The Opposition Division (OD) did not consider the information provided by Siemens on the TDR 460 train to be enough to substantiate the opposition. The OD particularly considered that the process of transfer between Bombardier and TDR should be considered as having taken place under conditions of implied confidentiality. The OD thus found that the burden of proof had been on Siemens to demonstrate the lack of such conditions of confidentiality between Bombardier and TDR in the Notice of Opposition. In the absence of any such proof the opposition was considered inadmissible.

The appeal case explored the issue of whether the burden of proof really should be viewed as resting with the opponent in such a case.

The public use of an invention prior to the priority date of a patent may be novelty destroying (Article 54(2) EPC). The approach of the UK courts to the issue of prior public use was considered by the High Court earlier this year (IPKat: Mr Justice Nugee and the Superhose: The potentiality of disclosure). The EPO and UK courts agree that an invention is judged to have been made available to the public even if only one member of the public was able to gain access to and understand the disclosure (T 0084/83). A rear-view mirror was considered by the EPO to have been disclosed, for example, because it was used in a public transport vehicle. However, a prior use disclosure is not considered public if it took place under conditions of confidentiality (T 1081/01).

In EPO proceedings, arguments such as an allegation of prior use must be supported by facts and evidence (as recently reviewed in another Boards of Appeal case, see IPKat: Shifting the burden of proof back to the patentee (T 1299/15)). In a case of alleged prior public use, the burden of proof is on the opponent to substantiate the when, what, and how of the public use (Case Law of the Boards of Appeal, I.C.3.2.4).

So who had the burden of proof in Siemen’s opposition of Bombardier’s patent (T 2037/18)? The Board (3.2.01) noted the Boards of Appeal case law that, if prior use is alleged, the burden of proof first falls on the opponent to demonstrate that the sale took place (T 326/93, r. 4.1). In particular, the when, what and by whom of the alleged prior use must be outlined by the opponent. The Board noted that this evidence had been provided in the Notice of Opposition. The question of whether the prior use had been public (i.e. whether it had been confidential) was considered irrelevant for the admissibility of the appeal. Such a question should be left to the assessment of the substantive merits of the case. The Board therefore considered the opposition to have been admissible.

The celestial teapot

The Board further noted that, in general, if the patentee argues in their defence that a prior use was confidential, the burden of proof is on the patentee to prove that conditions of confidentiality were, in fact, in place (T 221/91, r. 2). The shift of the burden of proof to the patentee is considered appropriate given that a buyer can typically be expected to disclose an item they have bought. As Siemens argued, TDR would have had a strong interest in using the train. TDR would also have assumed Bombardier to have made the arrangements for patent protection if any were necessary. As the Board put it “[a] search of the entire body of the Board of Appeal has not revealed a case, in which a chamber from the delivery of a rail vehicle to an operating company would have concluded the existence of a tacit confidentiality agreement” (r. 17, machine translation).

The Board further considered this reasoning to follow the legal principle of “negative non sunt probanda”, i.e. you can’t prove a negative. This Kat is reminded of Bertrand’s Russell’s analogy of the “celestial teapot”. Russell argued that if a party was to allege the existence of a small teapot orbiting the earth (where the teapot is too small to be spotted by telescopes) it would be impossible for another party to disprove the claim. Russell used this analogy to argue that the burden of proof must lie on the party making the unfalsifiable claim.

In the case of prior use, the Board argued, an opponent cannot be expected to prove that the sale didn’t take place under conditions of confidentiality. Such a requirement would require the opponent to prove a negative. The burden of proof would thus amount to the same as would be required to disprove the existence of the celestial teapot.

Burden of proof may shift again in view of prima facie evidence. For example, if a patentee presents what appears to be prima facie evidence of conditions of confidentiality covering a prior use, the burden of proof shifts to the opponent to disprove this evidence (r.10). However, the Board noted that such a shift in the burden of proof could only be triggered after the Notice of Opposition had been filed, i.e. in view of the patentee’s response to the opposition. The Board (obviously a fan of a pertinent Latin phrase), summarised that a shift in the burden of proof with respect to the existence of a confidentiality agreement could only occur ex nunc.

The reasoning provided in this decision is welcome, especially in light of the upcoming changes to the Rules of Procedure of the Boards of Appeal (RPBA) (see IPKat here). Under the new regime it will become significantly harder to introduce new arguments and evidence into proceedings as proceedings progress. Commentators are predicting that opponents and patentees will front load their cases in response to the revised changes. T 2037/18 provides some salient reasoning why it should not be required of an opponent to introduce all possible arguments and counter-arguments in the Notice of Opposition. The Opponent must also be able to respond to the patentee's position as proceedings progress. 

During opposition, the burden of proof very clearly rests with the opponent, at least initially. This applies to all forms of objection, from novelty to insufficiency. However, it is also possible for the burden of proof to shift back to the patentee in certain circumstances. In T 570/08, for example, the opponent and patentee had both submitted data relating to whether the objective problem had been solved. The opponent's data contradicted that of the patentee. The Board of Appeal was convinced that the opponent's data created sufficient doubt so as to place the burden of proof on to the patentee. 

The legal reasoning in this decision is similar to that of T 63/06, in which the opponent was able to shift the burden of proof with regards to the sufficiency to the patentee. In T 63/06 the Board found that the patentee was relying on only a weak presumption that the invention was sufficiently disclosed. The opponent was successfully able to shift the burden of proof by arguing that the skilled person would have difficult in putting the invention into practice. 

The recent case T 1299/15 is similarly a reminder that insufficiency attacks may be applicable even to the mechanical field. As is often the case with insufficiency disputes, the issue of how much work would need to be done by the skilled person (and whether inventive endeavour would be required) was a key issue. Following on from this, if the patentee has not provided an embodiment of the claimed invention, the burden of proof is shifted to the patentee if there is serious doubt that a skilled person could perform the invention without inventive skill.