A patent dispute for the Information Age
At 1 p.m. Monday, the Supreme Court will hear one hour of oral argument in Bilski, et al., v. Kappos (08-964). Arguing for two inventors will be J. Michael Jakes of Finnegan, Henderson, Farabow, Garrett & Dunner in Washington, and arguing for David J. Kappos, the head of the federal government’s Patent and Trademark Office, will be Deputy Solicitior General Malcolm L. Stewart. The briefs and other filings in the case are available at this link on ScotusWiki.
In 1981, the Supreme Court last decided a case spelling out the kinds of inventions that are eligible for patent rights under federal law. Now, in the wake of major changes in the world of commerce, the Court will try to provide a modernized definition of patent eligibility. Some experts say the outcome may affect the legality of more than 130,000 patents that already exist, and the legal fate of untold future inventions, especially in digital commerce. Background
Just because someone comes up with a bright idea, even a completely original one, does not mean that they can get a patent on it. To be eligible for the legal right to the exclusive use of an invention, or to license others to do so for a fee, an idea has to be made into something useful; in other words, the idea has to do something, in practical, real-world terms. If it doesn’t, the idea remains an unpatentable abstraction.
Under the federal Patent Act as it now reads, four kinds of inventions or discoveries are patentable, provided that they are “new and useful.” One is a process for doing something practical. One is a machine that produces an item or a result. A third is a creation, or “manufacture,” that turns raw material into something new. And the final one is a mixture or combination of substances. The Supreme Court is now focusing, in the case of Bilski, et al., v. Kappos, on the first one: a process. At an earlier stage of patent law, an invented process was considered to be one of the “useful arts,” as opposed to a “fine art,” such as a portrait.
In 1981, in its decision in Diamond v. Diehr, the Court provided a definition of “process,” saying the clue to a patentable process that did not involve a particular machine was that it transformed and reduced an article “to a different state or thing.” Picking up on both parts of that definition, the specialized appeals court that decides how to apply the Patent Act to new inventions, the Court of Appeals for the Federal Circuit, has now spelled out the only kinds of process it would find eligible for a patent: it has to be”tied to a particular machine or apparatus,” or it has to “transform a particular article into a different state or thing.” That is now commonly referred to by the shorthand phrase, the “machine-or-transformation” test.
That test emerged at the end of October 2008, in a Federal Circuit en banc case involving an invention by Bernard L. Bilski and Rand A. Warsaw. While some of the focus in the case necessarily remains on the details of their invention, and whether it fits within the Patent Act’s reach, the controversy has taken on a more fundamental character, asking the Act’s most basic question: what is patentable?
The case has generated high levels of interest here and abroad. A wide array of amici, totaling 38, joined in the case at the Federal Circuit, and that intense interest has been sustained as the case has moved forward in the Supreme Court, with 68 amici filings on the merits. It is the most-watched patent case in years, with many analysts speculating on how the outcome will affect innovation as the bounds of what is patentable are regularly being tested in an era of rapidly changing technology.
In seeking a patent, Bilski and Warsaw told the Patent Office in 1997 that their idea was a highly useful one: using complex mathematical formulas, they could tell a business how to hedge against risk due to the rising and falling of prices of raw materials that were used to produce something — say, to generate electricity. Commodities prices often fluctuate quite widely, because of market forces or even changes in the weather, so these two inventors figured out ways to manage what they called “consumption risk.” It is, they claimed, of benefit both to businesses and to their customers.
An example that lawyers for Bilski and Warsaw now use is this: “A school district with a fixed tax base and budget for heating or cooling requirements can be protected from yearly fluctuations in weather, while the suppliers are protected from the opposite effect of such fluctuations.” In other words, the school district can be insulated from spikes in demand — and rising prices — while the suppliers can hedge against a weakening of demand — and lower prices.
In effect, what these inventors claimed was that they had created a method of doing business by evening out risk among those in an ongoing economic transaction. One distinctive feature of it — and this became a legal problem for the inventors — is that it is not tied to a machine. While Bilski and Warsaw say that a computer, or modern telecommunications networks, may actually be used in employing their idea, that is not essential to its utility. It is thus being treated as if it were, technically, a “non-machine” invention.
A patent examiner rejected the invention, finding it merely allowed manipulation of an abstract idea, solving a math problem. Next, a patent appeals board agreed with the examiner, concluding that Bilski and Warsaw’s risk management formulas involved only non-physical manipulation, beyond the scope of the patent law. The case then moved on to the Federal Circuit; before a three-judge panel could rule on the case, the en banc Court took it on, and ultimately ruled against the Bilski-Warsaw invention.
If the ordinary meaning of “process” were used, the Circuit majority conceded, this invention would qualify because it does involve a series of actions or operations bringing about an end. But, it went on, the Supreme Court has given the word a narrower meaning in the Patent Act. With some hesitation, it translated the Supreme Court’s Diamond v. Diehr comment about “the clue to patentability of a process claim” into the machine-or-transformation test. It then found that the Bilski-Warsaw invention “entirely fails” that test. One of the three dissenting judges complained that the test “links patentability to the age of iron and steel at a time of subatomic particles and terabytes.”
In late January 2009, the large intellectual property law firm, Finnegan, Henderson, Farabow, Garrett & Dunner, filed its petition in the Supreme Court, challenging the new test.
Petition for Certiorari
Lawyers for Bilski and Warsaw raised two challenges to the validity of the machine-or-transformation test. First, they contended that the test contradicted Supreme Court precedents mandating broad interpretation of the patent grant. Second, they contended that the test ran afoul of action by Congress specifically allowing patents for methods of doing or conducting business.
The petition’s argument for review began with three questions: “What can be patented? Are patents only for manufacturing processes that are tied to a particular machine or produce some physical transformation? Or do patents also embrace modern business processes that do not depend on a particular machine or device?”
The Patent Act, it contended, is broad in scope, using the word “any” to introduce a concept of sweeping access to patent protection. “There is no exclusion for business methods or any other field of invention.” Only three categories are excluded: laws of nature and natural phenomena, neither of which, it said, can be invented at all, and abstract ideas, which are not useful until they become part of something practical.
The petition interpreted two prior Supreme Court precedents as actually declining to embrace the machine-or-transformation test. Moreover, it said that even the Federal Circuit had not done so previously, using a test of whether an invention produced a “useful, concrete and tangible result.” Using that approach, the filing said, the Circuit Court had “made patent protection available to such diverse fields as internet commerce, information technology, and business methods.”
Thus, it contended, the Circuit Court “abruptly changed course” in making the new test the sole one for patentability. “The Federal Circuit,” it asserted, “has essentially confined all process patents to manufacturing methods, using a test that may have been appropriate during the Industrial Age but no longer fits our modern information-based economy.”
The then-new team at the U.S. Solicitor General’s office twice got extensions of time to respond to Bilski and Warsaw. Then, in a filing on May 1, representing the Patent Office, the government’s lawyers urged the Court to deny review. The response defended the test, arguing that it actually was not new at all, but rather had been “drawn directly” from the Supreme Court’s 1981 decision interpreting patentability. Embracing that test, it added, the Circuit Court then properly cast aside its own precedents in order to conform to the Supreme Court’s approach.
“The decision below represents an unremarkable application of that machine-or-transformation test,” the Solicitor General asserted. The Bilski-Warsaw invention, the brief said, has no connection to “a particular device or article,” and does nothing to transform any physical object or substance. “The Court,” it went on, “has never suggested that a method of this kind, relating to purely human activity, may properly be the subject of a patent.” The invention is nothing more than a process “that is directed to the ‘abstract idea’ itself,” it commented.
What’s more, the Solicitor General asserted, the Federal Circuit “expressly left room to accommodate emerging technologies” and did not rule on the application of its test “to computer software, data-manipulation techniques, or other such technologies not involved” in the hedge claim behind the Bilski-Warsaw invention. And that invention “involves none of the frontier technologies” that the inventors’ lawyers had argued would be stifled by the test. There is no threat here, the brief added, to process claims that would describe a programmed computer’s operation, biotechnology advances, or chemical inventions.
The Court granted review on June 1.
Continuing to press the argument for an expansive reading of the word “process” in the Patent Act, the Bilski-Warsaw brief on the merits marshaled quotations from Supreme Court opinions back to the mid-to-late 19th Century, emphasizing two words in particular — “art” and “useful” — as the defining nature of a patentable process. In fact, the word “art” is traced back to one of the earliest Patent Acts, in 1793.
The brief, although attempting to persuade the Court that the modern Information Age needs an expanded concept of process, turned back to history for one of its most vivid examples of an “information-age process.” That was Alexander Graham Bell’s invention of the telephone, patented in 1876. In The Telephone Cases in 1887, the brief recalled, the Court validated Bell’s patent, and reinforced the view that “the Supreme Court’s broad framework for patent eligibility is flexible enough to accommodate innovations of every age.”
In more recent times, the brief asserted, “this Court has repeatedly cautioned against adopting special, rigid rules for patent cases where this Court’s precedents follow a broader, more flexible framework.” The Federal Circuit has had to be corrected before, the brief noted, when it attempted “to impose rigid rules for patents despite a flexible framework set forth by this Court.”
The brief goes to greater lengths than did the opening petition in marshaling recent enactments by Congress, supposedly demonstrating that the lawmakers had shown a special regard for “business method patents.” And, since the Bilski ruling by the Federal Circuit, the document said, the Patent Office and some courts have used the new test “to reject or invalidate business method claims in dozens of cases.” One judge who did so, it said, commented that “Bilski‘s holding suggests a perilous future for most business method patents.”
From a former position of neutrality on technological advances, the brief argued, the Federal Circuit has adopted a position of hostility to “innovation in the knowledge economy.” That, it added, is an abrupt policy shift of the kind that only Congress, not a court, should make.
The Patent Office’s merits brief has one frequently restated theme: patent protection simply is not available for “methods of organizing human activity.” By this, it said near the beginning, it meant “methods by which people conduct economic, social, or legal tasks, such as entering into contracts, playing poker, or choosing a jury.” To qualify for a patent, the brief summed up, a process must be “technological and industrial” in nature, “tethered to technology.”
The Bilski-Warsaw method of hedging commodity price risks, the Solicitor General argued, “relates solely to human conduct, untethered to any technology — any machine or transformation of matter.” Thus, the government presented an unqualified embrace of the Federal Circuit’s test.
Indeed, the brief went on to find the direct antecedents of that approach in the first patent laws, in 1790 and 1793. From that period to the present, it added, it has been understood that patent protection was to be available only in “the fields or technology and industry, as opposed to the fields of general knowledge and economic endeavor.”
That, it asserted, is what led ultimately to the Court’s 1981 decision in Diamond v. Diehr, spelling out that “the distinguishing feature of a technological process is that it concerns a particular machine or apparatus or effects a transformation of matter to a different state or thing. ”
Far from conceding, even impliedly, that the government was arguing for a narrow scope of patent protection for processes, the merits brief returned to Thomas Jefferson and the concept that “ingenuity should receive a liberal encouragement,” through what the government said are “broad and flexible patent laws.”
But, it stressed, ” ‘broad’ does not mean unbounded.” There are “social costs” in granting a monopoly as an incentive for research and innovation, including the exclusion of “would-be competitors who could produce a patented invention more efficiently, thereby inhibiting post-invention competition and innovation.”
The Bilski-Warsaw patent itself, according to the government, poses some of this risk: “It would preempt the abstract idea of hedging consumption risk.” Quoting a 1948 Court ruling, the brief asserted that “laws of thermodynamics, mathematical formulas, abstract ideas, and other phenomena of nature are part of the storehouse of knowledge of all men…free to all men and reserved exclusively to none.”
The Solicitor General offered the Court a two-step inquiry on the eligibility of an invented process for a patent. The first step is the machine-0r-transformation test. The second is a determination of whether the process would have the practical effect of wholly preempting public access to “the basic tools of scientific and technological work.”
Responding to the inventors’ assertion that the Federal Circuit’s test would impair the “knowledge economy” — a point echoed in a number of amici briefs voicing concern over the Federal Circuit’s test, the Solicitor General argued that the test is not a threat to computer software that handles data and information in novel ways. The risk-hedging invention here, it noted, is not software. But, in any event, it added, the Federal Circuit’s test “contemplates that many forms of ‘software’ inventions are patent-eligible.”
And, the brief sought to reassure the Court that the Federal Circuit remained open to modifying the test if it should present a threat to future developments in technology. “At present, however, no such ‘departure’ is necessary,” it concluded.
The government brief also interpreted Congress’s recent actions seeming to embrace business method patents as dealing only with defenses to infringement actions, and definitely not an endorsement of patenting “business methods directed solely to human activity.”
If nothing else about the case leads the Court to see it as highly important, the flow of amici filings surely should. In a volume that is usually reached only in the most highly visible, most deeply controversial cases, the friend-of-court briefs are stacked high. There are 68 overall, but the most remarkable fact is that nearly half of them — 26 — do not support either the inventors or the Patent Office. Briefs “in support of neither party” seldom are filed in any significant number. The 26 that seek to stand neutral between the contestants manifest a basic worry over the present state of patent law, and especially as it applies to the Digital Age.
In addition to those briefs, there are four others that urge that the Federal Circuit be upheld, but on grounds different from those argued by either party directly in the case. For the contestants themselves, Bilski and Warsaw attract 18 amici filings, while the Patent Office draws 20.
Perhaps there is no such thing as a truly representative argument from that long list of those supporting neither party, but the flavor of many of those filings is this opening statement in a brief by TELES AG, “a German high-technology ccompany”: “It is clear that the future economic strength of the United States will rely greatly on cutting-edge technologies — such as artificial intelligence, genetic programming, and human-machine communicatrions — that in the broadest sense are not ‘physical’ or ‘tangible.’ But the ‘machine-or-transformation test’ set forth by the Federal Circuit gives cause for serious concern that patentability will be statically limited to ‘physical’ or ‘tangible’ implementations of technical developments, largely excluding innovations in such cutting-edge technologies.”
Many of those briefs — indeed, many briefs from all sides of the amici briefing contest — offer their own formulations of a proper patentability test for process inventions. A group of 20 law and business professors offered one of the simplest: “Where an idea is claimed as applied, it is eligible for patentability, but if it is claimed merely in the abstract it is not.”
To a noteworthy extent, many of the briefs on all sides (including some in support of the inventors’ challenge) came prepared to say that, whatever the test, the risk-hedging invention of Bilski and Warsaw does not qualify. For example, the Boston Patent Law Association, while complaining energetically about the Federal Circuit’s test, suggested that the Justices may want to rule directly on the Bilski-Warsaw claims, and reject them as “lacking adequate limitation to a specific application,” and thus advancing only an abstraction.
Predictably, the Bilski-Warsaw invention comes in for especially aggressive challenge by the amici supporting the Patent Office. Many of these groups and individuals are deeply upset by a recent outpouring of patents on business methods and software, contending that this is locking up ideas that others should be free to exploit. One brief on this side, from entities ranging from Google to the Bank of America, argued that this trend is allowing the patenting of abstract ideas or methods “not to make productive use of them, but to extract licensing fees from businesses that apply and improve those ideas and methods in real-world products and services.” Belitting the work of Bilski and Warsaw, that brief said their invention “claims nothing more than the idea of hedging against the weather.”
Unless the Justices are now persuaded by an argument that did not keep them from agreeing to hear this case in the first place — that is, that the Federal Circuit has done nothing more than faithfully adhere to Supreme Court precedent, a daunting task awaits. It may involve no less than trying on its own to invent something: a new formulation of the concept, for patent purposes, of what is “useful.”
The briefs put before the Court not only a dizzying array of ideas on that very point, but throw in a high degree of alarm about where the patent system as a whole may be heading in a world increasingly dominated by those most elemental of electronic expressions, ones and zeroes — the language of digital commerce and conversation. Although the Patent Office and many of its supporters suggest that the Court’s own record on patentability is a quite consistent pattern since 1790, and that it has already adapted to the “knowledge economy,” there clearly is much disagreement with that.
There is a deep chasm between those who think too much innovation is being locked into patent monopolies, and those who think too little is getting the protection of exclusive legal rights, and yet the Bilski case seems to ask the Court to reconcile the two.
Even if the Court were to see the case as being limited solely to “business method” inventions, that is a concept that now has so many variables, present and future, that defining it does not appear to be any easier than saying — in legal terms — what the word “useful” means in the Patent Act.
Perhaps the only way that the Court’s work could be simplified is to examine the finite details of the Bilski-Warsaw invention and conclude that, whatever the test, this idea does not pass muster. That would end the case, but clearly not the controversy over patentability. On that, the Court is being pressed hard to provide a definitive resolution.