Brief outline of the case
The Legal Division rejected the applicant’s request for the transfer of the application from Gyrus ACMI Inc. to Olympus Medical Systems Corporation to be recorded in the European Patent Register.
The applicant appealed this decision
The appeal was dismissed.
The case is interesting as it deals with what can be considered as a signature under Art 72.
The legal situation
According to Art 72, the assignment of a European patent application must be made in writing and requires “the signature of the parties to the contract“. Under R 22(1), the transfer of a European patent application is recorded in the European Patent Register on request and upon production of documents providing evidence of such transfer. Pursuant to R 22(3), a transfer has effect vis-à-vis the EPO only at the date when and to the extent that the documents referred to in R 22(1) have been produced.
The case in hand hinges on how the term “signature” in Art 72 is to be understood and, in particular, whether it encompasses electronic signatures in the form of “text string signatures” without any further qualification.
The board held that, when interpreting the provisions of the EPC, the principles of interpretation laid down in Articles 31 and 32 of the Vienna Convention on the Law of Treaties 1969 (VCLT) must be applied, cf. G 5/83, Reasons 4; G 2/12, Reasons V.(3). The board also referred to G 3/14, Reasons 55.
The starting point of interpretation under Art 31 VCLT is the wording, i.e. the “objective” meaning of the term to be interpreted, regardless of the original “subjective” intention of the contracting parties, see G 3/19, Reasons XIV.2.
Definition of signature
From dictionaries
The Board provided dictionary extracts by way of example only, as a first approximation for establishing the most common usage of the term “signature” in the context in which this term is used in Art 72, i.e. in the context of signing a contract. The board referred to the Cambridge English Online Dictionary and to the Merriam-Webster’s Online Dictionary.
For the board, in the given context of a contract the dictionary definitions refer to the term “signature” mostly – although not exclusively – as a handwritten depiction of someone’s name.
From the different linguistic versions of Art 72
The board also referred to the various versions in all three official languages of Art 72.
The signature requirement in Art 72 ensures that the assignment contract becomes clearly attributable to the signing parties when they put their names on the contract in a distinctive manner, thereby creating an objectively verifiable link between signature and signatory.
It also highlights the legal significance of the act, thereby helping to prevent approval due to a lack of attention. This is commensurate with the importance of this legal transaction, which “has the effect of making a patent application the property of another person”.
Electronic signatures
If one were to interpret the term “signature” in Art 72 as also including electronic signatures, the question immediately arises as to which kind of electronic signatures could qualify as a “signature” within the meaning of Art 72.
The possibilities include simple electronic signatures, such as the “text string signatures” referred to by the applicant, advanced electronic signatures which are uniquely linked to and capable of identifying the signatory, or qualified electronic signatures which are additionally based on a qualified certificate.
In this respect, the board referred Regulation (EU) No. 910/2014 of the European Parliament and of the Council of 23 July 2014 on electronic identification and trust services for electronic transactions in the internal market.
The board considered that it would be at odds with this rationale of Art 72 if – without any explicit legal basis – any type of text in electronic form referring to the name of a person were considered a “signature” within the meaning of Art 72. Moreover, such an interpretation would not ensure that a “signature” makes the assignment contract clearly attributable to the signing party.
The EPC as signed and the later amendments
At the Munich diplomatic conference 1973, not only the Articles but also the Implementing Regulations to the EPC 1973 were directly concluded between the Contracting States.
The legal situation is less clear with regard to subsequent amendments to the Implementing Regulations. In Art 33(1,c), the contracting states to the EPC empowered the Administrative Council, as an organ of the international organisation created by the EPC, to amend the Implementing Regulations.
Accordingly, none of the subsequent amendments to the original Implementing Regulations were concluded between the contracting states to the EPC themselves but were adopted by the Administrative Council. This is true for the entire Implementing Regulations to the EPC 2000.
Against this background, i.e. in the absence of a separate agreement among the contracting states on the Implementing Regulations to the EPC 2000 and the subsequent amendments to these, the question arises whether the current Implementing Regulations are also to be regarded as part of the “text” of the treaty within the meaning of Art 31(2) VCLT.
Rule 2(2) EPC is not applicable in the present instance
Generally speaking the Implementing Regulations must be taken into account when interpreting Art 72. Hence, the applicant’s reference to R 2(2) – which in the applicant’s opinion provides meaningful context for the interpretation of Art 72 – cannot be disregarded from the outset.
For the board, R 2(2) does not have any impact on the interpretation of the term “signature” in Art 72.
The scope of application of R 2 in its entirety is limited to formal requirements for filing documents in proceedings before the EPO.
Formal requirements for filing documents in proceedings before the EPO concern a relationship between the EPO, a public authority, and a party to proceedings before the EPO, which is usually a private party.
Art 72, on the other hand, regulates formal requirements for assignment contracts. This usually concerns a relationship between two or more private parties.
Hence, R 2 and Art 72 regulate entirely different matters.
R 2 cannot be regarded as a provision implementing Art 72. Moreover, the mere fact that R 2 is part of the general provisions in Chapter I of the Implementing Regulations, as argued by the applicant, does not lead to an extension of its scope of application – which is clearly delimited in R 2 itself.
The applicant also invoked the “Decision of the President of the EPO dated 14 May 2021 concerning the electronic filing of documents” (OJ EPO 2021, A42), which is based on R 2. However, contrary to the applicant’s assertion, this decision does not provide meaningful context for the interpretation of Art 72 EPC either.
For the board, R 2(2), second sentence, does not empower the President of the EPO to specify what qualifies as a signature under Art 72 EPC.
The board concluded thus that, in the absence of a different definition in the Implementing Regulations, the term “signature” in Art 72 must be understood as referring to a handwritten depiction of someone’s name.
Comments
The distinction made by the board between electronic signatures in dealings with the EPO, allowable under R 2(2) and signatures in the meaning of Art 72, which can only be handwritten, is interesting and understandable.
It remains to be seen, whether in view of the progressing digitisation of the world, this distinction can be upheld.
On the other hand, the day when contracts signed in presence of a notary public will allow electronic signatures appears rather in the distant future. This will be first a question for national legislators in the EPC Contracting States, before it will concern the EPO. And for contracts between parties outside the EPC Contracting States, this day might never come.
https://www.epo.org/en/boards-of-appeal/decisions/j230005eu1
Comments
6 replies on “J 5/23 – What is to be understood under a “signature” in Art 72”
I am not sure that the decision’s logic is secure when it says that:
“The signature requirement in Art 72 ensures that the assignment contract becomes clearly attributable to the signing parties when they put their names on the contract in a distinctive manner, thereby creating an objectively verifiable link between signature and signatory”
– or at least not as far as this logic applies to wet-ink signatures.
For a wet-ink signature, anybody could have been holding the pen. There is no way of objectively verifying who actually held the pen, in the absence of additional evidence such as a film of the act of signature. It is possible to learn to forge signatures. And many people (myself included!) sign their name so rarely these days that the signature is never the same twice, which breaks or at least weakens any “objectively verifiable link”.
I can perfectly well understand that the EPO wants to preserve the old, tried-and-tested way of signing things (with the known level of risk that we have all been used to for centuries) until it is completely certain that electronic signatures are both secure enough and legally permissible. That is completely fair enough. But the logical reasoning *for* the old policy is weak, even if the logical reasoning *against* electronic signatures is stronger. Sometimes it is better simply to acknowledge that you want to stick with tradition because you like it, rather than trying to construct an elaborate justification.
@ Anonymous
I can fully agree with you that “For a wet-ink signature, anybody could have been holding the pen. There is no way of objectively verifying who actually held the pen, in the absence of additional evidence”.
Even if the signature changes slightly each time it is written down, there remain constants in a signature.
Furthermore, a signature can be completely illegible, but it will remain a signature. Just look at a physician’s signature on a drug prescription! And yet the chemist does not start questioning the legitimacy of the person signing.
It has to be accepted on balance of probabilities, that the person holding the pen is the person entitled to sign. Contracts like transfer of IP rights are often countersigned by notaries or even layers having helped setting up the contracts. Their role is to certify that the person signing was indeed present when signing.
This is quite different with an electronic signature where it is clear that the person is not present when signing as it all happens in the virtual world.
This is why the distinction made by the board about electronic signatures in dealings with the EPO cannot be compared to electronic signatures stemming from outside and not directly having to do with dealings with the EPO.
The attitude of the board has nothing to do with sticking to traditions but with as much as possible legal certainty.
When power of attorneys had to be produced in the original, i.e. during in-person OP, I remember refusing one. It was well signed, but the legal position of the person in the company having signed was not apparent at all. I was told it was the legal representative of the company, but this was not enough for me. It could have been signed the cleaning lady in the patent attorney’s firm.
As the representative was a qualified representative, we could hold the OP, but the decision was given in writing once a properly signed power of attorney was filed. If the representative would have been an employee of the party, then the OP would have had to be postponed.
This also has nothing to do with sticking to tradition, but on balance of probabilities, such a power of attorney was not admissible.
This decision is of interest to me because it is based on interpretation of provisions of the EPC according to the principles of the Vienna Convention.
I recently challenged the Guidelines interpretaton of Rule 51(3)EPC, according to which the EPO requires payment of all back annuities on a divisional.
The Receiving Section challenged my viewpoint. They stated that Rule 51(3) should be interpreted according to the principles of the Vienna Convention. However, the Receiving Section did not attempt to justify why the existing Guidelines practice complied with these principles.
I recently wrote a 30-page letter to the Receiving Section setting out what seems to me to be the only possible interpretation of Rule 51(3) that complies with the principles of the Vienna Convention, and that requires revision of the Guidelines.
We await their reply.
If you are interested, I can send a copy of the 30-page letter so you can form your own opinion. I can also explain why this change of practice is so important mainly in the context of the UPC.
Dear Brian,
It is indeed relatively rare that the VCLT is quoted in “normal” procedures before the EPO. This is more the case when the EBA acts under Art 112, e.g. G 1/21 or G 3/19. In the latest decision of the EBA, G 2/21, there is no reference to the VCLT.
This does however not mean that the Interpretation of the EPC by the EBA under the VCLT is at all compelling or convincing. It can be summarised by “its fits the purpose”, i.e. an expected result. This is especially the case when the EBA rewrites or interprets the questions referred like in G 1/21 and G 3/19.
I would be interested in your considerations on R 51(3) and will contact you by separate mail.
Interestingly you have mentioned the UPC.
It is interesting to note that in its decision on the provisional distribution of actions related to patents in IPC sections (A) and (C) pending before the central division, the Presidium of the UPC has carefully avoided mentioning the VCLT.
One wonders why?
The same applies to the decision dated 22.06.2023 of the Administrative Committee of the UPC under Art 87(2) UPCA.
One also wonders why?
It is difficult to see why the VCLT should not apply to the UPCA. There is no reason that it should not apply to the UPCA when it applies to the EPC.
In the following post, I have quoted the Art of the VCLT which appear important when it comes to interpreting the UPCA:
https://blog.ipappify.de/is-after-brexit-the-upc-agreement-compatible-with-eu-law/
Art 25, 31, 56 and 62 VCLT appear directly applicable to the UPCA.
Mr Steenbeek, a staunch supporter of the UPC has even gone so far as to consider that there is no need to refer to Art 31 VCLT.
One does not wonder why!
On a lighter note, wet signatures may have very peculiar risks, especially with fathers-in-law:
https://www.theguardian.com/lifeandstyle/2015/jan/20/gordon-ramsay-court-battle-pub-rent#:~:text=The%20chef%20accused,on%20legal%20documents
Dear Luis,
Thanks for your comment. For outsiders we can see what happened in UK on a lighter note, but the father-in-law concerned must not have ben amused when discovering that a ghost-writer machine had been used to forge his signature under a lease contract.
It simply shows that such machines cannot be given in all hands. The malevolent use of such a machine can have very dear consequences.
As it was on a contract the malevolent use of the machine did not appear soon. If it had been used on bank cheques, it would have become apparent much earlier. But even then, I doubt that the bank would consider the cheques not properly signed.
If used on mass mails or photographs, it does not normally not have such daring consequences, but even then a malevolent use cannot be totally excluded.