Application of G 1/24 by the Boards of Appeal of the EPO
G 1/24 has been issued on 18.06.2025. There are not a lot of decisions in which G 1/24 and the principles which it established are discussed, but it appears possible to derive some conclusions from those decisions.
This decision follows an appeal in opposition.
In Reasons 3.4.2, the board noted that, according to established case law, it is not permissible to read into the description any additional features implied by the problem to be solved. This has not changed as a result of G 1/24, which expressly refers to existing case law on the interpretation of claims in Reasons 10 of the decision.
G 1/24 can also not be used as a justification for the late filing of requests under Art 13(2) RPBA, cf. Reasons 4.4.1.
In Reasons 4.4.2, the board disagreed with the proprietor that the independent claim as granted must now be interpreted more narrowly in light of the description, with the result that claim1 is to be held as novel vis-à-vis the disclosure of D1.
This decision follows an appeal in opposition.
In Reasons 2.3, last two paragraphs, the board held that in accordance with G 1/24, , it saw no reason to deviate from the wording of the claim and cut down its scope by implying into it additional features which appear only in the description of a specific embodiment. This is particularly true given that the broader/unrestricted interpretation of the wording of the claim also makes technical sense and is also encompassed by the description as a whole.
Therefore, even after consulting the description and the drawings, and taking into consideration the embodiment of a dimension shown in Figure 4, the skilled person would in their act of interpretation of the language of the claim not find that feature X of claim 1 is restricted to the specific example disclosed in Figure 4 of the patent.
The lack of novelty over D1/D1t=JP 2005-334626, not mentioned in the ESR, was thus confirmed.
This decision follows an appeal in opposition.
In Reasons 2.4, when applying G 1/24, the board made clear that the description does not allow to give claims a narrower meaning and cannot be re-interpreted or understood in a more limited way in the light of the specific embodiments of the description.
The board also reminded that considerations of the EBA in G decisions are generally applicable, cf. G 1/19.
This decision follows the refusal of the application.
In Reasons 2.10, the board held that, G 1/24 did not define what it means in individual cases to consult the description and the drawings, but referred in this respect to the case law of the boards, cf. G 1/24, Reasons 1.
G 1/24 does not even expressly require that the definition of a term from the description must necessarily be used for the interpretation of a claim. The relevant question 3 was not answered as inadmissible, and it follows from the answer to question 2 that such a definition cannot be disregarded, since it must be taken into account as part of the description and the drawings.
In the present case, the board referred to the description and drawings in accordance with G 1/24 and explained why, in its view, this reference would not justify a narrower interpretation of the claim wording.
In particular, it took into account that the application does not provide a definition (‘or similar information’) for the disputed terms, regardless of whether or under what circumstances such a definition would then limit the claimed subject-matter.
The present application only gives examples of additional functions, but it cannot be deduced from this how “tasks” and “additional functions” would differ in concrete and fundamental terms from those of D5. Hence, claim 1 of the MR lacked novelty.
This decision follows an appeal in opposition.
In Reasons 1.5.4, the board held that, although the description and figures must be consulted when interpreting the features of the claims, cf. G 1/24, headnote, the description of a specific embodiment may be disregarded in accordance with the principles developed in case law with particular reference to T 2684/17, T 1871/09, T 1473/19, T 1465/23, T 161/24, and T 1999/23, cf. Case Law of the Boards of Appeal, 10th edition, II.A.6.3.4, and cannot, as a rule, be used to restrict the interpretation of an object claimed in more general terms.
This decision follows an appeal in opposition.
In his reply to the appeal, the proprietor referred to G 1/24, and argued that the subject-matter of the patent extends to what, after examination of the description and the drawings, appears to be the subject-matter for which the proprietor seeks protection, i.e. the will of the proprietor.
The board held that the intermediate generalisation was not allowable. This conclusion is not altered by the proprietor’s reference to G 1/24, Reasons 1.5.4. For the board, the description and claims as filed already express “the will” for the very reason that they were drafted by the applicant, and thus, there would not be any need for further speculation on the applicant’s the proprietor’s intentions, i.e. on what else might be the subject-matter to be protected.
Assuming arguendo that “the will” really was to be considered for examining the allowability of an amendment, this concept would identify multiple different wills in the present case.
The fact that in paragraphs [0040] and [0041] of the originally filed description some of the features are introduced by “may”, did not allow an arbitrary combination of features. The notion of “directly and unambiguously derivable” has not been superseded by G 1/24.
The Agfa/Gucci case before the UPC
The LD Hamburg – Order UPC_CFI_278/2023, found that the proprietor could not rely in infringement proceedings on parts of the description which were not any longer covered by the granted claim.
The granted claim had been limited, in order to avoid a novelty objection against the originally filed independent claim. The description had however not been adapted to exclude an embodiment originally covered, but removed from the scope of protection of the granted claim. .
This way of interpreting the claim by the UPC-LD Hamburg, is fully in accordance with G 1/24.
Conclusions to be drawn from the recent Board of Appeal decisions applying G 1/24 and of the UPC
It appears that applicants/proprietors want to use G 1/24 and to refer to the description, in order to read in a claim a different interpretation than that directly and unambiguously derivable by a skilled person from a claimed feature.
The way the boards interpret G 1/24, by referring to existing case law, as suggested in G 1/24, is nipping in the bud all those efforts of applicants/proprietors.
In other words, it is manifest after G 1/24, that the patent cannot be its own dictionary. This is a position which is clearly in contradiction with that of the German Federal Court (BGH). It will be interesting to see whether the UPC will continue to follow the line of the EPO or that of the BGH.
The notion of “directly and unambiguously derivable” has thus not been superseded by G 1/24. Any other interpretation of a claim would be contrary to the criterion of a direct and unambiguous disclosure, i.e. to the “gold standard” as defined in G 2/10.
Possible further consequences
Working within first instance divisions
One direct consequence is that G 1/24 will force EDs and ODs to look at the description before carrying out a documentary search and/or examining an application or an opposition, irrespective of whether a claim is clear as such.
In T 933/23, Reasons 3.2. in reference to G 1/24, Reasons 20, the board praised the ED’s scrutiny and diligence. The ED examined carefully whether the independent claim at stake contained all the essential features of the invention and did not merely formulate the subject-matter for which protection is sought in terms of the effects or problem to be solved (the so-called “result-to-be-achieved”). It also checked whether said claim was properly supported by the description.
It can be expected that the EPO Guidelines for Examination will be amended in due time, but it will not change fundamentally the approach of first instance divisions..
Guidelines B-III, 3.2 have not been amended since the early days of the EPO. According to this part of the Guidelines: “The search covers what appear to be the invention’s essential features and is adapted to take account of any changes in the objective technical problem solved by the invention that may later arise in view of the prior art found so far”.
In case the interpretation of a claimed feature, clear as such for a skilled person, receives a limiting interpretation in the description, the search and examination will be based on the broader interpretation given in the claim.
In case the interpretation of a claimed feature, clear as such for a skilled person, receives a broadening interpretation in the description, the search and examination will be based on the broader interpretation given in the description.
Following G 1/24, it simply means that the search has to be commensurate with the interpretation of the claims given in the description, but this will not mean that any inconsistency between claims and description, when it comes to interpreting claims, will be read into the claim. Such an inconsistency will have to be removed before grant or after maintenance in amended form.
Adaptation of the description to the claims
Although it is not part of the present topic, G 1/24 might have an influence on the decision to come in case G 1/25. In view of diverging case of the boards in this matter, in T 0697/22, Board 3.3.02, referred a series of questions relating to the adaptation of the description to the claims to the EBA.
It does not appear acceptable that
- on the one hand, a claim appears limited in scope for a skilled reader through its wording, but the description allows a much broader interpretation, and
- on the other hand, a claim appears broad in scope for a skilled reader through its wording, but the description allows a much more limited interpretation.
When a feature in a claim is clear as such for the skilled person, it appears contrary to the notion of primacy of the claims, that third parties have to dig deep in the description, in order to check, whether a a feature, clear as such for the skilled person, can receive a different interpretation in the description.
Any inconsistency between claims and description has thus to be corrected before grant or before maintenance of a patent in amended form.
Some adaptation of the description to the claims will always be necessary, for instance
- when some original embodiments do not any longer fall under the granted claims. Those embodiments do not have necessarily to be deleted, but will have to be marked appropriately;
- when the independent claim is limited with features from a dependent claim or with features from the description. In such a situation, the optional character of the corresponding features ought to be deleted.
It is therefore possible to say that G 1/24 has paved the way for G 1/25.
It is only with claims and description aligned with each other, that a common interpretation of claims before and after grant will be achieved.
This will however not hinder national courts and the UPC, to apply Art 69 and Art 1 the Protocol on interpretation of Art 69 EPC, to decide infringement, be it directly or indirecttly, by equivalence or contributory infrigement.
A correct alignment of claims and description will also avoid nurturing Angora cats.
A good example of what can happen is description and claims are not aligned is to be seen in case of EP 3 388 490 B1 before the UPC (Agfa/Gucci),see above. This decision has been quoted in T 0697/22, the referring decision in G 1/25.
Comments
Although lots of amicus curiae in G 1/24 wanted actually that the interpretation of the claims given in the description overrules the clear meaning for skilled person of a claimed feature, those people might be disappointed.
It appears to the contrary that the boards will not give up their usual way of interpreting claims and maintain the primacy of the claims.
It is to be noted that in G 1/24 it has not decided that a limiting or a broadening statement in the description should be substituted to the clear meaning for a skilled person of a claimed feature. In other words, G 1/24 has clearly not decided the patent can be taken as its own dictionary.
In lots of amicus curiae for G 1/24 this position was brought forward, directly or indirectly, as this position is often seen in decisions of the German Federal Court (BGH).
It should be reminded that decisions from national courts or from the UPC are not binding to the boards of appeal of the EPO, Conversely, the boards of appeal of the EPO cannot impose their views on national courts or on the UPC. See for instance T 722/24, commented in this blog.
If an applicant/proprietor wants to have a more limited or a broader interpretation of a claimed feature, he can bring the more limited or broadened interpretation from the description into the independent claim.
Applying G 1/24 will thus not give raise to any loss of scope of protection, it will just avoid third parties having to dig deep in the description to see what interpretation the applicant/proprietor wanted to give a claimed feature.
Comments
29 replies on “G 1/24, early following decisions and possible consequences”
Thus far, decisions applying G1/24 all seem to relate to situations where a proprietor tried to interpret a term in a claim based on an example or a specific embodiment disclosed in the description. I’m curious to see the BoA apply G 1/24 in a situation where a claim contains a (clear) term which has a dedicated, specific definition in the description, perhaps one that deviates from the usual meaning of the term…!
@ Anonymous,
Why should the way G 1/24 is applied to the situation you suggest, be different from the way G 1/24 has been applied in recent decisions?
If the applicant/proprietor wants the interpretation in the description to define the scope of protection of its claim, the solution is extremely simple: bring the interpretation/definition given in the description in the claim.
Why should third parties be forced to dig deep into the description to determine what the scope of protection of the claim should be? I do not hesitate in calling this way of hiding in the description what the scope of protection should be as cheating. This is not acceptable.
The situation is no different whether the dedicated, specific definition in the description is broader or more limited than that of the clear feature in the claim. By clear in the claim, I mean clear to the skilled person.
G 1/24 cannot be interpreted in allowing applicants/proprietors to hide information in the description which should be present in the claim.
G 1/24 has decided that the description is always to be consulted, and this is to be welcomed. for the rest, the EBA merely referred to existing case law. And this is what boards have done in their recently decisions.
G 1/24 did certainly not say that the definition in the description should overrule the clear meaning of a claimed feature. If this would have been wanted by the EBA, the latter would have stated it, directly and/or indirectly.
I am fully aware that this was the wish of lots of applicants/proprietors and their representatives, in other words to decide that the patent is its own dictionary, and what matters is the meaning given description but not the clear meaning of a claimed feature. At this rate, why do we then need claims?
Daniel, I love it when you dogmatically declare that “thus” ANY inconsistency (between description and claims) MUST be “corrected”. I don’t accept that. Yet. I think you haven’t proved your proposition, yet.
We both must wait and see what G1/25 has to say on the point.
My way of thinking is that if 1) the claim is clear in setting out the subject matter for which protection is sought and if 2) the description is that which Applicant thought was their invention on the filing date, then inconsistencies between 1 and 2 in the B publication are unremarkable, routine, normal, to be expected, when pre-grant search and substantive examination is thorough and robust.
Such inconsistencies do not get in the way of post-grant assessment of validity and infringement, as we see from the post G1/24 case law. Further, such inconsistencies help rather than hinder assessment of infringement by equivalent, under Art 69 EPC. Post grant courts can take them in their stride. Of that I am confident, based on what we see elsewhere in the world.
The EPO is the only one of the world’s Patent Offices which, up until now has been saved from the misfortune of being corrected by a supervisory court, and it is the only one that insists on strict conformity between description and claim. That’s not a coincidence. But with the advent of the UPC, that’s all changing. I look forward with interest to the decision in G1/25.
I could not have expressed this any better, Max Drei. I too look forward with interest to the decision in G1/25, and to the reactions and comments here on Daniel’s blog, which I always find extremely helpful in forming and modifying my own understanding. This is a great resource.
Dear HKJ,
Thanks for your comment.
I always considered a blog to be a forum for discussion. That we do not all agree on everything is good, as otherwise the patent world would be very boring.
The reason I published the present entry is to start a discussion, since in my opinion G 1/24 has given rise to lots of hopes which might actually not materialise, in view of the recent case law applying G 1/24.
Every opinion should be able to be expressed, as long as it is not done in an aggressive or defamatory manner. This is why there is no vetting before publication on the present blog, but I will not hesitate in deleting or editing comments which are simply attacking in an incorrect manner views expressed in the blog or in comments.
Dear Max Drei,
When I wrote my blog, I could have bet that I would receive a response such as the one given. There is nothing dogmatic about my statement, just a logical deduction from the situation. If there is an inconsistency between claims and description, it ought to be rectified, by aligning claims and description.
That you do not accept my statement does not come as a surprise as we have disagreed about this topic for a long time. I maintain my point of view. According to the interpretation of G 1/24 given by the bords, a different interpretation in the description of a clear claimed feature cannot overrule the interpretation of the claim by the skilled person. As I said to Anonymous, G 1/24 has not ruled that the patent can be its own dictionary.
It logically follows that, in presence of different interpretations of clear claimed feature between description and claims, there exists an inconsistency between claims and description. If the notion of primacy of the claim has any raison d’être, then, this inconsistency has to be resolved.
That the referring board in G 1/25 has quoted the decision of the UPC LD-Hamburg in the Agfa/Gucci case is not the result of a pure coincidence as the UPC CFI has clearly used the word inconsistency.
I agree with you that we have to wait what will be said in G 1/25. I do however maintain that G 1/24 has paved the way for G 1/25. Why leave an inconsistency between claims and description if the meaning in the description does not overrule the meaning in the claim?
That such “inconsistencies do not get in the way of post-grant assessment of validity and infringement, as we see from the post G1/24 case law”, is therefore beside the point. If the EPO grants a patent, with the aim of having the same interpretation before and after grant, it should make sure that there are no inconsistency between claims and description. In a comment in another blog, I called this a standard patent.
Removing an inconsistency between claims and description in no way hinders assessment of infringement by equivalent, under Art 69 EPC. Post grant courts are still free to decide on infringement by equivalent. It is also up to the proprietor to demonstrate infringement by equivalent, and not for a court to do the job of the proprietor.
That by removing any inconsistency, the EPO might not make the task of the proprietor easier, is acquired, but why should prima facie the proprietor benefit from an inconsistency between claims and description. Just look what Agfa tried to do.
Furthermore, it is interesting to look at the way the USCK has introduced infringement by equivalents after the disaster of the “Improver” decision. There was no G 1/24 at this time and the claim was clearly limited to pemetrexed disodium.
The opinion and case law of a national or supranational court might change, and you would like the EPO to follow all those changes?
Indeed, the “EPO is the only one of the world’s Patent Offices which, up until now has been saved from the misfortune of being corrected by a supervisory court, and it is the only one that insists on strict conformity between description and claim. That’s not a coincidence”. This is what makes the difference between the EPO and any one of the world’s Patent Offices. The EPO has to insure that, prima facie, the interpretation of a clear claimed feature cannot be altered by statements hidden in the description.
As far as the UPC is concerned, I do not think that’s all changing. Before the UPC we had national courts deciding on patents granted by the EPO, and neither the EPO could impose its views to national courts nor national courts impose their views on the EPO.
The UPC does not issue judgements valid for all EPC contracting states, although it would like to see it so, and makes any efforts possible to have a long-reach arm. The last word about this long-reach arm is not yet said, as the UPC is no more than a giant on very wobbly feet. It certainly cannot impose its views on the EPO and its boards for lack of a legal basis.
I also “look forward with interest to the decision in G1/25”.
Daniel, thanks for trying. I noted in particular your assertion that:
“The EPO has to insure that, prima facie, the interpretation of a clear claimed feature cannot be altered by statements hidden in the description.”
For you, it might seem self-evident, indisputable, that the EPC leaves the EPO with no alternative but to impose this duty on itself. But for me, despite your argumentation, it still isn’t.
After grant, a court cannot escape exhaustive study of at least i) the claim as granted and ii) the specification originally filed. To burden the court with a third document, namely the strictly conformed description, is to increase its workload by 50% at least. I am not yet convinced that it is wise of the EPO to impose so much extra burden on the courts. I am far from convinced that “the public interest” demands it.
The more you defend your position, the more ideas you give me, how to defend mine. Fun, isn’t it?
Dear Max Drei,
It is indeed fun as I can send back your own statement: The more you defend your position, the more ideas you give me, how to defend mine.
The reasons for a civil court to always have long discussion on the interpretation has been given by a former LQJ of the German Federal Patent Court (BPatG): having no, or hardly none, knowledge in technical matters, they are obliged to do so, as otherwise they could come to the wrong decision. This is why at the UPC a CFI can add a TQJ to the normal panel and in the CD which deals with revocation, a TQJ is always member of the panel. At the CoA two technical judges should be present, but the CoA has taken the right to decide with only three LQJ.
It seems to me that, probably due to your numerous US clients, that you are impressed by the US system. What you suggest corresponds to the notion of file-wrapper estoppel and no amendments to the description once filed. This is not the European system, and I will never get tired to say with the EPC as it is, the EPO has to do what it is doing.
Even the UPC does not follow your views.
In the famous Vusion/Hanshow case about EP 3 883 277 B1, the UPC-LD Munich wanted to interpret the claims starting with the claims as filed and to see how they have been amended during prosecution, cf. UPC_CFI_292/2023 of 20.12.2023. The CoA UPC, cf. APL_8/2024/UPC_CoA_1/2024-13.05.2024, decided that the interpretation of the claims is not to be done according to amendments during prosecution, but on the basis of the description, as stated in Art 69(1) and the protocol. In both instances the PI was refused.
In the famous Agfa/Gucci case, UPC_CFI_278/2023-30.04.2025, it was the description as granted which was determining and not the application as filed. This is also why the CFI-LD Hamburg saw an inconsistency between the claims as granted and the description as filed, as the latter had not been amended.
What matters in case of infringement or validity is the patent as granted as a whole, i.e. the description AND the claims as granted, and not what was there at the beginning of the procedure.
I do not remember seeing a decision of the German Federal Court in which the application as granted was taken into account besides cases of added matter.
In general, I have not yet seen that a national court resorting to a comparison between the application as filed and the claims as granted in order to interpret the claims. The contrary is true.
I can thus not agree at all with you that in order to assess validity and/or infringement, the court is burdened with a third document, i.e. the description as granted. It is rather your plea to look at the description as filed which represents a third document and would increase the workload of a judge. Only in cases of added matter it is necessary to resort to the original description, but this is not an extra burden, it is a bare necessity.
You will need much more efforts to convince me.
ADDENDUM
Prompted by L.T. to look at the decision BGH-Rotorelemente“-12.05.2015, Az.: X ZR 43/13, I discovered the following passage (free translation):
The content of the original documents or the publication of the application shall not be taken into account in the interpretation. Neither may the patent claim be interpreted in accordance with the original disclosure in order to avoid an inadmissible extension (BGHZ 194, 107, para. 28 – Polymerschaum I), nor, conversely, may its meaning be determined by comparing the wording of the patent claim with formulations in the application that deviate from it. At most, if it remains doubtful whether the patent claim and the description can be meaningfully related to each other, the ‘claim history’ may be used to further clarify the question of whether the claim protects an object that deviates from or falls short of that disclosed in the description (BGHZ 189, 330, para. 25 – Occlusion device; BGHZ 194, 107, para. 28 – Polymer foam I).
I would say the position of the BGH is abundantly clear: even for determining whether there is added matter, the BGH is adamant that the original disclosure is not to be taken into account. The door was left open in case of claim and description being at odds, but I do not associate this point with the statement on added matter.
This position is somehow surprising as I have always understood that, in order to determine the presence of added matter, the comparison ought to be made with respect to the original disclosure and not with the patent as granted.
I might have missed something, but I am not sure.
Well, Daniel, first I have to say that this dialogue gets more amusing with every exchange. When I read that bit about the BGH stating that for the consideration of added matter the description as originally filed is irrelevant I laughed out loud. Surely we both must be “missing something” here in the pearls of wisdom handed down by the BGH.
Having retired after more than 50 years in the patent business, I no longer have any axe to grind. I just want to see the good work done by the EPO since 1978 preserved, and patents in Europe litigated just as competently as the EPO has handled the patent grant process. I see the difference between a technical judge and a non-technical judge, even though there is in the English courts no such distinction. I worry that the litigation industry has an interest in making patent litigation as complex as possible and I am trying to visualise a way forward that reduces uncertainty and complexity to a minimum. Is anybody else lobbying in that way, I wonder.
We can use the pemetrexed and the “gathered” cases to think about the pro’s and con’s of conforming the description. Suppose the “gathered” case goes to grant with an unconformed description. The court can choose between declaring that the claim lacks novelty or declaring that the accused embodiment does not infringe. Those writing an FTO opinion can confidently tell their clients that the patent owner will not succeed, whether or not the description gets conformed.
Pemetrexed is more uncertain. If the accused embodiment was di-potassium instead of the claimed di-sodium, Art 69 EPC comes into play regardless of whether the description was “conformed” or not. Pemetrexed is a super example of a poorly drafted case that was destined for the courts as soon as its subject matter became a blockbuster.
The jurisprudence of the courts in Europe up until now is based on the assumption that the patents issuing from the EPO have had their descriptions strictly conformed. Court proceedings and court decisions follow that assumption. The litigation industrry lobbyists like it that way, because it increases the complexity of the cases in which they are engaged. Ivt doesn’t have to be that way but I see very little prospect of any change. But, Hey Ho, never mind. One can always dream.
I am surprised at your statement ‘The jurisprudence of the courts in Europe up until now is based on the assumption that the patents issuing from the EPO have had their descriptions strictly conformed. Court proceedings and court decisions follow that assumption.’
I have only been in the business 15 years , so not as long as you, but maybe I am missing something.
In my experience the EPO started demanding it,after the UK pemetrexed case. Before that, I rarely saw a request to conform the description, and even then only in extreme cases. Never in order to merely copy limitations introduced during prosecution into ‘the invention concerns…’. The first time I got one, I was completely surprised. I have the feeling it never happened before the pemetrexed case, so how the courts could have this assumption beats me. But maybe I have a grossly distorted view of the rosy past.
After all (at least that what I understood then, and still now) the claims (interpreted in the light of the description of course, and considering equivalents, of course) determine the scope of protection, so who cares if the description is at odds with the claims? The claims win anyway.
And actually, the handy list of decisions that Mr Thomas made confirms this: None of these decisions allowed the plain meaning of the claims to be interpreted differently because of the description. The system worked.
I would not go so far as to say that the patent can not be its own dictionary. If the patent explicitly defines what ‘gathered’ means, we can not ignore that. But if the description merely uses a wider or broader term in passing, without positive indication that this is intended to change the meaning of that term, it appears, based on the cited cases, that there is no chance that the EPO will interpret the claim differently (and rightly so, IMHO).
Dear BJ,
I can agree with you that the stricter requirement to align claims and description has been triggered by the “pemetrexed” case. In this case, the applicant bluntly refused to adapt the description and limit the description to what has been disclosed, i.e. pemetrexed disodium.
With T 2027/23 published today and already commented in my blog, the situation has become clear: boards will not read into a clear claim any different interpretation found in the description. You are right to say tat the claims win anyway, but if the description is at odds with the claims, it is for the applicant/proprietor to amend the claim should he wish the claim to be interpreted the same way as in the description.
I would say that with T 2027/23, the idea of the description is a dictionary for the claims has gone for good.
Dear Max Drei,
I confess, I was as well rather surprised when I saw the way the BGH handles added matter. On the other hand, the BGH is well known to be a repair shop for patents. As the patent at stake was granted by the EPO, it is possible to consider that examination was sub-optimal. EP 1 275 192 B1 should never have left the premises of the EPO.
I agree with you that the litigation industry has an interest in making patent litigation as complex as possible. The UPC is a prime example of lobbyism to the benefit of the litigation industry. And the UPC is helping this endeavour by tailoring the UPCA to its needs. I heard a legal member of the BA saying that, if London is Milan, then red can be green.
To me one way of bringing clarity in the matter is exactly what I suggest, align the claims with the description. This avoids proprietor to have in the claim a broad interpretation of the claim and a more limited one in the description.
As far as our discussion is concerned, please look at T 2027/23, published today and already commented in my blog. The decision is absolutely clear: boards are not prepare to reach alignment between claims and description by way of interpretative somersaults! What is there more to say?
By the way, the pemetrexed case was not the result of bad drafting. When fling, the applicant had only data about pemetrexed disodium and never attempted to bring in further data which could support a general applicability to other salts of pemetrexed. But is what the applicant implied by refusing to align the description with the claims before grant. The examiner simply gave up and granted the patent with a claim limited to pemetrexed disodium. With G 1/24 this would not have happened.
Daniel, you prompted me to have a look at the Agfa/Gucci decision on the matter of use of the descrn as filed and as granted. Do we not see here recourse to the EPO file wrapper? I mean, the court said:
“This is further supported by the fact, that during the granting proceedings, the Claimant deleted “chromatic colour” from the wording of original claim 1,”
I agree with everybody else, that the specification of the granted patent is to be considered in its entirety, and the meaning of the granted claims to be derived from that careful consideration. Courts all over the world do this, and are not stopped from doing that when the disclosure supporting the claims (that is to say, the content of the description and drawings) is that which the Applicant made to the Patent Office on the filing date. Of course the EPO must do its best to grant only claims which in themselves are “clear”. Of course, if a claim is not “supported” by the disclosure as filed, it should not be allowed to issue. Nor should any claim that includes any added matter. But if a court gives the claim primacy, as being the subject matter for which protection was sought, and reads the description as a record of what support there was in the application as filed for that claimed subject matter, it can then render clear decisions on i) what the claim means to an imaginary reader skilled in the art, and then ii) whether it is infringed and iii) whether it is valid. To qualify as a UK patent attorney one sits (for very good reasons) a written examination that tests one’s ability to write an opinion on infringement and validity. In real cases, there are always arguments that are a “squeeze” between infringement and validity. In real life, attorneys have to write FTO opinions based on the WO publication. In real life, conforming the description opens up more opportunities to run added matter attacks on the granted patent.
I am not yet convinced that the case law prevents courts from handling cases in this way, going forwards. But, as I said, I cannot imagine the EBA in G1/25 throwing away 40 years of accumulated case law, especially if the community of patent litigation law firms like it, the way it is now.
Dear Max Drei,
With T 2027/23, already mentioned in my previous reply, the idea “that the specification of the granted patent is to be considered in its entirety, and the meaning of the granted claims to be derived from that careful consideration”, belongs to the past.
That courts are not stopped to having their own interpretation of a claimed feature, clear as such for the skilled person, is not at stake. As any judicial body they are free in their assessment of evidence.
At least boards of appeal of the EPO, are not willing any longer to read into a claim a different interpretation found in the description. Post-rant courts can handle a case in the manner they think fit. In the Agfa/Gucci case the UPC-LD Hamburg did not buy the interpretation of the proprietor in its infringement attack in view of the limitation in the claim.
If the interpretation in the description is brought into the claim, I fail to see any potential of added matter, provided the wording, or at least the teaching, of the description is incorporated in the claim.
With G 1/24 as it is interpreted by the boards, the result of G 1/25 seems pretty manifest, but we will have to wait for it to be issued. The long line of case law on the adaptation of the description to the claims, which was only contradicted by a few decisions like T 56/21, will most probably be confirmed. I am not sure it will be to the liking of applicants/proprietors, but welcomed by third parties.
Dear Thomas,
Thank you for your excellent write-up on the recent case law involving G 1/24. Your work is – as always – much appreciated!
I remain unconvinced on your opinion that the patent cannot be its own dictionary and that the EBAs position is in clear contradiction with that of the German Federal Court (BGH) (and, for that matter, the UPC, cf. point III.2. of UPC_CFI_278/2023?).
In my mind, you and the judges of the UPC CFI have a different interpretation of what “the patent can be its own lexicon” means. Moreover, I think both interpretations can be brought in accordance with each other.
It appears that your understanding of “the patent can be its own lexicon” means that any statement within the description must be taken at face value and overwrites established technical meanings. I.e. if a patent defines that anything with more than two wheels is a car, you have to accept that planes are cars in the context of the application.
However, based on the Agfa vs. Gucci (UPC_CFI_278/2023), the CFI appears to understand under “a patent may be used as its own lexicon” that the patent serves as guideline which has to be consulted in order to construe the claim terms in the context of the claimed subject-matter. E.g. under point B.IV.2.b) and c) the CFI accepted the patent’s definition of “achromatic” while rejecting a definition proposed by the patent proprietor based on an overly broad prior art definition of the term “white” (as white is an achromatic color). Consequently, the CFI determined that certain embodiments still present in the description (ivory and pale yellow) are not achromatic and not part of the claimed subject-matter. The proprietor was bound by what he himself sought to define as “achromatic” by virtue of the patent being its own lexicon.
I ask you to consider that certain technical fields, and especially in chemistry, there are plenty of terms commonly used in claims language, which do not have clear-cut technical boarders but would be extremely cumbersome to define within the claim itself (cf. “protein” or “leaving group”). I consider it vital that applicants remain able to (more or less) precisely define such terms as to what they meant when using them so that the public can establish whether they fall into the scope of the claim or not. That does not mean that applicants should be able to ”overwrite” an otherwise clearly determinable claim scope, cf. T 1999/23 and the meaning of the term “excitation area”. In my mind, in the worst case a “bad definition” that unduly affects the claimed scope can be disregarded as inconsistent with the claim language.
To sum up my opinion in your words, I think that “a patent can be its own lexicon” does not mean that any inconsistency between the claims and description (and especially the definitions), when it comes to interpreting claims, will be necessarily read into the claim. I would rather interpret it as meaning that the skilled person will consider the definition of a term in the description and subsequently decide on a case-to-case basis whether this definition is compatible with his common technical understanding of the term and disregard it in case of an incompatibility (-> primacy of the claims). As such, the statement “a patent can be its own lexicon” as I understand it is not contra to G 1/24.
Therefore, I suggest one could clarify that “a patent can be its own lexicon, unless it seeks to unduly alter an otherwise clear and technically sensible claim scope or refers to subject-matter no longer encompassed by the claims.” Would you be inclined to agree with such a specified statement?
A further comment on the BGH position: While I agree with your opinion that the BGH often appears to function as “patent repair shop” (especially on added matter problem), one has to take care not take its decisions too far. While the often cited BGH – Rotorelemente literally states that while “the claims take precedence … this does not exclude the possibility that the description and drawings may lead to an understanding of the patent claims that deviates from the literal wording of the claims” (translation), that statement was made in relation to a claim that barely made sense in light of the description and certainly was not encompassed by it. I would be reserved to compare this BGH rationale to situations where the claimed subject-matter was encompassed by the description and technically sensible (cf. e.g. point 2.3.1 of T 1069/23). However, admittedly, I have no knowledge of whether there ever was a BGH case where a definition in the description overwrote a technically sensible claim term with a different technical meaning. I think to remember from PatG case law that a description cannot go contra to a clear technical understanding of a claim term (e.g. a definition of an “alkyl” cannot contain alkenes. Nonsensical descriptions are disregarded).
In sum, while I agree that the BGH is (a lot) more proprietor friendly in matters of claim construction than the EBA, I would not go so far as claiming that the BGH and EBA “clearly contradict” each other.
Finally, I ask to excuse myself for the overly lengthy comment and look forward to your comments.
Dear L.T.,
Thanks for your detailed comments. There is no need to apologise for a lengthy comment. My replies are also sometimes quite lengthy.
In spite of your comments, I am not in a position to change my views. The reasons are as follows.
I think that our divergence lies in the notion of dictionary. To me a dictionary is a reference source in print or electronic form containing words usually alphabetically arranged along with information about their forms, pronunciations, functions, etymologies, meanings, and syntactic and idiomatic uses, cf. Merriam Webster.
If a statement within the description does not have to be taken at face value, what is it then good for? I do not say that it must overwrite an established technical meanings, it must just be giving it a different meaning, and this is what dictionary a for: giving a certain meaning to a specific word. Your example with cars and planes is oversimplifying and hence no good.
If there is an inconsistency between the claims and description when it comes to interpreting claims, when will it the case that it is not necessary to read it into the claim, and when will it have to be read in the claim? This is all very confusing and hence gives rise to doubts about the actual scope of protection. This is not acceptable in general. Third parties have to know where they stand.
When you look at the case leading to G 1/24, there is no prima facie incompatibility between “gathered” in the claim and the definition of gathered in the description, and yet, depending on which definition one takes, that of the description or that of the claims, the claim lacks novelty or is respectively novel.
For me, the statement “a patent can be its own lexicon” as I understand it, is directly and unambiguously contrary to G 1/24. G 1/24 has for aim to sort out the contradiction or inconsistency between claims and description. What you suggest is simply reinforcing it.
Like every rule, it will certainly suffer some exceptions, may be in gene technology, but leaving doubts about the scope of protection is a no go.
For example, should the claim mention feature A, which has a clear meaning for the skilled person, and then the description gives a different meaning to that feature A, deviating from that the skilled person would give to feature A, then, if you consider the patent as its own dictionary, you end up not knowing what should be the scope of protection as it should be defined by the claims. What godd are then claims for, if third parties need to dig deep in the description to know were they stand?
I can therefore not agree with your specified statement, as it increases confusion and does not bring about the clarity needed. How can you bring your definition in accordance with the notion or directly and unambiguously derivable as it is used at the EPO?
For instance, when a more limited or a broader definition is given in the description from that the skilled person derives from the claimed feature there is a clear inconsistency between claim and description. This was the case in the referring decision T 439/22 with “gathering”. This is also clearly exemplified in T 1069/23 or in T 1561/23 when it comes to a limiting meaning.
In T 1069/23, the proprietor tried to give a more limited meaning to a feature clear as such for skilled person, as the meaning in the description was encompassed by the more limited meaning in the description. This is to me a clear attempt by the proprietor to use the description as a dictionary, since it gave another meaning in the description to a clear feature in the claim.
That in Point III.2, the CFI-LD Hamburg has mentioned that the patent can be its own dictionary is not at stake. It is simply part of a more general statement about how claims ought to be interpreted.
Drawing the conclusion that in the Agfa/Gucci case the patent was its own dictionary cannot be seen in said decision. In the latter, the LD Hamburg did not use the description as a dictionary, it merely construed the claim and came to the conclusion that there was an inconsistency between the claim as granted and the description. This has nothing to do with the patent being its own dictionary.
On the other hand, the notion of the patent being its own dictionary is clearly standard practice in German case law. See for instance OLG Düsseldorf, 28.06.2007 – 2 U 22/06 – Betonpumpe; BGH X ZR 85/96, GRUR 1999, 910 (911) – Spannschraube; BGH GRUR 1998, 133 (134) – Kunststoffaufbereitung; BGH GRUR 1991, 444 – Autowaschvorrichtung; BGH GRUR 1988, 896 – Ionenanalyse; BGH GRUR 1975, 422 (424) – Streckwalze II.
In this respect, the case law of the boards of appeal is to be held in contradiction with the case law of the BGH. The BGH decision Rotorelemente, X ZR 43/13 is a very specific example, and is actually the result of an examination, which I would call suboptimal as it lead to confusion. There was again an inconsistency between description and claim, and this inconsistency should have been removed at grant.
Be it for novelty, inventive step, sufficiency or added matter, nonsensical interpretations of clear claims are disregarded, whether looking at the claim on its own or taking into account the description. There is abundant case law on this topic.
I was certainly quite long, but it was necessary and worth it.
Dear Mr Thomas,
you wrote ‘but leaving doubts about the scope of protection is a no go’.
On this one, I fully, 100%, agree with you.
Where we don’t agree appears to be: whether not-conforming the description to the claims gives rise to doubts about the scope of protection. I say it doesn’t (usually, exceptions aside, like the ‘gathered’ case) because the Boards of Appeal and the courts are wise enough to nip such attempts in the bud (to use your words). And your position seems to be, but correct me if I am wrong: Let’s make sure that there can not be any doubt and be very strict on conforming.
Dear BJ,
With T 2027/23, already commented in previous replies, the boards have made clear that they will indeed nip in the bud any attempt to hide in the description interpretations which are different from that a skilled person will derive from a clear claimed feature. What matters is what is in the claims and not what is hidden somewhere in the description.
The necessary conclusion is that a strict conforming will be reached, if the interpretation hidden in the description will be brought in the claims.
In view of the board’s interpretation of G 1/24, it has actually brought a clarification which will not be liked by applicants/proprietors, but welcomed by third parties.
In other words, at the end of the day, boards are getting in the direction of requiring a strict alignment between claims and description in accordance with their long lasting case law.
Daniel, the thought occurs to me t hat there is a difference between i) a definition in the description as filed of a technical feature (say, “gathered”) in the claim and ii) a detailed description of an “example” in the description of the “invention” in the A publication which, following prosecution amendment of claim 1, is no longer an example of the subject matter for which protection is sought..
The definition endures, into the B publication, but the example falls away, exactly because the claim, during prosecution, is i) narrowed to avoid the prior art but ii) internally rendered patently “clear” as to its ambit, what it covers. Pemetrexed is history, but “gathered” is still inplay.
I expect the emerging case law on “consult the description” to reveal this difference, ever more clearly.
Dear Max Drei,
Thanks for a very interesting comment. It gave me food for thoughts
I fully agree with you that there is a difference between i) a definition in the description as filed of a technical feature (say, “gathered”) in the claim and ii) a detailed description of an “example” in the description of the “invention” in the A publication which, following prosecution amendment of claim 1, is no longer an example of the subject matter for which protection is sought.
If some examples do not any longer fall under the claims are granted, they will have to be marked as such before grant. This marking will then indeed appear in the B publication. To determine whether examples do not any longer fall under the claims, the description will necessary have to be consulted. G 1/24 does thus also apply in this situation.
The conclusion can therefore be drawn that G 1/24 will force examiners to consult the description before starting the search and during examination, as well as before grant. Any discrepancy between claims and description will have to be removed. This is a positive effect of G 1/24.
When the definition in the description is broader than that derivable by a skilled person from the claim, it will have a direct influence on search and examination as the broader definition will have to be taken into account. An objection of lack of novelty and inventive step is more prone to be raised in view of the breadth of the claim. The applicant cannot rely on the limited definition in the claim and has to accept that the relevant definition for search and examination is the broader one from the description. As a result, some examples might indeed not any longer fall under the claims and will have to be marked as such.
When the definition in the description is more limited than the claim, then the search has be carried out on the basis of the broader definition in the claims. If the applicant is then of the opinion that the limited definition has to be taken into account in order to avoid objections of lack of novelty and inventive step, then he has to bring the more limited definition in the claim. But even in this situation, some examples might not any longer fall under the claims and will have to be marked as such.
If the definition in the claim is the same as in the description, there is no reason for a deciding body to require insertion in the claims of the definition disclosed in the description. It can however still happen that examples do not fall under the claims as granted after search and examination. Those will also have to be marked as such in the description.
Whatever the situation might be, examples not any longer falling under the claims will have to be marked as such. I use purposedly the wording “marked as such” and I would not require deletion as what has been deleted has gone.
However the examples left in the description after grant should not allow the proprietor to claim, a broader scope of protection than that “derivable” from the claims as granted. I also use the word “derivable” purposedly, as this encompasses possible equivalents.
Whatever the situation might be, we end up with the necessity of the adaptation of the description to the allowable claims. From then on, we are arrived in the domain of the next decision of the EBA: the necessity to adapt the description to the claims, i.e. G 1/25.
Hallo Daniel and thanks for that kind reply. As you note, we are all agog for the EBA’s opinion in G1/25.
But meanwhile, I have to say, your forthright opinions remind me of a former UK Prime Minister, Margaret Thatcher, who acquired the nick-name Tina because of her prediliction for declaring There Is No Alternative! On the matter of “adapting” the decription, my suspicion is that the EBA is mulling over the possibility of at least one way to move forward other than the one set out in the Guidelines.
I just hope that, whatever the EBA decides, it will provide reasoning in support that is unassailably persuasive.
In my field, devices rather than molecules, I think it not so easy, to categorize every drawing Figure into just two categories, either “embodiment” or “not the invention”. For an example I return to my catheter shaft field.
My “invention” is embodied in a particular component of the device, say the inner of two co-axial components of a catheter shaft. My drawings show catheter devices in whole or in part, different sorts of catheter. However, only a few of the drawings illustrate the inventive detail within the inner shaft component. All the other drawings lack any detail in the inner shaft component. The description mentions that a catheter designer is free to incorporate an embodiment of the inventive inner shaft component in each of the various catheters illustrated.
The claim is to a catheter device. That’s because a claim to a shaft component as such, or to a co-axial shaft as such, lacks novelty. My description as filed declares every one of the drawing Figures is an “embodiment” of the invention. But is that true? The Figures that show the shaft component do not show any catheter. The catheter drawings do not show the inventive detail of the inner shaft component. So in fact, strictly speaking, NONE of the drawings is an illustration of an “embodiment” of the claimed invention.
Back to the statement in my (US-drafted) description that ALL of the drawings are of “embodiments” of the invention. Does that statement render the claim unclear or deprive it of support from the description?
If so, can I perhaps get to issue by amending it to read that NONE of the drawings show an embodiment of the invention? That’s closer to the truth, right? But will it help to make things clearer for the skilled reader? I have to say, I doubt it.
Patent attorneys have an aversion to deleting disclosure and also to labelling it as “not” the invention. By insisting on one or the other, the EPO is making the routine work of its Examiners harder, in my opinion unnecessarily. This is why I draw a distinction between i) definitions of claim terms embedded in the description and ii) descriptions of apparatus and methods said to be “of the invention” which as a result of a narrowing amendment of the claim during prosecution no longer fall within the strict wording of the narrowed claim. A court can easily finesse around the “embodiment” issue but finds itself in difficulty to ignore any such definition (because to do so offends the “Golden Rule” how to construe ANY given text document on the desk in front of it). Bring on G1/25.
It is unsurprising that the TBA’s are not letting proprietors save their patents by relying on the description. G 1/24 is vague enough to allow TBA’s to consult the description without using it.
The most interesting question left is whether EP3076804 will be revoked. If the line of argument which is being used by the TBA’s thus far is to be followed, the broad definition of “gathered” in the description should be ignored and it thus appears the patent will not be revoked. Would anyone offer a different prediction?
@ C 22,
G 1/24 might be vague, but it does by no means say that the interpretation in the description will overrule the interpretation given by the skilled person to a clear claimed feature.
G 1/24, does not mean that a limited interpretation in the description of a clear claimed feature, will be superseded by a broader interpretation of a claim.
When it comes to a broadened interpretation of a claimed feature, G 1/24 has not said that the broadened interpretation of the description overrules a clear claim feature. This is a clear Angora cat. During the OP in G 1/24, the opponent drew the attention to the fact that was the proprietor wanted was an Angora cat.
In both cases it is not acceptable that the proprietor is allowed to benefit from one interpretation when it comes to infringement, i.e. the broadest possible, and a different one when it comes to validity.
To reply to your question, the broad definition of “gathered” in the description should NOT be ignored and it thus appears the patent will be revoked.
The thing about Angora cat situations is that they can turn round and bite the super-cunning patentee. I mean, the court, looking at validity, can find ways to adopt the wider of the two available meanings of “gathered”. So, here, to the ordinary observer, the average consumer, the person in the street, “gathered” means “brought into closer proximity by any means under the sun” while to other more discerning readers, including all those in the fabric-based industries, “gathered” means something like “pleated”. So, the claim is incurably ambiguous, indefinite, not in compliance with Art 83, EPC. Further, or alternatively, the claim lacks novelty. Further, to the extent that deletion of the definition in the description changes what the reader derives directly and unambiguously, such deletion offends Art 123(2) EPC.
To do justice, the patent has to put out of its misery. One way or another, the Board will do justice, the Rapporteur choosing the despatch route.
Is it hard to imagine that the skilled person finds, based on the patent and his own knowledge, two specific alternatives (a wider one and a narrower one) both directly and unambiguously disclosed? I think not, so I don’t see the A123(2) issue.
I would say: both are sufficient disclosed, it is merely undecided which of the two is legally present in the claims, and therefore whether there is novelty or not.
Dear Max Drei,
I think you have given a clear definition of what “gathered” means. It simply means “pleated”. There is nothing else which would lead the skilled person to have another interpretation than pleated, even if relates to a sheet of tobacco product. In § [0035] of the patent, the applicant tried to give a much broader definition of what he understands under “gathered”, and this is why we ended up with G 1/24.
The claim is thus by no means incurably ambiguous, indefinite, and certainly fully in compliance with Art 83.
Furthermore, it is difficult to follow you when you consider that the definition in the description should actually be deleted. Taking into account G 1/24 and T 2027/23, the definition in the description should be incorporated in the claim. There would thus not be any problem with Art 123(2). The problem is then that claim 1 as granted will lack novelty.
This is a likely hypothesis, but will have to wait until the referring board has dealt with this problem and issued a decision.
The proprietor can put himself out of misery by amending its claim.
Dumb of me, Daniel, but I am not seeing what amendment of the claim would confer novelty without adding matter. I don’t accept that “gathered” means, as such, directly and unambiguously, “pleated”, even in the context of tobacco plant leaf fragments. Do you have in mind to amend “gathered” to read “pleated” or are you thinking of some other amendment? Is pleated, stated explicitly, the invention as-filed? Is that it?
Would you care to write a bit more, perhaps?
Dear Max Drei,
If the proprietor brings in the claim what he has defined in § [0035] of the description, I am at a loss to see how added matter could be the result of the amendment.
If the proprietor brings in the claim what he has defined in § [0035] of the description, the meaning of “gathered” will be clarified and it becomes then irrelevant what the skilled parson could understand under “gathered” as it stood in the claim.
The problem is that with the definition in § [0035] of the description introduced in the claim, claim 1 as granted lacks novelty. By keeping the claim as it stands, i.e. with a discrepancy between claims and description, claim 1 as granted could be considered novel.
Taking into account the definition in the claim, novelty could be given, but with the definition in the description, the proprietor could claim infringement if the tobacco sheet is simply pleated. Such a double standard is not acceptable and this is what G 1/24 aimed at removing..
This is the problem which is underlying the referring decision T 439/22. .
Ah! Thanks, Daniel, for replying. I think now that I misunderstood what you meant by “putting out of misery”.
Until your reply, I was thinking that you meant an amendment to claim 1 that removed the misery of languishing under potent threats to the validity of the claim. In other words, an amendment that overcomes the validity attacks. What other sort of amendment does a patent owner ever offer?
But now, having read your reply, I guess that you meant “putting out of misery” sarcastically, in that you had in mind an amendment that leaves the claim invalid, beyond any doubt.
There are five paragraphs in your reply. I have no quarrel with the first three. As to the fourth, I do not understand what “definition” of “gathered” the claim contains. There isn’t one, is there? To the contrary, the claim feature “gathered”, taken on its own, is indefinite to the extent that the word has a plurality of different dictionary definitions. In a crowd, human beings have “gathered” together for sure. But, for sure, there are no pleats or folds in the gathering. Conversely, when a dress- or curtain-maker speaks of gathering fabric, pleats are the consequence. No “gathering” takes place when one wrings water out of a wet towel.
As to the tobacco leaf squeezing industry, however, I would need a person skilled in the art to brief me what “gathering” means to them. Perhaps the EBA will tell us the answer. Roll on G1/25, I say.