CASELAW-EPO - reviews of EPO Boards of Appeal decisions

T 1138/20 – Standard of proof in case of a public prior use/public disclosure

chat_bubble 2 comments access_time 7 minutes

The patent relates to “Improved spot beam satellite systems”. It results from a divisional application of EP 07873929/EP 2 111 697.

Brief outline of the case

The proprietor appealed the decision to maintain the patent according to AR2.

Claim 1 as granted was lacking N over a PPU represented by the following documents:

NPL1=”DDSO – Digital Divide: the Satellite Offer”

NPL1-2=copy of the webpage

NPL1-3=ESTEC Contract Number 18194/04/NL/US

The opponent was the ESA.

The proprietor challenged the OD’s findings of fact in this respect, in particular that NPL1 was the presentation that was delivered to the European Space Agency (ESA) and then published on its website.

For the board, the proprietor has failed to demonstrate that the OD erred in its findings of fact. Thus, the board saw no reason to overturn the finding of the OD that NPL1 is state of the art within the meaning of Art 54(2).

The board thus confirmed the availability to the public of NPL1, as well as the lack of N of claim 1 as granted and of claim 1 of AR1 over NPL1.

In view of the prohibition of reformation in peius, the appeal was dismissed.  

The case is interesting as it reviews the rules applicable to fact finding and the evaluation of evidence.

The proprietor’s position

The proprietor argued that it had not been provenbeyond reasonable doubt” that NPL1 was the document that had been delivered to the ESA (European Space Agency) or published on ESA’s website (cf. NPL1-2). The contract NPL1-3 stated that the final presentation should not be marked as “proprietary”, but NPL1 was in fact marked as “proprietary”.

This suggested that the document provided as NPL1 was not the final presentation (cf. clauses 38.1 and 38.2 of NPL1-4). The standard of proof “beyond reasonable doubt” had not been met in the present case, in particular because the understanding that NPL1 was a draft document (rather than the final version) was consistent with the documentary evidence and with statements made by the witnesses confirming the existence of draft documents.

This constituted a plausible alternative explanation to that put forward by the opponent (and more consistent with the evidence than the opponent’s explanation). The existence of such a plausible alternative explanation had the inevitable consequence that the opponent’s explanation had not been proven to the required standard (i.e. “an extremely high degree of certainty”). Furthermore, there were fundamental deficiencies in the evidence adduced by the opponent.

The board’s position

There is only one standard of proof

The board noted that, for the issue concerning whether evidence demonstrates the prior art’s availability to the public, sometimes the standard of “balance of probabilities” and sometimes the standard of “beyond reasonable doubt” or “up to the hilt” is applied in the jurisprudence of the boards of appeal (see Case Law of the Boards of Appeal, henceforth “CLBA”, 10th edition, section I.C.3.5.2 a)).

This seems to imply that when establishing facts in proceedings before the EPO, different standards of proof are applied, depending on the circumstances. More specifically, according to the case law summarised in the CLBA, section III.G.4.3.2, it appears that in proceedings before the EPO, the general standard of proof is that of “balance of probabilities” which in certain circumstances changes to that of “beyond reasonable doubt” or “up to the hilt“.

Circumstances for applying a stricter standard are for example cases of public prior use where all, or almost all, evidence lies within the sphere of the opponent, while the general standard of “balance of probabilities” is typically applied if both parties had equal access to the evidence in question.

However, the board took from the decisions as cited in the CLBA that rather the circumstances of the case, for example that all pieces of evidence lie within the sphere of the opponent, are decisive for the deciding body to form an opinion as to what has been proven.

The present board considered that making a distinction between the above-mentioned standards is neither necessary nor mandated by case law. This approach is in line with several decisions of the boards of appeal. The board quoted T 545/08, Reasons 11, T 768/20, Reasons 2.1.2, T 660/16, Reasons 5.3.5 and T 1634/17, Reasons 19.

Therefore, the present board considered that there is only one standard of proof in the proceedings before the EPO: the deciding body, taking into account the circumstances of the case and the relevant evidence before it, must be convinced that the alleged fact has occurred.

The power to review the finding of facts

For the present board, the boards have the power, at any stage of the appeal proceedings, to establish the relevant facts of the case before them and thereby substitute the findings of fact of the departments of first instance. However, the boards have no obligation to establish facts de novo already established by the departments of first instance OD. In this respect the board quoted T 42/19, Reasons 3.3.

Neither the EPC nor any decision of the EBA limits the boards’ competence to establish the facts. Under Art 114(1), the EPO examines the facts of its own motion. This examination has no further restriction in the law.

The power to establish the relevant facts of the case before the boards of appeal is not limited to what the parties have provided or what the first-instance department has established.

In relation to Art 114(1), the EBA has restricted the boards’ power to examine in its opinion G 10/91. However, this restriction is solely related to the examination of fresh grounds for opposition (see Reasons 18), not the establishment of facts.

No restriction can be found in G 7/93, either. In this regard, it is noted that T 42/19 (see Reasons 3.4 and third Catchword) refers to the frequently cited obiter dictum in G 7/93, Reasons 2.6.

Fact-finding and exercise of discretionary power

The present board added that the board’s review of a fact-finding process should not be conflated with the review of discretionary decisions within the meaning of the obiter dictum in G 7/93, Reasons 2.6. The board also referred to T 1604/16, Reasons 3.1.8.

For the board, the review of findings of fact has simply nothing to do with the review of discretionary decisions. Discretionary decisions can only be reviewed for a particular type of errors, cf. G 7/93, Reasons 2.6, while a review of findings of fact has no such limitation.

No obligation to establish all facts de novo

Having the boards of appeal as final judicial arbiter of disputes arising in the context of the EPC is not a luxury to be had but it is a necessity to ensure compliance with international obligations, in particular that the proceedings under the EPC are in accordance with Art 6 ECHR. See also T 1604/16, Reasons 3.1.7 and the WTO/TRIPS Agreement, cf. Art 32 and 41.4 TRIPS.

If the boards did not have such power, in particular to examine all questions of fact, the users of the EPC system would have no recourse to a “tribunal” to review the decisions of the departments of first instance and, therefore, users would be deprived of an effective judicial remedy.

If the findings of fact of the departments of first instance rely on documents, the boards regularly make an own assessment of the evidence. The board might assess the piece of evidence (i.e. the prior-art document) de novo and replaces the findings of fact of the OD with its own findings.

In case of oral evidence given by a party, a witness or an expert, cf. Art 117(1)(a), (d), (e), or an inspection, cf. Art 117(1)(f), the boards will normally not assess the evidence de novo by, for example, re-hearing the witness.

In such a situation, the department having heard the witness is better placed to determine the probative value of the evidence given, in particular the credibility of the witness, i.e. the witness’ truthfulness and honesty, and the reliability of the testimony, i.e. its accuracy, e.g. how accurately the witness remembers and recounts the events. The boards may then defer to the first-instance department’s evaluation of evidence.

Burden of proof in appeal

The boards’ primary function is to review the decision of the first-instance department in a judicial manner, cf.see Art 12(2) RPBA20; see also G 9/91, Reasons 18. Hence, the parties on appeal may challenge any findings of fact made by the first-instance department in its decision, and the boards will review these findings.

However, the burden is on the party challenging a fact on appeal to demonstrate that the first-instance department erred in its finding of fact. In doing so, the party must specifically point to each alleged error in the department’s findings of fact or in its evaluation of the evidence and set out the reasons why this is considered erroneous.

If the party succeeds in discharging this burden and demonstrating such an error, the board will establish the facts on its own if this is necessary for reaching a decision. Depending on the circumstances of the case, the board may also remit the case to the respective department for a new finding of fact, cf. Art 111(1).

Reviewing errors in the fact-finding process

In the case at issue, the proprietor has invited the board to review the opposition division’s findings of fact.

In this regard, the board found helpful guidance in decision T 1418/17, Reasons 1.3, where the deciding board identified three typical situations in which a finding of fact may be overruled, namely

  • the failure to take into account essential points,
  • the consideration of irrelevant matters or
  • (iii) the violation of the laws of thought, for instance in the form of logical errors and contradictions in its reasoning

This list is not exhaustive. Since the boards have full competence to review findings of fact, there is no reason why a board should be limited in its review to these criteria.

For demonstrating an error in the fact-finding process, it will generally not be sufficient to argue on appeal that the evidence on file would have allowed a different conclusion and that this conclusion was equally likely, plausible or reasonable.

Such an argument is, by itself, normally not sufficient to meet the burden of showing an error. The mere fact that the body competent to establish the facts, be it the examining or opposition division, could have come to a different conclusion is typically not sufficient to demonstrate an error in the fact-finding process.

Rather, in order to show an error, the party must demonstrate that no body competent to establish the facts and acting in a reasonable way could have come to this conclusion.


The review of the case law about fact-finding and evaluation of evidence is interesting, but does not change matters fundamentally.

It is interesting to note that the board considers that under Art 114(1), the EPO examines the facts of its own motion. Further to the limitation under G 10/91, the board has simply forgotten to add a further limitation: the RPBA20.

The board came to the remarkable conclusion that there is only one standard which relies on the circumstances of the case. Actually applying either the “balance of probabilities” or “up to the hilt”, depends on the circumstances of the case, so the conclusion of the board cannot be seen as a scoop.  

It is true that the EPO boards of appeal are the last instance competent to carry out a judicial review of decisions of first instance,  but it has never been disputed that the boards have the power to come to a different assessment of facts and evidence.

That the onus of proof in appeal lies with the appellant is neither a scoop.

It is also rather rare that witnesses are heard for the first time by a board.

Whether heard for the first time by an OD or by a board it should be kept in mind that a witness can only confirm facts or evidence submitted during the opposition period and mentioned in the statement of grounds of opposition.

During its hearing a witness cannot supplement information which should have been filed during the opposition period, cf. T 241/99 or T 2010/08.

According to T 1100/17, it is the function of a witness to confirm what has been alleged and not to fill in the gaps in facts brought forward to support the case

The hearing of witnesses during an OP by ViCo is also not free of problems, cf. T 423/22 commented on this blog.

Share this post


2 replies on “T 1138/20 – Standard of proof in case of a public prior use/public disclosure”

Max Dreisays:

Daniel, thank you for the reference to T1100/07. Embedded within that Decision is a reference to J10/04. On the issue of “hearing” a witness, the decisions are interesting, especially in the context of doing it via a ViCo. Opinions might well vary, about the degree of credibility given, to what a witness says i) in person and ii) to a video camera. But (being provocative) does anybody care, any more?

As to the level of certainty it needs, before evidence testimony is accepted, you write that the tribunal must be convinced that the fact alleged has actually occurred. I see it differently. I think that the tribunal must be convinced that it is more likely than not, that the fact alleged really did occur.

I say that because of my own experience with a case not so very different from the facts in J10/04 but which, in my case came to the opposite conclusion on the facts alleged. Here they are:

Long before the days of e-filing, when the night letter box was in the Corneliusstrasse, a colleague posted an opposition at 8 pm on the last day of the 9 month period. It was a Friday evening. The EPO denied receiving the opposition. Evidence was filed, not only from the colleague but also from the formalities officer in the patent attorney firm, as to what papers were in the envelope. There was also an Affidavit from me, explaining how envelopes can be removed from inside the night letter box when it is full. On an earlier occasion, I had myself tried to post a large envelope on a Friday evening, only to find that I could not get it inside the flap because of blockage by a traffic jam of earlier-posted envelopes. I could remove from the box the most recently filed one or two of those earlier-filed envelopes. Perhaps somebody (a Friday evening reveller?) removed from the box my colleague’s duly posted opposition. Or another messenger from a different patent attorney firm, desperate to post their own envelope on time and then go home?

We asked the EPO for the architect’s drawings of the inside of the Corneliusstrasse night letter box. The EPO answered: there aren’t any. The Board decided that the opposition was not timely filed. We could not convince it that, more likely than not, the opposition papers were duly posted into the night letter box. Soon after that, the night letter box in the Corneliusstrasse was demolished and a new and more reliable one provided, in the Kohlstrasse.

In my case, then, despite the persuasive power of the witness evidence, the Board was not willing to be “convinced” by it. Instead, it “convinced” itself to find that the witness evidence was not convincing. In other words, they chose to disbelieve at least one of the three witnesses.

Is this not typical, for witness testimony. If it fits with the preconceptions of the Board, it is welcomed. But if it contradicts what the Board prefers to believe, it is dismissed as “unreliable”.

As you say, there are “problems” with witness testimony at the EPO, even though, under English law, it is the “Gold Standard” for resolving a dispute over the facts.

English and European mainland fact-finding processes are different. Such differences enrich patent law in Europe, I think.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Thanks for your comment.

I can agree that opinions might vary about the degree of credibility given to the testimony of the witness. This is however depending on the way the witness is questioned.

In principle, it should not make a difference whether the witness is heard in person or by ViCo. What I find problematic, is when a witness is heard by ViCo there is no guarantee that the witnesses has not listen in the OP before giving his statement or cannot be influenced or helped by documents lying in front of him. See my comments on T 423/22. When a witness is heard in person, it cannot be helped as he is sitting alone in front of the deciding body without any ancillary means at his disposal.

In the end what matters is the evaluation of the testimony. The basic question is then to decide whether the facts stated in the grounds of opposition have or not been confirmed? In this respect, the criterion is balance of probabilities, but in the end, it depends on whether the deciding body has been convinced about the sincerity of the witness. I used the term “convinced” within this meaning.

I would not go as far as to claim that if it fits with the preconceptions of the Board, it is welcomed, but if it contradicts what the Board prefers to believe, it is dismissed as “unreliable”. To my liking, this goes a trifle to far.

Your case was somehow particular and I can understand your frustration in the view of J 10/04. The odds were clearly against you, but you should not forget that the opponent has a more difficult position to start with. Any doubts about the admissibility of an opposition are not decided in favour of the opponent.

The only concession to opponents is that they can benefit from Art 122 if an opponent has failed to observe the time limit for filing the statement of grounds of appeal. He is then party to a proceeding before the EPO. See G 1/86, OJ 1987, 447. When filing an opposition in the letter box, an opponent is not even yet party to proceedings before the EPO.

On the other hand, it is frequent at the EPO that decisions can diverge. On certain topics, I can produce decisions going in one way or in the completely opposite way.

We should also not forget that the concept of cross-examination as practiced in the UK has not swept over The Channel. It might be regrettable, but so it is. I do agree with you, that at least at the EPO a kind of middle way has been developed by the boards. This has certainly enriched patent law in Europe.

Leave a Reply

Your email address will not be published. Required fields are marked *