights too narrowly, Congress and the states can always expand them through legislation and state constitutions. But if the Supreme Court reads constitutional rights too broadly, then there’s no possible “fix” short of amendment.
But if you take that view too far, it means that the Supreme Court should always err on the side of reading constitutional rights narrowly. That seems wrong to me. Constitutional rights are usually rights against the government, and supposed to be protected from the political process. If courts fail to prevent, e.g., viewpoint discrimination or general search warrants, it’s no answer to say that the government can always tie its own hands if the voters ask it to. If there is a constitutional right to abortion, and Dobbs is wrong on the merits, then that’s really bad. And it’s no answer to say that Texas and Mississippi can always create a right to abortion if their legislatures want to.
The real lesson, I think most would say, is that it’s critically important that the Supreme Court decide constitutional questions correctly, because of the consequences of being wrong, in either direction. Read rights too broadly, and you tie the hands of government and illegitimately take decisions away from the people. Read them too narrowly, and you defeat the purpose of having constitutional rights in the first place. That’s why the court at least sometimes overrules constitutional decisions when it’s convinced they’re wrong.
Other laws, e.g., statutes and common-law-type stuff are different. It’s not costless when courts get these questions wrong, but it makes more sense and seems more just to ask Congress to fix these sorts of things.
Your back peddling weakens your initial point.
I fear that you cannot have it both ways. Please choose one.
“If there is a constitutional right to abortion, and Dobbs is wrong on the merits, then that’s really bad.”
Fortunately for us there is no such constitutional right so we don’t have to worry about the rest.
“Read rights too broadly, and you tie the hands of government and illegitimately take decisions away from the people.”
They’ve been doing that in spades for like 70 years. Arguably there was some reason for them to 70 years ago, nowadays not so much.
It is a Tenth Amendment issue. The Constitution limits the power of the federal government. If the feds don’t have the power, it is left to the states. Or the people.
Congress cannot give the feds power that the Constitution does not.
This sounds right to me. I read a lot of articles urging a nationwide codification of Roe, and a lot urging a nationwide abortion ban, but a sane reading of the constitution should take both of those options off the table.
Neither homicide nor family law are specifically entrusted to the federal government in the constitution. Therefore, both should be regarded as reserved to the states. The feds should not be regarded as having power to set a nationwide policy on abortion in either direction.
That is how I see it.
“Congress cannot give the feds power that the Constitution does not.”
Congress can give themselves (aka the feds) whatever power they want by amending the constitution. Though there can be backlash.
“Congress can give themselves (aka the feds) whatever power they want by amending the constitution. Though there can be backlash”
We’ve been over this, 6.
It is far less “backlash,” and far more much higher requirements — on purpose — to make such a change to the Law.
Yes. So many people, including lawyers, do not understand this. Read Article V. Need 2/3s of both Houses or the legislatures of 2/3s of the states to convene a convention for proposing amendments and then you need ratification by the legislatures of 3/4s of the states.
It is not easy for a reason.
The Constitution does not spell out all the rights. It acts mainly as a brake on the Federal and State governments. Dobbs was decided by activist justices who did violence to the nature of stare decisis, and destroyed the myth that conservative Justices would abide by precedent. If the justices appointed by Trump could realistically be impeached and removed from office for lying to Congress, they would be. However, too many Republicans don’t have the guts or intellectual honesty.
In pieces due to hyper active filter
Wow — you are off on every single one of your points (after the first one):
“The Constitution does not spell out all the rights. It acts mainly as a brake on the Federal and State governments.”
Yay – you got one right (of course, this means Roe of course could not ESTABLISH the “right” as attempted).
“Dobbs was decided by activ1st justices who did v1 0 lence to the nature of stare decisis,”
Well, I can give you partial credit, as “activ1st justices” may be correct, but that is no more activ1st than those of Roe. As to v1 0 lence, there you are absolutely wrong.
“and des tr0 yed the myth that conservative Justices would abide by precedent.“
Nope – you cannot des Tr0y what was not there to begin with. Plus (reflecting your error on state decisis), precedent is not — nor should be — absolute.
“If the justices appointed by Trump could realistically be impeached and removed from office for l y 1ng to Congress, they would be.”
There was no l y 1ng to Congress (Alito wrote the decision, and NO to-be-Justice can pre-decide a case) — YOUR path would actually be a violation of judicial ethics.
“However, too many Repub licans don’t have the g it’s or intellectual h0n esty.”
There you go again, blaming a “party” while thinking your “party” must be correct. So, not only are you incorrect as to the substance here, your BLIND ide0 l0gue ways also run to error.
And yes, I am a “D.”
“the NEW stare decisis OF Dobbs is stronger than the stare decisis of Roe”
The new of Dobbs is never ever going away anon. Not in your lifetime anyway. Don’t even worry about that. That’s obviously why leftists are but blas ted. And of course that congress likely lacks the authority to establish a right to an abortion without it coming directly from the constitution (people think they do, but they probably don’t).
Never say never.
In count filter….
Never say never.
Yes. Even RBG admitted that Roe was not on firm ground.
The nominees for Supreme Court purposely avoid answers about the future. Because they cannot say what they would do in a particular case. Yes, Roe was the law at the time. Yes, they believe in stare decisis. But they did not say they would never vote to overturn Roe. They would never answer that question.
Justice Sotomayor said she was going to bring the views of a wise Latina to the court. Justice is supposed to be blind. Not based upon one’s POV. And that may be what is really disturbing to those who want it otherwise.
Even the latest (celebrated) NEW Justice refused to answer certain questions.
Of course, the absurdity of declining to answer what a woman was impacts the fickle Left’s Intersectionality agenda and it’s “SH UT UP MEN” agenda.
(…and do not even hint at fetal life being a separate human life – from a scientific point of view…)
Put differently, I’d challenge the petitioner to read Kimble v. Marvel and tell me how this case should come out differently.
Kimble v. Marvel was a DREDful decision [sic].
I’m ambivalent about Kimble, but I think the arguments for overruling Brulotte v. Thys were much stronger there than SawStop’s arguments for abolishing obviousness-type double patenting.
the ’64 case by Douglas…?
I will have to refresh my memory.
Kimble link to scholar.google.com
Brulotte link to scholar.google.com (yes, 1964, by Justice Douglas)
Brulotte was a terrible decision, but Kimble was exactly right. If Congress did not like the outcome that Brulotte creates, Congress has had ample opportunity to fix it in the many revisions that the Congress has made to Title 35 in the time since. Congress’ inaction on this point (and others) should be read as cementing Brulotte (and other points of the common law of patents) in place unless and until the Congress takes action.
Incidentally, this principle should not be understood as applying only to the SCOTUS. “[The Supreme] Court will not now lightly disturb a rule well settled by forty-five years of judicial consideration and conclusion in th[e circuit] courts.” Westinghouse Co. v. Formica Co., 266 U.S. 342, 349 (1924). See also, Merrill Lynch, Inc. v. Curran, 456 U.S. 353, 381-82 (1982). Where the circuit courts have laid down a consistent line of interpretation over many years and the Congress has—over that intervening time—amended the statute without changing the law to thwart the outcomes from the circuit courts, that similarly bespeaks a Congressional acquiescence that should not be lightly disturbed.
Kimble was horribly decided — and (interestingly enough) presages Alito’s Dobbs decision.
Kagan’s “I know it’s wrong, but meh, not going to change it” is simply atrocious.
“the Supreme Court has said a million times that stare decisis is stronger when courts aren’t the last word.”
Based supreme court listener.
6,
Take a gander at Alito’s dissent in Kimble.
I expect that we all agree that this cert. petition has little chance of success. Imagine, however, that it were to succeed, what would this mean for the improper Markush group rejection. As Mr. F. mentions below, there is at least some statutory warrant for ODP in §121. There is no statutory warrant for improper Markush rejections, however.
As a practical matter, the PTO needs the improper Markush rejection to police a fair application of examination fees. If the rejection were to disappear following a ruling that the courts may not invent new grounds of invalidity that are not explicitly provided in the text, could the USPTO continue to apply improper Markush as an objection?
Are you thinking of improper multiply dependent claims (112e?)? Improper Markush claims have a pretty clear 112b support.
Sorry, I am not following. If I claim “[a] machine comprising a first blade and and an attachment selected from the group consisting of a handle, a chain, and a second blade,” where is the §112(b) problem?
There isn’t one as that isn’t an improper markush. An improper markush would be “the group comprising:” But I don’t see how that has anything to do with examination fees, which is why I was thinking maybe you were talking about multiple dependent claims. Regardless of whether this is a rejection or not, what does it have to do with policing examination fees?
MPEP 2117.II: A Markush claim may be rejected under judicially approved “improper Markush grouping” principles when the claim contains an improper grouping of alternatively useable members. A Markush claim contains an “improper Markush grouping” if either: (1) the members of the Markush group do not share a “single structural similarity” or (2) the members do not share a common use.
Actually I get it from the example below. The “handle, chain, second blade” example is not really accurate. You’re talking about putting three different inventive ideas in a single claim as a markush group. I don’t see what is wrong with that, you’re not suggesting that because they’re included in a single claim that restrictable subject matter doesn’t become unrestrictable does it?
[Y]ou’re not suggesting that because they’re included in a single claim that restrictable subject matter doesn’t become unrestrictable does it?
In a claim in a PCT national phase application, the three are still restrictable. Caterpillar Tractor Co. v. USPTO, 650 F. Supp. 218, 220 (E.D. Va. 1986). However, in a §111 application, this is not permissible. In re Weber, 580 F.2d 455, 458 (C.C.P.A. 1978). Rather, the only option where an applicant presents a number of disparate elements within a Markush is for the office to impose an improper Markush rejection. The office may not impose a restriction requirement on that Markush because “[i]f… a single claim is required to be divided up and presented in several applications, that claim would never be considered on its merits.” Id.
Of course, a species election might still be proper, but a species election merely sets the starting point for a Markush group. If the examiner does not find art related to the elected species, then the examiner still must proceed to search the rest of the Markush.
“As a practical matter, the PTO needs the improper Markush rejection to police a fair application of examination fees.”
Why?
Why is an improper markush mote problematic for examination than any broad claim?
Improper Markush rejections exist to stop a person from claiming more than one distinct invention in a single claim. A broad claim on a single invention is easy to examine (the broadness makes it easy to find relevant art). An improper Markush group, however, forces the examiner to search the first Markush choice, then when that is clear of the art, to search the second Markush choice, etc.
“A broad claim on a single invention is easy to examine (the broadness makes it easy to find relevant art). An improper Markush group, however, forces the examiner to search the first Markush choice, then when that is clear of the art, to search the second Markush choice, etc.”
I sincerely don’t see how the broad claim to a “single invention” is different from a markush claim including a list of alternatives. The “single invention” frequently includes limitations which can be read on by alternatives. And the markush claim may be framed as limited to a single scope, where that scope includes an element which may be one of several alternatives. It appears to me to be distinction without difference.
Though I’ll acknowledge that I may be confused because MPEP 800 is a burning pile of rubbish and my approach to restriction is “Don’t.”
“ and my approach to restriction is “Don’t.””
That seems like a minority view (albeit restrictions are generally not super-abundant, when they ARE requested, they are often improvidently so requested).
Improper Markush rejections exist to stop a person from claiming more than one distinct invention in a single claim.
I sincerely don’t see how the broad claim to a “single invention” is different from a markush claim including a list of alternatives
Ah I see now. Imagine a claim to “a device comprising a processor and a memory, the memory storing code, wherein the code is selected from a group consisting of: [code for a network protocol], [code for implementing a machine learning algorith], [code for performing a business method on a computer]” – you’ve just put three tech centers into a single claim, so the claim can be species-restricted.
“forces the examiner to search the first Markush choice, then when that is clear of the art, to search the second Markush choice, etc.”
But only one alternative need be found to reject under prior art, right?
Of course to “fully search” a markush claim and anticipate the amendments, the examiner must search for all the alternatives.
But to “fully search” a broad “single invention” functional claim, the examiner must search for all the ways the specification supports for achieving the function.
Honestly, between an express list of alternatives (markush) and a vague suggestion of the alternatives (broad non-markush claim), the former sounds easier to examine.
Hopefully a concrete example will help to clarify. Imagine that I invent both a new hand soap and a new television. The soap has features A, B, and C, which I believe distinguish it from the prior art. The television has features D, E, & F, which I believe distinguish it from the prior art. I believe that we can agree that no single search will reveal the most relevant art for both of these inventions.
If I file both in the same application and claim the soap in 1–10 and the t.v. in 11–20, the examiner will impose a restriction. The elected claims will be examined and the unelected will have to be canceled before the case can grant.
If I write a single independent claim broad enough to read on both without using a Markush (e.g., “a composition of matter comprising carbon atoms”), this will extend so broadly that it is bound to fail until I start narrowing it toward one or the other of the real inventions that I have disclosed.
The real problem comes when I start using Markush language. If I claim “a composition of matter comprising: A, B, & C; or D, E, & F” then the examiner must search both inventions for only one fee. The improper Markush rejection is the solution to this situation. Without the improper Markush rejection, an applicant can get any number of inventions examined for only one fee.
I’m still, sincerely, not understanding how “a composition of matter comprising: A, B, & C; or D, E, & F” necessarily requires more search than “a composition of matter comprising carbon atoms”.
Yes, it’s very easy to find prior art that demonstrates that a tiny fraction of our non-markush claim scope is anticipated. It is perhaps less easy to find ABC, but if I find only ABC I can similarly reject the markush claim based on just a fraction of the claim scope being anticipated.
But just as finding ABC does not make a complete search of our markush claim, onlt finding ethane does not make a complete search of our non-markush claim.
What is a “complete search”? The office tells examiners a complete search includes the claimed invention and any subject matter that may be reasonably expected to be needed to handle the amendments.
So for our markush claim, I would ideally search for ABC, DEF, and anything that the specification suggests may be included in an amendment.
And for our non-markush claim, I would ideally find ethane and search for anything that the specification suggests may be included in an amendment.
For our non-markush claim to require less searching, the combined search space of “ABC, DEF, and potential amendments” must be larger than the search space of “compounds with 2 or more carbons, and potential amendments”.
That could happen depending on the relative specifications, but it is not necessarily the case, and it is not a function of how the claim is formatted.