Letter from AmeriKat II: Bilski, Baby (Justice Stevens - American Patent Law)
The birth of American Patent Law
During the Constitutional Convention (picture, left) in 1787, the Founders gave Congress a patent power so that it might "promote the Progress of...useful Arts" (aka the Patent Clause, Copyright Clause or IP Clause - the AmeriKat prefers the former). However, little is known about the Patent Clause because it was passed without debate or objection [ironically, given how often this Clause is now debated in IP and how cautious the Framers were in granting monopolies]. In 1790 the first Patent Act was passed that authorized patents for individuals who had "invented or discovered any useful art, manufacture, engine, machine, or device, or nay improvement therein not before known or used" if "the invention or discovery [was] sufficiently useful and important." In 1793, Congress modified the language slightly to cover "any new and useful art, machine, manufacture or composition of matter or any new and useful improvement on any art, machine manufacture or composition of matter."
Both statutes and early patent law had as their object of the constitutional patent power the term "useful art." Whether or not this meant "arts" which were also "useful" or a category known as "useful art" is unclear, however. Both avenues of definition encompass many meanings and with that, Justice Stevens, begins a 7-page undertaking into what was meant by "useful art."
The first edition of Noah Webster's American Dictionary of the English Language in 1828, only forty-sum years after the Constitutional Convention, is used by Justice Stevens as a reference point for the definition of the term "art". The dictionary defined "art" as the "disposition or modification of things by human skill, to answer the purpose intended" and differentiated between "useful or mechanic" arts on the one hand, and "liberal or polite" arts, on the other. Justice Stevens stated that Webster's definition was likely to convey a message similar to the meaning of the word "manufactures" in the earlier English statute. To support this view, Justice Stevens sites that just days before the Constitutional Convention, "one delegate listed examples of American progress in 'manufactures and the useful arts', all of which involved the creation or transformation of physical substances". Scholars suggest that the term "useful arts" would be understood today to relate to "technological arts". As such, fields like business and finance were not generally considered part of the 'useful arts" at this time.
However, Thomas Jefferson (the first administrator of the US patent system and author of the 1793 Patent Act), like many scholars now, have remarked that because the extent and scope of patentable subject matter was never seriously debated at the time of the drafting, one cannot say with any certainty what was considered to be patent-eligible under this definition. That being said, Justice Stevens makes it clear that although certain processes relating to technology might have been considered patentable, against the historical backdrop it would have "seen as absurd for an entrepreneur to file a patent" on methods of conducting business (Merges, Property Rights for Business Concepts and Patent System Reform, 15 Berkeley Tech L J 577, 585 (1999)).
American Patent Law learns to crawl
Initially during the first few years of the patent system no patents were issued for business methods and were in fact subject to much doubt as to their patent-eligibility. Jefferson later explained that the 1793 Act was drafted in a way which allowed it to be "turned over to the judiciary, to be matured into a system, under which every one might know when his actions were safe and lawful". Congress agreed with this view and during the next 160 years Congress steered clear, enabling the courts room to exercise their judicial discretion. This system worked well, notes Justice Stevens, with the courts consistently rejecting business-method patents. The reasoning of these decisions varied at times, but the theme was consistent and widely understood that "a series of steps for conducting business could not be patented" (at pages 26-27 of his decision Justice Stevens cited several such cases including US Credit Sys Co v American Credit Indem Co (1893), Hotel Security Checking Co v Lorraine Co (1908), Loew's Drive-In Theatres, Inc v Park-In Theatres, Inc (1949)). The Supreme Court never addressed the patentability of business methods during this time, either. Instead they focused the inquiry on whether the "art" was connected to a machine or transformation. Such an inquiry excludes methods of doing business and in any event, indeed methods of doing business was not an "art".
American Patent Law learns to walk and in doing so, walks "under the Sun"
By the 1950s, the courts were beginning to construe the term "art" with reference to words such as "method, process, system or like terms". With the reference of "process" entering into the mix, Congress updated the patent statute in 1952 to change the operative language of Section 101. This replaced the term "art" with "process" and added the much unhelpful definition of "process" in Section 100(b). At the time the change was only made for clarity-sake in recognition that "courts had been interpreting the category 'art' by using the terms 'process or method" and was therefore not intended to alter the scope of patentability. A House of Representative Report in 1952 had in fact explained that "the word "art" in Section 101, "has been interpreted by the courts as being practically synonymous with the process or method." Justice Stevens reiterated again that the change in the 1952 Patent Act did not actually change anything - it just codified a judicial interpretation deriving from the very courts who were also consistently excluding methods of doing business from the definition of "art".
But despite this strong legislative backdrop, the decision in Chakrabarty and the oft-quoted "anything under the sun that is made by man" phrase from the House Committee Report began to be the thorn in the Court's proverbial paw. Justice Stevens says that reliance on the Chakrabarty case is misplaced. Strict reliance on the case would have the effect of construing the 1952 legislation as meaning that "any series of steps is a patentable process." In context, the quote actually has a much less expansive language. In fact the language from the 1979 House Committee Report reads as follows:
"A person may have 'invented' a machine or a manufacture, which may include anything under the sun that is made by man, but it is not necessarily patentable under section 101 unless the conditions of [this] title are fulfilled."
It does not therefore purport that "anything under the sun" is patentable. In fact it may actually be construed to mean that a man can invent anything under the sun, but that does not mean it is necessary patentable under Section 101. In the Chakrabarty opinion, which relied on this quote, the Court cautioned that the 1952 Reports did not suggest that Section 101 had no limit on or indeed embraced every discovery. Further the language referred only to "manufacture[s]" and "machine[s]", not tangible objects made by men. It also does not refer to the "process" category of patent-eligible subject matter. In actuality, the Chakrabarty and Committee Report language is only to be understood as defining the term "invents". That is to say, that the invention "must be made by man" and not an abstract philosophical principle, for example.
The 1952 Act, concludes Justice Stevens in this penultimate section, cannot be understood as expanding the scope of patentable subject matter by the mere insertion fo the term "process" in Section 101. "If anything", he writes, "the Act appears to have codified the conclusion that subject matter which was understood not to be patentable in 1952 was to remain unpatentable...Our recent case law reinforces my view that a series of steps for conducting business is not a "process" under Section 101. The Court never ruled on whether the 1952 Act authorizes business method patents, but doubt was cast on such a proposition by giving "substantial weight to the machine-or-transformation test." Justice Stevens concludes:
"A business method is not a "process." And to the extent that there is ambiguity, we should be mindful of our judicial role. '[W]e must proceed cautiously when we are asked to extend patent rights" into an area that the Patent Act likely was not "enacted to protect' (Flook, at 596, 593), lest we create a legal regime that Congress never would have endorsed, and that can be repaired only by disturbing settled property rights."
American Patent Law Reacts to State Street and gives us Section 273
Although the "historical evidence" that business method patents are not patentable is strong, Bilski and the Court suggested that a subsequent patent law, the First Inventor Defence Act 1999, must be read together with Section 101 in order to make business methods patentable. The 1999 Act followed a Federal Circuit decision that intimated business method patents could be patented (State Street Bank). Congress swooped in to limit the possible fallout with the 1999 Act which provided a limited defence to patent infringement claims. Section 273(b) of the Act refers to certain "method[s] of doing or conducting business". The Act in no way ratified the State Street implication, but only limited a potential effect of the decision, i.e. that businesses might subsequently find themselves liable for innocently using methods of business they assumed that were not patentable. It had no effect on the categories of patentable subject matter in Section 101. To support this reasoning, Justice Stevens referenced the fact that Congress placed the defence in Part III, rather than Part II where Section 101 is found. Just because Congress thought it fit to create a new defence to address a potential problem, does not mean that the business-method patents were therefore at first instance patent-eligible.
Justice Stevens found particular fault, therefore, with the Court's reasoning that if it was to interpret the 1952 Act to exclude business methods, it would automatically render Section 273 meaningless. Although admittedly statue should be construed to give effect to all of its provisions, that does not mean that the 1952 Act should give effect to the 1999 Act. Although the Court was priding itself on the canon that "a statute is passed as a whole", the 1952 Act and the 1999 Act were not passed as a whole. That is not to say that the two are at odds with each other, rather that the 1999 Act sought only to limit some judicial creativity in the Federal Circuit.
The Constitutional Balance says "No" to Business Method Patents
In his final section, Justice Stevens argues that the Court must not exceed the constitutional limitation of the Patent Clause and subsequent patents statutes on Congress's behalf. It is for the judiciary to act within the the constitutional standard set down by the Framers, not to exceed it. Without any legislative guidance that suggests otherwise, holding that business metod patents are patent-eligible would place great pressure on the limits of the constitutional standard and stifle progress. Although Justice Stevens himself recognized that not all metods of doing business are the same and there is still a constitutional balancing exercise to undertake, he states that the balance supports the historic understanding that the term "process" excludes business methods.
Justice Stevens doubts that business-method patents are necessary for progress and innovation. His argument is four-fold. First, there are already ample incentives for companies to develop business methods without the need for patent protection, because "the competitive marketplace rewards companies that use more efficient business methods" (Burk & Lemley, Policy LEvers in Patent Law, 89 Va. L. Rev 1575, 1618). Second, he argues that business innovation does not entail the same amount and extent of risk in terms of research, time, and cost. Third, one of the patent system's goals is to promote public disclosure of useful inventions that may otherwise be practiced in secrete. Justice Steven argues that public disclosure is an unnecessary objective for business methods because many are practice in public anyway, so the benefits of such disclosure are thus nil. Finally, Justice Stevens argues that even if business methods patents were useful in encouraging innovation, "too much patent protection can impeded rather than 'promote the Progress of . . . useful Arts." (Laboratory Corp of America Holdings v Metabolite Laboratories Inc (2006)). To allow such patents may consequently prohibit "a wide swath of legitimate competition and innovation" and result in businesses living in "constant fear of litigation".
Justice Stevens stated that all these effects are "magnified" when on considers the vagueness of business method patents (eBay Inc v MercExhcange). Patents must pride themselves on clarity if they are to promote progress, however, patents for methods of conducting business are generally composed of breathy, intangible steps. Such vagueness of a business method patent signal a further danger to the patent system, innovation and competition if permitted.
"Patents on business methods are patents on business itself. Therefore, unlike virtually every other category of patents, they are by their very nature likely to depress the dynamism of the marketplace."
The AmeriKat feels, that in his last IP judgment as a Supreme Court Justice, Justice Steven's concluding paragraph must be reproduced here in all of its full glory. So here it is:
"The Constitution grants to Congress an important power to promote innovation. In its exercise of that power Congress has established an intricate system of intellectual property. The scope of patentable subject matter under that system is broad. But it is not endless. In the absence of nay clear guidance from Congress, we have only limited textual, historical, and functional clues on which to rely. Those clues all point toward the same conclusion: that petitioners' claim is not a "process" within the meaning of Section 101 because methods of doing business are not, in themselves, covered by the staute. In my view, acknowledging as much would be a far more sensible and restrained way to resolve this case. Accordingly, while I concur in the judgment, I strongly disagree with the Court's disposition of this case."
Justice Breyer's concurring judgment with Justice Stevens tried to unified the 'two courts' in Bilski by identifying grounds which both agreed. He found four grounds of agreement which, although not ground-breaking areas of agreement, were as follows:
- Section 101 is broad but it is not without its limits. Therefore the Court must be careful to only protect what is within the confines to do so.
- There are a series of decisions of the Court that identifies that the machine-or-transformation test is the clue that business method are not patentable.
- Although the machine-or-transformation test is a useful clue it is not the sole test.
- Although the machine-or-transformation test is not the only test for patentability,it does not mean that anything which produces a "useful, concrete, and tangible result" is patentable.
So what happened with Justice Stevens' opinion? Well, it is clear from his lengthy reasoning he was prepared to as far as possible unequivocally state that business methods were not processes under Section 101 and not patentable. The majority opinion did not want to tread that far, despite Justice Kennedy himself stating that business method patents were "of suspect validity" in eBay v MercExchange. One may guess that to go as so far as Justice Stevens & Co is to get too close the edges of then asking whether software should be patentable. In any event, it is clear that the Court, regardless of which side the opinions fell down on, are not a fan of the Federal Circuit's State Street case. This may signal, then, in the lower courts an opportunity for a new test case to be brought to give the Federal Circuit an opportunity to formulate a new test for patentability of "process" patents.
