Most law professors hold citation by the Supreme Court as a lofty goal. The best result would be adoption of a proposal made in an article, but most academics would settle for square consideration of a proposal, even if rejected. And then there’s the middle ground, when you can’t really tell if you’ve been accepted or rejected. It turns out that’s alright with me as well.
Today, the Supreme Court struck down a personalized medicine patent in Mayo v. Prometheus. I blogged about this case after oral argument. In short, the patent covered the following: administer a drug, measure the residual effects of the drug, decide whether more or less of the drug should be administered based on the findings. Anyone who has taken thyroid or blood pressure medicine has probably experienced a similar process.
As I predicted in my last post, Prometheus lost and the patent was struck down. After the Court’s decision in Bilski v. Kappos, the decision is hardly earth shattering. In short, the Court held that abstract ideas and laws of nature are not patentable subject matter, and they don’t become patentable subject matter by simply adding insignificant stuff to the claim.
There are continuing problems with this formulation that I’ll detail later but, first, there’s a lot to like in the opinion. First, it is minimalist – it does not strike down large chunks of subject matter but instead applies an old rule to to a particular patent. Many medical diagnostics and even personalized medicine patents may survive this opinion. Or maybe not. Second, the Court rejects the Federal Circuit’s application of the machine or transformation test (again) as to this case. As I’ve written elsewhere, the test is ridiculous, and its application to this case is especially ridiculous. I’m publishing a paper that shows that the test is not even based on historical patenting practices. Third, as discussed below, it adds a little more detail about how we might determine whether something is a natural law (or an abstract idea). Fourth, in doing so, it cites and adopts (in part at least) a Stanford Law Review article written by Mark Lemley, Ted Sichelman, Polk Wagner, and me called Life After Bilski. In that article, we argue that one should look to the application of abstract ideas to determine whether a claim’s scope is too broad.
But the Court does not seem to adopt our view entirely. This opinion adopts a long line of Supreme Court jurisprudence that says one may disregard insignificant post-solution activity when deciding whether a claim simply covers a natural law. Here, the Court makes it a bit more explicit: “To put the matter more precisely, do the patent claims add enough to their statements of the correlations to allow the processes they describe to qualify as patent-eligible processes that apply natural laws?” (emphasis in original)
The hard part is determining how much is enough. The Court gives us some indication, saying that, “Purely ‘conventional or obvious’ ‘[pre]-solution activity’ is normally not sufficient to transform an unpatentable law of nature into a patent-eligible application of such a law” and “Unlike, say, a typical patent on a new drug or a new way of using an existing drug, the patent claims do not confine their reach to particular applications of those laws.” It appears, then, that a key way to tell whether a claim is unpatentable subject matter is to see how novel and nonobvious it is in its application of the natural principle. The Court held that simple testing and comparing is not enough, and this is hardly assailable once you accept the notion that we should ignore parts of the claim we deem to be insignificant (I’m not buying into that view here).
The problem with the Court’s ruling is that, yet again, this is a subject matter decision that was easily decided, but the rule cannot be so easily applied to future patents. Of course, the Court need only rule on the case before it, but history has shown that the historic subject matter exclusions are very difficult to apply in all but the most simple claims. In fact, as I have said elsewhere, this whole line of inquiry is an unsolvable mess, and Prometheus helps only a little.
Because of this mess, I wrote a 2008 article called Everything is Patentable. In that article, I argued that every patentable subject matter decision could be explained by the Court’s concern about other patentability criteria, even if not explicitly so. The same is true in this case; the very discussion of how to identify insignificant pre- and post-solution activity appears to be based on novelty and obviousness.
Thus, I argued that we should jettison this historic exclusions such as abstract ideas and natural laws, and instead more rigorously apply rules such as novelty, obviousness, utility, and enablement. This, I argue, would result in the same outcomes but with much more predictable results based on well defined case law. As the Court notes, the government adopted this position somewhat in its briefing in this case: “But in [the government’s] view, other statutory provisions…can perform this screening function.”
This is one of those great times when the Court squarely considers your proposal…and rejects it. The Court in no uncertain terms states that the historic exclusions should apply:
This approach [relying on other sections], however, would make the “law of nature” exception to §101 patentability a dead letter. The approach is therefore not consistent with prior law. The relevant cases rest their holdings upon section 101, not later sections.
In short, the Court appears to firmly entrench the historic exclusions. So much for my argument. The Court goes on to explain how the other sections might not answer the concerns about natural laws and blocking future inventors that the Court has expressed. I disagree somewhat; in fact, I spend a good portion of Everything is Patentable showing how the very same patents would be rejected under my view. Oh well – you can’t win them all.
But here is the part where I unwittingly supported the Court’s rejection of my own argument. After the Court implicitly rejected my views in the Bilski case, Professors Lemley, Sichelman, Wagner and I wrote Life After Bilski. I think it is a really great article, but I’m biased. I’ll take multiple citations in Prometheus as vindication of that view.
However, there were four of us, each with different views of how the law should go, and we had to agree on a way forward, especially in light of the fact that (in my view) the better way (reliance on other factors) was being rejected. So, if the Court was going to apply subject matter exclusions, it should do so in a meaningful way that adds value. And so we argued in our article that subject matter is about overbroad claim scope, and we distinguished that breadth from the traditional way claim scope has been evaluated, namely enablement under Section 112.
Thus, I joined an article that backtracked from my position in Everything is Patentable. If we are living after Bilski, then we should consider Section 101 subject matter claim scope considerations differently than Section 112 enablement claim scope questions. I continue to think that the type of claim scope analysis that will now be done under Section 101 could have been done under Section 112, as we note in a footnote: “One of the authors has argued that we could accomplish the same end with a more capacious understanding of § 112… But we currently do not have such an understanding.”
The Court latched on to the difference in these two articles to support its position that patentable subject matter does work that the other sections do not do: “Compare Risch, Everything is Patentable, 75 Tenn. L. Rev. 591 (2008) (defending a minimalist approach to §101) with Lemley (reflecting Risch’s change of mind).”
On the one hand, I am flattered that someone at the Supreme Court read my articles closely enough to recognize the subtle difference in positions, and I’m giddy to have not one, but two articles cited in a unanimous opinion.
On the other hand, I hate to have supported the continued view that patentable subject matter is doing work that other sections cannot do, in large part because I have not changed my mind that the whole subject matter inquiry is a mess and should be abandoned.
So, I did express a nuanced change in view – in a world where the exclusions are in play and my broader view of Section 112 enablement is rejected, we should hope to apply exclusions consistently and our approach in Life After Bilski is a way to do that. However, I continue to believe that subject matter rejections are superfluous, and should still be handled first by more rigorous application other patentability criteria.
Two other notes about case law support for today’s opinion:
First, the Court approvingly cites to Mackay Radio for the proposition that application of a natural law is patentable. This is great, as that’s always been one of my favorite cases to show broad subject matter eligibility. But the problem with reliance on that case is that the patent in that case would surely be considered ineligible subject matter under today’s opinion. In that case, the patentee claimed an antenna that identically followed a known formula for antenna wire lengths. The antenna would probably be “routine” after today’s opinion, but the Court did not invalidate on that basis back in 1939.
Second, the Court, like many before it (dating back to O’Reilly v. Morse in 1853) cites to an old British case called Neilson in support of a limitation on process patents. As I argue in a forthcoming article called America’s First Patents, this reliance is unjustified now and was unjustified in 1850. At the time of Neilson, British law did not allow for process patents at all. Thus, early reliance on Neilson or any other British case is like applying apples to oranges. The article describes this in much more detail.
The item I would use as a classic algorithm test would be something from computer science, the RSA public key encryption algorithm. In the late 70s that was completely novel (notwithstanding NSA’s later claims to have invented the same thing earlier). As far as I known no one challenged the patent then but under the recent pronouncements of the Supreme Court it would seem like that issuance would be on shakey ground if it came up today. Certainly it is possble to perform RSA on paper, it is even possible to do in your head with stripped down versions. Yet despite all that I do think RSA should have qualified, they did bring something completely new into the world.
I have not read you paper “Everything is Patentable”, but as a practical matter its title happens to be for the most part spot on. Even Bilski, claimed with great specificity, would likely have passed muster before the court, with perhaps cert not even haven been granted.
What troubles me in general about 101 cases is that the entire notion of evidentiary requirements associated with the usual bases for patentability determination (i.e., 102, 103 and 112) are largely ignored in favor of what I charitably describe as an “I know it when I see it” test. Call it a “mental process”, a “law of nature” or an “abstract idea”, and further analysis screeches to a halt.
Adding to the confusion is that, as you rightly point out, this so-called “101 analysis” is dependent in many, if not all, instances on the examination of the claims in light of 102, 103, and 112.
Perhaps the court ultimately arrived at the proper conclusion, but the tortuous path if followed here does little, if anything, in my view, to provide definitive guidance and standard beneficial to practitioners of patent law.
In this instance I do not believe that the court once more struck down the MOT test. Quite the contrary. I believe it accepted the test as appropriate, but substituted its judgement for that of the CAFC by the simple expedient of proclaiming the “transformation” prong of the MOT test here was simply not transformative enough. In other words, there are “transformations” and then there are “sufficient transformations”.
I readily admit that mine is a simplistic reading of the decision. Nevertheless, if a goal of patent law is the encouragement of public disclosure, decisions such as this cut against the grain and may very lead to the discouraging of same.
Every patent on a compound is a patent covering a law of nature.
There is a fundamental inconsistency in section 101 itself.
And, frankly, Goodyear’s vulcanized rubber patent came much closer to covering the functioning of nature than does the Parker v. Flook patent on calculating alarm limits for vulcanizing rubber. I mean, evrytime a rubber tree falls in a hot sulphur spring, we will presumably get Goodyear’s rubber, even without human intervention. On the other hand, there is nothing in nature that is even remotely analogous to the calculation of alarm limits. There is no magic beaver that will pull the tree out of the sulphur spring after its sap has cured.
You could go on and on with these examples. the first US patent was on a method of making potash which was this: burn wood, then burn the ash, then soak the twice burned ash in water and then evaporate the water away. Nature does this all the time, and had done this for a long time before Samuel Hopkins came along and said that man could join in this sort of fun volitionally.
Of course the 1790 Hopkins patent predates O’Reilly and sction 101 and all that, but it is tough to make out any sort of limiting principle that can be applied when section 101 analysis, which has been developed in the context of software and business methods patents, is now applied to traditional technologies. It was nice to see you pick that up with the reffie to Mackay Radio.
Like when you first presented Everything is Patentable article in 2008, I still feel the approach you expressed there is the correct approach. Put it this way: how you determine whether the claimed method/process covers a “law of nature” (versus an application of that “law”) if you don’t first look at 35 USC 112, 2nd paragraphs which says the invention is defined by “one or more claims”?
After Bilski, I already considered Prometheus’ method claims to be problematic with the final data-gathering “determining step” followed by the “wherein clauses” to delineate what metabolite levels were too high or too low. Frankly, the administering step is unnecessary and irrelevant to what was potentially “unique” about the claimed method, namely what specific metabolite levels were too high or too low. Instead, and as my patent attorney brother Mark suggests, I would have dropped the “administering step,” started with the “determining step” (or a step which simply says what was provided by the “wherein clauses” in terms of when the metabolite levels being too high or too low), and follow with a final “indicating step” (or something similar) that suggests how the drug dosage should be appropriately adjusted (depending upon whether the metabolite level suggests the drug dosage is too high or too low) when being administered to the patient. But given Breyer’s muddled reasoning and conflicting statements in his opinion, I’m not sure even such an “indicating step” would make it to the “patent-eligible zone.
You are far more charitable about what “guidance” Breyer’s opinion provides on how to reach the “patent-eligibility zone” with such method/process claims. Breyer’s opinion is based largely on nonsensical reasoning (check my article which should soon be posted on IPWatchdog) and makes conflicting the statements that leave us “mere mortals” scratching our head on how to turn a patent-ineligible “law of nature” into a patent-eligible application of that “law.” SCOTUS never seems to see it as their responsibility to provide understandable guidance to us “mere mortals” when they completely upset our patent law “applecart” like they did here. That’s why my opinion of SCOTUS is very jaded and very low at this time (and I’m not alone that view).
For students looking for law-review comment topics, it might be interesting to compare and contrast the “original expression v. facts” requirement of Feist with this “enough more than a natural law” formulation in Prometheus since, although their statutory bases are different, their constitutional origin is the same… and Feist explicitly is a constitutional decision.
The solution is simple. No patents. Patents are a failed experiment. They don’t protect inventors; inventors are either forced to sell them or destroyed by people with more money for lawyers. They don’t advance the useful arts. Telephones didn’t become commonplace until after Bell’s patents ran out. Internet security was virtually nonexistent as long as the RSA patents were in force, and it has yet to recover. Patents do nothing but make unscrupulous trolls and their lawyers rich at the expense of the people who actually do the work. Patents are bad. They are wrong. They are immoral. Patents must be eliminated.
Interesting, but I think that you, and the Supreme Court have missed an important point. I’d missed it too, until I started writing up this case, and I ran some numbers. Did you know that the average earning per patent is only $575.00?
Seriously. If you want to check my numbers, you can download them in a spreadsheet from the article from this link:
So what good is the Patent system?