Patent Law: Baby Steps
Supreme Court Hears Arguments in KSR summarizes the oral arguments for KSR International v. Teleflex, an important patent case pending before the Supreme Court. The question involves whether it should be more difficult to obtain a patent for a claimed invention in a patent application.
In our patent system, a patent may be obtained if four criteria are satisfied: (a) if it is the type of thing that the patent statute covers, i.e. "statutory subject matter" (this is pretty broad, of late—it includes any practical invention—"anything under the sun that is made by man"—this is "Section 101" of the US Patent Act); (b) has "utility" (pretty easy to satisfy, unless it does not work at all, like a perpetual motion machine, or if it has only "negative" utility, e.g. a weapon that would destroy all life on earth—Section 101 again), (c) which is "novel" (this means new—section 102); and (d) which is "non-obvious" (section 103).
This last—non-obviousness (called "inventive step" in most other countries)—is usually the most difficult hurdle to overcome. This is because many things are "new" combinations of existing things or elements, but "would have been" considered "obvious" to a hypothetical "person having ordinary skill in the art" (PHOSITA).
In other words, an invention is not novel if you can find a single prior art "reference" (say, a previous patent or article) that fully describes it. But if the invention is a combination of elements that is a new combination of elements and not all found in a single reference, then it is novel. So then the Examiner has to find multiple references that teach all the elements.
Now originally, the Supreme Court held that simply aggregating old elements together does not pass the patentability test if the elements are each performing the same functions that they were already known to perform. (See pp. 36 et seq. of this brief.) But the Court of Appeals for the Federal Circuit (CAFC), the federal appellate court with exclusive appellate jurisdiction in patent law cases since it was established in 1982, has held that a claimed invention cannot be held obvious unless the patent office can show some proven "'teaching, suggestion, or motivation' [TSM] that would have led a person of ordinary skill in the art to combine the relevant prior art teachings in the manner claimed." That is, the CAFC's approach has allowed mere aggregations to be patented so long as there is no evidence suggesting that the elements be combined. (Here one gets a glimpse of the utter arbitrariness and inherent vagueness of IP law and its complete dissimilarity to real property law.)
So under current practice, using the example above, the PTO Examiner has to find two or more references that show all the elements of the claimed invention, AND he must somehow show some other teaching or reference that suggests combining the references. For example, suppose it's known to make brooms with straw bristles; and it's known to use plastic for a variety of purposes. Someone tries to patent a broom having plastic bristles. Let's assume this is novel—no one has done this before. Is it "obvious" to combine plastic (which is known to be widely used, and as a substitute for other materials) with the standard broom design (also well known)? What if the Examiner can't find an article where someone muses, "Say, you know, you could use plastic instead of straw in a broom if you want—..."?
The point is that quite often, you can get an invention patented by making sure it includes some fairly trivial feature that makes it "novel" and then, when the Examiner tries to argue well, this is just obvious, you tell him he has to prove it by finding an explicit TSM (teaching, suggestion or motivation) out there that suggests combining the elements. (And let's face it: all inventions are combinations of existing prior art; this is the nature of progress.) In other words, under the current TSM test, claims "cannot be held invalid . . . no matter how trivial might be the differences" from the prior art unless evidence suggesting the difference is found in the art.
In the current case, KSR manufactures gas pedals that can be adjusted for the height of the driver and uses electronic signals rather than a mechanical cable to accelerate when the pedal is pushed. Both features were developed separately but Teleflex sued KSR claiming that KSR's combination of the two features infringed on a patent it was issued in May 2001. KSR argues that the patent should be invalidated because the combination of the two features is obvious. They want the Supreme Court to overrule the CAFC, which "held that the prior art did not render the claimed invention unpatentable because they had not proven, beyond genuine dispute and by clear and convincing evidence, that there was a suggestion or motivation to combine the teachings of the prior art in the particular manner claimed by the patent at issue (U.S. Patent No. 6,237,565)."
So what will the Court do? It is possible the TSM test may be modified or dropped, and I hope it is. This would mean it is easier for the patent office to show that a claimed invention is obvious, since it will not have to show a TSM to combine well-known elements that the patent applicant has merely aggregated in an obvious way. The presumption may be shifted, so that if all the elements claimed are shown in multiple references, the patent applicant has to show that the invention is non-obvious, rather than the PTO (or opponent at trial) having to show some TSM to aggregate or combine the well-known elements.
It's interesting that some companies are against the TSM test (that is, they are in favor of making it more difficult to obtain patents), and some are in favor of it. "Critics of the test argue that this has made it too difficult to prove that a claimed invention is obvious." The Coalition for Patent Fairness, for example, which incldes high-tech companies like Intel, Cisco Systems and Hewlett-Packard, seems to be in favor of revision of the current test. "The tech companies [Microsoft Corp., Cisco Systems Inc., Intel Corp. and others] want to put a damper on all the patents on all the patents on business methods, designs and undeveloped technologies that casues them to spend lots of money to defend." (One of their concerns is "patent trolls - companies that exist primarily to make money from patents through litigation instead of commercialization"—but as I've noted, the hostility against patent trolls is misplaced. They are no different than any other patentee who takes advantage of the corrrupt and unjust patent system.)
But pharmaceutical and biotechnology research-based companies such as Johnson & Johnson, GE and DuPont, for example, have filed briefs "arguing that major changes to the patent system would jeopardize billions of dollars invested in product innovation. They want to protect pharmaceutical and chemical products from generic manufacturers and counterfeiters. These companies oppose any change to the test since it might make it difficult to patent known drugs for new uses.
I am hopeful the Court will decide to narrow or drop the TSM test, but even if they do, it's only a mild improvement to this horrible morass of artificial and unjust law.