CASELAW-EPO - reviews of EPO Boards of Appeal decisions

T 2968/19 – No recourse to the description in case of lack of clarity – Limited application of Art 69 in procedures before the EPO

chat_bubble 29 comments access_time 6 minutes

The patent relates to a process and an apparatus for the cryogenic separation of air in a distillation column system, in which the control of at least one process parameter is performed by a combination of an “ALC control” and an MPC controller.

Brief outline of the procedure

The application was refused as the ED considered that the subject-matter of claims 1 and 6 the subject-matter of claims 1 to 5 of the new MR did not satisfy the requirements of Art 84.

The refusal decision of the ED is in two parts. In the first part, the ED came to the conclusion that claim 1 of the MR lacked N over D1 and D2. In the following second part, after having decided on the lack of N, the ED then decided that the feature “ALC control” was “vague and unclear

The applicant appealed the refusal.

The board confirmed the lack of clarity of claim 1 of the MR: AR 1 and 2 were not admitted under Art 13(2) RPBA20.

The application was then refused for good.  

The whole discussion in examination and in appeal was about the possible meaning of the feature of an “ALC control“, which was considered unclear by the ED and the board, whereas the feature of an “MPC controller” was considered clear.

The ED’s position

In the contested decision, the ED concluded that claims 1 to 5 did not satisfy the requirements of Art 84 by using the “vague and unclear” feature “ALC control“.

The skilled person was uncertain as to its meaning, in particular because the term was used with different meanings in the prior art, for example in D2.

 However, the feature had to be defined in the claim – as also required by the Guidelines for Examination, F-IV, 4.2in such a way that the meaning was clear from the wording of the claim alone.

The applicant’s position

The applicant argued that, although the term “ALC control” is used for all sorts of things that have nothing to do with the “ALC control” of the patent application”, the correct meaning of the feature “ALC control” used in the claim was sufficiently clear from the description.

This was sufficient with regard to the requirements of Art 84, since according to general principles of interpretation and by analogy with Art 69 the description had to be used for interpretation. This was also in line with the general principle that “the patent application constitutes its own dictionary“.

For the applicant, Art 69(1) and its Protocol on Interpretation should be applied by analogy in examination procedures, and concluded that the feature “ALC control” is sufficiently clear in combination with the description.

Thus, the skilled person would directly and immediately understand what was meant by the term “ALC control” and how it was integrated into the claimed control concept.

The claim wording also satisfied the requirements of conciseness set out in Art 84.

In a balanced overall view of the above-mentioned legal requirements of Art 84, the claim wording thus proved to be allowable.

Moreover, the introduction of amendments for clarification would entail the risk of newly arising further objections, for example with regard to an inadmissible extension of subject-matter.

The board’s position


According to the established case law of the Boards of Appeal, the patent claims should be clear in themselves in such a way that a person skilled in the art does not need to consult the contents of the description in order to understand their subject-matter, cf. Case Law of the Boards of Appeal, 10th edition 2022, II.A.3.1. The board further referred to G 1/04, Diagnostic methods, Grounds 6.2.

The board did not agree with the “general principle” according to which the patent application constitutes its own dictionary.

For the board, Art 69(1) relates to the determination of the scope of protection of the granted patent and thus applies in proceedings before the EPO essentially in connection with Art 123(3).

Nevertheless, the description is not generally disregarded when it comes to the interpretation of claim features .

The board referred to T 1646/12, grounds 2.1 in which the board stated that two extremes must be avoided:

  • Firstly, it is not permissible to regard the claims and the description as communicating vessels, so to speak, for example, by reading into the claims limiting features which are disclosed in the description but not in the claims. Such a transfer of limiting features cannot be achieved by interpretation, but only by amending the claims.
  • Secondly, the claim cannot be regarded as completely separate from the description. The skilled person interpreting a claim must at least ascertain whether the expressions of the claim are to be understood according to their ordinary literal meaning, or whether the description defines a special meaning for these expressions.”

For the board, this balancing exercise is to be applied in an analogous manner to the review of the criteria of Art 84 in the case of amendments to the claims. However, here, in contrast to the interpretation of granted claims, the principle must first be satisfied that the missing essential features must be included in the claim and that the meaning of the features for the person skilled in the art should be clear from the wording of the claim alone, see G 1/04, supra.

The applicant’s recourse to the description in the sense of a separate dictionary may be admissible for interpreting the claims of a patent in opposition proceedings when features are no longer amenable to objection under Art 84 but must nevertheless be interpreted.

The board added that there are limits to such recourse in claim interpretation, cf. “primacy of the claim”: T 1473/19, grounds 3.16.2, commented in this blog, or T 169/20, grounds 1.3.4, also commented in this blog, in particular in the way that features missing in the claim, and not implicitly defined from the technical understanding of the terms used in the claim, cannot be read into the claim.

Similarly, in the examination procedure, the absence of such essential features which are not already immediately and clearly implicitly recognisable as belonging to the claim on the basis of general technical knowledge concerning the terminology used in the claim must be objected to as a defect under Art 84.

According to page 2 of the description the term “ALC control” stands for “Automatic Load Change Control“.

Whilst the term “MPC controller” has a definite meaning in the field of control engineering = “Model Predictive Controller”, this is not the case for the term “ALC controller“.

In D2, the term “ALC system” is indeed used in the sense of an “automatic load change“. However, according to the applicant, this system is precisely not to be regarded as an ALC control in the sense of the invention.

The term “ALC control” is thus unclear as such.

Missing essential features

In principle, the skilled person understands “Automatic Load Change Control to mean that when there is a change between different load cases of the cryogenic air separation plant, the system automatically switches to another load case with a different steady-state operating point in each case.

However, claim 1 additionally defines a set of measured values from test operation (feature [C]) that are included in the “ALC control”. Thus, the “ALC control” according to claim 1 no longer falls within the broad functional definition. This is not in dispute, as the applicant itself refers to further specific definitions on page 2, line 15 to page 3, line 5 of the application documents for the interpretation of the feature.


The requirement of conciseness in Art 84 is a condition in its own, according to which it is intended to prevent the clear intelligibility of the claim from being compromised, for example, by redundant features. However, it is not to be applied in such a way as to limit the requirements for clarity under Art 84.

Applicant’s argument that the necessary conciseness and intelligibility of the claim would no longer be maintained by a corresponding “comprehensive definition” is merely an assertion that is not substantiated in any detail and not readily comprehensible.

It is therefore not at all convincing that corresponding clarifications would “overload” the claim to the detriment of clarity.


The present decision belongs to what I have called in preceding blogs to the “second line of case law” with e.g. T 169/20, T 821/20 and T 1127/16.

For this line of case law, Art 69 is, in principle, only applicable before the EPO in opposition proceedings when it comes to the application of Art 123(3).

Art 69 is not to be applied in examination and is only applicable in opposition when features are no longer amenable to an objection under Art 84 but must nevertheless be interpreted.

In the present case, the board took the bother to check the claims of the MR against all criterion of Art 84, clarity as such and conciseness. It also added that essential features ought to be stated in the independent claim.

Even if one could consider that the leading line of case law is the “second line of case law” which limits the use of Art 69 in procedures before the EPO, the EBA should clarify the situation.

If claim 1 is manifestly unclear, as confirmed by the board, it is difficult to follow the ED that the subject-matter of claim 1 was at the same time lacking N. This would be possible, if the claim would be given a certain interpretation and with this interpretation, the subject-matter could lack N or IS.

In its decision, the ED dealt first with the lack of N to conclude in a second part that the claim was as well unclear. If a claim is unclear, how can it be compared with any prior art?

Share this post


29 replies on “T 2968/19 – No recourse to the description in case of lack of clarity – Limited application of Art 69 in procedures before the EPO”

Max Dreisays:

Daniel, you end with a question: if a claim is unclear how can one compare it with the prior art?

I am surprised. The answer is, I had thought: With no difficulty at all. The more unclear the claim, the easier it is to run a novelty objection.

I had thought that Examiners at the EPO have been using for decades a squeeze argument (at least in their WO-ISA’s addressed to those drafting PCT applications in US style). It goes like this: the claim is not clear in that it might or might mean X. If it does mean X, however, it lacks novelty over D1. (Americans would typically reply to the effect that the claim is perfectly clear; all you have to do is read it in the context of the description). Over here, laughs all round. Ha Ha!

Am I not understanding your question correctly? Would you like to clarify?

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

I can reassure you, but my comment needs some explanation. The decision of the ED starts with first considering that claim 1 lacks N over D1 and D2 whereby it mentions in the reasons that both contain an “ALC control”, but without properly determining any possible meaning or lack of meaning.

It is only after having decided that claim 1 was not novel, that the ED in a second part, considered that the claim was lacking clarity in view of the “vague and unclear” feature “ALC Control”.

In its decision the board notes: In D2, the term “ALC system” is indeed used in the sense of an “automatic load change”. However, according to the applicant/appellant, this system is precisely not to be regarded as an ALC control in the sense of the invention.

The reasoning of the ED can be summarised as follows: “ALC control” is not clear, but claim 1 is lacks N over D2 which contains the feature “ALC control”, whereby it can be added that it is not the same as that in the application according to the applicant. This kind of reasoning .is rather difficult to follow. The lack of N objection can be said to result only from a literal comparison of ALC control in the claim and in D2.

It would have been enough to decide that “ALC Control” and to refuse the application under Art 84.

The decade long practice of the EPO to first decide, when a feature is unclear, to give it a certain interpretation and on the basis of this interpretation the claim lacks N or IS. It is always more efficient for an EPO examiner to raise an objection of lack of N or IS, rather than an objection of lack of clarity which can easily end up in a deadlock as the two opinions are not to be reconciled. In my experience, an objection of lack of N or IS makes applicants move.

I hope I could clarify my position. It has not changed over the years.

Yesterday I met US patent attorneys, and the message I got was exactly what you said: look at the description, the inventor was in possession of the invention and nothing is unclear in the context of the disclosure. What they added is that this type of reasoning is OK before the USPTO and a patent can be granted, but it might not be so easy before a judge to accept.

Max Dreisays:

Thanks, Daniel. The interesting thing about the USA is that the law deems the patent issued by the USPTO to have a presumption of validity which can be displaced only by “clear and convincing” evidence to the contrary. But one ground of attack on validity, post-issue, is that the claim as granted is “indefinite” (what we call “not clear”). Just one more example of how different patent law is, in the USA.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Thanks for your comment about US practice.

If I understand you well, lack of clarity is a very pungent objection when it comes to litigation.

Patents granted by an examining patent offices are always presumed to be valid, be it on this side or on the other side of the Atlantic! In France under the old law of 1844, without examination, patents were granted SGDG “Sans garantie du gouvernement”, i.e. without any guarantee from the government. It is still the same today as you will never be able to sue a PO because it granted an actually invalid patent.

According to the IPQC patents granted by the EPO do not have the same quality as in early days.
When looking at some published decisions of the BA after opposition, it seems that the IPQC has a point, even taking into account that only a small number of patents are opposed and the opponent will make more efforts to find prior art.

Max Dreisays:

What I wanted to explain is that although lack of clarity is a ground of revocation in the USA, the attack on valdity succeeds only when it is supported by evidence that is “clear and convincing” which is closer to “beyond any doubt” than “more likely than not”. One can imagine that non-specialist patent judges are going to hesitate for a long time before they find the evidence of invalidity clear and convincing, contradict the expert Examiners at the USPTO, and condemn as indefinite a claim that the Patent Office has duly issued.

That’s the problem with Art 84, EPC, no? What is clear to one reader is not clear to another, and anyway, nothing is ever 100% clear. No wonder those who wrote the EPC excluded Art 84 as a ground of invalidity.

As to how much time and money the EPO should spend on searching, it’s a balance, isn’t it? Only a tiny proportion of issued patents are disputed. Must the EPO search every single patent application so thoroughly that a competitor with deep pockets can never, as Opponent, find any prior art as good as (or better) than that already brought to light by the ED? What would holding to that standard, on every single patent application, do to the levels of fees that the EPO would have to charge applicants, in order not to go bankrupt? Searching the art is ever more time-consuming. It seems to me more or less inevitable that well-funded Opponnents will be relying increasingly on prior art over and beyond what the ED found.

Which in turn means that those defending opposed patents will have to amend. Which results in Art 84 and Art 123 EPC attacks. Which results quite often in the opposed patent getting revoked, because the standards which the EPO case law sets under those two EPC Articles is so hostile to patent owners. One would hope that the EPO from now on, concentrates on the primary task, of upholding claims to good inventions, useful contributions to a technological art, even while it swiftly condemns claims directed to non-inventions, recognising the reality that the patent drafter (or ED) hardly ever has sight of all the art that an Opponent will invoke, after grant.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Having been myself a search examiner I know too well that the question as to when to stop the search is crucial one. It is manifest that after a certain time of searching, the probability to find a relevant document is rapidly decaying.

Nowadays electronic searches are actually rather fast when taken as such. The difficulty is to find the proper search query. The intellectual input for an electronic search is much higher than for an old fashioned paper search or its equivalent on a screen. When searching electronically, you are like a fisher on the side of a large lake comprising millions of documents. You have to find the right document which can be compared to a fish in the lake. Like the fisher you are dipping the rod with a bait at the end of the line. Nowadays most of the time is needed to properly define the bait for an electronic search.

I agree with you that only a tiny proportion of issued patents are disputed. At the EPO around 5% of patents are opposed. The number of patents ending in court is a magnitude lower. But as a former president of the EPO once said: a patent is a kind of insurance in case of success. But for this you need well searched and examined patents.

The problem is that presently, the number of patents revoked or maintained in amended form is way higher than the number of oppositions rejected.

What is it good for the proprietor to have had a patent granted which will not resist a deeper scrutiny? When looking at patents revoked or maintained after appeal, you can see lots of search reports with numerous X or A documents. A good search is not necessarily a search in which lots of X or Y documents are found. A search ending with only few A documents is not necessarily a bad search, but when it happens in lots of cases, one could start becoming weary.

When during opposition, the opponent comes with a document in one of the IPC classes of the patent is classified and which is N destroying or which represent a CPA much closer than that used in examination, you can honestly raise questions about the search. The same applies when the opponent comes with PA under Art 54(3), sometimes from the proprietor himself. Those cases are more frequent than you might think.

The number of cases in which the opponent comes with documents which could not be present in the search files of the EPO is very low. I am thinking here of public prior uses, oral disclosures and PhD theses. For the latter documents, you might need deep pockets to find them, but not for documents which are to be found in the classes in which the application/patent has been classified or in which the search has allegedly been carried out.

It does not seem more or less inevitable that well-funded opponents will be relying increasingly on prior art over and beyond what the ED found. It is only in a low proportion of the cases that this is actually the case. I do not blame examiners, they have to do what they can in the time allotted to them to carry out search and examination. The time to prepare the “bait” is limited and hence they do what they can.

Art 84 attacks are not possible in opposition. I do not have an actual figure, but the number of cases revoked under Art 123(2) is not as high as you think. May be a third of the cases. When the revocation is on Art 123(2), the ED/OD/BA apply the rules set by the EBA. They have no choice.

In case of a problem under Art 123(2) I would claim that the responsibility is shared between the ED and the applicant/proprietor. The ED should have spotted the problem (again within the time allotted) and the applicant should have checked more than once that what he got was worth more than the paper it is printed on. Most of the revocations are mainly for lack of N and IS with documents present in the search files. In some areas lack of sufficiency can also play a role, but in the absence of experimental data a lack of sufficiency is difficult to demonstrate.

In your opinion, the EPO should concentrate on the primary task of upholding claims to good inventions, useful contributions to a technological art. I can agree, but only to a certain point. It should not grant dodgy patents which fall into bits and pieces at the first contrary wind they meet. In the patent system, there is one player which is rarely mentioned: the public. The public, or society at large, has to be protected from undue monopolies. This is also part of the role of a granting patent office.

francis hagelsays:

The views of the Board in this decision are in line with what Mr Thomas has called the second line of case law and are generally welcome.

I have a comment though. The Board’s assertion that « the patent claims should be clear in themselves in such a way that a person skilled in the art does not need to consult the contents of the description in order to understand their subject-matter », while exactly quoting case law II.A.3.1 1st paragraph, should not be understood as denying any reliance on the description since this same paragraph, last sentence, states : « The description is taken into account for the purposes of interpreting the claims ». The term « interpreting the claims » here seems to merely mean « understanding the claims », as clarified by case law II.A.6.3 mentioned in this last sentence, which states : « the meaning of terms in a patent claim has to be determined from the point of view of the skilled person, who reads the claim in the context of the application ». This is sensible. When reference numbers must be mentioned in the claims, what is the purpose, if not helping in the understanding of the claims ?

On the question raised by Mr Thomas of the patent owner’s expectation as to the validity of the patent when reviewed in opposition or court proceedings, I agree that a review takes place for a small proportion of patents, ca. 5% for oppositions (but more than 5% in highly competitive areas). Add to that the cases subject to Article 115 observations, a few thousands per year. This is not so negligible.

In addition, a patent owner or applicant, depending on the context, may have strategic objectives which go way beyond the exclusive right conferred by the patent : freedom to operate for exploiting companies (generally rated as a major objective in industry surveys), marking its territory in R&E collaborations, strengthening its negotiating position in technology transfers, credibility, reputation, management incentive for innovation. This is why, even there are validity issues, patents and pending applications may bring about significant benefits businesswise.

Max Dreisays:

Daniel, your post of April 29 prompts a raft of comments, from somebody who advises corporations who request patents from the EPO but also, from time to time, ask the EPO to revoke a patent that it has issued to a competitor. One challenge I have is to provide the client with legal opinions on their “freedom to operate”, by which I mean whether or not to oppose the newly-granted European patent of a competitor. Opposition is a good idea only if it will remove the threat. An opposition that does not succeed serves only to increase the presumption of validity of the issued patent. No wonder then, that most oppositions succeed. That is an indication that the opposition system is working properly.

As to the grounds of invalidity that succeed, it is a tribute to the EPO if Opponent does it with prior art already on the EPO file. That shows that EPO searching is of uniformly high quality. It is a mistake to see such opponent success as a failure by the ED. An Opponent often has huge advantages over an EPO Examiner, when formulating reasons why D1 does in fact prejudice patentability. An Opponent knows the common general knowledge of the skilled person, and can demonstrate it with evidence. As you say, insufficiency can be difficult to prove if you can’t generate more experimental data. An Opponent can gather experimental data but not an ED. And so on. If an ED cannot prove that the claim under examination fails to meet the provisions of the EPC it ought to allow the claim. Or do you disagree?

Helped by the Opponent, presenting “the other side of the story”, an OD can see more clearly the issues decisive to patentability than can an ED. You agree, I hope. Again, if oppositions succeed that is in itself no evidence that the EPO is not “doing its job” competently.

On Art 84, again, it seems, I have failed to make my point. One situation I have in mind is opposition, with the Opponent forcing a narrowing of claim 1. As ever, the drafting of the application was not perfect, and to narrow the claim without offence to Art 84 or 123(2) the only literal wording of the amended claim available is one that fails to read on to Fig 2. Let’s take the Catnic lintel case as an example. I narrow claim 1 in opposition by adding into it that the member is “vertically extending” and Fig 1 indeed shows that feature. Fig 2 shows the member extending at 80° to the horizontal. From Catnic, we know that this is an infringement by equivalent. But the OD requires strict conformity and so refuses to approve the narrowed claim until the description is amended to state that Fig 2 is “not the invention”.

Is this right? Is this proper? Is this fair to the inventor and patent owner?

Next question: same narrowing amendment but made prior to grant, before an ED instead of an OD. Do you still say that my predicament is all my own fault, because of incompetent drafting and that I deserve the outcome, that in due course the courts will feel obliged to find the 80° “embodiment” NOT to infringe by equivalent?

Just to reiterate, what I am advocating is NOT that Fig 2 is said in the conformed description to be an “embodiment of the invention” and NOT that it is NOT according to the invention. Rather, Fig 2 is introduced in the description as showing “another design of lintel”. That much is true. That statement is not contradicting the claim. It is not unclear. Why then is it failing to comply with Art 84, EPC?

I do understand, Daniel, that you consider this topic already fully explored and that further discussion will bring no more clarity. I shall thereforenot be upset if you see no value in writing a reply. But I will nevertheless be a bit disappointed. Best wishes, MD

francis hagelsays:

Dear MaxDrei,

May I venture my reaction to your interesting scenariThe interpretation of “vertical” and whether an 80° angle from the horizontal is an equivalent to “vertical” is clearly a matter for the courts. It would be an utter overreach for the EPO to try to impose on the patent owner a description/drawing amendment to exclude the Fig 2 embodiment.

In addition, the understanding of the claim by the skilled person relies on the application as filed, consistent with the principle that the relevant date for validity assessment is the filing date. Substantial amendment to the content of the application risks being held new matter by a court because it would post-filing information which could alter the interpretation of the claim.

In my view, the principle should be that since art 84 does not apply post-grant, the use of art 84 in relation to post-grant claim amendments should be limited to the strict necessary, i.e. if the amendment really makes the terms of the claim unclear.

For your consideration, here is a symmetrical scenario : An alternative embodiment is disclosed but not claimed in the original application. The claims as filed are accepted by the ED. Since the claims have been amended, there is arguably no need to “adapt” the description by deleting the alternative embodiment, or stating it is not part of the invention. Would the EPO feel nonetheless entitled to require such amendment ?

In this scenario, another question is whether the unclaimed embodiment would be eligible as an equivalent before a court. But this is another story.

Avatar photoDaniel X. Thomassays:

As far as the discussion 80° vs. 90° is concerned, I refer to my reply to Max Drei.

It all depends what has been originally disclosed and the explanations given by the applicant during examination.

It is not the duty of an examiner to decide what is an equivalent or not. this is for a court. He can only decide on the basis of what is submitted in the originally disclosed application and during examination.

If the claims have not been amended, it should be easy for the applicant to show to the ED that what has not been claimed falls under the claim. Without any explanation, the ED is entitled to request amendment of the description.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

I can agree with you that an opposition should never be filed leisurely. It can remove a threat for a present item put on the market by the opponent or for a future item to come onto the market. But it also draws the attention of the proprietor towards the opponent and the items he produces and puts on the market.

I can also agree with you that an opposition that does not succeed serves only to increase the presumption of validity of the issued patent.

I do not necessarily agree that, if oppositions succeed it is in itself no evidence that the EPO is not “doing its job” competently and that the opposition system is working properly.

I do not wish to repeat what I said in my reply of April 29th, but you seem to have overlooked what I said there and the examples I have been giving.

When you look at lots of decisions of the BA after oppositions, you realise a few things:
– There are only very few successful oppositions based exclusively on documents mentioned in the search report
– Prior art not available in the EPO’s search file is not very frequently the reason for revocation or limitation. My guess is that it happens in less than 10% of the oppositions.
– Most of the prior art submitted by the opponent is actually to be found in the search files of the EPO. This is the case when the opponent comes with highly relevant documents in spite of the search report quoting lots of X or Y documents, which on top are to be found in the classes in which the search has been carried out and not in “exotic” classification units. Opponent’s come regularly with prior art under Art 54(3), sometimes from the proprietor himself. When looking at all those elements, it appears legitimate to question the search. this is often due to the fact that the search is stopped when allegedly X documents are found. It is only after those documents have been reviewed by the applicant that they can be confirmed in their category. But this does not mean that there are no further relevant documents to be found. If a claim is limited by features from the description you rarely see further documents brought in the procedure by the ED.
– Common general knowledge is rarely the reason for a patent getting revoked or maintained in amended form.

I repeat that I do not blame examiners, they have to do what they can in the time allotted to them to carry out search and examination.

As far as decisions of OD decisions are concerned, it seems here that roughly half of those decisions are set aside by the boards. This looks like an alarming fact. Using your own words, an OD might be presented “the other side of the story”, but when so little decisions withstands an examination by a board, this is not encouraging in spite of the creation of specialised “opposition directorates”.

I do not want to overrate the role of an opponent and the significance of the figures in opposition. It nevertheless turns out that the number of patents revoked which was in the past similar to that of the oppositions rejected or the patents maintained in amended form is now higher than the number of patents maintained in amended form and way higher than the oppositions rejected. And this mainly with prior art which was not found during the search.

I realise well that under those conditions it might become more difficult to give a reasoned opinion as to the “freedom to operate”, but it is difficult not being bothered by the complaints voiced by the IPQC.

As far as Art 84 is concerned, the EPO is not to tell what a national judge has to consider as being equivalent. If with 80° the effect is the same as with 90° this can be explained to the examiner if it was not originally disclosed. But stating simply that 80° is the same or has the same effect as the claimed 90° and merely represent an other embodiment goes a trifle too far for my liking. Why should the examiner swallow such a statement? Drafting is never perfect, but there are limits to what is acceptable. Furthermore, how do you want to write a reasoned freedom to operate opinion is there is such an unexplained uncertainty in the description?

francis hagelsays:

Correction of a typo : The claims as filed are accepted by the ED. Since the claims have not been amended,..

Max Dreisays:

There seems to be a difference of opinion between Hagel and Thomas on what a diligent ED should do about an aplication in which claim 1 recites “vertically extending”, Fig 1 shows extending at 90° and Fig 2 shows extending at 80°. Should it require an Applicant to “conform” the description to the claim by amending the description to state explicitly that Fig 2 is NOT an embodiment of the claimed invention.

F Hagel declares that this would be ED “overreach” whereas D Thomas (if I understand him correctly) declares that to allow the application without conforming the description to state clearly and explicitly that Fig 2 is not an embodiment of the invention would be going beyond what is “acceptable”.

Is that it, gentlemen? Or have I failed to understand one (or both) of you?

Daniel, as to searching, I accept your points, that: 1) ED’s could deliver better quality search reports, if they were given more time, 2) OD’s making their decisions on the basis of art that could and should have been found prior to grant is not a good look for the EPO, and 3) the rate of reversal of OD decisions by a TBA is also a quality metric which should weigh heavily with EPO management.

In defence of EPO management though, I would think that reversal of a national court by a national court of appeal is more to be regretted than reversal of an administrative instance by the first judicial instance that gets to consider the case. Isn’t it quite common, still, for a national appeal court to reverse a national court of first instance? That’s not good either, is it?

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

Making a difference between an administrative instance and a first (and only) judicial instance at the EPO appears to me rather semantic. When looking at R 115, 116 and Art 116, I fail to see a difference between an administrative and judicial instance.

It is only in Art 12(2) RPBA20 that the notion of “judicial review” has been expressed. I am not sure that the fathers of the EPC ever thought of making such a difference between the two instances. The notion of appeal that is of a revision instance was introduced in the EPC, but I am not convinced that its fathers thought much further.

A court, i.e. a judicial instance, should be free of any influence from the executive. When looking at G 3/19 and the saga of G 1/21, doubts are permitted. On the paper the BA look “independent” and their transfer to Haar was there to increase the perception of their independence.

If a revision instance, whatever label you put on it, revises an inordinate lot of decisions of the first instance, something seems to have gone wrong. That for a given case, there can be differences of opinion is nothing unusual, but when it goes, let’s say above a third of the cases, something seems wrong.

As far as Fig 1 is showing 90° and Fig 2 showing 80°, it all depends on what is in the claim, of the problem the applicant wanted to solve as well as the effect thereby achieved. If the claim states 90° and the applicant is not able to show that 80° has the same effect than 90°, then 80° does not fall under the claim for 90°. My question is then: why is the applicant then not claiming the range 80° to 90°? If he sticks to 90° in the claim it must have good reasons to do so.

Max Dreisays:

Daniel thanks but I am still not getting satisfaction on my 90/80 example. I didn’t draft the application but I am prosecuting it at the EPO.

Original claim 1 recited “a support member”. D1 (in the EPO EESR) renders that claim unpatentable so I add claim 2 into claim 1. Claim 2 states that the support member is “vertically extending”.

Turning to the description, Fig 1 shows it vertical and the text declares that Fig 1’s support member is vertically extending. Fig 3 shows it horizontal so I delete Fig 3 and its description.

The dilemma is what to do about Fig 2 which shows 80°. The description of Fig 2 silent about whether its support member is “vertically extending”. I leave Fig 2 and its description unamended. I do not want to excise Fig 2 or declare it to be “not” the invention. How should the Examiner/ED react? What does “vertically extending” mean, what is the subject matter for which protection is sought, what is the scope of protection given by the claim, what is “the invention” and, above all, what is “not the invention”?

You might say that my example is contrived and artificial but I have the feeling when I conform the description, day by day, that such dilemmas are not uncommon at all.

francis hagelsays:

Your example is a real-life example, there are plenty of such examples in real files.
My comment is a question : is it arguable on the basis of the description that the 80° embodiment achieves the same function/effect as the « vertical » embodiment ? I agree here with Mr Thomas. If so, why delete it ? You have the right to argue that « vertical » does not have to mean « strictly vertical ».

The idea of a 80°- 90° range seems to raise a new matter issue.

As to the distinction between administrative and (quasi)judicial instances, there is the position vs the Guidelines. The Guidelines are binding on the EDs, in practice if not entirely in theory. They are not binding on the BOAs.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

You might not getting satisfaction on your 90/80 example, but “vertically extending” means in the broadest sense, being orthogonal to a direction parallel to a base line which is not itself vertical.

If the description is silent about 80° and the possible effect to the deviation from vertical, I see no other possibility to decide that 80° does not fall under the claim which is clearly directed to 90°. Claiming only 90° is a deliberate choice of the applicant during examination, or a necessity in view of the prior art brought in by an opponent.

It is for you to convince the examiner that the principle established for 90° works also for 80°. Only then, you might be allowed to say, directly or indirectly, that 80° falls under the claim. If you cannot bring this proof, the logical consequence is that 80° does not fall under the claim. There is no way out of it. The examiner is not there to endorse or envisage further theoretical considerations which might be apparent for the applicant. The same applies later in front of national judge.

I agree that in such a situation you might be faced with a dilemma, but the cause of the dilemma is not the EPO and the examiner. An applicant is free to formulate its application/patent as he wishes, but it can also have not so pleasant consequences for the applicant/proprietor. In English, I would say, you cannot have your cake and eat it.

Max Dreisays:

@FrancisHagel, thanks for that. I don’t yet see an answer from Daniel Thomas but I suppose that what he will argue is that it is the duty of the Examiner to raise objection under Art 84 to “vertically extending in claim 1, for the reason that it is not clear whether “vertically extending” has a meaning wide enough to include 80°. Either yes or no. If No, delete Fig 2 (or declare that it is NOT the invention). If yes, and Fig 2 remains, what other embodiments might there be? Applicant must therefore make it clear how much beyond 80° is “vertically extending (because the purpose of Art 84 is to ensure that competitors know whether or not they infringe).

There is an answer to that, but let’s first see what M. Thomas writes.

As to an answer to your question, as you say, the example is one from real life and I was supposing that lintel engineers gave evidence to the court in England, when the Catnic case was litigated, that i) 80° works in the same way, and just as well, as 90° and that ii) readers of the patent would derive exactly that, from their study of the specification.

francis hagelsays:


If the ED required the deletion of the 80° embodiment from the description/drawing or the addition of a « this is not the invention » notice, you could brandish the 123(2) flag, with the argument that the amendment would potentially alter the interpretation of the claim by a court, specifically as to whether the 80° embodiment can qualify as an equivalent.

An applicant is entitled to resist an ED’s requirement for « adaptation » of the description based on a clarity objection if the adaptation would entail a 123(2) issue and the striking down of the patent by the court. It cannot be the duty of the ED to open a 123(2) issue. And the applicant could argue that the assessment of equivalents is none of EPO business and is only for the courts, in other words, the ED’s requirement would be outside the EPO’s remit. Again, it cannot be a duty for the ED to venture beyond the EPO’s remit.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

I could not reply earlier, but now you got my explanations.

I will most probably not have convinced you, but you will by no means have convinced me.

Catnic is a good example of what a court can decide much later, but without any further evidence, vertical means 90° and the examiner has no choice to react to 80° as it does.

I have however great difficulties to see that, in the circumstances described, marking 80° as not belonging to the invention, Art 123(2) can at all be infringed. The interpretation of the claim might be different, but there is no new information brought in by the amendment so to me Art 123(2) is not available. Interpretation of the claims is a matter for national courts and not for the EPO in spite of some decisions to the contrary wanting to bring Art 69 into examination.

The EPO is not to venture in grounds reserved to national courts, but it cannot accept some statements in an application/patent which might lead to theoretical considerations for which ab applicant is not even aware when filing his application. With a reasonable explanation, yes, but on the simple assumption of a possible possibility the answer is clearly no.

With those last two comment, I consider having taken due care to the comments of Mr Hagel.

Max Dreisays:

Ah, there we go, Daniel. In the Catnic case in England, long years ago, there was no Doctrine of Equivalents. The court, interpreting the claim purposively, found that “vertically extending” meant “in substance vertically extending”. The skilled addressee would not uinderstand “vertically extending” to exclude 89° or 91° or 80°. Don’t EPO Examiners construe claims purposively, with a mind willing to understand? Or does “vertically extending” to them exclude 89° and 91°? How about 89 1/2 degrees? In the context, the court found, 80° was “substantially” vertically extending.

Or must I these days preface every feature of my claim with “substantially”? Or (to ensure I cover 90° exactly) must I write “vertically extending and substantially vertically extending”? Depending how many features I recite in my claim, the claim soon becomes unreadable, very far from “clear” for its reader.

When I trained, in the 1970’s, I learned that judges read “substantially” into every feature of claim, unless the context requires otherwise. And therefore it is counter-productive to write “substantially” in front of every feature recited in the claim. Has that changed, these days, for the EPO?

Otherwise, for what it’s worth, I agree with you and doubt F. Hagel, on the 123(2) point.

Avatar photoDaniel X. Thomassays:

Dear Max Drei,

When a numerical value is given in a claim, it means that this numerical value is measured with the measuring instruments usual in the field, whereby the normal rounding rules apply.

This is why you do not have to specify “ca 90°”, “about 90°” or “substantially 90°”. This is all superfluous.

Removing “about” from the claim would not infringe Art 123(2), see T 875/11, Reasons 1.1.

I would not see any problem with 89° or 91°, even 89,5°, but to me 80° is quite different. Without any further explanation, even looking at the claim purposively, with a mind to understand and not to misunderstand, you name it all, 80° is not falling under the claim mentioning 90°.

If the court finds 80° falling under the claim, it is its good right.

You can however not require the examiner simply to accept 80° without any further explanation or information.

I therefore repeat my question when it works with 80° why not claim broadly 80° to 90°, or even 80° to 100° and then preferably 90°. But 90° in the independent claim and 80° in the description without any further explanation is a not acceptable.

What you have learned in the 70s has not changed at the EPO.

Max Dreisays:

An unfortunate misunderstanding, Daniel. Who said anything about a numerical value in the claim? Only you.

The claim we are debating does not say “90°”. I don’t know how you came to that idea. The claim says instead “vertically extending”. That is not the same as “90°”. Consider an umbrella, for example, it has a canopy supported by a vertically extending stem. That stem is never at exactly 90° to the horizontal.

Or are you saying that at the EPO an ED has no choice other than to give “vertically extending” the meaning “90° to the horizontal, not one degree more, not one degree less”? It is exactly this point that the English court addressed, and answered in the same way you did, by asking what the term meant to one skilled in the art. The difference is that the court had the advantage of witness evidence given under cross-examination by the respective expert witnesses put forward by the opposed parties. Perhaps you are saying that at the EPO an Examiner knows better than an English judge who has already heard both sides of the story from two different expert witnesses who have both sworn to the court that they understand that their duty is to the court (and not the party paying their fee) to tell the truth, the whole truth and nothing but the truth.

Avatar photoDaniel X. Thomassays:

Max Drei,

Enough is enough and my patience is at its end. There is no unfortunate misunderstanding whatsoever. When you say “extending vertically” and bring in the value 90, any person willing to understand and not to misunderstand what is meant has to conclude logically that the value at stake is the numerical value 90°. Anything else is blatant bad faith.

Rather than answering the question I have raised with the possible ranges, you come yet again with a further example which is of little value: the umbrella. Depending on the diameter of the umbrella you might be able to hold at a different angle than the vertical up to the moment the umbrella will not fulfil its function which is to protect from the rain. What do you want to demonstrate? I am at a loss to understand it and I can only repeat what I said before: no theoretical example can be used to define general rules. I will never enter again any discussion with you based on a theoretical example.

Your comparison with the Catnic case and the example you try to derive, falls short from the beginning. How can you come to compare a legal discussion in front a national court with the procedure before the EPO.

I have never said that at the EPO an Examiner knows better than an English judge who has already heard both sides of the story from two different expert witnesses who have both sworn to the court that they understand that their duty is to the court (and not the party paying their fee) to tell the truth, the whole truth and nothing but the truth.

What happened in the Catnic case occurred years later after grant! Wanting to transfer this knowledge to what can happen during examination is far from reality and anything but reasonable. You are neither an English judge! I understand that you want to defend an applicant, but there are limits not to be exceeded. In view of the EPC as it stands examiners have no choice than to act as they do when it comes to the adaptation of the description, whether you like it or not.

Max Dreisays:

Daniel I am sorry you feel so upset. I accept that a reply from you will not be forthcoming.

For other readers though:

My error was to suppose it unnecessary to write in my post that in the application as filed Fig 2, with the support member at 80°, is presented as an embodiment of “the invention”, the one claimed in the application as filed. That makes sense because it functions in the same way as the Fig 1 embodiment at 90°, and delivers the same result.

Then the application gets scrutinised in the EPO. The Examiner cites D1 with its support member extending horizontally, and correctly observes that originally filed claim 1 lacks novelty.. Then comes the claim amendment, adding claim 2’s “vertically extending” into claim 1. Any other amendment will offend Art 123(2) EPC.

Clearly, Fig 1 is still an embodiment of “the invention”. But what about Fig 2, and 80°? It is not clear to me that it is no longer an embodiment of the invention as now claimed, with its requirement that the support member be “vertically extending” because a term like “vertically extending” implies a range of angles centred on 90°. The question is, how wide is the range? Who knows? Who shall decide? Meanwhile, I am reluctant to delete Fig 2 or declare that it is “not the invention”.

I am persisting in my comments because I fondly suppose that the dialogue with M. Thomas will help to to explore why so much heat is generated by the EPO’s insistence on ultra-strict “conformity” under Art 84, EPC. I had thought that the facts of the Catnic case might provide a useful matrix for exploring the issue further. Readers will judge whether that was a mistake. I did not want to be rude to the operator of this excellent and wonderfully educational and topical blog. All I wanted was to stimulate the Art 84 confomity debate. But perhaps we are all better off without any further stimulation.

Max Dreisays:

Afterthought: Drawings are schematic and, in my Catnic-based hypothetical, the description does not state what angle the support member has, in Fig 1 or Fig 2. It is just that, looking at the drawings, it seems that the one in Fig 1 has 90° and the one in Fig 2 has an 80° angle.

So what does Fig 2 show? Is it an embodiment of the claimed invention? Or is it something that is NOT an embodiment of the claimed invention?

I guess this is one danger with hypothetical examples: one never states at the outset every one of the crucial facts.

Avatar photoDaniel X. Thomassays:

As I explained before, no reply will be given here on theoretical examples.

Bringing in crucial elements in a piecemeal manner and leaving crucial facts to later are of no help whatsoever.
Also for other readers.

If there is something important to be said, then this has to be done as early as possible.
It is impossible to guess what is going on in somebody else’s mind.
What can be discussed is what is written in a clear and unambiguous manner.

Anything else is wasting time for all.

The discussion on adaptation of the description under Art 84, last sentence, is, for my part, closed.

francis hagelsays:

I came across recently the file of a granted patent in which the ED had deleted in claim 1, at the R71(3 stage, the word « substantially » preceding a geometrical definition of the shape of a device. The ED invoked GL F-IV 4.7 which bans such words as « substantially » and « approximately ». Since the description never discloses the strict geometrical definition, this opened a new matter issue. And the applicant did not challenge this claim amendment, while there was ample basis in the description for an argument relying on a technical effect of the geometry.

This is not entirely relevant to the adaptation of the description discussed in the previous comments, but it illustrates the new matter issues which may arise from amendments when the ED does not carefully check whether the amendment it intends to enter at the R71(3) stage is supported.

Avatar photoDaniel X. Thomassays:

Mr Hagel,

The question you raise with amendments in the R 71(3) communication is indeed, as such not relevant to the adaptation of the description. That it can have dear consequences for an applicant if he accepts the amendment is nothing new.

I can only reiterate that this form of “arm twisting” from the side of the ED is not correct and should not be accepted by applicants.

The Guidelines GL F-IV 4.7.1 explain clearly under which circumstances such a term can be deleted.
If it is merely to define a form within the technical tolerance of the method used to manufacture it, it can be deleted.
If this is not the case, it should clearly not be deleted.
For instance if error margins are specified in the application, they must be used in the claims in place of the expression containing “about” or similar terms.

In GL F-IV 4.7.1, I further note the following:
“For example, “a substantially vertical seat back” is interpreted as allowing for a certain +/- variation around 90° where the skilled person can recognise that a functionality for supporting the sitting person’s back is present”.

I would allow myself to conclude that without a functionality being present, 80° does not fall under a claim for 90°.
With or without Catnic…..

Leave a Reply

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