KSR International v. Teleflex (On Petition for Certiorari)
Like that little blue engine
, KSR’s case continues to move forward despite significant obstacles. Now, the Solicitor General has filed its brief on behalf of the U.S. government supporting KSR’s petition for certiorari.
This case takes a direct assault on the Federal Circuit’s longstanding obviousness jurisprudence that requires some explicit reason as to why one of ordinary skill in the art would have known to combine two or more references to solve a particular problem. The question presented reads as follows:
Whether a claimed invention can be “obvious,” and therefore unpatentable under 35 U.S.C. 103(a), without proof of some “teaching, suggestion, or motivation” to modify or combine the prior art in the manner claimed.
The statute in question, 35 U.S.C. 103(a) is not so specific. Rather, the statute requires that a patent be denied if
the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art.
35 U.S.C. 103(a). The Solicitor General’s brief, filed last week, sees the disjunct between the CAFC rule, Section 103(a), and prior Supreme Court decisions such as Graham. In a blunt rebuke, the SG argues that “the Federal Circuit’s teaching-suggestion-motivation test extends patent protection to non-innovative combinations of familiar elements.” Rather, the SG argues for a return to the broad factors outlined in Graham for determining the level of one of ordinary skill in the art: “the scope and content of the prior art, the differences between the prior art and the claims at issue, and the level of ordinary skill in the pertinent art.”
The SG’s support for certiorari is rare in patent cases and is often the most cherished of a petitioner’s amici.
Notes: Of course, most inventions involve “a combination of old elements” where each element is found somewhere in the prior art. (In re Kahn). After the teaching-suggestion-motivation test is gone, what standard will a patent examiner use to decide whether a new combination is non-obvious? To be sure, the current test is formalistic and has been both under– and over-inclusive. However, has anyone proposed a better test — one that can be carried out by the PTO even in light of the high-turnover and young workforce in the agency’s examining corps? A patent attorney e-mailed his fear that a less formal test will give examiner’s permission to make the dreaded “because I say so” conclusory argument.
The EPO also has a rather formalistic approach known as the "problem-solution" test. That test also has its detractors, but is seen by many as a superior approach. In the UK, the “Windsurfing” test takes its name from the case of Windsurfing Int’l v. Tabur Marine. In that case, the court looked to whether, when viewed without knowledge of the alleged invention, the differences between what is known to a “skilled but unimaginative technician” and the alleged invention constitute steps that would have been obvious to the technician or whether they require any degree of invention. The hypothetical technician does not consider the commercial consequences which might follow if the step or process in question were found in practice to achieve or assist the objective which he had in view.
Another e-mail that I received focused on the low-tech nature of this case.
Like sectional couches with remote controls in the arm rest (recall that nightmare?), the technology [here] is so simple that everyone believes they understand it once it is laid out for them. The technology most [patent attorneys] deal with on a day to day basis is not this simple - drugs, catalysts, biotechnology, automation, software. The risk of a result-driven declaration of the law on obviousness is keeping me awake at night.
These comments at least have anecdotal weight — in the recent eBay oral argument, Chief Justice Roberts was heard to quip that even he could have invented invention in question (or at least created the software flow-charts).
Documents:
- Party Briefs
- KSR’s Petition for Cert (Including the CAFC and District Court decisions in Appx. A & B);
- Teleflex’s Brief in Opposition
- KSR Reply Brief (314 KB)
- Amicus Briefs
Links:
Mr. Peterson,
I have written many responses where I cite the exact mechanism you describe. In other words, I say in my response, “…Applicant notes that the Examiner appears to have found Smith and Jones by way of a keyword search. While applicant acknowledges the utility of the keyword search, is cannot act as a substitute for careful consideration of the individual references for proper motivation and teachings. Without such consideration, a presumption can arise, almost by definition of the act of keyword searching, that the Examiner used improper hindsight reconstruction and piecemeal application of the prior art in an attempt to arrive at the claimed invention.”
Fortunately, usually only novice Examiners fall into this trap.
Also, when Examiners provide the so-called lip service motivation, e.g. it would have been obvious to combine Smith and Jones because the world would be a better place, you can easily challenge such boilerplate assertions. The motivation has to: 1) appear in the references themselves; or 2) be in the knowledge generally available in the art. If the Examiner is relying on 2), then without an explicit teaching in the reference, the Examiner at some point must provide art, or be willing to execute an Affadavit supporting his contention that the motivation would have been within the general knowledge in the art. Few Examiners are willing to execute Affadavits and therefore, if they can’t find art to support thier lip service motivation, they must fall back on the references themselves. If they can’t find a motivation there, and I often try to foil that by finding a contrary teaching in one of the references, then the lack of motivtion argument has huge teeth.
In other words, don’t let a little lip service ruin a perfectly good day.
I have a pending patent application where an examiner has introduced prior art that he claims is reasonably pertinent.
if one skilled in the art is a perfume bottle designer, is it reasonably pertinent that he would be aware of what is going on in fish tank design? I fail to see how he can introduce this non-analogous art on the ground it’s reasonably pertinent.
with the introduction of patent databases that are keyword searchable, an examiner can now find read: cherry pick) elements of inventions based on his keyword searches, and then claim that it would be obvious to one skilled in the art to have combined elements as might be described in a new patent application. this is hindsight nonsense, where once you know how to combine the elements from non-analogous fields then it would be obvious to combine them in the manner described in the new application.
is this a legitmate way to introduce prior art? doesn’t the examiner have to look at how one skilled in the art actually works? a perfume bottle designer does not look to fish tank design to solve his problems. so why should an examiner be allowed to introduce prior art that is non-analogous by arguing his keyword searches have turned up “elements” of the new patent application and then state it would have been obvious to combine them.
How did patent examiners “find” prior art before the introduction of databases that are keyword searchable? If one skilled in the art “at the time of the invention” is relevant, then isn’t it reasonable to impose a restriction on the examiner that he can’t keyword search a database for prior art that is then used to suggest obviousness? after all, the person skilled in the art didn’t have the ability to search for elements in a database to “solve his problems.” Plus I would argue this is NOT the way one skilled in the art, especially for the period prior to databases, works. Doesn’t the examiner have to respect HOW one skilled in the art actually works when looking for prior art that is reasonably pertinent?
How can a patent examiner disallow your initial filing date? What precendent is there for this in the patent office – never heard of this before?
***If a rigorous theory of inventions is ever developed it may be found that no matter what the standard used there will always be some inventions that are obvious but for which the obviousness is impossible to prove.***
I agree that there will always be applications that look obvious, but cannot be proven obvious. We can never truly put ourselves in the shoes of the hypothetical one skilled in the art sitting in a room having intimate knowledge of all the relevant art that happens to be posted on the walls of the room.
I think the USPTO, which is supporting the Appeal to the Supreme Court, is (i) trying to fix a problem that will never be completely fixed and (ii) fixing the problem in the wrong way.
The “bad” issued patents that people laugh about and point to as being a result of these “strict” obviousness requirements set forth by the Federal Circuit are few and far between. However, for every “bad” issued patent that I’ve heard about, I personally know of an application that is languishing in prosecution (or has been abandoned) based poor obviousness rejections (i.e., rejections that do not reflect the current position of the Federal Circuit), which means that are a lot more patent applications out there that should be allowed, but are not, than there are of patents that should not have been issued, but have. If the Supreme Court comes down with looser obviousness requirements (comparable to what I already see in the Patent Office), there will be a lot more patents that I believe that should have been issued (because they really are inventive, in even the most strict sense of the word) but won’t be issued.
As I noted before, I think the real solution to the problem is to have better searches and better examinations. With better searches comes better prior art and nothing works better in rejecting a claim than good prior art. Also, making reexamination proceedings more accessible will also help weed out “bad” patent applications.
The more I think about it, the more I believe that no matter what way the Supreme Court comes down, the USPTO’s backlog will decrease because there will be a lot more certainty about the law.
That is all for now … I hope this discussion continues.
**If the combination in this case was so obvious, why couldn’t the accused infringer find a reference that provides a teaching, suggestion or motivation to combine? The time constraints of patent examination at the USPTO are one thing, but in most federal district courts in this country patent cases proceed so slowly that the accused infringer has sufficient time to scour the earth for the best prior art. I understand that for simple technology, patent documents may not be the best source for a teaching, suggestion or motivation to combine known elements. But it seems to me that for a truly obvious combination, SOME publication, e.g., a technicians manual or handbook, should provide such a teaching, suggestion or motivation.**
I completely agree with your statement.
I see the Solicitor General’s (i.e., the USPTO’s) support for overturning the Federal Circuit’s approach as just another way that the USPTO is trying to reduce their backlog without hiring additional Examiners.
From suggesting that Applicants do their own searches, to trying to limit continuations, the continual reapplication of 101 rejections to business-method-type inventions, and now to this support for a more flexible approach with 103 rejections, the Patent Office is trying to reinvent itself as the Rejection Office.
To me, the cure for the Patent Office’s backlog (and the “bad” applications that get issued), involves real training, better candidates (which will require better pay), better supervision, and more time for individual Examiners to search and examine applications. This could all happen if the all the funds being paid by Applicants are left in the patent system, which they are not since a good portion of the funds are currently being diverted.
I also noted the following comments made by Ed Lee in his blog (which he referred to earlier, so is fair game in my mind) that appear to be relevant to a point you made:
**My thoughts: The Federal Circuit approach effectively guts the nonobviousness requirement. Patent examiners (who are paid by how many patent applications they process) have no time or incentive to try to find a prior art reference that suggests a “motivation to combine” other prior art references. The Federal Circuit approach places too much stock in documentary references — prior art can exist in general knowledge and know-how, without being specifically written down. Some things are just so obvious — the fact that a wooden door knob can be also made of ceramic, metal, or other substance — that it’s not worth writing down.**
I just don’t understand how an administrative agency (i.e., the USPTO) and the Courts can review any decision based upon “general knowledge and know-how” if that alleged general knowledge and know-how is not written down. The ultimate determination of obviousness is based on the entire record, by a preponderance of the evidence, with due consideration to the persuasiveness of any arguments and any secondary evidence. In re Oetiker, 977 F.2d 1443, 24 USPQ2d 1443 (Fed. Cir. 1992). The legal standard of “a preponderance of the evidence” requires the evidence to be more convincing than the evidence which is offered in opposition to it. M.P.E.P. § 2142. However, if the “evidence” is “general knowledge and know-how” that resides in someone’s noggin (instead of in a published document), how can a court weigh that evidence (i) if the court doesn’t know if the evidence actually exists or (ii) which if existed, the exact nature of which is unknown.
I analogize this to Examiner taking “official notice” of a fact. As stated M.P.E.P. § 2144.03, the Examiner can take official notice of facts outside of the record, however, these facts must be of such a notorious character so as to be instantly and unquestionably recognized as “well-known” in the art. I always challenge the Examiner’s assertion … if the facts are so “well-known,” as alleged, the finding them should be easy. If the facts cannot be found (which is almost always the case), then perhaps the facts are not as “well-known” as the Examiner thought. I believe the same logic should apply to obviousness rejections.
**Great commentary. I’ve posted several on the case on my blog here: link to elee.cc**
I went to that blog, and noted the following comment:
**Now I believe the Supreme Court will substantially reverse the Federal Circuit’s approach and will hold that obviousness can be shown in numerous ways and does not require any “teaching, suggestion or motivation to combine” prior art. It must be decided on a case by case basis, and no factor is dispositive. Even more, the patent office can rely on the expertise of the patent examiner in making such a determination that an invention is obvious based on the prior art.**
If the first two sentences come to pass, I will be very nervous. If the last sentence comes to pass, I may look for a new profession.
I have thought of perhaps a dozen ways I could approach this issue, but I’m going to settle on this one … what exactly is the “expertise” of an Examiner? The “expertise” of an Examiner is very subjective and varies considerably from one Examiner to the next.
To understand the currently imposed requirements for “objective evidence” for a motivation to modify/combine (i.e., teaching, suggestion or motivation), you have to understand why the requirements are imposed. If a rejection is to be reviewed by the Board or by the Federal Circuit, how will the Board/Court review the record? Will the Examiner be required to be subjected to a deposition?
My favorite line of cases dealing with the objective evidence requirement is that of In re Zurko and In re Lee, which cites Zurko. Zurko discussed the Administrative Procedure Act (APA) vis-à-vis PTO fact-findings. The APA states that a Court can “(2) hold unlawful and set aside agency action, findings, and conclusions found to be … (A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law; … (E) unsupported by substantial evidence in a case subject to sections 556 and 557 of this title.”
In my mind, the “teaching, suggestion or motivation” test simply requires that the USPTO provide “substantial evidence” for a finding of obviousness. Otherwise, if the PTO was allowed to use subjective beliefs (e.g., the “expertise” of an Examiner), how could a court prevent an Examiner (who was irritated an Applicant or Applicant’s representative) from “arbitrarily and capriciously” denying a patent.
The USPTO is not much different than the FCC or FDA, whose decisions (positive or negative), have a financial impact of millions or billions of dollars. With the FCC and FDA, their decisions are sometimes accompanied my hundreds of pages of supporting documentation and frequently reviewed by the courts. Similarly, the decision to grant (or deny) a patent may also be worth millions and challenged in court. That being said, why should the USPTO be able to get away with decisions based, not upon substantial evidence, but instead on subjective beliefs? Such an implementation would be ripe for abuse.
**I have noticed that the Board of Appeals carefully considers the teaching-suggestion-motivation test, but examiners too often give mere lip service to the test. Abolishing the teaching-suggestion-motivation test will primarily affect those few applications that are appealed.**
I agree. However, the PTO is encouraging the filing of more Appeals, which makes this issue more relevant. Still, why should an Applicant be forced to file an Appeal to overcome a 103 rejection that is without merit? This raises another issue, which I’ll discuss at some other time, which is that of the training of Examiners.
**Aside from its effect of the prosecution process, I wonder how abolishing the teaching-suggestion-motivation test would affect the business of patent law. Abolishing the teaching-suggestion-motivation test will surely weaken patents–causing more to be invalidated in litigation. Would this weakened patent system lessen clients’ desires to obtain patents? Would this be bad for business for patent prosecutors?**
You raise a very, very interesting issue. If the Supreme Court weakens the test for obviousness, I would very seriously start thinking about transitioning my practice to focus on invalidating patents instead of obtaining patents.
I am generalizing now … but IMHO, without some “objective evidence” standard to reign in Examiners, I cannot see how any determined Examiner cannot come up with some reasoned (yet subjective) analysis as to why a particular invention is obvious. Unless an invention includes some completely novel feature, then it won’t be hard to combine references (showing the old parts) and assert that the claimed invention is obvious.
**When I was an Examiner I always favored the new and unexpected results test when considering obviousness. It seems unfair to require a Patent Examiner to provide an explicit motivation in cases when the applicant simply combines known elements in a common art but fails to provide a rational why the particular combination they are attempting to patent provides something greater than the sum of the parts.**
I am not in favor of this “new and unexpected results” test. One way to “invention” is through the recognition of the underlying problem, and once the problem has been recognized, how to solve the problem is easy, and may include well known parts that, in combination, produce expected results. In this regard, see M.P.E.P. § 2141.02(III) and In re Sponnoble.
A “new and unexpected results” test would consider how the invention was conceived (e.g., those inventions based upon recognition of the underlying problem would not be considered patentable), which would seem to me contrary to the portion of 35 U.S.C. § 103, which states that “[p]atentability shall not be negatived by the manner in which the invention was made.”
**Judge Howard T Markey who died this May famously said that only God creates new things and that man must work with known elements. As he rightly says, combinations of known elements should be treated with respect.**
You raise a point that I often try to make during prosecution using the following language:
As a general matter, “virtually all [inventions] are combinations of old elements.” In re Rouffet, 149 F.3d 1350, 47 USPQ2d 1453 (Fed. Cir. 1998) (quoting Environmental Designs, Ltd. v. Union Oil, 713 F.2d 693, 218 USPQ 865 (Fed. Cir. 1993)).
I am sorry that I am late to this discussion. As someone whose practice is mostly patent prosecution, I see this case as having far reaching consquences to my practice. In the following posts, I want to provide responses to some of the comments made by the previous posters. In many instances, I am going to generalize (for sake of brevity), but I would be more than happy to elaborate if so desired.
The growing difference between US and Euro patent practice is that the patent application in Europe is intended to not onlt describe the machine, but to define the outer limits of the aruments for patentability, or as some have called it, the “template”.
The US courts have expressly rejected this. You can file a description of a machine* together with a clearly way-broad claim, and then develop your claims and discover your inventions as prosecution goes along.
I suspect we take this approach because we want to award inventors for all the new and useful things they come up with and disclose even if they have no idea what is new about them at the time they filed their applications.
This is changing, I suspect, and that is the reason for the PTO’s new restrictive attitude toward continuing applications. Rather than discovering on the fly, we are forced to research in advance and make our most compelling case right in the original filing.
*Motorola was filing applications that were nothing more than detailed descriptions and claims for a while there. I don’t know if they are still at it or not.
Patent attorneys are supposed to be experts in the meaning of words, but sometimes I have the feeling about the US patent law system that it puts a premium on obfuscation.
Let me assure you: to quote the schmaltz rock band Boston, your feeling is more than a feeling. It is fact carved in stone and perhaps the most important fact for patent prosecutors in the US to bear in mind when drafting applications.
Great comment SF. I have trouble explaining to US clients why the EPO is so strict on amendments that “add subject matter” to the application. I say: think of it like an ultra-strict requirement for “written description”. In our “first to file” landscape such strictness is needed, if only to be fair to the guy who came second to the Patent Office but was first with an enabling and sufficient disclosure. Watch the CONOR multi-national litigation in Europe, on their TAXOL-eluting stent.
MaxDrei stated, “Never say what the invention is, seems to be the watchword in USA these days. That is a bit tough for Europeans like me to understand.”
In my experience, the EPO requires explicit statements identifying the invention. Consequently, in order to amend the claims in the EPO, you almost always need word-for-word language in the specification.
In constrast, word-for-word language is almost never necessary for amendments in the United States. In fact, US applicants can almost always amend their claims by arbitrarily broadening elements and omitting elements–even if the specification does not explicitly recite the broader terms and even if the specification does not state the omitted element is optional. All that seems to matter is whether the amended claim “reads on” the disclosure of the patent application.
Consequently, in US practice, explict statements of what “the invention” includes (or does not include) serve no useful purpose.
In fact, statements about “the invention” may restrict the applicant’s ability to broaden or omit elements.
Worse, statements about “the invention” may unintentionally and unexpectedly narrow the scope of their issued claims in litigation. For example, in Inpro II Licensing, S.A.R.L. v. T-Mobile USA, Inc., the Federal Circuit recently construed a broad term “host interface” to mean a narrower term “parallel host interface” in part because of the applicant’s statements about “the invention.”
In US practice, never saying what “the invention” includes is understandable when viewed in the context of US law. What might not be understandable to Europeans is why the US does not have stricter requirements for written description support.
Tom, I assumed that the POSITA is in the cattle business (BTW, not a rancher who knows his own beasts , but a more remote person, interested in marking cattle) because I posited a claim which starts from that technical field. Now let us assume that the inventor wrote a claim such as:
Bar code marking method, characterised in that: it’s on cattle.
Well then the POSITA is in the bar code field and the question is whether it would have been obvious to apply a barcode to cattle. As I wrote, the inventor defines his own template.
When I wrote last, I did not know whether, in your scenario, bar codes were a recent invention, or common general knowledge for “years and years”, and when “so much evidence goes the other way” (evidence that was in your imagination but unknown to me). As I wrote, when you and I have completely different visions of the prior art landscape for your hypothetical example, it is difficult to debate it meaningfully.
You write: why imagine that a bar code person would apply it to cattle? For this reason. The spec says that the tech effect of applying it to cattle is the usual tech effect of bar codes, namely, more readability and more info. So, when the skilled bar code person is asked, on the prio date, how could one improve readability and info content on cattle markings, is there in the art at that date a hint or suggestion to achieve more readability and info content by replacing an Arabic number string cattle ear tag with a barcode. I guess that the art would say YES (but maybe you are going now to put me straight on that too).
I am sorry you continue to view P&S sceptically, but you are in good company. Patents courts all over Europe do so too. As I wrote, I put it down to the “Not Invented Here” syndrome.
You may be right, that P&S is not “grounded in the psychology and methodology of creation and invention”. As I wrote, it is admittedly artificial. But is that a sufficient reason for dismissing it. Plenty have tried in Europe over the last 25 years. But it just goes on defying them, and getting the business done, fairly. I have an idea. Ask the big US corporate users of the EPO, who in the EPO oppose the patents of their competitors, as well as prosecuting their own EPO cases. How do they find it 1) as EPO Applicants and 2) as EPO Opponents.
As I wrote, it is a framework for a focussed debate on obviousness, which gets to a fair result in short order.
Finally, I don’t understand your remark about US innovators and European patent killers. If Europe is against patenting business methods, well it is against such claims regardless where they originate. It reminds me of the US client who was convinced that his EPO patent was doomed, just because the three members of the Opposition Division had “German” names and the Opponent company was resident in Germany. The Opponent had a good case but the opposition failed. Where I work, the treatment meted out to inventors is nationality-neutral, and I am sorry you ever suspected it might be otherwise. Dare I say it: it takes one to know one?
A “snark” of my own. I hate the use of “typically” to imply “predominantly”. Patent attorneys are supposed to be experts in the meaning of words, but sometimes I have the feeling about the US patent law system that it puts a premium on obfuscation. For example, you used to have in your specifications a section headed by the sub-title “Summary of the Invention” but now youngsters are taught to write instead “Summary”. Never say what the invention is, seems to be the watchword in USA these days. That is a bit tough for Europeans like me to understand.
MaxDrei,
Thank you for your comments. I wonder, though, if they really get at the heart of my objection.
You’ve spelled out a particular way in which the “problem” might be articulated for the cattle example, and particular assumed prior art, and a particular POSITA.
But all the issues I raised, at least hypothetically, remain.
You assume that the correct POSITA is, say, a cattle rancher. BUt why would this be the correct field from which to draw the POSITA? What if the inventor is himself in the bar code field, and indeed describes an invention with all the details known to such a person about how logistical systems can enrich a problem such as presented by cattle ranching, once that problem becomes present to his mind? Why does it make sense to settle on the POSITA as a typical cattle rancher?
And, while you argue that the patent examiner might declare that the cattle rancher POSITA would find the use of bar codes obvious, because of his general exposure to the concept of bar codes, isn’t it quite possible that there would be powerful countervailing evidence against this — such as, for example, the (hypothetical) fact that cattle ranchers have been aware for years and years of bar codes, yet not a single mention of using them on cattle is to be found anywhere, despite their clear viability for all that time? In short, aren’t you reduced to exactly the sorts of considerations used in the US to determine obviousness? If you just leave it up to the examiner to dictate that something like this is obvious when so much evidence goes the other way, isn’t that performing hindsight in the worst possible way?
I guess I would have thought that you might come to the example from the opposite direction, declaring that the designated POSITA would be the bar code person, and that the cattle coding scheme would be just a routine application of the general logistical schemes well known in the bar code industry. Of course, then the question would remain: why imagine that such a person would ever think to apply those schemes to the cattle industry, and why imagine that it’s fair to frame the whole analysis under the assumption that that is the very problem he has to solve?
Honestly, the more I look at the “problem-solution” approach, the more it strikes me that it’s just a conceptual straitjacket, either doing no real work at all, or getting the exactly wrong result.
What I certainly don’t see is how it is in any way grounded in the psychology and methodology of creation and invention. Virtually all studies of creation and invention see as an absolutely key feature of creation that it involves the juxtaposition of two frameworks previously regarded as separate. Sometimes, this is called lateral thinking, bisociation, or divergent thinking. Where is there in the “problem-solution” framework any appreciation for this basic fact of how innovation actually occurs in real people? If the framework can’t acknowledge that there are many, many instances in which people failed to see for long periods of time connections between different kinds of things, even if in hindsight they seem “obvious”, then of what use is the framework, except as a device to deny innovators the fruits of their invention? (And, allowing me a little snark, how surprising would it be if Europeans are happy to do that, when the innovators are typically American?)
Tom, here’s my response to your invitation. I think yr secenario gives me an opportunity to show how the crucial EPO insight “Obvious, for what?” can de-mystify the obviousness issue. But, first, some caveats:
1. I am just an ordinary coal face worker, no intellectual or academic, so maybe (we shall soon see) I am getting in above my head, and not reading the game six moves ahead;
2. P&S pre-grant is often a bit lenient on Applicant, with inter partes proceedings (like EPO opposition) needed to bring it to full balance between opposed parties. That’s better though than a system which, in pre-grant exam, weeds out some wheat along with the chaff;
3. We are not talking mathematical proofs here. If 99% of cases are dealt with efficiently and fairly using P&S, it is no fatal blow to the usefulness of P&S if 1% of cases are awkward;
4. When you mentioned cattle marking I was reminded of the original EPO MPEP from 25 years ago, which used “Freeze branding” as an example of a so-called “Problem” invention, in which perception of the problem is the non-obvious step, all subsequent steps being obvious. But your bar code scenario is easier, and not a problem invention (as far as I can see).
Your supposed invention is something like:
“Method of marking cattle, characterized in that: it’s a barcode.”
So, the specn presumably announces the tech field as that of marking cattle, and mentions as prior art the ubiquitous ear tag with a string of Arabic numbers, and explains that a barcode enhances information content and readability of the marking.
OK. We take the inventor at his word. He wrote the spec, starting with a white sheet. We give him full faith and credit. We start (D1) with the numbered ear tag. We postulate the skilled person as somebody in the cattle business,who has a motivation to mark cattle. That person on the priority date of the claim is deemed to have an interest in enhancing readability (and info content) of the markings. P&S deems him to be looking for ways to enhance readability (here’s the “for what” in “obvious for what”). When that person reviews the art for ways to make a marking more readable, does the art give him a hint towards barcodes? Yes, and the claim is obvious.
If I were the Exr, I would cite the prior art use of barcodes on blood samples and bunches of bananas, picked and weighed self-service at a supermarket. Why do cattle have to be marked? Not in order that the little hill farmer can identify his own cow (he does that by naming her Blossom). More likely by a Government Inspector aiming to control BSE, or an agent from Archer-Daniels or Cargill or McDonalds aiming to stop fraud. Such people do indeed have an interest in enhanced readability and info content.
If I were EPO representative, instructed by you, I would then be arguing that, specifically in the cattle business, there was no perception of any need to increase the info content of the numbered ear tag, or its readability and, even if there were such a perception, the correct skilled addressee would not know how to advance to a barcode.
The Examiner might then reply that a farmer is the wrong addressee, and even people in the connected with the cattle raising business know how to buy a bunch of bananas self-service, or have seen barcodes on TV. I can visualise having a tough time getting your case through the EPO.
Of course, the trouble with examples like yours is that you and I have different visions of what is in the prior art. One advantage of P&S for a multi-lingual EPO, where most everybody is using English as a foreign language, is that we can decide obviousness on the prior published docs alone, without needing witness testimony.
Tom, the point about P&S is not that it delivers a result ergo like a mathematical proof. Rather, it sets up rules for the debate whether something is obvious or not. With P&S, the debate is often pentratingly quick, and brutal in puncturing clouds of obfuscation. That’s why, in EPO circles, it has so many fans.
I think that the teaching, suggestion, motivation test is fine ONCE you are satisfied that you are dealing with a true combination of features and not with a mere collocation.
As regards collocations you cannot do much better than two old cases of the Supreme Court. In Pickering v McCulloch the Supreme Court said:
“In a patentable combination of old elements, all the constituents must so enter into it so that each qualifies every other: to draw an illustration from another branch of the law, they must be joint tenants of the domain of the invention, seized of every part per my et per tout, and not mere tenants in common with separate interests and estates. It must form either a new machine of a distinct character and function or produce a result due to the joint and cooperating action of all the elements, and which is not the mere adding together of the separate contributions. Otherwise it is only a mechanical juxtaposition and not a vital union.”
The same point was made positively in Carnegie Steel v Cambria Iron Company where the US Supreme Court affirmed that:
“It may be laid down as a general rule, though perhaps not an invariable one, that if a new combination and arrangement of known elements produce a new and beneficial result, never attained before, it is evidence of invention.”
Before the EPO and the UK courts, if you can point to such a result, your application is likely to be granted and your granted application is likely to be upheld, whereas if there is no such result the converse applies. The new function or result which we should look for is the basis for defining technical problem, and the methods of technical analysis which we use in Europe once mastered are so simple and complling that it is a shame that they are not better known and applied in the US.
MaxDrei,
OK, let’s just use a simple example so I can see how the problem-solution framework is supposed to work.
Suppose that we’re at a time the development of bar codes in which bar codes have been broadly applied to logistics and retail, but little beyond that. Now suppose some inventor realizes that it would be enormously valuable to bar code cattle, and treat them as a logistical system. Quite possibly, this application would introduce huge efficiencies into the cattle business.
Now here’s my puzzle. How do we describe the objective “problem” here? What is the correct characterization of the “technical effect” you talk about? And to whom do we pose this “problem”?
If we say the problem is to organize the cattle business in certain ways, and treat cattle as presenting as a logistical problem, and we pose the problem to a POSITA in the bar code industry, the answer might be extremely obvious ONCE the problem is so posed: he could rattle off where and how the cattle should be scanned, and simply enter the data into pre-existing data structures generically useful for logistics. But the thing is, there would be absolutely NO reason to believe that such a person might make the leap of inference that those techniques might apply to something like cattle or the cattle business. Yet simply by stipulating from the outset that the problem is to introduce certain kinds of efficiencies in the cattle industry, we have given away the game entirely as to what needs to be done. The very leap of intuition required to make the connection is assumed as given in the framing of the problem.
On the other hand, if we pose the same question to someone in the cattle business, likely they wouldn’t come up with a bar coding scheme, because they are never thinking about such things, and wouldn’t in any case know how to implement them. Again, the core of the invention is the very insight that the techniques of one discipline, bar coding, might be very useful in a very different discipline, the business of raising cattle.
So in the “problem-solution” technique, what WOULD be the correct “problem”, or “technical effect” here? To whom would it be posed? Would this application turn out to be obvious or not?
Tom, you are sceptical that EPO P&S can handle a cross-disciplinary invention. I wonder how much practical experience of EPO P&S you have. If you go to “Fire of Genius” Comment #8 in the item “New Obviousness Standard”, you can read how P&S is used in practice, and how it is equally applicable for inventor and for the attacker of validity. It may be artificial to start from a single prior art reference but it sure is a powerful way to set up a fair debate, free of hindsight, whether the claimed subject matter is or is not obvious. Of course you are right: when the claimed subject matter is cross-disciplinary, we have a fierce debate whether the advocated prior art starting point is “realistic” and who should be the skilled person. In such cases, the inventor would have good chances to survive. But, as I tried to explain in “Fire”, the objective problem falls out of the analysis “automatically”. It is an inevitable result of 1) what the inventor told was his/her contribution to the art (technical effect) and 2) what technical effect the prior art starting point delivers. The one, minus the other, corresponds to the objective problem (to achieve that tech effect). This renders moot the subjective problem.
With respect to the obviousness argument, and the European “problem-solution” approach, as I understand it, I don’t see how this can possibly handle correctly some of the most important, and revolutionary inventions possible, namely what I would describe as “cross disciplinary” inventions. These are inventions in which the true insight is that two distinct disciplines can be combined to achieve a certain useful effect not earlier envisioned. Indeed, this may present issues even for the definition of obviousness under the “teaching, suggestion, motivation” rule, but less so.
In such a case, how does even frame the “problem” being solved, without giving away the core of the insight? It may very well be that simply articulating the problem to the POSITA in either, or at least one, discipline allows them to put together the solution. For example, it may be a problem understood in discipline A, but not in the other discipline B, but simply posing the problem to someone in discipline B will enable them to come to the right combination of elements.
And how does one choose an appropriate POSITA without giving away the core of the insight and invention? Is it a person of ordinary skill in one discipline or the other? Is it a hypothetical, but at the time non-existent, person in BOTH disciplines? I see no effective way of forcing this into a “problem-solution” framework without gutting the essence of the invention, and rendering “obvious” what may be in fact revolutionary and breathtaking.
On the other hand, the “teaching, suggestion, motivation” rule most certainly will grant that such an invention is indeed non-obvious, because it will be impossible in either discipline to find prior art that teaches, suggests, or motivates the combination involved. There remains the problem of choosing which discipline from which the POSITA may be drawn, and whether one tries to sneak in a hypothetical POSITA of both disciplines. Yet if the “teaching, suggestion, motivation” rule is enforced, the invention will still stand as non-obvious.
And of course the problem of the granularity of what constitutes distinct “disciplines” is a considerable one that remains. Again, the “teaching, suggestion, motivation” rule is the best way, perhaps the only way, to keep hindsight from creeping in.
In mathematics there is an idea called “Godel’s Incompleteness Theorem” which basically states that in any mathematical system based on arithmetic there are always theorems that are true yet unprovable. If a rigorous theory of inventions is ever developed it may be found that no matter what the standard used there will always be some inventions that are obvious but for which the obviousness is impossible to prove.
Chris:
For inventions in unpredictable arts such as biotechnology, I understand that formulating and obviousness defense (during litigation) or rejection (during prosecution) has added considerations. But the fact is that the patent at issue in the KSR case claims an electromechanical invention. Electromechanical arts are generally deemed predictable.
And we are talking about a litigation here, not a prosecution where an examiner had 12 or so hours to find the best prior art. The accused infringer almost certainly had at least several months, and likely much more time, to develop an invalidity defense. For much less than the cost of hiring an experienced Supreme Court advocate, the accused infringer could have hired the best document searchers that money can buy.
As to your example, ignoring unpredictablity considerations, the “teaching” reference need not teach the full combination X, Y and Z. (If it did, it would anticipate.) If the art already teaches X and Y and X and Z, a reference that teaches, for example, that two stabilizers with X works better should suffice. (Again, importantly, I am ignoring unpredictablity considerations because I am not a biotech guy and the KSR case does not involve an unpredictable art.)
So much information is published. Not all of it is about ground-breaking stuff. Most of it is about routine, everyday matters. There are all kinds of published manuals and handbooks that get into the minutae of known components, substances, etc. When I was in college, my civil engineering friends took a whole class devoted just to concrete.
Whether the Federal Circuit crossed the line by ignoring Supreme Court precedent is a separate and important issue, but it seems to me that it is not unfair to require an accused infringer to come forward with a teaching, suggestion or motivation to combine known elements in order to invalidate a patent claim as obvious.
The MPEP states, “The references must be viewed without the benefit of impermissible hindsight vision afforded by the claimed invention.” MPEP § 2141 at 2100-125.
JH states, “In my opinion the issue of hindsight is ludicrous. One cannot begin to formulate a rejection without first having knowledge of the invention.”
Of course, all rejections are formulated in hindsight in the sense that the examiner must first read the claim and then look back to the prior art. By referring to hindsight rejections, patent agents/lawyers aren’t suggesting that the examiner can’t look at the claim. That *would* be ludicrous, ridiculous, etc. In contrast, the term “hindsight rejection” is generally used as shorthand for rejections that presumably rely upon the benefit of hindsight because the examiner has cited no prior art teaching, suggestion or motivation for making the proposed modification or combination.
The MPEP explains that the examiner has the burden of demonstrating this teaching, suggestion or motivation: “To establish a prima facie case of obviousness, three basic criteria must be met. First, there must be some suggestion or motivation, either in the references themselves or in the knowledge generally available to one of ordinary skill in the art, to modify the reference or to combine reference teachings….” MPEP § 2142 at 2100-134.
Because it’s the examiner’s burden to identify the teaching, suggestion or motivation, patent agents/lawyers often (as JH noted) “only cite the popular cases, and claim there is no motivation, or that the rejection is a hindsight reconstruction.” This is entirely proper if the examiner has recited no teaching, suggestion or motivation. More often than not, however, examiners provide one or two sentences reciting a motivation of some sort to avoid the virtually automatic “no teaching, suggestion or motivation” response.
JH states, “The idea that providing motivation for the combination somehow prevents a hindsight reconstruction of the invention is ridiculous to me.”
On paper, the teaching-suggestion-motivation test should prevent the use of the benefit of hindsight because the examiner must actually cite some teaching, suggestion or motivation in the prior art. But since the teaching, suggestion, or motivation may be found in either the knowledge of one of ordinary skill or the references themselves, examiners routinely opt for the knowledge-of-one-of-ordinary-skill route. Because most any examiner can (and will) cite the knowledge of one of ordinary skill to provide a two-sentence motivation, abolishing the teaching-suggestion-motivation test probably will not make a huge difference in prosecution–as I mentioned above.
AP
“If the combination in this case was so obvious, why couldn’t the accused infringer find a reference that provides a teaching, suggestion or motivation to combine?”
The answer to that question is simple and gets right to the heart of the problem with the explicit teaching test: some combinations (a whole lot of things that I see, in fact) are so freaking obvious that people simply do not bother to write every one of them down explicitly.
For example, take a “solution for stabilizing protein X.” The art teaches the use of X, X and Y, Y and Z, X and Z, and the general utility of the genus of Y and Z.
So Mr. Inventor puts X, Y and Z in his solution and says “Look, nobody to told anyone to combine these things!” and of course it’s oh-so-unpredictable because it’s “biotechnology.”
Anyway, long answer to your question but this explains why crappy obvious patents issue all the time.
What I found suprising about the Solicitor’s Amicus brief is the absence of any mention of the additionally CAFC required “clear and convincing evidence” test and jury instruction to challenge 103 obviousness in an infringement suit, even for undisputed evidence of prior art in the form of a prior patent, as in this case. Is it not the combination of both CAFC added requirements, not just the the subject requirement, that affects the statistics?
If the combination in this case was so obvious, why couldn’t the accused infringer find a reference that provides a teaching, suggestion or motivation to combine? The time constraints of patent examination at the USPTO are one thing, but in most federal district courts in this country patent cases proceed so slowly that the accused infringer has sufficient time to scour the earth for the best prior art. I understand that for simple technology, patent documents may not be the best source for a teaching, suggestion or motivation to combine known elements. But it seems to me that for a truly obvious combination, SOME publication, e.g., a technicians manual or handbook, should provide such a teaching, suggestion or motivation.
KSR Comments
Patently-O discusses KSR and non-obviousness, such as the vexed question: If the Federal Circuit formulation of the standard is trashed, then what is the new standard? What does the Court tell the PTO to start doing? The Comments, which include…
Ed gives a great explanation. I suggest a reading of the writ and the government’s response if one wishes to understand the issues.
In my opinion the issue of hindsight is ludicrous. One cannot begin to formulate a rejection without first having knowledge of the invention. The idea that providing motivation for the combination somehow prevents a hindsight reconstruction of the invention is ridiculous to me.
I can state my opinion quite simply. Decisions on obviousness, which I agree with, which apply the teaching/motivation test, would also be decided identically if they relied solely on the Graham factors. Decisions which I do not agree with generally only can be supported by agreement or disagreement with the presence of proper motivation, a concept which is extremely vague and inconsistently applied.
I would like to mention that one other thing that the teaching/motivation test seeks to remedy, in my opinion, is enabling one to obtain a patent by combining only prior art concepts together, in the situation in which the idea to combine such concepts is novel in and of itself. However, I believe Graham can satisfy these situations with a consideration of “unexpected results”. Certainly if the combination produces unexpected results, derived from a combination of known elements, it should be novel.
Also, regarding Ed’s statement that examiner’s issue rejections with no disussion of the teaching motivation test, I believe that examiners do all that is required to establish a prima facie case for obviousness, and that it is only in a resopnse to arguments, or an examiner’s answer, that a detailed discussion of motivation should be applied. Also, applicant’s/attorneys rarely provide clear rebutals of obvious type rejections. They generally only cite the popular cases, and claim there is no motivation, or that the rejection is a hindsight reconstruction. This simply perpetuates the vague inconsistent decisions which arise from the requirement for motivation.
Dennis,
I suggest a new way of analyzing nonobviousness at link to thefireofgenius.com
Toward a new nonobviousness analysis
In the wake of the SGs brief supporting review in KSR, Patently-Os Dennis Crouch poses the core questions:
After the teaching-suggestion-motivation test is gone, what standard will a patent examiner use to decide whether a new combination…
The EPO obviousness test is interesting because it is out of a Patent Office you could call itinerant, lacking a national Supreme Court to impose Precedent on it. Its equal rank Technical Boards of Appeal compete for the most persuasive line of obviousness logic and the fittest survived. Meanwhile, for all national patents courts in Europe, P&S is “Not Invented Here”. Few outside EPO prosecution circles bother to try to understand it, but using it is like riding a bike – impossible till you have learned how, then ridiculously easy. So much so that you wonder how you ever had a problem with it.
The wonder of it is that it is quick, clean, elegant and objective, all at once. Bad news for litigation lawyers, I guess. But no wonder it is now unassailable and eponymous within the EPO, including at Board of Appeal level in mega oppositions such as with blockbuster drugs, where the patent for the whole of Europe gets revoked at one “oral proceedings” session.
I suspect that P&S has got itself a bad name in USA, but I am curious to know what the criticisms are exactly, and whether they are well-informed. Within Europe, EPO P&S has seen off its detractors, but it took 25 years to do it.
I think the answer to what a new test should be is simple. Add the following to the existing statute:
Any combination of elements that performs no function beyond the functions, implicit or explicit, of the individual elements is presumed obvious.
This would prevent the granting of patents to combinations that previously were patentable only because, based on evidence, the examiner cannot mount a prima facie case against the combination. The presumption above can be overcome by sufficient proof of secondary considerations.
db:
You are correct that courts use the Graham analysis in litigation.
However, the courts must also use teaching-suggestion-motivation test in addition to the Graham analysis. Because many inventions may look obvious in hindsight, the Federal Circuit has repeatedly held that the prior art must provide some teaching, suggestion or motivation to make the allegedly obvious alteration/modification. Thus, while the Graham analysis is necessary, it is not entirely sufficient under current Federal Circuit law.
The government’s position is that the additional teaching-suggestion-motivation test is unnecessary and that the Graham analysis is sufficient.
You’re right that, if teaching-suggestion-motivation test is abolished, the examiners would have more power. Currently, examiners are subject to both Graham and the teaching-suggestion-motivation test. See MPEP 2141, 2143.01. I think, however, it may not make a huge difference in prosecution since examiners currently issue rejections with little or no discussion of the teaching-suggestion-motivation test.
Great commentary. I’ve posted several on the case on my blog here: link to elee.cc.
I think the Supreme Court will grant cert.
(Sorry, I wanted to do a trackback, but my feature is not working.)
rf-
IANAL (I’m an agent), but it’s my understanding that the Graham factors are applied by the courts when the patent is litigated and obviousness is raised, so this change will only effect the prosecution of the patent.
However, litigation is a much better venue for application of the Graham factors, allowing for experts, cross, etc. I think the above comments are accurate that Examiners would have much more power if the Court held that Graham should be applied during prosecution.
-db
Patents for “obvious” inventions
There’s an interesting patent case, KSR International v. Teleflex, bubbling up through the US courts. It’s been reviewed by the Court of Appeals and KSR has petitioned (90 page PDF) the Supreme Court for a writ of certiorari, i.e. for
I have noticed that the Board of Appeals carefully considers the teaching-suggestion-motivation test, but examiners too often give mere lip service to the test. Abolishing the teaching-suggestion-motivation test will primarily affect those few applications that are appealed.
Aside from its effect of the prosecution process, I wonder how abolishing the teaching-suggestion-motivation test would affect the business of patent law. Abolishing the teaching-suggestion-motivation test will surely weaken patents–causing more to be invalidated in litigation. Would this weakened patent system lessen clients’ desires to obtain patents? Would this be bad for business for patent prosecutors?
Judge Howard T Markey who died this May famously said that only God creates new things and that man must work with known elements. As he rightly says, combinations of known elements should be treated with respect.
When I was an Examiner I always favored the new and unexpected results test when considering obviousness. It seems unfair to require a Patent Examiner to provide an explicit motivation in cases when the applicant simply combines known elements in a common art but fails to provide a rational why the particular combination they are attempting to patent provides something greater than the sum of the parts.
As an examiner, making an obviousness determination is one of the hardest parts of the job, especially in the electrical arts. We try to be fair but given the choice between a bad allowance and an awkward obviousness type rejection, the awkward obviousness rejection is the option often taken. Office Actions are randomly chosen for review and an examiner can get a black mark for a bad allowance OR a bad rejection so sometimes it’s like being between a rock and a hard place.