Posted: October 18th, 2022
G 2/21 on post-published evidence and plausibility – Preliminary Opinion of the EBA, reported by Dr. Klaus Reindl and Dr. Georg Anetsberger, Bardehle Pagenberg
With the decision T 116/18, three questions were referred to the Enlarged Board of Appeal (EBA) of the EPO relating to post-published evidence. The EBA now issued its preliminary opinion as to these questions in preparation of the oral proceedings scheduled for November 24, 2022.
The preliminary opinion shows a strong tendency towards answering the first referral question with “No”: No exception to the principle of free evaluation of evidence should be applied when assessing post-published evidence. Such evidence cannot be disregarded per se, as this would deprive the party relying on such evidence of their right to be heard (Art. 113 (1) EPC and Art. 117 (1) EPC).
In principle, the EBA could have stopped there, as the referral questions two and three only concern the case of a positive answer to the first referral question. However, the EBA acknowledges that there is need to provide “some guidance” on the issues referred to in questions two and three. The EBA thus takes a pragmatic approach of trying to provide guidance and more legal certainty, without insisting on a formal rewording or reordering of the referral questions (which the preliminary opinion contemplates as well).
As a consequence to the answer to the first referral question, post-published evidence regarding a certain technical effect is considered as “potential source of evidence” to be taken into account when assessing inventive step. The core issue, according to the EBA, is that the purported technical effect needs to be encompassed by the technical teaching of the application as filed, as understood by the skilled person in view of common general knowledge. It must not embody a different invention. If the skilled person had “significant doubts” in that regard, an exception would apply. In such exceptional case, successfully relying on post-published evidence, according to the EBA, is “questionable”.
As a result, the EBA provides a clear roadmap for assessing inventive step. Any evidence has, as a general rule, to be taken into consideration and assessed. This also applies to post-published evidence for supporting the presence of a technical effect relied on in inventive step discussions. Only in the exceptional circumstance of “significant doubts” as to the purported technical effect based on the application as filed, such evidence may not be sufficient.
Notably, the EBA avoids any reference to the term “plausibility” used in the referral questions. Apparently, the EBA is not interested in coining a new criterion for assessing inventive step but instead tries to rely on other established principles for dealing with the matter. Nevertheless, the preliminary opinion comes close to what the referring decision termed “ab-initio implausibility” (post-published evidence cannot be relied on if the purported technical effect was implausible ab-initio, i.e., based on the application as filed). It thus finds middle ground amongst the various levels of “plausibility” outlined in the referring decision.
What has not been expressly commented on by the EBA’s preliminary opinion is the burden of proof. However, it should be assumed that it would be the patentee’s burden to show that the technical effect in question is encompassed by the technical teaching of the application as filed. It would then be the opponent’s burden to show that there are significant doubts as to the purported technical effect based on the application documents.
While the EBA’s preliminary opinion clearly goes beyond what would be needed to formally answer the referral questions, it also clearly indicates that it will limit the scope of the decision to inventive step, as the referring Board limited their questions accordingly. Hence, it will be interesting to see whether the EBA will stick to this indication or whether they will ultimately include an obiter dictum regarding applicability to sufficiency of disclosure as well.