by Dennis Crouch
In 1981, the U.S. Court of Customs and Patent Appeals (CCPA) issued its decision in In re Keller, 642 F.2d 413 (CCPA 1981), a case that continues to shape obviousness determinations over four decades later. As the predecessor court to the Federal Circuit, the CCPA’s rulings remain precedential and influential, with Keller cited 10,000+ times by the USPTO’s Patent Trial and Appeal Board (PTAB) and its predecessor Board, the BPAI.
In the typical USPTO vs Applicant negotiation, Keller favors the USPTO and is often cited to uphold examiner rejections. Although the case remains influential, it does have some fundamental problems. Notably, it creates a gap — what is not sufficient evidence — and then fills that gap with the amorphous language of “suggestion.” But this gap places the case in the good company of KSR v. Teleflex that similarly calls for flexibility in finding ways to judge obviousness.
Background: The patented invention in Keller related to a cardiac pacer with a digital counter driven by a high-frequency oscillator. Patent No. 3,557,796. At the time, analog radio-controlled pacers were common. And, in fact, Keller had previously patented an analog version for his employer Cordis. Patent No. 3,253,596.
The digital pacer patent issued in 1971, but was tossed back to the USPTO for reissue, with the applicant seeking to address new art that had come to light. During the reissue, the examiner cited Keller’s own prior analog patent alongside an article describing a digital timing system for controlled heart stimulation in research settings. The Board affirmed the examiner’s rejection, concluding that it would have been obvious to replace the analog timing circuitry in Keller’s earlier pacer with digital circuitry as taught by Walsh. Keller argued against this combination, contending that bodily incorporating Walsh’s digital system into his earlier pacer would require extensive modifications and therefore did not suggest the invention.
The Court’s Reasoning: Sitting en banc, as was its norm, the CCPA affirmed and, in the process, provided two key statements that continue to guide obviousness analysis today:
- Bodily Incorporation Not Required to Combine References: The court emphasized that “[t]he test for obviousness is not whether the features of a secondary reference may be bodily incorporated into the structure of the primary reference. . . . Rather, the test is what the combined teachings of the references would have suggested to those of ordinary skill in the art.” In Keller, the result was that – even though Walsh’s digital timing circuitry could not be directly inserted into Keller’s analog pacers, the broader teaching of using digital timing in heart-related devices was still relevant.
- Attacks on Individual References: The court noted that “one cannot show nonobviousness by attacking references individually where, as here, the rejections are based on combinations of references.” The court found that one of the affidavits submitted by the patentee was ineffective because it focused solely on whether a particular reference suggested the combination. The proper analysis required considering what the combination of references would suggest. This principle prevents patent applicants from overcoming obviousness rejections by simply pointing out missing elements in each reference individually, when the combination as a whole still renders the invention obvious.
In many ways, the refusal to accept mere lack of bodily incorporation serves as a pre-statement of the principles espoused by the Supreme Court in KSR allowing for “ordinary creativity, not … automation.” KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398 (2007).
But is it Relevant?: I like to think of bodily incorporation as relevant to the question of obviousness — if the parts of two references simply snap together then it is seemingly much more likely to be obvious. But in its 1985 en banc decision In re Etter, the Federal Circuit explained the arguments that one reference “cannot be incorporated in [the other] are basically irrelevant, the criterion being not whether the references could be physically combined but whether the claimed inventions are rendered obvious by the teachings of the prior art as a whole.” 756 F.2d 852 (Fed. Cir. 1985) (en banc) (emphasis added). For me, Etter improperly introduced one-sided rigidity, but that has continued. In Mouttet, the Federal Circuit confirmed that a prima facie rejection need not show how to physically incorporate a secondary reference into the primary “because obviousness focuses on what the combined teachings would have suggested.” In re Mouttet, 686 F.3d 1322 (Fed. Cir. 2012). This same language is found in more recent cases such as the non-precedential opinion of In re Universal Elecs., Inc., No. 2022-1716, 2023 WL 5219774, at *9 (Fed. Cir. Aug. 15, 2023) (“To the extent that [the patentee] is demanding bodily incorporation or is otherwise attempting to compartmentalize each individual reference’s teachings, it errs.”).
Nuances and Limitations: Despite its widespread application and thousands of case citations, Keller is not without nuance or limitation. As noted in a recent PTAB decision, Ex Parte Bayat (Appeal 2022-001941, 2023), when an examiner’s rejection is explicitly based on bodily incorporation rather than suggested combination, applicants can legitimately argue against the specific proposed combination. And, the case citation is something of a tale-tale sign of rejection affirmance. In my look at PTAB decisions from 2023-2024, fewer than 5% of cases citing Keller reverse the examiner rejection. Applicants will do well to avoid citation to Keller — often with a simple trick of always tying the physical differences to the conclusion that the combination would not suggest the invention.
But, a key problem with Keller is a similarity with KSR — both doctrines eschew clear rules in favor of amorphous standards. With Keller, I often see it cited by the PTAB when refusing to give any weight to the fact that two references would be difficult to combine. In the process the tribunal typically indicates that the focus should be on what is suggested by the combination — at that point though, the opinions very typically fail to explain what is suggested other than the conclusory hindsight statement that the invention being claimed is that suggestion.
At the Federal Circuit, the case and its progeny continue to see favor. In a 2023 non-precedential decision, for example, the Federal Circuit applied Keller‘s principles from to reverse a PTAB finding of non-obviousness. Intel Corporation v. PACT XPP Schweiz AG, 2023 WL 2198649 (Fed. Cir. 2023). The court found that the Board had improperly required Intel to show how the claimed caching mechanism could be bodily incorporated into King’s bus system. The court emphasized that the proper analysis should have focused on whether an ordinary artisan would have been motivated to combine the teachings and if the combination suggested the invention as claimed. The Federal Circuit found the PTAB’s approach to be legal error and proceeded to conduct its own analysis of the motivation to combine, ultimately concluding that the strong evidence of motivation to combine these references addressing the same problem could only support a finding of obviousness.
Out of the thousands of citations to Keller, Westlaw only identifies a small handful with negative treatment. Perhaps the most salient comes from a PTAB holding that Keller only applies to situations “where the prior art separately teaches all of the elements and the issue is whether they can be properly combined.” Ex Parte Lahti, APPEAL 2012-002768, 2013 WL 3294434, at *3 (Patent Tr. & App. Bd. Apr. 24, 2013).
Keller‘s invention is about keep pace, and the Keller decision appears to fit that model — staying in fashion for the duration.