After years of debate and conflicting decisions from the Boards of Appeal, we finally have a referral to the Enlarged Board of Appeal (EBA) on the description amendments! To be honest, after the first false alarm (IPKat), PatKat was half-expecting them to bottle it again. However, it is now official. In the first referral of 2025, the question of whether applicants and patentees must amend the description to conform to the allowed claims has now been referred to the EBA (G1/25).
Adaptation of the description: What is all the fuss about?
The description amendment saga began with the introduction of substantially stricter guidance in the 2021 EPO Guidelines for Examination on how much the description should be adapted in line with the allowed claims. The Guidelines currently require that, when adapting the description, the applicant must either delete subject matter not covered by the claims or explicitly state in the description that such subject matter is not part of the invention. However, the Guidelines are at odds with a number of Board of Appeal decisions finding no legal basis for the mandatory adaptation of the description, most notably T 1989/18 (IPKat).
The issue of description amendments causes controversy for a number of reasons. First, amendments made to the description of the patent application may influence interpretation of the claims under the doctrine of equivalents in national court proceedings. Additionally, the claims of the granted patent may be interpreted at opposition in view of the description of the patent as granted and not the description of the application as filed (see T471/20, IPKat). As such, amendments to the description may irretrievably add matter and invalidate the patent (IPKat). These risks, and the less than clear legal basis for the requirement, have led to considerable dissatisfaction with the EPO's uncompromising approach to description amendments. On top of all this, description amendments take up considerable patent office and attorney time, and thereby increase costs for applicants. Given that no other patent office in the world considers it necessary for patent quality or legal certainty that the description is amended in line with the claims, why does the EPO make such a big deal out of it?
 |
| Hydroponics |
Interestingly, the last time a referral on description amendments was considered, the Board of Appeal in the case back-pedalled, in view of what it considered the clear lack of legal basis for the requirement (T56/21). In this previous case, the Board of Appeal remarkably commented that "[i]f the legislator considers it justified to require that the description be aligned with the subject-matter of claims held allowable, the legislator should provide for the respective legal basis by way of amendment of the EPC."
G1/25: Case background
The case at the centre of this new referral is an appeal concerning EP 2124521 (Hydroponics growth medium). The patent, owned by Knauf Insulation, was opposed by Rockwool International. Both parties appealed the interlocutory decision from the Opposition Division.
The patent in question relates to a method of growing a plant in a special hydroponics growing medium. Claim 1 of the key request on appeal defined the organic binder in the growing medium with a great deal of specificity, including requiring that the binder was "based on reaction products obtained by curing an aqueous solution comprising citric acid, ammonia and dextrose". This feature was imported from the dependent claims of the patent as granted and significantly narrowed the scope of claim 1.
The Opponent argued that this amendment introduced a fatal flaw because it rendered the original description, and the amended description filed at first instance proceedings, inconsistent with the claims. Specifically, the description described the binder in much broader and more optional terms than the amended claim. For instance, the description stated that the binder "may be based on a combination of a polycarboxylic acid, for example citric acid, a sugar, for example dextrose, and a source of ammonia, for example ammonia solution". The Opponent argued that what was described as optional in the description had become mandatory in the claim. The Opponent thus submitted that, because of this inconsistency, the claims were not "supported by the description" as required by Article 84 EPC.
The Board of Appeal in the case (T697/22) agreed with the Opponent that the amendment had indeed "introduced an inconsistency" between the claim and the description. It found that the allowability of the request was therefore entirely dependent on the question of whether such an inconsistency was a legal impediment to maintaining the patent (r. 2.2)
The Board of Appeal's search retrieved 115 relevant decisions and concluded that there are two lines of clearly diverging case law (r. 12). The first line of case law affirms that the description must be adapted. However, the Board of Appeal noted that there is no consensus on the legal basis for this requirement. Some decisions have relied on Article 84 EPC alone (T1024/18), whilst others cite Article 84 EPC in combination with other provisions (T438/22). By contrast, the second (more recent) line of case law argues there is no legal basis whatsoever in the EPC for this requirement. This view holds that the support requirement of Article 84 EPC does not preclude the description from containing unclaimed subject-matter and that forcing adaptation is a policy choice for the legislator, not the EPO (T56/21).
Referral confirmed! (G1/25)
It has now been confirmed that the Board of Appeal in T697/22 has referred the following questions to the EBA:
- If the claims of a European patent are amended during opposition proceedings or opposition-appeal proceedings, and the amendment introduces an inconsistency between the amended claims and the description of the patent, is it necessary, to comply with the requirements of the EPC, to adapt the description to the amended claims so as to remove the inconsistency.
- If the first question is answered in the affirmative, which requirement(s) of the EPC necessitate(s) such an adaptation?
- Would the answer to questions 1 and 2 be different if the claims of a European patent application are amended during examination proceedings or examination-appeal proceedings, and the amendment introduces an inconsistency between the amended claims and the description of the patent application?
Notably, in its decision, the Board of Appeal makes a number of references to the recent EBA decision in G1/24 (IPKat). The Board of Appeal specifically found that G1/24 reinforced the need for a referral. Specifically, the Board of Appeal noted that, because G 1/24 requires that the description and drawings must always be consulted to interpret the claims, the issue of whether there can be an inconsistency between a claim and the description has "become of even greater significance". For an in-depth analysis of the context behind the referral (including the relevance of G1/24), see our latest GuestPost from Katfriend Greg Corcoran (IPKat).
The mandatory note of caution required with respect to any EBA referral should now also be sounded. As we have seen with other recent referrals, it is possible that G1/25 will not provide the clear answer that many practitioners are looking for. If the EBA holds true to its recent approach, it may also only superficially provide an answer, and leave it to the Boards of Appeal to interpret what this means in view of the existing case law (see all of G1/21 (IPKat), G2/21 (IPKat) and G1/24 (IPKat)).
In this Kat's view, EBA decisions should also always be interpreted and understood within the context of the existing case law. In the wake of G1/24, it was fascinating to see the divergence amongst the commentariat as to how the decision should be interpreted, with many choosing to ignore the key nuance of language chosen by the EBA (see for example the comments on this IPKat post). As this Kat pointed out at the time, the use of the term "consult", as opposed to "use" in the Headnote of G1/24 is critical to how the Boards of Appeal will interpret and apply G1/24, as has been recently proved (IPKat). On such a controversial topic as description amendments, it can be easy to forget that what really matters for day-to-day practice is what the law says, not what you think it should say.
Further reading
- Board of Appeal finds no legal basis for the requirement to amend the description in line with the claims (T1989/18) (Dec 2021)
- Can amending the description to summarize the prior art add matter to the patent application as filed? (T 0471/20) (Jan 2022)
- EPO Board of Appeal toes the party-line on description amendments (T 1024/18) (April 2022)
- Board of Appeal case law against the description amendment requirement starts to mount up (T 2194/19) (16 Nov 2022)
- Another Board of Appeal enters the fray on description amendments (T 3097/19) (Nov 2022)
- EPO tries to have its cake and eat it on claim interpretation (T 0169/20) (Mar 2023)
- When is it necessary to amend the description of a patent post-grant (T2391/18) (Jun 2023)
- Board of Appeal poised on the brink of a referral on description amendments (T 56/21) (July 2023)
- The risk of pre-grant description amendments (T 450/20) (July 2023)
- Adding matter by amending the description to exclude embodiments (Ensygnia v Shell [2023] EWHC 1495 (Pat) (Aug 2023)
- Deletion of claim-like clauses: Board of Appeal finds legal basis for adaptation of the description from "the EPC as a whole" (T 0438/22) (Jan 2024)
- Discrepancies in the description should be amended in line with the claims, but do not affect interpretation of the claims (T 0447/22) (March 2024)
- Board of Appeal back-pedals on referral in view of "unequivocal" lack of legal basis for the description amendment requirement (T 56/21) (Oct 2024)
- [GuestPost] Has G1/24 made the referral of the question of amendment of the description from T697/22 inevitable?
Reviewed by Dr Rose Hughes
on
Tuesday, July 29, 2025
Rating:
Follow the IPKat on LinkedIn 
![[Book Review] Decolonizing Intellectual Property Law – An Afrocenteric Approach](https://blogger.googleusercontent.com/img/a/AVvXsEiLxxU1FAFmN4Kz-lkqEmoEQPwj8Yi6OcAENC94bZ9A2-MObgI1bqc9y5AOtyDN89b8TXfzDXdMI03idZYOrThD3xgbQXv9nz9k9x_0bdrWHwp4tAKvHqzas8RgQxmdGuLqqJ9pel3t5XPuRPndnxaiGsfcC8YiRtiV5UlUhNFM9DC8Mtlcm526Ew=s72-w244-c-h320)
Posts
At first glance, it would seem that the referral already assumes what an inconsistency is.
ReplyDeleteIt seems to me, however, that the problem lies precisely in defining what an inconsistency is.
For example, is there an inconsistency if the description contains examples A+B and A+C, and the claims only claim A+B?
Since it is explicit in the EPC that the claims define the scope of the patent, I would say no. The page fees are the mechanism used to encourage applications to be concise and not discuss anything extraneous, but there is no explicit legal basis requiring that the description refrain from discussing more general or alternative solutions.
DeleteI agree that the question should have made precise what is meant by "inconsistency". In the case at hand, it seems the inconsistency consists in an embodiment of the invention (or just an example?) being broader than and thus not falling within the scope of the (independent) claims. (If it is just an example, then I see no problem with it.)
DeleteAnother problem I have with Q1 is that it asks whether the description has to be amended. This is the wrong question. The right question is whether the inconsistency violates a requirement of the EPC (which could only be A.84). If it does, then the patent cannot be maintained, Art.101(3)(b) EPC. Or the appellant may choose to amend the application documents, i.e. claims, description, drawings.
So this referral, with Q1 properly reformulated, is simply a question about Art. 84. Q2 and Q3 are then no longer necessary. And Q3 obviously is inadmissible anyway.
I can agree that the question is really about Art 84, and especially the part "support".
DeleteTo take the example, of "chalk" and "cheese", if this is not an inconsistnce, then, what will be an inconsistency?
If the claim only relates to A+B, then A+C is not falling under the claims. A+C does not have to be deleted, but acknowledged not to be covered by the claim on A+B.
If the applicant/proprietor wants as well protection for A+C he has to file a corresponding claim in the actual procedure or file a divisional. In any case, after grant he will not have the possibility to file a divisional.
Yes, but what if the applicant would like to argue in litigation before a national court that A+C falls under a claim to A+B according to the doctrine of equivalents? Being forced by the EPO to say explicitly in the description that A+C is not to be covered by the claim A+B will undermine the argument for equivalence in many jurisdictions (such as the UK). National offices do not require amendment of the description. It should not be the case that a patent application prosecuted via the national route performs better in litigation than a patent prosecuted via the EPO. However that is the situation faced by many patentees at the moment.
DeleteDear DXThomas, the "A+B" question regards of course Art. 84 but the part about clarity, not support, since the claim A+B is certainly supported by the description of A+B.
DeleteOne may argue that the disclosure of A+C might render A+B unclear (I am wondering why only the EPO is the only office to have this doubt), hoverver if and only if the disclosure adds that A+C is an "embodiment of the claimed invention" or the like.
Otherwise, I really cannot see how a "neutral" disclosure of A+C (i.e. without "claiming" A+C in the description) would cast doubts on the clarity of the the scope of proctection of claim A+B.
On the other hand, "disclaiming" a "neutral" disclosure of A+C in the description, as required by the EPO, might invalidate a patent because of Art. 138(1)(b) and/or 123(2) and/or (3) EPC.
@ Anonymous of Thursday, 31 July 2025 at 10:32:00 GMT+1
DeleteIf there is a remote possibility that A+C is meant to fall under a claim to A+B, then it why has this not been be made clear in the description at filing?
Another possibility would be to claim not only A+B but also A+C provided there is unity of invention, or do you want a patent with less than 15 claims in order to save claim fees?
If, during prosecution, you limit your claim to A+B, you cannot maintain that A+C is still covered by the claim. The two statements are not compatible. I would not require deletion of A+C, but a clear acknowledgement that A+C is not falling under the claims is necessary. If you have good arguments in national litigation, I do not see why a national judge would not follow you.
That national offices, like the USPTO or the French patent office, do not require amendment of the description is irrelevant as far as the EPO is concerned.
I have said it many times, with Art 84 as it stands, there is no way that the description is not to be amended, contrary to what has been said in T 56/21 and the few other decisions going along the same path. May be G 1/25 will change this, and then I will be happy for you and your colleagues.
The day Art 84 is amended, and for instance stripped of the part relating to support, and a kind of file-wrapper estoppel is introduced in the EPC, I will be able to agree that the description need not be amended. That an inconsistency between claims and description may still exist in case of misalignment of claims and description, can however not be denied.
I further invite you to show me and the readers a case in which a national patent has received a better treatment in litigation than an EP patent. For the time being, this is only a theoretical possibility.
If I am not mistaken, in EPC member states, a European patent and a national patent with identical claims cannot coexist. If the claims are different, then your scenario could be right.
@ Patent Robot of Thursday, 31 July 2025 at 15:35:00 GMT+
DeleteI first fail to understand in how far the disclosure A+C can be qualified as “neutral”, whatever “neutral” might mean.
Any feature or combination of features mentioned in a patent description is by essence anything but “neutral”, otherwise the applicant/proprietor would not have brought those in. If the applicant/proprietor has brought in A+C next to A+B he must have had his reasons.
However, when only claiming A+B the applicant/proprietor has, by its own volition, implicitly disclaimed A+C as he has not made A+C subject-matter of claims, for whatever reasons. What has not been claimed cannot be considered covered by the actual claims. This is plain logic.
Thereafter, simply stating in the description that A+C is not covered by the claims cannot therefore pose a problem. It is a matter of fact which finds its origin in the own behaviour of the applicant/proprietor.
Even if later, e.g. during litigation/validity the proprietor might want to claim A+B+C or B+C, those combinations will infringe Art 138(1)(b).
Secondly I fail to follow you when you claim that , "disclaiming" A+C might invalidate a patent because of Art. 138(1)(b) and/or 123(2) and/or (3) EPC. I agree with you that A+C does not have to be deleted, but it is necessary to acknowledge that A+C is not claimed.
I am still waiting to see the first patent revoked by court acting in post grant procedures under Art 123(2) because of the adaptation of the description. It happened once in the UK, but there it was the applicant/proprietor which has amended the description of its own volition.
"Yes, but what if the applicant would like to argue in litigation before a national court that A+C falls under a claim to A+B according to the doctrine of equivalents?"
DeleteThen the applicant may draft a claim that encompasses both.
Equivalents do have a role to play in infringement, but if the equivalent is already disclosed in the application there is no justification for telling the EPO that it does not fall under the claim for novelty and inventive step while telling the national court that it does fall under the claim for infringement.
That said, I don't see why the applicant should specifically disclaim (in the description) a particular example as that would unnecessarily invite Art. 123(2)/(3) traps. As long as the description does not suggest that the example is an "embodiment" of the invention, that should be enough. (And if the description anyway does mark it as an embodiment, then in opposition the EPO should interpret the claim as covering that embodiment.)
Dear DXThomas, I fully agree with you that "when only claiming A+B the applicant/proprietor has, by its own volition, implicitly disclaimed A+C as he has not made A+C subject-matter of claims, for whatever reasons. What has not been claimed cannot be considered covered by the actual claims. This is plain logic.".
DeleteI would add that it is not a task of the EPO to investigate the reasons why the applicant included A+C in the description. Maybe the applicant just wanted to donate his knowledge to humanity.
However, I fully disagree with you that "simply stating in the description that A+C is not covered by the claims cannot therefore pose a problem", and I give you two examples here:
1. B is not described sufficiently but C describes a feature common to B which renders B sufficiently disclosed: explicitly disclaiming A+C might nullify claim A+B for insufficiency
2. C contains a disclaimer relating to a feature common to B: disclaiming A+C might nullify claim A+B for added matter and/or claim broadening
Please note that the reality is far more complex than these two examples on A+B/A+C: often reviewing the description to disclaim features takes more time than drafting the description itself. This may become a real nightmare, with a very high risk of errors, at the end of oral proceedings.
The concrete risk of this useless procedure is that at the end a fully valid patent is revoked because of a questionable loophole.
Decades ago I was taught that the description was written on stone and the claims on sand: I would be glad if the EPO limited the requirement of amending the description only to cases in which the disclosure of a features (i.e. "gathered") renders a claimed feature unclear.
I would even not amend the description to cite the prior art, which is another useless requirement peculiar to the EPO, since the cover of the granted patent already cites the prior art (and, if needed, its relevance can be better understood by analysing the examination files on the EPO Register).
@ Patent Robot Friday, 1 August 2025 at 12:52:00 GMT+1
DeleteFor your example 1: “If B is not described sufficiently but C describes a feature common to B which renders B sufficiently disclosed: explicitly disclaiming A+C” might in no way “nullify claim A+B for insufficiency”, as what you claim disclaim is the combination of A+C, but you do not disclaim C as such. If you would disclaim C as such, then you would indeed have a problem with Art 123(2). The same reply is valid for your example 2.
What you have been told “decades ago” is not absolutely correct. On the one hand, you cannot bring in the description new features, and deleting essential features will end with problems under Art 123(2).
On the other hand, claims can be amended, but I add immediately that, according to G 1/24, and even before it, the description has to be consulted when interpreting claims. Claims cannot be considered independently from the description. As far as the search is concerned, just have a look at the Guidelines for search B-III.
It is your good right to refuse quoting the prior art revealed during the search, but then you are not helping your clients and the application will be refused.
In T 276/07 it was held that, the amendment to the description in order to add an acknowledgement of the prior art does not bring any new content to the disclosure of the invention even if it could be considered to put the invention in a different light. This decision goes back to T 11/82, quoted in the Guidelines F-II, 4.3.
Dear DXThomas, some quick replies:
DeleteSorry but I am not sure that " what you claim disclaim is the combination of A+C, but you do not disclaim C as such", since when you disclaim an embodiment, you disclaim it entirely.
Re G1/24: I was taught that "the description has to be consulted when interpreting claims" but I always thought that the description meant the description as initially filed, not as amended after filing, and that the claims had the primacy in case of contradiction beween description and claims.
I know the parts of the Guidelines and the T decisions you mention: I just think they are not supported by the EPC (let's agree to disagree on this, please).
For example, you write "It is your good right to refuse quoting the prior art revealed during the search, but then you are not helping your clients and the application will be refused." I explained why quoting the prior art is completely useless (and a cost for my clients): why should an inventor lose his patent rights just for not complying a useless obligation?
Even in the light of G 1/24, I think that there is a good argument that the claims are perfectly clear. This is on the basis of principles outlined in UK case law (Virgin Atlantic v Premium Aircraft [2010] RPC 8 at [5]), such as:
ReplyDelete"It is the terms of the claims which delineate the patentee's territory"; and
"if the patentee has included what is obviously a deliberate limitation in his claims, it must have a meaning. One cannot disregard obviously intentional elements".
Thus, given that there is only a need to interpret limitations that actually appear in the claims, broader disclosures in the description would, in principle, not appear to create a lack of clarity. Indeed, no such lack of clarity would appear to arise due to the alleged "inconsistency" in the case at hand.
The comments in the GL are evasive with regard to the limb of Art. 84 EPC (ie whether clarity or support) that is triggered by an alleged "inconsistency". However, they do indicate that an "inconsistency" should be dealt with (only) if "it casts doubt on the subject-matter for which protection is sought". Unless someone can persuade me otherwise, it seems that this points solely to the clarity limb of Art. 84. This is because it is impossible to envisage how an alleged lack of support could cast doubt upon the scope of the claims.
It is therefore a shame that the TBA did not engage with the patentee's submissions on what qualifies as an "inconsistency". This is not least because, as illustrated above, it is at least arguable that the TBA was wrong to conclude that there is such an "inconsistency" in the case at hand. However, it is also because it seems like a bad idea to refer questions to the EBA without providing any kind of definition (let alone a clear definition) for a key term used in those questions.
Proof, I'll bite. When the claim defines the matter for which protection is sought as A+B and the description announces that in one aspect the invention is A+B while, in another aspect, the invention is A+C, then reasonable minds can hold that out to amount to an "inconsistency". Does it cast doubt on the ambit of the claim? I think that will vary from case to case.
DeleteAs to whether a problematic inconsistency is objectionable under "clarity" or "support", again, reasonable minds will differ. Suppose we are in court and hearing witness evidence of fact. First witness up firmly declares that the car that knocked the pedestrian down was red. Then comes another witness who equally firmly declares that the car that knocked the pedestrian down was blue. Does the court have a problem with the "clarity" of the testimony from the first witness or with support for it? Likewise, if the claim recites "red" and the inconsistent description declares "blue", does the tribunal have a problem with clarity or with support?
Max, with absolutely no disrespect intended, I do not think that you are asking the right questions.
DeleteThe mere existence of an "inconsistency" (whatever that may be) between the description and claims does not equate to a ground for objection under the EPC. This is because there are no provisions within the EPC which address an "inconsistency" per se. Instead, the only seemingly relevant provisions in the EPC are found in Article 84, namely the requirements for clarity (of the claims) and support (by the description for the subject matter of the claims).
I therefore believe that identifying an "inconsistency" and asking why that should not be removed is putting the cart before the horse. Instead, it is first necessary to ask whether an "inconsistency" gives rise to an objection under Article 84 EPC. Also, for the reasons outlined in my previous comment, I believe that the GL indicate that an "inconsistency" is only objectionable if it gives rise to a lack of clarity (of the claims).
It is perhaps not inconceivable that a lack of clarity may arise in the scenario you described. For example, there could be a debate over whether, in the light of the disclosures of the application or patent, the skilled person would interpret "A+C" as being a sub-category of the genus "A+B". However, if there are clear distinctions between the two, I believe that the principles of UK case law that I cited in my previous comment will mean that no court could possibly reach the conclusion that "A+C" falls within the scope of a claim to "A+B".
The same logic applies to "claim-like clauses": it is inconceivable that any court of an EPC Member State would ever confuse such clauses for actual claims. Indeed, even a court unfamiliar with the anatomy of a patent will be provided with a clear pointer to the correct section - by the heading "Claims" that the EPO ensures is present in each published patent (application).
It is therefore nonsensical for the GL to suggest that Article 84 EPC requires the deletion of all "claim-like clauses", including even those which relate to subject matter of the claims. Such a suggestion is based upon an implausible hypothesis that finds no basis whatsoever in the black-letter law of the EPC.
Fair enough, but Proof I am inclined not to give up the argument quite yet. Making use of my red car/blue car hypo, what if the claim, in itself clear, says "red car" and the description has a red car worked example to support that claim. But the description also includes a blue car example of "the invention". Some would say that Art 84 is nevertheless satisfied. The ongoing presence of the blue car example does not, in itself, deprive the claim of its support given by the red car example.
DeleteBut others would say that, even when the red car claim is in itself clear, the description, in stating that a blue car is as much the invention as the red car, falls short of "support" for the claim, just like the second witness, swearing blind that it was the blue car that knocked the pedestrian down, deprives of "support" the testimony of the first witness who is equally sure that it was the red car. The evidence of the second witness removes any support for the bestowing the consequences of the knock down on the red car driver (or red car claim).
Max, I am a fan of purposive (ie teleological) interpretation of legislation. For this reason, the idea that "supported by" in Article 84 EPC means "completely aligned with" is a non-starter.
DeleteA WIPO study that considered the support requirement (as well as sufficiency and written description; see https://www.wipo.int/meetings/en/doc_details.jsp?doc_id=307341) concluded that:
"The meaning of the term “the claims shall be fully supported by the description” is largely similar in most jurisdictions. In general, the term means that there must be a basis in the description for the subject matter of every claim and that the scope of the claims must not be broader than is justified by the description and drawings. The examination guidelines of some offices also add that the scope of the claims must not be broader than is justified by “the contribution to the art”".
To me, that seems a very sensible interpretation of the meaning of, and the purpose behind, the support requirement. However, note the repeated emphasis on "must not be broader". This clearly indicates there is no lack of support in cases where the claims are narrower than justified by the description (and/or its contribution to the art).
As I have previously pointed out, it is difficult to envisage a scenario under which an alleged lack of support would cast doubt upon the scope of the claims. Therefore, provided that the claims meet the "provide basis for" and "not broader than" requirements mentioned above, it is hard to see what possible practical purpose could be served by insisting upon elimination of all "inconsistencies" between the description and claims.
Interpreted purposively, the support requirement of Article 84 EPC therefore does not support EPO practice as described in the first line of case law cited in the referring judgement.
That I find persuasive. When thinking of the EPO dogged insistence on the necessity of conforming the description you remind me of that old joke about the OAP driving the German Autobahn and receiving an anxious call from his wife warning of a Geisterfahrer driving the wrong side of the central reservation. Replies the man, it's not just one, dear. There are hundreds of them.
DeleteThe joke about the “Geisterfahrer” has been brought by Max Drei in a comment on T 1999/23 published on my blog. Let us give credit where credit is due.
DeleteWe will have to see at the end of the day, who will end up as “Geisterfahrer”-
To follow your example Max, both the red car and the blue car are supported.
DeleteThe fact that the applicant chose to only claim the red car, does not diminish that support. The fact that the applicant mentioned that the blue car is also according to the invention is not per se a problem, but a minor inconvenience, easily resolved by the understanding that 'the invention' in the inventor's mind (several years ago) does not need to be equal to the claims as they are now. All it takes is to accept the primacy of the claims.
A true clarity issue, on the other hand, exist if 'red' itself is undefined. For instance if dark-orange or reddish-brown also exist. Staying with the furry topic of the blog: A clarity issue also exists if a 'red squirrel' is not only a specific animal species, but also a grey squirrel with dyed fur.
If I were a person inclined to lay bets on outcomes, I would place a large sum on the outcomethat the EBA will dismiss Q3 as inadmissible. My hunch is that the EPO (management, working with EBA members) will prefer to sit on its hands and watch (for a few years) how the thorny issue of "conforming the description" to an otherwise allowable claim of a patent application plays out in the aftermath of G1/24's command that an ED, weighing up the allowability of any given patent application, must in the process be sure always to "consult" the description. In this context, what does that unremarkable and inoffensive little word "consult" mean? The longer time goes on, the better we shall all understand what shall be the consequences of any such consultation.
ReplyDeleteCompare "technical". Americans amuse themselves, lamenting that the EPO chooses not to provide any definition of the term. Does that matter though? Is it not a good thing, that there is no definition? We know what "technical" means, don't we, every succeeding year more so, even
in the absence of a definition handed down to us by the EBA.
Good point, Robot. Since the Guidelines were amended in 2021, the EPO has invented and is abusing the concept of “inconsistency” without a clear consensus about its meaning.
ReplyDeleteI wonder if this will address what I see as a lacuna in the EPC in that "claims not supported by the description" is not a ground for opposition, so if the examiner doesn't notice it during examination, nothing can be done.
ReplyDeleteAround 30 years ago the company I worked for was threatened with infringement of three of their main competitor's patents which, if upheld, would have prevented them selling a new product vital to their continued existence.
One of the patents had claims that were totally unsupported by the description, but which read on perfectly to one of the embodiments in one of their other patents, making me wonder if they had got mixed up before filing.
The claims were perfectly clear in themselves, and I presume that the examiner had searched for what was claimed and had only given the description a cursory examination. The embodiments involved electronic circuitry, specifically comparatore and logic circuits, where you needed to read the description, particularly the waveforms of the signals appled to the various elements, to understand how they operated. Before joining the patent profession I had been an electronics engineer designing exactly that type of technology, and without going into technical details, saw immediately that the claims related to "chalk" and the description to "cheese". The case was settled out of court, so that and a couple of other issues were never judicially considered by the UK courts.
I have commented the decision in my blog.
ReplyDeleteI can agree with Dr Hughes that a lot will depend on how the boards will interpret the decision of the EBA once G 1/25 is issued. I take bets that there will be a lot of amicus curiae. It also depends whether G 1/25 will elaborate on the topic or if the EBA will not say much like in G 1/24.
When we see how G 2/21 is interpreted in different ways, the future could look gloom. With G 1/23 we have not yet see any interpretation.
With G 1/24, there are already a few decisions and it appears that inconsistencies between claims and description are not ignored, but the boards consider that, when a claimed feature is clear as such for the skilled person, the description does not allow to give a granted feature, clear as such for the skilled person, a more limited or a broader meaning. In other words, the notion of the description being the dictionary of the claims has lost ground. A teleological interpretation of the EPC will not change a lot.
On the other there is only a handful of decisions which go against the general understanding of the boards according to which the description does not need to be adapted to the claims. Like for disclaimers or the golden standard, it could well be that the G 1/25 will confirm the general view of the boards.
When people claim that they do not understand what inconsistency might mean, let’s remain serious. Please have a look at the Agfa/Gucci decision of the UPC-LD Hamburg cited in the referring decision. This is not a decision from a TBA but from a different court.
When for a skilled person the claim means A, then trying to give in the description a more limited meaning, A-, or a broader meaning, A+, the net result is directly and unambiguously an inconsistency.
I have difficulties to see why Question 3 is not admissible. Examination and opposition procedures are separate procedures and it is not unreasonable to ask the EBA if a difference should be made. The question is formulated in such a way that the most likely reply should be NO. Why should a proprietor get a better treatment as when he was applicant?
ReplyDelete' When for a skilled person the claim means A, then trying to give in the description a more limited meaning, A-, or a broader meaning, A+, the net result is directly and unambiguously an inconsistency. '
Exactly there lies the problem. Of course, what you say is true. But what others are saying, including me, is that the inverse is not necessarily true.
And what you say is not even a problem:
If the description says that the meaning is A-, that is a risk the patentee took and we will see whether the court applies A or A-, the outcome for the patentee is neutral or negative (as is befitting)
If the description says that the meaning is A+, this is firstly an error by the examiner, and secondly an error by the court if it follows this (or a bit of both). The outcome for the patentee will be neutral or positive in this case.
Now you may argue that this potential 'positive' is unwarranted, and therefore unfair. But I don't agree. The issue of whether this 'positive' is or is not unwarranted has not been examined. Perhaps the applicant honestly feels that a positive opinion by the examiner is correct, even if using the A+ interpretation. Is it fair to the applicant that he should bear the consequences if the examiner misses something? Is it fair to the applicant that he should foresee that the court may make a mistake in future and already limit his (wider, A+) scope (to which he honestly thinks he is entitled) just to reduce it the potential of a future mistake by the courts?
@ Anonymous of Thursday, 31 July 2025 at 08:40:00 GMT+1
DeleteWhether A- or A+ has not been spotted by the examiner is an error and the examiner has not done his job properly, as he has manifestly not consulted the description, to stay with the wording of G 1/24. I would thus not limit the examiner’s mistake only to A+.
I hope that, thanks to G 1/24, the tendency of not reading the description by some examiners in order to gain time will come to an end. I do not blame examiners for playing with the system, sometimes encouraged by their superiors.
I can agree with you that the applicant/proprietor should not suffer from a mistake made by an examiner. However, in case the examiner has missed an objection of added-matter, the applicant/proprietor will, at least in opposition at the EPO, suffer from the examiner’s mistake. The mistake is however shared between examiner and applicant/proprietor, as the extension of object has been introduced by the latter.
In the A- situation, the proprietor will try to fob off a patentability attack by claiming that his claim should not be interpreted broadly according to A, but when it comes to infringement he will argue that A is the way to look at the claim.
In the A+ situation, the proprietor will try to fob off a patentability attack by claiming that his claim should be interpreted narrowly according to A, but when it comes to infringement he will argue that A+ is the right interpretation. This is what is called nurturing an Angora cat.
There is no asymmetry between both situations, and to my understanding both are as harmful.
Whether A- or A+ is taken compared with A in the claim depends indeed on how the court deals with it post grant procedures. It is actually in order to avoid a court to chose with possible variants of the meaning of the claims that it exists Art 84, support.
This will not hinder a court in post grant procedures to have its own interpretation of claimed features, and conclude to infringement by equivalents, but when the patent is granted by the EPO, there should be no inconsistency between claims and description.
This is why I consider it necessary to adapt the description to the claims before grant and after maintenance in amended form.
I agree to the extent that there should be no blatant contradiction between claims and description. But mere potential inconsistencies are simply a fact of life for the courts to deal with, as they do in many other walks of life.
Delete'This is why I consider it necessary to adapt the description to the claims before grant and after maintenance in amended form ': In my view you want to make 100 applicants jump through hoops just to avoid 1 problematic inconsistency. This 'jumping through hoops' comes at a financial cost to applicants (also the other 99 applicants), and at the expense of a (real or perceived) weakening of their legal position.
And all that while other, bigger problems (see: the proportion of patents at least partially revoked in opposition) are not solved.
@ BJ of Friday, 1 August 2025 at 07:38:00 GMT+1
DeleteI will not repeat what I have said here more than once, but the problem lies in the notion of “blatant” or not. I reminds me on the problem of added matter as dealt with by the EBA or the TBA. There is not a blatant or a non-blatant infringement of Art 123(2) for the EBA and the TBA.
The EBA and the TBA have set up a long line of case law, which might be held too severe by some, but it has the merit to be clear and easy to apply. Any new information added to an existing disclosure is to be considered added matter. The EBA and the TBA have taken a strict position in the interests of third parties.
If the EBA takes in future the position that the description has to be adapted, this will certainly be in the interests of third parties.
However, we should not mix the issues. The proportion of patents at least partially revoked in opposition has little to do with the adaptation of the description.
On the other hand, when looking at published decisions of the boards after appeal on opposition you have a point. Only 15% of patents survive an opposition, 40% are revoked and 35% maintained in limited form. The problem is that in less than 8-10% of the cases, opponents came with prior art which was truly not available in EPO’s search documentation. That means that in the vast majority of cases, the opponent came with documents which were present in the search files but not found.
This is a real problem, and certainly greater than the problem of the adaptation of the description, but this does not imply that the adaptation of the description is to be forgotten.
Let us give some thought to legal certainty, in the area of writing a "freedom to operate" opinion for Europe, and how it might change if the EPO forswears its present insistence on strict conformity.
ReplyDeleteDoes the opinion-writer have an easier job, here in Europe, because the EPO insists, as a condition of allowance of every patent application, that the description be strictly "conformed" to the claims? I'm not convinced.
Consider two situations, A and B, for the opinion-writer. In A, the case got to grant with narrower claims but an untouched description. In B, the case was granted with a fully conformed description. How much more confident can the opinion-writer be, in situation B, that any given device, accused of infringement by equivalent, is or is not going to be found by the court to infringe. Perhaps the notorious pemetrexed case is helpful, but perhaps not.
And then there's the validity prong. Given the potential that conforming has, to muddy the water on validity, does any increase in confidence about answering the question "Infringe Y/N?" that comes with full conformity outweigh all the negative consequences of insisting upon strict conformity?
Some of the answers in this thread provide food for thought.
Oh and Daniel, thanks for sticking up for Max Drei on the Autobahn. But I must confess that I was (accidentally) the poster who mentioned it above and forgot to identify himself as Max Drei.
Max, I take the position that it is the EPO's job to grant (or refuse) patents in accordance with the provisions of the EPC. Those provisions have been calibrated to balance the interests of third parties with those of patentees.
DeleteAs expressly stated in some BoA decisions, the purpose of the EPO's current description adaptation practice is to ensure that the claims are interpreted in the same way both before and after grant. However, there are so many fundamental problems with that objective that it makes my blood boil.
What happens to patents after they are granted is not the EPO's concern. It is simply outside of their remit. If they have done their job properly with regard to ensuring compliance with the EPC, they have done all that they can (and should) to ensure that justice is done.
The EPO is entitled to their own view on which subject matter is encompassed by the claims of a patent granted by them. However, as illustrated by the high combined rate of invalidation / amendment in opposition, the examination process at the EPO is far from infallible. We can therefore conclude that the claim interpretation adopted by the EPO will be wrong in a significant percentage of cases.
Given the limited time and resources allocated to the examination process, it would be a miracle if the claim interpretation adopted by the EPO were found to be precisely correct more often than not. This is not a criticism of EPO examiners, but instead merely a recognition of the fact that it is unreasonable to expect that, in every case, an examiner will be able to adopt the mantle of a person of ordinary skill in the very specific technical field to which the invention relates (or to be aware of all relevant common general knowledge).
So if it is reasonable to assume that the EPO will get things (at least slightly) wrong in most cases, why should applicants be forced to mangle their description to align with an interpretation that is likely incorrect?
By forcing applicants to do this, is the EPO not putting its fingers on the scales of justice and tilting the balance much too far in favour of third parties (by attempting to "simplify" FTO for them, at no risk to the EPO and third parties but by generating very significant invalidity risks for patentees)?
National courts will interpret claims by reference to the description of a granted patent. Especially in view of the likelihood that the EPO did not get everything 100% correct, the interests of justice dictate that those courts should be free to reach their own decisions on the subject matter encompassed by the wording of the claims. It takes quite some hubris for the EPO to decide that it would not be acceptable for the national courts to reach a different interpretation, and to effectively force those courts to agree with the EPO by presenting them with a description in which the EPO's claim interpretation has effectively been imprinted throughout.
We shall have to see whether the EBA agrees with me that there is simply no justification for a practice that not only improperly tilts the scales of justice but that also prevents (or at least hinders) fully independent reviews by national courts of the EPO's work products. I would have hoped that all TBAs would appreciate these concerns, so I find it alarming that many do not.
I fully agree with you and I could not have worded it better than you.
DeleteHowever, it occured to me that the EPO does, after all, have one reasonable interest in requiring amendment of the description. This is: to signal to other actors in the patent system what, exactly, is the EXAMINED scope of the claims (with or without examination errors, but that is an entirely different issue), irrespective of whether this aligns with the actual scope of protection offered by the claims.
All other users of the patent would arguably benefit from this knowledge.
Now whether that should override your important considerations....? I would say no, but other might say yes.
Hard to take issue, Pudding, with anything you have written today.
DeleteFor the avoidance of doubt on the issue of the Geisterfahrer, I do not say that one or other commentator here is one. My point is that, alone of all the world's Patent Offices, the EPO sees fit to enforce, as a condition of allowance, wholesale revision of the text of the description as filed to "conform" it with the allowable claim. That said though, there are cases where the description flatly contradicts an otherwise clear claim. To leave both the description and claim unamended in such cases strikes me as tantamount to prejudicing the clarity of the claim. A good example, I think, is the "gathered" case. I think I am clear as to what "gathered" in the claim means to me and it is not what the description declares it to be. Otherwise the pemetrexed case. If the claim recites "sodium" its clarity is not prejudiced if the description declares "the invention" to be something broader than that. There are three concepts here. They are i) the invention ii) the subject matter for which protection is sought and iii) the scope of the protection conferred by the claim after it grants. These three concepts are not co-terminous, not the same. Who says they are? The EPO? Anybody else?
Those of the Patent Offices of the world that purport to examine patentability substantively, as a service to the public, need help, with a methodology to judge whether any given claim is "clear" or "definite". The EPO has a role to play here but its present approach seems to me to be unlikely to find broad acceptance. Will the UPC encourage the EPO to continue on its present course? I doubt it.
Max, I think that you have correctly identified that clarity could be the (only) issue that arises when there is an "inconsistency" between the description and claims. In that scenario, the appropriate response would be to amend the claims, assuming that this is possible and that it is necessary to ensure a clear distinction over the disclosures of the prior art.
DeleteOf course, a clarity objection may not be available after grant. But that should merely prompt the EPO to read the description properly, and to check for unusual definitions and/or missing essential features. This ought to be perfectly possible for the EPO, not least because it is (and always has been) one of the tasks that they need to complete for the purpose of ensuring compliance with Art 84 EPC.
@ BJ,
DeleteIt will not come as a surprise that I do not agree at all with Proof of the Pudding, and hence with you.
The “important considerations” are to be respected for what they are: no more than an opinion. An opinion which might be shared by lots of representatives, but which does not correspond to the opinion of the vast majority of the boards of appeal.
G 1/25 might give an answer, which might displeased one or the other group, but at least then, it should settled the matter, unless the reply to the questions is neither fish nor fowl and gives raise to divergent interpretations of the boards.
At least for G 1/24, the boards are quite coherent in their views, the description has to be consulted.
Indeed MaxDrei, as you say there are three concepts: i) the invention ii) the subject matter for which protection is sought and iii) the scope of the protection conferred by the claim after it grants.
DeleteWhat EPO has been doing for some years is to force applicants to adapt i) to become the same as ii). But the justification, repeated time and time again, is that ii) should be the same as iii). That is simply not logical.
In an ideal word, ii) should indeed be the same as iii). But those that determine ii) seem to be all too keen to pre-emptively bend over backwards to please those that determine iii). And that's another fundamental wrong, in my view.
DXThomas, you seem to forget the following:
DeleteBy means of the EPC, the contracting states have signed away part of they souvereign authoritity to the EPO. They have signed away ONLY what is in the the EPC, not an iota more.
The only role of the EPO is to execute the EPC.
This means firstly that there is no scope for the EPO/President to play a role in defining the patenting system (outside the provisions of the EPC). It is simply of no concern to the EPO that judges may have difficulties in interpreting patents, or that the whole patenting system outside EPO collapses, they only have to execute the EPC. All the rest is for the legislator to handle. You may not like this, but this is the legal situation according to the EPC. The mere observation that there is a potential problem with the scope of protection of European patents is NOT enough for the EPO to take action.
Secondly, this means that, via Art 125 EPC, it is highly relevant how national offices handle such matters.
Thirdly, this means that if EPO wishes that applicants adapt their description (even though, see above, it is not up to the EPO to have such wishes), there needs to be a provision in the EPC that indicates that this should be done. An implicit, indirect provision, indicative provision does not count, because the authority of the of EPO is determined by a narrow interpretation of the EPC (narrow, because it is inconceivable that the contracting state should have given away more of their souvereignty than they really wanted, as expressed by the explicit provisions of the EPC). Such a provision is missing.
Perhaps a reasoning can be made based on Art 84, in combination with Art 69, and refering to a public interest to have an unambiguous scope of protection, but this is insufficient to justify taking more authority by the EPO than the contracting states were prepared to explicitly sign away.
Yes, of course non-perfect alignment between scope of protection and scope of examination is undesirable.
But the contracting states, by means of the EPC, did not allow the EPO to solve this, not even to have a desire to solve this. And even if EPO wishes to solve it, it legally has no choice but to conform its practice to what national offices do.
BJ, I would say that the EPO's situation is not as clear-cut as you make out. In opposition proceedings, for example, when a granted claim is amended, Art 123(3) EPC comes into play. When deciding whether the amendment is allowable, shall the EPO ignore Art 69 EPC?
DeleteAnd as to the relationship between the EPO and the UPC, up until now the EPO has had the pan-European playing field all to itself. What we are now seeing might be a power struggle. Who gets to make the law of patent validity in Europe, going forward? Has the EPO already given up the struggle to be top dog? I doubt it.
Max, as I have argued above, the EPO has no remit to even engage in such a power struggle, so I can not have any sympathy or understanding for them if do so.
DeleteAnd G1/24 clearly seems to want to align EPO practice with UPC decisions, which seems a peculiar way of fighting a struggle, if the first thing you do is 'we'll follow what you say'. But may it all goes above my head...
@ BJ of Friday, 1 August 2025 at 07:51:00 GMT+1
DeleteIf there is one person who is aware of the fact that, “by means of the EPC, the contracting states have signed away part of their sovereign authority to the EPO”, and that the “only role of the EPO is to execute the EPC”, it is certainly myself, having been there from the opening of it in 1978 until the end of 2012.
I would like to draw your attention to the exact wording of Art 125:
“In the absence of PROCEDURAL PROVISIONS in this Convention, the European Patent Office shall take into account the PRINCIPLES of PROCEDURAL LAW generally recognised in the Contracting States”.
For a start, Art 125 is thus only applicable in procedural matters, and then only if some procedural rules are missing in the EPC, general principles of procedural law recognised in the Contracting States shall be applied.
You are not the first person forgetting that Art 125 does only deal with procedural matters, and Art 125 can in no way be interpreted to apply on substantive matters, like for instance, patentability, clarity, sufficiency and added matter.
What we are discussing in G 1/24 and in the referral under G 1/25, are matters of substance. Art 125 is thus not applicable at all. Contracting states are free to apply and interpret the EPC as they think fit, and the EPO is by no means obliged to follow decisions in substance of jurisdictions acting in post grant procedures like nullity or validity.
I thus allow myself to tell you that you are fundamentally mistaken when you claim that the EPO has, “legally [has no choice] but to conform its practice to what national offices do”.
Should you have forgotten it, there is one legal basis, contrary to what T 56/21 has alleged, which obliges the EPO to request adaptation of the description, that is Art 84 with its two prongs, clarity and support. It is the duty of the EPO to avoid any inconsistency between claims and description.
I hope that the EBA will comfort the position of the EPO and its boards in this endeavour.
If the claims are narrowed down during examination, the original description will be broader but still necessarily covering (and supporting) them. One may then “consult” the description (G1/24), whatever this means, but still the claims have “primacy” and overrule. Where is the inconsistency?
ReplyDeleteMaxDrei, you write that for you the meaning of the term 'gathered' in the claim of the patent leading to G1/24 is clear to you and differs from that included in the description. The meaning of 'gathered' taken by the opposition division was, as I understand it, derived from terms used in sewing, rather than articles for smoking. Is 'rolling' not a standard process of gathering tobacco leaves together for manufacturing cigars? Are those Torcedoras and Torcedores to be put out of a job?
ReplyDeleteL’amour est enfant de Bohême
Il n’a jamais jamais connu de loi
Your point, Kant, escapes me. I am neither a person skilled in the art of sewing nor one inthe art of fashioning smoking materials out of tobacco leaves. But I do know what "gathered" means to me and it does not include "rolling". I am not an expert in the "gathered" case at the EPO but it was my understanding that i) gathered in the claim gave it patentability over conventional rolling but ii) the description defined gather in a way that included rolling. Is that not a situation crying out for clarification by an EPO Exami ner?
DeleteMy point was that the word 'gather' has multiple meanings depending on the context, from gathering your family together, forming a pleated structure, or simply co-locating. If the applicant indicates in the specification which of these meanings is meant in that document, the organs of the EPO should be wary of applying a different meaning based on their own particular interpretation.
DeleteRien n′y fait, menace ou prière
L′un parle bien, l'autre se tait
I'm still struggling. You are right, Kant, that the word "gather" is used in a plethora of different contexts. But in all of those, it signifies the same action, does it not? Carpet layers roll up and unroll a carpet but you cannot reasonably assert that in so doing they "gather" the carpet. Likewise those who roll tobacco leaves. So, when a description sets forth a definition of the claimed "gather" that includes the prior art and conventional "roll", we have in the specification an internal problem with clarity, don't we?
DeleteOn further reflection, Kant, on your "gathering" point, and about how cigars are made from tobacco leaves, perhaps one can begin the step of rolling the leaves only after one has completed a step of "gathering" them. In that case though, gathering, in this particular context, being notoriously old, is no help towards patentability.
DeleteI think that the problem is maybe that some people kind of equate "the invention" with "what is being claimed". Now, in what regards an invention, the EPC does not even require an invention to be novel and to involve an inventive step. Thus, I see no contradition between the description referring to "A+C" as an embodiment of the invention (that is, the invention that the inventor believed to have invented in the first place) whereas the claims (limited to what the inventor and the Examining Division agreed was novel and inventive) is limited to A+B. I think that the problem may reside in the fact that the requirement for "support" for the claims in Art. 84 actually mirrors the requirement for sufficiency of disclosure of Art. 83, and that some persons may have considered this duplication to be "non-logical" and then have strived to find a more "reasonable" interpretation of the requirement for support, ending up interpreting the requirement for support to be a requirement for "consistency" between description and claims. But Art. 84 does not mention consistency.
ReplyDeleteFor an engineer, a requirement for "support" typically implies that there is "sufficient base". There is nothing wrong with "a base that is wider than what is being claimed".
That is, I tend to think that the problem is a confusion between "the invention" (in broad terms: what the inventor originally inventived) and the invention as finally claimed. The reference to "A+C" as an embodiment of the (originally conceived invention) is, in my view, not incompatible with the fact that the claims are limited to A+B. I think that basically, the problem is artificial: at some stage someone thought that any reference to "an embodiment" (of the invention) had to be interpreted as an embodiment of the invention as claimed.
Now, Art. 69 indicates that the extent of the protecion conferred by the patent shall be determined by the claims. So if what is claimed is A+B, why should it matter that the description refers to A+C being an embodiment of the "invention" if this is not what is claimed?
It seems to me that the current situation (the alleged need for "consistency" between the description and claims, in the sense that if what is claimed is A+B, A+C should not be stated to be an embodiment) is the result of some kind of strange legal or pseudo-legal reasoning of the type that in other jurisdictions have made practitioners reluctant to even use the word "invention" in patent specifications.
So I really hope the BoA will state that there is no need for this adaptation of description to claims. Of course, as a patent practitioner, I should maybe hope for the contrary, as adaptation may give rise to more billing hours... and to Art. 123(2) issues that may give rise to even more billing hours...