PATENTS - Speculation on Alice v. CLS Bank

Speculation on Alice v. CLS Bank.

Those discontent with software patents are looking for a knock-out blow at the U.S. Supreme Court. After oral argument last week, the Court had a conference on this question:

Whether claims to computer-implemented inventions – including claims to systems and machines, processes, and items of manufacture – are directed to patent-eligible subject matter within the meaning of 35 U.S.C. § 101 as interpreted by this Court.

Patent subject matter eligibility is a murky morass. I don't anticipate a bright line rule about when an invention is an "abstract idea" and when it isn't. I do expect specific guidance to remedy the en banc Federal Circuit's split into 7-3 + 5-5 judgments with no majority opinion. At this point a clerk has been asked to begin drafting the opinion. 

Based on the oral arguments and briefs, my guess is that the district court judgment will be affirmed: Alice's asserted patents are abstract ideas without sufficient steps to identify patentable subject matter under section 101 of the Patent Act. The opinion will reason that identifying computer-implementation as an element, or "limitation" in a claim, does not sufficiently manipulate or alter an abstract idea, here of solvency, to create patent-eligible subject matter. 

Such a ruling would advance questions remaining after in Bilski (which found claims covering nothing more than hedging were unpatentable abstract ideas) with the reasoning of Mayo and Myriad, and the predecessor cases, going back to Benson and Flook.

It will be stated more eloquently by the Court's writers, but my guess is that the principle behind the opinion's reasoning will be this: 

Taking a law of nature, physical phenomena, or an abstract idea, and claiming an application of any of the three, without additional details as to how to alter, manipulate, or effect the use of law, phenomena, or idea, over and above well-understood, routine, conventional activity previously engaged in by professionals in the relevant field, is ineligible subject matter for patents. 

The effect will be to create a "quick review" obviousness determination, by stripping out from a patent claim any conventional computer-implemented activity, to see whether what's left is a business or financial concept like insurance, hedging, or solvency.

The line between an unpatentable abstract idea and a patentable application of an abstract idea, such as certain business methods, is that adding a computer to an economic concept is like telling doctors to engage in well-understood, routine, conventional activity previously engaged in by scientists in the field. In Mayo this reasoning was used to invalidate "drafting efforts designed to monopolize a discovery of a medical correlation." There, the Court found that "applying" the discovery was not sufficient activity to transform an unpatentable law of nature into a patent-eligible application of such a law. 

At bottom, Alice's asserted claims require use of a computer in a method or system, but those uses are well known in the art. Use of a computer simply tells software engineers, like the invalid claims that would have covered medical professionals in Mayo, to engage in well-understood, routine, conventional activity, previously engaged in by programmers in the software field. Without additional detail of how to manipulate the idea, this is conventional activity that cannot transform an unpatentable abstract idea into a patent-eligible application of such an idea. 

As for the positive effects, I suspect that the consequence of such a ruling would force patent drafters to create (marginally) more robust and descriptive claims. Drafters can still claim inventions related to email (data encryption), word processing (swype), or even business methods (frequent flyer miles, specific applications of credit default swaps). But by raising the exceedingly low hurdle of 101, business method and software patent drafters will have to claim more specifics of how the person having skill in the art to manipulate a natural law, physical phenomenon, or abstract idea, or risk rejection under 101. A result of more descriptive claims could be better-defined targets for examiners to reject patents under 102 and 103. If the claimant must be more specific, and because the drafter cannot defend claims solely under 112 enablement standards and written description standards, then those who believe there is insufficient patent notice in non-pharmaceutical, non-chemical patents will get the best ruling they can reasonably ask for, while leaving the uncertainty in the future of innovation as open as possible.

I will not attempt here to add more to the the normative discussion. Please see the excellent amici briefs on file with Scotusblog.

Here's how I would structure the draft brief. 

I. Background law. 

The parties agree that abstract ideas are not patentable. The opinion will first explain patent law's basic rules, including that the bulk of the work of rejecting patent applications is done by §§ 102, 103, and 112 of the Patent Act. But, before they get there, there is a "coarse filter." For several decades, the Court has said that "anything under the sun made by man" may be eligible for patenting under Patent Act § 101, with three exceptions. As Petitioner Alice noted, the Court has identified “three specific exceptions to § 101’s broad patent-eligibility principles: ‘laws of nature, physical phenomena, and abstract ideas.’” Bilski v. Kappos, 130 S. Ct. 3218, 3225 (2010). The “abstract ideas” exception provided the basis for the decision in this case.

See Pet. Br. 2.

II. Background facts.

Although Alice does not bring much attention to it, not all claims in the relevant patents claims at issue in this case. 

Alice is the owner of U.S. Patent Nos. 5,970,479, 6,912,510, 7,149,720, and 7,725,375. These patents have largely overlapping specifications (Pet. Br. 7; Pet. App. 154a), which describe multiple inventions that are covered by two distinct sets of claims. Most of the original specification, and the great majority of the drawings, relates to claims 1–32 and 35–39 of the ’479 patent, which involve the formation of certain multi-party contracts. Alice has not asserted these claims against CLS.

Resp. Br. 2. 

So, what did Alice assert against global entity CLS? A method of transacting business. 

Method claim 33 of the ’479 patent is representative, as Alice concedes (Pet. Br. 8) and the courts below recognized (Pet. App. 26a, 80a, 134a, 176a– 77a).  See J.A. 383–84.  In Alice’s method, a third-party middleman stands between counterparties to one or more transactions.  Pet. App. 25a–27a.  This intermediary maintains “shadow” accounts for each counterparty and, so long as these accounts have adequate value, adjusts them according to the terms of the parties’ agreement(s).  Id. at 26a–27a.  At some point in the day, the middleman sends final instructions so that real-world accounts reflect the changes to the shadow accounts.  See id. at 164a– 65a. 

Resp. Br. at 3.

III. The Procedural Posture:

CLS and Alice filed cross-motions for summary judgment on the question whether the asserted claims are patent-eligible under 35 U.S.C. § 101.  D.C. Dkt. Nos. 43, 54, 94, 95.  The district court held the motions pending this Court’s decision in Bilski v. Kappos, 130 S. Ct. 3218 (2010), and, after additional briefing, issued an exhaustive opinion analyzing every asserted claim.  Pet. App. 172a–238a. “

“For the purpose of deciding patent eligibility at the district court, the parties agreed to a broad claim construction that was favorable to Alice.  The district court concluded that each [asserted] claim … re- quires computer implementation.”  Pet. App. 155a.  No other claim limitations were disputed for purposes of the Section 101 analysis. 

Resp. Br. at 6 (quoting Pet. Br.).

The district court held that the method claims are not patent-eligible because they recite the “abstract idea of employing an intermediary to facilitate simultaneous exchange of obligations in order to minimize risk,” which the court determined to be “a basic business or financial concept much like” the hedging method deemed unpatentable in Bilski. 

At the Federal Circuit, the judges voted 7-3 to affirm the district court’s invalidation of the method and media claims, and invalidated the system claims because the court split 5-5.  There were six separate opinions, none of which commanded a majority of six votes. 

Judge Lourie, writing for five, explained that none of the asserted claims are patent-eligible under Bilski and the intervening decision in Mayo Collaborative Services v. Prome- theus Laboratories, Inc., 132 S. Ct. 1289 (2012). Alice’s method claims are ineligible, because “reducing settlement risk by facilitating a trade through third-party intermediation” is an abstract idea.  Pet. App. 28a.

Under Mayo, “adding generic computer functions to facilitate performance pro- vides no substantial limitation and therefore is not ‘enough’ to satisfy § 101.” That is, adding a computer is like saying here is a principle of nature, "apply it." Judge Lourie found a “striking level of generality” in Alice's patents, which “recited a handful of computer components in generic, functional terms that would encompass any device capable of performing” the computer-implemented methods.

Chief Judge Rader and Judge Moore dissented in part.  They agreed with the plurality that the method claims, as well as the media claims, are ineligible because they recite an abstract idea.  Pet. App. 80a–84a.  But they viewed the system claims as patent-eligible on the ground that “tangible” computer components, such as a “data storage unit,” could not be “abstract.”  Id. at 70a–71a, 86a–87a, 97a–99a. c.

Judges Linn, O’Malley, and Newman also dissented, they would have found everything met 101 due to the claim of computer implementation alone.  Pet. App. 123a–26a.  Judge Newman maintained that “when the subject matter is within the statutory classes in section 101, eligibility is established.”  Id. at 111a; see also id. at 101a. 

IV.A  Do the claim elements, together, transform the underlying ineligible abstract idea into an eligible invention?

According to Respondent CLS, no. 

The claims asserted by Alice recite the fundamental economic practice of intermediated settle- ment or escrow, in which a “middleman” stands between the counterparties to a transaction and effectuates the transfer of entitlement once all condi- tions are satisfied.  This is an “abstract idea” under Bilski, as both the district court (Pet. App. 214a) and a majority of the Federal Circuit (id. at 31a (opinion of Lourie, J.), 82a–84a (opinion of Rader, C.J.)) recognized.  See Part I, infra.  The plurality below (Pet. App. 29a–31a) also correctly determined that the asserted claims, while computer-implemented, lack an “inventive concept” that is a prerequisite to patent eligibility under the methodology prescribed by this Court in Mayo.  See Part II, infra.  Alice’s bid for reversal would thus require the Court to critically undermine, if not overrule, both Bilski and Mayo.  Doing so would harm innovation and the economy.  See Part III, infra.1 

According to Alice, yes. 

[N]ot even the plurality suggested that the supposed “abstract idea” it identified in Alice’s claims was a preexisting fundamental truth akin to a mathe- matical formula or law of nature that “reveals a relationship that has always existed,” Flook, 437 U.S. at 593 n.15. And it is evident that the alleged “abstract idea” the plurality identified bears no resemblance to the fundamental, mathematical truths at issue in this Court’s prior cases. Instead, the plurality found Alice’s claims ineligible by holding the “concept” upon which they supposedly “draw” to be “an abstract idea because it is a ‘disembodied’ concept.” Pet. App. 28a (emphasis added).4 But, as explained above, the Court’s cases addressing the abstract ideas exception apply that analysis only to fundamental truths—not disem- bodied concepts. It is only claims directed to funda- mental truths that raise the preemption risks the Court described in Benson, Flook, Bilski, and Mayo - and only those claims can, consistent with the statutory text, be deemed ineligible despite fitting within the four statutory categories.

Alice refers to its own claims: 

The computer is itself the intermediary. See JA958 (reciting, inter alia, “a computer … that is configured to (a) receive a trans- action; (b) electronically adjust [accounts] …; and (c) generate an instruction”); JA383-84 (reciting computer-implemented steps, including “creating” shadow records, “obtaining” balance information, “adjusting” the shadow records in chronological order, and “instructing” exchange institutions to provide delivery). The use of the computer to track multiple transactions, in chronological order, and to issue simultaneous instructions at the appropriate time— regardless of the geographic locations of the parties, their time zones, or the particular exchange institu- tions they employ—is central to the claimed methods. The invention as claimed will not function without a computer configured (i.e., programmed) to carry out the claim steps. Moreover, a computer and other hardware specifically structured and configured to perform the recited functions are essential to the claimed systems

If Method claim 33 of the '479 Patent is indeed representative of Alice's system and other claims, then Alice's patents will be invalidated as "abstract ideas," and the district court judgment will be affirmed. Here is claim 33:

A method of exchanging obligations as between parties, each party holding a credit record and a debit record with an exchange institution, the credit records and debit records for exchange of predetermined obligations, the method comprising the steps of:

(a) creating a shadow credit record and a shadow debit record for each stakeholder party to be held independently by a supervisory institution from the exchange institutions;

(b) obtaining from each exchange institution a start-of-day balance for each shadow credit record and shadow debit record;

(c) for every transaction resulting in an exchange obligation, the supervisory institution adjusting each respective party's shadow credit record or shadow debit record, allowing only these transactions that do not result in the value of the shadow debit record being less than the value of the shadow credit record at any time, each said adjustment taking place in chronological order; and

(d) at the end-of-day, the supervisory institution instructing ones of the exchange institutions to exchange credits or debits to the credit record and debit record of the respective parties in accordance with the adjustments of the said permitted transactions, the credits and debits being irrevocable, time invariant obligations placed on the exchange institutions.

Well before oral argument, patent law giant Professor Mark Lemley noted that deciding when something is "abstract" is a challenge. 

The problem continues to be line-drawing. Here, an abstract idea exists because, once the software implementation is removed, the steps amount to confirming credit solvency prior to executing a transaction. For future cases, once the rudimentary software implementation is removed, if all that remains is "a basic business or financial concept," an abstract idea exists. 

The Court will likely rule that preventing insolvent transaction is "well understood, widely used, and fairly uniform," whether it be Justice Breyer's examples at oral argument of a mother stopping a youth from writing too many checks, or the King Tut example, where the King is stopped by a man with an abacus who counts as the King gives away too many chits for gold, and tell the King, "Stop." 

The holding for the court should be that without any alteration of the abstract idea of solvency, Alice's patents are not patent eligible subject matter. 

Admittedly, this will leave the state of the law with many close calls. If hedging and solvency are abstract, what about uses of credit default swaps, credit, or securitization, or Bitcoin-like innovations? (I assume decentralized Bitcoin-like innovations would be able to marshal enough arguments to to escape the murky morass of 101.)

IV.B  Despite its flaws, the legal argument's solution is viable and consistent. It is to treat an abstract idea as if it were a natural law or physical phenomenon. Just like the correlation of medical properties in Mayo, or ownership of genes issue in Myriad. The argument falls outside a standard noscitur a sociis argument, where courts use the context of a list in a piece of legislation to derive content or questionable meaning of a part of the list. Here, the Court would be deriving limitations on the "abstract idea" from the three exceptions to 101 that are part of the Court-created precedent, rather than legislation, but the logic of how to apply "laws of nature, physical phenomena, and abstract ideas" exceptions to 101 will be consistent. In this way, the Court discovers how to deal with physical or natural phenomena and articles of nature. 

* Plainly, the Court has said that its cases should not be read as if they were statutes. In this case, however, long-standing precedent exists for the three exceptions. See Gottschalk v. Benson, Parker v. Flook, Diamond v. Diehr, along with Bilski, Mayo, and Myriad. The opinion will likely state that a different approach can always be taken by the legislature, if the Court's ruling eliminates too few or too many "abstract ideas" from patent eligibility. 

Regarding the recent case law supporting this approach, in Bilski v. Kappos (2010) the Court invalidated a method of optimizing a fixed bill system for energy markets as an unpatentable abstract idea. 

In Mayo v. Prometheus (2012), the Court invalidated drafting efforts designed to monopolize the scientific correlations: "Because methods for making such determinations were well known in the art, this step simply tells doctors to engage in well-understood, routine, conventional activity previously engaged in by scientists in the field. Such activity is normally not sufficient to transform an unpatentable law of nature into a patent-eligible application of such a law."

Later, in Association for Molecular Pathology v. Myriad Genetics (2013), the Court invalidated a patent because it was a product of nature that was merely isolated. 

Alice's use of a computer is a method well know in the art, and simply tells software engineers, like doctors, to engage in well-understood, routine, conventional activity previously engaged in by programmers in the field. This is insufficient to transform an unpatentable abstract idea into a patent-eligible application of such an idea.

V. The amicus briefs are excellent, but I doubt the opinion will address the normative considerations much. This is an application of ~40+ year old precedents.

VI. Stepping back from a draft opinion, here are some miscellaneous thoughts.

a. One of Alice's arguments is that demanding specificity should not be required under 101 because that is a 112 an enablement issue. I don't think that's right. First, all "abstract ideas" would be able to hide behind this argument. More importantly, the abstract idea protection exists so no one gets to monopolize a broad theory, like a physical law, applied to all uses by the public. The enablement protection is a demand by the government for the quid pro quo: teach the public your idea, patent applicant, or you get no patent. Since enablement requires the inventor to show a professional in the field how to make the claims work without undue experimentation, any claims challenged on abstract idea/mere computer implementation-type grounds can use this argument. It is precisely because Alice's computer-implementation is so easy to create by a number of Silicon Valley coffee-shop dwellers (Justice Kennedy's caricature, not mine), that the legislature would agree invalidity should be tested under 101 as well. In that sense, it is a complementary means of rejecting patents that are too easy to create, and where too little experimentation is required. 

b. A holding focused on the above should meet the concerns identified by Justices Kennedy and Breyer at oral argument, namely, that improvements to software systems, computers, robotics, etc., will pass muster under 101. Once you strip away the conventional software activities, there must be some alterations of abstract ideas like email or communication that are not conventional. See, e.g., Swype. I don't think Justice Kennedy's focus on what he termed "mechanical" should be a focus, as it's not based in the statute or case law.  

c. Alice arguued that a parade of horribles will occur if their patent is also obviated - that patents on cell phones, word processors, and other everyday objects, and uses of software on microchips, will be in doubt. But the ideas can be represented as something more than an abstract idea of "communicating," for example with an electronic interface or a data medium, as long as the inventor explains how the ideas of communicating (through manipulating 1s and 0s), securing privacy (through encryption), etc., are manipulated or altered. 

Relatedly, during oral argument, the government at one point said business methods like frequent flyer miles should not be subject matter eligible. I don't see the statutory basis for that. 

d. Ultramercial, LLC v. Hulu, LLC, 657 F.3d 1323 (Fed. Cir. 2011) will have incorrectly found that "The application of an abstract idea to a 'new and useful end' is the type of invention the the Supreme Court has described as deserving of patent protection." 


e. Many amici argued that software patents are a bad idea. While subject matter eligibility arguments are one attack, what is often forgotten is that patents need not all be one-size, and other ideas for the patent system could be worthwhile. 

Perhaps Swype does not need a 20-year patent to provide sufficient incentive to innovate. Would 5 years suffice? Maybe pharmaceutical innovation companies do, in order to recoup the investment of billions for FDA and related testing, and for funding of a portfolio of "losers" before finding the next big winner.

Legislation would be required, but shorter patent durations, and some other "dismal" ideas still exist, such as patent buyouts.